Talk:Do-support

Introduction

The introduction has been added to ensure that encyclopedic need (need to include information on a commonly mentioned topic) does not purport support of an erroneous concept.

I have undone the edit by USER:Angr (edited part of the introduction out citing removal of opinion) because the information provided is not opinion.

I understand that it is counter to the information provided in the body of the article and is in fact counter to what is included in the majority of grammar texts (especially those published by ESL publishers), but in fact it is not opinion. The idea of "do-support", "pro-verbs", and "dummy verbs" are however opinion. They are concepts based on fallacious analyses of the English language and are known to be non-functional in most instances.

Having no linguistic basis, and being the cause of so much confusion and frustration among learners of English, my preference would be to delete the article entirely as I have spent 20 years dealing with the damage caused to learners by promoting these ideas. However, these ideas are common, and as such, people will see this words out there and should be able to look them up somewhere. It is important though that when they do encounter an article on them, especially from a trusted sources such as WP, that a disclaimer be provided that both establishes these concepts as what they are while also providing a brief explanation of the correct analysis and justification for classifying them as linguistically incorrect.

Please do not revert without further linguistic discussion of the information provided. Drew.ward (talk) 14:30, 13 June 2011 (UTC)

Please read WP:BRD. Making a bold change is fine (like you did), but when someone else reverts that change (like I did), the next step for you is to discuss without reverting back to the contentious edit. As for the content of the page, it's simply a descriptive fact of English that do-support occurs in questions and negatives when no other auxiliary is present. You say it has "no linguistic basis", but it has a clear linguistic basis, namely the following data, whose accuracy I trust you do not deny:

(1a) John rides the bus.

(1b) John does ride the bus.

(2a) John rides not the bus.

(2b) John does not ride the bus.

(3a) Rides John the bus?

(3b) Does John ride the bus?

In each pair, the (a) sentence has no do-support and the (b) sentence has do-support, and that's all the term "do-support" refers to: the replacement of a finite form of a lexical verb with the corresponding finite form of do followed by the "base form" of the lexical verb. In English, the unmarked sentences in each pair are (1a), (2b), and (3b), while (1b) is used only for emphasis, and (2b) and (3b) are archaic but may still occur in poetry. This isn't fallacious, and it isn't a lie to ESL/EFL students; it's just a fact of English grammar. The term "do-support" by itself does not make any claim as to why or how this substitution occurs. It occurs in English under certain circumstances, and it occurs in other languages under other circumstances (a fact which needs to be discussed in the article; for example, in Manx, finite forms of lexical verbs occur very rarely: in most circumstances, a finite form of do or be is used in conjunction with the verbal noun, so in that language, do-support is more widespread than in English). Now, linguists may argue about the cause and analysis of do-support (is it caused by insertion where it appears, or is it there from the beginning and then suppressed where it does not appear?), but argument about the theoretical analysis of do-support doesn't change the fact that it exists as a phenomenon in the grammar of English (and other languages). —Angr (talk) 21:06, 13 June 2011 (UTC)

First, full disclosure: I created this page. That said, I agree that it is insufficient and needs a more world-wide view, among other shortcomings.

Second, to the substance of this dispute: I do not agree that material added by Drew.ward is an improvement in terms of the coverage or information of this article. Since the additions were unsourced and appear to reflect original research by user Drew.ward, they should therefore not be added to the article.

Third, to the notions of "encyclopedic need" and the role of subject-matter experts: Wikipedia policy is not to publish new analyses that its editors deem superior to previous ideas. (See also the essay Wikipedia:But it's true!) Instead, Wikipedia reflects the consensus view and main minority views of previously published work. (See also the policy statements Wikipedia:Core content policies and Wikipedia:Verifiability.) In this sense it is more like a textbook than a scientific journal. Subject matter experts (and by the way, it is my understanding that both Angr and Drew.ward are professional linguists, as am I) should verify that content on Wikipedia pages accurately reflects reliably published sources, and that those sources do not represent fringe theories within our fields. However, subject matter experts should never succumb to the temptation to underplay theories we disagree with nor to label them as fallacious. (See also the policy statement Wikipedia:Neutral point of view.) Cnilep (talk) 00:39, 14 June 2011 (UTC)

Two subtle questions

Since a recent correction, this page declares, "Except in copular sentences such as Is he here? or They are not banjo players (and except in have questions like have you any bananas, which are permissible in some but not all dialects), in English all questions and all sentences with negative polarity feature an auxiliary verb."

This is not strictly true, as exemplified by Angr's examples above, or literary usages such as "No, no, go not to Lethe". Is it worth getting into the particulars, I wonder? The pre-correction version was no truer: "In English, all questions and all sentences with negative polarity feature an auxiliary verb."

A second, related question: I added (originally in a footnote) the bit about "Have you any bananas?" I know that some speakers use main-verb have without do, and that it sounds slightly odd to me. I wonder, though, whether this is a case of WP:You don't need to cite that the sky is blue, or if some readers or editors may doubt it?　Cnilep (talk) 03:07, 1 August 2011 (UTC)

Regarding your first point, I actually thought of that when I wrote it, but wrongly decided to ignore the out-of-the-ordinary constructions in the interest of keeping the sentence from getting too long. I'll put in "almost" for strict correctness. However, I think that "John rides not the bus" and "Rides John the bus?" are poor examples -- those forms would be extremely rare. A better example would be "John rides the bus?" I'll put that in as an exception.

Regarding your second point: It sounds a bit odd to me too, but only because I'm American. But I do see it a lot from British speakers -- both the form "Have you any bananas" and, more commonly, the form "I haven't any bananas". I don't think it needs to be cited, because I think any Briton would say it's a sky-is-blue thing. Duoduoduo (talk) 15:21, 1 August 2011 (UTC)

I have to agree with Duoduoduo here, "Have you any bananas?" is grammatically incorrect and thus shouldn't be taken into account. It is commonly used by some speakers particularly in parts of the UK but it is not a correct form. It has come about due to speakers confusing have as a perfecting auxiliary "You have eaten -- Have you eaten?" with have as a main verb "You (do) have a cold -- Do you have a cold?" but not "Have you a cold?". Drew.ward (talk) 01:59, 12 October 2011 (UTC)

Under normal circumstances "commonly used by some speakers" is virtually equivalent to "grammatically correct", at least within traditions of descriptive and theoretical linguistics. See Linguistic description and Grammaticality, inter alia. Cnilep (talk) 04:12, 12 October 2011 (UTC)

Just because an error is common does not make it grammatical. I can go out into the street and here "She be hatin on him" but to include it in a reference to English grammar would be wrong because no matter how many people say it, it's grammatically incorrect. Drew.ward (talk) 13:41, 12 October 2011 (UTC)

No, we have to be descriptive here. There are dialects in which "Have you any bananas?" is grammatical and other dialects in which it is ungrammatical or at least very odd. It would violate both Wikipedia's NPOV policy and all standards of scientific endeavor for the article to call "Have you any bananas?" an error. (Incidentally, I'm not a speaker of British English, but I believe that most Brits would be more like to say "Have you got any bananas?" than "Have you any bananas?".) Angr (talk) 20:55, 12 October 2011 (UTC)

Being descriptive does not include misinforming readers. It's obvious that "Have you any bananas?" is incorrect in Standard English. This is not an article on language theory, and obviously there are too many "dialects" to describe here. How is it NPOV to describe only one when we can admit that it's not Standard English? I think we've already covered ourselves with the "almost" qualifiers, and we don't need to get into non-standard usage. Please remember that this page is very useful for ESL students and teachers, and that "Have you any something?" is a common mistake among beginners and intermediates. David.f.dana (talk) 16:21, 22 June 2012 (UTC)

Since there is no single monolithic "Standard English", you cannot say it is "incorrect in Standard English". English has more than one standard, and "Have you any bananas?" is correct in some standard varieties of English, and incorrect in other standard varieties. Angr (talk) 18:05, 22 June 2012 (UTC)

• OK I just reworded that paragraph to include the colloquial "Have you a..." example but to label it appropr'ately. I also rewrote most of it, but kept the same points. As most of you know, I don't buy into this idea of do-support and it's not something that could in the slightest bit be linguistically justified. That said, for some reason the consensus on here is that this weird idea is to be the official wikipedia version of English grammar so I didn't "correct" it as I feel would be the academically responsible thing to do. What I have done though it clarify some of the ways that first paragraph was worded to ensure that most of the missing understanding of what goes on in English verb forms is at least somewhat accounted for so that readers upon consideration of the article may themselves surmise the actual linguistic happenings that proponents of do-support attempt to account for with this theory. Hopefully this is an adequate compromise.Drew.ward (talk) 19:37, 22 June 2012 (UTC)

Do-support isn't a theory; it's a descriptive fact. You yourself just wrote "I didn't 'correct' it", using do-support. You didn't write "I 'corrected' it not". Anyway, I had to revert your changes because they were incomprehensible. I have no idea what "an unperfected, simple aspect verb" or "vocal auxiliary" means, and the sentence was long and rambling and very difficult to follow, including multiply nested parenthetical tangents. Finally, you called the American English constructions "Do you have a pen?" and "No, I don't have a pen" standard, although they are not standard in British English. Angr (talk) 07:06, 23 June 2012 (UTC)

Well now I'm unreverting it because saying you don't like something and calling it incomprehensible because you personaly don't understand the concepts are not grounds for reverting a change. If you don't know what these things are maybe you should either ask or research before rejecting them. A vocal auxiliary is an auxiliary used to signify voice (as in active, passive, or middle voice). In English the two vocal auxiliaries are BE and GET with BE being far more common than GET in that usage. Unperfected means that the verb has not been perfected - or in simple terms that the grammatical structure of the verbal construction has not been preempted by the perfecting auxiliary HAVE with the following verb put in the past participle form. If you don't know simple aspect is, I really don't think you need to be arguing anything on here...

I (and the others you're opposing on here regarding the 'have you bananas' issue) have not asserted preference over American or British varieties. The use of have without auxiliary do is neither standard in the UK nor in the US. It is however a colloquial construction found in both varieties albeit all but limited to older literary sources in American varieties these days. Just because you or people around you use this construction does not nor has it ever made it standard. No one's calling wrong, or incorrect, or less desirable. Get over yourself. It's not standard English; it's colloquial and when you consider the small number of speakers out of the 400-500 million native and near native English speakers in the world, it's rare. Providing reference to it while classifying it as a regional colloquialism (which it is) is the best compromise possible while not confusing readers or destroying the quality of this article.

Finally, if as you say, "Do-support isn't a theory; it's a descriptive fact," then prove it. I have neither edited out the content of this article asserting the acceptance of this theory, nor have I presented an alternative. I know it's not correct and I know that's true because it cannot be linguistically proven because the only way to buy into the idea of do-support is to ignore huge chunks of well-proven linguistic attributes of English grammar. I've already fought the fight on here to have this article label do-support' as the non-universally accepted theory that it is. The consensus on here has been that despite ample proof (easily explained proof) that it's incorrect, that wikipedia won't classify it as anything but true and the only wiki-approved explanation of what DO is in English whether it works or not. If you are so sure of yourself, why don't you do what no one else on here has ever been willing to do: provide irrefutable proof that do-support is correct. Proof by the way is not sources or citing it's being mentioned on websites or in books or in grammar guides. Proof is actual, accepted linguistic proof that provide an explanation that works 100% of the time, has no exceptions, cannot be disproven, and does not require a specialised version of syntactic theory that applies only to the English language. Go ahead. Let's see you do it.

I'm guessing you'll just take the easy way out, throw a fit approach that everyone else on here takes when called out and pack up your toys and say you don't have to prove anything because this is what it is or some other excuse. I'm hoping though that you'll be a better man (or woman) than that and take the effort to stand by your claim and provide proof. Let's see...

I am going to now revert back the change to the last version I added. In doing so, I will however look through it and see if I can reformat it to be clearer or more easily read. Until you can take the time to reply to the points above though here on the talk page, I think it apt that you not revert what I've written again. If what I've written is wrong and what you've written is right, then you should be able to easily prove it at which point I will gladly do the revert for you myself. Additionally, if the style or wording of my version bother you, then without changing the content, why not recommend or make some subtle changes as you feel would be required to rectify your perceived issues?Drew.ward (talk) 17:43, 23 June 2012 (UTC)

OK, I have reverted to the previous version, but have then moved the majority of inline examples to hover boxes and provided hover box definitions for all of the terminology that confused you. Hopefully this version is easier to read.Drew.ward (talk) 18:57, 23 June 2012 (UTC)

Once again, please see Wikipedia:Core content policies, Wikipedia:Verifiability, and Wikipedia:But it's true! It is not Wikipedia policy that English has do-support. It is Wikipedia policy that Wikipedia articles reflect reliable third-party sources; it is sources such as Traugott and Pratt 1980, Kaplan 1989, and most English grammars that assert the existence of do-support. It may be appropriate for Wikipedia editors to argue about the adequacy of those sources as sources, but Wikipedia is not the appropriate forum to argue that most existing grammars are wrong and that one or several Wikipedians know better. Cnilep (talk) 00:20, 24 June 2012 (UTC)

OK, if it's verifiable, find me a single source that can verify that do-support is linguistically sound by proving it to be correct. One source or a hundred sources that say something without proving it to be true have no more value than having no sources at all. If it's so easily clear and such a sound concept, why can't any of you find a single nugget of support other than source after source that says 'this is what it is' without themselves providing any support or defense at all? You guys all I'm sure have the best of intentions regarding the quality of articles on here, so if you do, find proof for do-support. And if you can't, either delete the article or reword it to point out that it is an unfounded theory. You guys need to be holding yourselves to the same standards as you're holding me in regard to this discussion. So?Drew.ward (talk) 01:53, 24 June 2012 (UTC)

[outdent] Here are a few.

• Brinton, Laurel; Brinton, Donna (2010). The Linguistic Structure of Modern English. John Benjamins. p. 234. So a verbal element must be supplied by inserting the dummy auxiliary do. This serves the function of an auxiliary when there is no other independent auxiliary present. This insertion transformation is called do-support.
• DeCapua, Andrea (2008). Grammar for Teachers. Springer. p. 5. Therefore, to make these sentences into questions, we need to add something to the position before the noun phrase. This "something" is the auxiliary do, which functions to "fill" the auxiliary slot before the noun phrase in question.
• Freidin, Robert (1982). Foundations of Generative Syntax. MIT Press. p. 171. The distribution of this auxiliary can be accounted for via a substituion transformation that inserts the auxiliary do (not to be confused with the verb do) into an empty auxiliary position in Infl.
• Heidinger, Virginia (1984). Analyzing Syntax and Semantics. Gallaudet U Press. p. 82. In the sample question, do must be inserted. Without do the questions would be The baby walks? and *When your guests arrived? Although The baby walks? is grammatical, and it asks a question, it does not have the syntactic form of a question.
• Huddleston, Rodney D.; Pullum, Geoffrey K. (2005). A Student's Introduction To English Grammar. Cambridge U Press. p. 152. Negative clauses of this kind require the presence of an auxiliary verb. If there is no auxiliary in the corresponding positive clause, formation of the negative involves the insertion of do as described in Ch. 3, §3.1, and illustrated in [8]
• Klammer, Thomas P. (2006). Analyzing English Grammar. Pearson/Longman. p. 261. It [the auxiliary verb do] appears only in sentences that have undergone some kind of transformation. {{cite book}}: Unknown parameter |coauthor= ignored (|author= suggested) (help)

These come from generative grammar, phrase structure grammar, and pedagogical grammars. They are not bound to a single school or approach and are certainly not fringe theories. I would suggest that any argument rejecting these and the many similar grammars of English is better suited to linguistics journals and not to Wikipedia. Cnilep (talk) 03:27, 24 June 2012 (UTC)

Those are all just statements. They don't justify it at all. They don't explain what the role of do is, how they justify treating this particular use of an auxiliary verb differently from all others in English, or how they justify using a unique version of syntax for English that can't be applied to any other language (and which only works in English sometimes). Can you provide actual research and proof?Drew.ward (talk) 05:41, 24 June 2012 (UTC)

You can't blame me for not understanding your text when you don't use words in their usual meanings. "Vocal" is never used as the adjective of "voice" in the sense of "active/passive/middle voice". The verb "to perfect" (from which you are deriving "unperfected") doesn't mean "to put into perfective aspect". When you invent your own personal technical vocabulary, you cannot be surprised when other people fail to understand you. Next, when you say I have to "prove" that do-support exists in English, I repeat what I said before: you use it yourself in your comments here. In your comments above, you wrote "Well now I'm unreverting it because saying you don't like something and calling it incomprehensible because you personaly don't understand the concepts are not grounds for reverting a change" and "why don't you do what no one else on here has ever been willing to do", with do-support at the underlined passages. How can you deny that do-support exists when you use it consistently yourself? I think you have misunderstood what the term "do-support" refers to: it does not refer to a specific syntactic theory of how "do" gets inserted into questions and negatives. Rather it refers only to the descriptive fact about English that the negative of "John goes" is "John does not go" rather than *"John goes not", and the question form of "John goes" is "Does John go?" rather than *"Goes John?". That's all "do-support" refers to. Denying that do-support exists in English would be like denying that nouns and verbs exist, or that vowels and consonants exist. Angr (talk) 14:43, 24 June 2012 (UTC)

I don't USE do-support. No one does because there's no such thing as do-support. All that do-support refers to is a misled attempt by grammarians (the main original push for spreading this idea coming from the ESL industry) who, lacking the basic linguistic competencies to understand the functions of the English verbal system in expressing aspect, mood, tense, and the like, to come up with an explanation for why (in their mistaken view) DO seemingly appeared out of nowhere in certain constructions (some questions, negatives, and some emphatics, as well as some "short forms".

This was during a time when attempts to decipher English grammar were still based off the now universally-abandoned view that English could and should be forced into the mold of the grammar of classical Latin. Further complicating things, it wasn't until around the second world war that our understanding of the ways in which Germanic languages convey aspect, aktionsart, perfection (which by the way is different from 'perfective aspect' which is an entirely different concept), tense and mood. The original works proclaiming do-support were written before this modern understanding became common knowledge. The English-grammar world ended up with two competing schools, on one hand you had philologists, linguists, modern grammarians, and experts on the language who equipped with modern scientific tools and understanding worked through much of the 20th century to rid documentation of the language including grammars of the many erroneous theories that ran counter to linguistic common sense (such as the prohibition on split infinitives, forbidding prepositions at the ends of a sentence, and many analyses based on explaining English in terms of Latin or Greek grammar). On the other side were educational publishing houses, grammar book authors, and ESL training companies all who had and continue to have a vested interest in protecting the "truth" of the fortunes they have riding on the thousands and thousands of course books, grammar books, textbooks, ESL curricula, etc that fuel their billions of dollars a year industry. The one thing an industry based on educating people is to admit they had gotten lots of it wrong and even worse to admit that they've known it was incorrect for decades yet have chosen to keep pushing the version of English grammar in their books and courses than to issue corrected ones because there was too much money at stake. Plus, because things like do-support and the ESL-specific usage guidelines on when to use which form of a verb don't make any sense, are confusing, and contradictory, they get to spend years longer selling class time to learners trying to get them past their natural justified confusion. That means more schools charging tuition, more companies charging franchise fees, more books to be printed and sold, and more teachers paying to be trained (because not even native speakers can intuitively understand this crap). I suppose what's good for all these guys is that Wikipedia and all of the editors like you guys who would rather maintain the status quo than go out and try to find something proving or linguistically justifying do-support, are being gladly complicit by treating this idea as a proven concept. I'm sure they're appreciative.

None of us use do-support. What we do use, is the appropriate aspectual auxiliary verb along with the specific form of the main verb it requires to put each verbal construction into the correct aspect we need to express the information we desire. Because in English, all verbs have as part of the intrinsic meanings qualities of duration and completeness and such (aktionsart) which establishes the primary force of these qualities, aspect and perfection are used only to emphasize of override those qualities. When a speaker needs to express information about a verb regarding duration, and that verb does not already naturally have a durational quality to it, he uses the continuous/progressive aspect to mark that durational quality via use of a grammatical form (the aspectual auxiliary BE with the main verb in present participle form). If however that main verb already has duration as part of its meaning, he needs not use an additional duration-marking form and instead uses the verb in the neutral 'simple' aspect which simply leaves the verb's inherent qualities as they are. The grammatical form used in this instance is the aspectual auxiliary DO with the main verb in base form. The using a verb in the simple aspect maintains its native qualities of duration and such. Using a verb in the progressive aspect adds duration if none exists, and if the inclusive verb already has a durational aktionsart, then is simply allows for emphasis or amplification of this durational quality.

A verb in either of these aspects may be similarly modified with perfection. Just as each very carries within its meaning duration or a lack thereof, they are also inherently naturally completed or not completed. Verbs like die, swallow, finish, etc obviously and naturally end as part of their basic meaning and thus have an easily perceived termination. Other verbs like live, play, eat, etc have no such natural termination. If a speaker has need to point out, discuss, draw attention to, or emphasize the completion (and particularly the temporal point of completion) of one of these naturally-non-completed verbs, he has to use a special grammatical form consisting of the perfecting auxiliary HAVE with the verb following it placed in past participle form. Together this combination of auxiliary plus specific verb form convey completeness and allow the quality to be added in a situation in which the meaning of the inclusive verb did not already express it, or to emphasize it (just as the progressive aspect adds or emphasizes duration) when the inclusive verb is naturally completed through the grammar of the sentence in which the verb appears.

It is this combination of these two tools of aspect and perfection, combined with the primacy of every verb's aktionsart that allow for the expression of the full range of temporal information possible in the English language (and in varying forms, this same process occurs in every other language as well with the shape of those processes mostly dependent on how strongly governing the role of aktionsart is in each particular language). When those who first pioneered these theories like do-support and the various ESL-centric versions of when to use which forms, they did so with no knowledge of what these very simple tools of aspect and perfection actually did, and without even realising that aktionsart even existed. That's why their ideas were wrong a hundred years ago and that's why no matter how many grammar guides or course books keep including them, they are just as wrong today (even worse of course wrong and they know it).

Now, why does DO suddenly appear out of nowhere? It doesn't. And had they recognized what aspect was at the time, they would have never assumed such a thing to be happening. Every verbal construction expresses aspect and this is done as mentioned above via an aspectual auxiliary followed by its subordinate in a specific form. Both are required and when a given auxiliary is followed by a subordinate in a different form, it means it's not actually acting to convey aspect. Consider BE: "BE + present participle" = progressive aspect; "BE + past participle" = passive voice; "BE + to infinitive" = necessatative/obligative mood. Similarly with HAVE: "HAVE + past participle" = perfected; "HAVE + to infinitive" = necessatative/obligative mood. BE and HAVE, in addition to their functions as aspectual, vocal, or perfecting auxiliaries, also represent non-auxiliary verbs representing such things as existence and possession when not followed by a subordinate. DO along with its usage as a verb of performance, may represent only a single auxiliary role, that of aspectual auxiliary when there is no need to focus on duration.

Just as auxiliaries may be classified by their function (modal, aspectual, perfecting, vocal), they may also be organised by their structure (single verbs with one form like may or can; versatile single form verbs like do, have, or be; phrases made up of a verb + an adjective be+willing, be+able, be+going, have+better, etc; an so on. A third classification is by strength. Some auxiliaries are 'stronger' than others and while they may subordinate additional auxiliaries, they may bot themselves be subordinated (shall, will, must, etc). Others can subordinate additional auxiliaries but can also be subordinated themselves (be+going, have, be, etc). Others still can be subordinated yet themselves cannot subordinate another auxiliary and are called 'weak auxiliaries'. The English aspectual auxiliary DO is an even further very weak type of auxiliary that is not only incapable of subordinating other auxiliaries, but in fact is displaced when it is itself subordinated, giving the appearance of it disappearing in the presence of any other auxiliary verbs.

As I've pointed out, auxiliaries alone have little value. They require a combination of auxiliary + subordinate to convey their purpose. Speakers of a language intuitively analyze for both elements as well as the various other restrictions on form and word order and such that go with them. Sometimes, when enough other clues give away such meaning or function, speakers may for matters of ease or expediency leave one or more parts out of the mix. We see this with contractions such as "I have eaten" yielding "I've eaten" or the less clearly obvious "He's..." which requires context or further grammatical components to decipher its represented form ("he's here -> he is here" = is as main verb; "he's leaving -> he is leaving" = is as aspectual auxiliary; "he's there -> he is there (now) / he was there (then)" = requires context; "he's seen -> he is seen / he was seen / he has seen" = present or past tense passive voice, or present tense perfecting auxiliary). Sometimes this creates forms where entire normally required chunks are left out as in "He (has) done it" or George Bush's infamous "(I'm) not gonna do it." Be it through context or otherwise, when speakers get the chance to express a more complex idea via a seemingly less complex form, they generally do; that's a basic tenet of language change. Sometimes however this results in mistaken analyses where speakers hear one thing and then conflate it with something that sounds or looks similar but has a different meaning or function. This is the root of your "Have you any bananas?" construction. "I do have bananas" becomes "I (do) have bananas" and further "I have bananas". Speakers intuitively know that when forming questions the subject and right-most auxiliary verb trade positions ("He (does) have bananas" becomes "Does he have bananas?") -- similar movement takes place with negation. Having more expediently taken the Do out of "I have bananas" and being quite familiar with moving HAVE around in the many common constructions in which it acts as perfecting auxiliary ("You have seen them" -> "Have you seen them?" and "You have not seen them" -> "You haven't seen them" or -> "You've not seen them"), changes in form for constructions in which HAVE were used as an auxiliary were mistakenly applied to constructions in which HAVE were used as the main verb ("You (do) have bananas." -> "You have bananas." -> "Have you bananas?" -> "I haven't..."). Things like this happen all the time, and it's not necessarily a bad thing although such changes rarely become universal or standard or stand the tests of time, even though speakers generally have no active awareness of what's happening with the language or why they chose one form over another.

DO represents one of the few situations in English in which this sort of linguistic shorthand not only took hold but quickly became a universal attribute of the language, so much so in fact that barely anyone retained awareness of there being something going on. As English moved away from its strictly declined roots and toward one that relied more on word order, the job of subject-verb agreement was given to the right-most auxiliary verb (that auxiliary closest to the subject itself). This auxiliary was tasked with handling agreement for person and number and eventually through its praeterite form, past tense or subjunctive mood. Depending on the auxiliary, some or all of these functions may be confined to a few or even single form. In regard to the aspectual auxiliary BE, such changes appear as (am, is, are, was, and were). For the perfecting auxiliary HAVE you get (have, has, and had). Other auxiliaries may yield things like (can/could, will/would, may/might) while others may show no change at all as with MUST. Because DO is that weakest of auxiliary and is so easily taken out of play without losing the neutral aspect quality of its auxiliary use, speakers easily and naturally developed a system through which they could leave out the auxiliary itself while 'inflecting' its forms into the main verb. Because they have to maintain the person/number agreement and tense expression handled by DO, the solution was to put the main verb into the equivalent old-style declined form which would have matched the form of DO to be omitted. This zero ending DO+verb became verb, while 3rd sing ending DOes + verb became verbS, and past marking DID+verb became verbED (and likewise for irregular verbs). Now of course the world of English speakers didn't sit down together and actively decide to do this, but this system did develop (as such things naturally do in languages) and the result was an abbreviated system so intuitive and easy that natives never even notice it's happening (while non-natives beat their heads against the wall trying to figure out what's going on).

This more efficient linguistic shorthand version of DO+verb however was limited. It's expediency only worked in situations where an auxiliary verb wasn't needed as part of the syntactic requirements of a grammatical construction. Because questions are formed by swapping the positions of the subject and auxiliary, and the negative marker not has to be placed immediately after the auxiliary, leaving the aspectual auxiliary DO out in these types of sentences is simply not an option. Thus the otherwise expedient omission or DO while appending its ending onto the main verb, was limited to only those constructions that did not require such movement. Because DO being such a weak auxiliary, is displaced by all other auxiliaries, including modal auxiliaries which are the most common manner of expressing future tense, this meant that this specialised shortcut was even further limited to only affirmative statements in the past tense or unmarked for tense (usually present tense). With the emphatic mood in English expressed by adding stress to whatever the target unit within a construction is, this meant that even in some otherwise acceptable past and unmarked affirmative statements, that DO had to be retained in original form ("I DID wash the car!" versus "I WASHED the car!" or "I washed THE CAR!" in contrast to nonemphatic "I did wash the car." which could just as well be "I washed the car.").

Without the required knowledge or understanding to comprehend what was going on with DO, and realising that its appearance was not consistent with similar auxiliaries which were fully regular, instead of trying to explain the disappearance of do (which being limited to only a certain set of constructions and thus being among the most rare situations in the entire language), they instead failed to recognise this and attempted to explain why DO suddenly seemed to APPEAR in all other situations. The resulting theory was do-support and further the erroneous classification of DO as having to roles, "dummy verb" and "emphatic auxiliary". Even back then, they knew that these ideas required ignoring everything else that was going on in the verbal syntax of the language and even that it required the development of a specialised set of rules of syntax that only applied to English (sometimes) and not universally which even at this early period ran counter to accepted practice. As is the tradition within the ESL and language education publishing industries, one person put their ideas out there, then the next copied them, the next copied them, and so on. And today here we are...

There has never once been a source that has provided proof that these ideas work, nor linguistic research justifying them. In fact, every source listed above and in the article could be said to fail WP's policy on references because, since they are basically secondary sources which no matter how far back you go, never lead to a primary source that proves the concept, are really little more than primary sources themselves and as has been pointed out to me several times in regard to linguistics debates on here, primary sources do not suffice. This idea has been around for about at least probably a hundred years now. If it's actually correct and viable and a part of English grammar, why has nobody managed in that time to actually provide any proof or linguistic justification for it? Everything I've written above is well-explained, clear, and once you know what you're dealing with, quite obvious. Everything regarding do-support on the other hand is based on putting your faith in an idea that cannot be proven and is based on false understanding while choosing to ignore all the easily discernible research out there that disproves it (is this the linguistic version of organised religion?). So what's it going to be?Drew.ward (talk) 19:50, 24 June 2012 (UTC)

It is as I suspected. You're not arguing against do-support, you're arguing against a particular analysis of it. Angr (talk) 19:56, 24 June 2012 (UTC)

No, read it again, I am explicitly arguing against do-support. It does not exist! I don't know how more clearly I can word that! Are you mixing up my discussion of DO with do-support?

"Do-support" is an unfounded theory that attempts to describe the use of DO as an aspectual auxiliary without understanding or acknowledging its role as such.

Does DO exist in the English language and is it a integral component of many verbal constructions? Absolutely! Does "do-support" successfully describe and account for its usage? Not in the slightest bit. I have presented the REAL explanation of the role of auxiliary DO in English above. Read it thoroughly. It's not congruent with "do-support". And unlike do-support, it is based on well-founded, well-proven, universal linguistic principles and unlike "do-support" it works ALL the time, accounts for EVERYTHING relating to DO, and does not have any exceptions! None of that can be said of "do-support".Drew.ward (talk) 21:01, 24 June 2012 (UTC)

No, you're arguing against what you call "do-support", but what you call "do-support" is not the same thing as what this article is about, or indeed what anyone else means by "do-support". Do-support is not a theory, unfounded or otherwise. It doesn't attempt to explain anything. It's the name for an observed fact of English grammar, namely that do appears in certain constructions and not in other ones. That fact can be explained in different theoretical ways, including both the theory you support and the theory you're opposed to, but this article isn't about any theory one way or the other. Angr (talk) 21:11, 24 June 2012 (UTC)

• "In questions and negatives the auxiliary do has no meaning in itself,[1] so it is sometimes called a dummy auxiliary."
• "In English grammar do-support or do-insertion refers to the use of the auxiliary verb do in negative or interrogative clauses that do not contain other auxiliaries." (these two from this article)
• Do also plays the role of modal auxiliary verb for sentences referring to completed (simple) actions. The basic job of a modal is to make actions into explanatory phrases. Unlike the other modal auxiliaries (have or be), "Do" is not used in affirmative statements because unlike in perfect or continuous, in the case of simple, the action actually does happen.
• Do is used in questions and negations however because in those cases, there is no real action, but rather only the "idea" of the action.
• The auxiliaries do, does, and did are also used for emphasis in positive declarative statements in which the verb otherwise contains only one word: "I do like this shirt!", "He does like this shirt", "I did like that shirt".
• Auxiliaries can be repeated at the end of a sentence, with negation added or removed, to form a tag question. In the event that the sentence did not use an auxiliary verb, a dummy auxiliary (a form of do) is used instead. (all those are from the auxiliary verb article)
• Finally, the dummy auxiliary verb do can be used when there is no other auxiliary verb, except if the main verb is be. (from the pro-verb article)

This particular article has recently been edited down considerably but still clearly expresses a theory and then in particular one that takes on a specific view of the English verbal system that excludes the actual function of DO as an auxiliary and instead purports a different view and creates additional erroneous concepts like "dummy verb" with all of then then being transferred into other wikipedia articles that further corrupt them as well.

There is nothing wrong with there being an article on do-support but unless it is properly worded to establish that it is only a theory and not something that has ever been proven and that there are totally different (proven) views out there, then the article and every part of an article that attempts to establish it as a fact of English grammar needs to be deleted.Drew.ward (talk) 21:35, 24 June 2012 (UTC)

I have seen the remarks about "have" without an auxiliary. I and people like me speak the very best English, and we would say "Have you any bananas?" quite normally. My mother used to laugh at my brother and me because we often said "don't have", as in "I don't have any money" instead of "I haven't any money", which she said we had picked up as a provincialism as boys in America long ago. Seadowns (talk) 10:45, 30 March 2021 (UTC)

Origin of do-support?

Hello,

Is there any interest in adding a section on the origin of this phenomenon (that is, how and why English developed this unique characteristic)? I'd think supplementing the informative but straightforward current desciption of do-support with a bit about the historical linguistic circumstances that led to its development would improve the article. Any thoughts? NORTHUMBRIAN SPRǢC 04:48, 21 October 2011 (UTC)

A few sources we might use for an origin section. I'll look them over and perhaps work on an origin addition this weekend, if no one objects. Cheers, NORTHUMBRIAN SPRǢC 15:32, 21 October 2011 (UTC)

Northumbrian should read Our Magnificent Bastard Tongue by John McWhorter. In Chapter 1, McWhorter demonstrates how periphrastic do is a feature of the Celtic languages Welsh and Cornish. It would behoove a future editor of the Origin section (before or after it is published) to demonstrate the use of do in these languages. In all likelihood the Celts were the first to have the periphrastic construction. I wonder whether anybody has attempted to analyze the salient features of the proto-Celtic spoken in the south of Britain when the Angles arrived. --Simonsa (talk) 12:09, 10 November 2012 (UTC)

The trouble is, periphrastic do doesn't show up in Welsh and Cornish until about the same time it shows up in English, so it isn't clear which language is influencing which. Angr (talk) 12:46, 10 November 2012 (UTC)

Angr is right. The Welsh/Celtic source of auxiliary DO is a false analogy. Our modern verb DO actually is the combination of two separate verbs with auxiliary DO coming straight from the Germanic root that gives modern German tun today. Use of auxiliary DO seems to have arisen primarily as a means of specifying aspectual qualities for foreign loan-verbs, particularly those from Romance languages. This also precedes by about a century and coincides with the early stages of the seeming loss of awareness of such information at the lexical level which seems to have been active knowledge (or at least active intuition) for English speakers up into the late 18th to mid 19th century. Because native verbs would have had most of their aspectual information given as part of the inherent meaning of the verb, they generally didn't get DO at first. With foreign verbs not entering the language 'equipped' with such qualities, explicit aspect-marking forms were needed (DO+verb for simple aspects and BE+verbing for progressive aspects). DO+romance verb seems to be the earliest pattern for auxiliary DO. It wasn't until around the 19th century at the same time that explicitly marking the progressive aspects grammatically became normal, that auxiliary DO (as the counter balance to aux BE in marking aspect) extends fully to all verbs, native (germanic) and romance alike. Remnants of the transition still occur with constructions such as have you a pen versus do you have a pen; most such forms have died out although this one has held on, most likely due to similarity with perfected forms.Drew.ward (talk) 01:44, 11 November 2012 (UTC)

To add to the confusion: do-support also appears in many Dutch dialects, at least in questions and usually optional. Anyway, the article still fails to give the information I was looking for: when did it first show up in English? Steinbach (talk) 17:49, 3 May 2014 (UTC)

Exactly. This information is also needed to evaluate the plausibility of the Celtic hypothesis. @Angr: According to Brittonicisms in English#Change from syntheticism towards analyticism, which references McWhorter, the equivalent of do-support is already frequent in Middle Cornish and used in Middle Breton as well, and on the English side, it first crops up in the southwest, which is eyebrow-raising. Middle Cornish is dated ca. 1200–1600, and at the end of that period, where we are already in the Early Modern English period and Shakespeare's time, do-support was still not regular (if possible) in English. If do-support was already frequent in Middle English, as frequent as in Middle Cornish, I agree that the direction of influence remains unclear. But is it so? And even if it was already present – or even frequent – in Middle English, the apparent regional limitation to the southwest is still suspicious. The possibility that the frequency of a possible (if perhaps "nonstandard", whatever that means in the medieval period) variant construction of native origin was boosted by Celtic influence appears attractive to me. (Unfortunately, McWhorter's paper, the cited source, is not available anymore, and archive.org does not have a copy, either; it does not seem to be freely – i. e., not behind a paywall – available online anymore at all.) --Florian Blaschke (talk) 19:36, 19 August 2015 (UTC)

See this abstract of McWhorter's paper for a counterpoint to Angr's assertion. Even if contact between Breton and Brythonic in Britain may not have broken off immediately after the emigration to Gaul, the presence of the construction in Breton is still a remarkable point. --Florian Blaschke (talk) 20:08, 19 August 2015 (UTC)

As a speaker of both Dutch & Breton, I must strongly disagree with 2 points here: do-support does NOT exist in Dutch, neither in any Dutch dialect. Some claimed that it exists in baby-talk but even if that were true, it's not part of the official language/dialects. Breton however DOES have a lot of do-support and was clearly not influenced by English. Lenn a ran ar journal (To read I do the newspaper = I read the newspaper). Even the verb to do itself (ober) can be conjugated with itself, also in statements (which is also not possible in English): ober a ran (to do I do = I do). Similar to English are the following:

Negative: komz saozneg ne ran ket (speak English I do not = I don't speak English).

Final: Gra? (comparable to English "does he/she/it?")

Questions and answers: "komz a ra saozneg? Ne ra ket" (speak he does English? He does not = does he speak E? He does not).

In other words: very normal and richer than in English, without being influenced by it. In Bretagne (Brittany) it is the other way around. It influences the local French: lire je ne fais pas = to read I do not ("proper French": je ne lis pas). Since it has more options than in English, and it exists on both sides of the "Mor Breiz" (The Channel), I suggest that it does belong to (P) Celtic. If it ever existed in Dutch, it could even have been Celtic influence as well. By now it is well known that the Roman-line between Germanic & Celtic was basically a fantasy... — Preceding unsigned comment added by 86.85.65.6 (talk) 21:48, 27 January 2019 (UTC)

A bit late to point this out, but do-support exists in at least one Dutch dialect: the dialect of the city of Nijmegen. I live there. The dialect isn't widely used any more. PiusImpavidus (talk) 07:50, 22 June 2021 (UTC)

A Celtic origin of do-support seems a crock to me; "to do" ("doon") is common in Low Saxon (Low German, Plattdüütsch, Nedersaksisch) in Germany and the Netherlands to produce the present continuous. An example from Low Saxon from Germany: "Ik do mi düftig fraien, dat wi düt Jaor wier in't Teaoter gaot!" (I do me greatly rejoice that we this year again into the theatre go / I'm really looking forward to us going back into the theatre this year); from Tweants Low Saxon: "Ik do nich geern stofsugen" (I do not happily hoover / I dislike hoovering). Ni'jluuseger (talk) 07:53, 27 June 2022 (UTC)

Draft started

I've started a draft of the Origin section here using the references above. It's very rough at the momentl I hope to continue polishing it over the next few days. Anybody else interested is free to work on it too. NORTHUMBRIAN SPRǢC 22:43, 25 October 2011 (UTC)

Who should we trust?

Two users (or at least, users editing from two different IP addresses) have removed or changed the example sentence Who did Jean flirt with? Although neither left an edit summary, I presume that each believes that only whom (not who) may serve as the object of a preposition in English. Although a few usage commentators insist on a "whom for objects" rule (e.g. Grammar Girl, various style guides), most (e.g. Strunk & White's The Elements of Style, Fowler's A Dictionary of Modern English Usage, Merriam Webster's Dictionary of English Usage, et alia) do not. MWDEU notes that objective who has been in use since Shakespeare.

The Wikipedia:Manual of Style does offer some guidance on grammar, but who versus whom is not part of that guidance. Anyone who believes strongly that this usage is unacceptable should probably argue the case at Wikipedia talk:Manual of Style, and be prepared to cite reliable sources for arguments in either direction. Such editors may also want to review this archived discussion.

Come to think of it, though, since this is an example sentence, and since English speakers actually do say the sentence in either of the two ways indicated on this page, such MOS arguments wouldn't really come into it. Cnilep (talk) 07:18, 16 June 2012 (UTC)

I'm not sure the intent of the previous unsigned edits but regardless of whether it's common or whether linguists argue for its acceptance or not, whom and not who as object is standard English and is the only grammatically correct option. Obviously quite a few speakers use who for all forms and for them whom (while known to be the prescriptive form and certainly understood) is not common usage. There are other articles in which the who vs whom dispute is constantly discussed, but seeing as this article is on the theory of do-support, this is not the place to make changes or have discussions over which to push as the norm. Whom is the grammatically correct form, so even if for many editors who is more common, the correct form should be the one used because doing otherwise risks detracting from the value of the given example in regard to the topic at hand (this goofy erroneous idea of do-support).Drew.ward (talk) 00:23, 17 June 2012 (UTC)

So to summarize, on one side of the debate we have Grammar Girl and Drew.ward. On the other side we have Strunk and White, Fowler, MWDEU, Shakespeare, Ursula K. Le Guin, The New Yorker, Willa Cather, and the British Parliament (see below), among others.

I note Drew.ward's suggestions above that experts, not the usage of English speakers and writers, determine what is Standard English, but I strongly disagree.

• "I want vengence. Who did he talk to here? I want them." Le Guin, Ursula K. (2001), Tales from Earthsea, Houghton Mifflin Harcourt, p. 89, ISBN 054754555X
• "'Who did you speak with at Swiss customs?' he asked. 'I don't recall the name.' 'Who did you speak to at German Customs?'" The New Yorker 68, 1992
• "Who did he trade with? Charley Fuller, in town." Cather, Willa (2008), Oh Pioneers!, p. 28, ISBN 1595476873
• JOHN CONSTABLE, sworn; Examined by Mr. Serjeant Pigott "Who did you meet with there? — There were different parties; Mr. Vickers and Mr. Sturdy." Reports from Committees, Great Britain. Parliament. House of Commons, 1860
• "Who did he talk to when he got there? Mr. Bond." William J. Tolar, Duncan G. McRae, Edward Graham Haywood, Charles Flowers, United States Military Commission (1867), Proceedings in the case of the United States against Duncan G. McRae, R. Avery{{citation}}: CS1 maint: multiple names: authors list (link)

Cnilep (talk) 01:40, 17 June 2012 (UTC)

"I note Drew.ward's suggestions above that experts, not the usage of English speakers and writers, determine what is Standard English"

ummm...where exactly did I say that??Drew.ward (talk) 16:28, 19 June 2012 (UTC)

You didn't say that. You wrote, "regardless of whether it's common or whether linguists argue for its acceptance or not, whom and not who as object is standard English and is the only grammatically correct option", which is simply untrue. Who as object is also standard English and is also grammatically correct. Angr (talk) 19:35, 20 June 2012 (UTC)

This is the wrong place to have this argument. I have replaced the original example with another one that illustrates the same point just as well, without unnecessarily provoking the peevers. Shout out to all my whomies. CapnPrep (talk) 18:33, 19 June 2012 (UTC)

Banners

@Cap, according to your very own statements on various ling debates, the sources you've added do not meet WP's standards for secondary sources. They support the existence of do-support purely by saying that's what it is and without referencing primary sources which prove its existence or validity. Because these sources can't manage to prove do-support (which would actually make them primary sources) and don't reference anything previously which does prove it, they're nothing more than opinion pieces.

The very fact that they are claiming do-support is a valid component of English grammar is pushing a POV and until this article is worded to reflect the fact that do-support is merely a theory with no base in research (or until one of you guys supporting this view can actually find some proof of it) and gives equal footing to alternative views, then it does without a doubt give undue weight toward one view while denying equal attention to others.

In the few times in the past that I have similarly removed a banner for the same sorts of reasons, you and others have quickly pointed out that banners should never be removed without proper discussion and never until the issues raised by those banners are resolved. While you guys disagree with my opposition to this supposedly-factual presentation of do-support, none of you have bothered to find a source validating it. If you feel the banners are unmerited, I would imagine you would be doing so based on confidence in the topic and thus should certainly be willing and able to find a secondary and primary (the important thing being that they have to eventually lead to a primary source that does in fact prove the concept) source that proves do-support.Drew.ward (talk) 20:47, 26 June 2012 (UTC)

May I ask, what in your opinion would count was "a source validating" the presentation here. You do not accept Analyzing English Grammar, Analyzing Syntax and Semantics, or English Grammar: Principles and Facts as valid support for statements about English grammar. You do not accept A Student's Introduction To English Grammar, Linguistics for Students of Literature, or Grammar for Teachers as support for the notion that this page presents the majority point of view within applied linguistics and English teaching. I honestly cannot imagine anything that would ever validate this page to your satisfaction. And if that is the case, I don't see that the maintenance tags perform any valid function. Cnilep (talk) 03:38, 27 June 2012 (UTC)

The answer's simple: an analysis. All of those works have "analyzing" in the title yet present no analysis backing up their proposition of do-support nor is any attempt made to justify their claims of such with linguistic evidence. Find a source that actually does that and successfully justifies do-support with grounded research and I'll gladly concede. Drew.ward (talk) 03:44, 27 June 2012 (UTC)

I don't know what statements of mine you are referring to. I can only suggest, as others have, that you have a closer look at the policy pages WP:SOURCE and WP:NPOV. Banners indeed should not be removed without discussion, but in this case there has been discussion about the issues you have with the article. You have had ample opportunity to make your arguments (many of them multiple times), but unfortunately they do not seem to have convinced anyone. Judging by the reactions of the active editors, both on this talk page and on the article page, consensus has been reached, and it is not in your favor. It may be time for you to accept that, and to consider more constructive ways to contribute to this article. CapnPrep (talk) 06:27, 27 June 2012 (UTC)

No, there is no consensus because I am the only one who has presented an argument. All anyone else has done is say things along the lines of "this is an established part of English grammar" yet none have taken the slightest effort to prove their case.My contention with this article is exactly in line with WP:SOURCE and WP:NPOV. The content of this article and that of the sources cited are not neutral point of view nor do they make any attempt at expressing one. And the sources, according to WP's own policy are all primary by nature because they do not reference proven research, rather just expressing opinions of the authors. According to WP:SOURCES, such sources are not acceptable support for an article. All you guys who disagree with me and support the validity of do-support have to do is find a single source of verifiable research to validate it, and then (in line with WP:SOURCE) secondary sources referencing that.

Have any of you even tried to find that yet?Drew.ward (talk) 18:02, 27 June 2012 (UTC)

The problem is, the argument you have presented appears to be your own. At least, you've never given any indication that any of it has been published in peer-reviewed journals or books. Neither the article nor this talk page is the place for original research. Saying "these sources aren't reliable because I can refute them" isn't the right way to approach the writing of an encyclopedia article. We're here to report what the published literature has to say about do-support. If you think you have a better analysis, you should be writing a doctoral dissertation, not an encyclopedia article. Angr (talk) 19:43, 27 June 2012 (UTC)

How, by your own definitions, is 'do-support' not original research? None of you seem able to find anything published that proves it and none of these cited sources reference such publications either. By this argument, do-support has no place on wikipedia either. Drew.ward (talk) 20:45, 27 June 2012 (UTC)

Drew.ward asks, "How, by your own definitions, is 'do-support' not original research?"

1. Wikipedia defines original research as follows: "Wikipedia does not publish original thought: all material in Wikipedia must be attributable to a reliable, published source. Articles may not contain any new analysis or synthesis of published material that serves to advance a position not clearly advanced by the sources." (WP:No original research)
2. Each major claim in each section of this page is attributed to one or more published sources.
3. The page accurately reflects what those sources say, and does not synthesize or reanalyze them. Most of the points were originally sourced to Kaplan (1989); subsequently other scholars making substantially the same claims were added to show that Kaplan's is not a fringe point of view.

By Wikipedia's definition, Do-support the page features no original research. Q.E.D. Cnilep (talk) 02:11, 28 June 2012 (UTC)

• By WP definitions:
  • Within WP:SOURCE, prohibition on original research: "Base articles largely on reliable secondary sources." -- everything listed in this article and in the quoted references above are primary sources because they don't draw upon primary sources on the topic.
  • from WP:NOR, primary sources: "Do not base an entire article on primary sources, and be cautious about basing large passages on them." -- with no one able or willing to find a primary source that validates this idea, and none of the cited sources further referencing such a primary source (making all of these sources themselves primary), this article fails WP policy.
  • also from WP:NOW, secondary sources: "(Secondary sources) rely on primary sources for their material, making analytic or evaluative claims about them." -- again, the cited references herein make claims with no analysis of or reference to a primary source.
  • This article also fails WP:Fringe Theories in quite a few ways as well.

There's nothing wrong with having an article on this if you really want to have it. But, as this is an unprovable theory pushed by certain people within a given industry and not universally accepted, the article should be written in a way that expresses that. Or, alternately, you should be able to find something out there to validate it but since none of your sources do that, I'm guessing you can't do it either.Drew.ward (talk) 05:15, 28 June 2012 (UTC)

Hello Drew, Angr and others, I have just read this entire exchange. Interesting stuff. I have a couple of thoughts and insights that may be helpful to the article and finding a solution.

My understanding of the term do-support is that it is purely descriptive. It has no theoretical bent to it. I therefore support Angr and the arguments he presents. I think Drew is confusing choice of terminology for a given phenomenon with the actual theory that helps explain the phenomenon. If we were to reject all the linguistics terminology that was originally chosen ineptly, we'd have to reject a significant percentage of the terminology that we as linguists use all the time. For instance, the term infinitive strikes me as poorly chosen. What is infinite about a non-inflected verb?

However, one side of me empathizes with what Drew is trying to accomplish. He is attempting to rectify what he deems to be a poorly chosen term, and he sees himself struggling against the machine (the ESL industry). I too have found myself in a similar situation. My particular issue is the putative binary division of the clause (into a subject NP and a predicate VP) that all phrase structure grammars (including chomskyan generative grammar) take for granted without ever demonstrating empirically that this division really exists. Too many syntacticians' research careers rely on the validity of the division. They've written hundreds if not thousands of books that are reliant on the existence of the division, for if it is not backed up by empircial fact, the basis for their entire understanding of syntax collapses at once. The fact that uncountably many research careers are reliant on a theoretical assumption does not in and of itself make that assumption correct. It does, however, explain why so many linguists are reluctant to really examine the issue - the consequences of changing their stance would be too detrimental to their research agendas. I think Drew sees himself struggling against a similar ingrained unwillingness to question fundamental assumptions.

As the article now stands, it is like many articles on syntax, grammar, and linguistics in Wikipedia: it is not well written. The phenomenon is not well presented. The examples are bunched into the paragraphs instead of set off. The number of examples should be increased. Examples of negative inversion that introduce do should be included. Unacceptable sentences that lack do-support should be produced to help illustrate the importance of the auxiliary. I may revise the article myself in a couple of days. Be aware, though, that I will revise it in a manner that is consistent with Angr's comments. It should not be difficult, however, to include a minor section or note that mentions Drew's view that the term do-support is poorly chosen.

What do you guys think? --Tjo3ya (talk) 04:40, 1 July 2012 (UTC)

I don't think that Drew disagrees about when "do" is used in English or what sentences containing it mean, but about the process of production of sentences or something similar. Perhaps some of the confusion is with expressions like "'do' is introduced" which seem to describe a dynamic process, although all it is really doing is explaining that "do" is present in one sentence and not in another sentence. I do not honestly see what his problem is with the article but maybe a little rewording would help, if this hasn't already been done. Count Truthstein (talk) 13:03, 16 July 2012 (UTC)

I guess you could that's close to my point. My biggest issue with this is that DO is NOT added; it's taken away. In fact, it's not really even taken away as its role still has to be accounted for by appending the markings for tense and person/number agreement onto/into the main verb if it is elided. And even then, such elision is only possible in a very narrow range of circumstances. "Do-support" attempts to provide a mechanism by which the ill-perceived addition of DO is justified but it attempts this under the guise of a fallacious understanding of English syntax in which those subscribing to such views fail to recognise the role of auxiliaries in establishing aspect in conjunction with specific subordinate forms.Drew.ward (talk) 15:06, 16 July 2012 (UTC)

Count and Drew,

I can perhaps agree with the original point that Drew was making when it is stated in this way. That is, "support" implies some sort of addition, whereas an explanation from a diachronic perspective sees the auxiliary being removed or fazed out of standard, non-negated declarative sentences over time. While that explanation may well be defensible from a diachronic point of view, it changes nothing about the fact that the term do-support is very widely employed in grammar and syntax textbooks to denote the core phenomenon. I think a compromise should maintain the basic term do-support, but a brief section that challenges the choice of terminology seems warranted. This section will need to cite a source or two where the issue is laid out. --Tjo3ya (talk) 19:56, 16 July 2012 (UTC)

Poor presentation

Hello Cnilep,

I find the content of the article appropriate. I especially support statements such as

(though terms such as "movement" and "insertion" should be understood merely as parts of a model in certain linguistic theories).

But the organization and presentation can, in my view, be improved greatly. Above all, what is needed are unacceptable examples. The unacceptable examples help the illustration of the core phenomenon, e.g.

a. They understand the point.

b. *They understand not the point. - Without do, the sentence cannot be negated via simple insertion of not.

c. They do not understand the point. - With do, the sentence is properly negated.

a. The puppy chewed it up.

b. *Chewed the puppy it up? - Without do (did), the sentence does not allow the subject-auxiliary inversion of a yes/no-question.

c. Did the puppy chew it up? - With do (did), the sentence is easily rendered as a yes/no-question.

When the examples are presented in this manner, I think the article becomes accessible to a larger audience. Non-linguists can begin to build an understanding of the phenomenon by contemplating the examples that are unacceptable.

I would like to revise this article myself. I will maintain the content and sources that are currently present, but I will improve the organization and provide examples like the ones just illustrated here. Is there any opposition out there to me doing this? --Tjo3ya (talk) 14:54, 18 July 2012 (UTC)

Revision up

I have revised the article in line with my suggestions. Ungrammatical examples now help to illustrate the importance of do-support. I have kept most of the citations, although I am not sure that they are placed entirely correctly. Furthermore, I find the the citation conventions employed (which I did not alter but tried to maintain) to be awkward. The citations should list the page numbers directly where the point is discussed. The convention employed in this article does not do this. I cannot, however, correct the convention since I did not provide the citations originally. Drew, if you are out there, please note that the last sentence mentions the role of do as a potential marker of aspect. I think this is where you might insert a sentence or two explaining the point, and of course, references where the issue is discussed at length should be cited. --Tjo3ya (talk) 23:15, 18 July 2012 (UTC)

I have left the body of the article as per your revision but reworded the introduction. I have left the bulk of it in line with what was there but added a few lines to specify (important) that this is only a theory and that it attempts to provide an explanation for the use of do as an auxiliary. I have also added a line pointing out the two primary critical flaws in the theory (lack of acknowledgement of understanding of expression of aspect and emphatic mood). If these changes are acceptable, I would agree to removing the unbalanced tags. However, I believe the citation and source tags should remain until someone can actually provide a source that leads back to original validated research that actually proves this concept which I think is fair as that is in line with what would be expected of anything else on wikipedia.Drew.ward (talk) 14:46, 20 July 2012 (UTC)

Inappropriate introduction

Drew,

The introduction you have provided is inappropriate. It is pushing one particular analysis of the phenomenon that is not (at least not yet), supported by sourced literature. Your point may be legitimate, but it needs to be sourced. More importantly, your point should not dominate the introduction. Most readers of the article do not care about the theoretical account of the phenomenon. They care, rather, about understanding what the term do-support is intended to describe. I suggest you express your point lower down in the article, approximately at the point I indicated in the message just above. And again, please provide a source a two that backs up your analysis! I am now going to revert your edit. --Tjo3ya (talk) 15:44, 20 July 2012 (UTC)

Tjo3ya, How is my introduction inappropriate? You say that the analysis is not supported by sourced literature yet you and everyone else have yet to be able to provide properly sourced literature supporting your version either! If you are willing to hold my introduction to this standard than you must within the even weakest academic integrity hold yourself to that same standard. If you cannot find sources that prove the validity of your version of do-support than you MUST point out IN THE INTRODUCTION that it is little more than an unaccepted and unprovable theory! as you said: PLEASE PROVIDE A SOURCE THAT BACKS UP YOUR ANALYSIS! Until then I am going to revert your revert. What's fair is fair.Drew.ward (talk) 17:24, 20 July 2012 (UTC)

As stated above, I have reverted this back to my last version. Tell me, what problem do you have with the following? You are demanding sources as if something is claimed that is not true yet of these statements, how do they not mesh with the intent of this article:

• "The term do-support or do-insertion refers to a theoretical explanation of the use of the verb do as an auxiliary verb in negative or interrogative verbal constructions in English that do not contain another auxiliary verb. "

Do you dispute that "do-support" is a theoretical explanation as described?

• The theory purports that the addition of do "supports" the formulation of questions or the inclusion of the negation marker not and often further that auxiliary do is required for expressing emphatic mood.

Do you dispute this statement either? If so you need to rewrite the body to remove such claims below as well as the other auxiliary verb and grammar articles you've been working on.

• Criticism of "do-support" as a valid explanation includes its lack of accounting for syntactic expression of mood and aspect in that it attempts to explain when and why auxiliary DO appears but does so without acknowledging the role of auxiliaries in expressing grammatical aspect in English syntax, and also ignores the use of vocal or written stress in expressing the emphatic, which results in problematic and cumbersome explanations requiring numerous "exceptions" to the rule in grammar guides featuring the "do-support" theory.

Do you dispute this at all? If so, how?Drew.ward (talk) 17:32, 20 July 2012 (UTC)

The issue isn't whether we dispute these statements, it's that the statements are not supported by sources. The threshold for inclusion in Wikipedia is verifiability, not truth. Angr (talk) 19:00, 20 July 2012 (UTC)

Either nothing in this article is supported by sources or everyone involved is just refusing to find them. I find it rather odd that so many supposed linguists can't manage or aren't willing to actually find proof of this theory.Drew.ward (talk) 19:44, 20 July 2012 (UTC)

There's a bunch of sources already in the article. They're down in the References section. They support the statements they're backlinked to by means of the <ref> tags. Your criticisms of do-support and your claim that do may have historically been an aspect marker are not supported by any sources. Angr (talk) 19:51, 20 July 2012 (UTC)

I have never claimed it has historically been an aspect marker. It IS (in combination with the bare form of its subordinate) an aspect marker today just as be (in combination with the present participle form of its subordinate) is an aspect marker. Regarding your "sources", WP's policies state that tertiary sources are most desired and that beyond that secondary sources are required. Tertiary sources must refer back to secondary sources to be tertiary. Secondary sources must refer back to primary sources to be secondary. Neither you, anyone else on here, nor any of the authors of these supposed sources have ever managed to find one single primary source proving or validating "do-support". Thus, they cannot be tertiary or secondary sources. Further, by WP's own standard's this article still, years after I'd first raised these arguments to the same groups of people, has not a single valid source! For this reason, it MUST be clearly noted that it is merely an unproven (and thus far unprovable) theory and nothing more. You guys are very critical of my opposition to treating this idea as fact without substantiation, but what I find considerably more suspect is the rest of your staunch opposition to classifying it rightfully as a theory and strange loyalty to something that you either cannot or are unwilling to find proof for.Drew.ward (talk) 22:31, 20 July 2012 (UTC)

Two Thoughts

Two thoughts on the article:

• The original article posted by Cnilep in 2010 is superior to the current one. If the inaccuracy with respect to copula sentences were to be remedied, I think that version should be restored.

• Drew.ward's behavior on this article has been in gross disregard to Wikipedia's standards of conduct, and border on vandalism. Is there anyway that this article could be locked until a resolution can be reached to end his repeated reverts? — Preceding unsigned comment added by 64.124.28.131 (talk) 06:40, 22 July 2012 (UTC)

The article posted by Cnilep did not employ ungrammatical examples, and it included one piece of information in the introduction that was mostly irrelevant and a bit misleading because it was elevating a marginal construction with have to too great a status. Such marginal examples belong lower down in the article. That said, I am not opposed to improving the article. I agree that the introduction as it now reads is too brief and does not provide a good overview of the article. In this respect, I suggest expanding the introduction. I may do this myself before too long if no one else does so.

Drew's behavior has been counterproductive here and elsewhere. --Tjo3ya (talk) 15:37, 22 July 2012 (UTC)

So let me get this right, if someone disagrees with you and compels you to hold yourself to the same standards you wish to hold everyone else, it's counterproductive?

Instead of whining why don't you just address my criticisms regarding the sources? Your user page lists a page of articles you've worked on, boasts of publications, and so on. Certainly you should be able to easily prove this concept. Right?Drew.ward (talk) 19:55, 23 July 2012 (UTC)

Nobody owes you anything as Wikipedia is created by volunteers. Even if he could easily prove the concept or provide better sources, he doesn't have to. I think you are misunderstanding where the burden of proof is for articles, which is that unsourced material may be challenged and removed, not that other sources must be provided to prove that the material is false. Count Truthstein (talk) 22:46, 23 July 2012 (UTC)

The article as it is currently worded, and also as it was worded prior to recent changes, claims that "The term do-support or do-insertion refers to the use of the auxiliary verb do in negative or interrogative clauses in English that do not contain another auxiliary verb." This is not only widely accepted but also supported by half a dozen sources that use the term and (in most cases) explain that these are generally accepted terms. Note that this is a claim about use of a term, and not an argument about grammatical theory per se. (I should perhaps note Arnold Zwicky's dictum that labels are not definitions.)

Drew.ward makes a more sophisticated theoretical argument that the term is inaccurate since it may not be the case that anything is inserted, and the function of the word do may not be to support anything. This claim is not supported, but neither is it actually made in the article (as opposed to this talk page). Therefore, no sources can be said to be missing on that account.

The argument made on this talk page, primarily by Drew.ward, that the theoretical controversies must be solved seems inappropriate to me, and certainly beyond the scope of Wikipedia. It is the role of Wikipedia or any encyclopedia to report and summarize the consensus of secondary literature (along with significant controversies), not to advance, challenge, or support scientific theories. Cnilep (talk) 01:35, 24 July 2012 (UTC)

Introduction (September 2012)

Hi Victor,

I support your recent expansion of the introduction. I think it is better than what I added, since it provides a more comprehensive overview. Now, let's see if Drew lets it stand. --Tjo3ya (talk) 08:21, 14 September 2012 (UTC)

Now I think it just became a bit two wordy. The last paragraph in the introduction is overkill. The introduction no longer matches the rest of the article because it is too long compared to the rest of the article. --Tjo3ya (talk) 08:31, 14 September 2012 (UTC)

I'm in the process of expanding the rest of the article as well; when that's done the introduction might not seem excessively long (or it still might, in which case it can be adjusted). Victor Yus (talk) 09:36, 14 September 2012 (UTC)

I trimmed the intro a bit. Victor Yus (talk) 11:33, 14 September 2012 (UTC)

I mostly support your revision of the article. You've included some additional content that is pertinent and accurate (e.g. contractions don't and the lack of do-support when an embedded predicate is being negated, e.g. I tried not to laugh. You've provided some explanations that are going to be helpful for many readers.

Where I think the article is lacking now more than before is in the area of organization and presentation. It does not seem like a thought-out and well-organized whole. The aesthetics of the article are also not pleasing: a lot of separate sections with a lot of examples, all run together somehow. Some organizational statements at the beginning of major subsections could overcome this problem. To get a sense of what I mean, see the main subsection "constituency tests" at this article constituent (linguistics) and see the introductory paragraphs in the main subsections in the article clause. If similar organizational measures can be put in place, I think this is going to be a good article. --Tjo3ya (talk) 20:21, 14 September 2012 (UTC)

I'm not sure what you mean - do you mean there should be some explanatory text between each main section heading and the first subsection heading? I can't think of much of value to put in those places that isn't already clearly implied by the headings themselves, but if you have some ideas, go ahead. Victor Yus (talk) 08:56, 15 September 2012 (UTC)

Other negations than "not"

I understand that there are other negations besides "not", such as "neither" and "never". However, I am not sure if "only" is considered as negation, too (I wonder, because the French equivalent "ne ... que" is a negation, the German equivalent "nur" isn't; I refer to occurances where inversion is not applied). I think this should be mentioned in the article.--94.223.158.149 (talk) 10:15, 3 February 2013 (UTC)

only can be viewed as containing a negation of a sort, since it elicits negative inversion, e.g. Only Tom do we think we saw. I do not see that this fact impacts the use of do-support, however. Standard varieties of French and German do not employ something like do-support. I don't think that data from those languages are relevant in this article. --Tjo3ya (talk) 15:10, 3 February 2013 (UTC)

Do-support with "have"

I am rather surprised that some commentators here regard a sentence like "Have you any bananas?" as non-standard. I always thought that do-support did not originally apply to "have" and was only extended to it later, especially in American English. ("Ba ba black sheep, have you any wool?") An American cartoon from the 1940's has a woman enter a butcher shop and ask, "Have you any Porterhouse steak?" The butcher answers, "Yes, we have." (This would seem stilted to most Americans today.) Kostaki mou (talk) 20:02, 15 April 2014 (UTC)

Searches of Shakepeare's texts on http://www.shakespeareswords.com/ would seem to bear me out on this. (Try it.) Kostaki mou (talk) 03:26, 26 April 2014 (UTC)

That sounds stilted to me, even pretentious. It's not natural North American English. Shakespeare died centuries ago. I suggest checking what people really say, for instance in the media. One hears Do you have.. much much more frequently than Have you... There are a couple of lexicalized exceptions, e.g. Have you no shame!. --Tjo3ya (talk) 08:41, 26 April 2014 (UTC)

This is exactly what I was saying. Nowadays, it does indeed seem stilted (to most Americans, at least). I do believe that it does reflect older usage -- and not from all that long ago, either! (Searching the transcript of the Lizzie Borden trial (1893), for example, I find both constructions used, though "have" without "do" is still more common.) I'm sure the question is amply covered in existing sources, if anyone can supply any. Kostaki mou (talk) 22:43, 28 April 2014 (UTC)

"Emphatic Do" use

How can I use "emphatic do"? The article is not clear.

For instance, is the "do + copula" (i.e. "be") form correct? The article only states it's correct in imperatives, although it do not state it is correct only in imperatives.

Some food for your brains:

[1]

[2]

[your links here, feel free to edit]

Thanks in advance. 2A05:3E00:1:55:602B:830A:A644:4A1F (talk) 17:34, 12 December 2015 (UTC)

Seems to me that in standard English at least, "do + be" is only possible in imperatives. W. P. Uzer (talk) 22:02, 12 December 2015 (UTC)

Dubious

The article makes some rather categorical claims about historical forms being "ungrammatical" today, for example that I know not is ungrammatical. I don't think that's true. It's certainly archaic, but not ungrammatical. --Trovatore (talk) 21:54, 19 August 2016 (UTC)

I have fixed the introductory section to my satisfaction. The rest of the article still needs to be looked over. --Trovatore (talk) 14:52, 22 August 2016 (UTC)

auxiliary do-support: do murder

The Cambridge Grammar of the English Language, page 98, reads

Auxiliary do was used more widely in earlier stages of the language, and in certain genres one comes across archaic uses that go beyond our do-support constructions – e.g. in legal language (on or about the 14 th day of June, 1997, he did murder one James Robinson)

What type do-support does "did murder" represent? --Backinstadiums (talk) 16:35, 6 October 2019 (UTC)

I am not a lawyer, but I have always understood this sort of usage as an example of Do-support#For emphasis. Cnilep (talk) 03:00, 7 October 2019 (UTC)

article should be only about do-support, not other uses of the verb "do"

In the article's introduction, it is stated that "the use of do as an auxiliary should be distinguished from the use of do as a normal lexical verb, as in They do their homework.". However, unfortunately, near the end of the article it starts talking in length (three sections) about "do so" and "do as a main verb". These are exactly what the introduction warns about distinguishing. These sections simply do not belong in this article, and just confuses its main point. 46.121.71.206 (talk) 08:26, 13 June 2021 (UTC)

Browsing the article's history, I just realized that for many years, this article just had a very short "main verb" section. Only last year, someone added and started growing all those sections on the unrelated pro-verb, "do so", ascii art, etc. I think those should be removed, or perhaps moved to another article (which?). 46.121.71.206 (talk)