Talk:Auxiliary verb/Archive 2

Archive 1

Archive 2

Article issues

I for one am getting a bit confused by all the threads we've been creating! I'd like to concentrate here just on the three main issues I see in the article and not on discussing other aspects of the linguistics. Peter coxhead (talk) 21:00, 19 September 2012 (UTC)

Issue 1

One difference in the literature is between those who classify words as auxiliaries (or not) and those who classify uses of words. How do we convey this?

A real problem is that all the sources given in the article which I can check are confused/confusing. Even the one that the current footnote 11 says is clear [it says "Jurafsky and Martin (2000:320), for instance, state clearly that copula be is an auxiliary verb"] isn't, if you look more closely. I think we must have different editions (mine is New Jersey: Prentice Hall), because there's nothing about be on p. 320 in mine. In my edition, on p. 294, they say "English auxiliaries include the copula verb be ...", but it's not clear that they mean "be when used as a copula" rather than that be is called by them a copula verb. On p. 295 of my edition of Jurafsky & Martin they say "In addition to the copula have mentioned above ..." where the earlier mention is of have as a perfect tense marker, which to me is an odd use of "copula", suggesting they are not using the word in a way that would distinguish the use of be in He is a man from the use of be in He is running.

So I don't know the answer to my question, just that the article isn't right as it is. Peter coxhead (talk) 21:00, 19 September 2012 (UTC)

Peter, This is splitting hairs in my view. In my view, you are unwilling to back away from a stance that you adopted at some point in the past that is proving to be untenable. I tell you what I will do to put this issue to rest once and for all: I'll spend some more time in the library. Perhaps I can find one or two more sources that state that the copula be is an auxiliary. If that is done and the sources have been added to Jurafsky and Martin, I hope you will let this issue go. --Tjo3ya (talk) 22:38, 19 September 2012 (UTC)

I'd like to be quite clear about your position: are you saying that there isn't a difference in the literature between those who classify words as auxiliaries (or not) and those who classify uses of words? If there isn't a difference, which of the two is present?

If you mean that my account of Jurafsky & Martin is "hair splitting", you're right, but in general their writing is just unclear so that without close analysis it's not clear what they mean. Peter coxhead (talk) 14:36, 20 September 2012 (UTC)

Issue 2

How can we be clearer about the different definitions in sources and their consequences?

Let me take have as a case-study and consider what should be in the table of auxiliaries. To use a source which for once is clear, on p. 296 Jurafsky & Martin write of have that it has three senses, (1) as a main verb (I have three oranges), (2) as a modal (There has to be some mistake) and (3) as a non-modal auxiliary (I have never seen this) [their examples]. It's clear that for Jurafsky & Martin, auxiliaries include modals – they say so explicitly on p. 294 of my edition. So both have₂ and have₃ are auxiliaries for them, and should be in any table of auxiliaries. Now in my dialect only the first and second uses allow inversion and contracted negation (Have you any wool? as in the nursery rhyme is fine for me as is Why haven't you any wool? but *Has there to be some mistake? is unacceptable compared to Does there have to be some mistake? as is *There hasn't to be a mistake compared to There doesn't have to be a mistake). So by the inversion and contracted negation tests, have₁ and have₃ are auxiliaries (1 only in some dialects) and should be in the table but have₂ isn't an auxiliary and shouldn't. So:

• have₁ is called by Jurafsky & Martin a main verb but is an auxiliary in some dialects by the inversion and contracted negation tests.
• have₂ is called by Jurafsky & Martin a modal and hence an auxiliary but is not an auxiliary in (any?) dialect by the inversion and contracted negation tests.
• have₃ is called by Jurafsky & Martin a non-modal auxiliary and is an auxiliary in (all?) dialects by the inversion and contracted negation tests.

I think that issue 2 can be resolved by writing up the above case study more clearly – I'm happy to do this, if it's agreed. We would also need to sort out the page numbers, because mine may be different from the edition referenced in the article. Peter coxhead (talk) 21:00, 19 September 2012 (UTC)

The way they are analyzing have as a modal auxiliary is the same way I have analyzed such modals above. Because have is the fully versatile periphrastic equivalent to must, it is still a single modal (and thus fills a single modal auxiliary slot). However, because unlike its defective twin must, have is a fully declinable verb, this means that it inevitably contains an 'auxiliary within an auxiliary' situation with auxiliary do ("inflected" onto its subordinate have and seemingly "omitted" in the indicative mood, unperfected, simple aspect) as its internal auxiliary. Just as with the non-auxiliary use of have, the verb handles subject agreement for person & number and can also be used to convey tense -- both of these functions possible through (seemingly) "inflected" versions of have (have, has, had), or via periphrastic (do have, does have, did have) with periphrases as "do-support" necessary in situations requiring the explicit presence of an auxiliary. Defective single-unit modal auxiliary must can convey only mood with no signification of tense, additional mood (aside from mood conveyed without auxiliaries, compare "I must leave" (obligative/necessitative mood) with "I really must leave" (obligative/necessitative mood via must plus emphatic mood via really) although some sources may classify the latter as modality rather than mood.) The equivalent periphrastic modal auxiliary have however can convey the same moods but can also do so with the same flexibility and versatility it allows for conveying "possess" when functioning as a non-auxiliary content verb and that includes not only passing its own modality onto the verb(s) it subordinates, but also accepting any combination of aspect, perfection, tense, or additional modality onto itself and through that internal syntax, as an added layer of meaning for that primary obligative or necessitative modality it conveys as a whole unit.

I am not trying to be argumentative, nor am I doubting the validity of the sources in arguing that only single-unit defective modals can be auxiliaries, but I am trying to point out that those authors have surely missed something in choosing to go that route because by their analysis, far too much ability to convey meaning in English would be lost of impossible. Certainly there is the ability to affect whatever modality for purposes of tense, aspect, perfection, and further modality discussed above, but then there are more easily obvious differences. Observe: (in the examples below, the modal auxiliary is in italics with everything it subordinates (passes that modality onto) in bold)

1. You must drink the water.

2. You must not drink the water.

3. You must not drink the water.

4. *You must drink not the water.

5. *You must not drink not the water.

1a. You have to drink the water.

2a. You have to not drink the water.

3a. You don't have to drink the water.

4a. You have to not drink the water.

5a. You don't have to not drink the water.

Examples 1 & 1a have identical meanings, as do examples 2 & 2a. However, examples 2 & 3 create a potential ambiguity because the form with must provides no means of differentiating negation of its mood from negation of its subordinate; thus 2 & 3 can mean the same thing or two different things. Example 3a on the other hand offers no such room for ambiguity with 2a. Further demonstrating the limits of must which justify its periphrastic equivalent have, 4a & 5a convey two additional applications of this same shared mood that are impossible with must and yield ungrammatical sentences as with examples 4 & 5.

See where I'm coming from in questioning the legitimacy of any definition of auxiliary or modal that precludes these periphrastic forms?Drew.ward (talk) 21:56, 19 September 2012 (UTC)

Drew, Please slow down! Please check your messages for typos and grammar before you hit click. Please spend some time ensuring that what you are writing is coherent.

Peter, I agree that the idiosyncracies you are mentioning definitely exist, although for me, a formulation like "Have you no decency?" is a lexicalized form and as such, it is no longer productive in the language. I also acknowledge that some dialects may be holding on to such formulations, so that main verb have has auxiliary status and is thus like copula be in those dialects. In general, I think an additional subsection that draws attention to these idiosyncracies would be overemphasizing their importance and would hence detract from the article in general. But if you insist, I am OK with it if you go ahead and draft such a section. I think it should appear immediately after the section "Diagnostics for identifying auxiliary verbs". I also think it should say something about the other marginal cases: dare, ought, and need. Its length should not exceed the length of the other sections.

Incidentally, I think a section on how parsers and taggers treat auxiliary verbs would be a more valuable contribution.

I'm spending too much time on this. Please do not be offended if a go silent for a couple of days. --Tjo3ya (talk) 23:07, 19 September 2012 (UTC)

I understand; we're all spending too much time on this. But when you write the "main verb have has auxiliary status and is thus like copula be in those dialects", then we are talking past one another and you're missing the point I want to get across, namely that whether have₁ is an auxiliary is not a matter of fact but a matter of dispute. Jurafsky & Martin clearly say that it's a main verb, which means it can't be an auxiliary for them. Those sources who use inversion/negation tests will say that in a dialect like mine it is an auxiliary and so can't be a main verb. You seem to want not to allow the article to say that the definition of an auxiliary and hence of what is included in this category varies from source to source. Peter coxhead (talk) 14:43, 20 September 2012 (UTC)

I don't think there's any intent on Tjo3ya's part to limit the article to only a specific view of auxiliaries or to disregard views here in the talk page. I think he's just mainly working within the view of auxiliaries that is most common within his primary discipline (dependency grammars which from what he's described herein, actually seem to at least try to handle auxiliaries far better than constituent approaches) and otherwise to ensure that whatever various alternatives we're proposing are similarly sourced like the views he's drawing upon are in the article.

I think that these discussion, and also similarly themed discussions on virtually every Wikipedia article dealing with things related to auxiliaries, do make it clear that "auxiliary" is neither an easily defined simple piece of terminology, nor a universally agreed upon concept even within these various individual frameworks. It seems regardless of whose definition is used and regardless of which framework that definition is built upon, that all of these sources listed and those being discussed define "auxiliary" in far too narrow a way to actually account for all the bells and whistles of auxiliary-related syntax/semantics (and in regard to our discussions, at the very least as relates to English). I don't know what the solution to this is, but it does seem to be at the core of these debates and in fact of all auxiliary-related discussions I've seen.

The section on diagnostics seems to have been to root of most of these current discussions, especially in that there are the various situations (copula be, modal have, auxiliary-absent lexical have, and the various others pointed out herein) where the diagnostics either don't recognize an otherwise valid auxiliary, or (as I believe has been the main point of contention) in which items generally recognized as not being auxiliaries (functionally or structurally) receive "false positives". Might I propose for the purpose of those points within this greater discussion that are centered around these issues with non-auxiliaries "passing" the diagnostics, that instead of using that situation as the focus of a discussion in general, that we instead just add in material to the diagnostics section pointing out these few situations in which they don't work (or work yet result in things not generally called auxiliaries meeting the qualities attributed to auxiliaries). Would that help?Drew.ward (talk) 16:36, 20 September 2012 (UTC)

Issue 3

How do we say that some sources (Huddleston in my interpretation, and certainly the Penn Treebank part of speech tags) don't regard auxiliary as a useful category at all?

Although I added only this issue to the article, when issues 1 and 2 are sorted, this can be relegated to a much less prominent position. Peter coxhead (talk) 21:00, 19 September 2012 (UTC)

Yes, we still disagree concerning Huddleston. Perhaps this issue will also be resolved if I find that Huddleston in his more modern works acknowledges auxiliaries. --Tjo3ya (talk) 23:14, 19 September 2012 (UTC)

Duked out

Victor,

You are very good at finding the middle ground between disputes on talk pages and I support most of your work. This article is currently being duked out here on the talk page, however, and the section you just added seems redundant to other information in the article. I encourage you to comment on the exchanges above, especially the three recent threads immediately above. --Tjo3ya (talk) 16:15, 25 September 2012 (UTC)

Sorry, I just came here and looked at the article, I had no idea there was any dispute going on. Anyway, I hope my contribution will be considered - I don't think it's particularly controversial (though it might technically count as original research). I'll look at the above discussion later. Victor Yus (talk) 16:23, 25 September 2012 (UTC)

I agree with your point that the article as it stands is overly English-centered. But the section you added was indeed making the article somewhat redundant. I would support adding a paragraph to the introduction pointing to the fact that auxiliary verbs are going to be identified and classified in various ways depending on the language at hand, and then I think at some point further down in the article a section (with subsections) can be added to address auxiliaries in other languages. One thing we should not lose sight of, however, is that this is English Wikipedia, and most every reader who comes to this page can immediately understand English data. --Tjo3ya (talk) 16:52, 25 September 2012 (UTC)

Yes, hence my preference for using examples from English if possible, when making general points. But we ought to address the general case before the particular - it's highly confusing to launch into a (confessedly controversial) list of English auxiliaries without first making clear what's happening and why the approach being taken here seems not to correspond exactly to that implied by the article's introduction. I thought that's what I was doing with my addition - can you explain again why you don't like it? (I don't get "was making the article redundant" - do you mean "was redundant to the rest of the article"?) Victor Yus (talk) 08:34, 26 September 2012 (UTC)

You're clearly right that the article's introduction doesn't correspond at present to the content, taken as a whole. That's because we can't quite agree on the content at present. The time to re-write the lead is when we have stable content. So I agree with Tjo3ya that you can best help by assisting with the content. Peter coxhead (talk) 10:32, 26 September 2012 (UTC)

I thought that's what I was doing... Victor Yus (talk) 10:39, 26 September 2012 (UTC)

I don't remember exactly which article was the base for the conversation or whether it resulted in any sort of official WP policy, but it seems that a couple of years ago there was pretty wide consensus that this being the English-language side of Wikipedia, that it was both assumed and justified that the bulk of examples and the centering of discussions about grammar and linguistics should when practical, focus on English. I don't think it was necessarily having to do with this article but the reasoning at the time was something along the lines of it making sense that if people were choosing to read the English-language page (especially when other language options were available), that they either would be more comfortable with or more interested in English-centric descriptions and examples in English. However it was equally agreed that focussing on English should in no way preclude discussions and examples from other languages, nor demote those non-English discussions at all, but that non-English examples do need to be framed in a way that they are comprehensible and useful to English-speakers having no knowledge of those other languages. Other editors may remember this and have more direct links of info...Drew.ward (talk) 14:25, 26 September 2012 (UTC)

That is not the case. It is a wikipedia policy at the most general level that articles should represent a worldwide view of a topic, including all aspects in all relevant languages and cultures. Examples in English are preferred only when there is a good reason to do so for example because it is easiest for readers to understand. English examples should be used to illustrate English usage, but not as the center around which descriptions or analyses should be constructed.·ʍaunus·snunɐw· 14:37, 26 September 2012 (UTC)

So, there seems to be little ongoing discussion on this page... Can we return to my addition (which was reverted previously)? Is there any substantial objection to it at this point? If so, what? (I realize the link to subject-verb inversion should be to subject-auxiliary inversion.) Victor Yus (talk) 12:43, 11 October 2012 (UTC)

Victor, my problem with it is the same as I have with other parts of the article, namely that it doesn't have inline citations showing its sources. For example, who says "Auxiliary verbs do not necessarily constitute an obviously distinct class in a given language"? Who says "Their characteristic features might be expected to be, firstly, that they add little lexical content besides grammatical information, and secondly, that they govern or 'help' another verb or verbal expression"? In particular, who might expect their characteristic features to be these, and why? Peter coxhead (talk) 15:27, 11 October 2012 (UTC)

Victor, citations are important, especially when there's disagreement. It is the only way that I can see that one can resolve disagreement and progress with producing helpful articles. In this regard, I'll repeat my comment from another page. Your work is generally good and helpful, but I am seeing little to no effort on your part to back up your contributions with citations from the literature.

Peter, Tesniere (1959) views the copula as an auxiliary. He says so in numerous places in his book. I have also communicated directly with two high level linguists (I mean really high) who have stated that they prefer to view the copula as an auxiliary verb. I have not yet made it to the library again to search through sources for more direct support that the copula is an auxiliary. --Tjo3ya (talk) 17:34, 11 October 2012 (UTC)

There are sources given in the article, are there not, for both positions? (i.e. copula is or is not an auxiliary). What is this disagreement you all keep talking about? If it's just about whether the copula is an auxiliary, I don't see that it's a question we can or should settle - we just say that it is according to some definitions, and not according to others. Is there any other point of disagreement concerning the substance of this article? The main problem for me is the presentation - it goes straight into the question of auxiliaries in English, although that language is highly atypical in this regvard (in that there is a set of words that get to be called auxiliaries based on their syntactic behaviour). Victor Yus (talk) 07:29, 12 October 2012 (UTC)

Victor, I agree with what you write about the copula. I think the article is fine as it is in this area. It acknowledges both sides of the issue with sources. Drew as well, he stated that he was fine with the article in this area. Peter is the one who does not want to let the issue go. I also disagree with what Peter has added concerning Huddleston's early work on auxiliaries, and I think the comment about the lack of auxiliaries in some tree banks should be put into a separate section about how auxiliaries are treated by computational linguists. That comment does not belong where it is.

Concerning expanding the article to address auxiliary verbs in other languages, I'm in favor of doing that. But again, I think these articles are going to be accessible to the widest audience if they first focus on English data. --Tjo3ya (talk) 08:34, 12 October 2012 (UTC)

As to the last point, this is a somewhat special case in that regard, since English is so atypical as regards what are counted as auxiliaries (given English grammar, it's so convenient to amend the meaning of the term to refer to all verbs that participate in that kind of inversion). We should perhaps start at least with some general definitions of auxiliaries, and then when we move on to English, make clear that we are being very language-specific. Victor Yus (talk) 09:15, 12 October 2012 (UTC)

Victor, This is what I fear is going to happen here if you get going. You're going to add a large amount of text that is completely unsourced. The text may end up being less than fully coherent, in part because you often prefer to include examples in the running text. --Tjo3ya (talk) 09:31, 12 October 2012 (UTC)

There's no need to be unpleasant, or to make an issue of the trivial matter of how examples are presented. It seems that others are perfectly at liberty to add incoherent and/or largely unsourced material to Wikipedia's linguistics articles - I've been spending much of my time lately tidying such material up. I just think that on a subject that is not specific to any one language, and for which the familiar language (English) is not a good illustration, we should not launch straight in to a discussion of the subject with respect to that language before we've said a little about the subject in general terms. Victor Yus (talk) 09:42, 12 October 2012 (UTC)

Victor, how the examples are presented is not a trivial matter. It is a basic question of how one presents information so that it is accessible to a wide audience. I want the articles where I have contributed to be coherent. I want them to be thought out and consistent in themselves. I want them to be accessible to a wide audience. When one adds information at a feverish pace, as you seem to be doing, I think the danger arises that what one ends up with is often simply poor quality.

Concerning what others have added, I agree that there is a lot of poor quality out there, and in this regard, I think your efforts are commendable. I also welcome your scrutiny to the articles I patrol. Concerning this article, why don't you make a concrete suggestion about what you want to do. I can foresee a couple of easy ways to maintain the current content of the article and at the same time, expand it so that it adds information about auxiliaries in other languages. --Tjo3ya (talk) 10:07, 12 October 2012 (UTC)

Since the notion of "auxiliary" in English is disputed (at least on this talk page!) it is of great importance that any material added is supported by inline sources. The fact that other linguistics articles are poor in this respect can't be used to argue that this one should be as well. So to that extent, I agree with Tjo3ya.

On the other hand, I think that Victor Yus may be right that it would be better not to use English as the first or main language in which to explain the term. It's clear that English "auxiliaries" are a disparate group. Even the most minimal "core" group common to most (?all) sources is not homogeneous as it consists of a reasonably coherent subset of modals (marked e.g. by the absence of third person singular agreement), plus uses of be and have to mark aspect or mood. The article might work better, and be easier to understand, if there is sourced material available for another language in which auxiliary is a clearer category and this were introduced first. Whether there is or not, I don't know. Peter coxhead (talk) 10:58, 12 October 2012 (UTC)

Peter, I mostly disagree. Beginning an article on auxiliary verbs with significant data from another language other than English is likely to reduce readership. I think a majority of readers are interested in English data, and English data are what every reader can immediately understand. The English data provide the orientation point for readers to understand the data from other languages. This is a basic component of making articles on linguistics accessible to the widest possible audience. This is English Wikipedia!

I think the introduction should be revised to accommodate a broader approach to auxiliaries. The article should then proceed with the discussion of mainly English data. It should then proceed to other languages. I can easily expand the article with information and examples about auxiliaries in German and French, and I can source my contributions. --Tjo3ya (talk) 18:38, 12 October 2012 (UTC)

We can certainly use English data in the initial explanation, but not the detail about English grammar rules (the stuff about inversion and negation and so on), since that is very specific to English and will give quite the wrong impression about how auxiliaries function (or what auxiliaries are) in general. Victor Yus (talk) 16:40, 13 October 2012 (UTC)

Tjo3ya, I have to say that I side with Victor here. He is is surely quite right that tests of inversion and negation (which are "very specific" not just to English but to modern English) cannot be features which define the category of auxiliary, otherwise auxiliaries would not have existed in earlier English when main verbs could be directly negated or inverted. In the King James bible for example, we find "ye know not..." (e.g. at Mark 13:35), "know ye not ...?" (e.g at 1 Corinthians 6:9), and "do ye not know...?" (e.g. at 1 Corinthians 6:2). Yet there can't be any doubt that know in all three examples is a traditional main verb and that do in the third is a traditional auxiliary. So I think the article needs to establish first the meaning of the traditional category of "auxiliary verb". The peculiar (and inconsistent) behaviour of auxiliaries in modern English should be secondary and much less prominent than in the present version. Peter coxhead (talk) 00:10, 15 October 2012 (UTC)

OK, here's my suggestion for a compromise of a sort. We revise the introduction so that it broader and perhaps vaguer so that it provides a broader sense of what an auxiliary verb might be. We then add an overview section immediately after the introduction that produces some statements and examples about auxiliary verbs in various language. We then launch in to the section on auxiliaries in English. After the English section, we might add sections about auxiliaries in various languages, e.g. German and French and whatever other languages you guys might know. What do you think?

That sounds OK with me, though perhaps the section on auxiliaries in English could also be reduced somewhat, with the detail being moved to the specific article on English auxiliaries. Victor Yus (talk) 10:34, 15 October 2012 (UTC)

I agree. Peter coxhead (talk) 12:38, 15 October 2012 (UTC)

OK, if Victor would like to draft the new section "Overview", I'm OK with that. Here are my suggestions for making the overview coherent and accessible to a wide audience:

1. Keep the number of paragraphs to a minimum. A new paragraph on every line breaks up the text and presents an unpleasant picture. The information should flow, not hop along.

2. The examples should appear separate from the running text as they are currently in the article. This is standard practice in linguistics texts.

3. If anything at all can be cited, that would be good, since inline citations increase the validity of the information.

4. Non-English examples should be glossed. You can see how this is done at the article right node raising, which I just created. Take a look at the German examples in the article. Each non-English word should be matched to its English counterpart. Doing this is time consuming, but it is essential for making the data accessible to a wide audience.

If this is too much work, I will draft the overview myself before too long. --Tjo3ya (talk) 21:13, 15 October 2012 (UTC)

New section

Victor, the new section you added is redundant. It restates what appears in the intro and in the section that follows. What is needed is an overview about auxiliaries in general across many more languages. The most common auxiliaries across languages are the equivalents of be and have. The overview can state that while the modals are auxiliaries in English, they generally do not qualify as auxiliaries in other languages, in part because they appear in full inflectional paradigms in other languages. It can state that the criteria for identifying auxiliaries will vary across languages. It can state that auxiliary verbs typically do not constitute separate predicates, which means they cannot be negated. It can state that there are numerous multi-word combinations that behave similar to auxiliary verbs in most languages but that strictly speaking do not qualify as auxiliary verbs because of their multi-word status. --Tjo3ya (talk) 15:26, 17 October 2012 (UTC)

You seemed to object before when I started talking about other languages - you implied our readers would be flummoxed by anything other than English... Anyway, I've no objection to any of this being added, but as you yourself were insisting not long ago, citations should be given for anything that isn't obvious. Victor Yus (talk) 09:01, 18 October 2012 (UTC)

Victor, what I agreed to was that the article should present a broader picture of auxiliary verbs that gives a more accurate impression about auxiliary verbs across languages. You have provided a section that basically repeats what is stated in the intro and what is already illustrated (more comprehensively) with English examples in the table further down. You also produce an example with a modal verb from English. That generates an inaccurate picture, since it suggests that modal verbs are auxiliary verbs in general. Modal verbs can be taken as auxiliaries mainly only in English. In the other languages I know, they behave more like full verbs since they can be fully inflected. This is a point that needs to be presented early on

Something else you do which I judge to be problematic is the number of paragraphs. Your paragraphs on average are very short, often they consist of just one sentence. What that does is it breaks up the flow of information unnecessarily and it generates a sense that the text lacks organization and oversight.

I also note that you opened a dictionary in order to produce a citation. Do you own any grammar or syntax book? Why not open it to produce a citation? Pick up a grammar book, look in the index for "auxiliary verb", and then produce a citation or two. --Tjo3ya (talk) 17:20, 18 October 2012 (UTC)

Well, we already do have citations from grammar books, so that doesn't seem to be a particular problem at this stage (though no harm done if you or I or anyone else can add a few more). I'm not sure why you think that being fully inflected disqualifies a verb as an auxiliary - it doesn't disqualify be or have or etre or avoir. But I agree that I get the impression that foreign equivalents of things like can and must tend not to be referred to as auxiliaries - we could do with some citations for that. [PS I just added a reference for French pouvoir and so on being classed as semi-auxiliaries.] (The short paras are mainly a result of doing what you said and writing the examples on separate lines - obviously when you do that, the text between them gets split into what looks like lots of short paragraphs.) Victor Yus (talk) 18:31, 18 October 2012 (UTC)

Yes, what you just added toward the end of the overview is good. You have a citation, which strengthens the point considerably. But what is at the start of the overview is still redundant. It is focusing mainly on English data that is addressed more comprehensively further below. The statement about mood, aspect, and voice already appears in the introduction. I suggest cutting out the first part of the overview. --Tjo3ya (talk) 19:40, 18 October 2012 (UTC)

I don't understand - a while ago you were insisting that we should begin by illustrating the topic using English examples (since foreign ones might put people off). Now when I do exactly that, you complain that the examples are redundant. I think a certain amount of redundancy is a good thing at this stage - we can expect that an "overview" will contain information that is expanded on (and in the process, repeated) later on. The point is to present it in such a way that everyone will be able to grasp the essentials of the topic. Victor Yus (talk) 19:45, 18 October 2012 (UTC)

Victor, you emphasized in this article and in the inversion article that the information should be broader, not so English focused. You managed to convince Peter to support your stance in this area. But when I then agree, you proceed to produce a section that hardly broadens the focus, since it involves mainly English data. So which do you want? A broader focus with data from more languages, or your specific interpretation of the English data? I can support the first change, but not the second.

A compromise solution from where we are now might involve changing the title of the section to "Traits of auxiliary verbs". We then list the traits associated with auxiliary verbs in general across languages. We avoid producing specific example sentences, since those already appear further below. What you have on semi-auxiliaries in French is just what I think should appear in the section, since it delivers a sense of variation in auxiliary status across languages. I know that German grammars tend to draw a distinction between auxiliary verbs and modal verbs, whereby modal verbs are not auxiliary verbs.

Why don't we produce a list of common traits associated with auxiliary verbs across languages. Here's is a list of traits that I associate with auxiliary verbs:

1. Help express grammatical meaning of aspect, mood, and voice,

2. Closed class, relatively small in number,

3. Most frequently occurring verbs in a language,

4. Most widely accepted auxiliary verbs across languages are the equivalents of be and have,

5. The modal verbs may or may not be classified as auxiliaries based upon the criteria employed,

6. Typically appear with a main verb, although

7. The verb be when it is used as a copula (and therefore appears without a main verb) can be classified as an auxiliary based upon how auxiliary is defined. --Tjo3ya (talk) 20:14, 18 October 2012 (UTC)

Again, I don't mind you adding any of this (though some of it might require citations), but I think it's useful for the article to start off roughly as it does now, with an overview that states the basic dictionary definition of auxiliary verb and gives a number of easily understood examples, to make the topic accessible to as many readers as possible. More technical and language-specific information can follow after that. (The use of mainly English examples is not intended to focus on English data, but to illustrate the general concept with examples that will be accessible to all readers.) Victor Yus (talk) 11:56, 19 October 2012 (UTC)

Victor, Please slow down. You are now taking too much liberty with your changes. --Tjo3ya (talk) 13:57, 19 October 2012 (UTC)

And what about you? You've made even more extensive changes than I did. You seem to think that you're the only person allowed to edit these articles - and that the requirement for controversial statements to be backed up by sources applies to everyone except you. Victor Yus (talk) 12:42, 20 October 2012 (UTC)

Victor, The section of examples you added is not what I thought we were agreeing to. I thought we (you, me, and Peter) were agreeing to add a section that delivered a sense of auxiliaries across languages. The examples in the section you have are redundant in part because they are mostly English examples. The needed English examples appear more comprehensively in the table. I will now move the table up in the article so that it is closer to the start of the article. Concerning your new found emphasis on citations, I welcome it. Both of us should be using more citations. I will give you a chance to respond before I take action concerning the redundant information you have added. --Tjo3ya (talk) 15:10, 20 October 2012 (UTC)

But examples are needed before we start talking specifically about English (which as we know, is an unusual case). We can have a selection of examples from several languages, if you like (a couple English, a couple French, a couple German...) Just to let readers know, in terms that they can readily understand, what auxiliary verbs are, as the term is generally understood. We can then go into the technical detail and the mess of the conflicting definitions, for those who may be interested. Victor Yus (talk) 17:35, 20 October 2012 (UTC)

Victor, The problem I am having here is that you seem to argue for one thing, but then you do something else. That makes finding a compromise very difficult. I generally agree that a series of examples should appear immediately after the introduction because the examples are crucial for orientation. I will add a couple of examples from German and French to the section. --Tjo3ya (talk) 15:02, 21 October 2012 (UTC)

The reason I do something else is mostly to take account of what other people have said (like you originally said you didn't want foreign stuff at the top). Anyway, it seems we now agree about how the examples should look. The following sections are still a mess, though - the "traits" thing I presume can be tidied up and supported with cites in due course, but the section on English is a logical quagmire. I started trying to sort it out, but you reverted for no expressed reason - can you explain what you want to do with it? As it stands it's hard to follow what it's saying - various would-be definitions and sets of auxiliary verbs are given, with no attempt to explain or relate them to each other. For me it seems a good idea to first talk about the set of verbs that have those special negation and inversion properties (since the existence and composition of this set, whatever people may call it, seems an undeniable fact of English grammar), and then go on to talk in detail about the various definitions of the set of auxiliary verbs (which may be, but are not always, precisely identified with the negation/inversion set). Victor Yus (talk) 07:03, 22 October 2012 (UTC)

Victor, I think the sections on auxiliaries in English can be rendered coherent without too much effort. An introductory paragraph needs to be added that provides an overview of what is covered and some transition sentences will tie it all together. Please let me do this, since I have already invested a lot time here and want to see the article take on a coherent and well-rounded form that I can support. If you don't like what I do, I will welcome your feedback and suggestions for improvements. I am busy with other things at the moment, I will get to this in a couple of days, perhaps over the coming weekend. --Tjo3ya (talk) 19:14, 22 October 2012 (UTC)

OK, will wait for your edits. When we have this sorted out (or perhaps simultaneously) there is similar work to do at the English auxiliaries article. Maybe some of the detailed English-specific information could be placed in that article and just summarized in this one. Victor Yus (talk) 10:31, 23 October 2012 (UTC)

"To be gone"

"English uses "be" only with "go" in some senses." Does that make any sense? We could just say that "gone" is an adjective. It's no use to say that "I'm gone" is a present perfect. —The preceding unsigned comment was added by Army1987 (talk • contribs) 23:22, 31 October 2004 (UTC)

• Weird, I don't know why this is suddenly popping up as the recent edit, but to answer: actually HAVE is only one of two perfecting auxiliaries in English with BE being the other. The use of BE in perfecting constructions has dropped considerably to the point of today being considered an historical relic, but it is still a valid part of the language and while preference for HAVE has greatly overtaken BE, it is still there. As with other Germanic languages, BE is limited to perfecting certain intransitive verbs, usually verbs of motion, and a handful of transitive verbs, usually having to do with changes of state or accomplishments.Drew.ward (talk) 03:03, 13 November 2012 (UTC)

It popped up because Wwoods added some invisible instructions to tell MiszaBot to archive it, but I guess it's too late now… This comment refers to a very very old version of the article (late 2004) that contained the quoted statement, which with no explanation, examples, or sources, was completely useless. It was finally removed in June 2010, thankfully. I don't think there's any reason to discuss it further on this talk page; it's a point of English grammar that could be mentioned in one of the English-specific articles, but not in this one. CapnPrep (talk) 15:57, 13 November 2012 (UTC)

Ah...did I mess up its archiving by responding to it? Agreed though that it's likely not worth giving too much attention to herein as it's likely that those who wouldn't be familiar with or willing to bother researching the topic would dispute it rather than discuss it. Perhaps it belongs more on the pages for perfection or perfecting auxiliaries (however those are worded on WP), or if within this article, included within the greater discussion of auxiliary BE as one of the possible uses. I think Tjo3a has been working on that sort of thing.Drew.ward (talk) 17:32, 13 November 2012 (UTC)

I think a form like It is all gone is indeed a historical relic, and my guess is that there are more than a couple such lexicalized forms. I am not aware of any claims that auxiliary be can be employed productively as an auxiliary of the perfect in modern English, and I agree that the observation does not belong in the article in its current state, although I can image that such information might be appropriate for an article that focuses on the perfect in English. --Tjo3ya (talk) 04:12, 14 November 2012 (UTC)

Was

You seem to have left out "was". drh (talk) 00:17, 9 December 2016 (UTC)

Missing auxiliary verbs

"Keep" is a glaring absence from the table. "He kept walking, but she couldn't keep going." etc, etc. --71.222.173.131 (talk) 16:37, 3 April 2015 (UTC)

Keep is not an auxiliary verb. It does not license subject-auxiliary inversion (*Kept he drinking the water?) and it does not license VP-ellipsis (*She kept drinking the water, and he kept, as well).

According to the Merriam-Webster and the Collins, get can be an auxiliary verb. Shouldn’t it get included in the list? Palpalpalpal (talk) 08:48, 3 November 2017 (UTC)

verbs that belong to both auxiliary and lexical classes: BE

According to the Cambridge Grammar of the English Language, page 92, the auxiliary verbs of English are:

[modals]: can, may, will, shall, must, ought, need, dare
[non-modals]: be, have, do, %use

Need, dare, have, do, use are dually categorised: they belong to both auxiliary and lexical verb classes.

Why isn't the verb BE included with those that belong to both auxiliary and lexical classes? --Backinstadiums (talk) 10:54, 6 October 2019 (UTC)