"Have" without "do"

[Phaedrus]

If auxiliary verbs are helping verbs, copula BE must be both a lexical verb and an auxiliary: it helps itself.

 

[jutfrank]

If auxiliary verbs are helping verbs, copula BE must be both a lexical verb and an auxiliary: it helps itself.

But in what sense is it lexical? What does that actually mean in syntactic terms? Doesn't it just mean it doesn't have NICE properties? It's lexical in virtue of the fact that it's not auxiliary?

 

[Phaedrus]

But in what sense is it lexical?

Copula BE is lexical in that it functions as the main verb of a clause: The tree is green. Lexical verbs function as main verbs. Progressive auxiliary BE and passive auxiliary BE cannot function as the main verb of a clause. They must be complemented by verbs in the -ing and past-participle forms, respectively, whether as main verbs (The tree is shedding leaves; The tree is blown by the wind throughout the day) or as additional "helping verbs" (The tree is being chopped down).

What does that actually mean in syntactic terms?

In Chomskyan generative grammar, a verb (V) with the [+AUX] feature undergoes movement to T in a finite clause when it is the uppermost (or first) verb in a sequence of verbs. Its presence at the T node is what allows it to undergo T-to-C movement (NICE property I) and Neg contraction (NICE property N), as well as to receive clausal emphasis (NICE property E) and license verb phrase ellipsis (NICE property C). Copula BE moves to T, just as progressive BE and passive auxiliary BE do.

Doesn't it just mean it doesn't have NICE properties? It's lexical in virtue of the fact that it's not auxiliary?

Lexical BE does have the nice properties (see Post #22). For those of us who lack British HAVE, copula BE is the only verb in the English language that is simultaneously a lexical/main verb and an auxiliary verb in certain circumstances. Those circumstances are as follows.

1. The clause must be finite. Copula BE is not an auxiliary in nonfinite clauses like the infinitival clause in "He believes the tree to be green."

2. The clause must be a root clause. Copula BE is not an auxiliary in, e.g., "He wishes the tree were green," "He thinks that the tree is green."

3. Even in finite clauses, copula BE will only be an auxiliary if it is the uppermost (or first) verb in a stack of verbs. Copula BE is not an auxiliary in "The tree has been green for many months." In that sentence, perfective HAVE is the auxiliary.

 

[jutfrank]

Copula BE is lexical in that it functions as the main verb of a clause: The tree is green. Lexical verbs function as main verbs.

Okay, so you share with Piscean the idea that an auxiliary must be 'auxiliary to' another verb. Is that right?

Progressive auxiliary BE and passive auxiliary BE cannot function as the main verb of a clause.

It isn't that they cannot function as the main verb of a clause, but just that they are not called 'main verbs' when they function as auxiliaries. If they did function as the main verb of a clause, they obviously wouldn't be auxiliaries. Right?

In Chomskyan generative grammar, a verb (V) with the [+AUX] feature undergoes movement to T in a finite clause when it is the uppermost (or first) verb in a sequence of verbs. Its presence at the T node is what allows it to undergo T-to-C movement (NICE property I) and Neg contraction (NICE property N), as well as to receive clausal emphasis (NICE property E) and license verb phrase ellipsis (NICE property C). Copula BE moves to T, just as progressive BE and passive auxiliary BE do.

I don't understand that very well. What's T? The subject position?

It seems that people are using different criteria for determining what does and what doesn't count as an auxiliary. What I'd like to know is what these criteria are.

 

[Phaedrus]

Okay, so you share with Piscean the idea that an auxiliary must be 'auxiliary to' another verb. Is that right?

No, I'm saying that, e.g., "is" is an auxiliary in "The tree is green," in which copula BE is the only verb present. It's not an auxiliary to any other verb.

It isn't that they cannot function as the main verb of a clause, but just that they are not called 'main verbs' when they function as auxiliaries. If they did function as the main verb of a clause, they obviously wouldn't be auxiliaries. Right?

I don't think progressive BE or passive BE have the capacity to be main verbs.

*It is. (does not have progressive or passive meaning outside of context; meaningless unless the meaning is "It exists")
*He is. (ditto)

I don't understand that very well. What's T?

It's the Tense node of TP (Tense Phrase), a.k.a. the I (Inflection) node of IP (Inflectional Phrase). You wanted syntax, so I mentioned it.

It seems that people are using different criteria for determining what does and what doesn't count as an auxiliary. What I'd like to know is what these criteria are.

Below is a link to a paper that clearly articulates the account of auxiliary-verb-ness that I follow. It was written by a former syntax professor of mine and the best grammarian I know (a longtime close colleague of Geoffrey Pullum's, incidentally; their offices were right nextdoor to each other for at least two decades). You'll also see the contemporary TP machinery at work. Enjoy!

https://babel.ucsc.edu/~hank/jh.papers/nfneg.pdf

P.S. The point of the above paper is not to analyze what auxiliary verbs are. Nevertheless, it gives a clear and succinct statement of their nature in passing.

 

[PaulMatthews]

PaulMatthews, do you consider copula BE an auxiliary verb?

Yes: copula "be" is an auxiliary; in fact (with one very minor exception) "be" is always an auxiliary.

 

[PaulMatthews]

For me, the prime characteristic of auxiliary verbs is the traditional function of 'helping' non-finite forms other verbs in the formation of verb phrases showing:

aspect: We are eating, We have eaten.
'code':
...ellipsis: I eat, and so does he,
...tag questions: He eats, doesn't he?
emphasis: We do eat.
interrogative inversion: Do we eat?
'mood': We can/could/may/might/must/shall/should/will/would eat.
negation: We don't eat.
voice: It is eaten.

Within those verb phrases, auxiliary verbs can also show:

person/number: He is eating, they are eating.
tense: He doesn't eat, he didn't eat.

Auxiliary verbs have the NICE properties, but these properties are not an exclusive marker of auxiliary verbs. If they were, every verb in earlier forms of English would be classified as an auxiliary, as would the verb in such idiomatic utterances as How goes it?, Ask not what your country can do for you.

The lexical verbs BE and HAVE [. . .] also conform to the NICE properties, but we will not regard them as auxiliary verbs. The reason is that they can occur on their own in clauses, whereas auxiliaries can't. Aarts, Bas (2011.69, Oxford Modern English Grammar.

It's the NICE properties that mark "be" as an auxiliary verb. In fact, leaving aside one minor construction, “be” qualifies as an auxiliary not just when it is marking progressive aspect or passive voice, but also when it's the only verb in the clause taking a complement with the form of an AdjP, NP etc.

Static "have" can behave as either an auxiliary ("Have we enough money?”) or a lexical verb ("Do we have enough money"). I don’t think there’s much more we can say about that.

Regarding your use of the term 'helping verbs':

Consider a pair of examples like:

[1] She was painting the house.
[2] She began painting the house.

"Be" in [1] is an auxiliary verb while "begin" in [2] is not. But this doesn’t mean that there is a difference in grammatical structure. "Was painting" is not a grammatical constituent any more than "began painting" is. "Painting the house" is a non-finite clause functioning as complement of the preceding verb, "was" and "began" respectively. The differences between "be" and "begin" with respect to negation, inversion, etc. (the 'NICE' properties) don’t provide any evidence for saying that there is a difference in grammatical structure, for these properties apply in examples like

[3] She was ill.
[4] She hadn’t any money.

No substance can be given to the idea that auxiliaries are "helping verbs": there’s no basis for saying that "was" in [1] is a helping verb while "began" in [2] is not. I would thus strongly recommend dropping the term "helping verb". In this connection compare

[5] You needn’t come if you don’t want to.
[6] You don’t need to come if you don’t want to.

"Need" is an auxiliary verb in [5] but not in [6]; there is however, no semantic difference between these, no sense in which "need" is "helping" in [5] but not [6].

"Auxiliary verb" must be defined in grammatical terms, not in semantic terms. Thus auxiliary verbs are verbs with the NICE properties, not verbs that help other verbs: you couldn’t identify them by asking whether or not they help the following verb. They are called auxiliary verbs because they characteristically express meanings similar to those expressed (either in the same language or in other languages) by verb inflection – but this property provides a basis for naming the grammatical class "auxiliary verb" not for identifying individual instances.

 

[PaulMatthews]

We were discussing primary verb-forms, not secondary (non-finite) ones like the plain form found in imperatives. The lexical "be" that I was referring to is the one found in examples like "Why don't you be more caring".

The point about the term "auxiliary" is that it should be used to identify auxiliaries according as they have the NICE properties, not as they function in individual instances, as the examples I gave illustrate.

 

[Phaedrus]

We were discussing primary verb-forms, not secondary (non-finite) ones like the plain form found in imperatives. The lexical "be" that I was referring to is the one found in examples like "Why don't you be more caring".

The point about the term "auxiliary" is that it should be used to identify auxiliaries according as they have the NICE properties, not as they function in individual instances, as the examples I gave illustrate.

Piscean's examples (1), (2), and (3) in Post #36 and your example "Why don't you be more caring" illustrate that copula BE cannot always behave NICE-ly. It doesn't have the NICE properties in imperatives or when do-support is used.

It is worthwhile to note that neither copula BE nor the other auxiliaries behave NICE-ly when a modal is involved (You should be more caring) or when they come after after another auxiliary (You have been more caring) or in a nonfinite clause (I want you to be more caring, I remember you being more caring).

So it doesn't really make sense to say that copula BE has the NICE properties. It has them only in certain circumstances. It has them (or all of them, anyway) when it (a) is in a finite root clause and (b) is the first or only verb of that clause. Only in such circumstances is it capable of behaving NICE-ly.

Consequently, the formalist approach exemplified in the paper to which I gave a link can be very helpful when one tries to see clearly in this area. The property [+Aux] simply picks out the ability of a verb to move to T. There's no need to have recourse to the NICE properties in the definition.

When a V moves to T, it can behave NICE-ly. Unlike HAVE and BE, modals originate in T (they are T heads) and so behave NICE-ly from the start.

 

[PaulMatthews]

I think I covered those points earlier:

1. I mentioned in #37 that there is an exception where "be" can be a lexical, and I gave the example "Why don’t you be more caring". Elsewhere "be" behaves as an auxiliary, subject to 2. below.

2. I also mentioned in #37 that we are talking here of the primary verb-forms, not the secondary ones used in non-finite clauses (including catenatives) and imperatives.

However, in a chain containing two secondary forms of "be" (progressive and passive), we can have: "They may (modal) have (perfect) been (progressive) being (passive) overlooked", where in addition to auxiliary "have", we have "been" and "being" as auxiliaries.


3. Yes: do-support indicates that "have" is behaving as a lexical verb. But the speaker first has a choice of lexical or auxiliary.

 

[Rachel Adams]

Other meanings as in "have a bath", "have fun", "have a shower" "have dinner" , "have a meal". Right? These should have "do".

I was rereading this thread. You answered my question, but I didn't ask if "do/does/" are required in questions as well. Are they?

As in, "Does she have breakfast?" and not "Has she breakfast?"

As you explained:
"When HAVE means, in the broadest possible sense, possess, some of us older Brits do not use auxiliary DO. With HAVE in its other meanings, we always use auxiliary DO."

 

[GoesStation]

As in, "Does she have breakfast?" and not "Has she breakfast?"

As you explained:
"When HAVE means, in the broadest possible sense, possess, some of us older Brits do not use auxiliary DO. With HAVE in its other meanings, we always use auxiliary DO."

"Have" doesn't mean "possess" in that sentence, so DO is required. But the question is odd unless you're asking about the person's habits. If you wanted to ask whether the person is currently eating, you'd say "Is she having breakfast?"

 

[PaulMatthews]

I was rereading this thread. You answered my question, but I didn't ask if "do/does/" are required in questions as well. Are they?

As in, "Does she have breakfast?" and not "Has she breakfast?"

As you explained:
"When HAVE means, in the broadest possible sense, possess, some of us older Brits do not use auxiliary DO. With HAVE in its other meanings, we always use auxiliary DO."

The distinctions between lexical and auxiliary "have", and between dynamic and static "have" have all been dealt with in great detail in earlier answers.

I don't see any need to go over them all again.