You generally have two types of "have":

(1) He has two sons. (stative)

(2) He has lunch alone. (dynamic)

A stative "have" can be followed by "got", whereas a dynamic "have" cannot:

(1a) He has got two sons. (stative)

(2a) *He has got lunch alone. (dynamic)

A dynamic "have" can appear in the progressive, whereas a stative "have" cannot:

(1b) *He is having two sons. (stative)

(2b) He is having lunch alone. (dynamic)

Now, is the following "have" stative or dynamic?

(3) I have to go.

(3a) I have got to go.

(3b) I'm having to go.

  • 8
    Neither one. First of all, it's not have, it's have to, an idiom, pronounced /'hæftə/. Second, it's not a meaningful verb, it's a modal paraphrase. Have to means must, and like must, it's folowed by an infinitive form of the verb (in this case, go). So it's neither stative nor active, but rather a modal, which belongs to a different category. Apr 14, 2014 at 3:35
  • 1
    Does this "have" have any syntactic attribute of a modal auxiliary at all?
    – JK2
    Apr 14, 2014 at 4:42
  • 1
    To repeat: It's not have. It's spelled that way, but it's not pronounced that way and it doesn't have anything to do with the verb have, except historically. The have and the to are fused together and not pronounced like two words; in particular, they are not pronounced like the two words have and to. It requires an /st/ in the middle, never /z t/. As for syntax, hasta has more possibilities than the real modals: it allows do-support (I don't hafta do ir; Do I hafta do it?), and it has a past tense (I hadta do it). You can't do that with must. Apr 14, 2014 at 13:52
  • 3
    If "have to" is fused as a single unit, why is it possible to say "have only to"? As for its pronunciation, how do you pronounce "have to" as in "Let's hear what they have to say"? Do you pronounce the "have to" here any differently?
    – JK2
    Apr 14, 2014 at 22:25
  • 1
    BTW, aren't things like 'do-support' and 'past tense' characteristics of the lexical verb and never those of the modal?
    – JK2
    Apr 14, 2014 at 22:40

3 Answers 3


Now, is the following "have" stative or dynamic?

  • (3) I have to go.

  • (3a) I have got to go.

  • (3b) I'm having to go.

It seems that -- if I'm understanding CGEL right -- that your examples are all using the stative "HAVE" (not the dynamic "HAVE"). And that it is a lexical verb in your 1st and 3rd versions (#3 and #3b), while it is an auxiliary verb in your 2nd version (#3a) due to the "have got".

I intend to show some excerpts from CGEL that is directly related to this topic, but first I'll have to provide some background info (that's also in CGEL). For instance, some background info on the properties of auxiliary verbs and modal auxiliary verbs.

Auxiliary verbs have syntactic properties that distinguish them from the open class of lexical verbs; and modal auxiliaries have an additional set of properties that distinguish them from the other auxiliaries (CGEL page 92). A quick summary and examples of these properties follow, which use "WILL" for the examples. CGEL page 108:


  • [A] Primary verb negation - - - It will not work.

  • [B] Subject-auxiliary inversion - - - Will it rain?

  • [C] Emphatic polarity - - - I WILL help you.

  • [D] Stranding - - - He won't attend, but I will __.

  • [E] Exclusion of "DO" in code - - - (*)Ed will go and I do too. ("I will go")

  • [F] Precede adverb/ quantifier - - - They will probably/all accept.

  • [G] Negative forms - - - It won't help.

  • [H] Reduced forms - - - She'll be here soon.


  • [I] Only primary forms - - - (*)It's expected to will finish soon.

  • [J] No agreement - - - She will win. vs (*)She wills win.

  • [K] Only bare infinitival complement - - - It will be over. vs (*)It will to be over.

  • [L] Can occur in remote apodosis - - - If it weren't for her I would give up.

  • [M] Modally remote preterite in main clause - - - I would ask you to treat it seriously.

In general, the auxiliary verbs will have the properties [A-H], and the modal auxiliaries will have those properties and the additional properties [I-M]. Although the prototypical central modal auxiliary verbs will have the properties [A-M], there are some modal auxiliaries that don't have all of them, e.g. "must".

Let's first see some info about the dynamic "HAVE". The dynamic "HAVE" is a lexical verb (not an auxiliary). CGEL page 111:

(b) Dynamic have

This is a lexical verb in all varieties of English:


  • i. He had a swim. He didn't have a swim. Did he have a swim?

  • ii. He had it painted. He didn't' have it painted. Did he have it painted?

It has none of the auxiliary properties [A-H] -- cf. (*)He hadn't a swim, (*)Had he a swim?, etc. As the label 'dynamic' indicates, it expresses an event rather than a state. In [i] it is used as a 'light verb' (Ch. 4, &7): the main semantic content is in the following noun. Similar is its use with the meaning "experience": I didn't have any difficulty in persuading her. In [ii] it is a catenative verb taking an object and a past-participial complement; it can also take a bare infinitival: Did she have you retype it?

The constructions that involve "have to V" use the stative "HAVE"; and those usages are of a lexical verb and not an auxiliary (except for some dialects). CGEL page 111-2:

(c) Stative have and the idiom "have got"

Where have expresses a state rather than an event it is replaceable by the idiom have got (subject to conditions outlined below):


  • i. a. She has a swim every day. - - - b. (*)She has got a swim every day. -- [dynamic]

  • ii. a. She has a swimming-pool. - - - b. She has got a swimming-pool. -- [stative]

  • iii. a. She has to swim each day. - - - b. She has got to swim each day. -- [stative]

Stative have occurs either with an NP object, as in [ii], expressing possession and similar relations (cf. She has [many virtues / two sons]), or as a catenative verb with a to-infinitival complement, as in [iii], where the meaning is of obligation or necessity, much like that of "must" (see &9.11). Note that in spite of its semantic similarity to must this have has none of the modal properties [I-M]. For example, it has secondary forms and hence can appear in the progressive (I'm having to work late tonight), the perfect (I've had to do it all myself), and with a modal (We may have to cancel it). There is therefore no case at all for including it in the syntactic class of modal auxiliaries.

That above info about [57.iii] seems to be directly related to what you were asking about in your post (because it uses the "have to V" construction). Also, note the progressive example: "I'm having to work late tonight" (which has a form similar to your 3rd example #3b).

Within this same section of the CGEL, there is this following bit which might be helpful too. Page 112:

. . . While dynamic have is invariably a lexical verb, stative have can behave as either a lexical verb or, in some varieties, an auxiliary. This means that for the negative we have either don't have or haven't (or the analytic forms with not), and analogously with inversion. If we include have got too, we find therefore the following possibilities:


  • i. a. I have enough tea. - - I don't have enough tea. -- Do I have enough tea? - - (%)I haven't enough tea. - - (%)Have I enough tea?

  • i. b. I have got enough tea. - - I haven't got enough tea. -- Have I got enough tea?

  • ii. a. I have to read it all. - - I don't have to read it all. -- Do I have to read it all? - - (%)I haven't to read it all. - - (%)Have I to read it all?

  • ii. b. I have got to read it all. - - I haven't got to read it all. - - Have I got to read it all?

This following excerpt also has info related to the OP's topic. On page 113:

Stative have: lexical or auxiliary (doesn't have vs hasn't)

In AmE stative have always behaves as a lexical verb (and is preferred over have got). In BrE the lexical use has become common too, and the auxiliary use is tending to sound relatively formal or old-fashioned (with have got or lexical have preferred). The auxiliary use is hardly possible with a habitual interpretation: Have you to mow the lawn?, for example, shows the same restriction as have got in [60.i]. (F.E.: [60.i] I've got to mow the lawn. - single obligation)

And then there's the topic of modality, where "HAVE" is involved too, as a lexical modal (and as an auxiliary in the idiom "have got to V"); and there's also a "dynamic" term involved: deontic vs dynamic vs epistemic. CGEL page 205-6:

9.11 Lexical modals have (got), and quasi-modal be

(a) Strong modality

Have and have got


  • i. You have (got) to come in now. - - [deontic]

  • ii. Now that she has lost her job she has (got) to live extremely frugally. - - [dynamic]

  • iii. This has (got) to be the worst restaurant in town. - - [epistemic]

Have and have got are most commonly used for deontic necessity, as in [i]. Here they characteristically differ from must in being objective rather than subjective: with [i] I'm likely to be relaying someone else's instruction but with You must come in it's more likely that I am myself telling you to. With dynamic necessity, as in [ii], they are more likely than must. The epistemic use of [iii] is widely found in AmE, but is still fairly rare in BrE. The differences between have and have got are dealt with in &2.5.6. (fn 65)

In that above excerpt, those uses of "HAVE" are of the stative "HAVE" (especially since they have alternations with "have got"), and they all involve the construction(s) "have (got) to V".

So in conclusion, if I'm understanding all this stuff in this post properly, it seems that the OP's three examples (#3, #3a, #3b) are all using the stative "HAVE"; and that those versions that use "HAVE got", the "HAVE" is an auxiliary verb, otherwise it is a lexical verb (and also a lexical modal). And all three involve strong modality due to using either the lexical modal "HAVE" or the idiom "have got" in their "have (got) to V" constructions.

NOTE: CGEL is the 2002 reference grammar by Huddleston and Pullum et al., The Cambridge Grammar of the English Language (CGEL).

  • Thanks for the detailed answer, F.E. I do appreciate it. I do own a copy of CGEL myself. But if this HAVE is stative, how could you possibly have it in the progressive as in (3b)?
    – JK2
    May 1, 2014 at 2:22
  • I mean, the very second you're using it in the progressive, you're using the verb as a dynamic verb, aren't you?
    – JK2
    May 1, 2014 at 2:24
  • 1
    The "dynamic vs stative" is actually "event vs state". The stative "HAVE" describes a state: "I'm having to work late tonight" describes a state where you are obligated now to work later tonight (close in meaning to "must" in meaning). The difference between the progressive and non-progressive versions probably isn't that much.
    – F.E.
    May 1, 2014 at 2:38
  • You're right that "I'm having to..." describes a state. But the issue is not whether the entire phrase is dynamic or stative but whether the single verb HAVE is dynamic or stative. If this HAVE is indeed stative, why would you use it in the progressive in the first place?
    – JK2
    May 1, 2014 at 2:46
  • @JK2: "the issue is not whether the entire phrase is dynamic or stative but whether the single verb HAVE is dynamic or stative". Sorry, wrong. Stative and active (not "dynamic", btw) are categories that apply to predicates. Predicates may consist of multiple words. Some of the words are spelled the same way as some other words, for historical reasons, like the to of today. That doesn't mean they are the other word, however, and they don't have the same properties. So, as I said at the beginning, "the verb have" in have to is neither active nor stative, but part of a modal. Aug 31, 2014 at 15:46

(3) I have to go.

(3a) I have got to go.

(3b) I'm having to go.

This use of "have" isn't the same word as your other examples. Here, it means "need to", "should" or "must":

(4a) I need to go.

(4b) I should go.

(4c) I must go.

(3) and (3a) would both be translated into (4) and all three have the same meaning. Note, however, that the "to" or "got to" are only related to "have" and removing "have" will not necessarily allow "to" or "got to" to remain.

(3b) sounds unnatural to me and I'm not sure what it means.

But to answer your question:

Now, is the following "have" stative or dynamic?

"have to" and "have got to" are stative in the sense that they are describing the state of needing to leave. This is evidenced by (4a-b).

There is no realistic way describe this meaning in a dynamic way sense the dynamic variants will completely remove "have to":

(5) I am going

  • Good answer except that there's a huge difference between 'should' and 'must'! 'Have to' is a modal of necessity, so it shouldn't be paraphrased as 'should'. Jun 23, 2014 at 3:08

I would see "I have to do" as a shortening of "I have the obligation to do", so there is no sense in saying "I'm (in the act of) having to do.The same for "I've got to do".

  • Are you saying it's not possible to say "I'm having to do" or "I've got to do"??
    – JK2
    Apr 14, 2014 at 4:38
  • I mean it is unnecessary to say "I'm having to do" or "I'm having got to do".
    – rogermue
    Apr 14, 2014 at 4:54
  • Where did "do" come from? The question is asking about "go".
    – MrHen
    Apr 28, 2014 at 15:57
  • 1
    @MrHen When I say "I have to do" "to do" stands for any verb that can follow. This notation is more practical than to write "to have + to-infinitive".
    – rogermue
    Apr 28, 2014 at 16:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.