On page 137 of First Certificate Trainer by Peter May (Cambridge Books for Cambridge Exams), the last paragraph in Test 4, Use of English Part 2 (a cloze test on a short text entitled Safe camping), reads:

After meals, pick up any bits of food that might be left on the ground, as these can attract insects – or larger creatures. It also makes sense, for the same reason to keep unused food in closed containers well away from the camp. You don't want a hungry bear or other animal suddenly appearing in your tent!

Why not the structure "(not) to want someone to do something"? Why not "You don't want a hungry bear or other animal to suddenly appear in your tent"? (Or "suddenly to appear in your tent", or "to appear in your tent suddenly"… it is not the place of the adverb I am interested in here.)

The structure "(not) to want someone doing something" is found in the Cambridge Dictionary online, with the example sentence "I don't want a load of traffic going past my house all night, waking me up." But no explanation is given as to when one structure should be used, and when the other.

  • 3
    You're right; it seems illogical to use a continuous tense. But it's quite idiomatic (ie commonly used and accepted) in this sort of construction. Perhaps it just sounds less po-faced-formal than the equivalent to-infinitive catenation. May 31, 2014 at 11:41
  • 4
    Admittedly it's a small sample size, but Google Books says it has 42 instances of don't want them turning up compared to just 7 for don't want them to turn up, which accords with my own expectation (and preference). So I'm don't really agree with "Why not the more usual structure?" above. The actual citation seems far more "normal" than OP's alternative to m e. May 31, 2014 at 11:57
  • @user58319: I don't think turning up is a gerund here. Since a gerund is effectively a noun, that usage would result in don't want their turning up - just about credible, but unlikely. May 31, 2014 at 12:00
  • 1
    @EdwinAshworth: thank you very much for this (to me) new adjective 'po-faced' – perhaps from po 'chamber pot' (19-20 centuries), from French pot, influenced by poker-faced – says my Longman Dictionary of Contemporary English. Anything that sounds dirty must be French, mustn't it?! This rich English of yours is the reason why I keep returning to this site and its bunch of irritating advocates of usage against any kind of rule, any kind of logic. I love-hate you all! To me, it seems logical to use a gerund or a continuous form here!
    – user58319
    May 31, 2014 at 12:03
  • 1
    @Fumble But with different verbs, the results are quite the opposite. “Don’t want you thinking” yields only 70 results, whereas “don’t want you to think” yields 507,000 (!). May 31, 2014 at 12:06

6 Answers 6


I commented that I don't think OP's turning up was a gerund, but I'm somewhat backtracking on that point now. Whatever - even if it's not formally a gerund, it looks quite like one. And there's been a general tendency over the past century or so for gerunds to replace infinitive forms.

The main factor controlling the gerund/infinitive choice seems to be the specific verb involved. Thus, looking at a simpler construction...

He enjoys swimming (not He enjoys to swim)
He wants to swim (not He wants swimming)
He likes swimming/to swim (both are okay)

There's also this from englishpage.com, pointing out that infinitives sound more abstract...

Gerunds sound more natural and would be more common in everyday English.
Infinitives emphasize the possibility or potential for something and sound more philosophical.

Learning is important (normal subject)
To learn is important (abstract subject - less common)
The most important thing is learning (normal complement)
The most important thing is to learn (abstract complement - less common)

To which I would add my own example...

I like you to kiss me (sounds dated, formal, starchy to me, and all 6 results are from long ago)
I like you kissing me (sounds perfectly natural, but all 58 results are from the past couple of decades)

Putting all that together, what it means is native speakers today like using gerunds wherever possible, whereas a century ago they liked to use infinitives. Actually, I just used those two forms to illustrate my point. Strictly speaking if we go back far enough, native speakers had no choice (because using gerunds like that simply wasn't grammatically possible).

As to whether there's any semantic difference, I think the answer is No, not usually. But when I consider the following pair...

I hate to eat alone
I hate eating alone

...I can easily convince myself it's more likely the first speaker doesn't actually eat alone very often. There's more "immediacy" in the gerund/continuous verb form used by the second speaker, which suggests to me he's currently eating alone (or at least, frequently does so).

Here's another related context where the infinitive/gerund choice definitely makes a difference. I know it's only because of the "when", but still...

Sports teacher: "Okay, you kids get changed and go outside while I finish my paperwork - I want you running/to run round the track when I come out".

EDIT: By way of explaining @user58319's assertion that the infinitive "implies some kind of control or influence". This just arises because in some contexts the implied subject is in fact the speaker...

"I don't like to smoke in bed" (so I don't)
"I don't like smoking in bed" (so please can you not do it)

  • @user58319: I do really think you're stretching a point there! As I hope I've shown above, the biggest factor by far is simply gerund = modern. Semantic distinctions are usually minimal (if they exist at all) and probably vary on an almost "case by case" basis. And if I'm forced to distinguish "He likes to work" from "He likes working", for example, I would say the infinitive version is better able to carry the sense that he [sometimes] chooses to work [on personal projects] simply because he likes the activity. The more familiar gerund just suggests he likes doing his [necessary] job. May 31, 2014 at 19:44
  • Your "immediacy" is just my idea of "powerlessness" in a different guise! The first speaker does all he can to avoid eating alone, and he is rather successful, whereas the second speaker does not like this situation – eating alone – but is powerless to change it. I agree that infinitives, like simple tenses, are – or have come to be felt to be – 'skeletal' (your 'abstract', 'philosophical') in comparison to gerunds, and continous tenses, which are 'fleshed out', alive… which is only logical since a continous tense shows an action in progress, 'presentifies' it…
    – user58319
    May 31, 2014 at 19:46
  • 1
    It seems to me you're trying to impose a single semantic distinction that simply doesn't exist. Crucially, the trend is to use gerunds more and more. And for any given context, once the gerund is "established", it takes over the "most likely" interpretation. So we look for a "slightly left-field" interpretation if the infinitive is used. But getting "established" is complex. Compare I like you doing that (normal), I like you to do that (bit weird), I liked you doing that (normal), I liked you to do that (very weird). Even tense affects what sounds "normal". May 31, 2014 at 20:07
  • 1
    @FumbleFingers Really did not expect that, considering the negativity my answer raised. I almost deleted it... Thanks for the appreciation.
    – fev
    Jul 19, 2021 at 10:58
  • 1
    @fev: You obviously put quite some thought into your answer, which definitely enhanced my perception of the issues involved here. I don't necessarily think everyone (including myself! :) should have the same opinion about possible nuances of meaning here. But you've given me (and hopefully others, now and on into the future) some interesting food for thought - which I figured was the best I could hope for anyway (so thanks for giving me something to be appreciative of! :) Jul 19, 2021 at 14:27

I agree; the use of the present participle here does seem rather illogical. However much we would like to think that English constructions are well-behaved, we'll quite often be disappointed. It might be instructive to examine the behaviour of other verbs in similar complex catenations here, after an example of a simple catenation:

He remembered paying.

is obviously (and famously – but I would use '-ing form' rather than 'gerund' here; see FF's comment) different in meaning from the catenation with the to-infinitive:

He remembered to pay.

So we see the participle 'paying' used in a punctive (completed short action) role.


I don't want him turning up.

I don't want him to turn up.

either is acceptable, and there is little change in meaning. This pair come close to OP's examples, but even the inclusion of the adverb 'suddenly' alters 'how it sounds'. 'I don't want him suddenly to turn up' sounds very formal, even starchy; 'I don't want him to suddenly turn up' sounds better but still not as idiomatic as 'I don't want him suddenly turning up'.


I didn't allow him to come.

the to-infinitive is mandatory.


I can't see/imagine him coming / turning up [...: he never goes to parties.]

the -ing form is mandatory, even though 'turn up' is punctive.

So, -ing form catenations are not restricted to continuous processes. Even where there otherwise seems a fairly straight choice, with the -ing form perhaps not as appropriate, it may be perversely preferred with some verbs. Choice is probably influenced by the over-formal sound of the to-infinitive alternative.

  • You agree… with yourself! Could you comment on your 'little change in meaning'because THIS is the whole point… sounds like 'there is something I know, but I won't tell you, because you wouldn't understand, anyway!' Highly irritating!
    – user58319
    May 31, 2014 at 13:32
  • I was agreeing with [the implications of] what you said in your original pre-edit question: Why not the more (usual)[predictable] structure "(not) to want someone to do something"? Why not "You don't want a hungry bear to appear ... The linked Wikipedia article discusses the differences in meaning of certain simple catenations. As far as I know, an equivalent list dealing with complex catenations does not exist. As for why certain catenations are permitted, others aren't, and some have counterparts with different meanings, I'd be surprised if the doctorate has yet been written. May 31, 2014 at 13:41
  • Going on to your "Could you comment on your 'little change in meaning'" I was leaving a little leeway here. Others might claim a nuance suggested by one but not the other in this particular case, but I'd say register swamps that. May 31, 2014 at 19:46

This Ngram of "don't want you *" shows that to (I'm sure it's not a preposition) is by far the most productive. So it's unlikely that not want someone/something to infinitive is somehow inferior to not want someone/something V-ing. If anything, it's the other way around.

I've heard somewhere that Ngram is different from Google Books, but I think the former carries more weight because it's more comprehensive. Having said that, I don't think there's any significant meaning difference between to-infinitive and V-ing after not want someone/something.

There's one thing I can think of, though, about how the V-ing can come after not want someone/something but not as readily after its positive counterpart want someone/something (unless you want to express the progressive aspect as in the example I want you running round the track when I come out suggested by @FumbleFingers). And that one thing is the possibility of native speakers parsing not want as a single verb such as dislike, mind or even resent. Note that these verbs allow someone/something V-ing but not someone/something to infinitive:

He disliked/minded/resented his daughter staying away from home.

*He disliked/minded/resented his daughter to stay away from home.

This does not mean that they mean "dislike/mind/resent etc." by saying not want. It's just that they could be treating not want as a single verb without changing its meaning. So there's no significant difference in meaning.

As long as they don't parse not want as a single verb, they still can freely use to infinitive. Therefore, the two structures can co-exist.


… and here is why:

'(not) to want someone to do something' implies some kind of control or influence over what the speaker tries to encourage someone to do or discourage them from doing, some kind of power over them.

In the case of the bear, obviously, you just do not want the thing to happen but you know perfectly well you do not have any influence on it happening or not! Apart from taking precautions such as not keeping any food in or near the tent…

And it is more often the case with things we do not want to happen than with things we do want to happen; the negative can carry the idea or powerlessness, which would explain why 'not to want someone doing something' can be found together with 'not to want someone to do sth'!

There is a reason behind usage!

  • 3
    I disagree with the statement that “‘(not) to want someone to do something’ implies some kind of control or influence”. Whether you use the infinitive or the participle, control or influence is entirely dependent on the context, not the grammatical structure. “I don’t want you to die!” is perfectly common, even though obviously we have no control or influence whatsoever over whether someone else will die or not. May 31, 2014 at 12:13
  • 4
    I see absolutely no such distinction. None at all. Imagine, for example, a parachute instructor saying, “I’ll come round and double-check that you all have everything set up properly and safely. We don’t want you getting hurt on your very first jump!”. Are you seriously saying that would be a case of expressing a feeling a kind of powerlessness over the situation, while if he had said “we don’t want you to get hurt”, he would feel in control? That is quite obviously not the case. He’s doing what he can to influence the situation in either case. May 31, 2014 at 12:28
  • @JanusBahsJacquet: 'I don't want you to die!' means 'I will do all I can for you not to die!'… and I want you to believe in my power to do so. 'I don't want you dying!' would sound fatalistic, I do not want this to happen but there is nothing I can do to prevent it… a very weak way of supporting anyone about to die! –
    – user58319
    May 31, 2014 at 12:31
  • 1
    I understand your point, and I agree there must be something governing the to verb/verbing choice in such constructions. I must be using some principle to predict reasonably accurately which pairs of verbs are more likely to favour -ing, before finding confirmation on Google Books. So far I'm not consciously aware what that "something" is, but I don't think it's a matter of whether you (or whoever the subject is) has any control over the activity. May 31, 2014 at 14:29
  • 1
    'I/We don't want you dying' is in a less formal, less sombre, more tongue-in-cheek register. You wouldn't use that register with someone who was near death's door. Jun 1, 2014 at 7:28

We are discussing the catenative verb don't want in a complex construction with a gerund-participial. The subject of the catenative verb is you and its object is the non-finite clause a hungry bear or other animal suddenly appearing in your tent, where the NP a hungry bear or other animal is the agent of the verb in gerund appearing. "Want" in the affirmative is generally said to be followed by a "to infinitive" (except for the concealed passives of the kitchen wants painting type). Even in the negative, if the subject of "want" is the same as the agent of the verb of the object clause, the to infinitive will be used ("I don't want to leave"). So it is good to narrow the issue you are raising to the particular instance of non-affirmative catenative "want" followed by an object clause where the agent of the verb is different from the subject of "want".

In theory, the difference in meaning between I don't want you doing and I don't want you to do is negligible. I found a short reference to the use of gerund-participials after want and don't want in Rodney Huddleston & Geoffrey K. Pullum (The Cambridge Grammar of the English Language, pp. 1231-1232):

The gerund-participial with want generally has a progressive interpretation, but in non-affirmative contexts it can be non-progressive.

  • I want them standing when the Minister enters

is equivalent to

  • I want them to be standing when the Minister enters,

contrasting with non-progressive

  • I want them to stand when the Minister enters.

However, in

  • I don't want you bringing your dog with you,

the meaning is to bring, not to be bringing.

However, almost 20 years passed since that grammar was published and anyway it was only concerned with the progressive or non-progressive nuance of the meaning. Gerunds have seen an increase in use, and although it is difficult to find more exhaustive explanations on this issue, I think people in their conscience may have begun to give it a slight nuance. In your context, it looks like the author of the statement explains generally a situation in which this sudden turning up of the bear might happen every time the situation is repeated. So

You don't want a hungry bear or other animal suddenly appearing in your tent [every time you leave bits of food on the ground or keep unused food in open containers in the camp].

The author may have felt that

You don't want a hungry bear or other animal to suddenly appear in your tent.

is more punctual, for one particular situation, rather than for repeated instances of the same circumstances.

In support of that, I found this post of a native English speaker:

Go for the gerund, if you want to stress repeated action: "I don't want you coming home so late". I don't want you to come home so late,(using the infinitive), is more likely to infer the action happening once, or occasionally, of course depending on the context. (WordRef)

Here is another thread that agrees with the two nuances:

There's usually no difference intended in sentences like this, but here are a few comments on nuances. "I didn't want anyone thinking I was trying to be a pet of the teacher" sounds a bit like the 'anyone' might be thinking this for a long time. I didn't want that. "I didn't want anyone to think I was trying to be a pet of the teacher" sounds a bit like maybe the 'anyone' might just think this briefly.

Another post from an MA Linguistics says:

The nuance is slightly different.

“I don’t want you to do that” has the slight implication that you haven’t done it before. Mother says to child, “I brought some chips home, but I don’t want you to eat them.”

“I don’t want you doing that” implies that you were caught in the act of doing it, or have been in the habit of doing it. Mother says to child, “That’s the third time all the chips have disappeared! I don’t want you eating those chips!”

But it’s just a slight nuance. Really, either form could be used in either case.

This last post adds another slightly different nuance, that of habit, but the emphasis on repetitiveness conveyed by the gerund-participial and non-repetitiveness conveyed by the to infinitive is obvious. It could point to the habit that bears have of coming near humans when they smell food.

  • Fine detective work. Jul 12, 2021 at 15:49
  • 2
    I don't understand how you could argue the OP's context supports a repeated action. Even if it's a repeated action by way of being a "habit", you cannot explain why the affirmative want doesn't allow V-ing for the same repeated action.
    – JK2
    Jul 12, 2021 at 15:53
  • @JK2 The OP did not ask to explain why the affirmative of "want" does not license V-ing. And the fact that it does not license it is very well documented.
    – fev
    Jul 12, 2021 at 16:07
  • 1
    No, OP didn't. Regardless, I was just pointing that out to show that the "very well documented" behavior of the affirmative want does not support your logic.
    – JK2
    Jul 12, 2021 at 16:09
  • Also, it's a very bold argument that the verb usage could change so dramatically within just two decades that CaGEL cannot be trusted. Actually, it's not even two decades. Two other posts you've cited are written in 2011, which is less than a decade from CaGEL's publication. So you're essentially arguing that the verb usage has changed in several years. Good luck with that. More importantly, even if that were all true, it's hard to believe that people would be making that kind of subtle distinction when speaking/writing these constructions.
    – JK2
    Jul 12, 2021 at 16:11

I would simplify this by adding both would be possible in this particular situation.

However the use of 'appearing' puts you into the perspective of it happening at that moment of hearing the phrase spoken, rather than being a future idea. It makes it feel more present, thus more possible to the listener.

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.