11

Bounty note:

This question is primarily about the acceptability of will in different types of environment. I have used the verb mind in my examples, but if you are an American English speaker, as opposed to a UK one, you might like to substitute this with the verb care (it doesn't make any difference to the question).

The question

Let's suppose that Maria the elephant is going to take part in a performance tomorrow. I am discussing potential acts for her, and my friend responds:

  1. I don't mind what she does tomorrow.

Now that subordinate clause what she does tomorrow uses a present simple construction. If he'd said:

  1. I don't mind what she will do tomorrow.

That would have been passable, but feels a bit wonky for this particular conversation. Here the subordinate clause uses the modal verb will (in what's often called a future simple construction) instead of the present simple.

Now, my friend could easily have said:

  1. I don't know what she will do tomorrow.

Here the future simple seems perfectly apposite. However this time, in contrast to the examples in (1-2), the present simple sounds positively wrong for this conversation:

  1. *I don't know what she does tomorrow.

Ok, so far so bad. However, if we deconstruct the differences in meaning between the two sentences, things become yet more confusing. Here's why. Remember that I was discussing Maria's potential act tomorrow, which means that what act she's actually going to perform is completely undecided. Now if we consider this against sentences (1-2), this seems to be part of the problem with sentence (2):

  1. I don't mind what she does tomorrow.
  2. I don't mind what she will do tomorrow.

It seems, to me at least, that sentence (2) would be perfectly fine if what Maria was going to do had already been decided, but is wonky because nobody, including Maria, knows what she's going to do. So the will here would be ok if what she was going to do tomorrow was already known.

We can compare this with (3-4):

  1. I don't know what she will do tomorrow.
  2. I don't know what she does tomorrow.

Here the version with will is fine whether or not we know what Maria's going to do. In contrast, (4) can only be used in some kind of other situation in which what Maria does has already been time-tabled. So if she does the cha-cha on Tuesdays and the high trapeze every Wednesday and so forth, for example, but the speaker cannot remember which specific thing Maria does on a Thursday—and today is Thursday—then sentence (4) is perfectly appropriate.

So my questions are:

  • Why does "will" sound wrong in the first pair of examples , and why does not using "will" sound wrong in the second pair? (given the situation described).

  • What is it about the verbs mind and know (or care and know) that causes this difference?

Araucaria - Him
  • 47,146

4 Answers4

7

The major difference between the predicates distinguished here with Wh-clause complements is that

  • not know belongs to a class of predicates that takes a Disjunctive Wh-clause complement

whereas

  • not care (in the US) belongs to a class that takes a Conjunctive Wh-clause complement.

The difference is in the pragmatic nature of the Wh-clause;
conjunctive clauses are factive -- they presuppose the identity and truth of their complement. E.g,

  • I am aware of/They'll be surprised by/He doesn't care what she does tomorrow.
    In this example, the clause refers to the set of all actions that she is to do tomorrow,
    as a settled matter; it's called conjunctive because the set consists of A and B and C and ...

while disjunctive clauses are indefinite, and presuppose nothing, E.g,

  • It's a mystery/I wonder/They don't know what she'll do tomorrow.
    In this example, the clause refers to an unknown set of possible events or actions;
    it's called disjunctive because the set consists of A or B or C or ...

Since the conjunctive clause is presupposed, there's no need for a predictive modal like will. But the disjunctive clause is indefinite, and its truth set may be empty ("..., if anything"), so the predictive will is necessary.

John Lawler
  • 107,887
  • Before I'd asked this question, I thought something along these lines. This theory doesn't seem to sit very well, though, with I know what she'll do tomorrow/ I know what she does tomorrow* or It's clear what she'll do tomorrow/ what she does tomorrow* or We have already established what will happen tomorrow ... Any further clarification? – Araucaria - Him Nov 21 '17 at 09:19
  • I know what she does tomorrow is perfectly fine, but it forces a generic reading, with a schedule involved: She always does the mouse trick and the big spray on Tuesdays, so that's what she does tomorrow. Again, factivity is involved; there is a presupposition. – John Lawler Nov 21 '17 at 15:16
  • The same goes for the don't know what she does tomorrow example, doesn't it? – Araucaria - Him Nov 21 '17 at 15:17
  • Generic is usually an available interpretation if the predicate is a repeatable activity. See here for more (but still far from all of the) information on generic VPs. – John Lawler Nov 21 '17 at 19:34
  • 1
    Thanks @john for the links. What I'm trying to work out is whether the fact that conditional protases resist the use of will to express normal futurity is just part of the normal behaviour of will in subordinate clauses. ... I don't know if you have any thoughts about that? – Araucaria - Him Nov 22 '17 at 09:21
  • @Araucaria-Nothereanymore. Have you figured that one out? – JK2 Nov 15 '22 at 06:24
  • @JK2 No, unfortunately not! – Araucaria - Him Nov 15 '22 at 14:22
  • @Araucaria-Nothereanymore. You've probly seen this link, then. I think it's just a feature of will, btw. And I'm not comfortable with "conditional protases" used as if it were a syntactic term; afaics, since context is created bit by bit from any little pieces we can find, everything in a discourse counts as a conditional protasis, if someone can infer something from it. I'm a syntactician and I want syntactic tests for syntactic constructions. – John Lawler Nov 15 '22 at 15:00
  • @JohnLawler I don't get "everything in a discourse counts as a conditional protasis, if someone can infer something from it"? I'm also wondering how to describe that feature of will? ('Cuz we wouldn't be able to say that it doesn't appear in protases/conditional adjuncts!). – Araucaria - Him Nov 15 '22 at 15:41
  • In if clauses, epistemic will does not occur. Modals are full of such disallowances; epistemic can is an NPI. As for protases, I need to unpack; sorry. – John Lawler Nov 15 '22 at 15:46
  • @JohnLawler EDIT: How about in "I don't know if it will rain tomorrow"? – Araucaria - Him Nov 15 '22 at 15:47
  • It's grammatical. It's asserting that the prediction is the speaker's. What's the question? – John Lawler Nov 15 '22 at 15:49
  • It's just that you just said "In if clauses, epistemic will does not occur." ... – Araucaria - Him Nov 15 '22 at 15:50
  • Sorry, misread it. That's not a normal if; that's the if that you can substitute whether for, in an embedded question. It's excluded from the rule because it's part of a special construction. – John Lawler Nov 15 '22 at 15:54
  • 1
    @JohnLawler I think it's more complicated than that. For example, we have "I don't care if/whether it rains tomorrow" where will seems to be similarly restricted. Also we have whether or not it rains versus I don't know whether it will rain. And we have As long as/ unless/ on the condition that/ provided that/ assuming it rains, where epistemic will is similarly disallowed. It would be fine in other adjuncts beginning with because, although, whereas, even though, much as etc – Araucaria - Him Nov 15 '22 at 16:29
  • 1
    If you listed all the environments where can is ability or those where it's epistemic, you might get a similar list. Indeed modals are complicated. And the differences between care and know and think, among other predicates, along with negation, does intrude difficulties also. – John Lawler Nov 15 '22 at 16:51
  • @JohnLawler On the other idea in your earlier comment, I still don't get the "everything in a discourse counts as a conditional protasis, if someone can infer something from it" bit. Could you explain/give an example? – Araucaria - Him Nov 15 '22 at 23:55
  • @Araucaria-Nothereanymore. Do you remember Jon Hanna's answer back in 2015? There his answer was that if I smoked here in Would you mind if I smoke here? is not a complement of mind but a conditional if clause, and you agreed with his answer in a comment below that answer. https://english.stackexchange.com/a/217941/27275 If the same analysis is applicable to "I don't care if/whether it rains tomorrow", doesn't it explain why "will is similarly restricted"? – JK2 Nov 16 '22 at 00:42
  • @JK2 I wasn't paying close enough attention and I didn't know as much as I do now! I was wrong-but with a caveat. You can turn any old clause into a conditional by just sticking an if-clause on it (with a few grammatical restrictions on the if-clause, for example, the presence of will/would). So even if mind can take an if-clause as complement, there's a homophonous sentence where the if-clause is a conditional adjunct. Having said that, mind requires the reconstruction of some sort of complement (null anaphora) if the complement is missing. – Araucaria - Him Nov 16 '22 at 09:44
  • @JK2 So if I meet you and say "I don't mind", you won't have the foggiest what I'm talking about unless you can guess what it was I don't mind from the context. In your JH example "If I smoke here, would you mind?" a) the meaning is different because mind there requires some kind of null anaphora. It could mean mind that I was smoking or it could mean mind something else. Anyhow, it's clunky. We can show that it is at least possible (in fact it's more likely) that the clause after mind is a complement by extracting a wh-word from it. – Araucaria - Him Nov 16 '22 at 09:58
  • @JK2 Conditional adjuncts, like other adjuncts, are islands for extraction. *Which book are you leaving if I read? <--That's not good. However, this seems to be OK --> Which books do(n't) you mind if I read?. Looks like that's a complement. Anyhow, GKP wrote a paper on sentences similar to your JH one where he argued that there is a third IF in English in sentences similar to (but not exactly the same) as yours. It's called Implications of extraposed irrealis clauses (1987). His student wrote a similar master's diss Rocchi (2010) A third if?. – Araucaria - Him Nov 16 '22 at 10:21
  • @Araucaria-Nothereanymore. I don't know if I'm talking about "a third if", suggested in the dissertation, but even if the if-clause now seems closer to a complement of mind (or care in your example), there's no denying that it does look somewhat like a conditional adjunct. Then, isn't it possible to hypothesize that's why "will is similarly restricted"? – JK2 Nov 16 '22 at 11:51
  • @JK2 There's a rule of interpretation for all those clauses (if you agree that the word if is not a preposition after all, but just an interrogative subordinator and that the 'conditional' meaning of if-conditionals is encoded constructionally, not in a magic word). So, yes, I believe you're right, the will restrictions could all relate to that single meaning/interpretation. – Araucaria - Him Nov 16 '22 at 12:07
  • 1
    @Araucaria-Nothereanymore. that sounds like a resolution to my problem with "conditional" being treated as a syntactic entity instead of a pragmatic one. Though I'd still like to encounter one of those homophonous ambiguous sentences with mind where the if clause is a conditional adjunct, as if "conditional adjunct" were a syntactic description. – John Lawler Nov 16 '22 at 15:30
  • 1
    @JohnLawler Girl: The reason I mind hanging out with you in public is you're just not cool. Boy: Would you mind if I smoked and died my hair green? <--- Where the intended interpretation is "If I smoked and died my hair green, would you still mind hanging out with me in public?" The homophonous sentence where the if-clause is a complement has a non-conditional reading and means something like "Do you have anything against my smoking and dying my hair green?" – Araucaria - Him Nov 16 '22 at 17:53
  • 1
    Thank you. Those two readings are almost certain to be intoned differently, though. Particularly if the guy has any brains. – John Lawler Nov 16 '22 at 17:53
  • 1
    @JohnLawler I still don't get the "everything in a discourse counts as a conditional protasis, if someone can infer something from it" comment! – Araucaria - Him Nov 16 '22 at 18:10
  • @Araucaria-Nothereanymore. That makes two of us. – JK2 Nov 18 '22 at 21:49
  • @JK2 Well, it might be because I didn't explain things very well in some writing I sent to John :) – Araucaria - Him Nov 19 '22 at 00:43
  • @Araucaria-Nothereanymore. Whatever the reason, I think it's odd that he should not respond to your comment in any way even though this is the second time you've asked the same question, while he did respond to my comments on another post twice already clearly after your last comment to him a couple days ago. – JK2 Nov 19 '22 at 03:10
  • 1
    I don't keep score. And I rarely look at who made which comment. I don't know anybody here personally, so it's all just text to me. I'm dealing here with the idea (which is novel to me) that "conditional" is a syntactic construction with syntactic properties (instead of a purely logical and pragmatic term, with no special syntax except certain governing words, like if). Use of will with if-clauses is a fact about will (and if), not necessarily predicted from any theory of modals or logic. – John Lawler Nov 19 '22 at 17:10
  • 1
    @JohnLawler If we're talking about the conditional construction, then all conditional adjuncts (which don't already require a different modal) have the very same restrictions on will, not just if-ones: "whether or not/ however/ as long as/ unless/ provided that/ supposing/assuming Bob wins the lottery, xyz." – Araucaria - Him Nov 20 '22 at 22:58
  • 1
    @JohnLawler Note that this applies to conditional constructions which don't have a meaning related to conditionality (e.g. so-called 'relevance' conditionals), and doesn't necessarily apply to other constructions which do. So, for example, neither clause in a sentence containing an otherwise adjunct is subject to these restrictions on will: "He will sign the declaration, otherwise they'll reject his application." Such sentences are not conditionals, because what might be thought of as the semantic equivalent a consequent/apodosis/result appears in the subordinate clause, not the main one. – Araucaria - Him Nov 20 '22 at 23:26
  • I see I am way out of date on "conditionals". I use scare quotes because I still don't see conditional as a well-defined construction, or construction type, with syntactic constraints and conditions, and a short list of lexical items (mostly verbs) controlling it. I'm still looking for basic syntax here. I can see they're logically some modal variety of P ⊃ Q, but that's not syntax and has no syntactic outcome. – John Lawler Nov 21 '22 at 13:45
  • @Araucaria-Nothereanymore. You've lost me in your last sentence: "Such sentences are not conditionals, because what might be thought of as the semantic equivalent [of?] a consequent/apodosis/result appears in the subordinate clause, not the main one." Are you suggesting that they'll reject his application is a subordinate clause in your example? – JK2 Nov 22 '22 at 21:22
  • @JohnLawler Could you please show us a few examples of "some modal variety of P ⊃ Q" that doesn't have a typical conditional construction that resists epistemic will (but not deontic will) in P? Because I can't think of any. – JK2 Nov 22 '22 at 21:27
  • @JK2 Hastily written. I was analysing otherwise as a prep there. Oops. But in any case, it's not a conditional because it doesn't have a conditional structure. – Araucaria - Him Nov 22 '22 at 22:37
  • Fergodsake, what's a "conditional structure"? All I see is references to it, rather like Jesus. Where are the definitions? Where are the restrictions? Where are the governing predicates? Where's the "structure"? We're all talking about meaning, not grammar. – John Lawler Nov 23 '22 at 16:33
  • @JK2: Does your request imply that you won't encounter any in the future? Oh, and btw, is because a conditional word? – John Lawler Nov 23 '22 at 16:35
  • @JohnLawler You mean in some future English that's different from Present-day English? Then, does your first question imply that you can't produce a single example in PDE of "some modal variety of P ⊃ Q" that doesn't have a typical conditional construction? // AFAIK, because cannot introduce a conditional clause. I can't even imagine why you'd even ask if because is "a conditional word". – JK2 Nov 24 '22 at 01:56
  • @JK2 You asked for an example of "some modal variety of P ⊃ Q" and I provided one. Or didn't you recognize it? And I'm sorry about your lack of imagination. – John Lawler Nov 24 '22 at 16:34
  • @JohnLawler Oh, you mean the example is this?: Does your request imply that you won't encounter any in the future? Please let me know if this is your example, because, as you notice, I do lack imagination. – JK2 Nov 25 '22 at 01:57
1

'I don't mind what ...' can have two meanings, and these are informed by the constructions used.

'I don't mind what you are doing' usually means 'I am not troubled by the things I know you are doing'.

'I don't mind what you do tomorrow' means 'Whatever you choose to do tomorrow is OK by me'.

.......

'I don't mind what you will do tomorrow' is unidiomatic and thus has no default sense.

........................

'I don't know what she does' is idiomatic for 'I don't know what her usual practice in these situations is / I don't know how she earns a living'.

'I don't know what she does tomorrow' must have the sense 'I don't know what the timetabled practice for tomorrow is'.

'I don't know what she will do tomorrow' is transparent.

Edwin Ashworth
  • 83,272
1

The question posed is about 'the acceptability of ‘will’ in different environments'.

In one of these environments I believe that language comes up against what is currently termed ‘cognitive dissonance’ - a conflict not of logic, as such, but a conflict between personal value and personal action. It is the conflict of being obliged to do something one does not, personally, feel one should be doing. (Like uttering something that it is just not possible to utter.)

This arises, in this case, when someone is obliged to picture themselves in the far future, looking back towards the near future, in order to report a future happening as though it was a completed event.

I believe this is why, in English, we have a situation which some would describe as “possessing no future tense”. The reason we do not have what they term “a future tense” is because no such thing can exist. And many languages, I am informed, are thusly. My suspicion is, they all are.

In order to confidently make predictions about the future one has to project oneself into the far future, then look back at the near future (as though it were a past event) and report the occurrence of that event as though it had been completed, had been documented, and could be looked upon as past history.

Some have not understood what is involved in making a future prediction and their misunderstanding created the controversy of the Waw Conversive, which is ably dealt with by Robert Young in the preface to his Literal Bible. He blasts it to smithereens.

‘I don’t know what she does tomorrow’ answers the question ‘What does she do tomorrow ?’

‘Does’ is a continuous matter. ‘Does she like ice cream ?’ ‘Does she visit the theatre ?’ ‘Does she have brown eyes ?’ It reports either a continuous state or a repeated activity.

The possible answers to ‘What does she do tomorrow ?’ are many and various :

‘I haven’t a clue; let me hop in my time machine and I’ll be back five minutes ago and one of us will tell you’ is one of the least exasperated outbursts which could result.

I would have to predict, in the future, what her state or her repeated actions will be and then I have to report them with the same degree of certainty that I report what I had for breakfast three hours in the past. Her eyes could be bloodshot, no longer brown. She may now hate both ice cream and theatre.

So my answer, properly punctuated, is ‘I don’t know what she ‘does’ tomorrow,’ (!)

How can I know ? How can I answer your question ? So ‘does’ is in quotes, because you said it, not me. And ‘know’ is in italics because I really do want to get across to you that I don’t know !

I don’t mind what she does tomorrow - my mind is completely detached from the future event and I am comfortable with that. No problem. She can do what she likes; polish the time machine, maybe. (Spoiler alert : we need it again.)

I don’t know what she does tomorrow - haven’t a clue, so I need not engage my sensitive cognitive faculties and they, thus, will not become dissonant. I am totally disengaged from tomorrow. It can be worded 'I don't know what she will do tomorrow' and it doesn't matter. I still do not have to predict anything.

I don’t mind what she will do tomorrow. ‘Will’ is her will, not mine. She will do fifty somersaults tomorrow (but only if she wills to do it). If somebody else wills to shoot her between the eyes, she won’t be doing a thing. So the blue button gets you into the future and the green one gets you back again (don’t touch the red one). Once you’re back and can tell me what she (and everybody else in the world) will be willing to do tomorrow - at that point I’ll tell you whether I mind or not.

We are creatures of the present (with memories). This is all we know. It is all we can know. It is all that language will let us report. And if we try to bend language, we become dissonant

Nigel J
  • 24,448
1

The following sentence is unidiomatic in English:

I don't mind what she will do tomorrow. NO

But the sentence can have a future meaning if we add the auxiliary "will" before the verb mind

I won't mind what she does tomorrow.

Today, the more common response would be

I don't mind what she's doing tomorrow.

Cambridge Grammar contains a note on the usage of mind

Warning:
When we refer to the future, we use present (not future) verb forms after mind:

I don’t mind what day they come and stay as long as it’s not Tuesday 12th because I’m away.

Not: … what day they will come and stay …

I suspect that do not mind, is a relatively recent development stemming from its more polite and formal equivalent would not mind, which is used to make polite (and sometimes hesitant) requests

A: Would you mind if I open the window?
B: No, but I should mind if you were to leave it open all night.

From Fairy Birds from Fancy Islet, or the Children in the Forest, dated 1846, the phrase I should mind has a future meaning. Today, the “I don't think” would probably be omitted, and “I should mind in the least” would be replaced with I don't mind in the slightest

“I do not think,” said Alice “I should mind in the least having my wings nailed on.” “Should you not?” said Jim, (half ashamed of his cowardice;) “ yes, but you would, though.” “No.” said Alice. “I am sure I shall not: I will have the wings nailed on directly, if the grey bird think it the best way.” […]

“Why,” said Jim, “you know very well, that your saying you would not mind having the wings nailed on your shoulders, must make me appear very cowardly, so if you did not feel afraid, you should not have said so, for my sake.”

For the "don't know" construction, the present progressive tense, which is also used to express future meaning, sounds far more natural and logical than “I don't mind/know what she does tomorrow”

I don't know what she's doing tomorrow
I don't mind what she's doing tomorrow

Mari-Lou A
  • 91,183