negative force
<Similarly, I perceive forms like "have to", "need to", "tend to", "be going to", "be supposed to", "want to", "use to", and so on to act as fixed verbal forms,>
Would you add the relatively new use of "set to" to those?
"Present-day English seems to be developing a new future auxiliary. It is very common, especially in journalistic writing, to read that something is 'set to' happen: interest rates are set to rise, pub opening hours are set to change. Not long ago, this was a metaphor (referring to a runner in the 'set' position just before the starting pistol is fired), used only for people who were ready to do something. Now it is losing its original meaning and becoming grammaticalized as an auxiliary, used not only for people but also for things and processes."
M Swan
<Damn, now that I think about this, this is actually a very significant syntactic difference between North American English dialects; the fact that your dialect places "not" before the "to" indicates that "to" is still treated as part of a "full" infinitive, whereas my very strong preference for placing "to" immediately after "tend" in all cases implies that forms like "tend to" have become truly lexicalized here, and that they really take "bare" infinitives, like classical English modal and auxiliary verbs. >
I agree. In your version "to" has a purposive function, it has lost its infinitive function.
Isn't it all about the degrammaticalisation of "to" in AmEng?
Where are you from, for the record? (I'm just wondering for the sake of figuring out the distribution of "tend to not" versus "tend not to".)
Olga Fischer, Anette Rosenbach, Dieter Stein, in Pathways of Change: Grammaticalization in English, all notice a growth in AmEng of infinitives split by negatives.
<At least here, such is extremely common, moreso than preceding negation, and furthermore is semantically very distinct from preceding negation.>
Not sure if the semantics are distinct. The pragmatics may be, but the semantics...?
To me, these entail the same:
She decided not to tell him.
(>>she did not tell him)
She decided to not tell him.
(>>she did not tell him)
Which one indicates why she was sensible?
She was sensible not to rise to his baiting.
She was sensible to not rise to his baiting.
So, all in all, forms such as tend to, be going to, set to, have to, about to, all show that "to" is becoming more and more bound to the verb, right?
>>Not sure if the semantics are distinct. The pragmatics may be, but the semantics...?
To me, these entail the same:
She decided not to tell him.
(>>she did not tell him)
She decided to not tell him.
(>>she did not tell him)<<
No, I meant like this:
"She decided (to not / not to) tell him"
versus
"She didn't decide to tell him."
Your examples are practically identical, and which one to use would probably be more a matter of dialect, idiolect, and register more than anything else, but the ones I have above here are quite distinct semantically...
>>So, all in all, forms such as tend to, be going to, set to, have to, about to, all show that "to" is becoming more and more bound to the verb, right?<<
You mean bound to the *finite* verb, I presume. And yes, they do show that.
<Your examples are practically identical, and which one to use would probably be more a matter of dialect, idiolect, and register more than anything else, >
And, for me, often a matter of negative force - the "to not" forms having more force.
>>And, for me, often a matter of negative force - the "to not" forms having more force.<<
Tis true - the examples with "not to" almost make it not clear just what is being negated (as it still echos Early New English negation and New English auxiliary/modal negation), whereas "to not" makes it very clear that it is the infinitive being negated.
<You mean bound to the *finite* verb, I presume. And yes, they do show that. >
That it.
Also, in 1998, Biber noted that even though the "to not" form appeared much more in American conversation than in British conversation, the "to not" form was still much more frequent than the "to not" form in American Eng ("not to" 80 in 1 million words: "to not" 10 in 1 million words).
So its not surprising that some AmEng users here find the "to not" form still a bit odd.
That must be a matter of differences between NAE dialects, as at least the dialect here amongst younger people has practically completely shifted to using "to not" rather than "not to" (I am not so sure about middle-aged and older people here, though, particularly because their speech can often be very conservative in some respects).
<Tis true - the examples with "not to" almost make it not clear just what is being negated (as it still echos Early New English negation and New English auxiliary/modal negation), whereas "to not" makes it very clear that it is the infinitive being negated. >
Still, which would we prefer here?
To be or not to be? That is the question.
To be or to not be? That is the question.
;-)
BTW, Travis, in you dialect, do you also combine two negatives?
e.g.
I can't afford to not do math.