It's now colloquial, if that fits your definition of "incorrect". Reasonably old, there are OED citations back to 1686*.
It fits with a larger pattern of using "and" to (as the OED again says) "[connect] two verbs, the second of which is logically dependent on the first". Like "go and see", "send and let her know", "be sure and tell us", ....
* Actually that's just in the "try" entry. In "and" there's one from 1599. And a 1671 Milton quote.
I'm a native speaker who uses "try and [verb]" extremely frequently. Never really thought about it before, which shows, I guess, that it's grammatical, at least in my internal grammar. I've never seen any prescriptivist rants about it, though I wouldn't be surprised if they existed.
I recognize and use "try to" too, but it feels different by a shade or two. I can't really articulate how. spamsink's idea might be the right one, but your example (which doesn't carry that implicature and even without the "I can't make any guarantees" part still sounds kind of doubtful to me) is fine, too. I dunno.
I figure it's vaguely like the resultative complement construction in Mandarin. Very colloquial, natch, but completely intelligible and completely acceptable.
Edit: I should say, I don't think that the resultative complement construction in Mandarin is overly colloquial, I believe it's more widely used than that. The two sentences above are not perfectly connected.
I've always rather disliked "try and [verb]," because the fact that the conjunction puts the two verbs in parallel seems to override the inherent uncertainty created by the first. I.e., if you tell someone to "go and see" something (to use nekokaze's example), you are directing him to do two things which are theoretically equally in his power. On the other hand, to tell him to "try and see" something has the sense (to me, at least) of both acknowledging the possibility of failure and almost willing him to overcome it.
That said, there are many colloquial-English-related pet peeves that are nearer and dearer to my heart, so I don't tend to make a big deal of this one. I do rarely, if ever, say "try and [verb]" myself, though.
Comments 15
Reply
It fits with a larger pattern of using "and" to (as the OED again says) "[connect] two verbs, the second of which is logically dependent on the first". Like "go and see", "send and let her know", "be sure and tell us", ....
* Actually that's just in the "try" entry. In "and" there's one from 1599. And a 1671 Milton quote.
Reply
Reply
Reply
Reply
I recognize and use "try to" too, but it feels different by a shade or two. I can't really articulate how. spamsink's idea might be the right one, but your example (which doesn't carry that implicature and even without the "I can't make any guarantees" part still sounds kind of doubtful to me) is fine, too. I dunno.
Reply
I have a sense that it's "not proper language for written use at work" but unfortunately no hard details.
Reply
Edit: I should say, I don't think that the resultative complement construction in Mandarin is overly colloquial, I believe it's more widely used than that. The two sentences above are not perfectly connected.
Reply
That said, there are many colloquial-English-related pet peeves that are nearer and dearer to my heart, so I don't tend to make a big deal of this one. I do rarely, if ever, say "try and [verb]" myself, though.
(AmE native speaker)
Reply
Leave a comment