Reader Tim wrote recently to ask:
Before we get to the methodological issues, I'll answer his question. No, it's not a well-recogni{s/z}ed difference between BrE and AmE. Every source I've checked lists the two phrases as variations on each other, without noting anything about dialect, and sources on BrE/AmE differences (like Algeo's British or American English?--my usual [chiefly AmE] go-to book for verb variations like this) don't mention it either. But I share his intuition that are nothing to do with sounds "less American" than have nothing to do with. Note that here we're only talking about the use of these phrases when they're describing states and not when they're describing intentional behavio(u)r. So, if I want to proclaim that something has no impact on me, I could say (if I felt equally comfortable with either phrasing): It is nothing to do with me or It has nothing to do with me (or it's nothing to do with me, which hides whether it's a be or a have--but see this post for BrE/AmE differences in contractions). But if we're talking about shunning someone, it has to be have and not be: She'll have nothing to do with me since I insulted her pig.
Back to the methodology: Tim has made do with the limitations of the internet in studying US versus UK: comparing the UK to the whole world and assuming that anything peculiarly BrE will stand out in the .uk sites. It's hard to get around that, since few US URLs end in a reliable country code. What one can do instead is to search something like .edu versus .ac.uk or .gov versus .gov.uk, or search a couple of reasonably similar newspaper sites, but Tim's .uk versus 'the whole world' analysis is telling...except for one little problem.
The problem is comparing are and have. Both are the plural forms of their respective verbs, but have is also the base form. So, many of the have examples have their equivalent in be nothing to do with rather than are nothing to do with (as in it could be/have nothing to do with). So, let's try re-running the searches with the third-person singular forms is and has, which won't have this dual purpose (remembering that some of the has examples will have the 'shunning' meaning):
So, the difference here is not as dramatic as Tim's figures would suggest, but what hasn't changed is that he's right in his intuition that be nothing to do with is more common on .uk sites than in the rest of the world. But much of the rest of the world is speaking Britishoid Englishes. To try to get out that influence, let's try it once more with .edu versus .ac.uk:
Now there's a big difference! (Not quite as big as Tim's original difference, but big enough.) I suspect that the 'shun' meaning is less common on academic sites than on general sites, which include lots of places where people can be relatively free about declaring whom they are shunning--so I believe this might be a truer reflection of the relative Britishness of be nothing to do with.
So, well done, Tim! (Despite my methodological quibbles, which we in the educational establishment call a "teaching opportunity". At least that's what we call it when we want to show off a bit.)
Read more
Now, I love queries that come with the homework already done. Farrrr superior to the ones with no greeting, no signature, no please and no thank you that demand that I explain why their spouse says such-and-such or why other people are idiots who say this-and-that. (I'd write "you know who you are", but such lack of self-consciousness in making demands of complete strangers is evidence of a lack of, well, self-consciousness.) And Tim did a good job of comparing the two phrases...except for one little problem. Can you spot it?I noticed the phrase "are nothing to do with" in David Crystal's The Fight for English (p168) and it seemed peculiarly British to me. I would have used "have nothing to do with" instead. Google backs me up, saying its relative frequency is 10 times higher on UK websites than on the Internet in general:
"are nothing to do with"= 115,000 / "have nothing to do with" = 8,700,000 = 1:76
"are nothing to do with" site:.uk = 21,000/ "have nothing to do with" site:.uk = 156,000 = 1:7Is it a well-recognized regionalism?
Before we get to the methodological issues, I'll answer his question. No, it's not a well-recogni{s/z}ed difference between BrE and AmE. Every source I've checked lists the two phrases as variations on each other, without noting anything about dialect, and sources on BrE/AmE differences (like Algeo's British or American English?--my usual [chiefly AmE] go-to book for verb variations like this) don't mention it either. But I share his intuition that are nothing to do with sounds "less American" than have nothing to do with. Note that here we're only talking about the use of these phrases when they're describing states and not when they're describing intentional behavio(u)r. So, if I want to proclaim that something has no impact on me, I could say (if I felt equally comfortable with either phrasing): It is nothing to do with me or It has nothing to do with me (or it's nothing to do with me, which hides whether it's a be or a have--but see this post for BrE/AmE differences in contractions). But if we're talking about shunning someone, it has to be have and not be: She'll have nothing to do with me since I insulted her pig.
Back to the methodology: Tim has made do with the limitations of the internet in studying US versus UK: comparing the UK to the whole world and assuming that anything peculiarly BrE will stand out in the .uk sites. It's hard to get around that, since few US URLs end in a reliable country code. What one can do instead is to search something like .edu versus .ac.uk or .gov versus .gov.uk, or search a couple of reasonably similar newspaper sites, but Tim's .uk versus 'the whole world' analysis is telling...except for one little problem.
The problem is comparing are and have. Both are the plural forms of their respective verbs, but have is also the base form. So, many of the have examples have their equivalent in be nothing to do with rather than are nothing to do with (as in it could be/have nothing to do with). So, let's try re-running the searches with the third-person singular forms is and has, which won't have this dual purpose (remembering that some of the has examples will have the 'shunning' meaning):
all sites | .uk sites | ||
is nothing to do with | 1,440,000 | 85,900 | |
has nothing to do with | 13,400,000 | 333,000 | |
ratio | 1:9 | 1:4 |
So, the difference here is not as dramatic as Tim's figures would suggest, but what hasn't changed is that he's right in his intuition that be nothing to do with is more common on .uk sites than in the rest of the world. But much of the rest of the world is speaking Britishoid Englishes. To try to get out that influence, let's try it once more with .edu versus .ac.uk:
.edu sites | .ac.uk sites | ||
is nothing to do with | 4340 | 1140 | |
has nothing to do with | 232,000 | 2030 | |
ratio | 1:53 | 1:2 |
Now there's a big difference! (Not quite as big as Tim's original difference, but big enough.) I suspect that the 'shun' meaning is less common on academic sites than on general sites, which include lots of places where people can be relatively free about declaring whom they are shunning--so I believe this might be a truer reflection of the relative Britishness of be nothing to do with.
So, well done, Tim! (Despite my methodological quibbles, which we in the educational establishment call a "teaching opportunity". At least that's what we call it when we want to show off a bit.)