Monday, 20 September 2010

On highest mountains

A correspondent writes to ask about the difference between tallest and highest in such sentences as Everest and K2 are the two tallest mountains in the world, which he has found on a BBC site. He quotes Michael Swan as an example of a grammarian who says that it has to be Mont Blanc is the highest mountain in Europe and not . . . the tallest mountain, and is understandably confused.

Usage is undoubtedly blurred, because we frequently see such listings (very common online) as 'The Tallest Buildings in the World', 'The Tallest Mountains in the World', and so on, alongside 'The Highest...' Tallest is four times more common than highest for buildings on Google. Highest is over ten times more common than tallest for mountains, but tallest still attracts a healthy 60,000 hits.

When we look for the reason, we enter a world of technical definition. Mountains are actually measured in three different ways. (1) Sea level to the peak. (2) Base to the peak. (3) Distance from the centre of the earth to the peak. The differences are significant. Under (1), the title goes to Mt Everest. Under (2) the title goes to Mauna Kea, Hawaii, which has only 4,245 m of its 30,000 m above sea level. Under (3) the title goes to Chimborazo volcano in the Andes, only 6310 m high but (thanks to the equatorial bulge caused by the earth's spin) further away from the centre than anything else. In this approach, (1) is referred to technically as the highest mountain, and (2) as the tallest mountain.

Most people would never need to refer to (3), and few to (2). We normally think of mountains as being above sea level, so this motivates the use of highest. But the fact that tallest is also a permitted collocation will have reinforced its appearance in everyday usage, resulting in the overlap noted by my correspondent.

Other domains raise similar issues. People normally talk about tallest buildings, referring to the dimension of the building from base to roof. The highest building in the world could be a single-storey house on some mountain-top somewhere. But again there is an overlap, and there are complications, well known in the field of encylopedia listings. Does the radio mast on top of a high building count as part of the height of the building or not? It can get quite controversial when competing claims are being made for the world record. And the complications reverberate linguistically when we encounter such usages as The highest part of the tallest building is well above roof level.

Thursday, 9 September 2010

On gonna

A correspondent writes to say she had read an interview in Rolling Stone with the cast of AMC’s Mad Men, where one of the actors said they had to be very careful with their pronunciation. 'There was no ‘gonna’ or ‘shoulda’ back then [in the 1960's].' Could this be true?

It certainly couldn't. It's easy enough to check the point. These two forms actually get separate entries in the OED, and there we find gonna with a first recorded use in 1913 and shoulda with a first recorded use in 1933. (Also woulda, 1913, coulda 1925.)

Both usages are undoubtedly much older. Joseph Wright's English Dialect Dictionary has a separate section under go giving examples spelled ganna, gauna, gaunna, and ginnie. The earliest is 1806. And under have there's an example of should ha' from 1899.

Why stop there? In the 1602 Quarto edition of Merry Wives of Windsor we find Nym saying 'I should ha borne the humor Letter to her', and there are several similar examples in the literature of the period.

When it comes to linguistic mythology, it's seems it's a mad Mad world.

Sunday, 5 September 2010

On updating and the OED

Several media correspondents have been in touch this week to discuss the report that the OED may not have a further print edition. Whether it will or not I can't say, but I do know that I've not looked at my print edition for years, and use the online edition pretty well every day. It includes an amazing amount of new lexical information. There are updates of various sections of the alphabet at regular intervals, as the OED team slogs on. And one of the unspoken messages to scholars that comes across from the updating is: revise.

Like many people, I've used the OED repeatedly for information about the earliest use of words. The first recorded date for a word is of special interest. One knows that such dates are artificial, because they're only as reliable as the sources that have so far been examined, but they're still the best information we have, and they point in the right direction. So I always look with extra interest when I find the OED has discovered earlier citations for words. It's especially important when people are discussing such matters as the originality of Shakespeare's vocabulary or the number of new words in an early edition of the Bible.

It's not just earlier citations that can change the totals. A different dating chronology can wreak havoc with statistics. For example, the dates of Shakespeare's plays used by the original lexicographers are now hugely out of date. Nobody these days would place Love's Labour's Lost in 1588, as the OED does; most people would opt for 1593-5. Similarly, Titus Andronicus is given as 1588 (probably 1590-91) and A Midsummer Night's Dream as 1590 (probably 1594-5). The day they revise these dates, the whole of the 'Shakespeare invented words' industry will have to be reviewed, as in many entries the Shakespearean usage will leapfrog over another citation into second place.

The totals already need serious revision, in the light of the earlier citations that have emerged. I spent a lot of space in my Stories of English reviewing the evidence for Shakespeare's invented words, based on the OED entries, and for years now I've written a regular piece for Around the Globe (the magazine of Shakespeare's Globe) on what we can deduce from such 'Williamisms', as I call them there. We now know that several of the first recorded usages assigned to Shakespeare have been antedated. Not all are in the OED files yet. Lonely isn't, for example. The OED still gives Coriolanus 1607 as a first use, but, as I point out in Think on My Words, Mary Sidney, the Countess of Pembroke, talks about ‘lonely ghosts’ in her Tragedie of Antonie, and that is 1592.

I don't know when will be the best time to do a complete revision of the Shakespeare assignments. It's going to be a long job, and best done, perhaps, when a bit more OED revision has taken place. But I have taken a look at an easier topic: the number of first recorded uses in the King James Bible. In Stories of English (pp. 328-9) I say there are 55, and give a list. I've now gone through that list and checked against the latest OED entries, and the total now stands at 47. There are some additions and some deletions. For the record, here is the current list. Three of of the items (marked with *) also appear in the translations to entries in Randall Cotgrave's 1611 Dictionarie of the French and English tongues:

abased (as an adjective), accurately, afflicting (as a noun), almug ('algum tree'), anywhither, armour-bearer, backsliding (as an adjective), battering-ram, Benjamite, catholicon, confessing (as a noun), crowning (as an adjective), dissolver, dogmatize, epitomist, escaper, espoused (as an adjective), exactress, expansion, Galilean (as a noun), gopher, Gothic (as an adjective), grand-daughter, Hamathite, infallibility*, Laodicean (as a noun), lapful*, light-minded, maneh (Hebrew unit of account), miscarrying (as an adjective), Naziriteship, needleworker, night-hawk, nose-jewel, palmchrist, panary ('pantry'), phrasing (as a noun), pruning-hook, rosebud, rose of Sharon, Sauromatian, shittah (type of tree), skewed, taloned* (as an adjective), way-mark ('traveller guide'), whosesoever, withdrawing (as an adjective)

I say 'for the record'. This is one of the less-publicized benefits of a blog. There's no way I'd be able to draw the attention of Stories of English readers to the update otherwise. There may never be a second edition of the book, and Penguin isn't going to publish an updated edition just because some new facts have emerged requiring revision on pp. 328-9. So, all hail to blogs, as an updating procedure.