toward(s) and other ward(s)

The interview I did with the Chicago Manual of Style people has brought me quite a few new readers. (Not to mention a 'Hey! I saw you in this newsletter I subscribe to!" during [BrE] the school run. Next thing you know, it'll be the paparazzi.) One of these new readers is Linda, a Washington, DC editor, who wrote to ask if I'd covered toward and towards. And since I've been rather embarrassed for some time that I haven't covered this, Linda's request has gone to the front of the (AmE) line/(BrE) queue.

The first thing to say about toward and towards is that both are found in both Englishes. What is different is which one is more common and standard in each place. In the US, toward is more common, particularly in published work; in the UK, towards is. This is shown in the ratios of the two variants in each dialect. The Corpus of Contemporary American English has about 6 toward for every 1 towards. But the British National Corpus favo(u)rs towards 23:1.

Towards is one of the things that I resisted for a long time after moving to the UK--because of the associations I had for it in AmE. My first teaching job was teaching remedial (AmE) freshman composition in Illinois, and that was where I first reali{z/s}ed that I was a toward-sayer but that there were a lot of American towards-sayers. And I took it upon myself beat the 's' out of these people. I perceived the 's' as something that marked people as unsophisticated hicks. Most advice you can find on the internet these days will tell you that it's fine to use either. I was a young east-coaster in the midwest. Mea maxima culpa.

So, when I came to the UK and was surrounded by those esses, I just had to grit my teeth, much as I've learn{ed/t} to do with the BrE use of reckon (which says 'HICK' in capital letters to my northeastern US self) and whilst (which says 'PRETENTIOUS' to my US self). Live and let live, speak and let speak, as we're taught in Linguist School. [If you want to talk about those two, please use the comments sections at their linked posts.] These days, if I'm writing for a British publication or if I'm proofreading for a British writer, I do use towards.

The reason I've not done toward and towards in seven years of blogging is that I knew it'd bring up all the other -ward(s) words--and that means work, because they're not as straightforward. Toward(s) is almost always a preposition. Something like backward(s) can be an adverb or an adjective. In my dialect, I'd allow the 's' much more easily for an adverb than for an adjective and I'd allow the 's' more for the figurative use of the adjective than the literal. You may have different instincts about these:

• Adjective (literal):  a backward(s) motion
• Adjective (figurative): a backward(s) idea
• Adverb:  You've got that on backward(s)

I am not going to do an in-depth analysis of all of these. Picking out figurative and non-figurative meaning would be just too labo(u)r-intensive. So, at this point, I'm just going to look at adverbs (since they're more like the preposition toward(s) anyhow). I'm using the Global Web-Based English corpus for this because I suspect that there's a high risk for mislabel(l)ing (or 'mis-tagging', in the corpus linguistics parlance) the parts-of-speech of these particular words. By using GlobWE, I at least know that the same 'tagger' did the tagging, so any mistakes should be comparable. In the table, the percentages are within-dialect. So the AmE numbers add up to 100% in each row and so do the BrE ones.

Adverbs	AmE ward	AmE wards	BrE ward	BrE wards
back-	23%	77%	13%	87%
down-	67%	33%	17%	83%
for-	98%	2%	94%	6%
in-	78%	22%	31%	69%
on-	59%	41%	20%	80%
out-	78%	22%	37%	63%
up-	40%	60%	13%	87%

So we can see here that:

• Both dialects prefer backwards and (especially strongly) forward.
• With the exception of forward, BrE prefers -wards, in keeping with its preference for towards.
• With the exception of backwards and upwards, AmE tends to prefer the 's'-less version, in keeping with its preference for toward. 
• AmE's preference for onward over onwards doesn't seem very strong, though.

Showing you the percentages made the numbers clearer, but it hides some interesting things. For instance, Americans use onward(s) (1868 examples, counting both variants) a lot less than the British (5233 examples). Why? A quick glance at the examples shows that many of the UK examples were things like

from 1833 onwards
version 1.5.2 onwards
from primary school onwards

AmE would tend to use on as an adverb in such cases, rather than the -ward(s) form.  So, for example GlobWE has 11 examples of from 2008 onwards and 5 of from 2008 on in BrE. Those numbers are reversed in the AmE portion of the corpus.

The other thing that interests me in those numbers relates to my day job, in which I study antonymy (opposite relations). Why do forward in AmE upwards have different endings from their opposites? I can't come up with any semantic explanation. I'll just have to conclude with something I've been heard to say elsewhere (and may be heard to say again in Ashford and Ealing in September):
If you're looking for logic in vocabulary, then you're looking in the wrong place.

In other news: My second (and last for the time being) contribution to the Numberphile video series is now available--on differences in how numbers are said and used in AmE and BrE. If you're interested in more on that subject, here's the link to my other 'numbers' posts.