A reader recently took issue with my long-standing recommendation to avoid the use of the “-st” variants of amid, among, and while (original 2013 post here). Our exchange about it was growing unmanageably long for the comments section, so I’ve gone and turned it into a full post for today.
The reader’s actual question was, essentially:
“is it more appropriate to talk about being amidst people or amongst people?”
My response: neither. You should avoid using both amidst and amongst and stick to amid or among. If you want to phrase it the way most people do, choose among (although I don’t have enough context, so perhaps amid works better). You could also find a different word entirely: With people? Surrounded by people? Hanging out with people? Again, without knowing the precise context, I can’t know if these are better or worse solutions.
In the past, I’ve pointed out that good style authorities largely take one of two paths with the “-st” forms of these words: they either ignore them as if they don’t exist (not a good sign) or they specifically suggest that you avoid using them (an even worse sign).
I’ve noted that, to me personally, a writer or speaker can sometimes get away with amidst or amongst in the right situations, but usually only for specific humorous or ironic effect. Whilst is always a joke word. (This is in the US and these are my opinions: I don’t mean to insult any speakers of Commonwealth English, which is another language entirely, who may treat these words as they wish — with the awareness that at least one UK newspaper style guide bans the “-st” versions, and even the OED says ‘probably not.‘). While we’re at it, unbeknownst is an equally comical word, which should also be avoided.
It’s worth stressing again, as I did to the reader, that on this blog I strive to discuss CURRENT BEST PRACTICE. My role is to explain what that is, why it is so, and provide support; but I don’t dictate (…despite the reader’s comparisons of me to Hitler). If you want to use amidst or amongst, that’s your choice. I will, however, continue to attempt to convince you, for your own benefit, that this is a mistake. Some in your audience won’t care which you use, and a smattering might approve of the “-st” endings. However, others will wonder which of these four conditions most likely apply: you are adopting a pretentious affectation, you are ignorant, you have a speech impediment, or you just got off a flight from Manchester (the one in the UK, not NH). As I have stressed in many previous posts with other dicey language issues: Why take this risk?
Why take the chance of creating a problem for yourself when there is no need? Mark Twain supposedly said that it is better to remain silent and appear ignorant than to open one’s mouth and remove all doubt. When it comes to these “-st” endings, it’s not even that stark a choice: if you don’t use them, you cause no trouble; if you do, you risk a negative reaction.
My position is based on the best available sources. To say that by referencing Chicago, Garner, and AP Style (as the commenter did) that I was seeking “facts that only serve [my] opinions” is…well, preposterous. I not only referenced contemporary style guides, but I also used data culled from several million printed books; I personally analyzed several manuscripts in electronic form; and I sought out the in-house style guides of several current publications. This amounted to at least a full day of work when I wrote the original post, plus several more hours for this response, as opposed to the commenter’s investment of…just enough time to write his or her comments? The sources I used were ridiculed and the research was dismissed based purely on…the commenter’s opinion? That’s really no way to handle an argument. I’m reminded of how a former co-worker used to silence those who chattered ignorantly around him by pointing out that “you’re not entitled to an opinion: you’re entitled to an informed opinion.”
The commenter promised to provide some kind of support for his argument, but this didn’t materialize. It’s not my job to find data to strengthen a debate opponent’s position (nor could I find any), but I take my own work seriously and while awaiting his response I uncovered some interesting additional information about actual everyday use of these words. Here are some — further — things to consider.
While Google Ngrams only uses books, and only those published through 2008, there is a data set of actual contemporary writing on the Internet: fivethirtyeight’s Reddit comment n-gram data. It contains over 1.7 billion comments from over 200 million users, posted between late 2007 and August 2015. It’s not the best sample, as it leans demographically to men (80%), under 35, and it’s something of a self-selected snapshot. But for our purposes, let’s assume it’s up to snuff. If the commenter wants to challenge it, he’ll have to provide an equivalent or better data set.
Check out these results. (The amid/amidst data look a little weird — particularly those squared-off early spikes — when compared to most other Reddit n-grams, so I’m not sure how clean that is. But let’s accept it, since it agrees with a second source, further below.)
[Note: If you can’t see each of these graphs, go to the root link and enter the word pairs yourself.]
- among is used close to 4 times more often than amongst. Among declined through 2012, but has been steady since; use of amongst has not noticeably changed.
- while is roughly 40 times more frequent than whilst. Use of while increased through 2011, and has held steady; whilst has never had much traction (in this data set).
- ah…here’s something interesting: amid and amidst buck the trend. Both were virtually unused early in the sample (with a weird spike around February 2008), but something happened around December, 2008, to bring both into the reddit comment fold. Amid spiked and took an early lead, then both vanished for a couple months. They reappeared in July, 2009, but neither seemed preferred, until around December, when amidst took the lead — and it hasn’t let go. In fact, it’s hardly been threatened in the last five years of the sample: June, 2010, was the last time amid came close.
But this is critically important: “close” in this case, is relative. Amidst wins the head-to-head fight, but we’re not talking while/whilst levels of dominance: it’s more like 5 uses of amidst for every 3 of amid. We’re also not looking at frequently used words, either: for every use of “amidst,” “while” was used around 800 times (and “whilst” about 20 times); for every “amidst” someone typed out, someone else typed “among” nearly 60 times (and “amongst” around 15).
The unpopularity of both amid and amidst was a surprise. I did a quick double-check in Google Ngrams to see if this was a bad data issue: I’d never put all these words on a single graph before. Google concurs: while and among are way ahead, amongst and whilst are well behind (indistinguishable in the data at this scale); amid and amidst are used so infrequently that they barely register.
So there’s another piece of advice for the original questioner: use among over amid. In actual use, it’s preferred. Between might also be acceptable (in context); avoid the temptation of betwixt, which reaches an entirely different level of silly archaic affectation than the others.
To sum up, once again, for those still playing along at home:
Best practice is not to use the “-st” ending forms of these words, especially in writing.
But if you’ve got a thing for those “-st” forms, go right ahead.
Don’t say you weren’t warned, though.