Afraid of what?

A few days ago the chapter of the textbook was about comparative and superlative adjectives, and one question was something like “What are you most afraid of?”. One student said “I am afraid of ” something that sounded like duck or dog. Was she afraid of ducks (the bird) or duck (the meat), or dogs (the animal) or dog (the meat, in some countries, see later)? I might have asked for clarification then, but decided to let her keep talking. She said that when she was young, the toilet was accessed from outside, so she always asked one of her parents to take her. So did they have ducks or dogs in their backyard? I finally said “I don’t know whether you said duck or dog”. She said “No – daakk”. Aha. “Afraid of the dark.” Why do we say “the dark” rather than “dark”. Would Dracula say “I am afraid of light” or “I am afraid of the light”? Google Ngrams shows that afraid of the light is about twice as common as afraid of light. Continue reading



Mistakes are interesting because they tell us something about what we notice easily and what we don’t notice without training.

– commenter ktschwarz on a Language Log post


I am trying to sell some old furniture through a ‘Buy, Sell, Swap’ group on Facebook. Someone in the group has advertised “furnitures” for sale. In current-day standard English, this is a plain mistake, but it may gain some usage under the influence of second-language learners and speakers. It makes sense, and there’s no doubt what people mean when they say or write it.

The more I investigated, the murkier it got. There’s a group of uncountable nouns which represent a collection of items, or more accurately there’s two groups of uncountable nouns which represent a collection of items. A flock of sheep consists of sheep (rams, ewes and lambs, a limited list), but furniture consists of tables, chairs, couches etc (a potentially unlimited list). Google Ngrams shows that a furniture appears overwhelmingly as a noun modifier of store, factory, manufacturer etc (and that its usage skyrocketed before 1890 and 1910, so I don’t know what people called it before then) and that furnitures is used just often enough for it may not to be a plain mistake. Among other things, it is used with the verbs are and were. Two of the most common collocations are furnitures thereunto and furnitures whatsoever, which suggests that it has a legal usage. Continue reading

A kind of affliction

Last Tuesday was an interesting day linguistically, even if it was a slow day work-wise. I noticed three separate issues twice each in different contexts. The first time each, I thought “Oh, that’s interesting” and the second time I thought “Hang on, I’ve seen that before”.

During a lull in my work, I was browsing through some of Geoffrey Pullum’s old Language Log posts. In one, titled ‘Another victim of oversimplified rules‘, he discusses a sentence which he found in a free newspaper on Edinburgh’s buses:

A record number of companies has been formed by Edinburgh University in the past 12 months.

Continue reading

More subeditorial accuracy

Following yesterday’s post about subeditorial accuracy, today I was subediting an article about nutrition in pregnancy. The external writer included the statistic that nine out of 100 full-term pregnancies result in the death of the baby shortly before, during or shortly after birth. That … can’t … be …. right. Fortunately, the source the writer cited says that the figure is nine out of 1,000. Unfortunately, every one of those is a tragedy for the family.

Maybe this is a good reason to include commas in 4- or more-digit numbers. If she’d written “1,00” it would immediately look obviously wrong. 

Relatedly, not long ago, I was subediting another article about pregnancy, and it mentioned the mother’s diet preconception. I just had to put a hyphen into that.

(I’ve also decided that in this blog from now, subedit and all related words will not be hyphenated. (WordPress accepts subedit, but not subeditorial.))

Sub-editorial accuracy

One of the purposes of sub-editing is to ensure accuracy of content. A few days ago, I was editing an article about complementary medicines (vitamins, minerals, herbs, weight management, sports nutrition etc), which stated that the value of the CM industry in Australia had increased from “$1.9million” in 2011 to “$3.5 billion” in 2015. Wow!  Really?

The author (an in-house journalist) provided a reference to a document published by a body claiming to be the largest CM industry organisation in Australia, and that had those figures. He agreed, however, that the first figure just had to be wrong, and that we should change it to “$1.9 billion”. An author and sub-editor agreeing is one thing; changing a monetary amount against a documented source is another.

I searched for “complementary medicine Australia 1.9 million 2011” and found another document from the same body also stating “$1.9million” but with a graph calibrated in “AUD $ mill”, showing a figure just below “2,000”, that is “$1,900 million”. My guess is that the original author typed that, thought “hang on, that looks awkward’, changed the “1,900” to “1.9”, but then forgot to change “million” to “billion”. (The lack of a space after “1.9” is another clue that something’s been edited there.)

So, this will appear in the next issue of one of our magazines as “$1.9 billion”. And none of our readers will ever know the editorial detective work which went on behind the scenes.


I try. I really try. I really really try not to notice mistakes in other people’s writing, and really really try not to blog about it here. If the blog in question was just anyone’s language-related blog, I’d say ‘ouch, then bite my tongue (or the digital (in both senses) equivalent), but it’s the blog of a Major Language-Related Website, so I’ll say ‘ouch’, then blog about it here.

One post referred to the A Song of Ice and Fire/Game of Thrones character Thormund Giantsbane and the British writer Road Dahl. Ouch  and double ouch.

In real life, the spelling Thormund exists, but Tormund is much more common. (Thormund more closely reflects the original Old Norse Þórmundr.) The ASoIaF/GoT character is Tormund, although the spelling Thormund is occasionally used, more often on websites less related to the books or tv series.

In real life, Roald is still used in Nordic countries and among Nordic emigrants (but less so than in the days of Roald Amundsen and Roald Dahl). Road isn’t a name anywhere, though there are several occurrences of Road Dahl on the internet, including goodreads and IMDb. (I have a vague memory of first thinking that his name was Ronald.)

I’ll be generous and say that both spellings are the result of momentary inattention and muscle-memory influence from the much more common words Thor and road. But please proofread, especially if you’re a blogger on a Major Language-Related Website.

(Before anyone points out any mis-spellings in this post (I have re-read very carefully!), please note that this is not a Major Language-Related Website.)