A reader wrote in to The Times complaining about one the newspaper’s word games. The game involves making as many English words as you can from a fixed set of letters. The reader complained that the permitted words did not include krill.
In its weekly Feedback column on 19 November, The Times, reported the complaint. It also gave the reason: the rules of the game do not accept plural nouns. To avoid arguments, the newspaper specifies a dictionary (the Concise Oxford Dictionary) to determine which words are acceptable. Apparently, the COD specifies that Krill is only a plural noun and has no singular.
I haven’t been able to look this word up in the Concise Oxford Dictionary, but the Oxford Advanced Learner’s Dictionary does indeed list krill only as plural. The definition there reads: ‘Krill [plural] very small shellfish that live in the sea around the Antarctic and are eaten by whales’.
I disagree with the Feedback column’s response for 2 reasons:
- the blanket exclusion of all plurals is unnecessary
- krill may be an uncountable mass noun, not a plural.
Excluding all plurals is unnecessary
I can understand why the rules of the game prohibit regular English plurals ending in -s. If the rules permitted such plurals, players would score 2 points for what is essentially the same answer. For example, they could score a point for word and another point (with minimal effort) for words.
The game’s rule of excluding plural nouns would exclude not just regular plurals formed with the suffix -s, but also:
- irregular plurals of native English words, such as mice (from mouse), teeth (from tooth), brethren (one of the plurals of brother, in a specialised meaning); and
- irregular plural forms derived from other languages, such criteria (Greek plural of criterion), alumni (Latin plural of alumnus), bureaux (from French bureau) and cherubim (Hebrew plural of cherub).
I am not convinced that it is sensible for the game’s rules to exclude irregular plurals. A player submitting, say, criteria and criterion would gain 2 points, but also has to know 2 things: the singular and the plural. It seems to me that a fair reward for this knowledge is quite a bit more than 1 point—though perhaps 2 points is a slight over-reward.
Some English words have only a plural form, and lack a separate singular. Examples include cattle, scissors, dregs and news—and, according to The Times, krill. I can see no reason why the rules of the word game should prohibit these words. 1 point seems a fair reward for knowing one of these words. That reward would also be the same as for a noun (such as fish or sheep) having a plural identical with the singular.
In summary, I think the game’s rule of excluding all nouns that are only plurals is too restrictive. The game should exclude only plurals formed by adding the regular suffix -s to a separate singular form.
In English, one noun that can’t be used in the singular is tongs. Want a tongs for eat it – Language Miscellany
Plural or uncountable mass noun?
The comment in the Feedback column says that krill is only a plural word. I’m not convinced. I think of krill as an undifferentiated mass (like soup), not as a collection of distinct individual animals. So, to my mind, krill is a singular mass (non-count) noun, not a plural that happens to have the same form as the singular.
Quick look at some evidence
Now although I’ve seen many wildlife documentaries that talk about krill as a staple foodstuff for sealife, I’ve not had much reason to talk about krill much myself. Thus, I can’t be sure that people who do talk and write about krill share my intuitions about this word.
So, I used the Google Books NGRAM viewer to test whether krill is always, sometimes or never singular in written English. I also tested whether krill is a count noun or a collective noun. For these tests, I searched for 3 word sequences: krill are; krill is; a krill. Figure 1 shows the results:
- the frequencies for all 3 sequencies follow a similar trajectory.
- the word krill was extremely rare until the late 1960s.
- frequency peaked around 1980 and has dropped back since then to levels last reached in the early 1970s.
- krill is (blue line) occurred a little more often than krill are (red line) from 1960 to about 1995, but since then krill are has been slightly more common. But there is not a huge difference in frequency between krill are and krill is.
- krill is occurs a little more often than a krill (green line).

Google Books NGRAM viewer is at https://books.google.com/ngrams/
I used the default settings on NGRAM viewer: English corpus [ie the whole database, not sub-divided into, for example, British English and American English], case insensitive, smoothing of 3. https://books.google.com/ngrams/graph?content=krill+is%2Ckrill+are%2Ca+krill&year_start=1960&year_end=2019&corpus=26&smoothing=3
On limitations and pitfalls of using NGRAM viewer, please see Guideline for improving the reliability of Google Ngram studies: Evidence from religious terms, by Nadja Younes and Ulf-Dietrich Reips (2019), at https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0213554
Conclusions
Although this evidence is only limited, it suggests that krill:
- is not just plural, contrary to the assertion in the Oxford Dictionaries. If it were only plural, it could never combine with the singular verb form is and could never combine with the indefinite article a, which can only be singular.
- is not just a non-count (mass) collective noun, contrary to my own intuition. If it were only a collective noun, it could never combine with the indefinite article a.
I haven’t collected evidence to assess:
- whether all writers treat krill as a noun that is sometimes plural and sometimes singular, or whether some writers treat it as always plural and others treat it is plural.
- whether all writers treat krill as count noun or whether some writers treat it as a a non-count (mass) collective noun
For other posts using data from NGRAM viewer, please see https://languagemiscellany.com/tag/ngram/
Etymology
The word krill comes from the Norwegian krill (in Bokmål) or Kril (in Nynorsk). https://ordbokene.no/bm,nn/search?q=kril Norwegian monolingual dictionaries show that the Norwegian word has 2 meanings:
- marine shrimp-like luminous planktonic crustaceans of the order Euphausiacea
- small fry (young) of herring—and, some English sources suggest, of other species of fish
Further reading
The Oxford Reference Guide to English Morphology, by Laurie Bauer, Rochelle Lieber and Ingo Plag (2013)
Sections 5.2 and 16.5 of A Student’s Introduction to English Grammar, by Rodney Huddlestone and Geoffrey K Pullum (2005, 1st edition). The 2nd edition came out recently.
Pluralia tantum nouns and the theory of features: a typology of nouns with non-canonical number properties, Greville G Corbett, Morphology (volume 29, 2019), open access at https://link.springer.com/article/10.1007/s11525-018-9336-0
And ‘data’ is the plural of ‘datum’……
Thanks for the comment, Richard.
Certainly true etymologically, though I don’t know how many people still think of ‘data’ as plural.
I tend to treat ‘data’ as a singular mass noun (‘the data is’). Personally, I find it quite unnatural to treat ‘data’ as a plural count noun (‘the data are’), though some people still do.
On the rare occasions when I need to talk about a single item, I use ‘datapoint’ or ‘piece of data’. The singular count noun ‘datum’ is very old-fashioned, to my ear, and I don’t think I have ever used it, though of course I understand it.