Word frequency variation: elicit vs. illicit
« previous post | next post »
In the comments on yesterday's post about a slip of the fingers or brain ("Elicit → illicit"), there was some discussion about which of the two words is more common.
Obviously, the answer to such questions depends on where you look.
So I looked in a bunch of places. Overall, illicit tends to be more common than elicit — but the relative frequency varies widely, and sometimes it's the other way round.
The english-corpora.org website lets us get word counts from 16 relevant corpora:
| Source | elicit | illicit | (illicit/elicit) |
| NOW | 24261 | 123370 | 5.09 |
| iWeb | 24185 | 41506 | 1.72 |
| GloWbE | 3614 | 7888 | 2.18 |
| Wikipedia | 2073 | 4971 | 1.38 |
| Hansard | 4342 | 4224 | 0.97 |
| COCA | 2697 | 4199 | 1.56 |
| TV | 158 | 446 | 2.82 |
| Movies | 61 | 133 | 2.18 |
| SCOTUS | 427 | 866 | 2.03 |
| TIME | 152 | 526 | 3.46 |
| SOAP | 20 | 62 | 3.10 |
| BNC | 240 | 256 | 1.07 |
| CORE | 130 | 238 | 1.83 |
| Strathy | 177 | 146 | 0.82 |
| Google Books (American) |
596697 | 407460 | 0.68 |
| Google Books (British) |
130941 | 116216 | 0.89 |
Checking the National Library of Medicine's PubMed database gives us article counts, not word counts, but FWIW:
elicit 256156
illicit 27563
illicit/elicit = 0.11
Similarly, Court Listener gives us (U.S. court) opinion counts, not word counts — but I was surprised to see illicit lose again:
elicit 119986
illicit 27025
illicit/elicit = 0.23
The NYT archive also gives article counts:
elicit 26903
illicit 24051
illicit/elicit = 0.89
The Atlantic Magazine archive, again with article counts:
elicit 474
illicit 581
illicit/elicit = 1.23
There are lots of other places to look, of course, but that should be enough to make the point.
Alexander Pruss said,
August 20, 2022 @ 10:03 am
Before this post, I would have sworn that "elicit" and "illicit" start with a different vowel sound. I rarely if ever use "elicit" in speech, but I think that if I had been asked to pronounce it, I would have given it the same starting vowel as in "egret". I just checked and to my surprise the standard online dictionaries all give the same pronunciation of "elicit" and "illicit". But, after getting past a paywall, Oxford's dictionary of Canadian English does give the egret-like pronunciation as an alternative, which may explain my conviction. I'm a Canadian ex-pat in Texas.
By the way, does anyone pronounce "elicit" starting the same as "egg"?
Philip Taylor said,
August 20, 2022 @ 12:20 pm
« By the way, does anyone pronounce "elicit" starting the same as "egg" ? ». Yes, I do. The LPD gives (for "elicit") /i ˈlɪs |ɪt / and (for "illicit") /ɪ ˈlɪs ɪt/.
David Morris said,
August 21, 2022 @ 7:17 am
Because illicit and elicit are different word classes, does the comparative frequency really tell us anything, anyway, compared with, say, illicit and illegal?
Charles in Toronto said,
August 21, 2022 @ 9:32 am
In the end, a lot of unstressed vowels in English can reduce to a schwa but have their own identity when you are pronouncing them with more intentional clarity. So elicit & illicit will sound the same when you schwa them, but different when you articulate them. If you learn the meanings from everyday conversation, the confusion would be understandable.