One of the first things most people do when learning a new language is to learn the numbers from 1 to 10, and shortly after that, from 11 to 20. And it’s generally quite easy to remember them in the case of European languages, as they’re somewhat similar to their English counterparts. Look at French, for example:
Un, deux, trois, quatre, cinq, six, sept, huit, neuf, dix.
Not necessarily all obvious individually, but seen all together, they’re clearly the first ten numbers. The same goes, I think, for 11 to 19:
Onze, douze, treize, quatorze, quinze, seize, dix-sept, dix-huit, dix-neuf.
Clearly a combination of ten and the relevant number, and the same applies to English (eighteen = eight + ten, for example). Except, that’s not quite true in every case, is it? What about 11 and 12? They don’t fit the pattern at all. Twelve has a clear link to two, but not ten, and eleven seems to be unrelated entirely to either one or ten. What’s going on there then?
The most-accepted theory is that while eleven and twelve might not follow the same pattern as the seven numbers that follow, they’re still constructed in a similar way.
Most linguists agree that eleven is derived from the Old English enleofan, which means one left over (i.e. ten and one left over). Which seems like a slightly convoluted way to go about it, but it does make sense, after a fashion. Twelve probably has a similar origin, coming from twelf, meaning two left over.
So that’s that oddness explained then. But why aren’t 13 to 19 constructed in a similar way? It might be that 12 was considered special, being the number of months, lunar cycles, and signs of the Zodiac, among other things. And mathematically it’s a useful number, being divisible by 1, 2, 3, 4, 6, and 12. Everything after that, well, that’s all pretty ordinary, so just use the same pattern for them and let’s call it a day.
And The Twoteen Days of Christmas just doesn’t sound right, does it?