Mommy I'm twelveteen now!

September 5, 2009 11:43 AM

Why are the teen numbers (13-19) named differently than the rest of the numbers, and what's up with eleven and twelve?

I heard a little kid trying to count today and they went ten, eleventeen, twelveteen etc and it got me wondering. Why are the numbers 11-19 named differently than the other numbers?

20-29 is all twenty + one through 9. 30-39 is the same all the way up to 100. So why is it eleven and not tenone?


Wikipedia says that 11 and 12 are derived from german words meaning ten and one left and two left respectively. Ok .. I guess I can understand that, so then what's going on with the teens?

posted by Arbac to writing & language (33 answers total)

This page will help you:
To summarise, English is a Germanic language, but heavily influenced by Norman French, or to put it another way, a mess!
posted by Pater Aletheias at 12:03 PM on September 5, 2009

Not directly related to the "teens", but a more general answer about language for you: The words that are most basic, and most likely to have naturally been created first, generally don't follow grammar or other rules very well, because they were created before those rules became entrenched. So, for example, most of the time, past tense is denoted by "-ed", but the exceptions are in the most basic words: eat/ate, sleep/slept, see/saw/ etc.

So, why thirteen instead of onty-three? Maybe because the convention of "-ty" wasn't settled when they were adding the first numbers above ten.

I am not a linguist, though I got the basic insight from a language teacher. I reserve the right to be wrong about this.
posted by kingjoeshmoe at 12:05 PM on September 5, 2009

And the teen numbers aren't that different. Teen means 10. So thirteen is "three-ten" and so on.
posted by Pater Aletheias at 12:05 PM on September 5, 2009

maybe this is an oversimplification, but does teen sound like any specific number to you? (try ten).
posted by deadcrow at 12:06 PM on September 5, 2009

When it comes to this kind of thing, usage is often arbitrary. No one can say why it happened a certain way, it just did.

There are a lot of things in human languages which make little or no sense at all. That's just how it is.
posted by Chocolate Pickle at 12:18 PM on September 5, 2009

Yep, English is a mish-mash, and that's pretty much just par for the course. For more on the history of the English language, read up on the history of Britain. Conquest after conquest, that is. It's really no wonder they got so Imperial towards the end there.

In terms of numbers, just look at their former monetary system and their lingering units of measurement. England was just a crazy jumble of crazy numbers for centuries; 11 and 12 aren't the half of it.

Now, if you really want crazy: In French, the number 1997 is read mille neuf cent quatre vignt dix-sept, i.e. 1000 9 100 4 20 10 7. Four digits become seven numerals that are variously added and multiplied. Is it any wonder that these are the people who invented the metric system?

It's also worth noting that French doesn't start their teens (dix-) until 17.
posted by Sys Rq at 12:28 PM on September 5, 2009 [1 favorite]

My favorite French ones:
From 11 to 19:
Onze, douze, treize, quatorze, quinze, seize, [oh that's right we're in the 10's so let's continue with] dix-sept, dix-huit, dix-neuf...
And while we're at it, 70 is not something like seventy, but soixante-dix, that is, sixty-and-ten (same for eighty and ninety). But we don't just say eighty, we say quatre-vingt (four times twenty) so ninety is... you guessed it! Quatre-ving-dix (four-times-twenty-ten)!!!
posted by ddaavviidd at 12:32 PM on September 5, 2009 [2 favorites]

I suppose it could be argued that -ze is the French for -teen.
posted by Sys Rq at 12:34 PM on September 5, 2009

To those who point out that "teen" means ten - that's all well and good, but it still does not follow the pattern of the rest of the numbering system. If thirteen was "teenty-three" it would be much more logical.
posted by Rock Steady at 12:41 PM on September 5, 2009 [1 favorite]

Come to think of it, a lot of stuff only goes up to twelve. Twelve is big in numerology, and also in eggs and donuts, for instance. I'd be willing to bet there's something to that. You'd count up the individual bits to twelve, and then count the dozens.
posted by Sys Rq at 12:46 PM on September 5, 2009

The ~teen seems derived from the ~zehn on the end of the equivalent German numbers so does in fact mean ten- in a roundabout way; and as I understand it eleven (elf) and twelve (zwolf) were derived from German compound words for 'one left over' and 'two left over' hence them being special cases.

Most older European languages had/have vigesimal counting systems (groups of twenty). The early Brythonic languages (shown still in the 'purer' form of Welsh today) did, French still does in 70 (soixante-dix, or sixty-ten) and 90 (quatre-vignts-dix, four twenties and ten).
English has the word 'score' for twenty (e.g. "Four score and seven years ago...") so you could really start mixing things up and start saying 'one score and eleven' for 31 if you wanted to.
posted by sid.tv at 12:58 PM on September 5, 2009 [1 favorite]

When I was a kid watching British kids TV, or rather TV broadcast for schools to show kids, there was a programme about numbers. One of the presenters, Fred Harris, tried to get kids to count in 'standardised' numbers. There is no ten, instead it is onety. Twenty is twoty, thirty is threety, etc. I don't know what age I was when we were forced to watch it but I do remember thinking it was insane.

Clip of it here
posted by vbfg at 1:00 PM on September 5, 2009

Good soundtracks though.
posted by vbfg at 1:01 PM on September 5, 2009

When it comes to this kind of thing, usage is often arbitrary. No one can say why it happened a certain way, it just did.

There are a lot of things in human languages which make little or no sense at all. That's just how it is.

I don't see thing that way at all. If we assume things are "just are" then we never seek an explanation for them. We also never confirm whether that assumption is true by testing out alternative explanations and see if they work better. Yet we only ever apply the "just are" label to those things that lie outside of obvious patterns, such as with this questions regarding numbers. The asker doesn't really need to know how fifty-six and thirty-two are formed in English, because they already confirm to our understanding of how numbers are formed. The question is really, "why does the expected pattern not come true?" To which, "it just doesn't", isn't a very good answer.

In language, better answers are found by assuming that all language conforms to patterns. The alternative is that we have to memorize every single aspect of language, and generalizing rules is never worthwhile. Children don't learn this way, but instead attempt to generalize plurals across nouns and tenses across verbs, inspite of what competent users of a language know to be correct. Of course, languages aren't perfectly patterned, but understanding that they should be leads towards an explanation of why they're not.

The truth is that languages do conform to patterns, but that those patterns change with time. Speakers of English know to form plural nouns with -s or past tenses with -ed, but that hasn't always been the case, and won't always be. The old way of forming plural nouns is still present in a word such as oxen, and the past tense can be seen in a word like fought. Why these words didn't change to the new pattern is another matter, but their existence isn't arbitrary or inexplicable, it's a very real part of the history of English.

Getting back to the question, this layering of patterns applies also to numbers. The numbers that aren't obvious according to the current pattern of forming numbers are simply those which conform to earlier and now-defunct patterns. Sadly, I don't know all of these patterns, and how they have changed throughout the history of English. As Pater Aletheias has pointed out, influences form other languages may have helped construct our current numbering pattern, and moved us away from the previous one common to Germanic languages. This explains the split between the teens and twenty onwards. Eleven and twelve may result from some earlier influence, perhaps from a language with a duodecimal system - I don't know. But really the explanation is out there if we look, and I'm sure somebody with a better understanding of Germanic linguistics already knows the answer.
posted by Sova at 1:03 PM on September 5, 2009 [4 favorites]

Oops, wrong link. This one.
posted by vbfg at 1:04 PM on September 5, 2009

Eleven and twelve are actually related to one another, if that helps. It's easier to see in German:

11 = elf
12 = zwölf, ie zwo-elf.

"zwo" is a lot like "zwei", and two-eleven is kind of like twelve, innit?

From the OED's etymology for "twelve", formatting not even close to preserved:

f. twa two + li{bbar}- or lif-, of uncertain origin, but generally considered to belong to the same root as OTeut. *li{bbar}an to LEAVE (q.v.), and thus to denote ‘two left or remaining over (ten)’; cf. ELEVEN. Analogous formations to eleven and twelve are the Lith. vên{uang}´lika 11, dvýlika 12, in which the second element, Lith. -lika, has also the meaning of ‘left over’. All other Indo-Eur. langs. have or had forms composed of ‘two’ + ‘ten’, like the numbers 13 to 19; cf. L. du{omac}decim, Gr. {delta}{gwacu}{delta}{epsilon}{kappa}{alpha}, Skr. dw{amac}daçan.

Interestingly, Latin not only has the one-ten and two-ten words for eleven and twelve (undecim and duodecim), but also two-from-twenty and one-from-twenty words for eighteen and nineteen (duodeviginti and undeviginti)

The OED's etymology for "eleven":

f. *ain- (shortened from *aino-) ONE + -lif- of uncertain origin. Outside Teutonic the only analogous form is the Lith. vënó-lika, where -lika (answering in function to Eng. -teen) is the terminal element of all the numerals from 11 to 19.
The OE., OFris., OS., and ON. forms represent a type *ainlifun, app. assimilated to *tehun TEN. The theory that the ending is a variant of OTeut. *tehun, Aryan *dekm TEN, is now abandoned; some would derive it from the Aryan root *leiq or from *leip (both meaning to leave, to remain) so that eleven would mean ‘one left’ (after counting ten.)

And look! A mistake in the OED! The period in "(after counting ten.)" should be outside the parenthesis.

The question is, why don't we have threeleven or thirlve or whatever? Whence the switch from two-left to three-teen? Or, if formerly a system of numbers like that did exist, why weren't eleven and twelve affected when the other number names were adopted, presumably under the influence of surrounding languages? And when did this happen, given that modern German is similar? Old Frisian, Old Norse, and Old High German already had three-ten forms for "thirteen".
posted by kenko at 1:44 PM on September 5, 2009 [2 favorites]

NB I am not a historical linguist. I just have OED access.
posted by kenko at 1:46 PM on September 5, 2009

I also didn't see the note in the OP about wikipedia. Oh well! "Derived from German words" doesn't seem terribly accurate, though.
posted by kenko at 1:50 PM on September 5, 2009

To elaborate a bit on what kingjoeshmoe said - frequently-used words tend to be more irregular than less-used words because old forms are better preserved in words that get used a lot. It seems natural to assume that low numbers get used more (and before decimals took over the world twelve was a common unit) and hence are more likely to preserve old forms while higher numbers follow the standard 'modern' pattern.
posted by Phanx at 2:35 PM on September 5, 2009 [1 favorite]

Related: Malcolm Gladwell touched on this in his book Outliers. The Chinese language has the ten, ten one, ten two, ten three... pattern. He *thinks* that's why Chinese kids are so good at math because of the head start they have in the efficiency of the language relating to math.
posted by querty at 2:59 PM on September 5, 2009

To elaborate a bit on what kingjoeshmoe said - frequently-used words tend to be more irregular than less-used words because old forms are better preserved in words that get used a lot.

I'm sorry, but that makes no sense. Individual words (where "boot" and "boots" would be different individual words) are not themselves regular or irregular; what might be irregular or regular are the ways words relate to each other. "Kine" as a plural for "cow" is irregular but "cows" is not. Presumably you mean something like that.

But "kine" is older than "cows", we now say "cows", cows are pretty common, and reference to cows was probably more common when the change was occurring than it is now. Likewise with "brothers" and "brethren". I would expect frequently-used words to become regular faster than infrequently used words, at least among language-learners who aren't forcefully enough corrected by their teachers—you learn the currently productive rules and start applying them, and to what will you apply them more frequently than the words you frequently use? But really I have no idea about this.

"To be" is pretty irregular. What's up with that? OED time!

An irregular and defective verb, the full conjugation of which in modern Eng. is effected by a union of the surviving inflexions of three originally distinct and independent verbs, viz. (1) the original Aryan substantive verb with stem es-, Skr. as-, 's-, Gr. {elenis}{sigma}-, L. es-, 's-, OTeut. *es-, 's-; (2) the verb with stem wes-, Skr. vas- to remain, OTeut. wes-, Gothic wis-an to remain, stay, continue to be, OS., OE., OHG. wesan, OFris. wes-a, ON. ver-a; (3) the stem beu- Skr. bh{umac}-, bhaw-, Gr. {phi}{upsilon}-, L. fu-, OTeut. *beu-, beo-, OE. béo-n to become, come to be.

This sure ain't a case of old forms being preserved in a word that gets used a lot (and the copula gets used a lot!), as if long ago all verbs had infinitive forms like "be" and first-person singular present forms like "am", but only this one was preserved because people said it a lot (why wouldn't that have sustained the form in other words?). (If I were going to tell a just-so story about this it would involved different language groups each of which had regular forms commingling, communicating with each other, and ending up with something that resembled the ancestors in various synchronically nonsensical respects.)

Kingjoeshmoe's theory that words that get used a lot were created before the rules became entrenched is also kind of bonkers. There may not have been grammarians back in the PIE days, but why should we think that the language was irregular? That people would pluralize nouns (for instance) however they pleased?
posted by kenko at 3:37 PM on September 5, 2009

My completely baseless folk-etymology theory of why eleven and twelve are different is that ten is the new twelve. We still count some things in dozens, there are twelve months in a year, twelve hours in a half day, etc. Supposedly some systems of counting were based on finger bones, of which there are twelve (not counting your thumb). People might have been using units of twelve even before a base ten numbering system was introduced.
posted by dreadpiratesully at 4:16 PM on September 5, 2009 [2 favorites]

That wouldn't explain why eleven and twelve are one and two left over after ten, though.
posted by kenko at 4:21 PM on September 5, 2009

That wouldn't explain why eleven and twelve are one and two left over after ten, though.

Counting on fingers would do so. Twelve is a really useful unit for buying and selling, because it is divisible by so many numbers. So you can get a half-dozen, a quarter dozen, a third of a dozen etc. If you are counting on fingers, enumerating 11 and 12 as the two numbers in addition to ten that are needed to make up a dozen makes sense.
posted by Susurration at 6:37 PM on September 5, 2009

Ah ha! But what about Lithuanian?
posted by kenko at 8:52 PM on September 5, 2009

German does the same thing, so it obviously comes from the germanic stem: Zehn, Elf, Zwolf, Dreizehn, Vierzehn. Wiktionary says it comes from Old High German.
posted by Ironmouth at 1:19 AM on September 6, 2009

But really the explanation is out there if we look, and I'm sure somebody with a better understanding of Germanic linguistics already knows the answer.

Without a time machine, we can only speculate on the origins of languages. This is a fact. But we can make very good guesses.
posted by Ironmouth at 1:22 AM on September 6, 2009

And while we're at it, 70 is not something like seventy, but soixante-dix, that is, sixty-and-ten (same for eighty and ninety).

As an aside, this isn't true in all dialects of French. Several Swiss French dialects use the "regular" forms septante, huitante, and nonante.
posted by Johnny Assay at 7:37 AM on September 6, 2009

Because, if eleven and twelve weren't named as such, this anagram wouldn't work:

ELEVEN + TWO = TWELVE + ONE
posted by TSGlenn at 10:43 AM on September 6, 2009 [1 favorite]

I would expect frequently-used words to become regular faster than infrequently used words, at least among language-learners who aren't forcefully enough corrected by their teachers—you learn the currently productive rules and start applying them, and to what will you apply them more frequently than the words you frequently use? But really I have no idea about this.

Isn't it interesting that one's intuition can be completely the opposite of what actually happens? Phanx had it absolutely right. In language after language, word frequency often plays a strong role in whether words have regular or irregular forms, and it is the more frequent forms that tend to maintain irregular forms, while less common words become regular. This does make sense, because in actual speaking, from a young age, you are hearing these irregular forms; thus if you are trying to recall the proper form to mind when speaking, common irregular forms should be front and center in your mind. On the other hand, if a word is not very commonly heard, to the extent that the irregular form does not immediately come to mind or doesn't "sound natural", it is more likely that the speaker will then fall back on the most general rule (if there is one), such as the -s ending for nouns or -ed for past tense verbs.

School, by the way, is not where any person attains their fundamental command of language -- cultures with no education system at all have complete, rule-based languages. Also, the concept of formal language education is a lot younger than pretty much all of the languages in the world, including English. (I have to say, even if the role of school is wrongly assumed as key, your theory still sounds a little bizarre... did your teachers attempt to teach you some sort of regularized conjugation for the verb "to be"?)

Kingjoeshmoe's theory that words that get used a lot were created before the rules became entrenched is also kind of bonkers. There may not have been grammarians back in the PIE days, but why should we think that the language was irregular? That people would pluralize nouns (for instance) however they pleased?

Absolutely not -- the rules were simply different and they evolved, and sometimes in languages there is not a general rule that applies to 90% forms. German today has different plural forms for different words, and the particular plural must be marked in each dictionary entry. "Haus" gets an -er at the end (and a vowel change), "Frau" gets an -en, "Mädchen" doesn't change at all, "Auto" gets an -s, and so on. Arabic pluralization has even more variation than German. That doesn't mean you can pluralize however you want. It just means you might have to go word-by-word in some cases.

The fact that eleven and twelve differ from the pattern of the rest of the teens, and that the -teen ending differs somewhat from the 20s - 90s, is very likely to be related to the fact that the smaller numbers are used with greater frequency than larger numbers. That doesn't tell us everything interesting about eleven and twelve, such as how it happened, where it came from, and why we stop doing that at twelve, but it definitely plays an underlying role.
posted by kosmonaut at 7:42 AM on September 7, 2009

School has nothing to do with it. Teachers include everyone one talks to.
posted by kenko at 10:37 AM on September 8, 2009

Anyway, I take the point regarding the first half of your comment, but you seem to have completely misunderstood the gist of what you're replying to in the second.
posted by kenko at 10:38 AM on September 8, 2009

(German pluralization isn't quite as bizarre as you make it out to be: all diminutives have identical singular and plural forms, for instance. Anyway, look at kingjoeschmoe's comment: the words that are most basic, and were created first, were created before the rules were? Present-day German pluralization won't help you decide that claim one way or the other!)
posted by kenko at 10:45 AM on September 8, 2009