Join 3,512 readers in helping fund MetaFilter (Hide)


Where did numbers get their names?
January 16, 2012 10:19 AM   Subscribe

Where did numbers get their names? I'm asking particularly of English, but I'm guessing the answer goes back pretty far linguistically.
posted by squaregear to Writing & Language (19 answers total) 16 users marked this as a favorite
 
Zero: early 17th cent.: from French zéro or Italian zero, via Old Spanish from Arabic ṣifr ‘cipher.’

One: rom Greek patronymic -ōnē

Two: Old English twā (feminine and neuter), of Germanic origin; related to Dutch twee and German zwei, from an Indo-European root shared by Latin and Greek duo. Compare with twain.

Three: Old English thrīe (masculine), thrīo, thrēo (feminine), of Germanic origin; related to Dutch drie and German drei, from an Indo-European root shared by Latin tres and Greek treis .

Four: Old English fēower, of Germanic origin; related to Dutch and German vier, from an Indo-European root shared by Latin quattuor and Greek tessares .

Five: Old English fīf, of Germanic origin; related to Dutch vijf and German fünf, from an Indo-European root shared by Latin quinque and Greek pente .

Six: Old English siex, six, syx, of Germanic origin; related to Dutch zes and German sechs, from an Indo-European root shared by Latin sex and Greek hex .

Seven: Old English seofon, of Germanic origin; related to Dutch zeven and German sieben, from an Indo-European root shared by Latin septem and Greek hepta .

Eight: Old English ehta, eahta, of Germanic origin; related to Dutch and German acht, from an Indo-European root shared by Latin octo and Greek oktō .

Nine: Old English nigon, of Germanic origin; related to Dutch negen and German neun, from an Indo-European root shared by Sanskrit nava,Latin novem, and Greek ennea .

Ten: Old English tēn, tīen, of Germanic origin; related to Dutch tien and German zehn, from an Indo-European root shared by Sanskrit daśa,Greek deka, and Latin decem

Eleven: Old English endleofon, from the base of one + a second element (probably expressing the sense ‘left over’) occurring also in twelve; of Germanic origin and related to Dutch and German elf .

Twelve: Old English twelf(e), from the base of two + a second element (probably expressing the sense ‘left over’); of Germanic origin and related to Dutch twaalf and German zwölf. Compare with eleven.

-Teen: Old English, inflected form of ten.
posted by jedicus at 10:23 AM on January 16, 2012 [21 favorites]


To add to what jedicus has already listed, if you can get your hands on an American Heritage Dictionary, the latest edition has a rundown on the Indo-European and Sanskrit roots, you can go back a little bit further. Unfortunately, their website doesn't show the etymological info that's available in the book or app.
posted by jquinby at 10:37 AM on January 16, 2012 [1 favorite]


Adding to the above, Indo-European numbers (Wikipedia).

They can't be traced back any further than that.
posted by nangar at 10:41 AM on January 16, 2012 [2 favorites]


I have to disagree with your explanation for 1, Jedicus.

The words for "one" in other modern Indo-European languages are almost universally similar to the English "one", and the root for "one" is far older than even ancient Greek, nor does "one" seem to pass through Greek except tangentially (the Greeks used α to represent "one" for their numerals, although the lowest throw on ancient dice was called the οἴνη - oine, similar to the Latin unus, "one"). Instead, "one" seems closer related to the Old Frisian ān, ēn, Old Saxon ēn (Middle Low German ēn), or Old High German ein, ehin, ēn.

(All this is from the OED; I'm not sure what the rules here are on copying and pasting from other, paywalled websites, or else I'd just post the whole article on "One", which is quite long.)
posted by Oxydude at 11:05 AM on January 16, 2012


Jedicus' definitions from the Oxford dictionary, I presume.
posted by Think_Long at 11:06 AM on January 16, 2012


I think maybe jedicuses explanation for one has been cut off short.
posted by Jehan at 11:07 AM on January 16, 2012


> I think maybe jedicuses explanation for one has been cut off short.

Apparently so, but it doesn't really matter, since the numeral is not from any Greek patronymic; it's from PIE *oi-no- and is related to Latin unus, among others.

There's nothing particularly interesting about the etymology of numbers as opposed to other basic word groups, but a starting point for any investigation would be Mark Rosenfelder's Numbers from 1 to 10 in Over 5000 Languages. Compare the numbers in any of the language families he helpfully divides them into and you can usually see what's going on.
posted by languagehat at 11:14 AM on January 16, 2012 [3 favorites]


I think maybe jedicuses explanation for one has been cut off short.

Well, only by one letter: "from Greek patronymic -ōnē" is the entire etymology in the New Oxford American Dictionary, which is where my answers came from. I'm not a linguist, so I've got no personal stake in my answer, and I'm happy to see a better one.
posted by jedicus at 11:15 AM on January 16, 2012


Ha, oh, I see what I did. I scrolled down too far and took the etymology for -one in the chemistry sense (e.g. acetone).

The etymology for the number is: Old English ān, of Germanic origin; related to Dutch een and German ein, from an Indo-European root shared by Latin unus. The initial w sound developed before the 15th cent. and was occasionally represented in the spelling; it was not accepted into standard English until the late 17th cent.
posted by jedicus at 11:16 AM on January 16, 2012 [1 favorite]


> "from Greek patronymic -ōnē" is the entire etymology in the New Oxford American Dictionary

Wow, that may be the single worst etymology I've ever seen in a modern dictionary.

*shakes finger at Oxford*
posted by languagehat at 11:17 AM on January 16, 2012 [1 favorite]


Wow, that may be the single worst etymology I've ever seen in a modern dictionary.

Yes, until I discovered my error I was starting to think it was one of those intentional mistakes designed to catch copiers.
posted by jedicus at 11:18 AM on January 16, 2012


me: They can't be traced back any further than that.

Correcting myself in case there's any confusion: English number names can't.

(languagehat's already provided a link to zompist for other groups of languages.)
posted by nangar at 11:34 AM on January 16, 2012


The reconstructions for PIE are the "pretty far" that you're after (I'm always tickled by the concordance between Welsh pedwar and Greek tetra) but what might be interesting is to focus on the numbers in languages with PIE origins that appear to have distinct names early, those that are almost always compounds, and those that exist somewhere in between. One to ten are pretty locked in, twenty stands out among the higher multiples of ten, and so on. Between 10 and 20, it's more up for grabs.

Zero as a numerical entity is, of course, the big standout.
posted by holgate at 11:38 AM on January 16, 2012


Apparently so, but it doesn't really matter, since the numeral is not from any Greek patronymic; it's from PIE *oi-no- and is related to Latin unus, among others.

That's true, though I wondered if it might be cut so short that an additional example of the same root's expression as a Greek patronymic was all that was left. Not that I've ever heard of Greek patronymics being derived from numbers, but still.
posted by Jehan at 11:44 AM on January 16, 2012


What's notable about number names in English is how complex to say 'hundred' and 'thousand' are. Ever thought about how often we say 3 and 4 digit numbers as sequences of digits ("Volvo Two-Forty", "Nineteen Seventy-Four", "Boeing Seven Four Seven", etc)? That doesn't happen in other languages where 100 and 1000 are pronounced with one syllable (e.g. French) or two simple syllables (Spanish (100), German).
posted by ambrosen at 11:58 AM on January 16, 2012


two simple syllables (Spanish (100), German).

"Hundert" isn't any more complex to say than "hundred", especially given the way the second, unstressed syllable of "hundred" is typically pronounced: "hunderd".
posted by Philosopher Dirtbike at 12:20 PM on January 16, 2012


er, *less* complex.
posted by Philosopher Dirtbike at 12:21 PM on January 16, 2012


Eleven: Old English *endleofon*

Having seen that, I can never look at "11" the same old way again.
posted by exphysicist345 at 3:14 PM on January 16, 2012 [1 favorite]


The best thing about English numbers is the set that exists just for counting sheep.

My favourites are bumfit (15) and jiggit (20).

So I'm just going to assume you are asking about their etymologies instead of the boring ones. They come from Celtic.
posted by lollusc at 5:53 PM on January 16, 2012


« Older Technically they could fire me...   |  What are the best interactive ... Newer »
This thread is closed to new comments.