Numbers are weird. By that I don't mean that numbers per se, that is 1, 2, 3, etc., are weird, though they are strange in many ways, as any mathematician will tell you. I mean numbers as represented in individual languages. After all, this is a language blog.
All languages have words for numbers, though they reflect their speakers' interpretation of numbers in a myriad ways (there, you see; I've just used myriad, which is the Ancient Greek for "ten thousand" - why have a word for ten thousand?). Some languages have over a hundred words for numbers, while others might have only two or three (one, two, many). A short comparison of such treatments will provide some numerical food for thought.
Have you ever noticed how similar the numbers from one to ten are in most of the European languages and many Asian languages? That's because most, if not all, of the numbers from one to ten in these languages share a common ancestry and still survive in the modern languages, showing their derivation from their Indo-European ancestor. Take the number "two": French deux, Russian dva, Greek dhio, Welsh dau and Hindi/Urdu do. We might want to add to that Indonesian dua, but actually, this is pure coincidence, and dua has no connection with the Indo-European forms, belonging to a separate language group altogether.
Staying with Indo-European, nine seems to be similar to new, and possibly not without reason, as it may well have been quite literally a new number thousands of years ago. However, some of the most fascinating ways languages play with numbers can be seen in modern languages. Here are a few nuggets.
In English, we have one to ten, thirteen to nineteen and twenty to ninety, with -teen and -ty clearly variants of ten. So far so regular. But where do eleven and twelve come from? Surely we should have oneteen and twoteen? It seems that our Germanic ancestors, and hence the ancestors of German, Dutch, Swedish, Norwegian, Danish and all the other long dead Germanic languages, thought it good to combine en (an alternative form of one) and two with an old form of leave, creating the forms we have today and an original meaning of something like “one left over, two left over”.
A similar thing can be seen in Finnish, except that they extend it to all the -teen numbers. The word for ten in Finnish is kymmenen, so to create twenty to ninety, they just join up one to nine with a form of ten: kaksi (2) + kymmenen = kaksikymmentä (20). However, for eleven to twenty, they join up one to nine with -toista, literally “of the second”, in the sense that eleven to twenty is the second series of numbers after one to ten: kaksitoista (12).
The French have their own weird way of counting when it comes to the higher -ty set, as most British schoolchildren have known and hated for years. Everything is fine up to sixty-nine, soixante-neuf (no puns intended here; this is a serious language blog - if you want to believe that). But then things go strange: soixante-dix, literally “sixty ten”, and so on up to soixante-dix-neuf, “sixty nineteen (79)”. You might think that things would return to normal for 80, but you'd be wrong; they get all nerdy-mathematical here with quatre-vingt, “four twenty”, which goes on till quatre-vingt-dix-neuf, “four twenty nineteen” (99). What do French schoolchildren say when they want an ice cream with a chocolate flake in it? “Can I have a four twenty nineteen please?” If only the French had listened to their Swiss-French neighbours and stuck with good old septante, huitante, and nonante.
Now, let's turn to Russian. If there's a language which loves to complicate counting, Russian takes the biscuit. Russian is an inflected language, which means that it changes the endings of words depending on what the words are doing in sentences. Let's take student, which means, naturally, “student”. “A student” is student; “of a student” is studenta; “students” is studenty; and “of students” is studentov. So far, so good. Now, let's start counting our students with literal translations from Russian: “one student” - odin student, “two of student” - dva studenta, “three of student” - tri studenta, “four of student” - chetyre studenta, “five of students”, pyat' studentov. This continues up to “twenty of students”, but then we get to “twenty one student” - dvadtsat' odin student, and it all starts again, the same format all the way through to infinity: “thousand of students” - tysyacha studentov, “thousand one student” - tysyacha odin student, “million of students” – million studentov, “million one student” - million odin student. And just when you thought it was safe to learn all the other numbers, the system runs over into the multiples of ten: twenty – dvadtsat', thirty – tridtsat', fifty – pyat'desyat, sixty – shest'desyat, seventy – sem'desyat, eighty – vosem'desyat, ninety – devyanosto, hundred – sto. When you say "fifty students", it's pyat'desyat studentov, but "of fifty students" becomes pyatidesyati studentov. literally "of fifty of students". Oh, did I miss something? I forgot to mention the bundle of furs, srak, from which comes sorok, forty. Presumably Russian trappers bundled up their furs in units of forty.
If we move east of Russia, we get into even stranger territory. In many Asian languages, it's not possible to say simply “one house, two men, three dogs”, and so on. You have to know how to measure out the quantity of the noun that you are counting. For example, in Mandarin Chinese, the word for three is san, and the word for dog is gou, so you could reasonably expect “three dogs” to be something like san gou. However, you need to add a special word known as a measure between the number and the noun in order to package up the quantity of the noun, rather as we would say “three loaves of bread”. OK, so you just learn the measure and put it in the middle, but it's not as simple as that. Nouns in Chinese can be classified by what measure they use in common. Added to that, the measure in question for a particular classification of nouns may not seem to be obvious, based on the meaning of the measure and the nouns that are classified with it. Dogs are classified with tiao, hence san tiao gou, “three dogs”, but tiao is used for long, winding, wriggly things, like rivers, so how did dogs get in there? Were they long and wriggly in the ancient Chinese mind?
Indonesian shares the idea of measures with Chinese, but one of the easiest things about this language, apart from its relatively uncomplicated pronunciation, is the number of separate words that need to be learnt in order to count from one to a billion, a total of 15: 1 to 9 - satu/se, dua, tiga, empat, lima, enam, tujuh, delapan, sembilan; the word denoting multiples of ten – puluh; the word denoting the numbers from 11 to 19 – belas; hundred – ratus, thousand – ribu, million – juta, billion – milyar, hence: 10 - sepuluh 11 – sebelas, 12 – dua belas, 20 – dua puluh, 39 - tiga puluh sembilan, 256 – dua ratus lima puluh enam. Easy-peasy.
Yet, across the Indian Ocean, the exact opposite occurs. While you only have to deal with fifteen words in Indonesia, if you want to learn Hindi/Urdu, for 1 to 100 you have to learn, well, literally one hundred words. This is because all the numbers from eleven upwards are each effectively fused into single units, with the unit number from 21 to 99 forming the first part. The numbers from twenty to thirty illustrate this: bis, ikkis, bais, teis, chaubis, pacchis, chabbis, sattais, attais, untis, tis. On the bright side, all the numbers ending in, say, “five” in English will begin pa- in Hindi/Urdu. But don't get your hopes up too much as there are yet more numbers to learn: hundred – sau, thousand – hazar, hundred thousand – lakh, million – mil or das lakh, ten million – kror (also spelt crore), plus some even higher ones.
I think that's enough on numbers now, even though I've only just scratched the surface, or else I feel my number will be up, I'll be at sixes and sevens and lose my place on cloud nine. If you have any more fascinating examples of human creativity and mental agility regarding numbers, I'll be happy to see them on this blog.