Non-etymologies
Sep. 7th, 2011 11:37 am![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
dr_whomOne of my... hobbies?... is finding words that look like they should be etymologically related but actually have completely different origins. In a few cases, like male and female, the similarity between them is actually the result of analogy; one of the words was remodeled to resemble the other out of, apparently, some feeling that they ought to sound similar. (Female was originally femelle.) In other cases, the similarity is apparently completely coincidental: pencil looks like it should be a diminutive of pen but isn't; can looks like it should be a truncation of canister but isn't. My favorite example is that even though bondage 'the condition of being bound' is transparently just bond 'that by which a thing is bound' plus a derivational suffix, they actually have completely unrelated origins. (In this case, though, the similarity in meaning may actually have been influenced by the similarity in sound.)
There's also a few well-known classic cases of words in different languages that have the same meaning and similar or identical sounds: English much and Spanish mucho, German haben and Latin habeo, English bad and Persian bad, English dog and Mbarabam dog. These are habitually trotted out in Ling 1 or Intro to Historical Ling classes to get across the idea that just having the same sound and meaning isn't enough to establish that words in two languages are cognates: English and Mbarabam obviously aren't even related to each other, and in the other cases the languages are related but the words don't exhibit the sound correspondences that are characteristic of the histories of those languages. (For instance, real cognates of Latin words that begin with h actually begin with g in German, like hortus and Garten.)
But I have a new favorite pair of unrelated words that do obey the sound correspondence. So, Latin s regularly corresponds to Greek h, as in Latin sex and Greek hex 'six', or Latin super and Greek hyper 'over'. The Latin word for 'sweat' is sudor. By regular sound correspondences, the Greek cognate of sudor should be hydor—which just happens to be the Greek word for 'water'. And yet there is no etymological relationship between these two words!
There's also a few well-known classic cases of words in different languages that have the same meaning and similar or identical sounds: English much and Spanish mucho, German haben and Latin habeo, English bad and Persian bad, English dog and Mbarabam dog. These are habitually trotted out in Ling 1 or Intro to Historical Ling classes to get across the idea that just having the same sound and meaning isn't enough to establish that words in two languages are cognates: English and Mbarabam obviously aren't even related to each other, and in the other cases the languages are related but the words don't exhibit the sound correspondences that are characteristic of the histories of those languages. (For instance, real cognates of Latin words that begin with h actually begin with g in German, like hortus and Garten.)
But I have a new favorite pair of unrelated words that do obey the sound correspondence. So, Latin s regularly corresponds to Greek h, as in Latin sex and Greek hex 'six', or Latin super and Greek hyper 'over'. The Latin word for 'sweat' is sudor. By regular sound correspondences, the Greek cognate of sudor should be hydor—which just happens to be the Greek word for 'water'. And yet there is no etymological relationship between these two words!
![[identity profile]](https://www.dreamwidth.org/img/silk/identity/openid.png){kind=small}
no subject
Date: 2011-09-07 03:57 pm (UTC)no subject
Date: 2011-09-07 09:18 pm (UTC)no subject
Date: 2011-09-07 05:54 pm (UTC)no subject
Date: 2011-09-07 09:33 pm (UTC)no subject
Date: 2011-09-07 11:41 pm (UTC)So okay: male comes from Latin masculus (as in "masculine"), via French, which is where all the consonants in the middle dropped out. Masculus is the diminutive of mas, the Latin word for 'male'. Female used to be femelle which is from the Latin diminutive of femina 'woman' (as in "feminine"), also via French but without losing any of its consonants. So actually I guess the L's in male and female do have a common origin; they're both the L from the Indo-European diminutive suffix. But the M's are totally unrelated.
Pen is more or less direct (well, via French) from Latin penna 'feather'. Pencil is a diminutive, but not a diminutive of penna; it comes from penicillus 'small paintbrush', which is a diminutive of penis 'tail' (which is also the source of the English word penis). Penna and penis themselves look like they might be related, but aren't; penis originally had an S in it, which it lost in the history of Latin (cf. Sanksrit pasas 'penis'), whereas penna: originally had a T and is cognate with feather.
Canister is from Latin canistrum basket, which is from Greek, from kanna 'reed' (what baskets were made of), which is itself perhaps of Semitic origin. Can is a native Germanic word whose origin is apparently unknown; the dictionaries I checked can't trace it back farther than proto-Germanic *kannōn-. (Cannon, on the other hand, comes from kanna—as does cane and canal.)
Bond comes from band and is actually related to bind and bound, and comes from a root that's had that meaning apparently all the way back to Indo-European. (The -bund in cummerbund is related to it—it comes from the Persian for waistband.) Bondage is originally 'the state of being a bondsman' (i.e., a householder or peasant farmer, which became essentially slavery over the course of British history). This isn't related to bind and band, but it is the same root found in husband; this is from a root apparently meaning 'live' or 'dwell', and is actually ultimately related to both be and build.
And here's a bonus non-cognate for you: minor and minimum and so on come from the Latin words for 'less' and 'least' and an Indo-European root meaning 'little'. But miniature comes from a Latin word meaning 'red lead', which is thought to be "probably of Iberian origin" and possibly related to Basque.
On the other side of the coin, here's a pair of words which are etymologically related: if you trace it back far enough, the word chair is literally just the root sit plus one prefix and one suffix.
no subject
Date: 2011-09-07 11:49 pm (UTC)I'm in three online classes through community colleges this semester, in a drive to finally complete the requirements for my English teaching license. American Lit 1 is incredibly dumb, but in exactly the way you would expect a community college course to be dumb (e.g., on a short research paper, the professor actively corrected my dates to say "September 2" instead of "2nd"). Intro to Public Speaking is... it's own series of livejournal posts. But the Intro to Linguistics class is really good, for the context! I like the book a lot, the assignments are both reasonable and rational, and the other students appear to have ever taken a college class before! So, yeah. I've just started the chapter on syntax.
no subject
Date: 2011-09-08 01:35 am (UTC)no subject
Date: 2011-09-08 01:37 am (UTC)no subject
Date: 2011-09-08 02:24 am (UTC)no subject
Date: 2011-09-08 01:41 am (UTC)no subject
Date: 2011-09-08 02:23 am (UTC)You've basically got the gist of it. Sit comes to us straight from Indo-European *sed-, with very little change on the way, as you can see: just the basic Germanic family patterns of *e becoming i and *d becoming t.
As I noted above, Indo-European *s becomes h in Greek, so *sed- becomes hed-. Then we add a prefix kat-, which means 'down', and a suffix -ra, which basically means 'okay guys this is a noun now', and we get kathedra 'thing you sit down on', i.e., 'chair'. (This is also the source of cathedral—the building where a bishop's seat is.) Then, as you say, we take kathedra and crank it through French—and the job of French is to take all the consonants in the middle of words and politely forget about them—and we end up with chair. So all that remains of the sed that we started with is one lonely i.