Aaaaay...Fonzie Conspiracy in Country Names?
November 6, 2007 1:02 PM Subscribe
Why do so many names of countries, cities, and states end in the letter "A"?
After checking the recent "-istan" thread, Wikipedia, and Google, I'm coming up short.
America. Atlanta. Ljubljana. Italia. Ukrayina (Ukraine). Ayutthaya. North Dakota.
These places would all seem to have different etymological roots, but the "a"-sound ending is consistent.
Is this just coincidence? Or is it just confirmation bias ("Yep, here's another a-ending... yep, here's another!") on my part as I look at an atlas?
Maybe (in Latin/European languages at least) the ending "a"denotes the feminine form of "land" (as in "mother land")?
After checking the recent "-istan" thread, Wikipedia, and Google, I'm coming up short.
America. Atlanta. Ljubljana. Italia. Ukrayina (Ukraine). Ayutthaya. North Dakota.
These places would all seem to have different etymological roots, but the "a"-sound ending is consistent.
Is this just coincidence? Or is it just confirmation bias ("Yep, here's another a-ending... yep, here's another!") on my part as I look at an atlas?
Maybe (in Latin/European languages at least) the ending "a"denotes the feminine form of "land" (as in "mother land")?
i don't know why, but i can tell you that when you play drinking geography bee (say a geographical name like a country, river, mountain etc that starts with the letter that the last person ended on) the "A"s are always the most common. any player worth their salt has a big reserve of obscure A places and things stored away. antwerp is my favorite.
posted by christy at 1:08 PM on November 6, 2007
posted by christy at 1:08 PM on November 6, 2007
Obviously we're talking about the English translations of country/city/state names. So maybe it has something to do with Latin? I'm just throwing shit at the wall and maybe it'll stick.
posted by HotPatatta at 1:21 PM on November 6, 2007
posted by HotPatatta at 1:21 PM on November 6, 2007
Oh, duh. You already posited the Latin explanation. I should have finished reading your post all the way through before posting.
posted by HotPatatta at 1:23 PM on November 6, 2007
posted by HotPatatta at 1:23 PM on November 6, 2007
Response by poster: Obviously we're talking about the English translations of country/city/state names.
Nope, I controlled for that in my question ("These places would all seem to have different etymological roots"):
Dakota (Native American)
Ukrayina (Ukrainian)
Ljubljana (Slovenian)
Ayutthaya (Thai)
These are, IIRC, what the people who live there call them.
posted by Rykey at 1:28 PM on November 6, 2007
Nope, I controlled for that in my question ("These places would all seem to have different etymological roots"):
Dakota (Native American)
Ukrayina (Ukrainian)
Ljubljana (Slovenian)
Ayutthaya (Thai)
These are, IIRC, what the people who live there call them.
posted by Rykey at 1:28 PM on November 6, 2007
Why does everyone that plays the Wheel of Fortune always guess the same letters, N S T R and L?
Because, statistically speaking, they're the most common consonants in the English language, which isn't relevant to the question at hand.
posted by mkultra at 1:30 PM on November 6, 2007
Because, statistically speaking, they're the most common consonants in the English language, which isn't relevant to the question at hand.
posted by mkultra at 1:30 PM on November 6, 2007
English transliteration is certainly a part of this question—it might be more accurate to say that a lot of those end, not with "a", but with a schwa-like syllable most naturally transliterated into roman/English characters as "a". "Okinawa", for example, doesn't end with "a" at all, except in transliteration; it ends with あ.
Running with that, the question could be considered to be why is that [set of] sounds as common as it is at the end of place names.
posted by cortex at 1:36 PM on November 6, 2007 [1 favorite]
Running with that, the question could be considered to be why is that [set of] sounds as common as it is at the end of place names.
posted by cortex at 1:36 PM on November 6, 2007 [1 favorite]
cortex is right to point out that it's not the letter "a" that is necessarily common, but the sound -ah and -ey. So, the interesting question is really two inches to the left but looks very similar.
(And you're speaking phoenetically, cortex, so あ was fine. :) )
posted by cmiller at 1:45 PM on November 6, 2007
(And you're speaking phoenetically, cortex, so あ was fine. :) )
posted by cmiller at 1:45 PM on November 6, 2007
I think the more interesting question would be what those respective countries are called in their native language.
posted by mphuie at 2:11 PM on November 6, 2007
posted by mphuie at 2:11 PM on November 6, 2007
It's not -ey, though, it's always -ah.
posted by alexei at 2:19 PM on November 6, 2007 [1 favorite]
posted by alexei at 2:19 PM on November 6, 2007 [1 favorite]
I don't think the bias you're asking about exists for Thailand. Of the 75 provinces it has, only 4 end with a (Phra Nakhon Si Ayutthaya, Phang Nga, Yala and Songkhla).
posted by kisch mokusch at 2:23 PM on November 6, 2007
posted by kisch mokusch at 2:23 PM on November 6, 2007
It's basically Latin and Greek, where -a is a feminine ending and -ia an extremely common place-name ending; hence Albania, Philadelphia, etc. There's another batch from other Indo-European languages where -a is also a feminine ending, for instance Slavic: hence Ljubljana and Україна. Everything else is coincidence/confirmation bias.
posted by languagehat at 2:29 PM on November 6, 2007
posted by languagehat at 2:29 PM on November 6, 2007
Slovenia, on the other hand... I think you're on a winner. Everything in that country seems to end with an a (7/8 provinces, by the same comparison).
posted by kisch mokusch at 2:30 PM on November 6, 2007
posted by kisch mokusch at 2:30 PM on November 6, 2007
I think you're right about Latin playing a significant part. Almost all countries had an -a ending in Latin and many cities did as well. (Ljubljana's Roman name was Emona -- the -a ending is pretty much the only commonality.) Latin's status as Europe's lingua franca probably sealed the deal.
posted by Ljubljana at 2:33 PM on November 6, 2007
posted by Ljubljana at 2:33 PM on November 6, 2007
This is a sort of "bad" (although interesting) question, because it makes assumptions that there's some connections which don't truly exist between these examples.
First of all, countries:
Many country names end in "-ia" or "-a" from a Greek-derived Latin ending from which country names were formed. It's for abstract feminine nouns - presumably the classification of "countries" or "nations." It's generally translated to mean "land of." That explains a lot of country names, especially for speakers of languages which were heavily influenced or descended from these languages. Most non-Romance languages I know don't make or use many country names this way, unless they're borrowings.
states:
Obviously, many state names were modeled in the same way - Virginia, West Virginia, North and South Carolina, Georgia, Pennsylvania, Indiana and Louisiana. That's 8 of the 21 states ending in "-a" or "-ia," all of them formed in the Latinate way to honor individuals, a people and a woodland named after someone. This leaves 13 states. Of those, four have these endings because they derive from Spanish sources which make place names this way (Montana, Arizona, Florida and California - although the last is somewhat contested, its ending, at least, appears to be formed by the Spanish.) That leaves 9 states. 5 states appear to derive from words which didn't end in "-a" or "-ia." Alaska (-q), Alabama (-o), Arizona (-k or -g), Iowa and Nebraska (both -e). One would assume that in the Europeanization of these names, the endings were altered to "-a" to fit a pattern, subconsciously or not. That leaves only four states: North and South Dakota share their etymology, obviously, which apparently is from a source which did end in an "-a" sound. As did Minnesota and Oklahoma.
It's not that many when one notes that many state names from Amerindian sources *don't* end in "-a" - Michigan, Illinois, Massachusetts, Idaho, Connecticut, etc.
Rykey, you didn't really "control" for anything by picking some non-English names which happened to end in an "-a" sound in the native language. For example, "Ukrayina" is from an Old East Slavic word which happened (possibly because it was a feminine noun, which often end in "-a" across many Indo-European languages) which comes from a root for the term "region," "edge" or "border." (Witness also the region "Krajina," which takes its name from the same root and which was part of the former Yugoslavia fought over by Serbia and Croatia.)
The Germans call Ljubljana "Laibach." German speakers know Vienna as "Wien." The Hungarians call the same city "Bécs" and they refer to Bratislava as "Pozsony." Come to think of it, there aren't many German or Austrian or Dutch or Swedish or Danish or Norwegian or British city names which end in "-a." (All of these are Germanic languages.) Or Hungarian city names, either. Actually now that I think about it, not a huge number of city names *do* end in "-a" or "-ia." A list of European capitals found at . . .
http://en.wikipedia.org/wiki/List_of_European_countries_and_their_capitals
. . . shows that only 11 out of 49 do, and several of those are their English, not native names. (Vienna = Wien) Subtract the ones formed in Romance or Slavic ways and there's only a statistically meaningless number. For the most part, I'd totally question the statement that an unusual number of city names end in "a."
Just before posting this, I notice that languagehat more or less is saying the same thing. But hopefully the info on state name origins will add a little something.
posted by Dee Xtrovert at 3:00 PM on November 6, 2007
First of all, countries:
Many country names end in "-ia" or "-a" from a Greek-derived Latin ending from which country names were formed. It's for abstract feminine nouns - presumably the classification of "countries" or "nations." It's generally translated to mean "land of." That explains a lot of country names, especially for speakers of languages which were heavily influenced or descended from these languages. Most non-Romance languages I know don't make or use many country names this way, unless they're borrowings.
states:
Obviously, many state names were modeled in the same way - Virginia, West Virginia, North and South Carolina, Georgia, Pennsylvania, Indiana and Louisiana. That's 8 of the 21 states ending in "-a" or "-ia," all of them formed in the Latinate way to honor individuals, a people and a woodland named after someone. This leaves 13 states. Of those, four have these endings because they derive from Spanish sources which make place names this way (Montana, Arizona, Florida and California - although the last is somewhat contested, its ending, at least, appears to be formed by the Spanish.) That leaves 9 states. 5 states appear to derive from words which didn't end in "-a" or "-ia." Alaska (-q), Alabama (-o), Arizona (-k or -g), Iowa and Nebraska (both -e). One would assume that in the Europeanization of these names, the endings were altered to "-a" to fit a pattern, subconsciously or not. That leaves only four states: North and South Dakota share their etymology, obviously, which apparently is from a source which did end in an "-a" sound. As did Minnesota and Oklahoma.
It's not that many when one notes that many state names from Amerindian sources *don't* end in "-a" - Michigan, Illinois, Massachusetts, Idaho, Connecticut, etc.
Rykey, you didn't really "control" for anything by picking some non-English names which happened to end in an "-a" sound in the native language. For example, "Ukrayina" is from an Old East Slavic word which happened (possibly because it was a feminine noun, which often end in "-a" across many Indo-European languages) which comes from a root for the term "region," "edge" or "border." (Witness also the region "Krajina," which takes its name from the same root and which was part of the former Yugoslavia fought over by Serbia and Croatia.)
The Germans call Ljubljana "Laibach." German speakers know Vienna as "Wien." The Hungarians call the same city "Bécs" and they refer to Bratislava as "Pozsony." Come to think of it, there aren't many German or Austrian or Dutch or Swedish or Danish or Norwegian or British city names which end in "-a." (All of these are Germanic languages.) Or Hungarian city names, either. Actually now that I think about it, not a huge number of city names *do* end in "-a" or "-ia." A list of European capitals found at . . .
http://en.wikipedia.org/wiki/List_of_European_countries_and_their_capitals
. . . shows that only 11 out of 49 do, and several of those are their English, not native names. (Vienna = Wien) Subtract the ones formed in Romance or Slavic ways and there's only a statistically meaningless number. For the most part, I'd totally question the statement that an unusual number of city names end in "a."
Just before posting this, I notice that languagehat more or less is saying the same thing. But hopefully the info on state name origins will add a little something.
posted by Dee Xtrovert at 3:00 PM on November 6, 2007
christy,
Antwerp has nothing on Appomattox.
It's the ending in -x places that will give you the edge:
Bronx
Delacroix
Essex
Fort Dix
Fort Knox
Middlesex
Sussex
I'm sure there are some more.
/derail
posted by SBMike at 3:32 PM on November 6, 2007 [1 favorite]
Antwerp has nothing on Appomattox.
It's the ending in -x places that will give you the edge:
Bronx
Delacroix
Essex
Fort Dix
Fort Knox
Middlesex
Sussex
I'm sure there are some more.
/derail
posted by SBMike at 3:32 PM on November 6, 2007 [1 favorite]
Response by poster: This is a sort of "bad" (although interesting) question, because it makes assumptions that there's some connections which don't truly exist between these examples.
Wrong. Whether there are connections, beyond the superficial sound of the endings of these place names, is my question.
You didn't really "control" for anything by picking some non-English names which happened to end in an "-a" sound in the native language.
Wrong. I attempted to head off the "That's just the Anglo/American name for that place" argument.
I notice that languagehat more or less is saying the same thing.
Right.
posted by Rykey at 4:09 PM on November 6, 2007
Wrong. Whether there are connections, beyond the superficial sound of the endings of these place names, is my question.
You didn't really "control" for anything by picking some non-English names which happened to end in an "-a" sound in the native language.
Wrong. I attempted to head off the "That's just the Anglo/American name for that place" argument.
I notice that languagehat more or less is saying the same thing.
Right.
posted by Rykey at 4:09 PM on November 6, 2007
Wrong. I attempted to head off the "That's just the Anglo/American name for that place" argument.
Simply by picking four unrelated, random place names which end in "-a" in the language native to them isn't any sort of control. An effective control would have been a truly random selection of place names in "their" languages (in other words, not pre-selected for "-a" endings.) Or pondering the same original question from a glance at a German or Hungarian (or some other) language map. Had you done that, you would have found (quite easily) that there isn't a preponderance of "-a" ending names outside of those from a few languages. So it's not a "control," because those places were chosen to further a "fact" which isn't true. I did bother to go to the few seconds of "work" to provide something closer to a control, by taking a random sample (in my case, of European capitals) and finding that there isn't any sort of "Fonzie conspiracy." Subtracting the already-explained Slavic and Romance "-a" ones, there aren't a meaningful number. One could take the world's ten most populous cities (another random sample, as far as names go), and see that none of them end in "-a" in their English versions (although Moscow does in Russian, roughly "Moskva"):
1 Mumbai
2 Karachi
3 Delhi
4 São Paulo
5 Moscow
6 Seoul
7 Istanbul
8 Shanghai
9 Lagos
10 Mexico City
Or one could have simply consulted a dictionary of etymological roots, knowing that this question was being asked this question from an English bias. This . . .
http://www.etymonline.com/index.php?l=i
. . . took me four seconds (one Google search) to find online. Check out the very first entry on that page.
You also claim that your list of places "would all seem to have different etymological roots, but the "a"-sound ending is consistent." But that's scarcely true - the feminized "a" ending in Greek, Latin and Slavic tongues probably stems from the same source; "Dakota" could easily have been "Europeanized" like many of the Amerindian state names I mentioned earlier. So ultimately, the etymological origins of the seven names you provided come from sisterly IE sources, except for the Thai one. Scientifically speaking, it's a very poor sample.
So if you're going to be snotty, yes it is a "bad" question from the perspective of being poorly (or not at all) researched. Part of the Ask MetaFilter creed is that one is supposed do a little checking first; none of what I've presented took me more than a few seconds to find online. And a preselected control is not a control.
posted by Dee Xtrovert at 7:01 PM on November 6, 2007 [1 favorite]
Simply by picking four unrelated, random place names which end in "-a" in the language native to them isn't any sort of control. An effective control would have been a truly random selection of place names in "their" languages (in other words, not pre-selected for "-a" endings.) Or pondering the same original question from a glance at a German or Hungarian (or some other) language map. Had you done that, you would have found (quite easily) that there isn't a preponderance of "-a" ending names outside of those from a few languages. So it's not a "control," because those places were chosen to further a "fact" which isn't true. I did bother to go to the few seconds of "work" to provide something closer to a control, by taking a random sample (in my case, of European capitals) and finding that there isn't any sort of "Fonzie conspiracy." Subtracting the already-explained Slavic and Romance "-a" ones, there aren't a meaningful number. One could take the world's ten most populous cities (another random sample, as far as names go), and see that none of them end in "-a" in their English versions (although Moscow does in Russian, roughly "Moskva"):
1 Mumbai
2 Karachi
3 Delhi
4 São Paulo
5 Moscow
6 Seoul
7 Istanbul
8 Shanghai
9 Lagos
10 Mexico City
Or one could have simply consulted a dictionary of etymological roots, knowing that this question was being asked this question from an English bias. This . . .
http://www.etymonline.com/index.php?l=i
. . . took me four seconds (one Google search) to find online. Check out the very first entry on that page.
You also claim that your list of places "would all seem to have different etymological roots, but the "a"-sound ending is consistent." But that's scarcely true - the feminized "a" ending in Greek, Latin and Slavic tongues probably stems from the same source; "Dakota" could easily have been "Europeanized" like many of the Amerindian state names I mentioned earlier. So ultimately, the etymological origins of the seven names you provided come from sisterly IE sources, except for the Thai one. Scientifically speaking, it's a very poor sample.
So if you're going to be snotty, yes it is a "bad" question from the perspective of being poorly (or not at all) researched. Part of the Ask MetaFilter creed is that one is supposed do a little checking first; none of what I've presented took me more than a few seconds to find online. And a preselected control is not a control.
posted by Dee Xtrovert at 7:01 PM on November 6, 2007 [1 favorite]
Response by poster: So if you're going to be snotty, yes it is a "bad" question from the perspective of being poorly (or not at all) researched.
Sorry, it was not my intent to be snotty.
Part of the Ask MetaFilter creed is that one is supposed do a little checking first; none of what I've presented took me more than a few seconds to find online.
I did some checking, and found nothing. That's part of my problem: I know nothing about languages or etymology (I was asking this question on behalf of a patron at the library where I work), so I didn't know how to start. The Googling I did was obviously faulty from the start.
And a preselected control is not a control.
Touche. Thanks for clearing that up.
posted by Rykey at 5:33 PM on November 7, 2007
Sorry, it was not my intent to be snotty.
Part of the Ask MetaFilter creed is that one is supposed do a little checking first; none of what I've presented took me more than a few seconds to find online.
I did some checking, and found nothing. That's part of my problem: I know nothing about languages or etymology (I was asking this question on behalf of a patron at the library where I work), so I didn't know how to start. The Googling I did was obviously faulty from the start.
And a preselected control is not a control.
Touche. Thanks for clearing that up.
posted by Rykey at 5:33 PM on November 7, 2007
This thread is closed to new comments.
posted by ORthey at 1:07 PM on November 6, 2007 [1 favorite]