I learned that sometimes Russian teachers (as in, RUSSIAN teachers, not teachers of Russian) will erroneously call it a vowel because it's not like the other consonants.
I can hardly imagine a RUSSIAN teacher (that is a person grown up in Russia) would call "Й" a vowel. Maybe you have a teacher that haven't been here for a looooong time, so he's practically forgot his native language.
Anyway, the question was fully answered by others.
yep. I figured that out when I started to see mens' names that ended with й. I knew mens' russian names had to end in a consonant, so I double checked with my prof.
::I knew mens' russian names had to end in a consonant Er, not necesssarily. E.g. Лука, Никита, Савва, Кузьма (most of these are not very popular these days, but they are mens' names anyway). To say nothing about diminutives, most of which end with a vowel: Ваня, Коля, Дима. But you figured out correctly - y is indeed a consonant.
My son is Никита Кириллович. In his elementary school class, there were four more Nikitas. Лука, Савва and Кузьма (as well as Фома) definitely are not that popular anymore.
It's a soft consonant because it's basically a consonant and a soft sign (ь) wrapped together in one package. Words that end in -й will take soft endings, for example.
I never said it was pronounced "fun-yuh." Y is always pronounced as eej, for example in "you," one would say "eeju." Although I don't like using js because they make it seem as if they're pronounced in another way. You know what I mean.
Besides, a letter should never be bi-consonovocal (made-up word). It is either a consonant or a vowel, and any letter saying so otherwise would just be lying to itself.
It's a consonant. In IPA, it would be /j/, or like the "y" in "boy", but a bit stronger. From a past discussion on this community, I learned that sometimes Russian teachers (as in, RUSSIAN teachers, not teachers of Russian) will erroneously call it a vowel because it's not like the other consonants.
-- so, ignore them.
и is a vowel. й is a consonant.
(The reasons why й is odd? It very rarely begins a word--and then only in loanwords--and never follows a consonant, as far as I know. The sound /j/ is written in a different way in those cases.)
In Russian words that would have started with й (necessarily followed by a vowel) the first character instead is a "jotted" (йотированная) vowel. For example, ёлка is not spelled as йолка which I thought would be more logical when I was a little boy.
"10 букв предназначены для обозначения гласных звуков и условно называются гласными (а, у, о, ы, э, я, ю, ё, и, е), 21 буква предназначена для обозначения согласных звуков и условно называется согласной (б, в, г, д, ж, з, й, к, л, м, н, п, р, с, т, ф, х, ц, ч, ш, щ), ъ и ь не относят ни к гласным, ни к согласным и называют графическими знаками."
(Е. И. Литневская. Русский язык: краткий теоретический курс для школьников)
It's a consonant. The best I've seen it transliterated as is "j," which makes sense to me, knowing Latin ... j was added to the Roman alphabet to represent the consonantal i (y in English, й in Russian). I think the reason that a transliteration of "y" doesn't work is because "y" in Eastern European languages that use the Roman alphabet can mean the same sound as "ы."
Probably more information than you wanted, but ... yeah. To make it short, й is a consonant.
As lunaqueen mentioned, /j/ is a semi-vowel, or more generally a 'glide'.
However, what it comes down to, is how speakers of a language want to classify their letters. The question asked was about orthography, not phonetics. So as irkin said, if school books say й a consonant, it's a consonant.
What is a "semi-vowel"? I mean, in Russian Й behaves like a consonant in all respects there are, as both a letter and a sound, so calling it a "semi-vowel" means what?
First of all, read what I wrote. I said /j/ is a semi-vowel. Й is a consonant.
A semi-vowel is a phonetic term describing a "vowel-like sound with consonant properties". This idea is separate from the alphabet of a language, where every letter is either a consonant or a vowel (or a soft/hard sign).
For example, the sound that corresponds to the Russian Й (and I assume all the time, unlike in English with our crazy orthogrpahy) is designated by the International Phonetic Alphabet (IPA) symbol /j/. It is articulated with the tongue body on the palate, it is voiced (your vocal chords vibrate as you say it). It is a glide (the air is not fricated enough to give you the sound of a German "ch"). And like vowels it had high sonority.
This is not Й. Й is a letter written down and placed within a word to guide pronunciation. And yes, it is pronounced like /j/. And yes, the pronunciaton of Й doesn't change like that of the English Y. But a letter represents a sound. It cannot behave like a sound.
Take that chart and flush it down the toilet. Й has always been a consonant and never had the dualism of the English Y.
Well there is no dualism of the English Y. The letter itself a consonant because we call it a consonant. What happens when it occurs in a word and we read that word does not affect its status as a consonant.
And on a side note, if you're thinking that Russian is a 'phonetic' language. Think again. You've got all these learners who know that the placement of stress in a word (which is only written in textbooks) changes the pronunciation of the vowels across the word.
And maybe yes, unlike English, the consonants in the alphabet seem to correspond to only one sound. But the letters by themselves don't tell you whether the sounds they correspond to are soft or hard, you need to look at the vowel following them to know. Ironically, Й is one of those few sure letters that always corresponds to a soft sound.
> a letter represents a sound. It cannot behave like a sound
Yeah, we should have started with that. Generally we cannot say whether a letter is a vowel or consonant, this division is only about sounds.
I have read in some book (i don't remember exactly the book) that Й and so called 'jotted' vowels had a real /j/ sound only when placed in the word's beginning. But being placed after a vowel the /j/ sound converted into a неслоговая (i.e. 'not producing a syllable') vowel (in IPA: undotted i, as in /mein/ being the reading of the word 'main'). And /j/ was considered a consonant sound, and /i/ (undotted) was a semi-vowel.
Actually, I just thought of a good reason why й is not a vowel. A vowel is something that can form a syllable on its own, e.g. о-гу-рец - "o-" is a legitimate syllable. Й cannot form a separate syllable; e.g. чай will take one "space" in a poetic meter, not two, and Николай - three, not four, therefore й is definitely not a vowel.
It seems like we've had every possible answer so far: vowel, consonant, both, semi-vowel... There's good reasons for the confusion, though. The main reason is the multiple meanings for "vowel" and "consonant"; basically a confusion between phonetic, phonology and conventional grammar. In my long answer below, I only use other, more specific words, avoiding saying "vowel" and "consonant" at all.
The short answer is: it's a consonant. This is the answer that you'd give someone who was studying a Russian textbook and looked up for a moment to ask "is й a vowel or a consonant?" if they had trouble deciding if, say "край" was a masculine or feminine word.
The long answer, not specifically concerning Russian letters: Of all the sounds that the speech apparatus can make, some sounds have more "sonority" than other. Roughly speaking, they have "more sound"; compare the sound of "i" with that of "f" - all other things being equal, "f" is much quieter. We can call the sounds at the high-sonority end of the scale "vocoids", and those of lower sonority (the rest) wa can call "contoids". Where is the dividing line drawn? A vocoid is basically held to be a sound produced without any closure or narrowing of the vocal tract sufficient to produce audible friction, while a contoid involves either complete closure of the tract or enough narrowing of the tract to produce some audible friction. Note that these are purely *phonetic* terms, they have nothing to do with how a particular language uses the sounds (phonology).
Samples of some vocoids: the sounds represented by the letters ...
"a" and "i", but not "e" in machine "u" and "r" in butter "u" and "y" in funny "w" and "e" in wet "y", "o" and "a" in yoga both of the "l"s in little, as well as the "i", but not the "e"
You can probably guess that the sounds represented by the letters "t" and "n" in "tin", "b" and "ck" in "back", and so on, are contoids. Note: I'm only mentioning words in order to make it easy to refer to the sounds they contain. These sounds are not vocoids or contoids simply by virtue of their position in the word, not should you confuse them with letters. When I say "like the 't' in 'tin'", I'm just giving you a recipe to isolate a particular sound that you can make, and that sound "t" on its own is a contoid. So, so far we're only talking about sounds, not their use in words or languages.
That was phonetics, which deals with the sounds the speech apparatus can make. Now some details from phonology, which is concerned with how languages put these sounds to use:
Syllables in languages like to arrange themselves to have a high-sonority sound at their nucleus (core), with lower-sonority sounds coming before and after. For example, in the word "tak", the "a" is high-sonority, and the "t" and "k" are low-sonority. The point is there there's usually a sonority peak in the syllable: and increase, a maximum, and a decrease (or there might be just a maximum in a syllable the the one you probably produce when you pronounce "o!" - the onset & coda are not required). The high point is called the nucleus, and the low points called either the onset or the coda (if it comes before or after the nucleus, respectively). Since the nucleus has high sonority, there's a good chance that it will be a vocoid. The onset & coda must then be either contoids or less sonorous vocoids. If the syllable nucleus happened to be a contoid, I guess the onset & coda would have to be lower-sonority contoids. I can't think of anywhere this happens, though (maybe I don't know enough Georgian: though I think Georgian just has very complex onsets....). I should say that it's the nucleus that acts, among other things, as the primary bearer of syllable stress (in langs that have it), the loudest part of the syllable, and ususally the longest part of the syllable. The other parts are more or less just decorations :).
And finally, somewhere in the vicinity of conventional grammar:
When it comes to writing down a language in a system that represents the language's phonology (not like, e.g. Chinese writing), you often try to give a separate graphical symbol to each different sound you find in the language (I'll pass over the subtleties of this), and you'll probably find that even if the language allows the same (presumably vocoid) sound to occur both in the nucleus and the onset (or coda) of syllables, you'll probably give them different symbols, or in some other way help to identify the nucleus. In fact, even the IPA tends to do this.
So, for example, in Russian we have a sound which, when it occurs in the nucleus of a syllable is written "и" and when it occurs elsewhere, is written with the symbol "й". In German, for the same purposes, you use "i" and "j". In English, you use "i" and "y" (Note: the converse isn't true: in English, the letters "i" and "y" have other uses. I only give English as an example because people reading here are familiar with it - it's a terrible example of my point). Sometimes, a vocoid used in an onset or coda position is called a "semi-vowel", and even imagined to be a separate sound from when it's used as a nucleus (they can sound rather different). As I mentioned the IPA does this too: it uses the pairs [i]/[j], [u]/[w], [y]/[ɥ] to distinguish nucleus/non-nucleus roles for some vowels.
At this stage we have a writing system, and we have the conventional distinction between "vowel letters" and "consonant letters". Stereotypically, the former refer to symbols that represent vocoid sounds in syllable-nucleus positions, and the latter refer to contoid sounds in the margins (onset & coda) of syllables. As for the other two possible combinations, there's no standard way in traditional grammar to describe them (maybe in Sanskrit grammar?): contoids in the syllable nucleus (like the second "r" in butter) might be called syllabic consonants, and vocoids in the margins of syllables can also be called "consonants", though if they're of high sonority (like [i] aka [j]), they might be called "semi-vowels".
In English, traditionally we say that "a" "e" "i" "o" and "u" are "the vowels" meaning that they're "the vowel letters". We could include "y" too, but you have to remember that usage of English letters isn't consistent: we use "y" in "yoghurt" (representing a vocoid - the same sound as "i" in "machine", in a "consonantal" (i.e. onset) position), in "my" (representing two vocoids ([a]+[i] in IPA) in a nucleus ("vowel") position), we use "l", which we call a "consonant", in the word "little" twice, first of all representing a vocoid in onset ("consonant"), second representing the same vocoid in nucleus ("vowel") position. This second use has another oddity, since we use the letter "e" after the second "l" in "little" to (perhaps) indicate that "l" is acting as a syllable nucleus ("vowel"), so that "e" has no sound or syllabic role of its own here. However, since the letter "e" is usually used to represent any of a number of vocoid sounds in nucleus position ("vowel"), we call it (the letter!) a "vowel", and if asked "what's the second vowel in 'little'?", most non-linguists would respond with the answer "e". Behind all this subterfuge is the traditionally belief (expressed in other ways) that English doesn't have syllabic consonants (perhaps because Latin didn't).
So, you can see where the confusion comes from! - even in Russian, where things are slightly better behaved. What it's hard to see is: what do people mean when they say "vowel" or "consonant"? If they're referring to a division of graphical symbols into two classes, then they're talking about a convention of the analysis of writing, something with no simple correspondence to the workings of the language's phonology. That's not so bad, but the real trouble arise when people use orthographic notions of "vowel" and "consonant" to ask questions about the phonology of a language.
So what is "й"? First of all, it's a letter. The sound it represents is the sound also represented by "и"; namely a vocoid - that's as close as you can sensibly get in phonetics to saying it's a vowel. Russian grammar calls certain *letters* (which are graphical characters with a complex relationship to sounds, not sounds themselves) "vowels", and certain others, "consonants", and this is used to encode certain bits of grammatical information in the spelling. For example, Russian nouns are either masculine, feminine or neuter, and membership of a particular class entails grammatical changes elsewhere in the sentence. Now, this gender isn't random, but is (usually) assigned on the basis of phonology: If the last syllable of the nominative singular has no coda, the noun is usually F or N (it doesn't matter which right now), and if it has a coda, it's usually M (exception: if the coda is a palatalized contoid, the noun can be either F or M). Treating "й" as a "consonant" allows you to explain this without getting into phonology and syllable structures: you can rephrase the gobbledygook in the last sentence as: if the noun ends in a "consonant", it's M, except for some which end in consonant + soft sign, which are F. Notice that there's been another redirection from phonology to "spelling", involving the "soft sign". (Imagine if there were two soft signs - one meaning "the noun that this sign ends is M", and another meaning "the noun this sign ends is F", but both pronounced the same! If these were as firmly entrenched in people's minds as "letters" in the same way that "й" and "и" are, we'd probably be having a very confusing discussion abouth them right now, and making phonological inferences from what would be basically nothing more than a spelling convention.)
Basically, the division of letters into "consonants" and "vowels" allow us to come up with simpler rules. Which, however, break down when people ask difficult questions. :)
Sorry it's so long. This issue comes up every now and again in various communities, and I thought I'd write a full answer to it, so that I can refer to it in future :). Hopefully I didn't stay from the point too much (some chance).
After spending some effort learning elementary Welsh, I just don't worry about orthography arguments any more. Learn how the language sounds. Figure out how to read and write with the rules they use, and be happy.
Is "y" (or "w" for that matter) considered vowel or a consonant, open or occluded? It doesn't matter. They're symbols with that stand for several kinds of sounds. We use them in several different ways, and millions of native speakers have already figured it out. So can learners.
no subject
Date: 2005-03-12 01:39 pm (UTC)no subject
Date: 2005-03-12 01:41 pm (UTC)no subject
Date: 2005-03-12 01:58 pm (UTC)no subject
Date: 2005-03-12 02:00 pm (UTC)no subject
Date: 2005-03-12 02:03 pm (UTC)no subject
Date: 2005-03-12 02:08 pm (UTC)no subject
Date: 2005-03-12 02:16 pm (UTC)Ok, nevermind. It is y as in Nikolay. Pronounced with a Russian accent, of course.
no subject
Date: 2005-03-12 02:20 pm (UTC)no subject
Date: 2005-03-13 06:39 am (UTC)Й is NEVER pronounced as "ee" in beet.
no subject
Date: 2005-03-13 03:09 am (UTC)no subject
Date: 2005-03-12 04:58 pm (UTC)no subject
Date: 2005-03-12 05:29 pm (UTC)I learned that sometimes Russian teachers (as in, RUSSIAN teachers, not teachers of Russian) will erroneously call it a vowel because it's not like the other consonants.
-- so, ignore them.
I have one of those teachers.
no subject
Date: 2005-03-13 03:25 pm (UTC)Anyway, the question was fully answered by others.
no subject
Date: 2005-03-12 01:50 pm (UTC)no subject
Date: 2005-03-12 03:34 pm (UTC)no subject
Date: 2005-03-13 06:43 am (UTC)Er, not necesssarily. E.g. Лука, Никита, Савва, Кузьма (most of these are not very popular these days, but they are mens' names anyway). To say nothing about diminutives, most of which end with a vowel: Ваня, Коля, Дима. But you figured out correctly - y is indeed a consonant.
no subject
Date: 2005-03-13 09:19 pm (UTC)no subject
Date: 2005-03-14 09:59 am (UTC)no subject
Date: 2005-03-12 01:59 pm (UTC)no subject
Date: 2005-03-12 02:35 pm (UTC)It's a soft consonant because it's basically a consonant and a soft sign (ь) wrapped together in one package. Words that end in -й will take soft endings, for example.
no subject
Date: 2005-03-12 02:02 pm (UTC)Remember the cereal Alpha-Bits: "A, E, I, O, U, and sometimes Y!"
...Sometimes Y? Really now...
Yeah, I go with mystache with the "'soft consonant,' whatever that is."
no subject
Date: 2005-03-12 04:45 pm (UTC)no subject
Date: 2005-03-12 05:19 pm (UTC)Besides, a letter should never be bi-consonovocal (made-up word). It is either a consonant or a vowel, and any letter saying so otherwise would just be lying to itself.
no subject
Date: 2005-03-12 02:34 pm (UTC)-- so, ignore them.
и is a vowel. й is a consonant.
(The reasons why й is odd? It very rarely begins a word--and then only in loanwords--and never follows a consonant, as far as I know. The sound /j/ is written in a different way in those cases.)
no subject
Date: 2005-03-12 08:40 pm (UTC)no subject
Date: 2005-03-12 03:09 pm (UTC)no subject
Date: 2005-03-12 05:38 pm (UTC)no subject
Date: 2005-03-14 03:25 pm (UTC)still, linguists can't stop arguing about this matter
no subject
Date: 2005-03-12 03:25 pm (UTC)"10 букв предназначены для обозначения гласных звуков и условно называются гласными (а, у, о, ы, э, я, ю, ё, и, е), 21 буква предназначена для обозначения согласных звуков и условно называется согласной (б, в, г, д, ж, з, й, к, л, м, н, п, р, с, т, ф, х, ц, ч, ш, щ), ъ и ь не относят ни к гласным, ни к согласным и называют графическими знаками."
(Е. И. Литневская. Русский язык: краткий теоретический курс для школьников)
no subject
Date: 2005-03-12 04:43 pm (UTC)Probably more information than you wanted, but ... yeah. To make it short, й is a consonant.
no subject
Date: 2005-03-12 05:15 pm (UTC)However, what it comes down to, is how speakers of a language want to classify their letters. The question asked was about orthography, not phonetics. So as
no subject
Date: 2005-03-12 05:41 pm (UTC)And if you call a tiger a mouse it still remains a fierce striped flesh-eator ;)
no subject
Date: 2005-03-13 03:23 am (UTC)no subject
Date: 2005-03-13 09:40 am (UTC)A semi-vowel is a phonetic term describing a "vowel-like sound with consonant properties". This idea is separate from the alphabet of a language, where every letter is either a consonant or a vowel (or a soft/hard sign).
For example, the sound that corresponds to the Russian Й (and I assume all the time, unlike in English with our crazy orthogrpahy) is designated by the International Phonetic Alphabet (IPA) symbol /j/. It is articulated with the tongue body on the palate, it is voiced (your vocal chords vibrate as you say it). It is a glide (the air is not fricated enough to give you the sound of a German "ch"). And like vowels it had high sonority.
This is not Й. Й is a letter written down and placed within a word to guide pronunciation. And yes, it is pronounced like /j/. And yes, the pronunciaton of Й doesn't change like that of the English Y. But a letter represents a sound. It cannot behave like a sound.
Take that chart and flush it down the toilet. Й has always been a consonant and never had the dualism of the English Y.
Well there is no dualism of the English Y. The letter itself a consonant because we call it a consonant. What happens when it occurs in a word and we read that word does not affect its status as a consonant.
And on a side note, if you're thinking that Russian is a 'phonetic' language. Think again. You've got all these learners who know that the placement of stress in a word (which is only written in textbooks) changes the pronunciation of the vowels across the word.
And maybe yes, unlike English, the consonants in the alphabet seem to correspond to only one sound. But the letters by themselves don't tell you whether the sounds they correspond to are soft or hard, you need to look at the vowel following them to know. Ironically, Й is one of those few sure letters that always corresponds to a soft sound.
no subject
Date: 2005-03-13 10:07 am (UTC)why can't we edit these things???
sigh.
IPA
Date: 2005-03-14 01:32 pm (UTC)no subject
Date: 2005-03-13 10:12 pm (UTC)no subject
Date: 2005-03-14 01:36 pm (UTC)Yeah, we should have started with that. Generally we cannot say whether a letter is a vowel or consonant, this division is only about sounds.
I have read in some book (i don't remember exactly the book) that Й and so called 'jotted' vowels had a real /j/ sound only when placed in the word's beginning. But being placed after a vowel the /j/ sound converted into a неслоговая (i.e. 'not producing a syllable') vowel (in IPA: undotted i, as in /mein/ being the reading of the word 'main'). And /j/ was considered a consonant sound, and /i/ (undotted) was a semi-vowel.
no subject
Date: 2005-03-13 07:12 am (UTC)Loooooooooooooooooooong answer
Date: 2005-03-13 12:58 pm (UTC)The short answer is: it's a consonant. This is the answer that you'd give someone who was studying a Russian textbook and looked up for a moment to ask "is й a vowel or a consonant?" if they had trouble deciding if, say "край" was a masculine or feminine word.
The long answer, not specifically concerning Russian letters: Of all the sounds that the speech apparatus can make, some sounds have more "sonority" than other. Roughly speaking, they have "more sound"; compare the sound of "i" with that of "f" - all other things being equal, "f" is much quieter. We can call the sounds at the high-sonority end of the scale "vocoids", and those of lower sonority (the rest) wa can call "contoids". Where is the dividing line drawn? A vocoid is basically held to be a sound produced without any closure or narrowing of the vocal tract sufficient to produce audible friction, while a contoid involves either complete closure of the tract or enough narrowing of the tract to produce some audible friction. Note that these are purely *phonetic* terms, they have nothing to do with how a particular language uses the sounds (phonology).
Samples of some vocoids: the sounds represented by the letters ...
"a" and "i", but not "e" in machine
"u" and "r" in butter
"u" and "y" in funny
"w" and "e" in wet
"y", "o" and "a" in yoga
both of the "l"s in little, as well as the "i", but not the "e"
You can probably guess that the sounds represented by the letters "t" and "n" in "tin", "b" and "ck" in "back", and so on, are contoids. Note: I'm only mentioning words in order to make it easy to refer to the sounds they contain. These sounds are not vocoids or contoids simply by virtue of their position in the word, not should you confuse them with letters. When I say "like the 't' in 'tin'", I'm just giving you a recipe to isolate a particular sound that you can make, and that sound "t" on its own is a contoid. So, so far we're only talking about sounds, not their use in words or languages.
That was phonetics, which deals with the sounds the speech apparatus can make. Now some details from phonology, which is concerned with how languages put these sounds to use:
Syllables in languages like to arrange themselves to have a high-sonority sound at their nucleus (core), with lower-sonority sounds coming before and after. For example, in the word "tak", the "a" is high-sonority, and the "t" and "k" are low-sonority. The point is there there's usually a sonority peak in the syllable: and increase, a maximum, and a decrease (or there might be just a maximum in a syllable the the one you probably produce when you pronounce "o!" - the onset & coda are not required). The high point is called the nucleus, and the low points called either the onset or the coda (if it comes before or after the nucleus, respectively). Since the nucleus has high sonority, there's a good chance that it will be a vocoid. The onset & coda must then be either contoids or less sonorous vocoids. If the syllable nucleus happened to be a contoid, I guess the onset & coda would have to be lower-sonority contoids. I can't think of anywhere this happens, though (maybe I don't know enough Georgian: though I think Georgian just has very complex onsets....). I should say that it's the nucleus that acts, among other things, as the primary bearer of syllable stress (in langs that have it), the loudest part of the syllable, and ususally the longest part of the syllable. The other parts are more or less just decorations :).
Re: Loooooooooooooooooooong answer, part XXXVI
Date: 2005-03-13 01:13 pm (UTC)When it comes to writing down a language in a system that represents the language's phonology (not like, e.g. Chinese writing), you often try to give a separate graphical symbol to each different sound you find in the language (I'll pass over the subtleties of this), and you'll probably find that even if the language allows the same (presumably vocoid) sound to occur both in the nucleus and the onset (or coda) of syllables, you'll probably give them different symbols, or in some other way help to identify the nucleus. In fact, even the IPA tends to do this.
So, for example, in Russian we have a sound which, when it occurs in the nucleus of a syllable is written "и" and when it occurs elsewhere, is written with the symbol "й". In German, for the same purposes, you use "i" and "j". In English, you use "i" and "y" (Note: the converse isn't true: in English, the letters "i" and "y" have other uses. I only give English as an example because people reading here are familiar with it - it's a terrible example of my point). Sometimes, a vocoid used in an onset or coda position is called a "semi-vowel", and even imagined to be a separate sound from when it's used as a nucleus (they can sound rather different). As I mentioned the IPA does this too: it uses the pairs [i]/[j], [u]/[w], [y]/[ɥ] to distinguish nucleus/non-nucleus roles for some vowels.
At this stage we have a writing system, and we have the conventional distinction between "vowel letters" and "consonant letters". Stereotypically, the former refer to symbols that represent vocoid sounds in syllable-nucleus positions, and the latter refer to contoid sounds in the margins (onset & coda) of syllables. As for the other two possible combinations, there's no standard way in traditional grammar to describe them (maybe in Sanskrit grammar?): contoids in the syllable nucleus (like the second "r" in butter) might be called syllabic consonants, and vocoids in the margins of syllables can also be called "consonants", though if they're of high sonority (like [i] aka [j]), they might be called "semi-vowels".
In English, traditionally we say that "a" "e" "i" "o" and "u" are "the vowels" meaning that they're "the vowel letters". We could include "y" too, but you have to remember that usage of English letters isn't consistent: we use "y" in "yoghurt" (representing a vocoid - the same sound as "i" in "machine", in a "consonantal" (i.e. onset) position), in "my" (representing two vocoids ([a]+[i] in IPA) in a nucleus ("vowel") position), we use "l", which we call a "consonant", in the word "little" twice, first of all representing a vocoid in onset ("consonant"), second representing the same vocoid in nucleus ("vowel") position. This second use has another oddity, since we use the letter "e" after the second "l" in "little" to (perhaps) indicate that "l" is acting as a syllable nucleus ("vowel"), so that "e" has no sound or syllabic role of its own here. However, since the letter "e" is usually used to represent any of a number of vocoid sounds in nucleus position ("vowel"), we call it (the letter!) a "vowel", and if asked "what's the second vowel in 'little'?", most non-linguists would respond with the answer "e". Behind all this subterfuge is the traditionally belief (expressed in other ways) that English doesn't have syllabic consonants (perhaps because Latin didn't).
So, you can see where the confusion comes from! - even in Russian, where things are slightly better behaved. What it's hard to see is: what do people mean when they say "vowel" or "consonant"? If they're referring to a division of graphical symbols into two classes, then they're talking about a convention of the analysis of writing, something with no simple correspondence to the workings of the language's phonology. That's not so bad, but the real trouble arise when people use orthographic notions of "vowel" and "consonant" to ask questions about the phonology of a language.
Part ye Last, in wch he Answers the Question.
Date: 2005-03-13 01:23 pm (UTC)Basically, the division of letters into "consonants" and "vowels" allow us to come up with simpler rules. Which, however, break down when people ask difficult questions. :)
Sorry it's so long. This issue comes up every now and again in various communities, and I thought I'd write a full answer to it, so that I can refer to it in future :). Hopefully I didn't stay from the point too much (some chance).
Phew!
Date: 2005-03-13 01:24 pm (UTC)/me falls over.
no subject
Date: 2005-03-13 01:29 pm (UTC)no subject
Date: 2005-03-13 07:33 pm (UTC)After spending some effort learning elementary Welsh, I just don't worry about orthography arguments any more. Learn how the language sounds. Figure out how to read and write with the rules they use, and be happy.
Is "y" (or "w" for that matter) considered vowel or a consonant, open or occluded? It doesn't matter. They're symbols with that stand for several kinds of sounds. We use them in several different ways, and millions of native speakers have already figured it out. So can learners.