Why isn't the letter 'R' a vowel--seriously?

[Ms2001]

*Originally posted by leenmi *
The letter “r” can’t complete a syllable.

Sure it can: grr.

[leenmi]

see, I knew someone would post word without vowels. Can you find an english word without vowels that isn’t onomonopaeic, or however you spell that thang?

Websters also states: “Apparently there are many, especially among writers of textbooks on phonics and teachers in the elementary schools, who are of the opinion that there can be no syllable without a vowel. Every dictionary in line of succession from, and including Noah Webster’s original dictionary of 1828 shows vowelless syllables.” p. 32a of my 1986 edition.

Okay, maybe I shouldn’t say vowels are needed to complete a syllable. Words like “grr” have no vowels. True. But “r” is still not a vowel. Grr is a word composed of consonants.

[MikeS]

Back when I took freshman linguistics, they taught us that the final syllables in words like “river”, “button”, “column”, and “little” were syllabic consonants. Here’s a good explanation of the phenomenon (scroll down to the last section.) Essentially, these consonants occur whenever you have an unstressed vowel followed by an /n/, /l/, /m/, or /r/ sound: for example, compare the words “batten” (stress on the first syllable, syllabic /n/) and “baton” (stress on the second syllable; the /o/ sound is obvious.) Or even better, pronounce the name of the Louisiana capital “Baton Rouge” the way the (English-speaking) locals do.

[Dragonblink]

Dang, I just missed getting to explain syllabic consonants.

The word “bird” is also an example of a syllabic r – when you say it, it’s just like you’re taking brrr (as in ‘I’m cold’) and throwing a d on the end. No true vowels in it at all, but the r acts as a vowel.

[Achernar]

*Originally posted by leenmi *
In English words (and I’m sure someone will come up with an obscure exception), a syllable can not be formed without one of the following letters: a e i o u y. (That’s why it’s a e i o u and sometimes y. Y is sometimes needed to complete a syllable. Such as in syllable. It’s not a vowel in the word Yellow.)

Okay, so atlatl is probably pretty obscure, but what about chasm and rhythm?

[jwest77]

Thanks for pointing out the syllabic consonants. It helps, but I still don’t get why it isn’t just called a vowel. I know I am starting to sound like a broken record, but based on the definitions, it acts like a vowel. I don’t understand why we have to come up with fancy names like syllabic consonants. All the definitions seem to start with the assumption that it is a consonant, and there are only 5 vowels.

When I say:

Hay
Her
Hi
Ho

My mouth, tongue, etc. are doing similar things. These are also very different from saying:

Hat
Had
Half
Has

Not only are these four different from the first four, they are also very different from each other.

What makes “her” so different from “hay” that it should belong in the second set?

[bordelond]

OK, you asked for it.

Jwest77, I’d suggest submitting your concerns as a query to the free Ask A Linguist service at Eastern Michigan University’s Linguist List website. Renowned linguists from all over the world will gladly take on and discuss at length questions just like yours.
Review the list of names at the bottom of the link. They are truly the heavy hitters in the contemporary world of academic linguistics. You may even e-mail many of them directly, if you wish – and most of them will answer you personally.

Actually, Ask A Linguist has dealt with your issue before. What follows is a sampling.

Larry Trask (University of Sussex) weighs in:

Q: How many phonemes do “star” and “fork” have? I thought 4 each. Is that right? How about “tern”? Would that be 3 and “ladder” 4, because “er” is a pure vowel. I would love to have the confusion about the r controlled
vowels cleared up.

**A: There are no absolute answers to these questions: the answers are different depending on the accent being described, and in some cases also on our analytical preferences.

In a rhotic accent (an accent that permits /r/ to occur in any position), the words ‘star’ and ‘fork’ have four phonemes each. But, in a non-rhotic accent (an accent that permits /r/ to occur only before a vowel), these words have only three phonemes each.

However, the word ‘tern’ is more complicated. In a non-rhotic accent, it has three phonemes. But, in a rhotic accent, the decision comes down to our preference in analyzing the r-colored vowel. If we decide to analyze this as a sequence of a vowel plus /r/, then ‘tern’ has four phonemes. However, if we decide instead to analyze it as a single r-colored vowel at the phonemic level, then the word has only three phonemes. Analytical tastes differ here.

The same happens with ‘ladder’: four phonemes or five.**

Trask on the difference between consonmants and vowels (cached on Google - search “syllabic”, “consonant”):

**It depends on whether you’re talking about phonetics or phonology.

In phonetics, a consonant (or contoid) is defined as a speech sound whose articulation involves noticeable obstruction of the airstream, while a vowel (or vocoid) involves no such obstruction. The grey area here is the approximants, which involve minimal obstruction, and especially the glides, which involve no more obstruction than some vowels.

In phonology, a vowel is typically defined as a sound which can be a syllabic nucleus, and a consonant as one that can’t. This makes glides consonants, and typically also the other approximants, but the grey area is that, in some languages, approximants, liquids, and sometimes even fricatives can be syllabic nuclei. This is the problem of ‘syllabic consonants’.**

James Fidelholtz (Benemerita Universidad Autonoma de Puebla, Mexico) addresses another related question:

Q: The folks who settled in the Appalachian Mountains of the Eastern United
States and migrated across the mountainous regions of the southern
states and into Texas and Oklahoma added an ‘R’ to many words that is
not in its spelling. Examples: holler = hollow (valley), winder =
window, idear = idea, warsh = wash and Etter = Etta. John Kennady said
‘Cuber’ and Paul McCartney (England) calls his wife Linder. So it is
not restricted to the mountain regions. My question is how did this
‘intrusive R’ originate? I have had explanations of: A lazy
tongue… Being unable to read (how a word should sound)… It
came from Northern Ireland… It came from the Elizabethan period of
England… It came from Cockney talk. Does anyone have a
professional answer to this? Many Thanks!

**A: The English ‘r’ is a rather complex sound phonetically, and, for
example, is one of the last sounds mastered by children acquiring the
language. (my ital)

This also means that it is more subject to changes in its
pronunciation over time than many other sounds. Indeed, in many
positions in the word (eg, at the end of the word), the ‘r’ has lost its
peculiar ‘retroflexion’ (that is, the curling back of the tongue) in many
different varieties of English, and, in at least some cases, apparently
independently. This has happened, among other places, in British
standard English, New England English, Southern (United) States English,
etc.

Now, notice that I said ‘in many positions’. Between vowels, the
‘r’ is never lost, so far as I know, in any variety of English. So
everybody everywhere pronounces ‘sorrow’ with a retroflex (normal) ‘r’.
However, the vowels surrounding the ‘r’ do not (in most varieties) have
to be in the same word. So, again in nearly all the varieties which
sometimes eliminate or change ‘r’, we would find, for example:

{transcription note: ‘v’ is the vowel sound of ‘but(t)’; ‘@’ is
the shwa, the last sound of ‘sofa’; ‘r’ is the retroflex (sulcal)
pronunciation typical of most varieties of English}

'butter is ...' [bvt@riz...]
'... butter.'   [...bvt@]

Thus, ‘butter is butter’: [bvt@rizbvt@], and the ‘r’ shows up in
one case of the word, and doesn’t in the other case.

Before going on, I should clarify that a good Midwesterner like
me actually pronounces the ‘r’ as a sort of ‘r-colored vowel’, what we
would call a ‘syllabic r’, in many contexts (just the ones where eg New
Yorkers might not have any r coloring). This we often just transcribe
with ‘r’, although technically we should indicate that it is syllabic
with some sort of mark, usually a small vertical line below the ‘r’ or
sometimes with a small circle below the ‘r’–this is hard to do in e-mail
for general purposes, so I will limit myself to ‘r’ for the (syllabic or
not) ‘full’ pronunciation of the ‘r’, and ‘@’ for the non-r-colored vowel
pronunciation that you asked about.

OK so far? Well, many varieties (like British English) nearly
ALWAYS ‘discolor’ the ‘r’ (make it into a simple shwa vowel sound) when
it is eg at the end of the word &/or before another consonant; but when
the following word begins with a vowel, the ‘r’ shows up. However, there
are many words like ‘sofa’, ‘Cuba’, et cetera, that historically never
had an ‘r’. But when pronounced in isolation by a speaker of one of
these varieties, ‘sofa’ rhymes exactly with ‘gopher’. So we would get:

I saw a sofa		[...@sof@]
I saw a gopher		[...@gof@].

Now, in many of these varieties (and probably also in the first
historical stages of eg British cultured speech, called Received
Pronunciation), we would still find a distinction between these two
words, eg in the following context:

A sofa is ... (over there)	[@sof@iz...]
A gopher is ...			[@gof@riz...].

But many speakers are insecure in their use of the language. They now
know that [gof@] must take an ‘r’ at the end in the second context, and
if they leave it off, then they may get corrected by their teacher, their
boss, etc. So, to be safe, they stick in an ‘r’ in the same context in
‘sofa’, giving [@sof@riz …]. In some varieties, such as Received
Pronunciation, this insecurity then became generalized, and now even the
most cultured Brits say [@sof@riz…], and this is called a ‘linking r’.
It appears even where historically there was never an ‘r’ in the word.

It also should be mentioned that the social value of dropping r’s
varies according to the region: in New York City, for example, it is
relatively ‘low class’ to drop too many r’s, while in many areas of
Southern United States English, it is relatively ‘low class’ to drop too
few r’s. In the areas where dropping is relatively ‘bad’, speakers who
wish to appear more refined often add r’s at the end of words like ‘sofa’
and ‘Cuba’, and in some cases, this becomes later the normal way to
pronounce such words. That is why Kennedy said ‘Cuber’ (especially when
the word was followed by another word beginning with a vowel).

Others may give you a slightly different perspective on this, and
some of my actual ‘historical’ facts might be slightly off, but this is
the general outline of how these different pronunciation arose (or at
least might have arisen).**

…

So after all that, Jwest77, the answer to your question is this: “R” is not, in fact, always considered a consonant. Sometimes “r” is considered a consonant. Other times it’s considered a vowel. And other times, it’s something in between.

[bordelond]

*Originally posted by jwest77 *
**Websters Third New International Dictionary (unabridged):

Vowel: one of a class of speech sounds in the articulation of which the oral part of the breath channel is not blocked and is not constricted enough to cause audible friction. Examples include o in the English “hot” and ü in the German word fünf.

Consonant: one of a class of speech sounds characterized by constriction or closure at one or more points in the breath channel.**

Keep in mind that these definitions are gross simplifications, and actually fall apart under academic scrutiny.

[Dogface]

“r” is neither a vowel nor a consonant, it is a LETTER! Letters are neither vowels nor consonants. SOUNDS are vowels, consonants, etc.

In some languages and some dialects of English, the sound often represented by “r” is a vowel. Some pronunciations of “bird” actually are /brd/. There is nothing inherently “consonant-like” about the sound of the rhotic “r”.

[bordelond]

*Originally posted by Dogface *
"r" is neither a vowel nor a consonant, it is a LETTER! Letters are neither vowels nor consonants. SOUNDS are vowels, consonants, etc.

You know … I overlooked this, but that is really a very important point.

[jwest77]

Bordelond:

Thanks for the great resource. I will check it out.

[AHunter3]

Letters are or are not vowels mostly on the basis of historical assignment. Letters do not have a one-to-one correspondence with sounds – not between different languages, and not within the same language in many cases and certainly not in the case of English.

The International Phonetic Alphabet has a symbol which looks like the schwa (upside down lower case “e”) with a tail on the top of it akin to the horizontal top portion of the lower case “r”.

This symbol represents the vowel sound in “word”, “bird”, “fur”, and “herd”.

And yes, it is essentially the same sound, produced in the same manner, as the “r” in “break” or “trill” or (yes) “river”, just drawn out and utilized as a vowel.

Seriously?
[IPAfont]
s´rt¨nlí
[/IPAfont]

[bordelond]

originally posted by AHunter3
**The International Phonetic Alphabet has a symbol which looks like the schwa (upside down lower case “e”) with a tail on the top of it akin to the horizontal top portion of the lower case “r”.

This symbol represents the vowel sound in “word”, “bird”, “fur”, and “herd”.

And yes, it is essentially the same sound, produced in the same manner, as the “r” in “break” or “trill” or (yes) “river”, just drawn out and utilized as a vowel.**

Just so. If I may pick a very slight and pedantic nit: the “turned schwa with r-hook” vowel symbol you mention is not the same symbol as that used for American English consonantal /r/. Instead, a “turned r” symbol is used for /r/ as a consonant, as in “red”. Also, you will note below that some phoneticists posit a separate symbol for the “er” sound in stressed syllables, using a “turned epsilon with r-hook” instead of a “turned schwa with r-hook”.

UCLA phoneticist Peter Ladefoged, in his book The Sounds of the World’s Languages, notes that the relationship between the “r” in “red” and the “er” vowel sound in “herd” is analagous with the relationship between the “y” in “yes” and the “ee” vowel sound in “need” (p. 325).

Somwhat related is this interesting seven-year-old discussion captured on a University of Vienna online log archive, in which Ladefoged’s work is referenced:

Q: If /u/ and /v/ were different phonemes in Latin, then why were they written the same, especially given that the model for the Roman alphabet, the Etruscan, did use two different signs.

A1: Here you seem to be getting into deeper water. A consonant and a vowel cannot, in my book, be “the same phoneme” (Consider “tweet” and “to eat” in English, for example. Is this not a minimal pair?)

**A2: It depends on one’s definition of “minimal pair”. “Tweet” is one syllable, “to eat” two. If a “syllable break” is counted as a phoneme somehow, the pair is no longer minimal ('twit ~ tu’it).

In a language like Spanish, that is usually not considered to have a separate /w/ phoneme, and [w] are indeed allophones (syllabic and non-syllabic) of the phoneme /u/. For instance:

"aun"      /'aun/               ['awn]
"au'n"     /a-'un/              [a'un]
"gru'a"    /'gru-a/             ['grua]
"guante"   /'guan-te/           ['gwante]

Likewise, most descriptions of English don’t consider [n] in “nose” and syllabic [n-] in “given” to be two separate phonemes.

On the other hand, introducing /w/ into the phonological description of Spanish would be defendable, as would be the introduction of a separate symbol for English syllabic “n”. Indeed, most transcriptions I have seen use separate symbols for consonantal “r” and syllabic “r” in those varieties of English that have the latter sound. Ladefoged, for instance, uses /(E*r)/ “turned epsilon with r-hook” for the vowel sound in AE “fur”, and /@r/ or /(@r)/ “schwa with r-hook” for the final sound in AE “father”. Transcribing them /r:/ and /r/ would, in my opinion, not be unreasonable.

In conclusion, “vowel” and “consonant” are not such clear cut concepts, and there is plenty of room for (allophonic) overlap. This is probably related to the fact that “syllable” is no clear cut concept either (To mention Ladefoged again: “Syllable: A unit of speech for which there is no satisfactory definition”). Whether it is the overlaps between /i/~/j/, /u/~/w/ etc., that prevent a satisfactory definition of “syllable”, or that it’s the lack of a definition for “syllable” that prevents us from deciding whether something is an /i/ or a /j/, a /u/ or a /w/, I don’t know.**

[bordelond]

*Originally posted by bordelond *
UCLA phoneticist Peter Ladefoged, in his book The Sounds of the World’s Languages, notes that the relationship between the “r” in “red” and the “er” vowel sound in “herd” is analagous with the relationship between the “y” in “yes” and the “ee” vowel sound in “need” (p. 325).

I screwed up the reference a bit here. The section paraphrased is actually on page 323 of the paperback edition.

[AHunter3]

ies, ouell said!

[saramamalana]

I don’t really have something to add, just that reading this thread was very satisfying. Earlier tonight I had fished out the notebook from one of my favorite classes, Linguistics of Uncommonly Taught Languages: Polish and it was fun to look over my notes. Reading this whole discussion was like a little refresher course.

[hijack]For any of you that know IPA, once you learned it did you find yourself writing random words and friend’s names using it? A friend and I were such dorks that for a while we’d pass notes writing in IPA. :D[/hijack]

[AHunter3]

Are you kidding? I created my own IPA font (back in the Macintosh System 4 days when bitmap fonts ruled the screen) so I could type in IPA.

Such a cool idea. One symbol per sound, one sound per symbol.

[Dreaming_of_Maria_Callas]

Sanskrit had vocalic R in addition to normal, consonantal R.

But even more interesting is that Sanskrit had vocalic L, as well. I don’t know any other Indo-European language that formally contained that sound. If you say the word “kleptomaniac” really fast, that’s what vocalic L is like. Indeed, the only common Sanskrit word that contains the sound is the verb root “klp”.

UnuMondo