Ravens are non-rhotic

« previous post | next post »

In the month of Poe's death (in October 1849), and of course the month of Hallowe'en, BBC television showed a program called "Edgar Allan Poe: Love, Death and Women". As I watched it, and listened to the speaker of standard American English that they used for the readings from his poetry (as so many radio and TV programs have done before), reminded me that I have always thought the dialect Poe had in mind when he wrote The Raven couldn't possibly have been one of the rhotic ones. It has to be one of the non-rhotic New England accents. That is, I don't think anybody who pronounces nevermore with an r sound at the end is reading The Raven correctly.

The sound made by ravens and crows is not [kɔɹ], like a typical Midwesterner saying core; it is much more like [kɔ:], reminiscent of a speaker with an Oxford accent saying the word (rhyming it with the educated Southern British pronunciation of law). The repeated nevermore has to sound like that, or the poem makes no sense as far as I can see. The raven's invariant repetition of "Nevermore!" at the end of each stanza has to sound like a raven cawing. Ravens don't have postvocalic r.

[Added later: Having expressed this opinion, let me now add very humbly that (1) I am not an expert on poetry (though I know enough to know that Poe's essay "The Philosophy of Composition" doesn't settle the rhoticity question); (2) I am not an expert on the details of Poe's native dialect (whatever it may have been, rhotic or non-rhotic — though he does seem to have had early experience of at least three places where the local accents are mostly non-rhotic); and (3) I am not an expert on corvids either (I have heard big black carrion-nibbling birds going "caw" without an r on three continents, but who knows, maybe some of the birds I took for ravens were crows or something else). What I do consider myself expert at, though, is expressing an opinion that will instantly get fifty or sixty people tussling with each other to disagree with me, or agree with me, or squabble wildly with each other. Read on and see… Mission. F***ing. Accomplished!]



82 Comments

  1. Morten Jonsson said,

    October 26, 2010 @ 5:00 pm

    Non-rhotic, yes, but why New England? Poe was a Southerner.

  2. Ray Girvan said,

    October 26, 2010 @ 5:01 pm

    On the other hand, there's Terry the Talking Raven, who manages a nicely rhotic "Aarrr" (0:40-1:26). I guess it depends on who taught Poe's raven to say "Nevermore".

  3. Peter Taylor said,

    October 26, 2010 @ 5:03 pm

    Do all ravens have the same accent?

  4. HP said,

    October 26, 2010 @ 5:09 pm

    Poe's own speech was a cultivated southern dialect (raised in Virginia in a wealthy family, studied at UVA and West Point, and spent most of his life in Baltimore). If I'm not mistaken, there are a number of non-rhotic English dialects to choose from in that part of the country, more so in the 19th c. than now. I can easily hear The Raven in a Chesapeake or Tidewater dialect (although I always read it in my native upper Midwest dialect). But a Brahmin or New England dialect seems wrong. (And Poe in a BrE dialect always sounds jarring to me.)

    I bet you anything, though, that there are Poe scholars — some of whom read LL — who can reconstruct Poe's speech in great detail and show you the evidence. And I'm always glad to be wrong.

    [(myl) Wikipedia says that Poe was born in Boston, moved to Virginia when he was two, went to Britain when he was six, and attended schools in Scotland and London before moving back to Virginia when he was 11. He moved to Boston when he was 18.]

  5. AJD said,

    October 26, 2010 @ 5:17 pm

    I don't really buy this argument—ravens can't say "neverm" either, so why should it matter whether or not they can say postvocalic r?

    James Thurber puts it this way: "It just happens that 'Arrk' is what ravens say. I have never known a raven that said anything but 'Arrk.'"

  6. Rubrick said,

    October 26, 2010 @ 5:25 pm

    I'm with AJD. It never even occurred to me that "nevermore" was supposed to sound anything like a raven's call (which it certainly doesn't, in any accent).

  7. Andrej Bjelakovic said,

    October 26, 2010 @ 5:29 pm

    In mid-19th century virtually all East Coast accents were non-rhotic, right? Richmond, as well as Baltimore etc.

  8. Kaviani said,

    October 26, 2010 @ 5:53 pm

    @ Peter Taylor – According to this, probably not.

    http://arstechnica.com/science/news/2009/10/birds-change-their-tune-to-adapt-to-life-in-the-city.ars

  9. ClockwerkMao said,

    October 26, 2010 @ 5:55 pm

    I expect it was a Ballmer raven, hon.

  10. Mark P said,

    October 26, 2010 @ 5:55 pm

    I have observed that at least some birds have regional accents. There is a bird that lives in my area that has a call very much like "Tweet." About 100 miles south, the call is something like "Ta-weet."

    I don't know whether crows have the same calls as ravens, but our crows' "caw" doesn't really sound that much like "caw." They also make a variety of odd gurgling and clicking sounds. A murder of crows in our back yard (gurgling and clicking) makes me think of Hitchcock's "The Birds." It's fairly disturbing, and I think a raven sitting in a study saying pretty much anything in pretty much any accent would be disturbing.

  11. michael farris said,

    October 26, 2010 @ 6:01 pm

    I could easily believe that Poe was non-rhotic and intended 'nevermore' to rhyme with 'caw' (my representation of a corvid vocalization). Or not, until we have time travel there's no way to know for certain.

    On the other hand, reading the poem aloud with a non-rhotic american accent would probably sound more than bizarre. And reading the poem in rhotic SAE with a non-rhotic 'nevermore' (southern or new england) would sound hilarious. Especially since my first that is that non-rhotic southern more is homophonous with 'moe' or 'mow' and a crow solemnly intoning 'nevuh moe' is comic gold.

  12. Richard said,

    October 26, 2010 @ 6:02 pm

    Ravens may not naturally have post-vocalic r. But they can certainly imitate it. I've uploaded the spectogram of Terry saying "arr" here. Note the clear change between the vowel and the r. The Linguistic Atlas of the US lists four possible types of post-vocalic r:1. Retroflex /r/ 2. Weakly retroflex /r/ 3. Vocalic /r/, also called unconstricted /r/ 4. Deletion of /r/
    Discounting the last one, that's a few different types that the raven could be reproducing from whatever demonic agent taught it to speak. Given the general ability for ravens to imitate speech fairly accurately, I don't doubt that the Raven would be unable to pronounce 'nevermore' accurately.

    What I would like to know is if the thrusting out of the syrinx that we seen in the Terry video has something to do with switching which side of the syrinx the vocalisation comes from between the vowel and the r. (Birds famously have two voices.) However, the syrinx isn't the only thing that is important here: the tongue is used in humans to lower the third formant, generally, in rhotic sounds (if I read The Handbook of Phonetic Sciences right), and we've seen elsewhere that the tongue is more important than we might think in vocal imitation in birds.

  13. Drago said,

    October 26, 2010 @ 6:06 pm

    Poe himself wrote about his creative process in developing "The Raven" in his "The Philosophy of Composition" and he suggests that the term itself came before the choice of the raven as a harbinger of doom. See here.

    The relevant passage:

    The question now arose as to the character of the word. Having made up my mind to a refrain, the division of the poem into stanzas was of course a corollary, the refrain forming the close to each stanza. That such a close, to have force, must be sonorous and susceptible of protracted emphasis, admitted no doubt, and these considerations inevitably led me to the long o as the most sonorous vowel in connection with r as the most producible consonant.

    The sound of the refrain being thus determined, it became necessary to select a word embodying this sound, and at the same time in the fullest possible keeping with that melancholy which I had pre-determined as the tone of the poem. In such a search it would have been absolutely impossible to overlook the word "Nevermore." In fact it was the very first which presented itself.

    The next desideratum was a pretext for the continuous use of the one word "nevermore." In observing the difficulty which I had at once found in inventing a sufficiently plausible reason for its continuous repetition, I did not fail to perceive that this difficulty arose solely from the preassumption that the word was to be so continuously or monotonously spoken by a human being- I did not fail to perceive, in short, that the difficulty lay in the reconciliation of this monotony with the exercise of reason on the part of the creature repeating the word. Here, then, immediately arose the idea of a non-reasoning creature capable of speech, and very naturally, a parrot, in the first instance, suggested itself, but was superseded forthwith by a Raven as equally capable of speech, and infinitely more in keeping with the intended tone.

    This suggests, I think, that the "r" is probably rhotic (and that ravens were thought to be capable of speech, or maybe they are and it's just my ignorance).

  14. Geraint Jennings said,

    October 26, 2010 @ 6:18 pm

    Growing up, I knew that ravens made the sound "nevermore" before I'd ever heard of Poe. I read Joan Aiken's childrens books about Arabel and her raven, as well as experiencing them being read on television (Mortimer, the raven, communicates almost entirely through expressive variations of "Nevermore").

  15. Nathan said,

    October 26, 2010 @ 6:21 pm

    In Poe's The Philosophy of Composition, he claims he chose the word nevermore because he saw "the long o as the most sonorous vowel in connection with r as the most producible consonant." He also says he first thought of a parrot, but this "was superseded forthwith by a Raven as equally capable of speech, and infinitely more in keeping with the intended tone."

    Of course a lot of people without phonetic training consider even a non-rhotic r a consonant, but I find reason here to doubt Dr. Pullum's interpretation.

  16. Ken Brown said,

    October 26, 2010 @ 6:41 pm

    Ravens can certainly imitate the sounds of human speech. Parrot-fashion, as that long and interesting quote from Poe confirms. Otherr crows can as well, famously jackdaws.

    British birdwatchers traditionally write down the most common raven's call as "prrrrk!" ordinary crows are thought to say "caw". Which rhymes with "law" for all of us, and is identical to "cor" for almost all of us, educated or not. Rlessness is triumphing on this side of the Atlantic – the barman in the pub I am posting this from is a non-rhotic Dubliner.

    As this is Language log, not Ornitho log, I can't resist pointing out that "crow" and "corvus" and "corax" and probably "jackdaw" are named for the sounds. And I'm pretty sure that the conceit of Poe's poem is that the raven had learned to say the human word. So the question is not whether or not Poe was rhotic but that the fictional previous keeper pf the raven was.

  17. Mark P said,

    October 26, 2010 @ 6:49 pm

    In my experience with a Southern non-rhotic accent (east Georgia below the fall line), "more" is pronounced more like "moe-ah" than "moe."

  18. F said,

    October 26, 2010 @ 6:49 pm

    How about a cross-lingual study of crow onomatopoeia? In Russian, crows say "кар" [kar] — the post-vocalic r is there, no ambiguity about it.

  19. Brett said,

    October 26, 2010 @ 7:08 pm

    I remember the Poe scholar Dennis Eddings telling me that if you made the Raven sound like an actual bird, it totally ruined the poem. This was followed by his wife saying (in a parrot-like voice), "Nevermore! Nevermore! Polly wanna' cracker! Nevermore!"

  20. Lee Stweasel said,

    October 26, 2010 @ 7:32 pm

    The poem's always made perfect sense to me, an American with rhotic, newscaster-type accent. And it seems to me that each word rhyming with "Nevermore" makes a true, perfect rhyme. Would that hold up to a change in pronuncation?

  21. Matthew Kehrt said,

    October 26, 2010 @ 7:57 pm

    On a related note, the story I heard was that ravens say "cras" in Latin, which means "tomorrow". An educated reader of Poe's day would expect the raven to say "tomorrow", which makes "nevermore" much more shocking.

    I don't know how true this is, but I know that "cras" is actually Latin for "tomorrow".

  22. Mr Punch said,

    October 26, 2010 @ 8:00 pm

    Pretty much all East Coast accents outside the Philadelphia area are non-rhotic today, and have been for a long time.

  23. Lazar said,

    October 26, 2010 @ 8:13 pm

    What Nathan said. I've always taken that passage in "The Philosophy of Composition" as an indication that Poe was rhotic.

  24. Bill Taylor said,

    October 26, 2010 @ 8:21 pm

    And then there's the wonderful Simpsons adaptation. James Earl Jones, the narrator, would surely have had an interesting way with "Nevermore," but the part of the raven is spoken by Bart (or rather a very Bartesque raven with Bart's voice). As I recall he has a pretty clear r on the end. The poet whom the raven torments is, of course, played by Homer (H. Simpson, that is, not the actual poet Homer).

  25. tablogloid said,

    October 26, 2010 @ 8:22 pm

    Although ravens resemble crows visually, the sounds they make are distinctly different. Ravens do not caw like a crow.

  26. GeorgeW said,

    October 26, 2010 @ 8:34 pm

    My South Carolina 'granmotha' (born early 20th century) was definitely non-rhotic.

  27. Ray Girvan said,

    October 26, 2010 @ 8:48 pm

    Just found this on YouTube: Ravenmania – Nevermore!. Irrelevant to Poe's accent, but …

  28. Adrian Morgan said,

    October 26, 2010 @ 9:08 pm

    What surprises me about people alleging that ravens say 'ark' is that I agree with them.

    What's surprising about that is that the ravens I know are Australian, and sound like this … http://www.anbg.gov.au/sounds/raven.mp3 … which is almost invariably transcribed as 'ark'. So I'd expect people presumably talking about Northern Hemisphere varieties to transcribe the calls differently.

  29. ignoramus said,

    October 26, 2010 @ 9:11 pm

    Many 'arvest moons ago there was a study of crows communicating in differing lands. East Anglian Crows and rooks did not respond to the California crow comments, thus I assume [bad to ass u me yep] that the tower raven would ignore requests from it's Baltimore "cosen" .

  30. Leonardo Boiko said,

    October 26, 2010 @ 9:18 pm

    I started browsing youtube for ravens and crows talking and I have to say I thought they were fakes. How can these birds produce such a wide variety of sounds, including near-perfect imitations of other animals and the human voice? And when will we have legal domesticated ravens so that I can have a pet and teach him to say “nevermore”?

  31. The Ridger said,

    October 26, 2010 @ 9:28 pm

    Indeed, AJD. When I walked past my first raven (last winter in Anchorage) and it called, it sounded nothing like a crow. I might have transcribed it as hrrrok, but arrrk does as well.

  32. John Chew said,

    October 26, 2010 @ 9:38 pm

    I'm raising my kids bilingually in Japanese and English, just like my parents did with me. This second time around, I'm noticing a few things I didn't pick up the first time. For instance, in Japanese, the sound a crow makes is カーカー ("kaakaa"), but the crows here in Toronto sound more like the [kɔ:] you mention. My sons just accept this as part of the weirdness of life, but on a recent solo trip to Tokyo, I was reminded that the crows there do actually go カーカー. Are there atlases of crow calls?

  33. Matt McIrvin said,

    October 26, 2010 @ 10:53 pm

    It sounds odd to claim that (almost) all US East Coast accents are non-rhotic, given the number of rhotic speakers in the region. Certainly the local accent here in Massachusetts is non-rhotic, but when I was growing up in northern Virginia, the people who did not have Southern accents were nearly all rhotic speakers, and, I'd say, an appreciable fraction of the Southern speakers were too.

  34. Clayton Burns said,

    October 26, 2010 @ 11:21 pm

    If we just examine the poem (without consulting any dictionaries of pronunciation or attempting to determine how Poe might have pronounced "nevermore" at various periods of his life), we can still establish with certainty that the raven pronounced the 'r' at the end of "Nevermore."

    The proof is in "morrow," "borrow," "sorrow," "sorrow," "unbroken," "broken," "sorrow," "unbroken," and "croaking" (croaking 'Nevermore'):

    "Doubting, dreaming dreams no mortal ever dared to dream before;

    But the silence was unbroken, and the darkness gave no token,

    And the only word there spoken was the whispered word, `Lenore!'"

    (We would also lose the full force of the reversal of "raven" in "never" if we failed to pronounce the final 'r' in "never.")

    Similarly, the reversal in "croaking" and "Nevermore" ('r' and vowel in the first word, and vowel and 'r' in the second) is foregrounded in the poem:

    If we read it out loud, we will become conscious of how "no mortal…before/ unbroken…token/ spoken…'Lenore!' forms a sequence in which it would be bizarre not to balance "unbroken" with the final 'r' in "Lenore."

    If you think in terms of what comes before and after /r/s in the poem, you will see that the final 'r' in "Nevermore" is impossible to avoid.

  35. Nijma said,

    October 27, 2010 @ 12:51 am

    From Michael Vanner's Encyclopedia of North American Birds:

    American Crow corvus brachyrhynchos "Its plumage is black and its call is a harsh caw.
    Chichuahuan Raven corvus cryptoleucus "Its call is a flat, drawn out craaaaak, higher pitched than that of the Common Raven."
    Common Raven corvus corax "Its call is a low, resonant craaak, deeper than that of the Chihuahuan Raven, but it also makes a variety of other noises including screams, whistles, and a melodious kloo-klok."

  36. Xmun said,

    October 27, 2010 @ 12:51 am

    It seems to me, having read GKP's post and all 34 comments, that neither the case for nor the case against has been proved. Is Poe's raven rhotic or non-rhotic? I don't know.

  37. Nijma said,

    October 27, 2010 @ 12:54 am

    The key to all of this is obviously Lenore. Find out how Poe pronounced "Lenore", and you will know what the raven said.

  38. Chad Nilep said,

    October 27, 2010 @ 1:24 am

    I was going to point to The Philosophy of Composition, but Drago and several others beat me to it. Since Poe had in mind a raven that could mimic human speech, a rhotic or non-rhotic accent is conceivable. A non-rhotic one might seem more emotionally compelling, but that's largely down to a matter of taste.

    Poe's description of his writing process and his decision to use a bird capable of mimicking human speech may actually have been post-hoc. See here:
    http://linganth.blogspot.com/2007_10_28_archive.html

  39. J. Goard said,

    October 27, 2010 @ 1:37 am

    What Clayton said.

    Poe is a master of sound patterns, if anybody is. The poem loses a lot without those "r"s.

  40. maidhc said,

    October 27, 2010 @ 1:39 am

    There's a interesting book about ravens, "Ravens in Winter" by Bernd Heinrich. He has many interesting observations about ravens — ravens will guide human hunters to deer so that they can get the offal.

    He built an aviary next to his office where he kept a couple of rescued ravens. They used to do a perfect imitation of a ringing phone. When he came rushing into his office to pick up the phone and got nothing but a dial tone, they would laugh at him.

    There's also a talking raven in "Barnaby Rudge".

    Here's a video of a talking raven:
    http://www.youtube.com/watch?v=rGeOpjxx2ho&feature=related

  41. Nijma said,

    October 27, 2010 @ 2:58 am

    The poem loses a lot without those "r"s.

    I have to admit, "Once upon a midnight dweawy, while I pondewed, weak and weawy" sounds a lot like Elmer Fudd.

    [Postvocalic r, you nitwit! The ones at the ends of the syllables! Neawly evewyone pwonounces them befowe and between vowels. What vawies is occuwwence following vowels. But I suspect you knew that and wewe just leading evewyone on, to get a wise out of them… —GKP]

  42. Joyce Melton said,

    October 27, 2010 @ 3:42 am

    A friend's daughter asked me what the difference between a raven and a crow was, so I took her to the zoo. In one cage sat two crows, smaller, shinier and with squared off tails compared to the raven in the next cage larger, dustier and with a pointed tale.

    Besides looking different, I told her, they sound different. Crows, I told her say "caw". Ravens say "Morning, Albert."

    And the bird, which I had met before, obliged by saying, "Morning, Albert." In a perfectly rhotic Western General American accent, too. :)

  43. Melodye said,

    October 27, 2010 @ 3:44 am

    @Clayton Burns @Nijma

    But would non-rhotic pronunciation have to be universal to every word r-word in the poem? Hypothetically, some of the r-words could take rhotic and some non-rhotic pronunciations; it could be a case of lexical diffusion, for instance, or the vocal idiosyncrasies of a man raised all over the English-speaking earth.

    Though — that might well destroy the rhyme scheme.

  44. Liz said,

    October 27, 2010 @ 7:46 am

    A PARROT? I will never be able to read the poem again in the same way. Doomy blackness replaced with a parrot?

    Goodness, the things I learn here! Crows with regional accents, yet.

  45. Neil said,

    October 27, 2010 @ 8:34 am

    "Missing. F***ing. Accomplished" indeed

    Check out this excellent comment in reply to someone complaining about the improper use of bacteria/ium in an article about giant viruses:
    http://arstechnica.com/science/news/2010/10/giant-virus-found-in-tiny-predator.ars?comments=1#comment-20945281

    Inspiring.

  46. Alen Mathewson said,

    October 27, 2010 @ 8:40 am

    I wonder if there is merit in looking at the reference in the poem to Lenore and tracking back to Poe's equally famous poem of that name which seems to require a postvocalic r, taking the end of the first verse as a guide:

    A dirge for the most lovely dead
    That ever died so young!
    And, Guy De Vere,
    Hast thou no tear?
    Weep now or nevermore!
    See, on yon drear
    And rigid bier,
    Low lies thy love Lenore!

    Perhaps Poe's time in decidedly rhotic Scotland, where ravens are seen in a very dark light (the old collective noun was an unkindness of ravens) and feature in many myths and legends, shaped his views on the speech patterns of ravens.

  47. AJD said,

    October 27, 2010 @ 8:56 am

    Mr Punch: Pretty much all East Coast accents outside the Philadelphia area were non-rhotic 60 years ago, but non-rhoticity has been on the decline since then; I've been told that it's almost gone in the South, though still hanging on to variable extents in New York, Providence, and Boston.

  48. Mary Sweeten said,

    October 27, 2010 @ 8:59 am

    Fish crows — a distinct variety that is common on the East Coast — don't say "caw": they have a double call that sounds like "Oh, no." Maybe that's what Poe had in mind.

  49. Mr. Fnortner said,

    October 27, 2010 @ 9:04 am

    For decent audio clips of raven and crow, check this site: http://www.shades-of-night.com/aviary/difs.html. I think Thurber is right.

  50. Dan T. said,

    October 27, 2010 @ 9:14 am

    That's So Raven!

  51. S. Norman said,

    October 27, 2010 @ 9:32 am

    Haha, have a discussion on the Internet and that's what you're going to get- conversation hijackers and tanget-tizers. Barring a PhD Only sign in, what can you do

  52. SimonMH said,

    October 27, 2010 @ 9:47 am

    I don't know if it is rhotic or not, although I do remember playing many games of backgammon in Poe's school (now a wine bar in Stoke Newington, London). The ravens there certainly had a non-rhotic accent.

    It's not much of a poem anyway. Poe is more highly regarded in France, where he had Baudelaire as a translator.

  53. richard howland-bolton said,

    October 27, 2010 @ 10:15 am

    In advance, sorry>

    http://howlandbolton.com/essays/read_more.php?sid=378

  54. Theophylact said,

    October 27, 2010 @ 10:45 am

    Pratchett fans will be familiar with Quoth the Raven, who serves as the mount for The Death of Rats, and who steadfastly refuses to use the "N-word"; but he's not in the least non-rhotic.

  55. Matt McIrvin said,

    October 27, 2010 @ 11:06 am

    I always hear Vincent Price's voice when I read the poem, and he was definitely a rhotic speaker, almost exaggeratedly so.

  56. ClockwerkMao said,

    October 27, 2010 @ 11:08 am

    Mr. Punch and AJD: Baltimorese is related to Philadelphian, and is also very rhotic. Has always been. It even leaks into adjacent areas: Frederick County Maryland has the same basic dialect (my old barber is a fantastic example), and I would expect that old-timers in much of Maryland between the mountains and the bay would also have it.

  57. Nijma said,

    October 27, 2010 @ 11:27 am

    "nitwit"??!? Why so mean, GKP? I'm crushed. And no, I don't speak linguistese, but I can google "postvocalic r".

    The non-rhotic accent I'm most familiar with is Boston, and the closest I can come to figuring out how they speak is that if a word has an r they drop it; if a word doesn't have an r they add one. (Vowels don't seem to figure in there anywhere.) In other words, "lobster" becomes "lobstaw", "tartar sauce" becomes "tawtaw sauce", and if they want to "go pawty" you don't know if they want to party or if they need a toilet (potty). On the other hand, something like "idea" becomes "idear", as I seem to remember from JFK's speeches.

    I'm not prejudiced against you persons of the non-rhotic persuasion either, at least I don't think I am. I even dated a non-rhotic guy once. :)

  58. Mr Fnortner said,

    October 27, 2010 @ 11:28 am

    @Mary Sweeten, I thought our fish crows were saying "uh-oh". Either way, what are they fretting over so?

  59. Tim Friese said,

    October 27, 2010 @ 12:29 pm

    @Clayton: you raise good points about the abundance of 'r's and Poe's skillful use of sounds. However, I'd like to toss out the possibility that he used a non-rhotic pronunciation but STILL deliberately took advantage of all of these 'r's (be they pronounced or silent). Here are two factors, and I admit that I'm reaching a bit:

    1) Like a number of posters have mentioned, Poe had a multi-dialectal background. He may have been non-rhotic but he knows how other people pronounce these words. Perhaps there is a trace of an 'r' in the brain from all these rhotic speakers even if there's not a trace on his lips?

    (A weak example of that is that I, an American, often mishear glottal stops in foreign languages as 't's, probably because I'm familiar with the British dialects with lots of glottals.)

    2) We need to remember the strange psychology (from the perspective of thousands of years of human speech) of hyper-literacy. The fact that two distinct sounds are written the same way leads many speakers to be oblivious to the difference, and despite homophony between two distinct words (writer / rider comes to mind) speakers claim to maintain a difference.

  60. Axel Svahn said,

    October 27, 2010 @ 1:00 pm

    There is an account of Poe being a non-rhotic speaker. The link is in my name.

    "Mabbott notes, “He [Poe] spoke with a slight Southern drawl. Hence he rhymed sister and vista, ha‘nted and enchanted. . . . Poe did not drop the final letter of words like hunting“ (Mabbott, Poems, 1969, p. xxv)."

  61. richard howland-bolton said,

    October 27, 2010 @ 1:08 pm

    @Tim F
    "and despite homophony between two distinct words (writer / rider comes to mind) speakers claim to maintain a difference."

    I'm pretty certain my British dialect does. In a related phenom, my ('Mercan) ex was a teacher, and after years of listening to her talk about it I still thought that there was a drug called *Ridalin: then one day I saw it in writing…

  62. The Ridger said,

    October 27, 2010 @ 1:28 pm

    @richard howland-bolton: I always maintained there was a difference, but I had to re-evaluate that when a friend asked me, out of the blue, had I seen "Ghost Rider" and it turned out she meant "The Ghost Writer". Context influences our perceptions much more than we think.

  63. Clayton Burns said,

    October 27, 2010 @ 1:41 pm

    Tim Friese: What you say about the "strange psychology" of "hyper-literacy" is interesting. We see in Blake's lyric the sequence "O Rose," "storm," "secret," and "destroy." Poe, despite his limitations as a poet (I would rather study "The Fall of the House of Usher"), would have been sensitive to this sequence in a way that the average reader would not be:

    1 O Rose, thou art sick!
    2 The invisible worm
    3 That flies in the night,
    4 In the howling storm,

    5 Has found out thy bed
    6 Of crimson joy:
    7 And his dark secret love
    8 Does thy life destroy.

    Online text copyright © 2009, Ian Lancashire (the Department of English) and the University of Toronto.

    The question that I have is whether we could implant in students something of the same sensitivity with an official database of 40 lyrics from 1600-1900 and "Macbeth." So that in phonetics we could input the grid of sound/sound symbolism sensitivity.

    At John Wells's phonetic blog ("prevalence," 27 October 2010), I have commented on that.

  64. KevinM said,

    October 27, 2010 @ 1:51 pm

    Now I have no idea what to do on "Talk Like A Raven Day." Arr(k?).

  65. Jim said,

    October 27, 2010 @ 2:16 pm

    1) You have never heard a crow make an r sound
    2) Some non-crows can make r sounds

    Therefore: all crows must be black….

  66. Xmun said,

    October 27, 2010 @ 2:36 pm

    richard howland-bolton

    Of course your British dialect maintains a distinction between "rider" and "writer". So does mine and everyone else's.

  67. Xmun said,

    October 27, 2010 @ 2:38 pm

    Oops. I wrote that before reading The Ridger's comment.

  68. Ellen K. said,

    October 27, 2010 @ 5:25 pm

    I think with rider and writer many of us Americans maintain a difference in our heads between the two, because of the difference between write and ride, in which the d and t are pronounced differently. That is, even though it's one sound, it's two phonemes, which we can tell apart in other words where it doesn't turn into a flap. We don't notice they are the same anymore than we notice they are the same as how some people say an r.

    There may also be a different in vowel length (ride definitely has a longer vowel the write for me), but I don't think that's consistent.

  69. John Cowan said,

    October 27, 2010 @ 6:23 pm

    GKP: Yes, and she got it, too.

    Nijma: Non-rhotic accents drop /r/ at the end of a syllable after the vowel ("post-vocalic") but not in the beginning. However, the /r/ reappears at the end of a word if the next word begins with a vowel ("linking /r/"), and sometimes /r/ appears in that context even if historically there was none ("intrusive /r/"). This is why JFK said Cuber if a vowel followed, but Cuba if not. (Idear is a special case; some people who are completely rhotic nevertheless use it instead of idea; it's really a separate pronunciation rather than an accent.)

    However, Bostonians don't in fact pronounce party and potty the same way, because they don't have the father-bother merger that makes these words rhyme in the rest of North America (but not anywhere else in Anglophonia). However, it can be difficult for other North Americans to hear the distinction they do make.

    Xmun: Some North Americans have no distinction between writer and rider; others do, but the distinction is in the preceding vowel.

  70. Jerry Friedman said,

    October 27, 2010 @ 7:03 pm

    @GKP: What's a nitrit?

    @maidhc: I too enjoyed Ravens in Winter. I think it was in that book that Heinrich repeated a naturalist's saying: If you're in the woods and hear a sound you don't recognize, it was probably a raven. (If you were within the raven's range.) Which is just to amplify Nijma's point that ravens don't make just one sound. I think I remember that over fifty have been described.

    I'm terrible at hearing human sounds in bird vocalizations. I don't think even chickadees and peewees "say their names". So though I've heard lots of ravens, don't ask me what they say.

    Regional variations in some birds' songs have been well documented, by the way, and they're known as "dialects".

  71. Jerry Friedman said,

    October 27, 2010 @ 7:06 pm

    I meant to recommend Heinrich's book The Mind of the Raven, too. (To allay the fears that some here may have: he doesn't claim they have language.)

  72. GeorgeW said,

    October 27, 2010 @ 8:43 pm

    @Ellen K.: "There may also be a different in vowel length (ride definitely has a longer vowel the write for me), but I don't think that's consistent."

    I think I do as well without the /-er/ prefix which causes the flap to occur. I am not so sure about rider/writer in unmonitored speech:

    1. She is a good rider.
    2. She is a good writer.

    How about in matter/madder, bitter/bidder, butter/budder?

  73. Ignoramus said,

    October 27, 2010 @ 9:27 pm

    rhotic -rhotacism
    My computer dictionary fails the test.
    On droppin' or over pronouncing the r to get ones idea across, is to be part of the acceptance by ones peer group. No wonder Rooks gather at night to chatter while crows will meet in clusters in the day when trying to scare away a chicken hawk and then go home alone, While the raven will keep the Beefeater hungry eating his beef.

    no wonder there be the saying "Birds of a feather flock together"
    to r or not to r

  74. Ignoramus said,

    October 27, 2010 @ 9:32 pm

    PS nobody mentions Poe's adopted father be a Scot, they be rather famous for rolling their "rrs".

  75. Nijma said,

    October 28, 2010 @ 10:35 am

    @John Cowen, thanks, I guess that leaves out r in consonant clusters. So I suppose a better approximation of the non-rhotic pronunciation of the poem would be "Once upon a midnight dreawy, while I pondewed, weak and weawy"…but now all I can hear in the poem is that voice of Elmer Fudd–curse the diabolical GKP for putting that voice in my head–fowevewmow!

  76. Ellen K. said,

    October 28, 2010 @ 11:24 am

    Ninma: People with non-rhotic accents do NOT turn r's into w's, and they do NOT drop an r that's between two vowels. An r between two vowels is pronounced as a beginning of the 2nd sylable.

  77. Ellen said,

    October 28, 2010 @ 12:16 pm

    P.S. Nijma (sorry for the typo above), you seem to be confusing non-rhotic accents with the speech of those who can't pronounce r's. Two different things.

  78. MJP said,

    October 28, 2010 @ 1:13 pm

    @Bill: from memory, that raven pronounces the word written "nevermore" as "eat my shorts". The first time, anyway…

  79. Julie said,

    October 29, 2010 @ 5:46 pm

    @Nijma: Try this:
    There aren't many r's affected, so I used the whole first verse.

    Once upon a midnight dreary, while I ponded weak and weary,
    Ova many a quaint and curious volume of forgotten loah,
    While I nodded, nearly napping, suddenly theah came a tapping,
    As of some one gently rapping, rapping at my chambeh doah.
    `'Tis some visiteh,' I mutted, `tapping at my chambeh doah –
    Only this, and nothing moah.'

    This is a very crude conversion, which does not consider the vowel quality, which is also different. But it will give you a starting point.

    Note that "weary," "dreary," and "rapping" are not affected. Those r's are at the beginning of syllables. In words like "there" it makes an extra syllable (or half-syllable), almost like "they-ah." In words with "er" endings, just drop the r. I put in an h so you can see it: "chambeh."

  80. Julie said,

    October 29, 2010 @ 6:15 pm

    Next question:

    Prophet!' said I, `thing of evil! – prophet still, if bird or devil!

    It looks as though "evil" and "devil" should rhyme; in fact, it's hard to read that line without rhyming them. They don't rhyme in my accent–does anyone know what Poe intended?

  81. J. Goard said,

    October 30, 2010 @ 1:17 am

    @Julie:

    It seems like Poe had no problem related words that are "slant-rhymed" in a typically rhymed position. Consider the last lines from the four stanzas of "The Bells", which follow this template:

    {From/To/In} the ____ and the ____ of the bells

    jingling/tinkling
    rhyming/chiming
    clamor/clangor
    moaning/groaning

    Two rhymes, and two slant rhymes with consonant changes. (I remind myself that Poe would have perceived the first vowel in "clamor" and "clangor" as the same, despite myself having a California shift in which "clamor" has the same vowel as "ran", but "clangor" has the same vowel as "rain".)

  82. Julie said,

    October 30, 2010 @ 3:29 am

    @J. Goard: I understand what you're saying. The assonance in those lines is perfect for me. I pronounce both "clamor" and "clangor" with "ran." But "evil" and "devil?"

RSS feed for comments on this post