Not So Fast with the Funny Fading Dialect Stuff

« previous post | next post »

This is a guest post by Josef Fruehwald, commenting on Daniel Nester, "The Sound of Philadelphia Fades Out", NYT 3/1/2014.

I might live in Edinburgh, Scotland now, but I haven't exactly left Philadelphia yet. When I moved here this past September, it was the first time in my life I ever lived outside the city limits of Philadelphia, and aside from my personal life, I have a fairly vested intellectual interest in the city's dialect. I just spent the last 3 years or so, together with William Labov, Ingrid Rosenfelder, Sue Sheehan, and many others, compiling decades worth of fieldwork into an electronic collection of dialectal information called the Philadelphia Neighborhood Corpus. I even wrote a dissertation on the topic(ish). Here’s a flashy bit of output from that research. Press play (at the bottom left), and watch 100 years of vowel sounds dance around.

Needless to say, my friends and family are keeping me up-to-date on what's going on in town, and this includes the New York Times Sunday Review piece by Daniel Nester called The Sound of Philadelphia Fades Out. In fact, one more person e-mailed it to me now as I was writing this. To be honest, my reaction to it has been more or less, "Eesh."

Nester's main complaint is that Hollywood films that portray Philadelphians rarely do so with accurate accents, and I totally agree. It either has to do with ignorance of film makers, that they don't know what Philadelphian sounds like, or the film makers worrying about the ignorance of the audience, that if the actors were using authentic Philadelphia accents, no one would understand what they were trying to portray.

But except for this scrap of common ground, I just don't like the piece very much. It starts out describing the accent like so:

No vowel escapes diphthongery, no hard consonant is safe from a mid-palate dent. Extra syllables pile up so as to avoid inconvenient tongue contact or mouth closure. If you forget to listen closely, the Philadelphia, or Filelfia, accent may sound like mumbled Mandarin without the tonal shifts.

Not an auspicious start. If you're not familiar with linguistic terminology, this is is all gobbledygook, but it does contain an interesting reversal of most folk dialect descriptions which has us putting in more syllables than usual, rather than dropping them out. I have no idea what it is about syllables as a unit of description and their presense or absence which fixates so many writers like Nester. "But wait," you say, "this is just an opinion piece, not science journalism! Maybe 'facts' and 'accuracy' don't matter." Ok, it sure would be nice to have a bit more of the science journalism treatment for linguistics, but even when it comes to the more touchy feely stuff, I've seen better. Take, for example, Samantha Melemed's piece which appeared in the Style and Soul section of the Philadelphia Inquirer two weeks ago called Translating Philly-ese. It's generally dialect-positive and doesn't resort to simply making stuff up. Or, there's the All Things Considered piece produced by Zack Seward called Dialects Changing, But Not Disappearing In Philadelphia, which I have to say is better than I could have hoped for when it comes to reporting on our recent Philadelphia studies (really, go listen to it).

On that point, the All Things Considered piece, which Nester linked to, is called Dialects Changing, But Not Disappearing In Philadelphia while Nester's is called The Sound of Philadelphia Fades Out. Well, both can't be accurate. The point me, Bill and Ingrid tried to make in our 2013 paper was that the Philadelphia dialect has started moving in a different direction than it was moving before. That means that some things are becoming a bit less distinctive, while others are becoming a bit more. For example, the word days has been raising and fronting over the past 100 years in Philadelphia, to sound a bit more like "deez", while day hasn't budged from having a mid, lax nucleus, creating a fairly distinct split between the two that is (as far as I know right now) unique to the dialect. Another fairly distinctive example is the merger of owl and Al (from which I got the name for my blog, Val Systems from "vowel systems"). It's an old feature of dialect, described all the way back in 1944 by Whitney Tucker, and it doesn't seem to be going anywhere.

It's actually really difficult to support a claim like Nester's, that the Philadelphia dialect is "fading out", with objective facts. It all depends on what point in time and in what social group you want to fix as being the reference, gold standard dialect. Go ahead and replay the the motion chart above from 1888 to 1950. In 1950, a lot of vowels have shifted around from where they were at the turn of the centry, and people might have accurately described the dialect as "fading away" with regard to how it sounded in 1900. Press play again, and in 1991, the dialect sounds different from how it did in 1950. It's more accurate to say that the Philadelphia dialect (and all dialects) is a moving target, made up of multiple component parts, some of which are more different from surrounding dialects than others, many of which are subject to change. Unfortunately, that message doesn't cash in on the combination of dialect mania that recently struck the Times, and Oscars night.

But what really bugs me about Nester's piece is how it revels in the eye-dialect (e.g. "sewda" for "soda", to indicate the pronunciation), which isn't helped by the graphics department totally whiffing on what a Philly soft pretzel looks like. I'd have to begrudgingly compliment him on his ear, except most of it seems to be borrowed from Sean Monahan's YouTube videos. I didn't feel the same way about Sean's videos back when they were originally posted, partially because he actually knows what he's talking about, and partially because you get the sense that he has a positive attitude and intellectual curiosity about the dialect. Nester's use, like you might expect from the opening paragraph, is really more about exoticizing the dialect.

This is, in fact, a real dialect spoken by real people (and so's Mandarin by the way) who have been generous enough to the researchers at UPenn to have invited us into their homes and told us about their lives. It was important to me, when we were talking to reporters surrounding our 2013 paper, to say that I think it would be a violation of those people's trust for them to be made fun of, or have their accents dragged out into the media like it's some kind of freak show. While Nester tries to make gestures that "I'm just one of yous," the piece is clearly about making a spectacle of the dialect, almost verging on treating Philadelphians as noble savages, their strange and incomprehensible language now a dying tongue.

I normally wouldn't spend so many words on a relatively contentless opinion piece, but I feel like I owe it to the speakers of Philadelphia being lampooned to say something. It is really only through the generosity of their time and stories that I've been able to launch my academic career, landing me in a pretty nice academic position in a pretty wonderful place. I fundamentally consider myself a scientist, not a politician, but as I told Zach Seward for the NPR story, to the extent we can spark any pride in people for the dialect they have, that's a bonus.

Above is a guest post by Josef Fruehwald.


  1. J. W. Brewer said,

    March 4, 2014 @ 11:19 am

    Maybe outsider-lampooning versus insider-jokiness is in the eye of the beholder, but I don't see why this guy's use of eye dialect is any different from Dr. Fruehwald titling his blog with "val" for "vowel." For a lay audience who would be completely at sea with IPA or phonological jargon like "lax" or "fronted" etc., I don't know that there's any alternative to eye dialect for describing the phenomena at issue.

    In particular, "sewda" for "soda" is as good a way as I can think of to convey to a lay audience in writing that extremely distinctive pronunciation which is intermittently present in my own idiolect (I grew up in northern Del., very close to the Pa. state line and maybe 20 miles south of South St. – I don't have all of the other features mentioned but certainly had schoolmates who did; as I may have mentioned in a prior thread that distinctive pronunciation of the GOAT vowel is what outed me, regional-originwise, to an NYC work colleague who'd grown up in Baltimore). Although then there's the oddity that I think "sew" and "so" are homophones in most AmEng varieties . . .

  2. AJD said,

    March 4, 2014 @ 11:28 am

    It's not as if Daniel Nester was avoiding phonological jargon (he writes "Nowhere but in the Delaware Valley can you hear those rounded vowels — soda is sewda, house is hay-ouse"). He's using the jargon, and using it incorrectly and confusingly (a feature of the Philly accent he's describing is that the "soda" and "house" vowels are less rounded than they are elsewhere).

    My experience of reading that article was not unlike what I imagine it must be like for a marine biologist to read an article about humpback whales that says "These majestic fish are the world's largest living animal species."

  3. Coby Lubliner said,

    March 4, 2014 @ 12:22 pm

    J.W. Brewer: Are there dialects in which "so" and "sew" are not homophones? Or did Nester mean "sew" as in "sewage"? Either way, a very poor use of eye dialect. Why not "sooda" if that's what was meant?
    "Val," on the other hand, is unambiguous.

  4. Josef Fruehwald said,

    March 4, 2014 @ 12:22 pm

    JW Brewer: Yeah, I tried to figure out while I was writing this up what distinguished Nester's use of eye dialect from other pieces. Comparing it to the Inquirer piece, the NYT piece uses it much more gratuitously. The phrase-book style "translations" pushed it over the edge for me, because they don't just provide ordinary orthography for the phrases, but "translate" all sorts of other things, like can -> may, it ain't -> it's not. The inquirer piece deployed the eye-dialect only to convey some specific pronunciation differences.

    So what distinguishes Nester from the PhillyTawk videos? That might be a bit closer to being a difference in the eye of the beholder. I actually think I might have reacted to Sean's videos similarly at first, but wouldn't have written a whole LL post, cause it ain't the New York Times. But Sean's follow up videos redeem any defaults with the original, because he really really knows what he's talking about. His posts on on the resistance to the low-back merger ( and the split-short-a system ( are very good.

  5. J. W. Brewer said,

    March 4, 2014 @ 12:30 pm

    I think maybe the closest IPA for the Delaware Valley (and perhaps broader Mid-Atlantic) diphthongized GOAT vowel (although I do not claim fluency in IPA . . .) might be "ɛu"? To the extent it's not a sequence that occurs in most AmEng varieties, I suppose that does make eye dialect a bit of a challenge. For some intuitive reason I am having trouble justifying or explicating, "sewda" for "soda" conveys it to me pretty accurately, just as "hewm" for "home" does, despite the fact that the typical AmEng pronunciation of freestanding "hew" would be just as misleading/irrelevant as that of freestanding "sew."

  6. Josef Fruehwald said,

    March 4, 2014 @ 12:58 pm

    It also doesn't help that he's now retweeting stuff that says about his piece.

    "Were they the only ones who noticed that [the Philadelphia accent] was offensive? Were they the only people who could identify it in a second? Are they the only ones to whom the mere sound of it is, although charming, as annoying as nails on a blackboard?? Finally, SOMEONE understands!! Hallelujah!"

  7. Mark F. said,

    March 4, 2014 @ 1:51 pm

    My problem with the eye dialect was that I couldn't make heads or tails of it. Like J. W. Brewer said, "sew" and "so" are homophones for me, so without being familiar with the Philadelphia accent I really just had to guess what he was getting at.

  8. Bill W said,

    March 4, 2014 @ 2:13 pm

    The Wall Women Bridge:

  9. Jerry Friedman said,

    March 4, 2014 @ 3:43 pm

    Isn't that the original meaning of "eye dialect"—that the misspelling indicates the same pronunciation as the correct spelling, like "sez"? "Val" is something different.

    I agree with J. W. that if you ignore the actual pronunciations of "ew" in English and you already know what the Philly GOAT vowel sounds like, "sewda" is a good representation of it (as I remember it from thirty-some years ago). I've also heard it or something similar in Pittsburgh, by the way, and I think from further down the Ohio Valley, though I can't remember why I think that.

  10. Ben Zimmer said,

    March 4, 2014 @ 3:43 pm

    When I was interviewed a few years back by the now-defunct Philadelphia offshoot of the Onion's AV Club, they spelled the Philly /o/ as "ayoh" (as in "hayohgie") and likened it to "Mike Myers as Wayne or Dr. Evil saying 'Oh,' in a bemused, sort of abashed way."

  11. AJD said,

    March 4, 2014 @ 4:12 pm

    Jerry, that fronting of the GOAT vowel is found throughout the entire Midland and South regions—i.e., from Philly west to Wichita, and points south.

  12. Jonathan D said,

    March 4, 2014 @ 5:30 pm

    I didn't think that "Extra syllables pile up so as to avoid inconvenient tongue contact or mouth closure." meant putting extra syllables in. A strange way of putting it, whatever was meant.

  13. Adrian Morgan said,

    March 4, 2014 @ 6:25 pm

    Even though I have Flash, and it works for everything else, in the specific case of the animation above, Firefox tells me that "a plugin is needed to display this content" (it does not elaborate).

    It worked in IE, though.

  14. Emily said,

    March 4, 2014 @ 6:48 pm

    Interesting piece, but: "The point me tried to make"? And the video wasn't playable for me, either–disappointing.

  15. Rodger C said,

    March 5, 2014 @ 8:57 am

    By "rounded" he seems to mean "diphthongized."

  16. Dennis Brennan said,

    March 5, 2014 @ 10:34 am

    I'm a Philly native, and I found Australian actress Toni Collette's version of a working-class Gray's Ferry Philadelphia accent in "The Sixth Sense" to be pretty good.

  17. Jerry Friedman said,

    March 5, 2014 @ 10:45 am

    AJD: Thanks. I was listening for it yesterday from a student from East Tennessee, and heard it a bit. By the way, I'm impressed that someone could tell a Philadelphia accent from a Baltimore accent by the GOAT vowel.

  18. Colin Fine said,

    March 9, 2014 @ 8:06 am

    Emily: he didn't say "the point me tried to make". He said "the point me, Bill and Ingrid tried to make". I too noticed this, and thought, Hooray, an academic linguist writing in English, instead of what Emonds calls "a grammatically deviant prestige construction"
    The argument that you are implicitly implying, though very common, is far from compelling. Number doesn't percolate into a coordination (you can't reason from "John goes" to "John and Jane goes") so why should case do so?

  19. Colin Fine said,

    March 9, 2014 @ 8:13 am

    I too was confused by "soda" vs "sewda". Once explained that "ew" is meant to represent the two sounds separately, rather than either of the customary pronunciations of the digraph, it becomes clear. Some Welsh accents have this characteristic.

  20. bianca steele said,

    March 9, 2014 @ 2:37 pm

    I watched one of the Monahan videos and they were fairly good (my husband, who grew up near New York and prides himself on ferreting out Philadelphia accents all over the world, most of which I usually can't hear myself, thought it was very accurate). The idea that primarily people in one city leave out consonants seems ridiculous to me, though. Replacing "of" with "o'" is obviously a "thing" that's pretty extensive. And I happened to watch the video a few hours before the Oscars were on, and Jonah Hill, whose character is absolutely not supposed to be from Philadelphia, very clearly says "I qui'" in the clip they showed.

  21. David said,

    April 3, 2014 @ 12:44 am

    Listening to the NPR piece, I found the young man's ay-diphthong quite remarkable. I wasn't familiar with the Philadelphia accent at all. The newer ay-sound immediately calls to mind the Newfoundland one, and I imagine many Canadians would react to it the same way that I do.

    Just curious – is there a significant gender difference in use of the newer ay-sound?

  22. Eneri Rose said,

    October 2, 2014 @ 9:54 am

    "The point me, Bill and Ingrid tried to make in our 2013 paper.." I had to read this a couple of times before I understood. Isn't a function of using standard grammar to make one's self understood? It is not just "deviant prestige construction".

    Colin Fine: Number doesn't percolate into a coordination (you can't reason from "John goes" to "John and Jane goes") so why should case do so?

    In the first example sentence, the subject John is a singular taking a singular verb. In the second sentence, the subject is John and Jane, which is plural, so it takes a plural verb. By similar analogy, in Fruehwald's sentence, "me, Bill and Ingrid" is the subject of the clause whose verb is "tried". As such, any pronoun in that subject should be in the nominative case. I'm sure Fruehwald would not have said "Us tried", so I can only assume his use of me instead of I was an oversight.

    Of course the way to avoid appearing to use a “deviant prestige construction" is to also use the more polite noun/pronoun order, "The point Bill, Ingrid and I tried to make…"

    Please forgive me if I have used sloppy terminology. I have not studied grammar since the 8th grade at St Katherine of Siena's in Philly's far North East.

RSS feed for comments on this post