Royal filled pauses
« previous post | next post »
In a comment on "Yair" (2/14/2025), Philip Taylor asserted that he routinely pronounces the orthographic "r" in the typical British filled-pause spelling "er":
« some Americans adopt a mistaken spelling pronunciation, rendering "er" with a final [r] » — well, speaking as a Briton, my "er" pauses, if prolonged, also end with an phoneme, although where exactly in the mouth I produce it I cannot be sure. Certainly it is totally unlike the trilled/r/ with which I might say "Run, rabbit, run, rabbit, run run run", but I think that it might be fairly close to the phoneme in my Maigret or Rien de rien.
If Philip actually trills the /r/'s in "Run, rabbit, run", and he's not from Scotland, this is a big dialectological surprise. And it's equally unexpected if he produces something like a French uvular /r/ at the end of his filled pauses. More likely, this is an extreme example of why sociolinguists are skeptical of how people think they talk.
Still, it's worth a bit of time to confirm the OED's r-less assertion (audio) about British filled-pause pronunciation. This would be a big task, overall, given the wide range of British speech across geographical, social, ethnic, and temporal variables, so I thought I'd start with the Royal family. (And I'll also end there, unless Philip's peculiar perceptions return in another form…)
The Good, The Bad & The Rugby has a 2023 interview with "The Prince of Wales, The Princess of Wales, & The Princess Royal at Windsor Castle". It's 50 minutes long, so there's plenty of time for filled pauses.
Here's the first example from Princess Anne, at around 2:16, with a bit of the context:
And by itself (with a bit of the /t/ from the preceding "but"):
Here are four of her filled pauses (leaving out the /m/-final versions), concatenated:
Here's the first example from Prince William, at about 4:31:
By itself:
And four of his filled pauses, concatenated:
The vowel quality of the Royal filled pauses seems somewhat higher and fronter than typical American versions, though I don't have time this morning to quantify that impression. But it's clear that the Royal's filled pauses have no final rhotic sounds, of whatever phonetic variety.
Philip Taylor said,
February 16, 2025 @ 6:20 am
To clarify: (a) I trill my "r"s because as a child, I realised that otherwise they came out as "w"a — "wun, wabbit, wun, wabbit, wun wun wun". Seeking to avoid the derision of my classmates, I took to trilling them. And (b) as to whether or not my prolonged "er"s end with a French uvular /r/, I think the answer is "possibly not", but the more prolonged ones definitely end with a repeated voiced vibration coming from the back of the throat.
Philip Taylor said,
February 16, 2025 @ 7:23 am
Perhaps I should add that it is only my prolonged "er"s (upwards of 0,5 seconds) that end in this way — a short "uh" (less than 0,2 seconds in duration) has no final <r> at all.
Coby said,
February 16, 2025 @ 4:43 pm
I wonder is Philip's "r" in "er" or "erm" is something like /ɦ/, which I have heard from some English folk.
Robert Coren said,
February 16, 2025 @ 7:17 pm
Christopher Robin's assertion that his bear's name was "Winnie-ther-Pooh", which puzzled American me as a child, only makes sense if we assume a non-rhotic "ther".
Mark Liberman said,
February 16, 2025 @ 8:46 pm
@Coby: "I wonder is Philip's "r" in "er" or "erm" is something like /ɦ/, which I have heard from some English folk."
Unless Philip provides a recording of his casual conversation, rabbits and filled pauses and all, we can't tell anything much about any of this.
Philip Taylor said,
February 17, 2025 @ 6:42 am
I will try (to provide such a recording, that is) but my office PC doubles as my audio recording kit, and as few people come into my office I would need to remember to start recording when I initiate a telephone call …
Mark Liberman said,
February 17, 2025 @ 8:07 am
@Philip Taylor:
It would work just as well, I think, to record yourself telling a story, for example about a trip that you once took, or describing a building and its surroundings, or whatever.
Philip Taylor said,
February 17, 2025 @ 8:20 am
Will do, Mark, but based on feedback from various conference talks that I have given, I rarely use filled pauses in normal conference-style delivery so may need to deliberately insert a few …
Ross Presser said,
February 18, 2025 @ 3:44 pm
@Robert Coren: I was puzzled by exactly the same thing as a child and it did not make sense to me until reading what you just posted, some 50 years after the confusion began.
Mark Young said,
February 19, 2025 @ 8:25 am
@Ross Presser: I had been reading Winnie the Pooh stories to my own boys for quite some time before I realized that "Eeyore" was a non-rhotic representation of donkey braying.
I was also very puzzled by one fellow's amusement at finding a census record showing a "G. Hardist" working at a WWI munitions factory.
Robert Coren said,
February 19, 2025 @ 9:27 am
@Mark Young: Yeah, same for me with Eeyore, minus the reading-to-kids part.
Philip Taylor said,
February 19, 2025 @ 3:25 pm
"Eeyore" was completely transparent to this non-rhotic Briton, but Richard Adams' "hrududu" was obvious completely opaque to those making the film of Watership Down as they had the characters pronounce it /hruː·duː·duː/ — it should, of course, have been pronounced /hrʌ·dʌ·dʌ/, ths sound of a very low-revving internal combustion engine as used in (for example) John Deere tractors of the period …
Robert Coren said,
February 20, 2025 @ 10:20 am
@Philip Taylor, by remarking on filmmakers missing obvious pronunciation clues, has sent me off on a tangent.
In Hilary Mantel's trilogy about Thomas Cromwell, she not only has the King nickname him "Crum", but also indicates that native speakers of French have trouble with the first vowel of his name; she renders their pronunciation as Cremuel, in which I assume the first e to be a kind of extended schwa. Both of these strongly suggest that she considered that first vowel to be /ʌ/ (which does not occur in French), but the makers of the TV series "Wolf Hall" ignored these clues and had everyone pronounce it as /ɑ/ as it would be in modern English – a vowel which would be perfectly familiar to French-speakers, and would make the name unlikely to be shortened to "Crum".
David Marjanović said,
February 25, 2025 @ 11:22 am
ther also shows that [ə] is the unstressed allophone not of /ʌ/ (as in America), but of /ɜ/, in non-rhotic southern British English.
For Thomas Cromwell, his Wikipedia article gives the pronunciations "/ˈkrɒmwəl, -wɛl/", i.e. with the rounded LOT vowel, from the Collins English Dictionary. (No pronunciation is mentioned for Oliver Cromwell, BTW.) This is, however, followed by the following footnote:
"Contemporary native speakers of English most often wrote the surname as Crumwell, suggesting that this spelling was an attempt to represent a commonly used pronunciation.[2] Based on the phonetic challenges to native speakers posed by this and other variants of the frequent Cromwell spelling (e.g. Cromuello) adopted by correspondents, Diarmaid MacCulloch proposes common usage of pronunciations missing the w and shifting to "a short 'o' sound", as reflected in spellings such as Crummle or Cromell (that used by the king when he stripped Cromwell of all his titles).[2]"
This opens the possibility that -om- was written instead of -um- just to avoid the mınım problem. In that case, the question becomes how far along the FOOT-STRUT split was in the 1530s. However, if this is not the case and the first vowel was LOT, then it would indeed be effortlessly represented by the French /ɔ/ as in comme. Indeed, the 21st-century Standard Southern British pronunciation of LOT isn't [ɒ], it's [ɔ] (and THOUGHT has moved on to [oː], but that definitely wasn't the case in the 1530s).