Connected speech problems?

jsk   Wed Aug 15, 2007 4:41 am GMT
Hi, first of all let me tell you a little background. I've been living in Canada for about 5 years, and according to people here I don't have an accent or anything like that. I never really learned English pronounciation formally in school since I was already past that stage of schooling when I arrived. However I have some challenges when I'm speaking.

I find it hard to pronounce the word "the" preceeded by an d sound. For example, phrases such as

"closed the.."
"hated the.."
"connected the.."

The last two examples were easy for my, but when I say the first one, it's really awkward. It's say these words individually quickly, but when I try to connect them, it sounds like I'm saying "close the", which loses the verb tense? Any advice?
Travis   Wed Aug 15, 2007 4:57 am GMT
In these cases, very many North American English dialects actually do not realize [D] but rather form a geminate [d:] or [t:] with the preceding /d/ or /t/.
Guest   Wed Aug 15, 2007 5:23 am GMT
Hi, Travis. As I pointed out before, I have not undergone formal training in English pronounciation and therefore do not know the system with which you represent these sounds, could you give an example of [d:] and [t:] ? Thanks!
Gwest   Wed Aug 15, 2007 7:24 am GMT
Travis does it again! We are not all versed in such technical language, Travis! The questioner even made that clear, but you went right on using such terminology.
Guest   Wed Aug 15, 2007 9:42 am GMT
Can you, Gwest, answer in a comprehensible way, please.
Gwest   Wed Aug 15, 2007 9:58 am GMT
If you are speaking fairly rapidly, it is hard to pronounce the /d/, but don't worry about it. Many, many native speakers also do not sound the /d/ in "closed the", etc. Most everyone will understand from context. Those that do not, or insist on sounded the /d/ are merely thick, pedantic and/or prone to hypercorrection.

Context will reveal all.
Travis   Wed Aug 15, 2007 1:29 pm GMT
The transcription system that I am using is called X-SAMPA. More information can be gotten on it at:

http://en.wikipedia.org/wiki/X-SAMPA

[d:] and [t:] are long consonants, which do not exist in English as specific phonemes but rather show up at morpheme boundaries in English. Examples of such include:

"Carrot Top" ["k_he:R1?%t_h:a?p]
"fad diet" ["fE{:%d:a:I1?]

Note that the [t_h:] here is just the aspirated version of [t:].

(Note that the pronunciation above is specifically in my dialect, and is not General American pronunciation.)
Travis   Wed Aug 15, 2007 2:36 pm GMT
The matter is that using such technical terminology very often allows one to express complex ideas in few words. If I did not use such language, such would probably have come out as something more like:

"In these cases, very many English dialects spoken in the US or Canada actually do not realize the 'th' sound at the start of 'the' as such but rather pronounce it like a 'd' sound (as in 'dog') when following a word that ends in a 'd' sound or like a 't' sound (as in 'stop') when following a word that ends in a 't' sound. In these cases, the two adjacent 'd' sounds or 't' sounds merge together rather than being actually pronounced twice."

Even then, that description is very imprecise, and is actually rather inaccurate, as it does not address the difference between phones and phonemes, which is relevant here. For instance, words that end with a glottal stop here (and not a "'t' sound", that is, [t]) will actually be pronounced together with a following /D/ as [t:]. Contrast this with my original sentence:

"In these cases, very many North American English dialects actually do not realize [D] but rather form a geminate [d:] or [t:] with the preceding /d/ or /t/."

This sentence is succinct and precise, explaining exactly what occurs in such dialects without needing roundabout language like that in my reworded paragraph. Furthermore, it properly addresses the difference between phones and phonemes, unlike my reworded sentence which is stuck with the ambiguous word "sound"* which is used in some contexts to mean "phone" and in others to mean "phonemes".

* It is ambiguous because many use the term to really mean "phoneme", even though most people do not really understand the difference between a phone and a phoneme. Due to the general lack of understanding by laypersons of this distinction, I had to use this term despite its clear inadequacy here.
Travis   Wed Aug 15, 2007 5:33 pm GMT
>>It seems that in these situations, I do something like Definite Article Reduction. For "close the door", I often say something like [klo:z?1"dO`:], with the final /d/ and initial /D/ both reduced to [?]. For "hated the..." I tend to just elide the /D/: ["he:4I46]. /D/ is a very strange phoneme in my dialect.<<

In cases like "close the door", I assimilate /D/ with the preceding /z/ except in very careful speech, to result in [z:]; note that final devoicing does not fully apply here due to the assimilation, but the resulting [z:] is often being not fully voiced either here. Consequently, I tend to get [%k_hM\o:z:@:"dO:R] for "close the door".

Note that similar assimilations frequently show up in other places here, such as word-final /s/ assimilating with word-initial /D/ to form [sT] (careful) or [s:], word-final /n/ assimilating with word-initial /D/ to form [n_d:]. For example, I have:

"pass the ball" [%p_hE{s:@:"bQU]
"man the vehicle" [%mE{~:n_d:@:"vi:I?kM:]
Travis   Wed Aug 15, 2007 6:09 pm GMT
I do not myself have assimilation with /r/ and /l/, but on the other hand I do have general voicing assimilation of /D/ with preceding obstruents. Note that such is along with the "hardening" of /D/, such that one can get word-initial [t] (not [t_h]) from /D/ following a fortis obstruent.
Jeff   Thu Aug 16, 2007 3:05 am GMT
<<Travis does it again! We are not all versed in such technical language, Travis! The questioner even made that clear, but you went right on using such terminology.>>

Travis could always post under Guest and I'd always clearly know it was him, due to his odd writing style.
Travis   Thu Aug 16, 2007 3:40 am GMT
Do you guys even realize that it is often difficult to impossible to express such ideas without such terminology? Such ideas need linguistics-related concepts, terminology and notation to express, and trying to only use concepts and terminology familiar to laypersons to express such clearly is difficult to impossible.
Guest   Thu Aug 16, 2007 8:00 am GMT
<The matter is that using such technical terminology very often allows one to express complex ideas in few words.>

Using few words would be a first for you.
NNES   Thu Aug 16, 2007 8:01 am GMT
<Do you guys even realize that it is often difficult to impossible to express such ideas without such terminology? Such ideas need linguistics-related concepts, terminology and notation to express, and trying to only use concepts and terminology familiar to laypersons to express such clearly is difficult to impossible. >

Gwest did it.
Travis   Thu Aug 16, 2007 12:50 pm GMT
But non-technical versions often gloss over important details, such as the difference between phones and phonemes, which are critical to really understanding what is going on in many cases.