Having seen the title of the discussion about 'smile' syllables, I thought it was going to be the same thing I have recently( or so!) thought of introducing to the world of phonology! But it was not.
I have noticed that Americans pronounce words such as 'smile' and 'slang' in two ways. In one, the following letter immediately flows after the /s/, as if they were strung together. But in the other, it's pronounced like /s/-mile, you pronounce /s/, a very short pause, and then -mile (Did I hear correct?), which is the way almost always non-natives, at least my country, choose to pronounce (in case they notice the difference).
The position of and the way to pronounce /s/ in the latter way seems to be similar to that of /n/ in words such as 'button' and 'carton'. phonetically the /n/ is called 'syllabic /n/'. So, why not have 'syllabic /s/',as well?
('Student' seems to be always pronounced with 'syllabic /s/', because of the stop /t/.) Rafee
I think what Rafee is talking about here is how some one-syllable words are "extended" to two for audible emphasis. Like, for example, the word crash. I've heard people pronounce this as kuh-rash when describing a "large" crash. This example is not quite onomatopoeia, but it might be close.
Another example is sweet, used as an exclamation for "nice" and pronounced suh-weet.
Never heard that affectation used with smile though.
Neither have I—but I do hear it, often, from those speaking English whose native language is something else, especially Spanish or Arabic. In both those languages, words don't start with sm- or sk-, while it's very common in Germanic languages. So Spanish speakers who are just starting to get comfortable with English tend to say "estone" for "stone"—not because they don't know better, just because they're not used to starting a word with that combination of letters.
Anglophones have the same problem with unfamiliar combinations, just different combinations. There's nothing inherently difficult about "Tlingit", for example, but it sure feels funny at first. And Americans are constantly mispronouncing "mbutu", I think for the same reason.
(In some African languages it's common to use a nasal—'m', 'n' or 'ng'—as a separate syllable. We never do that in English, so, confronted by word such as "ndami", most Americans would try "nuh-DAHM-ee" rather than "n-DAH-mi". At least, that's how I thought of it, until I learned better.)
On second thought, though, maybe I have heard Americans do what Rafee describes after all. I don't remember hearing someone say /s'maIl/; but I think I may have heard (without thinking about it) the slight hesitation in an initial /s/ in other words.
When I first read his post, I thought of something not exactly the same but perhaps related: When I first moved from the Midwest to the east coast, I noticed that people there pronounced "Wisconsin" differently. Midwesterners say "wih-SKAHN-sin"; Pennsylvanians, and others, say "wess-KAHN-sin". The vowel change I dismiss as noticeable but unimportant; what really changes the word, for me, is moving the 's' to the first syllable, because it changes the 'k' sound from unaspirated to aspirated. (I had to look up that term; I hope I've used it right.) In IPA/ASCII—what Wikipedia calls the Kirshenbaum system—that would be a change from /wI'skansin/ to /wEs'khansin/. I don't know IPA, but looking it up on the web that might be the difference between /wI'skɒnsən/ and /wɛs'khɒnsən/.
That isn't quite the way Rafee described it, but it occurs to me that the same thing may be going on. When I hear someone say /s'nel/ instead of /snel/, it sounds to my ear like a slight speech impediment. Maybe the same person would say /s'khIl/ (with a slightly aspirated /k/) instead of /skIl/. I'm just speculating, though.