Does [n] turn into [m] before [m], [p] or [b]? I'm a native Spanish speaker but have a C1 level in English. I'm trying to assimilate the native speech (no, I don't want a specific accent) and wish to know this, since Spanish does it.
Asked
Active
Viewed 602 times
2 Answers
2
In English, /n/ may be realized phonetically as [m] (or a similar phone? perhaps [n͡m]) before /p/ or /b/, as in the words "inborn" or "unprovoked", but this is described by some sources as a "gradient*" or "partial" phenomenon rather than a categorical assimilation. This seems to constitute a difference between English and Spanish.
*(Greg Lee seemed to disagree with the use of the word "gradient" to describe this phenomenon when I used that word in another answer, but nevertheless, it seems to be used in the literature. The main point is that it the phenomenon is variable and doesn't constitute a complete phonological neutralization of /n/ with /m/ in this context, or a completely identical phonetic realization of /n/ and /m/ in this context.)
I think it should be distinguished from the categorical alternation between /n/ and /m/ in the negative prefix "in-"/"im-" that occurs in Latinate words like "impossible". As far as I know, English speakers never use partially assimilated pronunciations like [nn͡mp] in words like "impossible" (although "nc" in Latinate words may be pronounced as either /nk/ or /ŋk/ based on somewhat variable and complicated factors).
It may be the result of "gestural overlap"
The English phenomenon is often considered to be the result of "gestural overlap" and co-articulation. There seems to be a relevant passage in the Blackwell Companion to Phonology. That said, there doesn't appear to be absolute consensus on the "gestural overlap" explanation; some phonologists may have other analyses.
Similar phenomena
Another similar phenomenon is the realization of non-nasal coronals like /t/ and /d/ as something like [p] or [b] before bilabial plosives.
Likewise, /n/ may become more like [ŋ] before a velar plosive, more like [ɱ] before a labiodental fricative, and more like [n̪] before a dental fricative. I don't remember if there are any significant differences in behavior based on the POA of the following consonant; e.g. if the realization of /n/ as [n̪] or [ɱ] is less variable than the realization of /n/ as [m] or [ŋ].
My understanding is that some languages (if I remember correctly, certain Australian languages), are even less permissive of this kind of "assimilation" than English. But I'm not sure. I found a paper that describes a similar kind of "gradient" assimilation in French: Gradient assimilation in French cross-word nasal+stop sequences (Laura Colantoni, Alexei Kochetov and Jeffrey Steele)
-
The way we say input as [ˈɪɱpʊt] is exactly what a native speaker of Spanish would do naturally to a written ‹np› sequence, even between words. I'm aware that English speakers perceive that as /ˈɪnpʊt/ phonemically, but the phonetics in natural speech always produces regressive assimilation, at least for me. What I'm less certain of is whether this occurs across morphemic boundaries in English, like talking about an in-pouring of support, let alone any risk of in-patient surgery coming off as impatient surgery. I rather think there's never a risk of that, but I can't say why or why not. – tchrist Dec 22 '17 at 02:30
-
You might note that in words like pink or tank or conquer, there is mandatory assimilation to [ŋ] before the velar stop. You can't not do that in English; Spanish speakers can't not do it in Spanish, either. And yet pincushion doesn’t do that at all, while concussion may or may not. My considered appraisal of these phenomena between the two languages’ phonologies is that nasal assimilation is stronger in Spanish than in English but not unknown here. – tchrist Dec 22 '17 at 02:50
-
@tchrist: I think the analysis of words like "pink" is complicated because I don't know of any clear evidence that these have the phoneme /n/ to begin with. Spelling doesn't seem to be an infallible indicator of phonology, so the fact that it is spelled with the letter "n" is not sufficient. Dictionaries will generally give the IPA transcription as /pɪŋk/, not as /pɪnk/. – herisson Dec 22 '17 at 03:07
-
It's possible to assume that the [ŋ] here is derived at some level from /n/, but that's a somewhat more abstract analysis, and it's hard to know how far to go along these lines (for example, some people go so far as to deny that /ŋ/ is a phoneme of English, and analyze word-final [ŋ] as phonemically /ng/). – herisson Dec 22 '17 at 03:07
-
@tchrist Pincushion does it for me. Hold on a sec ... Does it for my school director too, and my other-side-of the -ponder colleague too. What happens for you if you put it in a sentence, like How many pincushions did you get, for example? – Araucaria - Him Dec 22 '17 at 09:37
-
-
@tchrist No, not for me, 'cuz 'pincushions has just the one stress on the first syllable whereas 'pink 'cushions' has two, one on each word. Also there's a double /k/ in pink cushions (and the second one's quite aspirated - not that I can hear that in this noisy office, I'm just assuming it) – Araucaria - Him Dec 24 '17 at 12:52
-
1@tchrist, Also, a [j] glide is inserted between a front vowel and /ŋ/, but not before /n/ (even when /n/ is [ŋ]). Compare "Bank notes" and "Ban cameras". – Greg Lee Dec 29 '17 at 17:38
0
For the prepositional prefixes in and en the ‘n’ becomes an ‘n’ before a labial with a few exceptions. So ‘imbue’, ‘embellish’, ‘immure’, ‘implacable’. These are all derived from Latin.
There are exceptions, such as ‘inborn’ and ‘inbred’, as already mentioned in a comment. Personally, I think the difference is likely because the words born’ and ‘bred’ iare teutonic in origin, rather than because of any difference of emphasis in speech.
The main exception, however, is the privative prefix ‘un’ in ‘unbalanced’, ‘unbecoming’, ‘unmitigated‘ ‘unpalatable’
Tuffy
- 11,165
-
3You seem to be addressing spelling; the question is asking about phonetics. In imbue, etc., there is no /n/ to begin with—that change happened thousands of years ago. The change does happen in inborn, etc., which are generally (though not always) pronounced as though written imborn as well. – Janus Bahs Jacquet Dec 22 '17 at 01:04
-
@JanusBahsJacquet Of course you are mainly right, including my careless error over ‘imbue’. However, I have been trying for ages to work out whether I say ‘inborn’ or ‘imborn’, and I still cannot be sure. Spelling is so much more definitively ‘out there’. Is there research on this? My only, rather crude and inevitably flawed, check involves whether the tip of my tongue touches my palate on the ‘n/m’ or not. I’m pretty sure that with Latin derivations prefixed by ‘in’ it doesn’t, but with the rest, it does, however slightly. This would need an MRI scanner! – Tuffy Dec 22 '17 at 08:51
-
1It doesn’t really matter whether your tongue touches the palate or not. It’s generally the most distant articulator that defines the auditory perception of a sound, so it’s hard (if not impossible) to tell whether your tongue touches the palate if your lips are closed. In cases like inbred, it is, as you say, hard to tell exactly how much assimilation there is, and it varies even within the same speaker. – Janus Bahs Jacquet Dec 22 '17 at 09:41
-
@JanusBahsJacquet Yes indeed. You know vastly more about phonetics than I do. Perhaps the migration of pronunciation into spelling (discounting careless social media) declined with the rise of dictionaries and grammars in the enlightenment. It may now (for this generation of British English schoolchildren) be cancelled by the concentration on phonics in the teaching of reading. – Tuffy Dec 22 '17 at 10:01
nstraight before a labial or with a vowel intervening? Like inborn? – Dan Bron Dec 21 '17 at 23:38