baths, paths, moths, myths, houses etc.

How do you pronounce these plurals? I pronounce them:

baths - /b{Ts/
paths - /p{Ts/
moths - /mATs/
myths - /mITs/
houses - /haUsIz/

I have no voicing in these plurals. I only have voicing in plurals where it's supported in the spelling as in ''knives'', ''leaves'' etc.

SpaceFlight, do you mean to use [] for those? I assume you meant the phonetic level since you mentioned your pronunciations, not the phonemic representations.

I pronounce them:

<baths> [b{Ts]

<paths> [p_h{Ts], rarely [p_h{:Dz]

<moths> [mATs]

<myths> [mITs]

<houses> [h{UsIz] or [h{UzIz]

<<SpaceFlight, do you mean to use [] for those? I assume you meant the phonetic level since you mentioned your pronunciations, not the phonemic representations.>>

I mentioned my pronunciations on the phonemic level. On the phonetic level they'd be more like:

baths - /b{Ts/ - [b{Ts]
paths - /p{Ts/ - [p_h{Ts]
moths - /mATs/ - [mATs]
myths - /mITs/ - [mITs]
houses - /haUsIz/ - [haUsI:z]

I myself have:

"baths" : /b{Dz/ -> [b{:D_0z_0]
"paths" : /p{Dz/ -> [p_h{:D_0z_0]
"moths" : /mOTs/ -> [mOTs] or /mODz/ -> [mO:D_0z_0]
"myths" : /mITs/ -> [mITs]
"houses" : /haUzIz/ -> ["ha:U.zI:z_0]

Note that it is clear that it is underlying /Dz/ in a number of cases here because of the produced vowel length and the fact that the realized consonants are still lenis even though they are unvoiced.

That [Dz] sound is very difficult to me. I can't make it. I prefer to say [Ts] or just [z]. :)

For me,

<baths> [b{Ts]

<paths> [p{Ts] or [p{Dz]

<moths> [mQTs] or [mQDz]

<myths> [mITs]

<houses> [haUzIz]

<<<baths> [b{Ts]

<paths> [p{Ts] or [p{Dz]

<moths> [mQTs] or [mQDz]

<myths> [mITs]

<houses> [haUzIz]>>

It's interesting that there are some people here that preserve the voicing in some plurals, but not others. For me, ''mouths'' /maUDz/ is the only plural that I have the voicing in, besides the plurals where it's supported in the spelling as in ''knives'', ''leaves'' etc.

- Like you, I preserve voicing when it's supported by spelling, and I would definitely only ever pronounce the plural "mouths" as [maUDz].

- Likewise I would only ever pronounce "houses" as [haUzIz].

- Words ending in [T] are variable with me, with some allowing only [Ts] and some allowing either [Ts] or [Dz].

Yes, this is one area of considerable variation amongst speakers today. It's interesting how for some people the voiced versions may be interchangeable with the unvoiced ones and some only have one or the other and it often depends on specific words (like, I never have [Dz] in "moths" but I might rarely have it in "paths").

"Houses" is an interesting case because /z/ there was the historical form but it's been reanalyzed as /s/ in analogy with /haUs/ for many speakers (including me). However, I find I can have the /z/ or /s/ version there interchangeably--either one sounds fine to me and both sound like pronunciations I'd use.

I keep them all unvoiced, for whatever that is worth.

Roughly speaking, square brackets in phonetic transcription are descriptive; slashes are prescriptive. At least that's a general rule. You put things inside square brackets when you are describing how something is actually pronounced (often in greater detail than just the phonemes, or in contrast with the nominally correct pronunciation); you put them inside slashes when you are describing how they _should_ be pronounced, in which case you normally would just transcribe the phonemes.

<<Roughly speaking, square brackets in phonetic transcription are descriptive; slashes are prescriptive. At least that's a general rule.>>

That's not really true at all.

In accurate linguistic phonetic transcription, slashes are phonemic while brackets are phonetic.

It has nothing to do with prescriptivism vs. descriptivism. Linguists, who are pretty much by definition descriptivist, use slashes (phonemics) and brackets (phonetics) to describe language while prescriptivists could conceivably use both phonemic and phonetic transcription to prescribe a certain pronunciation. Of course it is my experience that most prescriptivists clearly lack a solid understanding of most basic linguistic concepts so I wouldn't be likely to expect accurate phonemic or phonetic transcription from such sources, anyway.

<<Roughly speaking, square brackets in phonetic transcription are descriptive; slashes are prescriptive. At least that's a general rule.>>

What makes you think that? That's definitely not true.

Because, if you are prescribing pronunciation, you usually only prescribe the phonemes, not all the details (unless you're trying to eliminate an accent, which is rare). If you are describing pronunciation, you are often doing so to illustrate deviations from the norm, in which case mere phonemic transcription is insufficient.

<<Because, if you are prescribing pronunciation, you usually only prescribe the phonemes, not all the details (unless you're trying to eliminate an accent, which is rare).>>

According to what you say, a prescriptivist would prescribe that the ''p'' in ''path'' should not be aspirated, which is not what any prescriptivist would ever do.

path - /p{T/ = [p_h{T]

<<If you are describing pronunciation, you are often doing so to illustrate deviations from the norm, in which case mere phonemic transcription is insufficient.>>

Why is it insufficient?