Talk:Navajo phonology

j_obs vs j_son

I recommend that we use <ʝ> for the obstruent and <j> for the sonorant. The subscripts are a bit clunky and non-IPA. See Nauruan language#Consonants for a precedent. — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 04:58, 5 June 2008 (UTC)

But how is the distinction made in the published sources? —Angr 06:09, 5 June 2008 (UTC)

McDonough writes the [j] from /ɣ/ as Yd, the [j] from the [s~j] alternation as Yc, and the other [j] she calls the "true" glide but since it is rare she doesnt discuss much. Kari writes the [j] from /ɣ/ as ɣ (which would be palatalized), the [j] from the [s~j] alternation as ɣ̑, and the other [j] as y. Hoijer's structural analysis doesnt distinguish between the [j] from the [s~j] alternation and the "true" [j], but he considers the [j] from /ɣ/ to be [ɣ̑] (he indicates palatalization with italics actually). Young & Morgan dictionaries dont symbolize a difference between these since they use orthography, i.e. they are all y. But, in their verb stem inventory, they always note what the stem-initial consonant alternates with, so you can figure out what is what. Proto-Athabascan generally uses ɣ̇, ɣ̑, and y.

I was using j_obs (= ɣ̑) and j_son (="true" j) because I didnt want any confusion about them being pronounced any differently. There's only a distinction in their phonological patterning, which is:

	voiced before i,e	voiced before a	voiced before o	voiceless	d-effect
/ɣ/	j	ɣ	w	x	k
jobs	j	j	j	s	ts
json	j	j	j	j	j’

So, I was following Young & Morgan's conflation of symbols but using the subscript to distinguish between the phoneme of the [s ~ j] alternation and the "true" j. Using <ʝ> is closer to using ɣ̑ (which is the Americanist equivalent).

I was also debating not puting the /j/ with the [s ~ j] alternation in the chart since it is a more abstract phoneme. – ishwar  (speak) 06:48, 5 June 2008 (UTC)

I'd be in support of the last option you mentioned. I think the chart should stick to "surface phonemes" (yes, that's a contradiction in terms, but you know what I mean). But it's fine to mention in the body of the text that various scholars have hypothesized on the basis of morphological evidence that /j/ actually corresponds to two distinct phonemes. (It's sort of like /ɑ/ in American English – in a phoneme chart I'd only list it once, but some people have argued there are two phonemes, a lax /ɑ/ and a tense /ɑ_t/ that merge on the surface, i.e. "father" and "bother" rhyme on the surface but underlyingly they still have separate vowels. I might mention that in the text, but I wouldn't list them separately in the chart of sounds.) —Angr 20:51, 5 June 2008 (UTC)

Ok, I'll do that then. – ishwar  (speak) 21:50, 7 June 2008 (UTC)

Q about syllabic N

Many sources mention syllabic N in Navajo, and the orthography also includes this. Listening to speakers, however, it seems much less like a true syllabic N, and instead sounds like /ni/ with a very short vowel. Native-speaker pronunciation of ńdeezid or wikt:ńdízídígíí, for example, sounds to my ears like there is a vowel gap between the ń and the dee, or the ń and the dí. Likewise for Chʼíńlį́ -- there is a very short vowel sound between the ń and the lį́. Sometimes I've even seen it spelled with the vowel explicit, as Chʼíńílį́, as it is indeed given at the top of the Chinle, Arizona page.

Other languages do have a syllabic N, such as Japanese or various African languages, where there is no vowel gap between the N and the following consonant. With this in mind, is "syllabic N" really the appropriate nomenclature for this very short /ni/ in Navajo? -- Eiríkr Útlendi | Tala við mig 06:05, 2 April 2012 (UTC)

If you find a source that says what you've said, we can include it. But we can't include original analysis. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 21:29, 2 April 2012 (UTC)

Indeed. That's why I'm posting here and not just editing the page.  :) Is anyone aware of such research? And what of audio file evidence, would that count as source material for WP purposes? -- Eiríkr Útlendi | Tala við mig 03:36, 3 April 2012 (UTC)

I think we would need more than an audio file to back up the claim that syllabic N really has an epenthetic vowel. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 12:08, 3 April 2012 (UTC)

Sure. I was thinking of 1) more than one, and 2) including audio files as evidence examples, not as the sole evidence. -- 205.166.76.15 (talk) 15:10, 3 April 2012 (UTC) ← That was me, my session timed out. -- Eiríkr Útlendi | Tala við mig 03:09, 4 April 2012 (UTC)

Yeah, audio examples would be illustrative. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 17:00, 3 April 2012 (UTC)

proper apostrophy question

My thing is the {{lang}} templates but that does not mean that I know anything about the various languages that {{lang}} wraps.

In this article, I replaced this:

{{spell-nv|''x da’ání''}} → [x da’ání] Error: {{Lang}}: text has italic markup (help)

with this:

{{Lang|nv|x da’ání|italic=yes}} → x da’ání

for two reasons:

1. the original has italic wiki markup ('') that {{spell-nv}} passes on to {{lang}} which {{lang}} then rejects
2. when the italic markup is removed, if all of the characters in the text handed to {{lang}} fall within that template's definition of the 'Latin' character set (most of the Unicode Latin character set) then {{lang}} will render the text in italic font. But it doesn't because the text includes the character U+2019 (right single quotation mark) which is outside the bounds of the Unicode Latin character set:

   {{spell-nv|x da’ání}} → x da’ání

There are a couple of alternate fixes that could be made. We could, easily enough, modify {{spell-nv}} to accept and pass on |italic=[yes|no|unset]. We could also modify Module:Lang to accept U+2019 as a 'Latin' character. Thinking about this last is what led me to this post.

In Navajo, is U+2019 the correct character for its use in day-to-day Navajo writing? Or, is U+02BC (modifier letter apostrophe [ʼ]) or some other character more correct?

I suppose that it is relatively unimportant because there are about a dozen articles that include U+2019 in {{spell-nv}} so perhaps no fix is needed.

—Trappist the monk (talk) 14:23, 27 April 2018 (UTC)

I refined my search criteria:

{{lang-nv}} with U+2019 about 10 results

{{lang-nv}} with U+02BC about 100 results

These results would suggest that U+02BC is preferred. Are there any other similar characters that might have been used where these two are used?

—Trappist the monk (talk) 14:46, 27 April 2018 (UTC)

[h] or [x]

while the article talks in depth about many topics, it does not answer whether navaho has [h] as a separate phoneme, or at least as a phone at all. some of the transcriptions use [h] but some use [x], in the same phonological contexts, so that one even can't by themself deduce some rule for allophony as the article does not provide any. the summary table of phonemes does not include it, while in the text it is often talked about sets like "/k, g, x, h/" giving it its own place besides /x/. this begs the question if navaho in fact uses [h] at all, or if it's shown here in IPA sometimes just out of sloppiness Hyperbaton (talk) 00:06, 18 March 2023 (UTC)