Wiktionary talk:About Navajo

Glosses for verb stems in etymologies

As of the present (2012-04-25), glosses for verb stems included in NV term etymologies tend to give just the form of a stem, such as imperfective verb stem. This is useful, but strikes me as incomplete -- it gives no information about the meaning of the stem, which is generally what one expects to find in a term gloss. The lack of any meaning information in stem glosses can be confusing to users, as noted at Talk:naalnish#Etymology.

It seems to me that it would be best to include a short gloss in addition to the stem form. This would align with current practice for Navajo verb prefixes and infixes listed in etymologies, and also align with the convention for term glosses in etymologies for other languages, such as at English know#Etymology or Spanish saber#Etymology_5 or Japanese 司る#Etymology.

Any thoughts, objections, alternate approaches? -- Eiríkr Útlendi │ Tala við mig 18:30, 25 April 2012 (UTC)

If you weren't already thinking of this, I suggest leaving the grammatical information outside of the quotation marks, as quotation marks to me imply "this is what the term means". I'd suggest using the slot other languages use for transliterations to hold grammatical information (like this), or adding a grammar slot to {{term}}... or, yeah, even what you already did is more helpful than no gloss at all. - -sche (discuss) 19:46, 25 April 2012 (UTC)

Actually, prompted by -sche to look further into it, I find that {{term}} already has a pos parameter that is ostensibly intended for exactly this kind of use. I've just edited the etyl at naaltsoos to use this param. -- Eiríkr Útlendi │ Tala við mig 20:37, 25 April 2012 (UTC)

Very nice fix to both issues. I wholeheartedly feel that the stems should have a gloss; my impression is that there have been objections based either on the idea that the stems can't occur independently, or that sometimes a stem can have multiple (related) meanings, but it's not clear which is original. But neither of these would mean the stem just doesn't have meaning. Ewweisser (talk) 21:12, 25 April 2012 (UTC)

Ta, thanks.  :) For glosses for terms that have multiple possible meanings, in JA entries, I've often either listed the few most common meanings, or given the meaning most relevant to the context of the current page, and then just made sure that the main page for the glossed term lays out all of the full meanings. The gloss field in {{term}} is really only intended for a short blurb; I think the idea is to give the user a general idea of what's going on, so short-and-sweet is fine in my opinion, even if it doesn't do the glossed term justice -- that's what the main entry is for, no?  :) -- Eiríkr Útlendi │ Tala við mig 01:25, 26 April 2012 (UTC)

How to handle dialectical differences?

Reading around on the NV WP and elsewhere, it's become clear to me that Navajo has lexicographically significant variation in some terms. How should the EN WT account for this?

One such variation is the /-aoh-/ > /-aah-/ shift in the second-person duoplural suffix infix. I've added the /-aoh-/ forms as alternates alongside the /-aah-/ forms in the conjugation tables at naalnish and neiłtsoos. Is this acceptable going forward? If so, should we add a footnote to the conjugation table to explain this difference to users? -- Eiríkr Útlendi │ Tala við mig 20:48, 25 April 2012 (UTC)

I'd been thinking about this too. I like what you did. Another variation that's been bugging me is the reduction of ni- to n- and ní- to ń. Thoughts? Ewweisser (talk) 21:12, 25 April 2012 (UTC)

Oo, yes, that's another excellent issue to bring up.

My first impression is that the shorter n- and ń- forms are abbreviations contractions, and not even full abbreviations contractions at that, as the speakers that I've heard still pronounce a vowel sound after the n and before the following consonant, albeit a shorter, clipped vowel sound. It took me a while of listening to figure out what was going on in words like Chʼíńlį́, as my background in Japanese led me to think that what's described by some as a syllabic N in Navajo really indicated something more like the moraic N in Japanese, but the audio clearly demonstrates something different.

One possibility is to list the full ni- and ní- spellings as the lemma forms, and list the shortened spellings as alternative forms, including explanations about the abbreviation contraction and pointing to the lemmata. But that could well be an error if the shorter form is considered the standard spelling for a given term.

Perhaps we could get Stephen or even Seb to chime in? (Seb doesn't have a user page set up here, but s/he does have one at w:nv:Choyoołʼįįhí:Seb_az86556 and at w:User:Seb_az86556.) -- Eiríkr Útlendi │ Tala við mig 23:40, 25 April 2012 (UTC) -- Edited, Eiríkr Útlendi │ Tala við mig 01:21, 26 April 2012 (UTC)

Comparative forms of descriptive stative verbs

Numerous verbs in Navajo that describe a state, such as nitsaa or nineez or nidaaz have comparative forms, such as áníłtso or áníłneez or áníłdas. This kind of information seems like it should go in the POS header, such as:

Verb nineez (intransitive, stative, comparative form áníłneez)

What does anyone else think? -- Eiríkr Útlendi │ Tala við mig 17:29, 11 June 2012 (UTC)

Font size

I'm curious as to why the template {{nv}} produces words in such a large font size. Looking at e.g. Appendix:Swadesh lists for Dené-Yeniseian languages and other threads on this page, the large size of Navajo words looks rather intrusive alongside normally-sized words. I have not encountered such sizing for other languages. My first guess was that it was used to make diacritics more apparent, but even diacritic-heavy languages like Vietnamese don't seem to use the same sizing here. Is there a legitimate reason for this? If not, can this be changed? ~~ Lothar von Richthofen (talk) 20:03, 17 August 2012 (UTC)

I can't address the reasons, let alone their "legitimacy" (?), but taking your last question literally — yes, it can be changed by changing this part of MediaWiki:Common.css:

    /* Navajo [[Template:nv-Latn]] */
    .nv-Latn {
        font-family: "Calibri", "Aboriginal Sans", "DejaVu Sans", "Arial Unicode MS", sans-serif !important;
        font-size: 130%;
        }

Alternatively, and more radically, we can simply remove that bit of MediaWiki:Common.css, and change Template:nv/script from nv-Latn back to Latn. —Ruakh_TALK 20:19, 17 August 2012 (UTC)

See User talk:Stephen G. Brown#Font specifications. --Μετάknowledge^{discuss/deeds} 20:24, 17 August 2012 (UTC)

The apparent size will depend on the fonts that you have installed. When I added the .nv-Latn font-family to MediaWiki:Common.css, I set the font-size to 100%. Later, User:Mglovesfun changed it to 130%, presumably because the font he uses has a very small apparent size. See [1]. For me, this causes the Navajo to look very large indeed. —Stephen ^(Talk) 20:29, 17 August 2012 (UTC)

However, can’t change Template:nv/script from nv-Latn back to Latn. Latn does not work for Navajo. —Stephen ^(Talk) 20:34, 17 August 2012 (UTC)

Thanks for the quick responses. Technical issues were another possible culprit in my mind. Is there anything that can be done (or that I can do for myself) to address this issue, then? ~~ Lothar von Richthofen (talk) 20:44, 17 August 2012 (UTC)

Yes, the font-size in MediaWiki:Common.css could be reset to 100%. Since User:Mglovesfun changed it to 130%, I suppose it would be up to him to agree to reset to 100%. —Stephen ^(Talk) 21:06, 17 August 2012 (UTC)

Well, it would seem that Mglovesfun is on board with this. If someone with the relevant userrights could do the honours, that'd be great! ~~ Lothar von Richthofen (talk) 22:47, 2 September 2012 (UTC)

 Done. I've just removed the font-size specification, so it's back to 100% now; but if desired, we can set it somewhere in between. By the way, the site-wide CSS is subject to client-side caching, so if you want to get the updated version immediately, you'll most likely need to visit http://bits.wikimedia.org/en.wiktionary.org/load.php?debug=false&lang=en&modules=site&only=styles&skin=vector&* (changing vector to whatever skin you're using — vector is the default, but if you've customized the site, it could be monobook or any of various other skins) and do a "hard" refresh (hold down the Shift key while clicking the refresh button). —Ruakh_TALK 23:27, 2 September 2012 (UTC)

Pronunciations

Here is a zip of 645 pronunciation clips (words + a few examples) in ogg format: https://dl.dropboxusercontent.com/u/29377071/Navajo%20Pronunciation%20Collection%20%28Ogg%29.zip After stumbling around the main Wiktionary page and looking at a confusing number of docs on what to do with them, I've decided that I'd rather not disrupt current entries, and instead let more competent contributors handle this. Some of these are rerecordings of words here that already have clips by navajowotd. This collection is, objectively, of better quality, and for the purposes of wiktionary and wikimedia falls under the same license as the existing clips (http://navajowotd.com/about#copyright). If you need to review the audio, I've fused them together and threw them up on soundcloud - https://soundcloud.com/navajowotd/1-navajowotd-navajo. Bcsho (talk) 20:04, 29 April 2013 (UTC)

Thanks, Byron. I’ll download it and see what I can do. —Stephen ^(Talk) 12:20, 30 April 2013 (UTC)

I downloaded the zip file, but some of the files were password protected. I don’t know if those were important or not. —Stephen ^(Talk) 13:25, 30 April 2013 (UTC)

That's interesting. Do you mind listing the files that need a password? I used a batch MP3 -> OGG converter that may have required a license before all files could be used. If it helps, here are the same files in MP3 before they were converted: https://dl.dropboxusercontent.com/u/29377071/Navajo%20Pronunciation%20Collection%20(MP3).zip Bcsho (talk) 23:17, 7 May 2013 (UTC)

They were No.’s 111, 112, 235, 236, 493, 645, 642, 91, 90. I’ll have a look at the unconverted ones. —Stephen ^(Talk) 23:36, 7 May 2013 (UTC)

No, the "Navajo Pronunciation Collection (MP3)" had the same problem, with the same file numbers. There must be something about those 9 files. —Stephen ^(Talk) 23:45, 7 May 2013 (UTC)

Okay. It seems there are some irregularities with the file names. I'll work on it and post the updated files. Bcsho (talk) 02:52, 10 May 2013 (UTC)

When I got to file No. 63, the file names became mixed up. File 63 is átʼahálo, but its name is "Shí Dóó Jerry Bá'áłchíní Éí Mary Bighandi Da'iidą́". File 64 is atiin, but its name is "átʼahálo". File 65 is bitiin, but its name is "át'ah". File 66 is atsá biyáázh, but its name is "t'ahálo", and so on. —Stephen ^(Talk) 11:41, 25 May 2013 (UTC)

Possessed form?

Moved from [[Talk:chʼiyąʼ]]

Presumably that means that this is a bound form and only ever seen following a possessive prefix? E.g. shichʼiyą / nichʼiyą / bichʼiyą etc.? -- Eiríkr Útlendi │ Tala við mig 06:58, 6 February 2013 (UTC)

Yes. —Stephen ^(Talk) 07:27, 6 February 2013 (UTC)

Thank you. A couple questions arise, then:

• Should possessed-form entries include declension tables showing the full paradigms?

I've been away from my NV studies for a while, but my dim recollection is that possessed nouns are much more regular than verbs in terms of how the prefixes work. If so, it should be possible to put together a template that covers such a declension table and requires fewer input parameters than the verb conjugation table.

• Should possessed-form entries start with a hyphen, since they are always prefixed?

I note that NV infixes and verb roots take hyphens in etymology sections, like -ł- or -gai at łigai, so it might make sense for consistency's sake to follow suit and start possessed noun roots with a hyphen.

I look forward to your thoughts. -- Eiríkr Útlendi │ Tala við mig 17:08, 6 February 2013 (UTC)

• Sample mock-up: a call to {{nv-pos-lowtone|chʼiyąʼ}} produces:

possessives of About Navajo

	singular	duoplural	plural
1st person	shichʼiyąʼ	nihichʼiyąʼ	danihichʼiyąʼ
2nd person	nichʼiyąʼ	nihichʼiyąʼ	danihichʼiyąʼ
3rd person	bichʼiyąʼ
4th person (3o)	yichʼiyąʼ
4th person (3a)	hachʼiyąʼ
Indefinite (3i)	achʼiyąʼ

I have no idea what the plural 4th-person possessive is, or even if it exists. The current code just leverages {{nv-verbtable}}. If there is no plural 4th-person possessive, I think it would be relatively easy to modify the current code to leave that cell blank. -- Eiríkr Útlendi │ Tala við mig 22:58, 6 February 2013 (UTC)

I think it’s a good idea. I have thought about it from time to time, but there has always been so much else that needed to be done. As far as the best way to make a page for these forms goes, I don’t really know what would be best. That’s why I have not made many pages like this. One way to go is like chʼiyąʼ, tsoh. Another way would be with an initial hyphen. A third way is to avoid pages like this and just make pages for each possessed form.

Yes, a template would be simple except for part of the 4th person. A template would be simpler if we leave the ha-, hw- forms off altogether. Also, I think we could leave the a- form off, since we’ve been using that form as a lemma (ajéí, abid, ajáád). The 3rd and 4th persons don’t have separate plural forms. The possessive prefixes are:

Navajo possessive pronouns

			singular	dual	plural
1st person			shi-	nihi-	danihi-
2nd person			ni-	nihi-	danihi-
3rd person			bi-
4th person (3o)			yi-
4th person (3a)			ha-, hw-
Indefinite person (3i)			a-

Note that some possessed forms cause the prefixed pronoun to have a high tone, as with álaʼ (hands). I see that I have already added the possessive pronoun prefixes at álaʼ.

—Stephen ^(Talk) 07:47, 7 February 2013 (UTC)

Concerning the use of a- form as a lemma, it is also very common to use the ha- form as the lemma:

hatsʼíís or hatah (body), hatsiistsʼiin (head), haniiʼ (face), hanííʼ (waist), hakʼos (neck), hadáyiʼ (throat), hawos (shoulders), hagaan (arm), hachʼoozhlaaʼ (elbow), hálátsíín (wrists), hálaʼ (hands), hajéítsʼiin or hayid (chest), habid (stomach), hazhiʼ (trunk), hwííshgháán (back), hatłʼaaʼ (buttocks), hajáád (legs), hagod (knee), hakétsíín (ankles), hakeeʼ (feet), hakágí (skin), hadoh (muscles), hatsʼin (bones), hatsiiʼ (hair). —Stephen ^(Talk) 07:47, 7 February 2013 (UTC)

Ain't nothing simple when it comes to language, and Navajo's a great illustration of that.  :)

Some other questions that arise then:

• Are the ha- forms also indefinite, like a-? Or are these true 4th-person forms?
• Do nouns that take ha- also take a-, or are these mutually exclusive?
• Do nouns that have an unbound form, like chʼiyáán or béésh, also have indefinite a- forms? Or do the unbound forms take that place?
• Is there any noteworthy east-west difference in whether a speaker uses ha- or a-?

I really wish we had a proper scripting language for templates; this would allow for *so* much more flexibility in creating these kinds of declension and conjugation tables. Ah, well. -- Eiríkr Útlendi │ Tala við mig 16:59, 7 February 2013 (UTC)

Yes, the ha- forms are like "one’s" (one’s mother), and the a- forms are like "someone’s" (someone’s mother). Nouns can take both forms:

atsʼíís or atah (body), atsiistsʼiin (head), aniiʼ (face), anííʼ (waist), akʼos (neck), adáyiʼ (throat), awos (shoulders), agaan (arm), achʼoozhlaaʼ (elbow), álátsíín (wrists), álaʼ (hands), ajéítsʼiin or ayid (chest), abid (stomach), azhiʼ (trunk), ííshgháán (back), atłʼaaʼ (buttocks), ajáád (legs), agod (knee), akétsíín (ankles), akeeʼ (feet), akágí (skin), adoh (muscles), atsʼin (bones), atsiiʼ (hair).

As far as I know, any noun can take these forms, but if they are unbound forms, then it becomes a matter of being SoP. Bound forms require possession, so you have to say "one’s mother" at the very least...but unbound forms may be possessed, too, when need be, and you can say "one’s metal" if you need to. The ha- forms are true 4th person forms (alternate 4th person), but the a- forms are more of a 3rd person form. There is no difference in this in regard to east and west dialect areas. —Stephen ^(Talk) 09:29, 8 February 2013 (UTC)

• Thank you, Stephen. For nouns then, it seems there are two broad classes, grammatically speaking -- bound, and unbound. Bound nouns always require a possessive prefix, and are never found in isolation, while the unbound nouns can be used without a possessive prefix, and generally undergo a kind of declensionary change when taking a prefix (such as the change in béésh to -béézh).

If I understand your comment here correctly, chʼiyáán would have hachʼiyąʼ, but not *achʼiyąʼ? Or would the achʼiyąʼ (“someone's food”) form still be valid Navajo, in the proper context? -- Eiríkr Útlendi │ Tala við mig 18:27, 11 February 2013 (UTC)

I would not characterize it as a declensionary change, it is more like consonant harmony. Navajo has several types of sound harmony, such as coronal harmony. In this case, the syllable sometimes changes as a result of something similar to the construct state of Arabic. It affects the whole syllable, not just the final consonant, but only certain consonants can be altered. For example, hééł becomes biyéél (h > y; ł > l). It’s the same thing that changes saad to bizaad. I don’t know exactly what this is called or how to describe it any better. Navajo phonology is an extremely complex matter.

I think hachʼiyąʼ and achʼiyąʼ are both valid in the proper context. —Stephen ^(Talk) 03:22, 12 February 2013 (UTC)

Many thanks again for the detail. Since both ha- and a- forms exist, my sense is that both should be included in any such table. I hope that this phonological change in a word from unbound to bound form is a consistent change? I.e., hééł > -yéél after all possessive pronouns? Or are there some nouns that have different bound forms depending on the possessive prefix? -- Eiríkr Útlendi │ Tala við mig 07:16, 12 February 2013 (UTC)

Well, ha- is a bit tricky, because it takes different forms. ha- + asdzáán = hweʼesdzáán; ha- + ííshgháán = hwííshgháán; ha- + anaʼí = hweʼenaʼí. I think it becomes hw- before most nouns that begins with a vowel (but that isn’t all there is to it). Of course, the other possessives occasionally are tricky, too, because they sometimes shift their vowel from -i to -e or -a to indicate a secondary type of possession: bi- + áłchíní = baʼáłchíní; bi- + hastiin = bahastiin; bi- + atsįʼ = beʼatsįʼ. Actually, the possessives often can take both vowels, -i and -e/-a, depending on the meaning: bi- + adeeʼ = bideeʼ (its horns that grow from its head, primary possession) or beʼadeeʼ (his horn that he took from an animal, secondary possession). I think the bound form is the same regardless of the possessive prefix. —Stephen ^(Talk) 08:10, 12 February 2013 (UTC)

Fun, fun stuff, layers within layers.

C.f. ha- and bi- alterations, the initial vowels in asdzáán, ííshgháán, and áłchíní are all verb prefixes, no? I dimly recall reading an etymology for áłchíní that described the word as ultimately meaning "one who has been brought out" (of someone else unspecified, hence the inspecific a-), i.e. "one who has been born"; and I thought I saw somewhere that asdzáán was ultimately a- + -s- (si-perfective) + -d- (classifier) + záán < same root as in sání? ííshgháán seems to be íí- (inspecific object a- + yíní) + -s- (si-perfective) + gháán (perfective root) < related to -ghá (“through”)... Do any deverbalized nouns that start with vowels have bound forms prefixed with inspecific a-?

Regarding secondary possession, I think bideeʼ decomposes into bi- + -deeʼ as the root bound noun, "his horn", whereas beʼadeeʼ decomposes into bi- + a- + -deeʼ, "his horn that originally belonged to some unspecified other party". I love how that fits together; it's quite elegant.

That said, I'm not sure how to account for all of these forms here in Wiktionary. I suspect that nouns with such secondary possessive forms are less common. Would it make sense to try to shoehorn all noun forms into one table? If so, it's sounding more and more like the irregularities of vowel and consonant shifts mean that any such template would require the user to put everything in manually, much like the current situation for verb conjugations. -- Eiríkr Útlendi │ Tala við mig 20:15, 12 February 2013 (UTC)

Well, ał- is yet another prefix. It’s a reciprocal duoplural object prefix that means something like "each other": áłká, ałkʼi. Yes, the secondary possessive forms are less common, and they seem to be becoming even less common. It appears that young Navajo speakers who are not truly fluent are likely to use the wrong prefix. It’s the same story with coronal harmony, as in sikʼis...many young speakers now say (and write) shikʼis. The shi- is supposed to become si- in harmony with the -s of -kʼis. We still consider this a mistake, but it is becoming very common.

I don’t think it would be very helpful to combine all forms into one table to appear on every page. I really don’t know the best way to handle it. I try to add an example or two on pages that take unexpected prefix forms (that is, I hope to do it, but it doesn’t always get done). —Stephen ^(Talk) 06:21, 13 February 2013 (UTC)

Verb forms

Moved from [[Talk:chʼiyąʼ]]

Further on the matter of the choice of lemma for Navajo verbs: I know this question was briefly discussed before at least once, but I do not remember where the discussions were, and I have not been able to locate them. For that reason, I’m just going to tack this additional note on it right here.

I happened across this brief mention of the reasoning that explains why it is the 1st-person singular imperfective that is normally considered to be the citation form for Navajo verbs. See How to use Young and Morgan’s The Navajo Language, part 5: Summary (almost the last page), where author Joyce McDonough says that "it is most helpful to know what the 1st singular ∅-imperfective form of a Navajo verb is". She adds that (speaking of Young and Morgan’s book) "all dictionary entries are given, if possible, in the ∅-imperfective (I), 1st singular form of the verb. If a verb does not have a 1st singular form, then it is given in the 3rd singular."

Seb az86556 has been rather ticked off that we use the 3rd-person form as the lemma here, and he thinks it’s important to correct it. I don’t know whether it really makes all that much difference, since all the forms will (eventually) be linked to the main entry anyway...but if we ever want to correct this, it would be better to do it while the number of verbs is still fairly low. —Stephen ^(Talk) 14:00, 17 March 2013 (UTC)

I also don't recall at the moment where that thread is. I do dimly recall not understanding why 1st-person forms were preferred. The link you give also doesn't really explain the reasoning, simply saying "[i]t is most helpful to know what the 1^st singular ∅-imperfective form of a verb is", but without stating any reason why. I haven't seen Y&M's reason for organizing their dictionary this way, and no book of theirs that I've been able to get my hands on so far seems to include any such explanation.

Meanwhile, there is an objective reason why not to use the 1st-person form as the lemma form for the imperfective mode -- 1st-person forms are lossy. The subject prefix obscures the verb classifier. Assuming knowledge of how Navajo infixes work, then when one knows the 3rd-person form, one can correctly derive the 1st-person and other forms. But if one only knows the 1st-person form, one cannot necessarily derive the correct 3rd-person or other forms.

By way of example:

• naashnish -- 1st person, classifier obscured. We cannot tell if this is -l-, -ł-, -d-, or -∅-. Any further conjugation of this verb requires that we know the classifier, which we cannot find in this verb form.
• naalnish -- 3rd person, classifier -l- clearly visible, allowing us to derive 1st-person naashnish, 2nd-person-duoplural naołnish, etc. With the classifier readily apparent, we can conjugate all other person forms for this verb mode.

In addition, some verbs semantically do not have 1st-person forms, but do have 3rd-person forms. It would be more consistent to give all verbs in 3rd-person forms, rather than to give some in 1st and some in 3rd.

I also note that Y&M's Analytical Lexicon tends to list verbs in both 1st-person and 3rd-person forms, one right after the other on the same line, presumably because the 1st-person form obscures the verb classifier. However, this listing format would only really work in the English > Navajo direction, which for our purposes would be restricted to translation table items under English term entries.

The only source I've seen so far that might target speakers of Navajo, rather than English speakers, would be Alyse Neundorf's Navajo/English Dictionary of Verbs. It's organized in an EN > NV direction, but the EN headword is immediately followed by a breakdown of the Navajo verb structure based on the 4th-person form, and then a brief explanatory sentence also using the 4th-person form. 4th-person does not obscure the verb classifier, but then again there are verbs that don't have 4th-person forms.

So far, I haven't come across any sound lexicographic reason to use 1st-person forms. If you or Seb or another editor has a reasoned basis for using 1st-person forms as the lemmata, I'm all ears. (Should we move this thread to WT:BP or Wiktionary_talk:About_Navajo?) -- Eiríkr Útlendi │ Tala við mig 15:43, 18 March 2013 (UTC)

Moved. ‑‑ Eiríkr Útlendi │ Tala við mig 17:23, 11 April 2014 (UTC)

Has this discussion been reopened (and possibly, the previous decision reversed)? The recent moves to 1st singular as lemma for several verbs seem to be the work of one person, but I wanted to check. If this is just a rogue user, I'm not looking to reopen the discussion. Ewweisser (talk) 13:14, 6 May 2016 (UTC)0

No, it has not been reopened. The recent moves were the work of one person, but I wouldn’t call him/her a rogue user. It could be anyone who knows Navajo, including on an academic level. Probably studied at Diné College, Navajo Language Academy (Diné Bizaad Naalkaah), or at any of the big universities that teach Navajo, such as University of New Mexico, University of Arizona, Northern Arizona University, University of California, Stanford University, California State University, University of Southern California, etc. —Stephen ^(Talk) 17:57, 7 May 2016 (UTC)

I'm guessing this will continue to come up every few years or so since you guys refuse to apply the academic standard. The lemma is the 1st person singular, and this cannot be argued by "logic" or your personal preferences. The major dictionary by Young and Morgan, the 40-year standard for academic study, lists it that way, and the textbook by Parsons Yazzie, officially adopted by the State of New Mexico for all High Schools and used at the University of Northern Arizona, introduces every verb that way. Most minor dictionaries follow this standard when listing a lemma. You can of course argue that the people who authored or co-authored these sources should not have a say in this and that you know better because you are more intelligent or have had a vote among each other. I don't know how wiktionary works. Wikipedia, at least, claims to rely on sources. Seb az86556 (talk) 06:51, 10 May 2016 (UTC)

Verbs: indefinite objects (i.e., a- forms)

Do we have any thoughts towards a policy or guideline regarding verb forms containing indefinite object prefixes? I ask mostly because of several such forms we have in Category:Navajo_verbs and Category:Navajo_lemmas. In some cases, no counterpart without the indefinite object exists, as is the case for alizh and alzhish. For some verbs, though, counterparts do exists, for example, yichʼiish for achʼiish, yidlą́ for adlą́, yitłʼó for atłʼó. Are these three verbs used with indefinite objects particularly often? I'd like to stress that I come to Navajo from a grammar approach, not a usage approach. I see the importance of considering usage, but have little to no knowledge thereof.

For this matter, we don't, as far as I know, have any forms with first-, second-, and fourth-person objects. They would qualify for their own entry pages, of course, but it's my impression that 1) no such form would "deserve" to be a lemma or separate verb, and 2) listing all of these (along with all their combinations with mode) on a lemma page would overwhelm the page and look terrible. I suppose tables that are compressed (or shrunk, closed, or whatever the terminology is) could work. I doubt this is anyone here's priority, but it does keep me thinking. Ewweisser (talk) 03:18, 25 July 2014 (UTC)

Another way of looking at verbs with indefinite objects is to think of them as transitive verbs: gohwééh łaʼ yishdlą́ = I’m drinking some coffee (not *I’m drinking it some coffee). The other verb is intransitive: doo ashdlą́ą da = I don’t drink.

A few months ago the WT:CFI was changed, and only attested forms and spellings are accepted now. I don’t know how familiar you are with the state of Navajo dictionaries, literature, and orthography, but there is no single dictionary that even attempts to list every Navajo lemma. Also, in most Navajo dictionaries, grammars, and texts, it is not at all uncommon to find typos and other misspellings, so they cannot be considered as infallible sources. Also, because of the extreme complexity of the language, most if not all Navajo dictionaries and glossaries also include a lot of nonlemma forms.

If you know a lot about the language, then you are aware that Navajo approaches language in a fundamentally different way...where English is object-oriented (every thing and object has a specific name), Navajo is description-oriented, and things and objects are described rather than named. For example, the word for planet: many people have gone against the natural grain of Navajo and created the term kéyah yádiłhił biiʼ hólónígíí (land/country the-one-that-exists in the jet-black-sky). It is very odd to think of Mars or Venus as lands/countries. In the natural way of speaking Navajo, you could say jóhonaaʼéí yináádáłígíí (the one that is walking around the sun). Thus: transit of Venus = jóhonaaʼéí yináádáłígíí Biinis wolyéhígíí éí jóhonaaʼéí yichʼą́ą́h ííyáago yichʼá̜á̜h siʼą́ą́ ńtʼééʼ (the one that is walking around the sun, the one that is called Venus, it blocks the sun as it travels, it was blocking it). I know that sounds awkward and foreign in English, but in Navajo it is natural and perfectly understandable. One of the upshots of this kind of system is that there is not just one way to say anything, but many ways. If ten Navajos formulated this sentence in their respective minds, they would have ten superficially different sentences, but all perfectly correct and understandable. Another upshot is that most Navajo words are not to be found in any existing Navajo language resource; therefore, most Navajo words are not acceptable under current CFI.

The best Navajo dictionaries, those compiled by Young and Morgan, adhere to the most common Navajo tradition of listing the 1st-person as the lemma of most verbs (some verbs do not have 1st-person forms), while 3rd-person forms are listed as modifiers (like adjectives). This means that a lot of Navajo verbs are attested only in the 1st-person, sometimes in the 3rd but with a different translation and explanation. This also relates to the CFI, since most of the verbs are attested only in the 1st-person.

Sorry about the long, rambling explanation. My point is that, because CFI rejects most Navajo words, although CFI does accept some misspelled Navajo words (because they can be found as misspelled in some book or other), and since Navajo is description-based rather than object-based (which runs counter to our editors’ concept of SOP), it seems to me that the Navajo language is simply incompatible with English Wiktionary. When the CFI was changed and some good Navajo terms were deleted by editors that know nothing about the language, I pretty much stopped creating any more Navajo entries. I add some Navajo words in the Translation sections, but I rarely create Navajo entries and have stopped adding conjugations to the verbs.

I was adding audio files for all of the entries, but since many of the existing entries are potentially deletable as unattested, I no longer mess with audio. —Stephen ^(Talk) 08:44, 26 July 2014 (UTC)

Digression about transitivity

Digression about transitivity

• It's a minor thing, but drink is still transitive (and probably adlą́ too), even if the object isn't explicitly stated. C.f. w:Transitivity (grammar) for more details. ‑‑ Eiríkr Útlendi │ Tala við mig 18:30, 30 July 2014 (UTC)
  • Perhaps I misunderstand you, but in sentences like "you can't make a horse drink" or "I don't drink", "drink" is certainly not transitive. - -sche (discuss) 21:57, 30 July 2014 (UTC)
    • I think we must be using different ideas of "transitive". The perspective on "transitive" that I'm coming from is that, semantically, certain actions require an object, even if that object is left unstated. One cannot drink without drinking something. Similarly, one cannot eat without eating something, and likewise for write, or take, or watch, etc. The something need not be stated, but it remains a logical requirement in order for the action to take place. ‑‑ Eiríkr Útlendi │ Tala við mig 08:49, 31 July 2014 (UTC)
      • I was looking at it from a grammatical perspective. From that perspective, there's a basic distinction between transitive verbs and senses of verbs, which have (actual, stated) objects, and intransitive verbs and senses, which don't. This is the distinction that {{context|transitive}} and {{context|intransitive}} and their associated categories make, especially on entries like die and be where they both occur on different senses (and sometimes even on the same—ambitransitive—senses). I can see what you're getting at about some verbs seeming to semantically require objects, but it seems like it would be inherently hard (and subjective) to pin down which verbs those are. For example, does the one transitive sense of die effectively point out that die is semantically transitive and always requires "a death" as its implied object? Does this transitivity carry over to its synonym "perish"? (It seems unlikely that anyone would go along with the idea that "perish" is transitive.) Or consider sense 2 of breastfeed, the sense used in sentences like "the baby liked to breastfeed". It seems functionally identical to the senses of eat and drink that are used in "I like to eat" and "I don't drink", so if eat is semantically transitive, that sense of breastfeed would seem to be semantically transitive, too... yet it seems to be outright impossible to supply it with a stated object. (Obviously, sense 1 can be supplied with an object, but sense 1 is comparable to feed (“provide with food”) rather than to eat.) - -sche (discuss) 10:01, 1 August 2014 (UTC)
        • There is a lot of middle ground for many verbs, as you note. For die and similar verbs, for instance, I would argue that these are primarily intransitive verbs, that have some idiomatically transitive senses. For verbs like stop or break, transitivity is context-dependent: is the subject performing the action on something, or is the action occurring solely within the scope of the subject? Some verbs like feed are even more complex grammatically and semantically. Is feed used to mean "to provide sustenance to another" as in I feed the mouse to the snake, or is it used to mean "to provide another with sustenance" as in I feed the cats, or is it used to mean "to sustain oneself" as in the hamsters feed on the grain, or is it used to mean "to pay out, to direct along a course" as in he feeds the cable down the pipe, or is it used to mean "to run into, to flow into" as in the river feeds into the lake? Ultimately, these are different concepts involving different kinds of action, some applied to an object and some not. Other verbs are much more conceptually cut and dried, like drink or say or write, where something is always the object, whether or not it's explicitly stated or even relevant to the context.

          I was taught that any verb or verb sense where you can meaningfully ask the question what was [verb]ed? is transitive. I've learned through the course of this thread, and the reading it's prompted me to do, that many folks use “transitive” instead to mean specifically that there is an explicitly stated direct object. I'm happy to leave it at that, and know for future discussions that I'll want to first clarify what “transitive” means to discussion participants. ‑‑ Eiríkr Útlendi │ Tala við mig 18:10, 1 August 2014 (UTC)

      • adlą́ is not transitive, it cannot take a direct object. Your definition of transitive is not the definition that most people use. —Stephen ^(Talk) 12:15, 31 July 2014 (UTC)
        • The initial a- prefix appearing here in adlą́ is described by Young and Morgan, Faltz, and Goossen as an unspecified or indefinite object. Since this verb form has an object built right into it, that would suggest transitivity. The action of drinking itself requires an object, regardless of language, again suggesting transitivity. And what "most people" would use is largely irrelevant when it comes to specific technical terminology, such as grammatical labels. What grammarians or linguists would use is much more pertinent, which is why I linked through to the [[w:Transitivity (grammar)]] article in my previous post. In light of your terse reply, I'm not sure if you disagree with that article, or if you just didn't read it.

        Reading around online, it seems like there are two broad camps: 1) those who view verbs as transitive only if they have an explicitly stated object, and any other context as intransitive; and 2) those who view verbs as transitive based on the kind of action involved and whether that action involves a direct object, regardless of whether that object is explicitly stated. I don't think either is necessarily "more" right than the other, as ultimately it boils down to how one defines transitive. Until that is made clear, however, there is room for confusion, as amply demonstrated in this thread. ‑‑ Eiríkr Útlendi │ Tala við mig 08:38, 1 August 2014 (UTC)

        • I don’t know what to tell you. adlą́ does not take a direct object. It is intransitive. It can be used with postpositions that may be translated into an English sentence that uses a direct object, but in Navajo it is not a direct object and not transitive. yidlą́ has the object built into it and yidlą́ is the verb that is transitive. Many American Indian languages have overt marking for transitivity on the verbs, and no matter how you define transitive for Indo-European languages, American Indian languages that mark transitivity have only the definition that I use. Verbs that are marked as intransitive are intransitive no matter how you may think of verbs in English. With these languages, there is no confusion over transitivity, no room for doubt. It is pointless to argue about it, you can’t use adlą́ with a direct object. In the sentence, tó yishdlą́ą́ʼ (I drank some water), you cannot replace the verb yidlą́ with adlą́. If you want to create Navajo example sentences using adlą́ with direct objects, go ahead. They will be incorrect and ungrammatical. Let’s drop this argument here, it is going nowhere. Think of Navajo verbs as you like. —Stephen ^(Talk) 10:07, 1 August 2014 (UTC)

          • Thank you for stating your position. I now understand better where you and -sche are coming from and what your use of “transitive” means here. I'm happy to drop this. ‑‑ Eiríkr Útlendi │ Tala við mig 18:10, 1 August 2014 (UTC)

WT:CFI and Navajo

• I've been mostly inactive over the past several months due to time constraints, so perhaps I missed something. Stephen, what was the change to WT:CFI that led some editors to delete Navajo terms? The most likely candidate edit I found when going through the first chunk of history listings is this one from September 2012, but even that makes allowances for languages that aren't well documented online, and I think Navajo would have to belong to that category. I'm hoping that the deleting editors were just confused, and that there isn't any actual policy that would dictate the removal of otherwise valid Navajo entries. ‑‑ Eiríkr Útlendi │ Tala við mig 20:45, 30 July 2014 (UTC)
  • If I may comment, as someone who has followed the recent evolution of CFI:
    There has been no change to how inflected forms are handled. As in the past, inflected forms are allowed if they meet CFI, and in highly inflected languages, specific inflected forms may have entries even if they are not attested, so long as the basic inflection is valid. (For example, mitternachtsblau has some 52 distinct inflected forms. If the masculine singular dative mixed declension form is found to have only two — or even zero — Google Books hits, but enough other inflected forms are attested to confirm that mitternachtsblau does indeed inflect, then it's not like we're going to create a one-off inflection table for use in mitternachtsblau that has a gap in that one slot, and it's not like we're going to delete mitternachtsblauen or redefine it to say "masculine singular strong genitive and accusative, weak genitive, dative and accusative, and mixed genitive and accusative but not dative, and feminine weak genitive and dative, and mixed...". If necessary, I can dig up the discussion where the community affirmed this over the objections of only one user.)
    A change was made in the past few years to how lemmata are handled, but that change was to allow more words, namely to allow any words that are even so much as mentioned in a single durable reference. Terms like Talk:jádí dághaaʼígíí which are not even mentioned in any references continue to be disallowed, like always; this last part seems to be what some user(s) dislike(s). - -sche (discuss) 21:54, 30 July 2014 (UTC)
    • Interesting, thank you. I thought CFI restrictions had been eased somewhat over time, but I know I've gotten the wrong end of the stick in the past, so thank you for the corroboration. ‑‑ Eiríkr Útlendi │ Tala við mig 08:49, 31 July 2014 (UTC)
      • Until Wiktionary:Votes/2012-04/Languages with limited documentation, languages with limited documentation (and especially polysynthetic languages, since none of them have complete dictionaries and will never have complete dictionaries) depended on editors with knowledge of the language and the standard orthography to determine whether a given word existed and was properly spelled. Since that vote, knowledge of the language is no longer a factor (not even if the language is the editor’s mother tongue), and only words found in existing dictionaries or texts are accepted as entries (even if they are misspelled, which is often the case with languages such as Navajo). There is also the problem of SOP ... Navajo is fundamentally a descriptive language, and it usually favors SOP descriptions over European-style one-word nouns. —Stephen ^(Talk) 12:15, 31 July 2014 (UTC)
        • Thank you for linking through to that vote. I note the line, the language community should maintain a list of materials deemed appropriate as the sole source for entries based on a single mention -- is there any such list for Navajo? Could that list include the NV WP, for instance, given that this might well be one of the largest curated collections of text written in Navajo and accessible online? That would be one way to avoid the problem of inadequate and incomplete dictionary materials. ‑‑ Eiríkr Útlendi │ Tala við mig 08:38, 1 August 2014 (UTC)
          • There is no list of materials for Navajo. Such a list cannot include NV WP. It is believed by the other editors, although incorrectly, that printed Navajo dictionaries and glossaries may be trusted to use correct spelling and have correct translations. In fact, all of them that I have examined contain typos and misspellings to some degree, and the translations are often misleading, or even incorrect. Navajo words taken from any source should be vetted for correct spelling, translations, usage, etc. As far as I’m concerned, you can enter Navajo words that you find in printed materials and treat them as you like. I don’t do it myself and I do not look at such entries. —Stephen ^(Talk) 10:07, 1 August 2014 (UTC)
          • @Eiríkr, re "Could that list include the NV WP": no, the requirement that citations be durably archived has not changed. Furthermore, WT:CFI#Attestation explicitly disallows Wikipedia as a reference for attestation purposes. - -sche (discuss) 10:17, 1 August 2014 (UTC)
            • I was wondering if those same requirements apply to less-documented languages. It isn't terribly clear from the WT:CFI page. I also note that the WT:CFI#Attestation list of three requirements appears to be an “or” list, making the “use in permanently recorded media” requirement just one condition that could be met to comply with WT:CFI, rather than one that must be met. This raises the question of whether extensive use of a term on the NV Wikipedia, which may be the only available, current, and online resource discussing certain subjects, would count as “clearly widespread use”, given that “widespread” for less-documented languages is much harder to gauge for those of us not living in those language communities. This seems to loop back to Stephen's concern that WT:CFI is ultimately being used in a way that devalues native-speaker knowledge. That doesn't seem right to me, and I'm wondering if there's a way past that. ‑‑ Eiríkr Útlendi │ Tala við mig 18:10, 1 August 2014 (UTC)