User talk:Julien Daux

There was some talk a while back about adding entries for the verb stems themselves. I think this would be a fabulous development -- it would vastly improve the usability of the site for Navajo, and allow a user to see all the different verbs built from a single root. The root entries could include the various conjugations, perhaps even with etymologies (on each appropriate entry) showing how these developed (for instance, I recall reading that many of the perfective and neuter imperfective forms arose through fusion with a terminative ni suffix, probably originally of the same origin as the terminate ni prefix; also, that future forms arose through fusion with an l suffix of some sort).

I don't have the expertise, resources, or time to engage in any such effort. Would you be interested in this? ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:06, 9 September 2016 (UTC)Reply

This is actually my ultimate goal :) but it surely is a daunting task... I hope I'll get enough courage to get through the first phase at least, while the number of verb bases is still low : linking between all verbs, stems and roots. Eventually, each root page would contain all the information you mentioned : the Proto-athabascan root etymon, the stem sets, the stem matrix (with the aspectual additional dimension),... The derivation of the stem sets is yet another very ambitious story, as it has been very little researched so far (cf. works by Leer and Hardy, both dating from the 70's...).

I myself do not know how much energy I'm willing to put in this endeavour, but I'll do my best.

One fear I have though, is that as per the copyright restrictions, I don't know how much I can copy from my sources. The stem matrix one can find in "The Analytical Lexicon Of the Navajo Language" by YMM, are for instance very tentative hypothesese by the authors, that varied greatly across the past 50 years.

But I'm not there yet. You may have noticed I started playing around with the categories, the links, etc.. trying to work out the best way to represent the full data. I'm very new to the Wiktionary world, and I have still a lot to learn. Once I get my POC ready, I'll expand to the other existing bases. Julien Daux (talk) 21:49, 9 September 2016 (UTC)Reply

We don't have a "Meaning" header- you can just put the definition on a line starting with "#". DTLHS (talk) 00:47, 10 September 2016 (UTC)Reply

Hi, Julien. For Navajo language, please always use ʼ (Unicode u+02bc) (not '). Sometimes it may not be easy to see the difference, but if you double-click a word that includes ʼ, it will highlight the entire word: eʼeʼaah. If the word contains ', the highlighting will be blocked by the ', as in: e'e'aah. This is because the Navajo letter ʼ is an actual letter, just like a, b, or c. On the other hand, ' is punctuation, and punctuation acts as a barrier to highlighting a single word.

Also, note that the ==Etymology== heading goes above the part of speech heading (such as Root, Noun, Verb, etc.). Thanks. —Stephen ^(Talk) 16:46, 11 September 2016 (UTC)Reply

Sure, I saw all your edits this morning and I started adopting this convention before your message. I also made a pass on all the root entries and fixed the Etymology section's location or formatting. Thank you for letting me know! Julien Daux (talk) 19:24, 11 September 2016 (UTC)Reply

I suggest you make a template for this rather than putting raw table markup into entries. If you're not familiar with templates, you can just make a page such as Template:nv stem set, or another title of your choosing in the template namespace, with the table code in it. Parameters can be passed with {{{1}}}, {{{2}}}, {{{abc}}}, etc. DTLHS (talk) 20:18, 11 September 2016 (UTC)Reply

The problem is that the stem set table can have a varied numbers of row every time. I'm not sure if this can be templatized, unlike conjugation tables... Julien Daux (talk) 20:37, 11 September 2016 (UTC)Reply

I see, well if you want you can use {{#if: {{{1|}}} | text if parameter 1 is defined | text if not defined }}. If you don't want to bother that's fine. DTLHS (talk) 20:45, 11 September 2016 (UTC)Reply

I'm curious why you're creating verb root entries in all-caps? ‑‑ Eiríkr Útlendi │^{Tala við mig} 04:13, 12 September 2016 (UTC)Reply

Lower case for stems, upper case for roots... That's how I got used to. Roots are taken conventionally as their perfective Momentaneous stems as they are considered the most conservative. The star (*) symbol will be kept for unattested, reconstructed roots of Proto-Athabaskan. Julien Daux (talk) 04:56, 12 September 2016 (UTC)Reply

• The star for unattested and reconstructed roots is a good practice. I believe that reconstructions also go in a special namespace -- have a look at entries like Proto-Indo-European *wódr̥ (“water”) for an example.

As far as capitalization, I don't think all-caps is a convention that Wiktionary uses at all -- I believe that the Wiktionary standard is to keep everything lower-case, unless a term is actually used in upper-case. For instance, all Proto-Indo-European roots are given in lower-case. Stephen is much more active in Navajo, and in fact in numerous other languages, than I am (I keep mostly to Japanese). I suspect he might have more insight. ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:26, 12 September 2016 (UTC)Reply

Stephen has been updating a number of my root entries over the past days (mainly for layout and typographical conventions like curly apostrophe instead of straight apostrophe for the electives), and never said anything regarding me using all uppercase... I actually started up with lower case then switched to uppercase, requested for deletion of the few lower case I had created. Stephen kindly deleted these ones. So I believe it's fine. Julien Daux (talk) 11:08, 12 September 2016 (UTC)Reply

Also, is there a space in Wiktionary where I can put up a page regarding the conventions I made, how to use the templates (I created a couple new ones), etc... ? Julien Daux (talk) 11:25, 12 September 2016 (UTC)Reply

Every template has a documentation subpage. There is also Wiktionary:About Navajo (you may want to discuss changes to this page with other editors before committing them). DTLHS (talk) 15:02, 12 September 2016 (UTC)Reply

• One functional concern about capitalization of roots is that roots are often identical to at least one form of a verb's stem paradigm. Take smell, for instance. The root is currently given as -CHĄ́Ą́ʼ₂. Numeric subscript aside (incidentally, this subscript is confusing and isn't explained anywhere), this is identical to the perfective and optative verb stems in all but capitalization. One should be able to click on a link to the perfective or optative stem -chą́ą́ʼ and see the root entry on that same page.

Are there any lexical reasons for capitalizing the roots? ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:13, 12 September 2016 (UTC)Reply

PS: Entry pages can be moved to different spellings. This is usually easier than re-creating an entry under a different spelling, and then having to delete the old spelling. ‑‑ Eiríkr Útlendi │^{Tala við mig} 19:15, 12 September 2016 (UTC)Reply

No lexical reasons, except that roots are different from stems, and are proxied by the mom-perf stem... What is the use case / context where you would have a perfective or optative -chą́ą́ʼ, and would liked to click on it ? and then, let's say this type of context exists (within wiktionary), in the case where it would be the imperfective or future stem or from some other aspect, no link would be provided to the root? The fact that the optative happens to be equal to the perfective shouldn't give it a better status. I need to work out a better system for linkage between stems and roots. So far, I worked out a way for verb to root, root to derived verbs, but I didn't do stems to root. Ideally, even a disambiguation page on stems would be needed as a stem can originate from many different roots of different aspect and mode.

The subscript is because some roots happen to be homophone too with different etymological derivation :). At some point, they'll need to be distinguished. I didn't want to put both roots in the same wiktionary page because it would have made by verbs-to-root redirection very messy and confusing. So I decided to distinguish them from the start.

Julien Daux (talk) 20:14, 12 September 2016 (UTC)Reply

Can I use the "About Navajo" page to explain these conventions, what type of links exist between verb entries, prefixes, stems, roots, and conjugation paradigms ? I'm building more and more categories right now. I'll keep on a limited number of verb entries (letter A so far), and see what other editors think of it ? Thank you. Julien Daux (talk) 22:41, 12 September 2016 (UTC)Reply

• (after edit conflict...) About the links, the mode is irrelevant, and I'm sorry that confused you -- my basic point is that any page that includes an entry for a verb stem form should also include an entry for the verb root on that same page, if that verb root happens to share the same spelling as the verb stem form. Because we were discussing the root -chą́ą́ʼ above, I mentioned the perfective and optative since those two verb stem forms match the verb root form. As you note, all verb stem forms should ideally have their own entries as well.

Incidentally, this -chą́ą́ʼ verb root seems related to the noun chąąʼ (“excrement”).

About use case, I think any verb or noun form should generally be a link: the user can learn a lot by clicking through and reading about the details. I, myself, often find I learn a language faster when I can understand the building blocks of its vocabulary.

About the numbers, I'm not aware of any other case here where homophones that share the same spelling are distinguished in the headword by means of numbers -- at Wiktionary, this is done instead by creating separate etymology sections, such as ===Etymology 1===, ===Etymology 2===, etc. See also Proto-Indo-European *dʰew- for one such example page. When linking, a gloss or other indicator can be provided to tell the user explicitly which portion of the targeted page is intended, and/or some other device is used to link directly to that section (compare the formatting of the links to *dʰew- at die#Etymology_1 and θέω#Etymology). This same structure would be used when multiple verb stems from different verbs all happen to share the same spelling -- the page would have multiple etymology sections, with the related stems grouped under the appropriate section. For reference, one extreme example of a single spelling with lots of different component parts is the Japanese entry 柄. ‑‑ Eiríkr Útlendi │^{Tala við mig} 22:43, 12 September 2016 (UTC)Reply

OK, I see. I'll rework my entries and will add additional interlinks between all these part of speech. I'm trying to do it template-based to help me avoid to have to copy many entries in many places. For example, I see that the roots in proto Indo-European have their own list of derived words broken down by family. This is horrendous to maintain : upon every new verb entry, you'd have to update the root entry as well, and other categories as well, all manually.

I myself got confused between chąąʼ and chą́ą́ʼ, that's how the subscript comes from ;) The two roots may be related (feces vs smell) but they are kept separate in Young. I'll check their Proto-Athabaskan etymology too.

Back to the point : What do you suggest to distinguish roots from stems? Inasmuch as I understand I can squeeze multiple etymologies in one root entry for multiple homophonous roots, I fear that adding stems in the same root entry would start to be very messy especially when there are so many homophones. I also fear it would make it more difficult to streamline and templatize. But if that was your original concern, it's my goal that clicking on stems brings you to roots (actually, clicking on stems will bring you to a disambiguation page that'll show all roots sharing that same stem). Julien Daux (talk) 23:36, 12 September 2016 (UTC)Reply

Also, I wanted to add something regarding concerning my choice for the upper case for roots: in other languages which also have a similar root system (abstraction that is synchronic to the language), roots are distinguished in Wiktionary in a specific way. Hebrew uses dashes between the 3 consonants, Arabic uses the isolated shape for each of the 3 consonants. Since stems (a real syllable) and roots (an abstract entity) both start with a dash to indicate their left-bound nature, what else can I do to distinguish between these two categories ? Julien Daux (talk) 02:10, 13 September 2016 (UTC)Reply

• Hi, sorry for the delay -- I'm sick at home and a bit loopy, so this reply is just to let you know that I'm not ignoring your questions.  :)

I think a mock-up entry might be useful as an illustration. Given other things IRL at the moment, and the fact that my sick day today means I'm behind at work, it will be several days before I can pull something together. Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 00:09, 14 September 2016 (UTC)Reply

Concerning the use of uppercase for roots, I originally considered using that to distinguish roots from stems, which is how Y&M did it. I really like that idea, but it seemed to be a departure to some extent from previous Wiktionary practice, so I didn't go that way. However, I like the idea, and I don't see that it should cause confusion, since roots are not words, and capitalization is irrelevant for them.

As Julien says, Navajo roots are complex and confusing, in more ways than one. For one thing, there are numerous roots that are spelt the same but that are not related. And often a given spelling might correspond to one tense or aspect of one root, and another tense or aspect of a different root. Y&M differentiates homophone roots by using numeric subscripts. Example: -CHĄ́Ą́ʼ₁ (“exhausted”), -CHĄ́Ą́ʼ₂ (“smell”), -CHĄĄʼ (“excrement”). I don't have a strong opinion about whether we use the numeric system that Y&M prefer or if we list all homophones on one page. However, without numerics, it seems more difficult for a user who is sent to a page with multiple unrelated roots listed, and who has to study all the roots to find the one that pertains to his case. Also, it is common that some stems of a root are differentiated only by tone, as -chééh (momentaneous) and -cheeh (continuative), and that a stem such as -chąąʼ could logically be thought to belong to either -CHĄ́Ą́ʼ₁ (“exhausted”) or -CHĄ́Ą́ʼ₂ (“smell”), when in fact it is neither. So I am open to the use of uppercase to set off roots, and numerics to distinguish between unrelated homophones, but other solutions are possible.

Yes, explanations of these conventions should go on the Wiktionary:About Navajo page.

My preference is to follow the Y&M style, keeping uppercase for roots (perhaps without the initial hyphen since roots are theoretical) and lowercase for the stems (with initial hyphen and classifier, if any ... these are the trickiest forms, in my opinion, since different classifiers can have a big effect on the spelling and pronunciation of the stem). —Stephen ^(Talk) 11:18, 15 September 2016 (UTC)Reply

So, I finally decided to drop the subscript in the roots, for several reasons:

1) it is not a Wiktionary convention. That's the least of them

2) the number itself is pretty arbitrary: when I'm entering my first XYZ root, should I use a subscript "1" just in case a "2" make come later? Should I use no subscript for the first entry, and use subscript starting the second one only? If I'm entering a CHĄ́Ą́ʼ root for the first time, should I use YM's subscript number even if it's the first root I'm entering it in Wiktionary? This may cause a lot of confusion

3) Young himself complains about very disparate meanings in some roots, but didn't deem necessary split them further down and number them separately. Comparing YM1987 with YMM1992, it also shows that Young changed mind a couple times on some roots. Even YM1987 alone is full of contradictions : between the root index, the root / stem / theme index and the dictionary body, some roots are treated in quite inconsistent ways.

On these premises, I decided to keep one root (proxied by the Momentaneous perfective stem) in one page, to separate etymologies in two or more sections if YMM1992 does so (this latter book seems somewhat more consistent throughout, but not always), and to attribute glosses to the various core meanings a root can have. Below the gloss will be shown the derived themes. Link to root categories must include the gloss (gloss replaced the arbitrary subscript). The only technical issue I have is that clicking on a root in a verb entry doesn't jump directly to the appropriate gloss (working in it...).

I went by great lengths deciding how much material needed to be shown in the root pages, but I think I'm stabilized by now. Some roots showed to be a real pain to neatly lay out and summarize. I believe that the point of a root page is to show its meanings + thematic material, and I got lost recording all valence nuance depending on classifier shifts (causative, transitive, mediopassive, reflexive, detransitive,...). I believe valence doesn't totally belong in the root page.

In the root page, after the glosses, appears the stem set matrix. Are shown only the aspects that I came across while refactoring the A-letter verbs, not all aspects shown in YMM1992 (which again is thousand times more consistent than YM1987 in this matter also; there are cases where both books do differ substantially). Clicking on a stem brings you to a category page showing all roots sharing that stem.

Finally, I also refactored parts of the verb pages. The Etymology section has been heavily templatized as a lot of repetitive material appears there that is not consistent throughout. Categories will be added that'll show all verbs sharing the same prefix. To the conjugation section was added a "Paradigm" indication (durative (Ø/si), semelfactive (yii/si), Neuter imperfective (ni),...), categories will be added that'll show all verbs sharing the same paradigm. Once again, precedence has been given to YMM1992 since a huuuuge clean-up has been performed there, aspects were quite messy in YM1987, and never mentioned in the dictionary's body itself. Additional categories will be added for irregular verbs, plural-only verbs, 3rd-person only verbs. I also need to sort out that ni-imperfective vs ni+Ø-imperfective vs ní+Ø-imperfective thing as both YM books, YM2000 ("An Overview") and Faltz show contradictory denominations. I guess Navajo mostly fails to enter into neat boxes and we just need to accept its uncategorizability ;)

Thank you for reading up to here and feel free to tell me what you think! Julien Daux (talk) 04:48, 19 September 2016 (UTC)Reply

The root for both is given as -ZĮĮD, and the classifier for both is effectively given as -Ø- (bééhózin has no classifier mentioned at all). However, most verb stem forms for transitive yééhósin start with S, not Z.

Is the classifier for yééhósin perhaps -ł-? -ZĮĮD preceded by a -ł- classifier would devoice to -SĮĮD forms, as we see in the yééhósin entry. Or is something else going on that can explain the -z- to -s- shift in the verb stems? ‑‑ Eiríkr Útlendi │^{Tala við mig} 01:14, 23 September 2016 (UTC)Reply

Yes, yééhósin gets its hard s because of the ł classifier. I'm currently formatting my edits, so you may have seen intermediate results where I had an issue with one of my templates (hence the missing classifier). This verb is very confusing, with 5 distinct bases: yééhoosįįh, bééhoozįįh, yééhósin, (doo ádił) yééhólzin, bééhózin. Was figuring my way out through those while editing... Julien Daux (talk) 01:27, 23 September 2016 (UTC)Reply

The usual syntax is "X verbs suffixed with -Y" (for example Category:Hungarian verbs suffixed with -ekedik). Do you mind if I edit {{nv-prefix}}? Or is there a distinction between "prefixed with" and "with prefix"? DTLHS (talk) 18:31, 24 September 2016 (UTC)Reply

(by the way this will allow you to create the categories just by using {{auto cat}}) DTLHS (talk) 18:36, 24 September 2016 (UTC)Reply

The "problem" is that some prefixes have a more specific denomination, like a classifier, a postposition complex, or a personal prefix (subject or object). Using "with", while admittedly ugly, allows me to keep parallel formulations like "verbs with prefix X", "verbs with classifier X", "verbs with postposition X", "verbs with subject prefix X", "verbs with object prefix X". "Verbs prefixed with classifier X" sounds weird (because even though classifier are technically prefixes and do occupy a "prefix slot" (number 9),they are rarely named so in the literature, I believe due to their bound nature to the root), as does "verbs prefixed with object prefix X", which is a little repetitive ... You may want to only change the "verbs with prefix X" and keep the others untouched, but I think it allows for more clarity to keep categories parallel.

Regarding autocat... Do I still have to create the new category's page, or can these be automatically created as soon as a word entry points to a category? Julien Daux (talk) 19:02, 24 September 2016 (UTC)Reply

If the category name fits with the expected pattern I automatically create it after a few days (I have a bot that analyzes Special:WantedCategories) DTLHS (talk) 19:05, 24 September 2016 (UTC)Reply

{{rfd}} is used when the deletion should be discussed. {{d}} can be used for uncontroversial deletions (a category is obsolete or misspelled, etc.) like the ones you are currently requesting. —suzukaze (t・c) 23:34, 24 September 2016 (UTC)Reply

Thanks for the tip! Julien Daux (talk) 23:41, 24 September 2016 (UTC)Reply

What are you using this for? I can't see any reason why it should exist; we don't need templates to strip diacritical marks from a word. —Μετάknowledge^{discuss/deeds} 07:22, 25 September 2016 (UTC)Reply

Because otherwise my words are not properly sorted. Ą comes after Z instead of A. Julien Daux (talk) 07:29, 25 September 2016 (UTC)Reply

Isn't the -s- infix the perfective "s"? ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:33, 1 October 2016 (UTC)Reply

The si-perfective conjugation (or "s-P", by Faltz's nomenclature), goes this way : 1. sé- / sis-, 2. síní, 3. s- / z-, 1dpl. sii(d)-, 2dpl. soo(h)-. So this -s- infix is actually the shape taken by the modal-aspectival si in the 3rd person. The linear decomposition in the etymology section sometimes doesn't allow to properly render some non-linear phenomena, like the fused nature of the modal-aspectival prefixes (Ø,ni,si,yi) with the personal prefixes. Additionally, there also is a si-imperfective conjugation, albeit much rarer, so si is not limited to the expression in the perfective "mode". Julien Daux (talk) 11:29, 1 October 2016 (UTC)Reply

Do they have definitions of their own? Also does it make sense to put the table in "This stem appears in the following roots:" in a derived terms section? DTLHS (talk) 18:58, 15 October 2016 (UTC)Reply

Well, stems are the forms roots take for a given aspect + mode combination. In a way, they are to roots what verb forms are to verbs, and verb forms have their own page in the non-lemma section. The table I'm showing is just a way of stating the below in a tabularized hence more easily graspable fashion:

1. Momentaneous Imperfective form of Root1
2. Durative Optative form of Root2

They have a meaning inasmuch as the roots they are the derivative of have a meaning. Like verb forms, the same stem shape can be shared by multiple roots for different aspect and mode combinations, just like English "I found", present of "to found" and past of "I find", but in a more horrendous way because these overlaps occur for almost all stems (hence the need of a reverse stem-to-root index).

Also, using pages as opposed to categories will benefit people wishing to look up a verb by the stem (which is how verbs naturally occur), while categories are more difficult to browse. The stem page will then redirect the user to all applicable root candidates. Julien Daux (talk) 19:36, 15 October 2016 (UTC)Reply

Please deal with these categories that I believe you have created. --kc_kennylau (talk) 16:08, 11 November 2016 (UTC)Reply

Sure, I will. Sorry for letting this stand out for a while. Julien Daux (talk) 18:03, 11 November 2016 (UTC)Reply

Julien, in ałtso, there is a problem with {{nv-prefixes|á-comp|ni-3s-Ø|ł}}. In the etymology, it shows the classifier -Ø-, but if you click on it, it takes you to Ø-, which doesn't exist. —Stephen ^(Talk) 12:50, 12 November 2016 (UTC)Reply

As I see it, classifier is correctly shown as -ł-. The -Ø- segment is the "third person subject prefix" of the ni-imperfective paradigm, which still doesn't have its own entry (just like many many other prefixes, but I'll create them all soon hopefully, after finishing the verbs). Julien Daux (talk) 14:08, 12 November 2016 (UTC)Reply

Oh, that's right. I must be half asleep. For some reason, I did not notice -ł- and I was reading -Ø- as the classifier. Nevermind. —Stephen ^(Talk) 14:34, 12 November 2016 (UTC)Reply

We do not usually use colon-space in category names. Please allow me to change the names. --kc_kennylau (talk) 10:23, 8 December 2016 (UTC)Reply

You want to replace them with colon without space or without colon at all? For instance, what would "Navajo verbs: durative (Ø/yi)" become? Julien Daux (talk) 12:45, 8 December 2016 (UTC)Reply

Something like "Navajo durative verbs (Ø/yi)". You can explore the sub-subcategories of Category:Verbs by inflection type by language to see how this is done in other languages. --kc_kennylau (talk) 15:00, 8 December 2016 (UTC)Reply

Actually, except a couple of these categories for Navajo (like Category:Navajo irregular verbs, which are already without colon), most of the other categories are actually not "inflections" types, but aspects. They should probably better be categorized in a supercategory called "Navajo verbs by aspect". If I'm recategorizing them in this new super-category, am I ok with the colon in the name? — This unsigned comment was added by Julien Daux (talk • contribs) at 18:42, 8 December 2016 (UTC).Reply

Please avoid colons in all category names. --kc_kennylau (talk) 09:41, 9 December 2016 (UTC)Reply

 Done Julien Daux (talk) 15:51, 10 December 2016 (UTC)Reply

Thanks. --04:36, 11 December 2016 (UTC)

Please deal with the categories on these two pages. Please also check if other pages are categorized in similar categories. --kc_kennylau (talk) 10:40, 11 December 2016 (UTC)Reply

Julien, if you ever find that you don't have anything to do, or if you just need a change in scenery, we would love to have some help at w:nv:Special:Watchlist. There are thousands (maybe hundreds of thousands) of misspelled words and other mistakes that need to be fixed. —Stephen ^(Talk) 22:31, 14 January 2017 (UTC)Reply

Stephen, I clicked on the link but can't see any words in the watchlist?

Anyway, I'm far from being idle at the moment:) I initially wanted to finish the entry pages for prefixes and roots, but it came to my attention after many typos from my side trying to add new verbs, that the next, much-needed thing to work on was a conjugation module. It is much harder and more intricate than I thought, and I haven't finished the imperfective yet... (planning on working on perfective and future next, but not more). That's why you may have seen my typo hunt game-- these came from me running the module against all current 700+ verb entries of the Navajo Wiktionary.

Back to work now :) Julien Daux (talk) 23:25, 14 January 2017 (UTC)Reply

Thanks. No rush, just if and when you have the time and the inclination. The watchlist doesn't show you anything because it will only show the pages that you have edited or otherwise have marked for watching. A good place to look is w:nv:Special:AllPages, which will show everything (even redirects).

Yes, I saw that you were working on typos. I did not think that a translation module would be possible for Navajo. An exciting possibility! —Stephen ^(Talk) 10:08, 15 January 2017 (UTC)Reply

Actually, this had me wonder for a while... Where do all those technical terms for species coming from in the Navajo Wikipedia? Like mollusque, Bijáád Ádaadinígíí, etc? Julien Daux (talk) 01:29, 18 January 2017 (UTC)Reply

Some of them are from the common way of forming translations for unusual terms by looking at examples and taking into considering the meaning in English, and then forming a descriptive native term that fits (as in the ones that are without legs). Some of them required quite a bit of work and input from a number of speakers. For instance, in modern Navajo society there are a lot of Navajo people who marry outside the tribe (either to bilagáana or other foreigners, or to Indians of other tribes), and when their children start growing up, they want to find out how to introduce themselves in the Navajo way. For tribes outside of the Navajo area, they usually ask what a tribe calls itself, and also what the name means, and then they choose a term based on that information that a person can use to introduce himself. Often they will use a broad term such as Naałání or Bitsiiʼ Yishbizhí, but sometimes people want to be more specific. In the case of the Dogrib people (their name for themselves being Tłı̨chǫ), they can call them Łį́į́ʼ bítsą́ą́ʼ dineʼé (which means the same thing). —Stephen ^(Talk) 14:31, 19 January 2017 (UTC)Reply

Hi Julien! Thank you for your advice! I've seen your improvements to the page dedicated to the root -DZIL. Which dictionary (or grammar) are you using to provide the verbal bases listed in the table? I used YM1972 (The Navaho Language - The Elements of Navaho Grammar with a Dictionary in two parts containing Basic Vocabularies of Navaho and English), but as far as I can see, it doesn't give so much information about the various stem sets. I also have YM2000 (The Navajo Verb System: An Overview) and I'm currently reading Faltz's The Navajo Verb: A Grammar for Students and Scholars. What book would you recommend me for an exhaustive presentation of the stem sets alternations, modes, aspects etc.? For example, I've seen that you added two rows to the table of the stem set of -DZIL: one for the durative aspect and one for the continuative aspect. I didn't know that this root had these alternations too. Where can I find a detailed source about stem sets analysis? Again, thank you in advance! Sorjam (talk) 19:17, 30 April 2017 (UTC)Reply

For the stem sets, I'm relying on YM1992, An Analytical Lexicon of Navajo. I would have bet you already possessed it, since the listing of verb you provided for -DZIL very much looked like the one found in YM1992. But according to what you're saying, you're relying on the root / stem / theme index appendix of Young's first dictionaries (1972,1980,1987). This appendix is terribly messy and riddled in typos. Literally. Look at the change you did on the conclusive iterative set of -CHĄ́Ą́ʼ. If you look in the appendix, you indeed see : łchin, łchį́į́h, łchą́ą́ʼ, łchį́į́ł, łchą́ą́ʼ (yishchin), but if you look at this verb in the dictionary part, youʼll find: yishchin, náshchįįh, shéłchą́ą́ʼ, deeshchį́į́ł, wóshchą́ą́ʼ, at least in my 1987 copy. The 1992 version of the dictionary, organized by roots rather than by imperfective first person, is a huge improvement over the previous versions. Still, YM1992 still has some major pitfalls and insoncistencies that my work here on Wiktionary is to try to address (mainly, themes and theme categories, and aspect identification). —Julien D. (talk) 19:40, 30 April 2017 (UTC)Reply

These entries are currently placed in Category:head tracking/unrecognized pos, because they all use "stem" as the part-of-speech category. This part of speech is not currently recognised by Module:headword. What do you think should be done with them? Either "stem" could be added as a possible part of speech, or it could be changed to something else. Would "root" perhaps be more appropriate? —CodeCat 15:18, 7 August 2017 (UTC)Reply

Roots and stems are two distinct notions in Navajo (and in all other Athabaskan languages). A root is an abstraction, while stems are the actual shape that a root takes for a given Aspect and a given Mode. The resulting matrix is called the stem set. For instance, -KID, pertaining to "sliding", shows the following stem set:

An actual verb will be the concatenation of one or several prefixes (that can be postpositions ;) ) and one stem, generating dozens of verbs around the idea of sliding, for various modes, aspects and subaspects. Navajo being a "morpheme-heavy" language, this prefix vs stem vs root distinction has wonderful descriptive power (as opposed to postulating say a stem parl- in French when only verb really uses it), therefore I believe stem should be added to the POS list. —Julien D. (talk) 04:51, 9 August 2017 (UTC)Reply

Ok, I've added it as a POS. —CodeCat 15:49, 9 August 2017 (UTC)Reply

Julien, I made these possessive templates, but I am not good at designing or writing templates. The ones I made are clumsy, and sometimes they even give incorrect forms for the 4th person (3a) (ha-, hw-). Their names are probably not very logical. If you ever have extra time, it would be great if you could do something with them so that they are better.

If you are too busy, don't worry about them. In most cases, they seem to be okay. —Stephen ^(Talk) 18:38, 16 October 2017 (UTC)Reply

{{nv-pos-hightone}} shí-

{{nv-pos-hightone-si}} sí-

{{nv-pos-lowtone}} shi-

{{nv-pos-lowtone-secondary}} sheʼa-

{{nv-pos-lowtone-si}} si-

{{nv-pos-vowel}} sh-

I can have a look into it... I think only nv-pos-vowel causes some issue? Could you please point to some examples of wrong predictions? Thx. —Julien D. (talk) 00:39, 19 October 2017 (UTC)Reply

One is ádí. There are a few, not many, where I decided to list the possessed forms individually without a template, because I could not make a good template that covers all cases. However, I can't remember the entries where I did that. I think that any noun which begins with the vowel «a» or «á» (when a/á is not an indefinite possessive prefix) is a problem. {{nv-pos-vowel}} will prefix them with hw-, but it should be h-, as far as I know. I occasionally encounter words that require another new template, and every new template makes the Inflection section harder to do. Also, I have not addressed the problem of coronal harmony (si- instead of shi-) in vowel-initial words, nor the problem of secondary possession (sheʼ-) in hightone or vowel-initial words. The number of different tamplates needed was growing out of control. —Stephen ^(Talk) 10:25, 20 October 2017 (UTC)Reply

I see... Would you have an example word for secondary possession in high tone and secondary possession of vowel-initial words? Isn't atsįʼ addressing the latter case? I assume these cases would be exceedingly rare.

I remember also one day you reverted a commit on íígháán (btw ííshgháán has the same issue), saying that the template's output wasn't correct (*hwíígháán vs hanígháán). I don't remember having ever come across this irregularity in my research, and the Bible, for what it's worth, has 2 occurrences of the former, and none of the latter, so this got me curious. —Julien D. (talk) 01:28, 21 October 2017 (UTC)Reply

Julien, I would like to run this by you. I can't find the Navajo forms for teacher anywhere online or in any books, but I have attempted to make them into a series of tables. Can you look these over to see if anything seems wrong? Thanks. (Oh, and please ignore the Navajo technical terms such as 1st person, imperfective, perfective, future, iterative, optative, etc. I was just experimenting with these technical terms. I'm not suggesting that we use them in our verb tables.) —Stephen ^(Talk) 15:54, 10 December 2017 (UTC)Reply

báʼóltaʼí = teacher

báʼííníshtaʼí = my teacher

báʼííníłtaʼí = your (singular) teacher

yáʼółtaʼí = his/her/their teacher

báʼííníiltaʼí = our (dual) teacher

báʼíínółtaʼí = your (dual) teacher

 

Funny that you brought up this very topic, I'm currently experiencing huge difficulty in theorizing such "conjugations" of nouns for a personal project I'm currently conducting on Navajo. As you know, what is a "word", or at that, a "noun" or a "verb" is a very very blurry notion in Navajo. When does it become a set phrase, when is it just sum of parts. It is very frustrating.

 

The very example Iʼm fighting against as I speak: assembly / gathering. There are: áłah náʼádleeh, áłah ńdaʼaleeh, áłah náʼádleehí ( or náʼádleehé), áłah náʼádleehii, áłah náʼádleehígíí, áłah náhdleeh,... and all combinations thereof. And then it expands further: a religious assembly or celebration can be: hodílzingo áłah náʼádleeh (one becomes repeatedly together while holiness is observed), dahodíínohsingo áłah náhdleeh (you (pl.) become repeatedly together while you observe holiness / keep it holy),... the list is virtually inifinite, but behind that is the same mental image: a religious celebration. But they donʼt always fall into neat tables as for báʼóltaʼí. In the mental lexicon, this "word" is more a recipe in the lines of: a form of "hodíínísin" (I make it holy, I observe holiness) + "áłah" (together) + a form of "yishłeeh" (I become) / "aleeh" (one becomes), preferably in the iterative ("ná") + a relativizing particule (-í, -ii, -ígíí or even Ø). That would be the archetypical form (lemma form?) of that "noun", in a form of dictionary a lot more adapted to the language.

 

In a way, the above is just reproducing at a syntactic / lexical level what already happens at the morphological level of the verb: the verb "to teach someone" for instance is not nanishtin, or neinitin, but really the recipe: na#O+ni+(Ø/si):Ø=tin (conversive) (as Kari does it in his Ahtna dictionary), allowing formes like: nináánihinétą́ą́ʼ (I taught you again), etc...

 

To your question: the table seems fine to me. Iʼm just not sure of the practical value of the Optative or Iterative forms, if thatʼs not a little too overreaching. Iʼm wondering if other combinations might not also be considered, like "ná ííníshtaʼí" (Teacher, when adressing him, lit. "you for whom I read"), etc, etc! Same as above, this should probably best be handled as a recipe.

 

I googled for báʼííníshtaʼí just now, and geez, there are only 2 hits: the wiktionary article under báʼóltaʼí, and your Quora answer! Je perds mes mots... —Julien D. (talk) 03:23, 12 December 2017 (UTC)Reply

Thanks. Yes, it's a big problem. My immediate concern is that the number of fluent speakers is rapidly falling. In 40 years or less, Navajo will probably be virtually extinct (it seems that almost all speakers are currently at least 40 years old). In the meantime, there are many Navajos who know a little of the language, and they try to learn more and try to use it, but they make serious mistakes. For example, I keep seeing or hearing people using "shibáʼóltaʼí" to mean "my teacher." As much as I would like to do something to remedy this situation, I feel helpless. I don't think I can do anything to slow down the inevitable. —Stephen ^(Talk) 02:57, 13 December 2017 (UTC)Reply

I asked a fluent friend of mine about using "ná ííníshtaʼí" as a vocative form for teacher. This is what he told me: "My instinct would be to say, "Báʼóltaʼí" because that is less direct and maintains a professional distance. It also recognizes their position (like "Dr."). Báʼííníshtaʼí is a little too direct. You would use that form to tell someone, "He/she is my teacher (Éí báʼííníshtaʼ / há ííníshtaʼ)." —Stephen ^(Talk) 04:02, 13 December 2017 (UTC)Reply

Regarding the shibáʼóltaʼí issue, thatʼs one thing I never really understood. There are numerous example in the language of such forms, when simply conjugating the nominalized verb would have sounded to me so much more conform to the genius of the language. For example: bee adziilii (power, lit. that by means of which one is strong) - shibee adziilii (my power), instead of bee sidziilii (that by means of which I am strong), bee hazʼáanii (law) - shibee hazʼáanii (my law), instead of bee haséʼáanii (the Bible shows both: bibee hazʼáanii and yee hasʼáanii). And same for other tool nouns in bee + (3i-verb form) + -í/-ii. As long as we accept those, we should accept shibáʼóltaʼí...

It is also said that speakers of a language by definition never do mistakes, they just speak the next form the language... It is probably not worth fighting that specific issue, when fluency as a whole is the problem... Also, sadly, no amount of wiktionary will address the issue if people don't speak the at home :/ —Julien D. (talk) 13:17, 14 December 2017 (UTC)Reply

Julien, can you check the conjugation of haiłgéésh? An anon added the original conjugation, and I think he made some mistakes. I made corrections, but I would appreciate your opinion and/or corrections. —Stephen ^(Talk) 12:37, 6 February 2018 (UTC)Reply

That is something I had already noticed while cleaning up the conjugations and writing the auto-conj module: all ha- verbs whose perfective was filled out were done following a full da-shift, when YM mentions that the shift should only be partial (3rd person only), and that only a- (away out of sight) verbs should follow a full shift.

Initially I wanted to update them all to turn 1st and 2nd person back to yi-perfective, but I can't remember where I found a paper that legitimated full shift for ha- verbs. Any case, there seem to be a great deal of inter-personal variation.

Also, for what it's worth, the Bible translation seems to favor the full shifted paradigm, but there are only a few of such occurrences.

Alice Neundorf shows both (I don't own a copy but could look up on Google books): while haigééd fully shifts, haiłgéésh only partially shifts.

Kari in his Verb Prefix Phonology (1976), citing Young, indicates a much higher tolerance for non-shifted forms nowadays compared to "30 years ago, and even 60 years" when comparing with the Franciscan Fathers' description. Yet another 40 years later, it would interesting to see where this all stands today. One thing that Kari does mention is that ha- would favor a shifted paradigm to avoid homophony with hada- verbs, with totally opposite meaning (hada- "downwards")

So in light of this, I had decided to keep the fully shifted paradigm. —Julien D. (talk) 05:01, 7 February 2018 (UTC)Reply

Hello Julien --

I've watched as you've built up the infrastructure here for NV entries. Thank you for your efforts.

One thing I'd like to suggest is linking from the NV templates to the relevant terms for certain structural or linguistic labels. For instance, at a full entry like łibá, there are some descriptive terms that are left unexplained for the reader: things like neuter and imperfective. Is there anywhere that explains these terms in the context of Navajo lexicography? Could these terms in the tables be linked through to the appropriate entries, or to a section on a page like WT:ANV that would explain these?

At a root entry like -BAʼ, there are the five aspects along the top (imperfective, perfective, future, iterative, optative), which can sometimes appear as potentially cryptic abbreviations (IMP in other contexts could be construed as imperative, for instance), and the other aspect types along the left (continuative, momentaneous, semelfactive, etc.). Could these also be linked to definitions appropriate to Navajo?

I hadn't yet seen the Appendix:Navajo roots and stems derivation page until just today. That is an excellent presentation of verb-root phonological shifts as a system. I am very happy to have this information so clearly laid out; thank you very much for pulling this together.

Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:48, 27 March 2018 (UTC)Reply

Hey, I want to add verbs in Ojibwe, Unami and some other algonquian languages. Do you think you could help me a bit? I know the verb forms, but I’m not good enough at wikipedia to figure out how to add verb paradigms with prefixes, suffixes, modifying particles, pre-verbs etc... Is there any way you could either personally teach me how to make paradigms with sketch-up or connect me with someone or some link to teach me? Hk5183 (talk) 14:34, 14 July 2018 (UTC)Reply