Maikxlx

Latin parasyllablic declensions

I am curious as to what source you have that suggests there are three different 3rd declension parasyllabic declension patterns. Even the Latin Wiktionary has only one such pattern. --EncycloPetey 04:19, 19 December 2008 (UTC)

Actually, I wouldn't say that there are three different 3rd declension (M/F) parasyllabic (noun) declension patterns. There is only one regular pattern - the one of "auris", which is defined in Template:la-decl-3rd-PAR. However, there are also several dozen parasyllaba (including some place names) which show, either variably or invariably, the pure i-stem sg. acc. -im in place of the regular -em, and/or the pure i-stem abl. -ī in place of the regular -e. As far as my sources: Harkness's A Complete Latin Grammar and D'Ooge's Concise Latin Grammar. The info is also presented in Allen and Greenough's New Latin Grammar in a less perspicuous fashion. Unfortunately, there are some minor discrepancies among these (and other) authorities, but they are in general agreement about the existence the "tussis", "navis", and "ignis" patterns. There are, marginally speaking, other patterns as well.

I will write more about this soon. I will try to enumerate the peculiarities (perhaps on my Wiktionary user page) more clearly in the near future. --—maikxlx (talk) 06:40, 19 December 2008 (UTC)

Umm... those aren't different patterns. What you're seeing is differential coverage from different authors. The ablative variation in the "ignis" table of D'Ooge reflects the fact that the ablative endings changed in the Middle Ages. This is not a different pattern, nor is it found in Classical Latin. Harkness is a Victorian-era work, lacking in recent scholarship. I would advise against creating new inflection templates on the basis of Victorian scholarship. In most academic areas I have studied, such works have since been shown to be error-prone. Occasional but known variants can be described in the text of the inflection section; modifying the inflection table for every documented exception and temporal variation would be overly and needlessly complex. --EncycloPetey 09:10, 19 December 2008 (UTC)

Umm... the simple fact of the matter is that the sg. abl. of "ignis" in classical texts is usually "ignī", and sometimes "igne" (esp. in poetry, for metrical purposes). After Augustine, it was usually "igne". This is not a controversial point, much less some sort of error of the Victorian scholars; our friends at the Latin Wiktionary seem to agree with us. It's appropriate therefore (inconvenience to modern wiki editors notwithstanding) to put the attested _standard_ forms in the declensional table. If we're going to produce declensional tables, why not make them accurate based on what actually appears in the classical text? Otherwise, we might as well just write down the dictionary citation forms and let Wiktionary users work out the actual declension as best they can. --maikxlx (talk) 15:09, 19 December 2008 (UTC)

Yes, after Augustine was the Middle Ages. A shift occurred in the standard ablative then, and that's what I was indicating. If it did occur also on a regular basis for metrical purposes in earlier poetry, then I was not aware of that. I did not blame this variation on a Victorian error; I cautioned you against trusting D'Ooge because (in general) Victorian scholarship is not infrequently suspect. What I'm questioning is not the fact that you've amended the tables for (deprecated template usage) ignis, (deprecated template usage) navis, etc., but that you've created different tables for each of these. Both words have the same inflectional tables on Latin Wiktionary, and it's the same table as for all other parasyllablic third declension nouns. Why then do we need three different tables? Shouldn't we just update the one table? --EncycloPetey 17:01, 19 December 2008 (UTC)

1. Okay, I am glad that we agree that the current "one-size-fits-all" template is not appropriate for all classical 3i substantives like (deprecated template usage) ignis and (deprecated template usage) navis. We also agree that the declensional tables should contain the standard classical forms, and not the medieval variants.

2. I am not just trusting D'Ooge. D'Ooge agrees with Allen and Greenough, and Harkness, and one or two other sources that I have, on all the edits that I have made so far. I am currently trying to obtain Wheelock, too. These guys are well-known scholars of Latin who spent their lives studying the classical corpus. Who should we rely on if not them?

3. Upon a second look, the Latin Wiktionary "one-size-fits-all" 3i-template is inaccurate in the direction opposite of the English Wiktionary: they decline too many forms, including forms (sg. acc. *ignim, *aurim) that probably would have been non-standard or even non-existent during the classical period. I don't think that's a very good improvement over the current English Wikipedia situation.

4. I like your idea of extending the current template to allow optional fields that will handle the peculiarities. I am capable of taking a look at that in the next day or two, and willing if you're amenable to the idea.

Certainly I'm amenable to the idea. The noun declension templates have not had the same attention as the adjective, verb, and participle tables. I know for a fact that some of the noun declension templates do not work the way they should, and that there are some tables we need but do not yet have. Doing something about this problem has been on my to-do list for a while, but I've had other issues (both on- and off-line) that have taken priority. the off-line activities are well...mandatory, and I've committed myself to finishing certain Wiktionary projects before beginning additional ones. So, I can offer input or suggestions as a long-time editor here, but I won't necessarily have the time right now to work actively in revising alongside you.

I am not completely opposed to the idea of noting medieval variants, but don't have any strong desire to burden the tables with them either. For a noun like avis, noting the key ablative difference in Ecclesiastical Latin is important, but in most cases, it doesn't seem so useful to note these differences. I'm more inclined to relegate that information to an Appendix that could be linked from the inflectional tables, so that users who desire the most detailed explanations can find it, but the general or beginning user will not be overwhelmed on first glance. There are appendices already on each of the five major declension patterns, including some information about variation, but they too are in need of revision. --EncycloPetey 18:50, 19 December 2008 (UTC)

Glad to see how much progress you've made in so short a time. One note: I see here that Example 2 and Example 3 take parameters in a different order from each other. That is bad news for most editors, when the parameter order varies from usage to usage. It is preferable to have a single, consistent sequence for parameter entry. The usual order for Latin templates is 1st-form-w/o-macrons, 1st-form-with-macrons, 2nd-form-w/o-macrons, 2nd-form-with-macrons, etc. --EncycloPetey 05:51, 21 December 2008 (UTC)

I have been experimenting this weekend, and essentially designed my own complete declensional template system for research purposes. The templates I wrote are backwards compatible (assuming rearranging of parameter order, which is easy) and allows a good number of things that Wiktionary could use including:

• Arbitrarily change a cell's contents using a named parameter.
• Arbitrarily append to a cell's contents using a named parameter.
• Arbitrarily suppress either the singular or plural column using a named parameter.
• Arbitrarily display the locative case row using a named parameter.
• Suppress all links where the given form is similar to the pagename (without macrons). (This makes the declensions look awesome! Look here.)

You are a keen observer to have noticed that I have been tinkering with the required parameter order. I understand your concerns, but let's have that conversation after I make my case here. --maikxlx (talk) 06:44, 21 December 2008 (UTC)