Category talk:Hungarian verbs by argument

Groups of arguments by from-to direction or at position

@Panda10: I noticed that the arguments are specified for certain verbs (e.g. lerohan but I'm 100% sure I've seen more examples) that don't actually take specific arguments, but any (whichever) separative/ablative (-bÓl, -rÓl, -tÓl), lative (-bA, -rA, -hOz, maybe also -nAk, -ig), or locative arguments (-bAn, -On, -nÁl, perhaps also -vAl). I think these verbs should not be categorized in such a specific way as "Hungarian verbs taking -ba/-be", because it's misleading. Instead, these verbs should have their own overarching hu-case templates for the separative/ablative, lative, or locative argument, and their categories could include the more specific categories. I think it would be too long to include the individual endings for these groups but they should be explained or defined somewhere so that readers should know what suffixes each group includes in terms of Hungarian. Where do you think these explanations or definitions could be placed and how could we phrase the link to these argument groups? For the first group, Hungarian terminology uses "ablative" pretty consistently, but the entry of lative mentions "separative" as its opposite. Maybe "separative" is better for our purposes, since "ablative" is already reserved for Hungarian -tÓl. What do you think? Adam78 (talk) 15:38, 25 February 2023 (UTC)

@Adam78 I have some initial thoughts and doubts:

1. While linguistically there may be a difference, how will a learner benefit from this separation? Will this change make things clearer or muddier? After all, whether it's a specific argument or a general one, users will have to learn them anyway.
2. Instead of creating a new hu-case, maybe adding a new parameter could direct the entry to a new category.
3. A new category name could be "Hungarian terms used with lative suffixes". The suffixes that belong here could be listed and explained at the top of the category.
4. Let's try to imagine the category tree. The first row is the current tree which will stay. The second row would be the new addition:

Hungarian terms by lexical property --> Hungarian terms taking arguments --> Hungarian terms taking -ba/-be --> Hungarian verbs taking -ba/-be‎

Hungarian terms by lexical property --> Hungarian terms used with lative suffixes --> Hungarian terms using -ba/-be --> Hungarian verbs using -ba/-be

I think my problem is that I don't see how this structure/change will make anything easier, clearer, more self-explanatory. Panda10 (talk) 22:22, 25 February 2023 (UTC)

@Panda10:

1. The situation is muddy at the moment, by having verbs that strictly take -ba, together with verbs that might take -ba just as much as they can take -ra or -hoz. Look at the list of verbs in these categories: they're growing larger (which is a good thing in itself), and I don't think it's good to expand them with misleading items. Instead, these items should be placed where they actually belong, and these categories could become more informative and easier to overview. (Maybe I should have looked up the other examples in advance so that you can realize for yourself how different they are from the rest and how helpful this distinction will be.)
2. Yes (although I can't imagine it completely), but we also need to decide what should appear as the link description.
3. Yes, it sounds good.
4. I think the lative etc. layer would be inserted in this categorization basically as an additional element, without interfering with the primary structure that we have now. It would only add an extra element for verbs with less specified arguments. Of course it could include the latter categories as well, so e.g. the category for -ba/-be would be available directly under "…verbs taking arguments" as well as "…verbs used with lative suffixes".

Adam78 (talk) 22:42, 25 February 2023 (UTC)

@Panda10 Another example is felmegy, where currently various possible "arguments" are listed, even though the situation is that the verb simply takes a lative argument. So the placement of this verb in the categories "verbs taking -ba/-be", "verbs taking "-ra/-re", and "verbs taking -hoz/-hez/-höz" practically does more harm than good, and it also makes the grammar information in the entry look more confusing. This problem could be eliminated with categories for lative etc. arguments. Adam78 (talk) 22:49, 25 February 2023 (UTC)

@Panda10: Or take lakik: you can say "ebben a városban lakom", "ezen a fenyőfán lakom", or "ennél a sorompónál lakom" – any locative argument will equally do. There is no need to specify which one needs to be used (because the verb itself doesn't specify it in the mental lexicon), and suggesting otherwise is misleading for language learners.

@Adam78 Okay. Do you want to go ahead and make a test change for the above verbs just to see how they will look? You can use manual categories for now instead of hu-case. It might be a good idea to create the category as well, to demonstrate the tree structure. Panda10 (talk) 22:56, 25 February 2023 (UTC)

@Panda10: Okay, I constructed another table that illustrates these relationships and added a category and a link to lakik. I'll continue later. Adam78 (talk) 15:31, 26 February 2023 (UTC)

@Adam78 The table looks good. I made some minor modifications in lakik. I also created Category:Hungarian terms with locative suffixes. Are you still planning to have subcategories or locative in the category name explains the suffixes? Panda10 (talk) 20:12, 26 February 2023 (UTC)

@Panda10: I'm not convinced if subcategories are necessary. – I've added the link and the category to some more entries. (By the way, for entries not after E/É, Nagyszótár can provide good examples, often with more variation and with more genuine instances for the suffixes than what we can devise by ourselves.) I wonder if these few are enough for you to see how it works. If so, we could use a template instead, and also a more specific category name that has "verbs" instead of "terms". (I'm not sure if the separative, locative, and lative are applicable to adjectives or other non-verb parts of speech.) Adam78 (talk) 21:09, 26 February 2023 (UTC)

@Adam78 Yes, these examples are sufficient to see the structure. About the template. I wonder if adding three new parameters (separative, locative, lative) would work. To be on the safe side, the existing pos parameter could be re-used to indicate other POS-specific categories (if any). For verbs, we can assume a default. Panda10 (talk) 21:21, 26 February 2023 (UTC)

@Panda10: I especially recommend Mazsola to check if a verb can have other (lative, locative etc.) arguments aside from the one(s) already supplied. You need to register only once and then you can log in and use it afterwards for free. There are examples illustrating how to use it, but it's pretty straightforward. (Currently it gives an error message after registration, but in fact the registration does take place successfully; you'll get an email in a few seconds. I noticed it after a few attempts when I was about to give it up. I suggest you don't use a sensitive password there because it's sent back as plain text.) With the same registration, you can also use the corpus of Nagyszótár with more than one billion running words. Adam78 (talk) 13:13, 28 February 2023 (UTC)

@Adam78 Thank you for the reminder. I was able to log in to both places with my old registration. I have not been using them lately, but they are very useful. And thanks for updating the template! Panda10 (talk) 17:58, 28 February 2023 (UTC)