Template talk:verbal noun of

RFD discussion: April 2019–September 2021

The following information passed a request for deletion (permalink).

This discussion is no longer live and is left here as an archive. Please do not modify this conversation, but feel free to discuss its conclusions.

---

Template:verbal noun of

Again, like the above. "Verbal noun" is not a well-defined concept anyway, and we are not in the habit of defining nouns derived from verbs like in shortening. Instead, a full definition and etymology is provided since they are considered lemmas of their own. —Rua (mew) 21:31, 1 April 2019 (UTC)

Not all languages work like English. Several languages have forms conventionally called "verbal nouns". Arabic is an example; every verb has one or more associated verbal nouns, which function somewhat as infinitives (Arabic has no infinitives as such). These verbal nouns are listed in dictionaries next to the verb, because you need them in many syntactic constructions. I'm really not sure what you would replace them with if you removed this template. Benwing2 (talk) 07:05, 2 April 2019 (UTC)

Arabic rarely uses this template (see the census below), so it must use {{ar-verbal noun of}} almost exclusively. {{ar-verbal noun of}} has the |form= parameter, which makes it harder to convert to {{verbal noun of}}. — Eru·tuon 07:25, 2 April 2019 (UTC)

I take it that, for English the idea of having "verbal noun" as a definition of an -ing-form of an English verb is that it thereby includes all current and future definitions of the verb lemma. As we put "present participle" alone as a definition under the Verb PoS, there is no explicit acknowledgment that -ing-forms can be used as nouns (eg, 'these belated shortenings of […] '). This notion applies to all English -ing-forms of verbs. In fact, the two English uses of the template in question also put shortening and blottoing in Category:English verbal nouns with some 90 members.

IMO, for English, the underlying issue would need to be addressed before we could take any action that would limit uses of this template and the category in English. If verbal noun is a term that is well defined as it applies in any language, as the census table below suggests that it does, then the template should be kept. Whether we should have the template throw an error message for languages for which we assert that the 'verbal noun' concept does not apply is a separate matter, linked to whether the corresponding categories should be deleted.

If this complicates our template/module practice, so much the worse for our uniformitarian template/module/category architecture. At least the module data tables allow for the complexity, though I wonder how much of the module's memory is consumed by invoking the tables. DCDuring (talk) 12:20, 2 April 2019 (UTC)

If verbal nouns are a well-defined inflection of verbs in some languages, then it might be ok to keep it. However, I have noticed that many entries, even in Irish and Arabic, give additional idiomatic definitions alongside the verbal noun definition. They are also categorised as noun lemmas in those languages, not as verb forms. That suggests to me that they are still derivational rather than inflectional. The fact that Irish has more than 10 different ways to make a verbal noun again suggests that this is word derivation, not inflection. —Rua (mew) 21:05, 2 April 2019 (UTC)

For all stems except the base stem, with rare exceptions that are still regular (I only know تَجْفَاف (tajfāf) from جَفَّفَ (jaffafa) and تَكْرَار (takrār) from كَرَّرَ (karrara), normally form II has the verbal noun taKLīM but a few geminate verbs deviate), all verbal nouns of the stems II and higher are utterly predictable. For base stems they are usually too depending on the meaning and the past vowel: transitive verbs with the pattern KaLaMa are likely to have the verbal noun of the pattern KaLM, even if other forms also exist (for example كِتَابَة (kitāba) is normally the verbal noun of كَتَبَ (kataba) but كَتْب (katb) is understood too), verbs of the pattern KaLiMa that have a meaning that would be or is the passive of a transitive verb with the measure KaLaMa are likely to have the verbal noun form KaLaM. But not even plurals are wholly predictable. This shows that the predictability or the different ways to form certain needed forms does not say anything about their categorization.

It does not need to be a “well-defined concept”. It is a lexicographical device to refer to “all current and future definitions of the verb lemma”, since that’s the point, that these forms stay linked to the verb meanings, may they have additional idiomatic meanings or not. The link is there in all the languages I know: Arabic with you know what, the forms that are given in the conjugation tables, German with -en or -ung (the former related to a more concrete or singular act and the latter to procedures), English with -ing, Latin with -tiō, its descendants also or by a clipping with a minor added vowel (Spanish rechazo for Spanish rechazar, Spanish toque for Spanish tocar), Slavic especially *-nьje and some specialized minor ones.

Grammatically Arabic verbal nouns accept accusative objects and genitivus objectivus, the analogous in English, while Slavic, Latin, German only take genitivus objectivus, by I don’t think this distinction is relevant here.

As a side note, the etymology is not relevant either for the categorization: The verbal noun may be formed from the verb or the verb be denominal.

I note a false dichotomy that being a “lemma” entails having own definitions, or in a compulsion to shed “derivation” and “inflection”. But it must not be like that we either attach forms like an appendage to “true, full” entries or cut any knot: this sacrifices reality for schematism, with little gain – not even for the dogmatics, for who told you that derivation and inflection aren’t ambiguous in languages? You see, it’s even in English ambiguous and at the same time something that one would not care about for anything.

Now I see why I {{perfective form of}}/{{imperfective form of}} has been deprecated, it is for the same reason while not being “not well-defined”, which I would have had difficulties to agree with. It would have been better to recognize that one item of such a pair is not “form of” but it still would be wise to refer from one to the other in the meaning. These forms live and die together, while they can as well live separately or only represent a part of the other (like пере́ть (perétʹ) which has different perfective forms according to sense). Fay Freak (talk) 22:20, 2 April 2019 (UTC)

@Fay Freak It isn't a false dichotomy if you understand what the categories of "lemma" and "non-lemma" mean in the first place. A lemma is the kind of form you'd expect to find in the average dictionary, along with a definition. A non-lemma is a form that isn't given its own entry, but is instead grouped with the lemma it belongs to. Another way to see the split is between independent and dependent. An inflection can't exist without the verbal paradigm it belongs to, whereas a derivation can exist independent of other lemmas. If a verbal noun has a sense that it does not share with its verb, then that sense is its own noun, because it's independent from the verb. But the verbal noun itself is still dependent, attached to the verb, so the verb form and the noun are distinct. Moreover, a key difference between inflection and derivation is whether the category is optional or required.

@Rua This is exactly where the average dictionary is ambiguous, the nominalizations being mentioned in varying fashion, variously headword-like.

Be it so that a sense makes an own noun, it would mean that only because of this we would have two headers, two times “Noun“, from which the reader does not benefit. We might use a headword template that categorizes both as lemma and non-lemma. But this has nothing do with whether the definitions should have “verbal noun of”. The definitions and their organization shall not be based on how one will categorize them. So no, T:verbal noun of can’t be deleted. I disown the notion that because “a sense” “is a lemma” or “a non-lemma” we can’t refer to other senses to define it – {{synonym of}} is also handy, just by the way. You are committing a great paralogism here exemplified by your utterance “sense is its own noun”: Nay obviously, a sense is not a noun, this confuses subjects and predicates fundamentally. There are nouns and we see senses in them (hence their “senses”, because humans sense them), and the senses of one form might be either independent of another term, or wholly dependent on it as a mere inflexion, or loosely dependent on it. The lexicograph can tell people how usage is without deciding whether something is “derivation” or “inflexion“, or “enclitic” or whatever, however nice there is a distinction in principle. I don’t know or don’t think what a form like عَلَيْهِ (ʕalayhi, “to him, to it”) with so-called “pronominal suffix” is dogmatically but I know without this that nouns with these personal suffixes should not be included. For other things one knows that they should be included and how to gloss them but wits not how they should be classified or related to other terms: but one knows what is economical, knows a good effort-information ratio. The question for including or writing something is always: Will it provide essential information rather than just be expensive? Be it that the distinction is imprecise in this matter, so is language, and still if this is bad it does not follow that we can’t gloss this way. Semantics trumps grammar. The distinction has also nothing to do with predictability and lexicalization: it can vary by language and in a language what one has to memorize or if there are special forms that one has to know, with varying degrees of predictability. Fay Freak (talk) 00:03, 25 April 2019 (UTC)

@Mahagaja Would a verb without a verbal noun in the Goidelic languages be considered defective, or is that a normal state of affairs? @Benwing2 The same question but for Arabic. That's not a rhetorical question, it's something I'd actually like to know because it's important in deciding whether to call these lemmas or non-lemmas. Based on the model of English gerunds, the standard treatment of such cases is to use the Verb heading for inflections, which are part of the verbal paradigm and whose sense is tied to the verb, and the Noun heading alongside that Verb heading for cases where a verb form has developed special senses that do not arise from the verb. If verbal nouns in these languages are inflectional, then this is the precedent we should follow for them. —Rua (mew) 22:10, 24 April 2019 (UTC)

@Rua: Of the verbs in CAT:Irish defective verbs, only ar(sa) and feadair don't have verbal nouns; the others do. (The fact that Irish is doesn't have a verbal noun is one of the many reasons why syntacticians, unlike language teachers, consider it a particle and not a verb, not even a defective one.) So even defective verbs tend to have verbal nouns; and nondefective ones always do. On the other hand, there are some verbs, like caint and magadh, that exist only as verbal nouns and have no finite forms. I do agree that from a purely morphological point of view, Goidelic verbal nouns are probably derivational rather than inflectional, but for speakers (both L1 and L2) the verbal noun is still an essential part of the paradigm, so you might say that they are "semantically inflectional" while being morphologically derivational. If you want to delete {{verbal noun of}} to reduce the overall number of form-of templates, then that's okay, as long as I'm still allowed to use {{inflection of|...|verbal noun}}. (Incidentally, the verbal noun is just as essential in Brythonic languages as in Goidelic ones; the reason there are no Brythonic languages in the table below is that for Brythonic languages we use the verbal noun as the lemma.) Lower Sorbian is in the list too; in that language verbal nouns are definitely inflectional and clearly part of the paradigm. As for the other languages I know in the table below, we can IMO eliminate "verbal noun" as a category for German, English, and French. —Mahāgaja · talk 07:46, 25 April 2019 (UTC)

We can keep the template if they are indeed a verb inflection for Goidelic. I'm not sure about the need for categorising them, though, but that is part of my suspicion of categorising non-lemmas in general. I don't find Category:Irish verbal nouns particularly useful, anymore than something like Category:Irish first-person singular present indicative forms would. More pressing, though, is that verbal nouns aren't in Category:Irish verb forms, when they clearly should be. I'm also not sure how to deal with an entry like atriail. Given a strict separation of lemmas and nonlemmas, there would not be a "verbal noun of" sense here at all, because the verbal noun would be treated as a verb form and thus gets a Verb header and not a Noun header. On the other hand, verbal nouns have noun inflections and even genders, and it would be somewhat redundant to have two sections with the same inflections, one for the verbal noun and one for the remaining senses. But it still does not sit right with me that we treat verbal nouns as lemmas, when they are also a part of the verbal paradigm. Looking at English participles gerunds as an example, we don't have nouns defined with {{gerund of}} either, nor do we have adjectives defined with {{past participle of}}. Instead, whenever there are senses that are separate from the verb, they receive their own header, while those belonging to the verb get Verb. But that doesn't work as neatly with languages where verbal inflections can have genders and inflections, not just Irish but also Dutch and German. I'm not sure what to do here, and the discussion in WT:BP about gerunds stalled with no useful conclusion. I would like situations like this to have a standard treatment that fits many languages. —Rua (mew) 10:30, 25 April 2019 (UTC)

I think verbal nouns are similar to participles in being in sort of simultaneously lemmas and nonlemmas, and we do have categories like CAT:Dutch past participles and CAT:Dutch present participles, as well as CAT:Dutch participle forms, and I think we should keep them. So I don't think we can commit to "a strict separation of lemmas and nonlemmas" anyway. —Mahāgaja · talk 12:59, 25 April 2019 (UTC)

Census of language codes
in {{verbal noun of}}

language code	count
ar	3
bg	5
de	4
dsb	18
en	2
fo	1
fr	1
ga	660
gd	506
gv	442
ha	2
he	3
jaa	1
ka	17
mga	4
oge	3
pl	1203
sga	63
sh	10
te	99
tr	9
xmf	1

• Kept due to lack of delete votes for two years. Vox Sciurorum (talk) 15:28, 20 September 2021 (UTC)