Gerunds

[Richard W]

How did the term 'gerund' come to embrace such seemingly different things as a Latin verbal nouns that denote objects in the accusative and mark arguments the same way as the verb, the Russian adverbial participles and the Sanskrit absolutives?

Is there much currency to the notion of a Dutch or Afrikaans gerund, and if so, what are they?

[bradrn]

How did the term 'gerund' come to embrace such seemingly different things as a Latin verbal nouns that denote objects in the accusative and mark arguments the same way as the verb, the Russian adverbial participles and the Sanskrit absolutives?

Because linguists can’t stop themselves from using terms inconsistently if they try. Same with terms like ‘infinitive’, ‘participle’, ‘serial verb clause’, etc. etc. etc. (I swear, 90% of the literature on serial verb clauses is devoted to trying to figure out a definition which would encompass everything that term is used for. More recent works give up entirely, and call them ‘multi-verb constructions’.)

If I had to specify one ‘central’ meaning of the term, I’d go with something like ‘any verbal form that can function like a noun’. This explains some of the diversity, given that the same verbal form can also have other functions (e.g. English -ing can be a converb, participle, infinitive or gerund, according to Nedjalkov in Converbs in Cross-Linguistic Perspective). But I’m too unfamiliar with Latin, Russian and Sanskrit to say anything more about those specific cases.

[Moose-tache]

The clue is in the name. "Infinitive" does not mean dictionary form; it means any form that is not conjugated. For some European languages, that mostly leaves the infinitive, and some participles which... don't worry about it. So if we extend "infinitive" to mean things like "the infinitive of sumimasu is sumu," we've no one to blame but ourselves if the result is confusing. Languages that don't conjugate their verbs should probably have their own terminology, which is what most grammar have been trying to do for a century.

Likewise, "gerund" basically means "do-y sort of doing-ness." Semantically it's a wet paper bag. So of course it's not going to be used consistently; how would it? Most academic grammars therefore use terms like "verbal noun."

[zompist]

I would add that grammarians name things mostly by analogy. The Latin grammarians called amāre an infinitive, for the good reason Moose suggests. That nicely fits French aimer, Spanish amar, Italian amare. By the time you get to Portuguese, what else are you gonna call amar? Only, oops, the Portuguese "infinitive" is conjugated. (Or can be. It's way, way in the back pages of your Portuguese grammar.)

In English we can say "to love" is the infinitive, on the grounds that you use it where French would have aimer, except not always. (French infinitives take prepositions too sometimes.) Or just "love" is the infinitive (what else is it in "I can love"?), though it's a bit embarrassing that it's also the verb root.

And the thing is, it's fine. For studying and learning a language, you should terms that are simple and useful, and don't confuse you when you switch to a different textbook or dictionary. Comparative grammar, or theoretical linguistics, or conlanging, is not of interest to 99% of the users of a grammar.

FWIW, you get exactly the same problem in biological anatomy, which is also mostly descriptive. E.g. what do you call the female hyena's sex organ? Comparatively, it's the clitoris, but it looks exactly like a penis, and it functions like a vagina. And that's nothing compared to the nightmares of insect anatomy.

[Moose-tache]

It's always bothered me that chemists and biologists use "exothermic" and "endothermic" in exactly opposite ways from each other.

[Raphael]

It's always bothered me that chemists and biologists use "exothermic" and "endothermic" in exactly opposite ways from each other.

How do biochemists use it?

[chris_notts]

I would add that grammarians name things mostly by analogy.

Given the fuzzy nature of meaning and the lack of a small finite set of perfectly replicated cross-linguistic categories, I'm not sure what the alternative is, except to invent a unique terminology for every language. The way we do it is probably the least bad option, although you can debate whether any given labelling choice is the least confusing one possible.

[WeepingElf]

I would add that grammarians name things mostly by analogy.

Given the fuzzy nature of meaning and the lack of a small finite set of perfectly replicated cross-linguistic categories, I'm not sure what the alternative is, except to invent a unique terminology for every language. The way we do it is probably the least bad option, although you can debate whether any given labelling choice is the least confusing one possible.

Yep. (Where is the Like button when you need it? )

[Moose-tache]

It's always bothered me that chemists and biologists use "exothermic" and "endothermic" in exactly opposite ways from each other.

How do biochemists use it?

Hilariously, the first result for "biochemistry endothermy" uses the term the biology way, while the first result for "biochemistry exothermy" uses it the chemistry way.

[Raphael]

Hilariously, the first result for "biochemistry endothermy" uses the term the biology way, while the first result for "biochemistry exothermy" uses it the chemistry way.

Thank you!

[vegfarandi]

I would add that grammarians name things mostly by analogy.

Given the fuzzy nature of meaning and the lack of a small finite set of perfectly replicated cross-linguistic categories, I'm not sure what the alternative is, except to invent a unique terminology for every language. The way we do it is probably the least bad option, although you can debate whether any given labelling choice is the least confusing one possible.

Yep. (Where is the Like button when you need it? )

Linguist Martin Haspelmath has been on a mission to retro-define as many commonly used terms for use in cross-linguistic comparison as possible, and he recommends using a capital initial when referring to a language-specific category in a cross-linguistic context. So he has a definition root (a root is a contentful morph (i.e. a morph denoting an action, an object or a property) that can occur as part of a free form without another contentful morph.) and if a particular language's analogous concept differs from that definition, you can refer to it as Root in x-language.