My powers of inference probably not being what they were, it's not clear to me if you are saying that singing is a noun or a verb in the sentence: Sarah made singing a priority. I'm also interested in your statement Depending on how we expand the word "singing" we may think of it as either a noun or a verb (but certainly not as a sort of a mongrel in between these two). Are you referring solely to the sentence in question or does that apply to all -ing forms?

We don't have any specific syntactic or semantic reasons whatsoever to think about "singing" in the OP in any other way than as a noun. When we think about verbs we think predominantly of the predicator in the sentence structure - it is what verbs normally do. When we think of nouns , the first thing that comes to mind is object or subject. If we try any other verb form in place of "singing" that will obviously not work. For this reason we don't have any more trouble distinguishing the noun "singing" from the verb "singing" than we have in understanding "work" as a noun and not a verb

..in the same sentence: She made work her priority. How do we know that "work" is not a verb when it has identical form as the corresponding verb? (Oddly enough, nobody asks this question, while the -ing form has become notorious for its dual analysis.) We know it mainly because of its functional position within a very familiar syntactic pattern, and we know it intuitively. We set apart word classes according to their peculiar characteristics, and the lack of characteristics typical of other classes.

Thanks as always for the quick response. So as I understand it the presence of opera turns the noun into a verb. But it seems to me that all singing implies some object (something has to be sung), so why would the implied object not render singing a verb?

And I am by no means convinced in the non-existence of mongrels if by that you mean that all words can be definitively placed into one or other of the word classes.

I don't think that the word "singing" in itself associates us necessarily to the object of singing. I'd rather think that in isolation, we think about it as an abstract concept, not directly related to any particular object. As I said above, -ing nouns are no different than any other noun derived from a verb. Following that criteria we could say that "destruction" is intuitively associated with the thing destructed etc.

Ok, thanks again. They want to kick us into chat. So let me just note that the very frequent and largely inconclusive discussions here on the lines of Is word X a (Word class) Y? confirm me in my elsewhere stated view that usage and grammaticality are more important than terminology and classification.

I don't know Shoe. People have different views on language, I won't get into that. This answer has already been downvoted once (and also upvoted once), and I got the same on other threads where I submitted my answers. I can only say that those who upvoted obviously understood what they were reading and those who downvoted are obviously lost to grammar stuff forever :)

Maybe the downvoter (who wasn't me by the way) disagreed with your classification of singing here as a noun. It would be nice if downvoters could explain their reasoning, but that is not going to happen on this site. And assuming that the time and effort taken to craft a good answer correlates with the number of upvotes it receives is a recipe for disappointment. You didn't respond to my question about mongrels (which I understand as subsective/intersective gradience). But let's return to this when we both have time to chat.

I don't have any doubt that the downvoter has some illuminating ideas on the subject, so I hope he'll be kind enough to contribute them here soon. I'm short of time as it is, so I will let others take over from here for now. We can schedule some time for a quick chat, it would be nice. As for this point, the idea of gradience of -ing forms is illustrated in examples in Quirk's grammar. If you ask me, I don't find it satisfactory and I suggest you change the perspective. There are moot points in word classifications but they are few and far between considering the bulk of the vocabulary.

Yes, but one of the mootest of points concerns the gerund-particpial which is at issue here. I'll move this to chat now and we can maybe pick up on it later.

Hi Shoe! I'll just briefly discuss the point now, and I'll get back to it later okay?

My point is that verbs and nouns are not clearly differentiated in regard of their orthographic form. But that is not in any way peculiar to the -ing nouns and verbs. Probably a better part of nouns and verbs are used in one or the other way. Some are typically nouns and others are typically verbs. We can illustrate this in any example. Off the top of my head, we will use the word "explore" as a verb, rather than a noun, but not exclusively so. "Let's have an explore of the area".

Or the other way around, "table" is typically a noun, but not excluded as a verb either. The same form of the word doesn't prevent us from easily recognizing whether the word in question is a verb or a noun. Of course, lexical processes and suffixes will do away with such dilemmas even orthographically so we'll have "destruct" and "destruction", "identify" and "identification" etc.

In the example from the thread, we could replace "singing" with "work" and nobody would raise a question about dual nature of the word "work". The question is why not? The answer is beats me. People are just stuck at this "gerund" thing and they won't budge. The upvoted answer on the thread, is just wrong on so many levels, I just have to briefly address the logical inconsistencies that it contains.

1. I really wonder, in which way the word "swimming" "describes the action of the verb" differently from any other verb-derived noun which doesn't end in -ing. Does the word "identification" describes the action of identifying? or "Activation" the action of "activating" or any other possible verbal noun that we can think of.

2. The second point is more problematic. We created a cross-breed of a noun and verb and now this word is modified by adverbs (which no noun is ever) and functions as the subject or object (which no verb does ever). So in this analysis, "swimming quickly" is a really strange PHRASE.

The head of the phrase is "gerund", modified by a n adverb. And what happens, if I just put a comma after it, to cut it off from the rest of the sentence - no more subject, no more gerund, then what is it in that case? Like, "Swimming quickly, he reached the shore". "Swimming" still describes the action, but the other crucial point for the definition of this class is missing now - it is no more the subject of the sentence. This is what happens when form and function are jumbled together.

To account for this fact, our "swimming" in "swimming quickly" has to be reclassified. You change the function, you changed the form. The problem starts compounding the further we get into the logical analysis of the form. The biggest problem, to cut the long story short, is that by understanding "swimming quickly" as a gerund phrase or whatever, is that we fail to see its strikingly clausal properties. It is not a phrase at all, it is a clause. It forms a system of clauses together with

ed participials and to-infinitivals. Non-finite clauses have so much in common in terms of their syntactic properties and distribution. In "Swimming quickly, he reached the shore" , the part "swimming quickly" is a subjectless clause (also tenseless, lacking the auxiliary) . In "Swimming quickly keeps you healthy" it is no less so, the only difference being in the function that this clause performs. In the supplementive function it borrows the subject from the main clause.

We could rearrange the clause into a finite one : He swam quickly and soon reached the shore. Anyway, that is as much as I had to say on the subject here Shoe. You should take a look at CGEL, it will change your perspective completely, I can promise you that.