Postpositions?

[zompist]

These things are often called verb particles and I've also heard verb extensions. Some have to be contiguous with the verb, like "I worked out this morning" "I finished up painting the house" vs. *"I worked this morning out" and *"I finished painting the house up". Others can appear in either position: "I figured out the system" and "I figured the system out". They're usually etymologically/morphologically identical to prepositions and adverbs, but like people have pointed out, they're quite specific in their distribution and usually a part of the verb in a sense, because the verb will often change meaning without them.

"Particles" is fine. But there are several subcategories:

1) where the particle retains its prepositional meaning and perhaps an implied object
Are you coming with (me)?
He ran out (the door) and then back in (the door).
I'm voting against (the bill).
Let's go around (the obstacle).
2) where the particle retains its prepositional meaning, but has no object:
Can you help me up?
I've fallen behind in my work.
3) where the particle seems to have a new, often idiomatic meaning, and has no object:
I worked out today.
Let's look it up in the dictionary.
We ran out of beer.
4) Ditto, but the particle can be backed.
We helped each other out.
The police brought the criminal in.
Those guys really worked him over.

Now, in syntax, there is a case to be made that a preposition may just not have an object. This may sound weird and wrong, but I'd point out category 1 above, where there's an optional object— there's absolutely nothing syntactically shocking about deletions. Plus, we're not surprised when a verb has no object. In X' theory, N / V / P have a lot of similarities, and this may be one of those.

I'd also note that historically, European languages love to stack prepositions up. English has into, about < 'on without', until < 'up to to', within, without. French has dedans, or the amazing au-dessous de ('at the of under of'). My favorite example tho' is Spanish en adelante 'henceforth' < 'in at of in before'. This sort of thing suggests that it's not correct to look at prepositions as just appearing in the P + NP frame.

[WarpedWartWars]

I'd also note that historically, European languages love to stack prepositions up. English has into, about < 'on without', until < 'up to to', within, without. French has dedans, or the amazing au-dessous de ('at the of under of'). My favorite example tho' is Spanish en adelante 'henceforth' < 'in at of in before'. This sort of thing suggests that it's not correct to look at prepositions as just appearing in the P + NP frame.

I'm reading the Conlanger's Lexipedia, and in it somewhere, you pointed out above coming from a bufan 'at above', bufan itself coming from be ufan (I think) 'be above' (I think). You also pointed out the dedans and en adelante ones there.

[evmdbm]

But if that is right, why does sentence 2 mean the same thing and why we can we not just agree on where the separable prefix goes, because it really does look like a preposition in sentence 2.

Why would we need to do that? Since put is a transitive verb, there's no ambiguity. For on my coat to be interpreted as a prepositional phrase in I put on my coat, the sentence would need an explicit direct object (e.g. "I put it on my coat".)

Fair point: that's not ambiguous, but
"I put my coat on..." might be. You might reply "You put your coat on what?" and I could reply either, "No, I put my coat on." or "Sorry, I put my coat on the table."

In any case it seems a bit odd to me that in some of these cases the particle/preposition floats about and sometimes it doesn't. Obviously it can't float if it's really a "preposition"
1. I climb up the mountain (not I climb the mountain up), but what really is the difference between
2. I put my coat on
3. I put on my coat

and
4. I believe in Boris Johnson (but not)
5. I believe Boris Johnson in

Both seem to me to examples of Kath's and others' complex verb. "To put on" means something different to "to put" and likewise "to believe" means something different to "to believe in". Strangely we allow free float in the case where there might be some ambiguity, but not where there can't be; there's no motion going on in sentences 4 or 5 so not possible confusion.

[KathTheDragon]

I think you've missed the point of positing a class of complex verbs. "to believe in" is not a complex verb because it does not behave like other verbs in the class. You can't insist that a composition must be in the class because the meaning differs from the base verb. That's not how the class is defined. Descriptively, a complex verb *requires* the adverb/preposition component to freely float.

Also I don't believe that you actually could get any ambiguity in saying "I put my coat on" and then getting cut off. The prosody of "I put my coat on" and "I put my coat on X" are different and that would make it clear which is intended.

[evmdbm]

I think you've missed the point of positing a class of complex verbs. "to believe in" is not a complex verb because it does not behave like other verbs in the class. You can't insist that a composition must be in the class because the meaning differs from the base verb. That's not how the class is defined. Descriptively, a complex verb *requires* the adverb/preposition component to freely float.

Also I don't believe that you actually could get any ambiguity in saying "I put my coat on" and then getting cut off. The prosody of "I put my coat on" and "I put my coat on X" are different and that would make it clear which is intended.

You did say

Such verbs (all, or only some?) have the rule that the adverb/preposition can be split from the simplex verb by the object.

But ok, so the class of verbs requires a free float to be a complex verb, so what about
1. "I believe in Boris" Is that metaphorical motion or positioning and in the same "class" as
2. "I climb up the mountain."

[KathTheDragon]

You did say

Such verbs (all, or only some?) have the rule that the adverb/preposition can be split from the simplex verb by the object.

Looking back, I have no idea why I added that qualifier. Disregard it; the only coherent approach is to use the syntactic behaviour as defining.

But ok, so the class of verbs requires a free float to be a complex verb, so what about
1. "I believe in Boris" Is that metaphorical motion or positioning and in the same "class" as
2. "I climb up the mountain."

I don't understand what you're asking here.

[Linguoboy]

Fair point: that's not ambiguous, but
"I put my coat on..." might be. You might reply "You put your coat on what?"

Only if I were a smartass. As kath says, the prosody indications whether the speaker has failed to complete the thought or not, but even if I heard this sentence with trailing-off intonation, I would assume the next part was another action. I.e.
"I put my coat on..."
"And then what? What happened next?"

In any case it seems a bit odd to me that in some of these cases the particle/preposition floats about and sometimes it doesn't. Obviously it can't float if it's really a "preposition"
1. I climb up the mountain (not I climb the mountain up), but what really is the difference between
2. I put my coat on
3. I put on my coat

and
4. I believe in Boris Johnson (but not)
5. I believe Boris Johnson in

I don't see how this is much different than German, where you find a similar pattern despite also having compound verbs with both separable and inseparable prefixes:

Ich glaube an Boris Johnson. "I believe in Boris Johnson."
*Ich glaube Boris Johnson an. *"I believe Boris Johnson in."
Ich glaube Boris Johnson. "I believe Boris Johnson."
Ich (ver)traue Boris Johnson. "I trust Boris Johnson."
Ich traue es Boris Johnson zu. "I give Boris Johnson credit for it."

[evmdbm]

nly if I were a smartass. As kath says, the prosody indications whether the speaker has failed to complete the thought or not, but even if I heard this sentence with trailing-off intonation, I would assume the next part was another action. I.e.
"I put my coat on..."
"And then what? What happened next?"

OK you can stop hitting me over the head now

I think we've established that they are different in the behaviour of the preposition/particle and that my own suspicion that they might be like separable prefixes in German is right. But that just establishes that they are different syntactically. I was wondering why - what semantic difference (say) explains the different syntactic behaviour of these words. Maybe I'm asking a question with no answer

[Moose-tache]

It all seems very simple to me. There is a historical spectrum from preposition to adverb.

Stage 1: a preposition frequently appears with a certain verb, subtly changing its meaning. It is still a preposition. For example, "look to."

Stage 2: since the preposition has some modifying effect on the verb now, speakers begin to apply some of the syntactic rules that apply to adverbs, such as limited (usually optional) movement. But not all the features of adverbs; it still can't appear without an object. For example, "put on."

Stage 3: the word is now treated more like an adverb than a preposition, to the point that objects are completely optional. For example, "give up." Since this combination of verb and adverb is its own lexical unit and not a productive combination, the adverb does not have the full range of syntactic movement of a normal adverb like "quickly" or "actually."

You'll notice some words appear more in some stages than in others. "To" rarely appears in stage 3 (the only example I can think of off the top of my head is "snap to"), while many examples from stage 3 use directional words like "up" that are already not limited to prepositional use. It's hardly surprising that words considered more dedicated to prepositional use would stay closer to one end of the spectrum. The whole process dates, at least to some extent, to Proto-Germanic, but most of the details are language specific, so it may have been fairly rudimentary until well into the first millennium. The point is, we don't need any special terminology about "particles," or "separable prefixes," or "postpositions." It's all explainable using the basic blocks of English grammar.

[bradrn]

It all seems very simple to me. There is a historical spectrum from preposition to adverb.

Stage 1: a preposition frequently appears with a certain verb, subtly changing its meaning. It is still a preposition. For example, "look to."

Stage 2: since the preposition has some modifying effect on the verb now, speakers begin to apply some of the syntactic rules that apply to adverbs, such as limited (usually optional) movement. But not all the features of adverbs; it still can't appear without an object. For example, "put on."

Stage 3: the word is now treated more like an adverb than a preposition, to the point that objects are completely optional. For example, "give up." Since this combination of verb and adverb is its own lexical unit and not a productive combination, the adverb does not have the full range of syntactic movement of a normal adverb like "quickly" or "actually."

You'll notice some words appear more in some stages than in others. "To" rarely appears in stage 3 (the only example I can think of off the top of my head is "snap to"), while many examples from stage 3 use directional words like "up" that are already not limited to prepositional use. It's hardly surprising that words considered more dedicated to prepositional use would stay closer to one end of the spectrum. The whole process dates, at least to some extent, to Proto-Germanic, but most of the details are language specific, so it may have been fairly rudimentary until well into the first millennium. The point is, we don't need any special terminology about "particles," or "separable prefixes," or "postpositions." It's all explainable using the basic blocks of English grammar.

I disagree with this idea. The problem is that there’s still a big difference between your Stage 3 and full adverbs. Adverbs are fully semantically independent from the verb: the meaning of the adverb+verb complex is purely compositional. By contrast, combinations such as ‘give up’ are idiomatic and lexicalised: not only does ‘up’ have no freedom of movement (the opposite of adverbs), it also has no meaning of its own independent of ‘give’ (and similarly ‘give’ has no independent meaning of its own independent of ‘up’). For this reason, I find it extremely difficult to consider these words ‘adverbs’ in any way, except insofar as ‘adverb’ gets used as a meaningless cover-all word in traditional English grammar.

[Moose-tache]

The problem is that there’s still a big difference between your Stage 3 and full adverbs. Adverbs are fully semantically independent from the verb: the meaning of the adverb+verb complex is purely compositional. By contrast, combinations such as ‘give up’ are idiomatic and lexicalised: not only does ‘up’ have no freedom of movement (the opposite of adverbs), it also has no meaning of its own independent of ‘give’ (and similarly ‘give’ has no independent meaning of its own independent of ‘up’). For this reason, I find it extremely difficult to consider these words ‘adverbs’ in any way, except insofar as ‘adverb’ gets used as a meaningless cover-all word in traditional English grammar.

With respect, you are falling into a common misunderstanding. "Adverb" in formal English vocabulary is a catch-all term, within which there are countless subcategories and subclasses, each with their own syntactic rules. If you read what I wrote, I did say that the "up" in "give up" does not behave the same as an adverb like "actually," hence the relatively fewer number of allowable positions the word can take. When I say the "up" here is being reanalyzed as if it were more adverby and less prepositiony, I am not implying that Adverb is a clear and scientifically consistent category either in formal linguistics or in the minds of English speakers. I am merely saying that English speakers recognize that the word is modifying the verb more than it is modifying a noun phrase, and treat it accordingly, borrowing traits from word classes that spend most of their time modifying verbs. As for the idea that a word cannot be an adverb unless is has complete semantic independence from the verb, I think this is a very tricky proposition to apply to any part of speech. Can we say that the word "tea" in the phrase "spill the tea" is no longer a noun because its meaning cannot be understood indepenedently like most nouns? Semantic independence is a spectrum, and a poor measurement of which part of speech a word should be placed into.

To reiterate, English speakers do not divide words into perfect categories, but assign various syntactic functions to them, often overlapping. A preposition can become more adverby in its function, even if there is no clear definition of what an adverb is. Perfectly regular parts of speech and semantic independence are illusions that may be helpful shorthand in grammar documents, but they do not describe real language.

[Estav]

It all seems very simple to me. There is a historical spectrum from preposition to adverb.

Stage 1: a preposition frequently appears with a certain verb, subtly changing its meaning. It is still a preposition. For example, "look to."

Stage 2: since the preposition has some modifying effect on the verb now, speakers begin to apply some of the syntactic rules that apply to adverbs, such as limited (usually optional) movement. But not all the features of adverbs; it still can't appear without an object. For example, "put on."

Stage 3: the word is now treated more like an adverb than a preposition, to the point that objects are completely optional. For example, "give up." Since this combination of verb and adverb is its own lexical unit and not a productive combination, the adverb does not have the full range of syntactic movement of a normal adverb like "quickly" or "actually."

You'll notice some words appear more in some stages than in others. "To" rarely appears in stage 3 (the only example I can think of off the top of my head is "snap to"), while many examples from stage 3 use directional words like "up" that are already not limited to prepositional use. It's hardly surprising that words considered more dedicated to prepositional use would stay closer to one end of the spectrum. The whole process dates, at least to some extent, to Proto-Germanic, but most of the details are language specific, so it may have been fairly rudimentary until well into the first millennium. The point is, we don't need any special terminology about "particles," or "separable prefixes," or "postpositions." It's all explainable using the basic blocks of English grammar.

I don't think this kind of progression is how English "particle verb" constructions developed. What you're saying here is that "[put] [on] [the hat]" would come from an earlier "put [on the hat]", right? But in this hypothetical "Stage 1" where "the hat" is contained in a prepositional phrase headed by "on", the verb would have no direct object (put what on the hat?) and the meaning has no clear pathway to the meaning of the present-day construction. It seems more likely to me that, if these structures actually developed at all from structures with originally transitive prepositions, it is by omission or elision of a separate object of the preposition, causing "on" by itself to have a similar meaning to a prepositional phrase: "[put] [the hat] [on yourself/your head/your body]" / "[put] [on yourself] [the hat]"> "[put] [the hat] [on]" / "[put] [on] [the hat]", similar to how some modern varieties of English have structures like "Are you coming with?" where the preposition appears to speakers of other varieties to have "lost" its easily inferred original complement.

[bradrn]

"Adverb" in formal English vocabulary is a catch-all term, within which there are countless subcategories and subclasses, each with their own syntactic rules.

I am fully aware of this, and have complained about it numerous times. However, if we using this catch-all definition of ‘adverb’, your statement becomes meaningless and vacuously true. For this reason I assumed you were using the term in the stricter sense.

When I say the "up" here is being reanalyzed as if it were more adverby and less prepositiony, I am not implying that Adverb is a clear and scientifically consistent category either in formal linguistics or in the minds of English speakers. I am merely saying that English speakers recognize that the word is modifying the verb more than it is modifying a noun phrase, and treat it accordingly, borrowing traits from word classes that spend most of their time modifying verbs.

And which traits would these be? ‘Adverb’, as you yourself said, is not a coherent word class, and has no coherent defining traits.

As for the idea that a word cannot be an adverb unless is has complete semantic independence from the verb, I think this is a very tricky proposition to apply to any part of speech. Can we say that the word "tea" in the phrase "spill the tea" is no longer a noun because its meaning cannot be understood indepenedently like most nouns?

Of course we can’t say that, because ‘tea’ does have an independent meaning in this sentence. Specifically, it means ‘a drink extracted from the leaves of the eudicot Camellia sinensis via submersion in hot water, or for some speakers, any other drink prepared from leaves in a similar manner’. A much better example would be ‘spill the beans’, where ‘beans’ can no longer be analysed separately from the rest of the sentence — but I think we would all agree that this is an idiom, not to be analysed in terms of individual words. It is the same with ‘give up’.

EDIT: Actually, on reflection, we can still get some useful word class information from such expressions. After all, ‘beans’ is still syntactically acting as a noun in ‘spill the beans’ — e.g. applying a syntactic test, it can take an adjective, ‘spill the small beans’, even if it stops being an idiom at this point. I’ll have to think more about this.

Semantic independence is a spectrum, and a poor measurement of which part of speech a word should be placed into.

I tend to disagree. A certain phrase is either compositional or non-compositional — if you prefer, non-idiomatic or idiomatic. It is a logical impossibility to have something with a meaning which is ‘a bit deducible from its components, but also a bit non-deducIble’. I do agree that it’s a poor measurement of part of speech, but only because there are no words found solely in idioms.

I realise this is a dangerous thing to say; feel free to supply me with counterexamples if you can find any.

EDIT: Never mind, found a counterexample. ‘Herbal tea’ is more than the sum of its parts, since it specifically excludes Camellia sinensis tea… yet it is still mostly compositional, since herbal teas are both herbal and tea (at least for many speakers).

To reiterate, English speakers do not divide words into perfect categories, but assign various syntactic functions to them, often overlapping. A preposition can become more adverby in its function, even if there is no clear definition of what an adverb is. Perfectly regular parts of speech and semantic independence are illusions that may be helpful shorthand in grammar documents, but they do not describe real language.

But they do describe real language! Very well, in fact. ‘Syntactic class X’ is merely a shorthand for saying that ‘words belonging to X are distinguishable by sharing such-and-such traits’. In any given language, there should be very few words which do not fit into one or more word classes.

I highly recommend reading François’s paper on word classes in Hiw, which presents a unified approach to defining word classes. It really helped me understand the usefulness of the concept, and is perhaps one of the best papers I’ve ever read (which is why I keep on linking it here).

[Richard W]

I think up in give up started out as an adverb. Indeed, in 'climb up the hill', the development seems to be the opposite - from adverb to preposition. This is assumed to be the usual IE direction of development. Remember that the earliest meaning of 'preposition' is something (loosely) prefixed to a verb. I say loosely, because in the branches with the augment, that goes between the prefix and the verb; in the reduplicated form, the preposition is prefixed to the reduplicated form.

For put on, if on therein originally be a preposition, its implied object is the location, which is frequently not expressed. For example, a fuller form is He put the bra on his head. Perhaps more typical is to repeat the word on, as in 'He put on the sock on his left foot' or 'He put the sock on on his left foot'. As an adverb, on goes at least as far back as Proto-Germanic.

[bradrn]

It all seems very simple to me. There is a historical spectrum from preposition to adverb.

Stage 1: a preposition frequently appears with a certain verb, subtly changing its meaning. It is still a preposition. For example, "look to."

Stage 2: since the preposition has some modifying effect on the verb now, speakers begin to apply some of the syntactic rules that apply to adverbs, such as limited (usually optional) movement. But not all the features of adverbs; it still can't appear without an object. For example, "put on."

Stage 3: the word is now treated more like an adverb than a preposition, to the point that objects are completely optional. For example, "give up." Since this combination of verb and adverb is its own lexical unit and not a productive combination, the adverb does not have the full range of syntactic movement of a normal adverb like "quickly" or "actually."

Having thought some more about this, I don’t even understand why you give special names to Stage 2 and Stage 3. It seems to be that Stage 2 verbs are simply transitive, whereas Stage 3 verbs are ambitransitive. As I said already, talking in terms of ‘adverbs’ is meaningless unless you use the word in the strict sense, in which case this is false.

I think the best distinction to make here, as I have already said, is that between ‘preposition’ and ‘particle’. The key syntactic test here is ability to behave as copula complement:

It is in the door
It is on the door
It is for the door

But:

*It is up the door
*It is down the door

(There are marginally acceptable cases, e.g. ?It is up the stairs, but here I think I would prefer It is at the top of the stairs, with a ‘true’ preposition. To is an interesting case here, since by this test it is not a preposition, but it patterns with them in obligatorily taking a complement; for this reason I am inclined to treat it as a separate word class of its own.)

Beyond this, for phrasal verbs themselves, there is a distinction between phrasal verbs allowing the separable part to come after the object, and phrasal verbs not allowing this:

I put on my shoes
I put my shoes on

I gave up smoking
I gave smoking up

You can rely on her
*You can rely her on

There is also a distinction between fully transitive and ambitransitive verbs (perhaps with a difference in meaning):

I gave up smoking
I gave up

I put on my shoes
*I put on

Out of everything, this appears to be the least important distinction for classifying English phrasal verbs, since transitivity is not phrasal verb-specific, but rather applies to every single English verb we know of.

[Richard W]

To reiterate, English speakers do not divide words into perfect categories, but assign various syntactic functions to them, often overlapping. A preposition can become more adverby in its function, even if there is no clear definition of what an adverb is. Perfectly regular parts of speech and semantic independence are illusions that may be helpful shorthand in grammar documents, but they do not describe real language.

But they do describe real language! Very well, in fact. ‘Syntactic class X’ is merely a shorthand for saying that ‘words belonging to X are distinguishable by sharing such-and-such traits’. In any given language, there should be very few words which do not fit into one or more word classes.

I have come to suspect that generative grammar is only an approximation to how humans function. Rather, what we see is generalisation from other sentences, so we see alternative constructions as well as faulty productions from those who have not mastered the constructions. Where templates are not available, we see hesitation, not unlike the difficulty many have in saying stridden.

[bradrn]

To reiterate, English speakers do not divide words into perfect categories, but assign various syntactic functions to them, often overlapping. A preposition can become more adverby in its function, even if there is no clear definition of what an adverb is. Perfectly regular parts of speech and semantic independence are illusions that may be helpful shorthand in grammar documents, but they do not describe real language.

But they do describe real language! Very well, in fact. ‘Syntactic class X’ is merely a shorthand for saying that ‘words belonging to X are distinguishable by sharing such-and-such traits’. In any given language, there should be very few words which do not fit into one or more word classes.

I have come to suspect that generative grammar is only an approximation to how humans function. Rather, what we see is generalisation from other sentences, so we see alternative constructions as well as faulty productions from those who have not mastered the constructions. Where templates are not available, we see hesitation, not unlike the difficulty many have in saying stridden.

I totally agree with all of this. But since word classes can be be deduced from the data only, they do act as a good description of how human languages function. Perhaps this is because they are in fact more like templates than anything else — ‘words belonging to X can be used in contexts A, B, C’.

[zompist]

I think the best distinction to make here, as I have already said, is that between ‘preposition’ and ‘particle’. The key syntactic test here is ability to behave as copula complement:

It is in the door
It is on the door
It is for the door

But:

*It is up the door
*It is down the door

(There are marginally acceptable cases, e.g. ?It is up the stairs, but here I think I would prefer It is at the top of the stairs, with a ‘true’ preposition. To is an interesting case here, since by this test it is not a preposition, but it patterns with them in obligatorily taking a complement; for this reason I am inclined to treat it as a separate word class of its own.)

Nothing wrong with "up the stairs" for me. Also up the street, up the mountain, up the creek, up the food chain, up one's ass, down the line, down the hatch, down the drain, down the hole, down the page. Your test is not a test of prepositionhood... I don't know what it is a test of, since lots of prepositions fail to work in the frame "It is ___ the door": between, across, among, except, from, into, onto, since, to, until, without...

I read François’s paper, and don't see that he makes a case against fuzzy categories-- he doesn't even seem to be aware of the concept. He talks about prototypical English nouns and verbs, but then just asserts that these categories are clear-cut and exceptionless. E.g. he claims that "know" can't be used as a noun... what about "to know"? "know how"? "know-nothing"? "know-it-all"? Or that fact that "knowing" falls somewhere in between nouns and verbs?

Also, Moose's "spill the tea" is an idiom, in American English at least (with the meaning "gossip").

[bradrn]

Nothing wrong with "up the stairs" for me. Also up the street, up the mountain, up the creek, up the food chain, up one's ass, down the line, down the hatch, down the drain, down the hole, down the page. Your test is not a test of prepositionhood... I don't know what it is a test of, since lots of prepositions fail to work in the frame "It is ___ the door": between, across, among, except, from, into, onto, since, to, until, without...

Some of those prepositions do work there: across, from, without. Others work if you alter the noun, since with door they’re not ungrammatical but merely semantically incoherent: It is between/among the doors. In fact, up and down also work if you alter the noun; contrary to what I said earlier, by this test they are truly prepositions (or at least, whatever this test is testing). But you’re correct in saying that this doesn’t work with several uncontroversial prepositions.

If I were to hazard a guess as to what exactly this is testing, on reflection it’s probably delineating a distinction between prepositions which don’t imply movement and prepositions which do — stative and dynamic prepositions, I might call them, respectively. In, on, for, at, up, down, between, across, among and from work, and none imply movement. Contrast the prepositions which don’t work: from, into, onto, to, which necessarily imply movement of some sort… but also except, since, until, without, which don’t. Either way, this is certainly a test of something, though I’m not quite sure what.

I read François’s paper, and don't see that he makes a case against fuzzy categories-- he doesn't even seem to be aware of the concept. He talks about prototypical English nouns and verbs, but then just asserts that these categories are clear-cut and exceptionless. E.g. he claims that "know" can't be used as a noun... what about "to know"? "know how"? "know-nothing"? "know-it-all"? Or that fact that "knowing" falls somewhere in between nouns and verbs?

‘Know’ is not acting as a noun in to know, which is a construction applicable to all verbs without exception — and also no nouns I know of. ‘Know-how’, ‘know-nothing’ and ‘know-it-all’ are all compounds: they are nouns, which happen to contain an (etymological) verb inside them. The first two don’t even obey normal English syntax. ‘Knowing’ fits in its own class, with some properties characteristic of nouns and some characteristic of verbs — but all verbs can occur in this form, meaning that ‘can take -ing’ is a characteristic delineating verbs specifically.

Also, Moose's "spill the tea" is an idiom, in American English at least (with the meaning "gossip").

Huh, I didn’t know that.

[zompist]

I read François’s paper, and don't see that he makes a case against fuzzy categories-- he doesn't even seem to be aware of the concept. He talks about prototypical English nouns and verbs, but then just asserts that these categories are clear-cut and exceptionless. E.g. he claims that "know" can't be used as a noun... what about "to know"? "know how"? "know-nothing"? "know-it-all"? Or that fact that "knowing" falls somewhere in between nouns and verbs?

‘Know-how’, ‘know-nothing’ and ‘know-it-all’ are all compounds: they are nouns, which happen to contain an (etymological) verb inside them. The first two don’t even obey normal English syntax.

They're lexicalized, of course, which is why I can cite them for all English speakers. And they do obey English syntax-- compare "I know how" and "I know nothing"!

The immediate point is, you can take a VP and treat it as a noun; and this isn't restricted to English-- cf. French gratte-ciel 'skyscraper'. This is exactly the sort of grammatical flexibility François talks about. The larger point is that languages are really complex, and looking at the most prototypical behaviors of word classes leads to many errors.

‘Know’ is not acting as a noun in to know, which is a construction applicable to all verbs without exception — and also no nouns I know of.
[...] ‘Knowing’ fits in its own class, with some properties characteristic of nouns and some characteristic of verbs — but all verbs can occur in this form, meaning that ‘can take -ing’ is a characteristic delineating verbs specifically.

But this is just pure dogma. Why do you call infinitives and participles "verbs"? Because some Latin grammarian did so 1500 years ago?

Compare: "To know is better than to guess" / "Knowledge is better than guesses." / "Knowing is better than guessing." These all act like nouns and take up an NP slot.

This isn't to say that nouns and verbs can't be distinguished-- only that there's a gradation between them. An infinitive is way more nouny than a verb, and it's arbitrary where we draw the line.