Linguistic Miscellany Thread

[bradrn]

I would call both of those inceptive as well. (Although the term seems to be subtly wrong for the Mandarin example — maybe ‘resultative’ would be a better term?)

Maybe, but "resultative" is used for something else in Mandarin.

Yes, I was a bit suspicious that I’d heard the name somewhere else…

In naming aspects, I'd suggest a Saussurian approach. That is, don't get too attached to the universal terms; think about what contrasts exist in your language.

I actually prefer the opposite approach: standardised names which mean the same thing no matter what language you’re looking at. That’s already how it works in phonetics: terms such as ‘voiceless’, ‘palatal’ and ‘lateral’ are standardised by the IPA (the Association, not the Alphabet), and so always mean the same thing (or at least very similar things) across multiple language. By contrast, terms such as ‘participle’, ‘adverb’ and ‘ergative’ have been used in so many ways across various languages that if you use them without qualification, they become nearly meaningless.

Of course, you should always consider the contrasts in each individual language… but please don’t change terms to have language-specific meanings! Try to find universal words instead.

With the "put on clothes" example, I don't think there's a universal answer to "what aspect is this".

Of course not; I was just talking about thethief3 and your examples, which are pretty similar in terms of aspect.

[Linguoboy]

I actually prefer the opposite approach: standardised names which mean the same thing no matter what language you’re looking at. That’s already how it works in phonetics: terms such as ‘voiceless’, ‘palatal’ and ‘lateral’ are standardised by the IPA (the Association, not the Alphabet), and so always mean the same thing (or at least very similar things) across multiple language. By contrast, terms such as ‘participle’, ‘adverb’ and ‘ergative’ have been used in so many ways across various languages that if you use them without qualification, they become nearly meaningless.

Of course, you should always consider the contrasts in each individual language… but please don’t change terms to have language-specific meanings! Try to find universal words instead.

But what works for phonetics doesn't work for syntax (and vice versa).

With phonetics, there's an essentially finite set of possibilities. There's only so many different significant positions your tongue can take in your mouth or so many different laryngeal positions that are worth differentiating. These can all be given more-or-less objective names. (Things get hairy on the level of phonology, where languages have all sorts of different ways of bundling these features, but the actual phones themselves can still be described with terrific accuracy.)

But that's just not the case with syntax. There are upper limits on, say, observed numbers of noun cases in living languages, but as Ithkuil demonstrated, that's nowhere near the limit of noun cases it's possible to learn to distinguish. More to Zomp's point, however, given the immensely varied and open-ended nature of human communication, it's impossible to ever definitively state how a particular inflection or construction is used. Sure, I could set out catalog every single known usage of the English progressive and produce a comprehensive list--which will be valid until the moment that some speaker somewhere reacts to a novel situation with a new usage.

So really, the best you can ever hope for is some sort of loose consensus on the "core" meaning of a usage, on the basis of which it gets a name of convenience. But it's always going to be the case that different languages (and different varieties of those languages and even different speakers of those varieties) are going to extend that "core" in different directions. So expecting a loose general term like "progressive" or "perfect" or "accusative" or "translative" or whatever to describe the same range of usage in whatever language it's applied to is really a mug's game.

[zompist]

I actually prefer the opposite approach: standardised names which mean the same thing no matter what language you’re looking at. That’s already how it works in phonetics: terms such as ‘voiceless’, ‘palatal’ and ‘lateral’ are standardised by the IPA (the Association, not the Alphabet), and so always mean the same thing (or at least very similar things) across multiple language. By contrast, terms such as ‘participle’, ‘adverb’ and ‘ergative’ have been used in so many ways across various languages that if you use them without qualification, they become nearly meaningless.

But in 3 of my 4 examples, those are the standard meanings! The only one that's maybe nonstandard is L&T's "durative".

Now, you could make a case that English "progressive" and French "imperfective" should be called the same thing. Except that "etre en train de" is sometimes called the progressive...

[Linguoboy]

Now, you could make a case that English "progressive" and French "imperfective" should be called the same thing. Except that "etre en train de" is sometimes called the progressive...

Except in Louisiana French, it's more of an inchoative (roughly equivalent to SF être sur le point de) while the closest match for the English progressive (and it's a fairly close match, being used much more frequently than être en train de in SF) is être après, e.g. "Quoi t'es après faire?" "What are you doing?"

[Kuchigakatai]

Personally, I think that ‘participle’ is a truly horrible piece of Indo-Europeanist terminology which is of no help and which should be scrapped.

That being said: as far as I’m aware, the active and passive participles are mainly used to construct the active and passive voice. Ergative languages with the passive (of which there are a few) could easily have them. But it’s also fairly common for ergative languages to have an antipassive rather than a passive, in which case they would presumably have stative and antipassive participles.

"Participle" is pretty useful when talking about European languages. (But zompist and linguoboy already gave you the response I intended about terminology.)

I have no idea about attested erg-abs langs really, but I imagine some of them probably ignore the topic of erg-abs alignment entirely when it comes to non-finite forms and imperatives...

I'm getting this suspicion from Latin, where the ever-famous deponent verbs (the we-only-have-passive-morphology verbs) treat the usual finite subject argument as the subject of all participles, surprisingly even the present-tense active-voice participle. So for lābor 'to slip, trip', present active lābēns is 'slipping' ([something] that is slipping), and past passive lāpsus is 'slipped' ([something] that slipped). Contrast mittēns 'sending [sth]' ([something] that sends sth) and missus 'sent' ([something] that was sent).

[Kuchigakatai]

3-SG-M wear-PROG clothes
"He puts on clothes"

"He puts on clothes" is a pretty bad example for the progressive, because of the various interpretations it can have in English (habitual vs. inchoative/inceptive vs. the imperfective+dynamic+progressive reading you intended). (I think Zompist and bradrn misread it above when they suggested "inceptive".) Note that Wikipedia correctly uses the "be V-ing" construction instead when translating the Cantonese/Mandarin examples, which is clearer in English for the intended meaning: "he is putting on clothes".

(Please note I'm using "dynamic" in the sense of something that creates a change on the subject or object and is typically fairly voluntary and conscious and a physical action. This is opposed to "stative", which doesn't imply a change in the arguments and/or is fairly involuntary and more about experiences or background environments. The term "dynamic" has another less frequent use as a synonym of "inchoative/inceptive" sometimes.)

What could be done about inherently telic actions (like say to ignite something). Could you use the progressive to mark failure to complete an action or is that just marking telicity.

Both continuous and progressive are typically strongly atelic (they're used for actions that last a while, and where the bound/limit/goal of a completed action is not the focus), so you probably want to do something else. Mandarin and Cantonese have more aspect markers than those two. Mandarin, for one, often uses 了 -le instead to mark the perfective+telic aspects (although 了 -le can also express some atelic stuff), and similarly Cantonese uses 咗 -jo2. Both languages can also leave verbs unmarked for aspect.

Tangential question: what exactly is the difference between progressive and continuous aspect? I’ve never understood the difference.

When I took ESL classes, both textbooks and teachers treated "continuous" and "progressive" as interchangeable synonyms for the "be V-ing" construction.

I have sometimes seen grammars where continuous is used for more stative notions and progressive for more dynamic ones, which is what Wikipedia tries to stress in that article. But the two aspectual notions are pretty similar, and I'm sure some linguists have used the terms the other way around out of confusion, protest, etc. "Durative" is another word, rarely used, that is sometimes a synonym or a near-synonym (and I've seen it in other works than Li & Thompson too).

You could define these as two aspects, but I emphasize that Li & Thompson do not; they consider it one aspect with two types of verbs. In the case of the dual-sense verbs, the senses are predictable: an action (take, put on) vs. a state resulting from that action (hold, wear).

I think it would've been better if Li & Thompson had said that both -zhe and (zhèng)zài share the aspect of a notion (action or state) that lasts for a while (which opposes them to the core meanings of -le and -guo), which seems to be what they mean when they say -zhe and zài are "one aspect". Then, it's another aspectual axis that's being used to contrast -zhe vs. zài, namely stative (-zhe) vs. dynamic (zài). But oh well, they seem to have pursued a model where one particle corresponds to one aspect, even if it meant creating a category of "dual-sense" verbs, whatever merit that had...

[Richard W]

But what works for phonetics doesn't work for syntax (and vice versa).

With phonetics, there's an essentially finite set of possibilities.

I'm not so sure a similar problem doesn't occur with vowels, where there's an essentially infinite set of possibilities, and each vowel has a range of possible realisations.

[Travis B.]

But what works for phonetics doesn't work for syntax (and vice versa).

With phonetics, there's an essentially finite set of possibilities.

I'm not so sure a similar problem doesn't occur with vowels, where there's an essentially infinite set of possibilities, and each vowel has a range of possible realisations.

Even then, the vowel space can only get so crowded (e.g. how common is it to have more than five distinctions of vowel height?).

[bradrn]

Oh, dear. I woke up this morning, and it looks like quite a few people disagree with the point I made last night. Now I’ll have to respond to all of them at once! Well, here I go:

I actually prefer the opposite approach: standardised names which mean the same thing no matter what language you’re looking at. That’s already how it works in phonetics: terms such as ‘voiceless’, ‘palatal’ and ‘lateral’ are standardised by the IPA (the Association, not the Alphabet), and so always mean the same thing (or at least very similar things) across multiple language. By contrast, terms such as ‘participle’, ‘adverb’ and ‘ergative’ have been used in so many ways across various languages that if you use them without qualification, they become nearly meaningless.

Of course, you should always consider the contrasts in each individual language… but please don’t change terms to have language-specific meanings! Try to find universal words instead.

But what works for phonetics doesn't work for syntax (and vice versa).

With phonetics, there's an essentially finite set of possibilities. There's only so many different significant positions your tongue can take in your mouth or so many different laryngeal positions that are worth differentiating. These can all be given more-or-less objective names. (Things get hairy on the level of phonology, where languages have all sorts of different ways of bundling these features, but the actual phones themselves can still be described with terrific accuracy.)

But that's just not the case with syntax. There are upper limits on, say, observed numbers of noun cases in living languages, but as Ithkuil demonstrated, that's nowhere near the limit of noun cases it's possible to learn to distinguish.

I do agree that syntax (and morphology etc.) does have a nearly limitless number of possibilities. However, in practice, most of these possibilities aren’t used by natlangs (as us conlangers are painfully aware). There are certainly few enough common possibilities that there should be no problem defining standard terms for them. (And I’m fine with coinages for new terms.)

More to Zomp's point, however, given the immensely varied and open-ended nature of human communication, it's impossible to ever definitively state how a particular inflection or construction is used. Sure, I could set out catalog every single known usage of the English progressive and produce a comprehensive list--which will be valid until the moment that some speaker somewhere reacts to a novel situation with a new usage.

So really, the best you can ever hope for is some sort of loose consensus on the "core" meaning of a usage, on the basis of which it gets a name of convenience. But it's always going to be the case that different languages (and different varieties of those languages and even different speakers of those varieties) are going to extend that "core" in different directions. So expecting a loose general term like "progressive" or "perfect" or "accusative" or "translative" or whatever to describe the same range of usage in whatever language it's applied to is really a mug's game.

If a language uses a construction for one thing in a new way, I’m fine with calling that by the same name (although only when you’re talking about that language). And if the boundary between two uses is blurry, I’m fine with that blurriness as well. What I’m not OK with is names which are so vague in the first place as to be useless, or names which have been misused in so many ways outside the original definition that they have become vague through overuse. (And once you’ve gotten rid of all these vague terms, that automatically lets you re-use terms across languages, since they now have a single main definition, even if it gets a little bit blurry around the edges.)

I actually prefer the opposite approach: standardised names which mean the same thing no matter what language you’re looking at. That’s already how it works in phonetics: terms such as ‘voiceless’, ‘palatal’ and ‘lateral’ are standardised by the IPA (the Association, not the Alphabet), and so always mean the same thing (or at least very similar things) across multiple language. By contrast, terms such as ‘participle’, ‘adverb’ and ‘ergative’ have been used in so many ways across various languages that if you use them without qualification, they become nearly meaningless.

But in 3 of my 4 examples, those are the standard meanings! The only one that's maybe nonstandard is L&T's "durative".

I’m not sure I understand what you’re saying here… which 3 examples are you talking about? Because if you’re talking about ‘progressive’, ‘continuous’ and ‘resultative’, I don’t follow your argument:

• Progressive is… um… some sort of imperfective?
• Continuous is… um… another sort of imperfective? Certainly no-one seems to know how it’s different to the progressive.
• Resultative does indeed have a standard definition. (I misused it because I wasn’t aware that it was a previously-used term.)

So, two of your examples seem to have no standard meanings at all… so I’m not quite sure how you can argue that you use them according to their meanings. (This is an example of what I was talking about above, with ‘names which are so vague in the first place as to be useless’.)

Personally, I think that ‘participle’ is a truly horrible piece of Indo-Europeanist terminology which is of no help and which should be scrapped.

That being said: as far as I’m aware, the active and passive participles are mainly used to construct the active and passive voice. Ergative languages with the passive (of which there are a few) could easily have them. But it’s also fairly common for ergative languages to have an antipassive rather than a passive, in which case they would presumably have stative and antipassive participles.

"Participle" is pretty useful when talking about European languages. (But zompist and linguoboy already gave you the response I intended about terminology.)

I know it’s useful for European languages. But the majority of ergative languages are non-European. (Although I do believe Hindi has participles, as an Indo-European language.) So unless you can come up with a universal definition of ‘participle’ which works outside Europe, Zju’s question is meaningless.

I have no idea about attested erg-abs langs really, but I imagine some of them probably ignore the topic of erg-abs alignment entirely when it comes to non-finite forms and imperatives...

Dixon notes that even the most thoroughly ergative languages always behave in a nominative-accusative way with regards to imperatives. I’m not too sure about non-finite forms, but I think there’s a couple of languages which treat them in an ergative way. (I’ve been meaning to read up about this topic though; I don’t know much about ergativity in syntax.)

(But how is this related to participles? I thought participles are usually finite and non-imperative, although admittedly I don’t know much about them.)

[Estav]

(But how is this related to participles? I thought participles are usually finite and non-imperative, although admittedly I don’t know much about them.)

The term "participle" is always used to refer to a non-finite verb form, as far as I know. Basque has a non-finite verb form that can be called a participle, although Wikipedia enigmatically states that "not all its uses are really participial".

Another language with ergative case marking and participles is Sirenik Eskimo (extinct).

[bradrn]

(But how is this related to participles? I thought participles are usually finite and non-imperative, although admittedly I don’t know much about them.)

The term "participle" is always used to refer to a non-finite verb form, as far as I know.

Oh, right. I really should have checked that I actually know the definition before I commented. I’ve never really understood participles — but looking at the Wikipedia article on them, it looks like they’re just an adjectival form of a verb. Is that correct?

Another language with ergative case marking and participles is Sirenik Eskimo (extinct).

That’s interesting! I had thought that participles were unknown outside Europe. I’ll have a look at it then.

[Estav]

Oh, right. I really should have checked that I actually know the definition before I commented. I’ve never really understood participles — but looking at the Wikipedia article on them, it looks like they’re just an adjectival form of a verb. Is that correct?

I think that's the general idea; the specifics of how participles are used can be complicated and the analysis of what they are is disputed. In English (and apparently also in Basque) some words with the form of a participle behave pretty much exactly like adjectives. An English example is "exciting" in a context like "that seems very exciting". Words like this can be analyzed as adjectives derived from verbs--it's commonly assumed that conversion from one part of speech to another implies a process of derivation instead of inflection, although that's been argued.

But there are other participle forms that generally can't be used in all the places where an adjective can be used: e.g. in English, we can say "the child is playing" or "a playing child", but it doesn't sound good to say things like *"a very playing child" or *"the child seems playing". Forms like "playing" that have a more restricted distribution than adjectives are often considered to be inflected forms of verbs, rather than belonging to the adjective POS (although some linguists have argued that even this second type are in fact adjectives). In English, there are also contexts where participles can be used but adjectives can't: e.g. in the "have + past participle" perfect construction in English.

The word "participle" is sometimes reserved for the second type, with the first type (the type that more completely shows typical adjective behavior) called things like "participial adjectives" or "departicipial adjectives". Other authors use "participle" to refer to both types of words, and call the first "adjectival participles" and the second "verbal participles".

[bradrn]

Oh, right. I really should have checked that I actually know the definition before I commented. I’ve never really understood participles — but looking at the Wikipedia article on them, it looks like they’re just an adjectival form of a verb. Is that correct?

I think that's the general idea; the specifics of how participles are used can be complicated and the analysis of what they are is disputed. In English (and apparently also in Basque) some words with the form of a participle behave pretty much exactly like adjectives. An English example is "exciting" in a context like "that seems very exciting". Words like this can be analyzed as adjectives derived from verbs--it's commonly assumed that conversion from one part of speech to another implies a process of derivation instead of inflection, although that's been argued.

But there are other participle forms that generally can't be used in all the places where an adjective can be used: e.g. in English, we can say "the child is playing" or "a playing child", but it doesn't sound good to say things like *"a very playing child" or *"the child seems playing". Forms like "playing" that have a more restricted distribution than adjectives are often considered to be inflected forms of verbs, rather than belonging to the adjective POS (although some linguists have argued that even this second type are in fact adjectives). In English, there are also contexts where participles can be used but adjectives can't: e.g. in the "have + past participle" perfect construction in English.

The word "participle" is sometimes reserved for the second type, with the first type (the type that more completely shows typical adjective behavior) called things like "participial adjectives" or "departicipial adjectives". Other authors use "participle" to refer to both types of words, and call the first "adjectival participles" and the second "verbal participles".

That’s the clearest explanation of what participles are that I’ve ever seen — thanks for explaining! I can see how those pseudo-adjectival forms end up as progressives, perfects etc.

One question: is there any way to predict whether a given verb will end up as an adjectival or verbal participle, or is it irregular?

[zompist]

But in 3 of my 4 examples, those are the standard meanings! The only one that's maybe nonstandard is L&T's "durative".

I’m not sure I understand what you’re saying here… which 3 examples are you talking about? Because if you’re talking about ‘progressive’, ‘continuous’ and ‘resultative’, I don’t follow your argument:

• Progressive is… um… some sort of imperfective? [...]

OK, I'll put this down to not having explained clearly, and I'll try again.

My point is, if you look at a situation, or at a single verb, there is no God-given answer on what aspect is involved, or what name to give it.

First, look at a situation: putting on something. I wasn't wearing a hat; now I am; I'm still wearing it. What aspect is that? There's no general answer; it depends on the aspectual resources of the language, and my pragmatic choices.

In English, we have at least 3 choices:
* Emphasize the present relevance of the changed state. That's a perfect. Thus, "I've put on a hat."
* Emphasize an ongoing, in-progress state. That's progressive "I'm wearing a hat."
* Emphasize that the action of putting it on is over and in the past. Thus simple past "I put on my hat", which isn't normally called an aspect at all, but is roughly being used as a perfective. (I say roughly because it could be completive or punctual as well.)

Mandarin has at least 5 options:
* No marking. A teaching website has the sentence 她喜歡穿深色的衣服 "She likes wearing dark clothing." There's no aspect particle at all, just the verb 她 chuān.
* Use chuān as an active verb, 'put on'. This uses the particle 在 zài.
* Emphasize the result of this action verb as a continuing state, using 著 -zhe.
* Emphasize the change of state: 她穿了皮靴 "She put on leather boots."
* Emphasize the continuing relevance of the action (perfect): 她穿皮靴了. Same particle 了, different position.

In a language with an inceptive, you could use that ("I started wearing the hat").

And that's before getting into the really messy ways people can use a verb form. I'm learning some Biblical Hebrew right now, and there's no simple way to name the conjugations. Many grammars use perfective/imperfective, but that's not it at all. The suffixing conjugation (SC) is relatively easy: it's prototypically past tense and completive. E.g. kāṯaḇ is usually "he wrote." The prefixing conjugation (PC) is most easily described as not the SC. That could mean that it's past tense but imperfective; or present habitual; or future; or some sort of vague modal. Any name you give it (besides a morphological one like PC) will be misleading...

Li & Thompson take ten pages to explain the durative (在 and 著). If there's a takeway here, it's that a single example on Wikipedia can be misleading. Its 她 chuān example isn't very helpful for explaining how the language works— it doesn't even mention, for instance, the contrast with the other choices above. I don't think you could read that article and guess how 拿 ná works. But L&T's explanation makes the pattern clear:

ná + zài = take
ná + zhe = take and continue in that state, i.e. hold
chuān + zài = put on
chuān + zhe = put on and continue in that state, i.e. wear

This is not very much like English's distinction between e.g. "I wear a hat" and "I'm wearing a hat." Just naming some aspect isn't enough; you have to know what options the language gives you.

[aporaporimos]

And that's before getting into the really messy ways people can use a verb form. I'm learning some Biblical Hebrew right now, and there's no simple way to name the conjugations. Many grammars use perfective/imperfective, but that's not it at all. The suffixing conjugation (SC) is relatively easy: it's prototypically past tense and completive. E.g. kāṯaḇ is usually "he wrote." The prefixing conjugation (PC) is most easily described as not the SC. That could mean that it's past tense but imperfective; or present habitual; or future; or some sort of vague modal. Any name you give it (besides a morphological one like PC) will be misleading...

I'm learning Syriac, which has analogous/cognate suffixing and prefixing conjugations, also conventionally called perfect/imperfect... but the "imperfect" almost always seems to be future tense or else modal or irrealis—and there's a dedicated separate construction for a past imperfective, which uses a past tense copula + a participle, just like English. You could call the Syriac SC "past" and the PC "future" and it'd be close enough for government work.

[bradrn]

But in 3 of my 4 examples, those are the standard meanings! The only one that's maybe nonstandard is L&T's "durative".

I’m not sure I understand what you’re saying here… which 3 examples are you talking about? Because if you’re talking about ‘progressive’, ‘continuous’ and ‘resultative’, I don’t follow your argument:

• Progressive is… um… some sort of imperfective? [...]

OK, I'll put this down to not having explained clearly, and I'll try again.

My point is, if you look at a situation, or at a single verb, there is no God-given answer on what aspect is involved, or what name to give it. […]

Alright, let me see if I can summarize what you said:

• Many (possibly all) languages have a set of verbal aspects, used to focus temporally on certain parts of the verb.
• Each language divides up this ‘aspectual space’ differently (e.g. Mandarin has -zhe for durative, but English uses the perfect for the same thing).
• Some verbal forms can be used ‘messily’ (as you put it): they may represent multiple different aspects, tense/aspect/mood combinations etc.
• All this means that languages divide up aspects in such different ways that, even if you can find two aspects in two different languages which can take on the same prototypical meaning, giving them the same name will almost certainly be misleading in one way or another.

Does this summary seem accurate? (I’m asking because I want to make sure that we’re both interpreting your post the same way.) If so, then it makes a lot of sense to me: I hadn’t really thought about that last point before, but if it’s true (and it doesn’t seem obviously false to me), then there would be no way of assigning any sort of ‘universal’ name to aspects in a way which isn’t misleading.

On the other hand, none of this helps with my original question: what is the difference between the continuous and the progressive? (Although I don’t think you meant to address this question at all in your post.)

[zompist]

I'm learning Syriac, which has analogous/cognate suffixing and prefixing conjugations, also conventionally called perfect/imperfect... but the "imperfect" almost always seems to be future tense or else modal or irrealis—and there's a dedicated separate construction for a past imperfective, which uses a past tense copula + a participle, just like English. You could call the Syriac SC "past" and the PC "future" and it'd be close enough for government work.

Yeah, that's basically shared between Imperial Aramaic and post-Biblical Hebrew.

[zompist]

• Many (possibly all) languages have a set of verbal aspects, used to focus temporally on certain parts of the verb.
• Each language divides up this ‘aspectual space’ differently (e.g. Mandarin has -zhe for durative, but English uses the perfect for the same thing).
• Some verbal forms can be used ‘messily’ (as you put it): they may represent multiple different aspects, tense/aspect/mood combinations etc.
• All this means that languages divide up aspects in such different ways that, even if you can find two aspects in two different languages which can take on the same prototypical meaning, giving them the same name will almost certainly be misleading in one way or another.

Does this summary seem accurate?

Yes, except for the specific example-- the English and Mandarin perfects are pretty comparable. I've always found the perfect hardest to explain; one way is that it sets up a consequence: "I've put on my hat (so let's leave)" Or "I've eaten (so I'm not hungry right now)".

On the other hand, none of this helps with my original question: what is the difference between the continuous and the progressive? (Although I don’t think you meant to address this question at all in your post.)

As I said earlier, I don't think we've found one, unless someone can find one in a language besides Chinese.

[bradrn]

• Many (possibly all) languages have a set of verbal aspects, used to focus temporally on certain parts of the verb.
• Each language divides up this ‘aspectual space’ differently (e.g. Mandarin has -zhe for durative, but English uses the perfect for the same thing).
• Some verbal forms can be used ‘messily’ (as you put it): they may represent multiple different aspects, tense/aspect/mood combinations etc.
• All this means that languages divide up aspects in such different ways that, even if you can find two aspects in two different languages which can take on the same prototypical meaning, giving them the same name will almost certainly be misleading in one way or another.

Does this summary seem accurate?

Yes, except for the specific example-- the English and Mandarin perfects are pretty comparable. I've always found the perfect hardest to explain; one way is that it sets up a consequence: "I've put on my hat (so let's leave)" Or "I've eaten (so I'm not hungry right now)".

To me, it looks like the English perfect covers all the meanings of the Mandarin perfect + durative. (e.g. English ‘I’ve put on my clothes’ is nearly synonymous to ‘I’m wearing my clothes’, which is the Mandarin durative.)

I also find it pretty interesting that you find the perfect hard to explain: I’ve always felt that it’s the only aspect for which I have a good intuitive understanding. I would define it as specifying an action which occurred in the past (relative to the time of reference), but has consequences with continuing relevance at the time of reference.

On the other hand, none of this helps with my original question: what is the difference between the continuous and the progressive? (Although I don’t think you meant to address this question at all in your post.)

As I said earlier, I don't think we've found one, unless someone can find one in a language besides Chinese.

I had thought this was the case, but I wasn’t quite sure yet, since no-one had explicitly said ‘there doesn’t seem to be a difference’.

[akam chinjir]

On the other hand, none of this helps with my original question: what is the difference between the continuous and the progressive? (Although I don’t think you meant to address this question at all in your post.)

As I said earlier, I don't think we've found one, unless someone can find one in a language besides Chinese.

I had thought this was the case, but I wasn’t quite sure yet, since no-one had explicitly said ‘there doesn’t seem to be a difference’.

I may have missed something relevant, but I think I've seen people distinguish continuous and progressive by saying that if a stative verb can take a particular form, then it could be continuous, but couldn't be progressive.