Linguistic Miscellany Thread

[Linguoboy]

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.

In previous discussions, Mark's highlighted the weasel-word quality of "relevance". After all, if an action isn't "relevant" at the time of speaking, why are you even mentioning it? I can't tell you how many times someone in my presence has made a personal statement in the present perfect and my response (spoken or unspoken) has been "Good for you!" because I can't for the life of me work out why they thought that particular statement was "relevant" to the current discourse.

(Moreover, any brief explanation of the perfect trips over the fact that millions of native American English speakers apparently disagree about the "continuing relevance" of most of these actions because they happily use simple past in contexts where it jars Commonwealth speakers, e.g. in sentences with "yet".)

[Kuchigakatai]

The English present perfect, like all TAMs in European and non-European TAM systems, is best explained as a list of particular uses anyway. The explanation "event in the past that's still relevant" is just the unifying metaphor of the various uses, but there's more to the story than that.

Spanish is also fun to talk about regarding the perfect, because there you get a range of dialects from Spain (where the perfect is very heavily used) to Argentina (where it is rarely used).

[Linguoboy]

Spanish is also fun to talk about regarding the perfect, because there you get a range of dialects from Spain (where the perfect is very heavily used) to Argentina (where it is rarely used).

Even within Spain, there's a tremendous range. The situation in most of Latin America reflects what you find in the Spanish of northwestern Spain and the Canaries (where it doubtless originated); I tend to follow Catalonian usage myself, since I find trying to keep one set of rules in my head for Catalan and a different set for Castilian just too confusing.

I wonder how closely the Argentine usage mirrors that of the southern Italian dialects.

[bradrn]

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.

In previous discussions, Mark's highlighted the weasel-word quality of "relevance". After all, if an action isn't "relevant" at the time of speaking, why are you even mentioning it? I can't tell you how many times someone in my presence has made a personal statement in the present perfect and my response (spoken or unspoken) has been "Good for you!" because I can't for the life of me work out why they thought that particular statement was "relevant" to the current discourse.

Oh, that’s a good point. I wonder how come I never noticed that before? Now I’m confused about the perfect again…

The English present perfect, like all TAMs in European and non-European TAM systems, is best explained as a list of particular uses anyway. The explanation "event in the past that's still relevant" is just the unifying metaphor of the various uses, but there's more to the story than that.

Well, of course. I would take this as a given for any aspect (although I do occasionally forget, so thanks for the reminder!).

[bradrn]

Does anyone know of any good resources on syntactic ergativity, as opposed to morphological ergativity? I want to write a section about it as the next post in my ergativity series, but none of my sources (mainly Dixon, McGregor, Oxford Handbook of Ergativity) seem to have much information about it. (It would be nice to get something not too theoretical, although given the state of most syntactic research today, I fear that this may be impossible.)

[Zju]

It just so happens that I read an article about ergativity today and it has some points on syntax, but from what I could gather it seems more like a (novel?) attempt to describe split-ergativity rather than a study of syntactic ergativity per se.

[akam chinjir]

bradrn actually talked about that account of split ergativity in his latest post

[bradrn]

It just so happens that I read an article about ergativity today and it has some points on syntax, but from what I could gather it seems more like a (novel?) attempt to describe split-ergativity rather than a study of syntactic ergativity per se.

As akam chinjir points out, I’ve already read that one. (Actually, akam chinjir was the one who recommended it to me.) Anyway, it’s not really what I’m looking for: that one doesn’t cover ergativity in syntax, but simply tries to explain an ergativity split (which is morphology rather than syntax) using syntactic techniques.

bradrn actually talked about that account of split ergativity in his latest post

Actually, if you look closely, I didn’t. It looks like that article (well, the bit on aspectual splits, at least) is a condensed version of Coon’s book Aspects of Split Ergativity, which is much more detailed, so I used that as my source instead. But yes, that article is where I started from.

(I know that this is a nitpick, but I couldn’t resist…)

[Qwynegold]

Thanks for all the answers to the question I asked a while ago!

As far as "the firewood was made to be chopped by me" goes, I agree it's totally grammatical to me, but not with the semantics you're looking at (< ~"Someone caused me to chop the firewood"). Rather, it clearly means "This firewood came into existence for the sole purpose of getting chopped up by me! I am soooo into chopping this wood!!!"

I can see how it could have the meaning you are talking about, but I think one can easily understand that that's not what's meant, especially if you read the sentence in its context. However, I'm concerned about that "by me" is ambiguous as to whether that's a causee or a causer. Hmm, I'll try to think of some kind of little explanation I could add in parenthesis.

[akam chinjir]

Does anyone know of any good resources on syntactic ergativity, as opposed to morphological ergativity? I want to write a section about it as the next post in my ergativity series, but none of my sources (mainly Dixon, McGregor, Oxford Handbook of Ergativity) seem to have much information about it. (It would be nice to get something not too theoretical, although given the state of most syntactic research today, I fear that this may be impossible.)

Have you looked at Manning, Ergativity?

[bradrn]

Does anyone know of any good resources on syntactic ergativity, as opposed to morphological ergativity? I want to write a section about it as the next post in my ergativity series, but none of my sources (mainly Dixon, McGregor, Oxford Handbook of Ergativity) seem to have much information about it. (It would be nice to get something not too theoretical, although given the state of most syntactic research today, I fear that this may be impossible.)

Have you looked at Manning, Ergativity?

Nope, haven’t seen it before. But this looks pretty much like what I was hoping for — thanks for the recommendation!

[fusijui]

Thanks for all the answers to the question I asked a while ago!

As far as "the firewood was made to be chopped by me" goes, I agree it's totally grammatical to me, but not with the semantics you're looking at (< ~"Someone caused me to chop the firewood"). Rather, it clearly means "This firewood came into existence for the sole purpose of getting chopped up by me! I am soooo into chopping this wood!!!"

I can see how it could have the meaning you are talking about, but I think one can easily understand that that's not what's meant, especially if you read the sentence in its context. However, I'm concerned about that "by me" is ambiguous as to whether that's a causee or a causer. Hmm, I'll try to think of some kind of little explanation I could add in parenthesis.

Sure, in context I'd probably extract your intended meaning — after stopping and going back to re-parse it, as it's such awkward and unnatural phrasing, not even remotely fluent native speech. With enough context, really, who needs morphosyntax, after all?

[Kuchigakatai]

Basilicata Neapolitan has a word for "October" that was borrowed from Oscan: attrufu, with intervocalic -f-.

[Qwynegold]

Sure, in context I'd probably extract your intended meaning — after stopping and going back to re-parse it, as it's such awkward and unnatural phrasing, not even remotely fluent native speech. With enough context, really, who needs morphosyntax, after all?

Lol and sigh. >_<

[Kuchigakatai]

I was just remembering that back in 2010 I used to see Taiwanese young people using the zhuyin markers for the 4th and 2nd tones together as an angry smiley.

Like: ˋˊ

I feel that should get an award for "tiniest smiley" or something.

I asked someone from Taiwan about it and I was told it's basically not used these days anymore. It has fallen out of fashion.

[Kuchigakatai]

The Japanese honorific suffix -sama comes from sa '(demonstrative)' + ma 'likeness, way, manner'. Sama was formerly an honorific pronoun.

The Japanese adopted the Chinese custom of using 貴 'valued, expensive' as an honorific 2nd-person prefix, and created 貴様 kisama ("valued that-way"). Although it was originally a very polite term of address, the Japanese pronoun euphemism treadmill has been such that now it is mostly used as an insult between men...

Basically: "your valued likeness" > "you motherfucker/asshole". Euphemism treadmills are all too real.

Does anyone know of any good resources on syntactic ergativity, as opposed to morphological ergativity? I want to write a section about it as the next post in my ergativity series, but none of my sources (mainly Dixon, McGregor, Oxford Handbook of Ergativity) seem to have much information about it. (It would be nice to get something not too theoretical, although given the state of most syntactic research today, I fear that this may be impossible.)

I think this is one of those parts where you'd benefit a lot from actually studying (learning) an ergative language. If you knew an ergative language or two, you'd be able to give examples of how the morphological cases or markers are used (and could be used) in syntactic transformations. Languages typically combine morphology and syntax into one thing anyway, so that a typical "syntactic" transformation is in reality a mixture of selecting inflections, rearranging words in a different order, and adding or removing function words (auxiliary verbs, prepositions, etc.) to express a different meaning.

Imagine you and me were writing a thread explaining case for people who're uncomfortable with case. One thing that could be helpful would be mentioning that languages with case can perfectly have more than one way of expressing something, one with a bare case (even if it's ambiguous) and one with unambiguous adposition + totally redundant case. For example, the Latin bare ablative can sometimes be used to express 'in X' in some contexts usually handled with preposition + ablative noun: in locō pōnere 'to put [sth] in its place' can also be locō pōnere. The bare dative is also sometimes used to express 'to/towards [a place]', even though that's usually handled with preposition + accusative noun. These little syntactic variants are probably things you wouldn't find in a more general typological discussion on case, but if you study Latin you know some of them...

(Old French has some similar examples, which are all the more fascinating because case is a lot more ambiguous in that language, since it only retained two, nominative and accusative, and in very phonetically decayed forms at that. The Old French accusative could still be used in bare form to regularly mark possessors, and occasionally also indirect objects and instrumental adjuncts...)

(This ties in with the post I said I would write in the SAE features thread, I know... Apologies to akam chinjir, but Richard W sent me down an interesting rabbit hole with his counter-example about Latin.)

[bradrn]

Does anyone know of any good resources on syntactic ergativity, as opposed to morphological ergativity? I want to write a section about it as the next post in my ergativity series, but none of my sources (mainly Dixon, McGregor, Oxford Handbook of Ergativity) seem to have much information about it. (It would be nice to get something not too theoretical, although given the state of most syntactic research today, I fear that this may be impossible.)

I think this is one of those parts where you'd benefit a lot from actually studying (learning) an ergative language. Part of what made Whimemsz's old thread on polysynthesis great was that he actually studied Ojibwe and had some general interest in the Algic family. If you knew an ergative language or two, you'd be able to give examples of how the morphological cases or markers are used (and could be used) in syntactic transformations. Languages typically combine morphology and syntax into one thing anyway, so that a typical "syntactic" transformation is in reality a mixture of selecting inflections, rearranging words in a different order, and adding or removing function words (auxiliary verbs, prepositions, etc.) to express a different meaning.

You’re right — learning a language is of course the ‘best’ way to learn about the linguistic features it uses. Unfortunately, there’s two problems with that with regards to syntactic ergativity:

• Firstly, one of the biggest problems with syntactic ergativity is its definition — there seems to be no consensus about exactly which features it includes. Some people say that it includes a whole variety of features; some people say that it only encompasses Aʼ-movement (whatever that is); Dixon seems to believe that it only includes coordination. There is similar debate over which languages are syntactically ergative: some people say that Austronesian languages are syntactically ergative, while others have apparently said that the only syntactic ergative language is Dyirbal. Reading a language grammar isn’t going to help me with any of these problems. (You might wonder why I need to know such definitional details in order to write about the topic, but I can’t write about syntactic ergativity without first knowing what it is!)
• Secondly (and more importantly), learning a language — or even reading through its grammar and understanding every aspect of its syntax — takes time and effort. I already have university (now online and with more assignments!), research about syntactic ergativity, and my own personal projects slowing me down; if I need to learn a whole other language I’ll never get around to writing the next post!

Imagine you and me were writing a thread explaining case for people who're uncomfortable with case. One thing that could be helpful would be mentioning that languages with case can perfectly have more than one way of expressing something, one with a bare case (even if it's ambiguous) and one with unambiguous adposition + totally redundant case. For example, the Latin bare ablative can sometimes be used to express 'in X' in some contexts usually handled with preposition + ablative noun: in locō pōnere 'to put [sth] in its place' can also be locō pōnere. The bare dative is also sometimes used to express 'to/towards [a place]', even though that's usually handled with preposition + accusative noun. These little syntactic variants are probably things you wouldn't find in a more general typological discussion on case, but if you study Latin you know some of them...

(Old French has some similar examples, which are all the more fascinating because case is a lot more ambiguous in that language, since it only retained two, nominative and accusative, and in very phonetically decayed forms at that. The Old French accusative could still be used in bare form to regularly mark possessors, and occasionally also indirect objects and instrumental adjuncts...)

That’s pretty interesting — I never thought about that. Which I suppose confirms what you’re saying.

(On the other hand, this is the sort of thing I personally would like to see more of in typological discussions, and which I’m trying to do in my ergativity series: discussions of things which may only occur in a couple of languages, but are still pretty interesting. I wish more typological discussions included this sort of stuff.)

[Kuchigakatai]

[*] Firstly, one of the biggest problems with syntactic ergativity is its definition — there seems to be no consensus about exactly which features it includes. Some people say that it includes a whole variety of features; some people say that it only encompasses Aʼ-movement (whatever that is); Dixon seems to believe that it only includes coordination. There is similar debate over which languages are syntactically ergative: some people say that Austronesian languages are syntactically ergative, while others have apparently said that the only syntactic ergative language is Dyirbal. Reading a language grammar isn’t going to help me with any of these problems. (You might wonder why I need to know such definitional details in order to write about the topic, but I can’t write about syntactic ergativity without first knowing what it is!)[/list]

Oh, I guess those researchers are talking about pure syntactic ergativity? Like, ergativity marked only with syntax, with no accompanying inflectional morphology (or case-marking adpositions/particles, I guess). Which would explain why Dixon says it must involve coordination. What I had in mind was more discussion about the interaction of (morphological) ergativity and syntax.

• Secondly (and more importantly), learning a language — or even reading through its grammar and understanding every aspect of its syntax — takes time and effort. I already have university (now online and with more assignments!), research about syntactic ergativity, and my own personal projects slowing me down; if I need to learn a whole other language I’ll never get around to writing the next post!

Well, yeah... But ya know, you can always come back to it in the future with a richer view of the thing to talk about those things.

That’s pretty interesting — I never thought about that. Which I suppose confirms what you’re saying.

(On the other hand, this is the sort of thing I personally would like to see more of in typological discussions, and which I’m trying to do in my ergativity series: discussions of things which may only occur in a couple of languages, but are still pretty interesting. I wish more typological discussions included this sort of stuff.)

Yeah... Hallow XIII likes to often say that this is a downside of typology for conlangers. Typology tries to create "grids" that can model linguistic diversity adequately "enough" (as in this thing of morphosyntactic alignments accounting for different combinations/alignments of S vs. A/O markings), which necessarily forces languages to be shelved within stereotypes, but a lot of conlangers are in practice more interested in quirkiness...

Under the WALS's strict definitions, Standard Arabic would qualify as using the "gap" strategy in feature 122A (Relativization on Subjects), as opposed to the "pronoun-retention", "non-reduction" and "relative pronoun" strategies. But Standard Arabic is a bit more interesting than that, using a subordinator adjective that agrees in case (besides gender and number) with the head noun (this is very un-European!) if the head noun is definite (otherwise, bare juxtaposition (with no subordinator) is used if the head noun is indefinite). From Ryding's grammar (page 323):

li=z-zaudʒ-aini l:að-aini jantaðˤira:ni ħadaθ-an saʕi:d-an
to the-spouse-DUAL.GEN REL.MASC-DUAL.GEN await.PRES.3DUAL event-ACC happy-MASC.ACC
'to the couple that is waiting for a happy event'

Here, the 'two spouses' are the subject of jantaðˤira:ni 'they (both) await sth'. In stereotypical European grammar, a relative pronoun would appear here in the nominative case (les époux qui attendent..., versus oblique que), but in Arabic the relative adjective al:aðaini is in the genitive in order to agree with the head noun zaudʒaini 'two spouses', which is in the genitive because of the preposition li 'to'.

Saying Standard Arabic uses the gap strategy would be perfectly fine really, but something good is lost in the process...

[bradrn]

• Firstly, one of the biggest problems with syntactic ergativity is its definition — there seems to be no consensus about exactly which features it includes. Some people say that it includes a whole variety of features; some people say that it only encompasses Aʼ-movement (whatever that is); Dixon seems to believe that it only includes coordination. There is similar debate over which languages are syntactically ergative: some people say that Austronesian languages are syntactically ergative, while others have apparently said that the only syntactic ergative language is Dyirbal. Reading a language grammar isn’t going to help me with any of these problems. (You might wonder why I need to know such definitional details in order to write about the topic, but I can’t write about syntactic ergativity without first knowing what it is!)

Oh, I guess those researchers are talking about pure syntactic ergativity? Like, ergativity marked only with syntax, with no accompanying inflectional morphology (or case-marking adpositions/particles, I guess). Which would explain why Dixon says it must involve coordination. What I had in mind was more discussion about the interaction of (morphological) ergativity and syntax.

No — syntactic ergativity is about syntactic processes which identify S with O. That much everyone can agree with. (I don’t believe there is any language which marks ergativity purely with word order, and in fact Dixon explicitly cautions against coming to that conclusion.) The problem is that — as with most things in syntax — everything’s a bit blurry. Some syntactic processes universally identify S with A (e.g. imperatives and reflexives), so some people have used this as evidence that ‘subject’ is universal and there is no such thing as a syntactically ergative language. But there are also some processes (albeit less clearly syntactic ones) which universally identify S with O (e.g. when a verb has multiple senses, the choice of meaning can depend on S or O but not A). And, depending on the language, there can easily be other processes which identify S with A or O — e.g. in the ergativity thread Xwtek pointed out that English nominalisations of the form ‘Troy’s destruction by the Greeks’ can be seen as identifying S with O. So the trick is to find a set of constructions which always identify S with A in morphologically accusative languages, and which can only identify S with O in morphologically ergative languages — and then those constructions are the ones which mark syntactic ergativity. Except that even here there’s disagreement: Dixon seems to believe that coordination is the only marker of syntactic ergativity, while the article akam chinjir linked classifies Tagalog as a syntactically ergative language (which seems impossible if you take the definition I proposed above). One paper comes to the conclusion that Aʼ-movement constructions are the ones which indicate syntactic ergativity, which is a very nice generalisation that’s useless to me because (a) I have no idea what that means and (b) the rest of the paper uses some complicated Minimalist approach I don’t understand.

• Secondly (and more importantly), learning a language — or even reading through its grammar and understanding every aspect of its syntax — takes time and effort. I already have university (now online and with more assignments!), research about syntactic ergativity, and my own personal projects slowing me down; if I need to learn a whole other language I’ll never get around to writing the next post!

Well, yeah... But ya know, you can always come back to it in the future with a richer view of the thing to talk about those things.

Well, of course! If I learn more, I probably will come back to it.

That’s pretty interesting — I never thought about that. Which I suppose confirms what you’re saying.

(On the other hand, this is the sort of thing I personally would like to see more of in typological discussions, and which I’m trying to do in my ergativity series: discussions of things which may only occur in a couple of languages, but are still pretty interesting. I wish more typological discussions included this sort of stuff.)

Yeah... Hallow XIII likes to often say that this is a downside of typology for conlangers. Typology tries to create "grids" that can model linguistic diversity adequately "enough" (as in this thing of morphosyntactic alignments accounting for different combinations/alignments of S vs. A/O markings), which necessarily forces languages to be shelved within stereotypes, but a lot of conlangers are in practice more interested in quirkiness...

Yes, I also really dislike this aspect of typology. (Although I will admit that occasionally it seems to work out well.) I suppose that what I like more is simply surveys of a topic which cover a variety of different languages (e.g. Dixon’s Ergativity, Palmer’s Mood and Modality), without being ‘typological’ per se by putting languages in lots of little boxes.

[Pabappa]

I was surprised to learn just now that Koyukon, featured in the children's show Molly of Denali, uses ee oo for /i: u:/. I had thought that the kids' show was simply using a nonstandard spelling to make it easier for kids to pick up, but it is in fact the proper orthography.