Linguistic Miscellany Thread

[Richard W]

And what about

• I was him in that play
• I was tired and
• If that be true...

?

What about them? Copulae are often weird with regards to alignment. Specifically, in English:

• Unlike all other verbs, the copula exhibits agreement with the number of the subject (counting you as a plural), as well as a special form for first person singular.
• Case-marking becomes a bit wonky as well: the subject is consistently in the nominative, but the object can be either nominative or accusative depending on how learnèd you want to be. (IIRC Dixon has noted that the copula taking a nominative subject and accusative object is a characteristic of marked-nominative languages.)

Alternatively, English has accusative verbal alignment, end of story.

[zompist]

Verbal agreement[/b] is accusative for 3SG, direct/absent for everything else (which interestingly violates the animacy hierarchy!)

Not quite. We and peas show direct/absent verbal agreement, while water shows accusative agreement:

We drink water
Water drink-s us
Peas drink us

I see how you got there, but this strikes me as an odd analysis. It amounts to saying that there are no null markers, or that the lack of an ending means that the distinction in question is not expressed. Of course it's expressed! It's expressed by the lack of ending.

A clearer example would be Russian академия наук "academy(s.nom) of sciences(pl.gen)". наук has no ending; this (pl.gen) is the only such form in the declension, so if you see a "naked" наук you know it's pl.gen. Compare академия дураков "academy of fools", where дурак 'fool' has an explicit ending. (The difference is gender.)

Surely it would be strange to say that Russian marks number and plural for the word наукa except when the syntax demands pl.gen, and in those situations it doesn't mark number and plural at all. It does, наук really is the plural genitive.

Now, what about Latin adulter 'adulterer'? As it happens, every case form but s.nom* adds something to this. Do you say that Latin adulter "doesn't mark case for nominatives"? Again, that seems silly— if you see adulter, you know it's s.nom; adulter is not unmarked, it simply is the s.nom.

Null marking is a marking, at least when it's unambiguous, as in these two cases.

* I'm leaving out treatment of the vocative here.

Things get more complicated when you have case syncretism. E.g. Latin animal could be s.nom or s.acc. I do think it would be strange to say that the word doesn't mark case and number. It certainly does— we know it's not abl/gen/dat and that it's singular. It just sloppily doesn't distinguish nom/acc.

So, I think it would be more accurate to say of English that shows accusative verbal agreement in the 3rd person. "He drinks/ They drink." Using "drink" with a 3s/3p subject automatically gives you the number, even if the noun obscures it ("The deer drink/drinks"). The null marker is how the 3p agreement is expressed, and it can be easily confirmed that the trigger is the subject, not the object. ("He sees the dog/dogs").

I purposely left out 1st/2nd person, because here I think it's arguable and even correct that indeed there is no agreement. (Except for "be", but here the concept of overdetermination comes to the rescue. I'm fine with saying that "be" works differently. That's more or less how I'd handle the Latin vocative, too.)

I understand that the idea of "null marking" may itself seem odd. But its less odd than the positions forced on you by the above examples if you reject it.

[bradrn]

Alternatively, English has accusative verbal alignment, end of story.

…except that’s not ‘end of story’ at all, because, for instance, the following sentences don’t demonstrate any sort of overt accusativity at all, except in word order:

People like paintings
Paintings please people

I see how you got there, but this strikes me as an odd analysis. It amounts to saying that there are no null markers, or that the lack of an ending means that the distinction in question is not expressed. Of course it's expressed! It's expressed by the lack of ending. … Null marking is a marking, at least when it's unambiguous, as in these two cases.

In these cases, of course I agree! Null marking is just as much of a marking as anything else.

Things get more complicated when you have case syncretism. E.g. Latin animal could be s.nom or s.acc. I do think it would be strange to say that the word doesn't mark case and number. It certainly does— we know it's not abl/gen/dat and that it's singular. It just sloppily doesn't distinguish nom/acc.

Saying this word ‘doesn't mark case and number’ would be stupid — it evidently does distinguish them. It’s just that, instead of nom/acc/abl/gen/dat, it distinguishes direct/abl/gen/dat, with direct case marked by an absence of ending. And this is in fact a good analysis: it means we can divide Latin nouns into those with a nominative alignment (masculine and feminine), and those with a direct alignment (neuter), which is exactly in line with the animacy hierarchy.

So, I think it would be more accurate to say of English that shows accusative verbal agreement in the 3rd person. "He drinks/ They drink." Using "drink" with a 3s/3p subject automatically gives you the number, even if the noun obscures it ("The deer drink/drinks"). The null marker is how the 3p agreement is expressed, and it can be easily confirmed that the trigger is the subject, not the object. ("He sees the dog/dogs").

I purposely left out 1st/2nd person, because here I think it's arguable and even correct that indeed there is no agreement. (Except for "be", but here the concept of overdetermination comes to the rescue. I'm fine with saying that "be" works differently. That's more or less how I'd handle the Latin vocative, too.)

Yes, this is an analysis I can agree with — no agreement for 1/2, and agreement for 3, where third person plural is unmarked.

[Richard W]

Alternatively, English has accusative verbal alignment, end of story.

…except that’s not ‘end of story’ at all, because, for instance, the following sentences don’t demonstrate any sort of overt accusativity at all, except in word order:

People like paintings
Paintings please people

Note the suppression of the present indicative ending -s. It and its analogues (cognates?) serve as TAM markers in Danish and some English dialects, as in You pays your money and you takes your choice.

[zompist]

And this is in fact a good analysis: it means we can divide Latin nouns into those with a nominative alignment (masculine and feminine), and those with a direct alignment (neuter), which is exactly in line with the animacy hierarchy.

Just as a minor point, this is not really a good analysis of Latin, because Latin gender is much, much more lexical than semantic.

[bradrn]

Alternatively, English has accusative verbal alignment, end of story.

…except that’s not ‘end of story’ at all, because, for instance, the following sentences don’t demonstrate any sort of overt accusativity at all, except in word order:

People like paintings
Paintings please people

Note the suppression of the present indicative ending -s. It and its analogues (cognates?) serve as TAM markers in Danish and some English dialects, as in You pays your money and you takes your choice.

I’m not quite sure how this is relevant. What point are you trying to make here?

And this is in fact a good analysis: it means we can divide Latin nouns into those with a nominative alignment (masculine and feminine), and those with a direct alignment (neuter), which is exactly in line with the animacy hierarchy.

Just as a minor point, this is not really a good analysis of Latin, because Latin gender is much, much more lexical than semantic.

Is it not at least true that more animate nouns, particularly human nouns, tend to be masculine or feminine? Even a small trend in this direction would place masculine and feminine to the left of neuter on the animacy hierarchy.

[Kuchigakatai]

I’m not sure whether my use is standard or not. On the other hand, I certainly find it a lot clearer than simply saying ‘English nouns use accusative marking’, since that lumps together all the different ways in which accusative marking can occur. (You wouldn’t be able to get away with saying this in a language like Komnzo.)

I think it's unusual / non-standard, although it's also more precise and probably more adequate. And it looks like right now you're confusing Richard and zompist too, due to that. When Richard says "English has accusative verbal alignment, end of story", he's referring to the common definition. When zompist says "this is not really a good analysis of Latin", he misses that you're purely talking about the morphological angle, in terms of just declension, not the morphosyntactic angle that involves distinguishing function roles. Saying masculine/feminine nouns have a morphological nom.-acc. alignment, unlike neuter nouns, is pretty unusual, I'd say.

By the way, why would my usage be (even) worse for Komnzo than for English? Not know anything of Komnzo myself... The Wikipedia article says nothing relevant too.

There’s also the fact that VO is often analysed as forming a constituent. Except that Movima’s the opposite — if anything, it looks like the verb and the transitive subject form a constituent, with O/S being ‘added on’ to that core bit. It’s certainly unusual in the scheme of things.

Ah. Well, that is one weird analysis, if it is true of Movima. I mean, subject-verb compounds rarely occur with animate subjects for good reasons (and even when they do it's either animals, or humans in some very involuntary notions such as 'to die'). I would've thought the limitation is due to subjects being too peripheral, not due to being too close to the verb. Hmm, weird.

Given the use of "direct" in direct/inverse, it would be better to refer to "no distinctions" as "neutral" marking.

That is one excellent and amusing point!

[Richard W]

Note the suppression of the present indicative ending -s. It and its analogues (cognates?) serve as TAM markers in Danish and some English dialects, as in You pays your money and you takes your choice.

I’m not quite sure how this is relevant. What point are you trying to make here?

You're saying that accusative alignment occurs with 3rd singular subjects, and only with them, because they cause the third singular ending -s to appear in the most important tense and mood combination. I'm pointing out that one could reasonably argue that non-third-singular subjects remove the -s TAM marker, so that it is in those instances that one has accusative alignment.

[Estav]

And this is in fact a good analysis: it means we can divide Latin nouns into those with a nominative alignment (masculine and feminine), and those with a direct alignment (neuter), which is exactly in line with the animacy hierarchy.

Just as a minor point, this is not really a good analysis of Latin, because Latin gender is much, much more lexical than semantic.

Is it not at least true that more animate nouns, particularly human nouns, tend to be masculine or feminine? Even a small trend in this direction would place masculine and feminine to the left of neuter on the animacy hierarchy.

There is a definite semantic component to gender in Latin (there is supposed to be at least some semantic component to gender in all languages).

The core connections between semantics and grammatical gender show up in that that proper nouns referring to male/female beings take masculine/feminine gender agreement on the basis of their semantic reference, very few nouns that specifically refer to animate beings are neuter (animal is an obvious exception, but as a general rule, it's stronger than just a minor trend), and nouns referring to animate beings usually match the semantic gender in cases where gender is salient (cases where they do not match include many animal names when not referring to an animal of a specific gender, and a small set of nouns such as persona, victima where the reference to a person appears to be transferred from another meaning).

The basis for saying lexical assignment is "much more" important than semantic would have to be the gender assignment of nouns referring to inanimates, which are very frequently masculine or feminine (their gender is usually not a purely arbitrary lexical fact, but the explanations/reasons are often apparently unrelated to their level of animacy, instead being more clearly related to the morphology or phonology of the word's ending).

The Romans themselves, as far as I understand, considered neuter words to have a nominative and accusative case form that were the same (syncretism) rather than having a single case form that was neither nominative nor accusative, but served the function of both. However, there are absolutely no exceptions to this syncretism in the neuter gender. In any case, though, this feature of the case marking of neuter nouns is supposed to go back to the PIE animate/inanimate distinction, so you don't need to treat animacy as a major part of Latin gender in order to have it be the ultimate explanation of the distribution of forms in Latin.

[zompist]

The basis for saying lexical assignment is "much more" important than semantic would have to be the gender assignment of nouns referring to inanimates, which are very frequently masculine or feminine (their gender is usually not a purely arbitrary lexical fact, but the explanations/reasons are often apparently unrelated to their level of animacy, instead being more clearly related to the morphology or phonology of the word's ending).

Indo-European is virtually the poster child for lexical gender. Probably the majority of words have entirely arbitrary gender, something every English speaker learning French, German, or Russian has rued. (My favorite example is French la verge (cock) and le vagin, the last despite Latin vagina being feminine).

There is a semantic gender system shoved on top, yes. But the system is far less semantic than (say) the Algonquian family or Dyirbal. (In those, there are quirky exceptions, but they are due to underlying or metaphorical gender, so there are rules you can follow.) And there are purely semantic gender systems like Tamil.

[bradrn]

I’m not sure whether my use is standard or not. On the other hand, I certainly find it a lot clearer than simply saying ‘English nouns use accusative marking’, since that lumps together all the different ways in which accusative marking can occur. (You wouldn’t be able to get away with saying this in a language like Komnzo.)

I think it's unusual / non-standard, although it's also more precise and probably more adequate. And it looks like right now you're confusing Richard and zompist too, due to that. When Richard says "English has accusative verbal alignment, end of story", he's referring to the common definition. When zompist says "this is not really a good analysis of Latin", he misses that you're purely talking about the morphological angle, in terms of just declension, not the morphosyntactic angle that involves distinguishing function roles. Saying masculine/feminine nouns have a morphological nom.-acc. alignment, unlike neuter nouns, is pretty unusual, I'd say.

Yep, I think you’re correct in saying this; my use is indeed a bit different to normal, and as you say that’s most probably what’s been confusing.

By the way, why would my usage be (even) worse for Komnzo than for English? Not know anything of Komnzo myself... The Wikipedia article says nothing relevant too.

Oh, I just chose Komnzo as an example of a language with different alignments in different parts of its grammar — you could say the same thing about any language like that. (Jacaltec and Warlpiri also come to mind, though other examples are numerous.)

(BTW the Komnzo Grammar is really good and open-access, so do take a look! It’s a fascinating language.)

There’s also the fact that VO is often analysed as forming a constituent. Except that Movima’s the opposite — if anything, it looks like the verb and the transitive subject form a constituent, with O/S being ‘added on’ to that core bit. It’s certainly unusual in the scheme of things.

Ah. Well, that is one weird analysis, if it is true of Movima. I mean, subject-verb compounds rarely occur with animate subjects for good reasons (and even when they do it's either animals, or humans in some very involuntary notions such as 'to die'). I would've thought the limitation is due to subjects being too peripheral, not due to being too close to the verb. Hmm, weird.

Movima definitely is one of the weirder languages I’ve come across. Here’s what Haude (2010) has to say about it:

Transitive clauses are identified by the fact that they may take two core arguments, i.e. two DPs or pronouns that are not marked as oblique. In a canonical transitive clause, both arguments follow the predicate … Apart from linear order, several other factors distinguish the two arguments of a transitive clause. The argument that comes first after the predicate [i.e. the more animate one] is obligatorily expressed and phonologically closely attached to the predicate through “internal cliticization” … The argument in second position, in contrast, has the same properties as the argument of an intransitive clause: its overt realization is not grammatically obligatory; it has a freer position in the clause; when represented by a pronominal enclitic, it is attached through “external cliticization”.

I’m not entirely sure V+A is actually a constituent in the strictest syntactic sense of the word, but there’s certainly some sort of close bond between the two.

Given the use of "direct" in direct/inverse, it would be better to refer to "no distinctions" as "neutral" marking.

That is one excellent and amusing point!

I also prefer ‘neutral’ to ‘direct’ as a name for this alignment; I’ve been using ‘direct’ here purely for consistency with everyone else’s terminology.

Note the suppression of the present indicative ending -s. It and its analogues (cognates?) serve as TAM markers in Danish and some English dialects, as in You pays your money and you takes your choice.

I’m not quite sure how this is relevant. What point are you trying to make here?

You're saying that accusative alignment occurs with 3rd singular subjects, and only with them, because they cause the third singular ending -s to appear in the most important tense and mood combination. I'm pointing out that one could reasonably argue that non-third-singular subjects remove the -s TAM marker, so that it is in those instances that one has accusative alignment.

But still: in what way exactly is it accusative with non-3s subjects? Let’s have a look at this paradigm, in detail:

	S	A	O
1s	-∅	-∅	-∅
2s	-∅	-∅	-∅
3s	-s	-s	-∅
1p	-∅	-∅	-∅
2p	-∅	-∅	-∅
3p	-∅	-∅	-∅

No matter whether you analyse it as absence of marking or null marking or removing -s or whatever, one thing is clear: S and A and O are all marked the same way for everything except 3s, which has S/A marked one way and O marked another. How is this not an instance of accusative alignment with 3s and direct alignment everywhere else?

The core connections between semantics and grammatical gender show up in that that proper nouns referring to male/female beings take masculine/feminine gender agreement on the basis of their semantic reference, very few nouns that specifically refer to animate beings are neuter (animal is an obvious exception, but as a general rule, it's stronger than just a minor trend), and nouns referring to animate beings usually match the semantic gender in cases where gender is salient (cases where they do not match include many animal names when not referring to an animal of a specific gender, and a small set of nouns such as persona, victima where the reference to a person appears to be transferred from another meaning).

This is pretty much what I was talking about: sure, lots of inanimates have ended up as masculine or feminine, but nearly every neuter noun is inanimate, meaning that the neuter gender as a whole is to the right of masculine or feminine on the animacy hierarchy. It doesn’t have to be wholly semantic for that to be true — just semantic enough, which the Latin system is.

[KathTheDragon]

How is this not an instance of accusative alignment with 3s and direct alignment everywhere else?

It is, at least in terms of person indexing. Argument flagging and syntax present two different pictures, a vestigial pronoun/noun split and consistent accusative alignment respectively, and all three are equally valid pictures, none overruling the others. You just need to be specific in which of the three you're talking about.

[bradrn]

How is this not an instance of accusative alignment with 3s and direct alignment everywhere else?

It is, at least in terms of person indexing. Argument flagging and syntax present two different pictures, a vestigial pronoun/noun split and consistent accusative alignment respectively, and all three are equally valid pictures, none overruling the others. You just need to be specific in which of the three you're talking about.

Thank you! This is exactly what I’ve been saying all along. I was responding in the quote to Richard W, who seems to have a different opinion.

[linguistcat]

I'm trying to learn some basic Mandarin right now. I've been making better progress than in past attempts, especially with tones (I tried in middle school and then again on my own after college), but I've found I tend to remember phrases instead of single words, especially when it comes to tones. I can repeat tones on their own but they don't seem to stick to the words in my brain without the context of a phrase or sentence. I'm sure it has to do with the fact I'm not a tonal language native, but have had a lot of experience with music from a young age.

Has anyone experienced similar while learning a tonal language? Or have tips to learn vocabulary as distinct words WITH the tones? Should I continue learning phrases/sentences the way I have, as almost mini-songs?

[Richard W]

The accusative alignment shown in I hate this argument has got nothing to do with one of the arguments being 3s. Note that the trigger for different form in the verb is the form of the subject.

[quinterbeck]

Has anyone experienced similar while learning a tonal language? Or have tips to learn vocabulary as distinct words WITH the tones? Should I continue learning phrases/sentences the way I have, as almost mini-songs?

I had the exact same experience learning Thai. Phrases even of two words were much easier to memorise correctly than individual words. (However I can't say I have any tips, since I only had two weeks of lessons and was only in Thailand for two months overall)

[zompist]

I'm trying to learn some basic Mandarin right now. I've been making better progress than in past attempts, especially with tones (I tried in middle school and then again on my own after college), but I've found I tend to remember phrases instead of single words, especially when it comes to tones. I can repeat tones on their own but they don't seem to stick to the words in my brain without the context of a phrase or sentence.

The biggest help for me was listening to tapes corresponding to my textbook. These days there are undoubtedly videos that are even better. There is a jarring gap between saying words in isolation, and saying entire sentences. If you learned a sentence for each word you'd actually be in better shape.

If it were 30 years ago I'd suggest gwoyeu romatzyh, a romanization that spells tones with letters. But that would probably just interfere with your pinyin. I wonder if you could adapt the idea for taking notes, though. E.g.:

* for mā write m^a
* for mǎ, the only compound tone, write maa
* for mà write m_a or ma\
* that leaves má, which can be left unmarked or written ma/

The idea is just to go to some extra trouble so your brain remembers the form more.

[bradrn]

The accusative alignment shown in I hate this argument has got nothing to do with one of the arguments being 3s. Note that the trigger for different form in the verb is the form of the subject.

…yes, I know very well that the subject (i.e. S and A, but not O) triggers verbal agreement. That’s what I mean by ‘verbal agreement is accusative’. Except that verbal agreement is only triggered when S/A is 3s, otherwise there is no marking. So it is more correct to say that verbal agreement is accusative for 3s, and direct for non-3s, with the direct agreement marker being the same as the 3s accusative agreement marker. Your example sentence I hate this argument exhibits the usual English word order, which has accusative alignment, and the usual English case-marking, which is marked-nominative for I and direct for this argument; in terms of agreement, the subject is non-3s, so the verb takes the direct agreement marker -∅.

Perhaps a non-English example may assist. The Lower Sepik language Murik has the following person-agreement affixes in the plural:

	S	A	O
1p	e-	e-	ŋe-
2p	o-	o-	ŋo-
3p	g-	bo-/mbwa-	g-

(The singular has a similar but less consistent pattern; there’s also a direct-inverse system on top of this, but that’s irrelevant here.)

This verbal agreement system clearly has accusative alignment in first and second persons, but ergative alignment in the third person. My claim is that English has a similar system, but with direct alignment in non-3s and accusative alignment in 3s.

If it were 30 years ago I'd suggest gwoyeu romatzyh, a romanization that spells tones with letters. But that would probably just interfere with your pinyin. I wonder if you could adapt the idea for taking notes, though.

I’m sure I saw a paper just a couple of weeks ago which tested this, and found that Gwoyeu Romatzyh has no positive effect on learning Mandarin. (I can’t remember whether they found that it had no effect or if they found it to be actually harmful.) I’ll post a link here if I can find it.

[zompist]

I’m sure I saw a paper just a couple of weeks ago which tested this, and found that Gwoyeu Romatzyh has no positive effect on learning Mandarin. (I can’t remember whether they found that it had no effect or if they found it to be actually harmful.) I’ll post a link here if I can find it.

I saw that too, but I'm not suggesting GR, but adapting the idea to supplement pinyin. English speakers can use all the help they can get. It's up to linguistcat to see if they like the idea and if so find out if it's useful.

[linguistcat]

I wonder if you could adapt the idea for taking notes, though. E.g.:

* for mā write m^a
* for mǎ, the only compound tone, write maa
* for mà write m_a or ma\
* that leaves má, which can be left unmarked or written ma/

The idea is just to go to some extra trouble so your brain remembers the form more.

That's a fair point, and even just writing things out would probably help, even if I use the usual notation as long as I keep the tones straight.