But they're approaching it from a perspective that's something like: "We believe grammar has a number of universal elements, and we can analyze every language using the same basic theory, if only we can figure out what that theory is"

And you'll see a lot of elements in that paper based on that concept.

I personally tend more toward the idea that we should abandon universality as soon as it makes things more complicated, and instead describe things in a language-particular fashion.

Oh, my apologies!

(I meant to write "it doesn't seem to me" above, edited :-)

I think -(r)are- is actually significantly more complicated than that.

In the traditional analysis, it has four basic meanings: 自発・受身・可能・尊敬 (spontaneous, passive, potential, honorific)

And the 受身 can be subdivided into two types, direct and indirect passives

So there's discussion about whether we can talk about those two passives as though they're really the same thing, or if they have to be analyzed as two thingies.

Yes, that's right.

So, this paper isn't concerned with unifying the four "basic meanings" listed above, but it is concerned with unifying the direct and indirect passive. Can they be considered one thingy?

@ogicu8abruok The paper attempts to tackle that question, though in more specific terms :-)

I haven't read the whole thing, though.

But it argues in essence that the distinction between the two is not only unnecessary, but leads to an inadequate analysis.

And that if you start with a more general theory, you can account for the data (better!) without drawing a line.

That is what the paper is about.

I haven't read the whole thing.

I can't claim that it's convincing or unconvincing :-)

It's certainly interesting!

Hmm... You might want to start with a summary of the unified versus separate debate. See Shibatani's The Languages of Japan pp.320-333

(This paper is newer than that.)

Although the paper itself has its own summary (in less detail) of existing analyses

Each section has its own numbered examples, so there's more than one 8a. :-)

You mean on page 10, I assume

But honestly, I need to spend a lot more time with it before I start talking about things like "convincing" or "unconvincing", so I will have to leave my opinions out of it :-)

I'm hardly an expert on the subject.

As an aside, some people made up a thing they call "Spritz", in apparent defiance of reading science

Yeah.

I love how they have a section on their site called "science"

It's actually a pretty funny read.

> Traditional reading involves publishing text in lines and moving your eyes sequentially from word to word. For each word, the eye seeks a certain point within the word, which we call the “Optimal Recognition Point” or ORP. After your eyes find the ORP, your brain starts to process the the meaning of the word that you’re looking at.

> When reading, only around 20% of your time is spent processing content. The remaining 80% is spent physically moving your eyes from word to word and scanning for the next ORP.

Hoo boy.

> When your eyes encounter punctuation within and between sentences, your brain is prompted to assemble all of the words that you have read and processes them into a coherent thought.

Hahaha.

Actual fluent readers don't pause their eyes on every single word, and they don't read a word at a time. They also don't read at a constant rate. The idea that we spend 80% of the time with our eyes moving (and are unable to process information while our eyes are moving!), and that we assemble words into thoughts when we hit punctuation? I'm not sure that really merits a response

Some introductory material for anyone who's interested: microsoft.com/typography/ctfonts/WordRecognition.aspx

@ogicu8abruok Can you give me a page number so I can follow along?

Oops, it's -(r)are-. I goofed earlier when I tried to talk about it :-)

（ら）れる is still the basic form of it in my mind.

For starters, are you familiar with the basic concept of raising?

(I don't know which things you understand and which you don't.)

Raising is a metaphor. A syntax tree has upper and lower positions. When something is moved from its expected location (lower in the tree) to a higher position, it can be described as "raising".

If you look at the chart on page 23 that started this discussion...

And take TP (Tense Phrase) as meaning "tensed clause", DP as meaning "noun phrase" (the DP hypothesis is irrelevant to this paper)

Well, you see the raising in the chart, right? :-)

In this case, there's something called a VP shell which I think is there because it takes multiple arguments, but the theory is designed around the idea that the tree should only have binary branching.

So extra nodes are inserted so there can be multiple arguments.

Sigh. This analysis is so complicated.

I'm not a huge fan of Chomsky's minimalist program.

(That's where this sort of structure is coming from.)

Yeah, any upward movement in the tree is raising.

(Any downward movement is lowering.)

Wikipedia's article looks like it describes a specific thing that gets referred to as "raising".

But the term is used more generally.

@ogicu8abruok The term VP = Verb Phrase, which is basically a fancied up term for predicate. It means a phrase with a verb as its head.

Verb ~ predicator

Let me go back to page 10 :-)

Ahh, when I click the (2) hyperlink it takes me to the wrong example! :-) Hold on, I have to get back to where I was supposed to be

(2) Naomi-wa hahaoya-ni sin-are-ta.

Yeah, it's saying there's no active source for passivization that would result in (3,4,5)-a. It's saying that -(r)are- doesn't introduce an external argument, because if it did, there would be active source sentences that could be passivized resulting in (3,4,5)-a. Instead, if you assume that there's another source for the nominative argument (a form of raising), you can come up with a generalization that allows (2) to be formed via passivization but not (3,4,5)-a

Therefore, it makes more sense to assume it's generated by raising than to assume it introduces an external argument. It says.

By the way, I hate this unaccusative/unergative stuff.

You need to figure out how a verb being unergative relates to it not having a big VP shell.

I don't have enough time to spend on this paper to work out the rest of the details at the moment.

Oh, yes.

死ぬ is unaccusative. 踊る is unergative.

The unaccusative/unergative stuff is always confusing. Different people mean different things by those terms, and a lot of linguists are unhappy with them :-)

Oh, that's just what the paper says.

I mean, I was trying to help work through what the paper says.

I don't know if that's really how it works. I haven't read the whole paper :-)

I don't know if it's correct or not.

My personal feeling is that there is probably a simpler explanation lurking somewhere.

This stuff all feels like it's more complicated than it needs to be. But the most important thing is probably whether it accurately describes the phenomena, right?

Oh, does it?

I misread it then. Just a moment

> However, the indirect passives that are said to require rich context exhibit a considerable amount of interspeaker variability with respect to acceptability ..., and passives like (3-a), (4-a), and (5-a) are all unacceptable to me even when contextualized.

It says they are not acceptable with context.

Good night, everyone!

@Earthliŋ Unfortunately, there's no way to tag stuff without bumping it. One possibility is to tag a little bit at a time, instead of bumping them all at once