Another question

[Daniel]

freknu,

Of course, I can, silly... :\ *scratches head*

Hmm.... I re-read it. It seems like normal English, but I get a little confused partway through it. It certainly could work if I already understood the context (maybe in the middle of an argument-- "no, it's actually THIS WAY: ..."). It's also right-attaching (rather than center-embedding) so that might make it a little easier. It's on the border of making sense, but still, if you have 100 levels if could be very confusing. (Some interesting examples of this are early American legal documents (a favorite of mine, where multiple paragraphs are one sentence is http://www.archives.gov/exhibits/featured_documents/emancipation_proclamation/transcript.html ).

There is also nothing wrong with "the cat that the dog chased that the rat bit, ran", then again, maybe native English speakers don't use structures like that. I might just be using my native logic in speaking non-native English.

Is there a limit, though? Can you go to 4 levels? or 5? or 10? Some languages are said to more easily do 3 levels than English, but few if any can naturally do 4. If you can supply data (in any language) on this I'd be curious. I wouldn't be shocked by 4-5 levels, but beyond that would be interesting! Even 4-5 would be worth publishing about though I think.


--

This is what I mean by "widest impact":
A sentence like "I know [whom you will ask] or [whom to ask]" can, I feel, be replaced by "I know [him/her]". This tells me that the CP "whom you will ask" or "whom to ask" is basically expressed in terms of the [spec, CP]. If something is embedded in this CP (like another CP), it too will be expressed in terms of the first [spec, CP] (and so on). Is that basically right?

The nature of recursion is that each level is independent and may internally be recursive as well. So I don't really follow. You can embed another level within that, or, also importantly, you could embed that in another level:
"I know that you know whom to ask." And so forth.
So "widest" is relative.

[casey61694]

This really cleared things up for me, thank you!

[casey61694]

I have another question. I have to admit that this one's pretty trivial, but I still think it's worth sharing:

 Are "semi"-wh-questions ("How about..." or "What about...") another way of asking yes/no questions?

For example:

"Do you like the color purple?"  "No."
"What about green?"  "No."
"How about red?"  "Yes."

[Daniel]

No. They introduce topics and imply (elided) yes/no questions.

"How about red, do you like red?"

(Certainly they function to do so, just like "You?", but there's no particularly interesting structural relationship between semi-wh- and yes/no- questions.)

[casey61694]

Pardon the odd example.

"The storm ate up September’s cry of despair, delighted at its mischief, as all storms are."

In this sentence, is "as" acting as a "relative pro-participial phrase" introducing an adverbial-type clause?

Any clarification/help is appreciated. Thank you all!

[Daniel]

Can you explain a bit more?

"As" in that case is what is traditionally called a subordinating conjunction. In formal syntax it's probably best analyzed as a Complementizer. Either way, it links clauses together.

The meaning of that whole clause then is as an adverbial, specifically a manner adverbial, which could answer the question "How [was it delighted]?".

There are pro-forms that aren't pronouns. For example, "do" is a pro-verb (actually more like a pro-VP, though I guess pronouns are usually pro-NPs), and "so" is a pro-adverb[ial] (or sometimes pro-adjective?). But "as" here I don't think counts. Instead, the sentence just involves ellipsis. "As" certainly supports that by suggesting parallelism.

[casey61694]

That is just how I've always parsed sentences like that. It makes semantic sense to me. It also behaves like other relative clauses. For example:

"This is the green toad, which no one else found."

The relative clause cannot be preposed (unlike a subordinate clause).

"Which no one else found, this is the green toad."

Similarly,

"This toad is green, {as all toads are}."

what is in curly braces cannot be preposed.

"As all toads are, this toad is green."

I think that "as" in this case is functioning as a relative pro-adjective introducing a relative adverbial clause.

Also, what do you mean by the last sentence in your last post?

Thank you djr!

[jkpate]

Similarly,

"This toad is green, {as all toads are}."

what is in curly braces cannot be preposed.

"As all toads are, this toad is green."

The preposed sentence is fine for me. I suppose I might be able to get a comparative reading with degrees of green-ness on the non-preposed version ("this toad is as green as all toads are") that is not available on the preposed version:

"This toad is (as) green as all toads are"
"*As all toads are, this toad is as green"

[Daniel]

Yes, the preposed version is fine, possibly not as natural.

As for my last sentence, usually under coordination there are parallel structures and there is ellipsis:
He sang and [he] danced.
He sang and she did so too.

So when we see ellipsis there, it probably represents something about those events/clauses being parallel. I don't have any specific analysis for 'as' in mind though.

[casey61694]

Why can't "am" contract with the negative particle?

Is this sort of thing not possible because "am" ends with a consonant?  Don't "is" and "does" and "should" also end with consonants? I know nothing about phonology or phonetics, but I'm guessing there are different "degrees" of consonant.  Is "m" a "harder" consonant than "s" and "d"?

He's brave, isn't he?

You're brave, aren't you?

----------

I'm brave, aren't I? (?)

I amn't scared!

[jkpate]

There is a contracted form for "am not" in some varieties of English:

I'm brave, ain't I?
I ain't scared!

However, this form is stigmatized as uneducated, so many talkers avoid it. Another factor may be the high frequency of "am not," since very frequent forms tend to be irregular. However, they also tend to be short, but "am not" of course is longer than "ain't." In any event, I don't think there's any regular phonological rule involved.

[casey61694]

very frequent forms tend to be irregular

It's kind of strange to me that the opposite isn't true instead. Shouldn't frequently used forms (like the verb "to be" for example, with its eight forms ) become more streamlined with time and predictably follow "the rules"?

Could you maybe give another example of this? I appreciate your help!

[jkpate]

Perhaps the tendency is easier to understand if I state it in the other direction: less frequent forms tend to be more regular. It's well-known that language vocabularies tend to be Zipfian: most of the word types (i.e. lexical entries) are rare. A system is easier to learn or remember if the rarest 95% of forms follow a rule than if the most frequent 5% of forms follow the rule. You can more easily memorize an irregular 5% and a rule for the other 95% than a rule for 5% and irregular 95%. So, if you want to streamline the system over time, you get more bang for your buck by streamlining the rare types.

[Daniel]

Any word can be irregular, but over time the less frequent ones will be forgotten and regenerated by the grammar as regular. You can, if you want an example, look at the history of irregular verbs in English over the last 1000 years or so. As words are used less frequently, they regularize. We don't know exactly what the past or participle forms of "smite" should be, but they're in early versions of the bible and other texts.

More frequent forms are preserved as irregular, and over time they may end up irregular, so eventually this means they probably will be. The verb "be" is irregular in almost every language (I'm thinking about a broad sample, but this is at least literally true throughout Europe), at least if there are any irregular verbs at all.


There's also something similar where "complexity" (=irregularity?) is preserved by small communities with minimal contact with others. Several recent theories about social dimensions of complexity have focused on this, such as work by Trudgill.

[casey61694]

So the natural tendency is for verbs (for instance) to be irregular?

If all people hypothetically had unlimited memory space, would it then be reasonable to say that this pattern would've continued and endured for pretty much every verb? Would "irregular" have become the new "regular"?

P.S. I've never thought of verb forms as being somewhat determined by limits on our memory. That's interesting.