zompist bboard

I've looked all over the place and can't figure this out... I'm basically trying to figure out where the adpositional adverb goes in this complex sentence and I can't find anywhere that discusses this topic. It is the very few times I think that people are trying to be too simple in their examples that they never show an example of things that only happen in more complex formulations. Anyways, the sentence is:

"One day as they were walking along they looked down on the ocean and wondered what was beneath it."

-----------------------------------------------------------------------------------------------------------------------------------------------------

I've broken it down as such:

One day as they were-walking along
((num adjective One) (noun day)) ((prep as) (noun they) ((verb-propst walking) (adverb along)))

they looked down on the ocean
(pn they) ((verb-pst looked) (adverb down)) ((prep on) (noun Ocean))

and
(conjunction and)

wondered what was beneath it.
(verb wondered) (pn what) (i-verb was) ((prep beneath) (pn it))

-----------------------------------------------------------------------------------------------------------------------------------------------------

I'm trying to make a SOV language and it seems that conjunctions should be head, so working with what I have, the word order without the initial adverbial preposition should be something like:

"they (the)ocean(on) down looked wondered what it(beneath) was and"

-----------------------------------------------------------------------------------------------------------------------------------------------------

The question I have is where does this go? This is modifying both clauses if the complex sentence so it seems like it should go with both clauses, but that would be dumb imo, so should it go before the "and"?

"One day, they(as) along walking(were)"

And come to think of it, should "One day" and the rest of the sentence be separated?

-----------------------------------------------------------------------------------------------------------------------------------------------------

This is the Sentence structure I'm working with:

Sentence Structure Order : [Subject] [Indirect Object] [Direct Object] [Verb]
Noun Phrase Order : [Adjective] [Noun] [Possessive] [Adpositional Phrase] [Complementizer]
Verb Phrase Order : [Adverb] [Verb] [Copula] [Adpositional Phrase] [Complementizer]
Adjective Phrase Order : [Adjective Adjective?] [Adjective]
Adpositional Phrase Order : [Noun]

It still needs some work probably.



-------------------------------------------------------------------------------------------------------------------------------------------------------

edit: I thought about it a bit more, still not sure, but I think the correct answer might be the following;

[(As)They]
[(on)(the)ocean)] [down looked]
[what (beneath)it was] [wondered]
[[one day] along (were)walking] and

"(As)They (on)(the)ocean) down looked what (beneath)it was wondered one day along (were)walking and"

note: moved the () words to read better, in english, I don't think it matters where they are since they're all not there in the SOV language I'm translating to.

edit2: thought more about it... changed to another order I think is correct. I didn't realize that "what (beneath) it was" should be regarded as an object and I was regarding it as a adposition, but then I realize it so I flipped that around and then also correct the mess up of "one day" being considered a second adverb to the entire sentence when it was really modifying the "were walking along" adverb and so I corrected that... I still don't know if that whole adosition is in the right position for that sentence though..

Many strict SOV languages don't have a conjunction "and", but embed one of the joined sentences into the other, using participle constructions:
In a typical Turkic language (classical SOV language), your sentence

One day as they were walking along they looked down on the ocean and wondered what was beneath it.

would look more or less like this (/ marks clauses):
One day they along walking / ocean-on down-looking (or: ocean-on-down looking) / what it-beneath being {quotation particle} / wondered.
Now, some Turkic languages have borrowed (Turkish ve, Uzbek va) or otherwise developed (Kazakh және) a conjunction "and" under the influence of Persian and Arabic. They are mostly used to join noun phrases. When they join clauses, their position is no different from that in English as as their syntax is borrowed from non-strict SOV languages.

I've never seen a language that allowed sentences (or any long constituent) to be coordinated by an "and" at the end, after juxtaposing both elements, and I doubt that they exist. There's just not much of a point of having an "and" if it's not at or near where the sentences or constituents are separated.

I also think your final answer sounds wrong from a pragmatic point of view. In the English original, "one day as they were walking along" appeared first and set up background information for the actions that seemed to be the communicative focus: looking down at the ocean and then wondering what lied at the bottom. In most if not all languages, I think speakers would highly tend to use a similar clause order.

I also think "one day" affects the whole sentence, although you could put it in [... along were walking] if you want...

I would suggest:

[They]
[
[as [they] [one day along were-walking]] [on the ocean] [down] [looked]
and
[what [beneath it] was] [wondered]
]

That's quite a number of brackets so I'm also going to provide a tree:




There's other ways to organize the sentence. For instance, in Classical Latin (a language often said to be "non-configurational but largely SOV, with certain SVO tendencies") you could say the following:

Quodam

certain.ABL

die,

day.ABL,

una

along

ambulantes

walking.PL.NOM

in

in

mare

sea.ACC

despexerunt,

look.down.PRET.3PL,

curiositas-que

curiosity.NOM-and

sciendi

know.GRNDV.GEN

quod

what.NOM

subter

beneath

erat

be.IMPF.3SG

eos

3PL.ACC

tenuit.

hold.PRET.3SG

(GRNDV = gerundive.)

A somewhat literal translation: "On a certain day, walking along they looked down on the sea, and curiosity to know what was beneath it held them." Latin does not have a verb equivalent to 'to wonder'. Also note that -que 'and' appears attached at the end of the first word of the second sentence (in my syntax tree below I had to move it to the middle, because in syntax trees lines cannot cross and constituents must be continuous).

Structure of my translation:
[On a certain day]
[along walking] [on the sea] [they looked down]
and
[curiosity [of knowing [what [beneath it] was]]] [them] [held]

Tree:


The provided "(ii)" '3PL.NOM' is provided because in syntax trees an S normally has a subject, and in this case it is said that the speaker (me) just dropped it, leaving the AdjP hanging in there. Latin is a pro-drop language.

EDIT: Fixed stupid typo in the second tree.

Serafín wrote:in my syntax tree below I had to move it to the middle, because in syntax trees lines cannot cross

Just as an aside, James McCawley would have disagreed. He played with trees where lines could cross, connecting to somewhere else than their linear order. So you could put your -que in its proper linear order, and connect it up to the same spot.

(To me that kind of seems like cheating— but the rules of syntax trees are not God-given and can be changed if we find something that makes sense! Something like McCawley's twist might be appropriate for a rule that operates on the surface level, like "place this item just after the first constituent".)

zompist wrote:

Serafín wrote:in my syntax tree below I had to move it to the middle, because in syntax trees lines cannot cross

Just as an aside, James McCawley would have disagreed. He played with trees where lines could cross, connecting to somewhere else than their linear order. So you could put your -que in its proper linear order, and connect it up to the same spot.

(To me that kind of seems like cheating— but the rules of syntax trees are not God-given and can be changed if we find something that makes sense! Something like McCawley's twist might be appropriate for a rule that operates on the surface level, like "place this item just after the first constituent".)

I do personally think that lines should be able to cross at least some of the time, and that there's some merit to dependency grammar, but alas this makes me non-mainstream.

I'd also like to interject that it's really the first word that -que attaches to, not the first constituent, which means that it's often found breaking the continuity of NPs, VPs, PPs and whatnot. Latin speakers loved discontinuous constituents.

• Dardanorum gentem delere propositum erat inque eorum agro sedes dare Bastarnis. (Livy, Ab Urbe Condita 40.57)
  The plan was to exterminate the Dardani people and make a home for the Bastarni in their territory.

Here -que coordinates the constituents Dardanorum gentem delere and in agro sedes dare Bastarnis, attaching itself to the preposition in.

Proper names can be broken too (which, by the way, also applies to the so-called "postpositives" enim, autem, vero, quidem, igitur, etc.).

• Gaiusque Tuditanus (Cicero, Brutus 95)
  Quintus enim Pompeius (Cicero, Brutus 96)

Sometimes you even find -que attaching to the second word, when the first word is a preposition. The first sentence quoted in this post could've contained in agroque sedes dare Bastarnis instead.

• Dionysius omnia, quae moveri poterant Dionis, in navis imposuit, ad eumque misit. (Cornelius Nepos, Vita Dionis)
  Dionysius put everything belonging to Dion that could be moved into the ships, and sent them to him.

Thanks for the help ^.^
The "and" at the end was what someone told me on another site months ago. It didn't look right.
I think I understand how to do it now.