List of Articular Infinitives

Stephen Carlson
Posts: 2825
Joined: May 11th, 2011, 10:51 am
Location: Melbourne
Contact:

Re: List of Articular Infinitives

Post by Stephen Carlson » August 11th, 2015, 7:10 pm

Jonathan Robie wrote:
Stephen Carlson wrote:
Romans 4:13 wrote:τὸ κληρονόμον αὐτὸν εἶναι κόσμου
As for Rom 4:13, it looks to me like τό construes with the det/noun phrase τὸ κληρονόμον ... κόσμου, not with the infinitive εἶναι. Again, I haven't checked the treebank.
Am I missing something? τό is neuter, κληρονόμον and κόσμου are both masculine. To me, it looks like τὸ has to construe with the infinitive εἶναι.
Well, I was missing my second cup of coffee... Sorry for that. If Burk doesn't identify it as an articular infinitive, maybe he explained himself (I no longer have his book). I'm not seeing it now.
0 x


Stephen C. Carlson, Ph.D.
Melbourne, Australia

Jonathan Robie
Posts: 3605
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: List of Articular Infinitives

Post by Jonathan Robie » August 11th, 2015, 7:28 pm

Stephen Carlson wrote:
Acts 4:18 wrote:καὶ καλέσαντες αὐτοὺς παρήγγειλαν τὸ καθόλου μὴ φθέγγεσθαι μηδὲ διδάσκειν ἐπὶ τῷ ὀνόματι τοῦ Ἰησοῦ.
I'm sure Burk discusses this (I returned his book on ILL), but according to BDF § 399(3), "A 4: 18 παρήγγειλαν τὸ (om. S* B) καθόλου μὴ φθέγγεσθαι: the art., if correct, is to be taken with καθόλου, cf. § 160, Diodor. 1.77 (I 130. 17 Vogel)." I don't know, however, how the treebank parses Acts 4:18.
My graphics aren't as nice as Randall's, but here's what it looks like in the lowfat:
acts4-18.png
acts4-18.png (35.01 KiB) Viewed 1150 times
With this interpretation, you could take καθόλου out entirely without doing much damage to the structure, the following sentence has almost the same structure:
καὶ καλέσαντες αὐτοὺς παρήγγειλαν τὸ μὴ φθέγγεσθαι μηδὲ διδάσκειν ἐπὶ τῷ ὀνόματι τοῦ Ἰησοῦ.
And both φθέγγεσθαι and διδάσκειν are articular here.

Fortunately, BDR's section numbers often correspond to BDF, and it looks like they are taking this as an adverbial accusative. What I don't see, though, is a discussion of why that's the right interpretation instead of the interpretation given above, it just tells me how they interpret it. I assume both interpretations are possible and reasonable?
0 x
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/

Stephen Carlson
Posts: 2825
Joined: May 11th, 2011, 10:51 am
Location: Melbourne
Contact:

Re: List of Articular Infinitives

Post by Stephen Carlson » August 11th, 2015, 8:22 pm

Jonathan Robie wrote:Fortunately, BDR's section numbers often correspond to BDF, and it looks like they are taking this as an adverbial accusative. What I don't see, though, is a discussion of why that's the right interpretation instead of the interpretation given above, it just tells me how they interpret it. I assume both interpretations are possible and reasonable?
Well, that's the problem with BDF/BDR: often their interpretations are merely identified, not justified. That they came down that way, though, is sufficient to explain why Burk came down the way he did.

As for the justification, I suppose one can go to the bigger grammars, Mayser, Kühner-Gerth, etc. and see what they say. Looking at BDAG, it appears that the article is often (but not always) present with this adverb. I don't have any feel, however, for why. A serious corpus analysis might answer it (in conjunction with a decent theory of the article), ... but that takes a lot of time I don't have and we may still be in the dark after all that work.
0 x
Stephen C. Carlson, Ph.D.
Melbourne, Australia

Randall Tan
Posts: 23
Joined: June 30th, 2011, 12:44 pm

Re: List of Articular Infinitives

Post by Randall Tan » August 11th, 2015, 8:53 pm

Jonathan,

Thanks for continuing to work on this & for posting your discussions of the issues.

All cases below that you discussed are actually meant to represent what you expect to find--articular infinitives in all cases. However, limitations of representation with the head attribute has apparently caused a problem with search. I included attached jpegs with the 1st 3 cases to provide a visual aid (I'm apparently limited to 3 attachments per post):

Mat 20:19 (2nd & 3rd ones missed),
Mat20_17.jpg
Mat20_17.jpg (42.2 KiB) Viewed 1126 times
Luke 9:18 (periphrastic),
Luk9_18.jpg
Luk9_18.jpg (33.41 KiB) Viewed 1126 times
Luke 11:1 (periphrastic),
Luk11_1.jpg
Luk11_1.jpg (37.86 KiB) Viewed 1126 times
Luke 24:15 (2nd one missed)
Php 1:23 (2nd one missed)

When multiple infinitives are modified by the same article, we purposely combined the infinitival clauses together first before we connected the article to the combined series of clauses. This is, in fact, what we have done for all the above cases. However, when any 2 nodes (which may contain multiple nodes within themselves) are combined, there can only be 1 head (a machine limitation). When 2 or more infinitival clauses are connected together first before they are connected to the article, one node had to be assigned as head, even if it is arbitrary. The first infinitive is automatically assigned as head in this type of multi-clause connection situation. So, if you are using the head attribute, only the first infinitive will be found with the article.

I should point out that I expect that the same problem will appear if you are searching for nouns/participles or anything else that shares one article. The head attribute will only point to the first of the conjoined nouns/participles, etc.

With Luke 9:18 (periphrastic), Luke 11:1 (periphrastic), there is an additional factor. With periphrastic constructions, we had chosen to make the participles the head of periphrastic constructions & not the “to be” verb. The “to be” verb happens to be the infinitive in periphrastic infinitives & since the participles are the heads, the infinitives are missed if you are searching for article with infinitive as head of the clause the article.

What becomes apparent is that the head attribute alone will not solve this search problem. With the problem of articles modifying multiple elements, maybe one way around the problem is to add some post-processing to add an "articular" attribute to all sub-nodes at the next level down directly connected to an article node. This should solve all problems finding any element that is "articular," even the 2 periphrastic cases if you are looking for infinitives within the clause. I added the limitation "all sub-nodes at the next level down directly connected to an article node" because there may be embedded infinitival clauses within the infinitival clause that you would not want to be taken as "articular." This is because the whole infinitival clause is articular, but any embedded clauses are not articular on their own embedded clause level.

I will discuss the 2 real remaining problems--Rom 1:11 & Rom 15:8 in 2 separate replies so that I can include visual aid attachments with the discussion.
0 x
Randall Tan

Randall Tan
Posts: 23
Joined: June 30th, 2011, 12:44 pm

Re: List of Articular Infinitives

Post by Randall Tan » August 11th, 2015, 9:12 pm

I'm aware that the current tree representation for Rom 1:11 is problematic. I've attached the current tree & 2 possible alternatives to start a conversation on which might be better. There might be other possible alternatives; however, I can only have 3 attachments per post (hence only 2 alternatives):

Current tree
Rom1_11Current.jpg
Rom1_11Current.jpg (112.07 KiB) Viewed 1115 times
Alt 1
Rom1_11Alt1.jpg
Rom1_11Alt1.jpg (104.37 KiB) Viewed 1115 times
Alt 2
Rom1_11Alt2.jpg
Rom1_11Alt2.jpg (103.57 KiB) Viewed 1115 times
0 x
Randall Tan

Randall Tan
Posts: 23
Joined: June 30th, 2011, 12:44 pm

Re: List of Articular Infinitives

Post by Randall Tan » August 11th, 2015, 9:15 pm

I'm aware that the current tree representation for Rom 15:8 is problematic. I've attached the current tree & 2 possible alternatives to start a conversation on which might be better. There might be other possible alternatives; however, I can only have 3 attachments per post (hence only 2 alternatives):

Current tree
Rom15_8Current.jpg
Rom15_8Current.jpg (106.16 KiB) Viewed 1115 times
Alt 1
Rom15_8Alt1.jpg
Rom15_8Alt1.jpg (108.1 KiB) Viewed 1115 times
Alt 2
Rom15_8Alt2.jpg
Rom15_8Alt2.jpg (109.71 KiB) Viewed 1115 times
0 x
Randall Tan

Jonathan Robie
Posts: 3605
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: List of Articular Infinitives

Post by Jonathan Robie » August 12th, 2015, 2:40 pm

Randall Tan wrote:What becomes apparent is that the head attribute alone will not solve this search problem. With the problem of articles modifying multiple elements, maybe one way around the problem is to add some post-processing to add an "articular" attribute to all sub-nodes at the next level down directly connected to an article node. This should solve all problems finding any element that is "articular," even the 2 periphrastic cases if you are looking for infinitives within the clause. I added the limitation "all sub-nodes at the next level down directly connected to an article node" because there may be embedded infinitival clauses within the infinitival clause that you would not want to be taken as "articular." This is because the whole infinitival clause is articular, but any embedded clauses are not articular on their own embedded clause level.
That would definitely solve the current problem. As discussed in email, I'd like to talk about this and see if this is one instance of a more general issue that affects queries needing to identify corresponding constituents in recursive structures. I don't know if it is or not - for participles, it currently looks like head is sufficient (now that wgs can also be heads), but I don't yet have enough experience to be sure.

I'll probably start a separate thread on participles in the Rijksbaron section in the next week or so, using examples drawn from your treebanks using XQuery. Micheal and I are focused on that for our SBL 2015 presentation.
0 x
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/

Jonathan Robie
Posts: 3605
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: List of Articular Infinitives

Post by Jonathan Robie » August 12th, 2015, 3:26 pm

Either of these alternatives work for me, I prefer Alt 1, but it could obviously be read either way. In general, I think the relationship among subclauses within a clause is often a little subjective, and that's not usually the part that my queries depend most heavily on.
Randall Tan wrote:I'm aware that the current tree representation for Rom 1:11 is problematic. I've attached the current tree & 2 possible alternatives to start a conversation on which might be better. There might be other possible alternatives; however, I can only have 3 attachments per post (hence only 2 alternatives):

Alt 1
Rom1_11Alt1.jpg
Alt 2
Rom1_11Alt2.jpg
0 x
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/

Jonathan Robie
Posts: 3605
Joined: May 5th, 2011, 5:34 pm
Location: Durham, NC
Contact:

Re: List of Articular Infinitives

Post by Jonathan Robie » August 12th, 2015, 3:49 pm

Here I'm having problems with the representation of "τὰ δὲ ἔθνη" in both versions. I was expecting a single CL containing the conjunction δὲ, followed by perhaps a CL that contains the rest of the clause that contains τὰ ἔθνη, κτλ. But I could easily be missing something here.
Randall Tan wrote:I'm aware that the current tree representation for Rom 15:8 is problematic. I've attached the current tree & 2 possible alternatives to start a conversation on which might be better. There might be other possible alternatives; however, I can only have 3 attachments per post (hence only 2 alternatives):

Current tree
Rom15_8Current.jpg
Alt 1
Rom15_8Alt1.jpg
Alt 2
Rom15_8Alt2.jpg
0 x
ἐξίσταντο δὲ πάντες καὶ διηποροῦντο, ἄλλος πρὸς ἄλλον λέγοντες, τί θέλει τοῦτο εἶναι;
http://jonathanrobie.biblicalhumanities.org/

Randall Tan
Posts: 23
Joined: June 30th, 2011, 12:44 pm

Re: List of Articular Infinitives

Post by Randall Tan » August 12th, 2015, 4:46 pm

Sorry, I'm not sure I understand what you mean, Jonathan. Can you clarify? Conjunctions that join clauses actually are always in between the clauses they join whenever we can show all the clauses involved in the same tree. When conjunctions function in relation to sentences, as a way to quickly check if there are any orphaned trees (every tree should always only have one topmost node), we had to place the conjunction with the second sentence it conjoins & connect it to that second sentence. We removed the Head attribute for the node where the conjunction connects to the second sentence as one way to indicate that the conjunction doesn't really belong with the sentence--there is no head when it connects to the sentence. The conjunction is still considered to function between the two sentences to conjoin them.

Sorry that we had to make this compromise. Because the trees are automatically machine-parsed, with any manual changes also needing to be parsed (we intervene in the parsing process with manual instructions according to rules the machine can understand), trees quickly become unwieldy & might even exhaust machine resources & cannot finish parsing if the tree just too big. The current trees do clause level & below well & they can mostly handle sentences. But keeping long speeches together or dealing with higher text linguistic units above sentence level would require building on top of the trees in a different way & with different tools.

As for why the 3 visualized options are the way they are:

Current: The conjunction de conjoins 2 infinitival clauses with both infinitival clauses modified by the article & the preposition.

Alt1: The conjunction de conjoins 2 infinitival clauses (same infinitival clauses as with the Current tree), the change in subject for the 2nd infinitival clause is considered to make the second infinitival clause sufficiently distinct that it is not modified by the article & the preposition. In other words, both infinitival clauses function in the same way as ADV, but the second infinitival clause has a different structure, without the article & the preposition.

Alt2: The conjunction de conjoins 2 infinitival clauses (the 1st infinitival clause is different than in the case of the Current tree & Alt1) with both infinitival clauses the speech introduced by "I say".
0 x
Randall Tan

Post Reply

Return to “Syntax and Grammar”