Syntax Search Example: Preposition with Dative Object

On the Logos Newsgroups, a user asked a question about syntax searching:

I’d like to search for every instance of the construction in Heb 1:2 — ἐν υἱῷ – i.e. ἐν followed by noun without article … Also (I think) in 1 Thess 1:5 – ἐν λόγῳ — our gospel did not come to you not simply “by means of word\speech”

I could do a normal search, but is this a category of construction that I could find with a syntax search? If so, could someone perhaps suggest how to go about it?

The answer is a resounding “YES!” It was like a slow-pitch softball that I couldn’t resist swinging at. So I did. You can watch the video now (Flash, 9 megs, with sound) but be sure to read the rest of the post too.

I should note that I’m running 3.0a beta 2, and you may see some visual changes inside of the Syntax Search Dialog.


The query I show in the video is in two parts. First, I solve the problem of finding a prepositional phrase with the preposition of ?? and an anarthrous (so, no article) dative object. There are a few hoops to jump through to specify the structure.

Hoop 1: Prepositions in OpenText.org word groups
In OpenText.org word groups, prepositional phrases function two different ways. The most common way is for the whole prepositional phrase to also be a word group; this whole structure then functions as an Adjunct clause component, usually modifying the Predicator in the clause. This is specified by searching within the head term for a modifier that is a specifier that modifies a word in the head term.

However, it is also possible for the prepositional phrase to function within the word group, acting as a whole to modify a word within the word group. When this happens, the prepositional phrase is called a relator. The OpenText.org folks define it like this:

A Relator is a modifier which is specified by a preposition (i.e. the Relator is the object of a preposition) that modifies another element within the word group. For example, in the word group ?? ???? ??? ????????, the term ??? is in a relator relationship with the head term ????????. This relationship only applies to prepositional phrases within word groups and not when the prepositional phrase functions as a clause component.

Porter, S., O’Donnell, M. B., Reed, J. T., Tan, R., & OpenText.org. (2006; 2006). The OpenText.org Syntactically Analyzed Greek New Testament Glossary. Logos Research Systems, Inc.

Basically, we’re searching for the same structure (specifier that is a preposition modifying a word) within different containers (one as a head term, the other as a modifier). Both of these need to be specified in the query with an OR serving to join them as alternates.

Hoop 2: Occurrence Options
Another hoop involves the anarthrous-ness of the dative prepositional object. When a preposition and an article act as specifiers modifying a word, it usually looks something like this:

Here you can see that the specifier has two words as its content, the preposition and the article. We’re only interested in finding when there is one thing in the specifier: the preposition. So we use the

Occurrence option Must be only child of parent. This means that the parent structure (the modifier) only has one child (the word). If there are two words, it will not qualify as a hit. Checking this option will allow us to find anarthrous (i.e. article-free) prepositional objects.

Hoop 3: Optional Modifier
One more hoop involves accounting for an optional modifier between the preposition and the object. I’ve done this by adding a modifier that can be either a definer or a qualifier (though I suppose it could also be a relator … ) that May Repeat. This means that the modifier may not occur at all, or it may occur any number of times between what precedes and what follows. This as well is set in the Occurrence options for the modifier, selecting the May be repeated any number of times option.

Hoop 4: Multiple Possible Words
The last hoop, shown in the second part of the video, involves specifying multiple possible lemmas. This is easy, it is done by typing the necessary word into the Lexeme box and hitting enter or hitting the green go button. Enter then next word, and the list is there.

Conclusion
Hopefully this helps in explaining some issues in syntax searching. In short, yes, this is the sort of question that syntax searching helps answer.

Now that you’ve read all of this, hopefully the video will make more sense. Here’s the link again:

Comments

  1. Superb! Well swung! Thanks for that.
    Cheers
    Richard

  2. Rick,
    The explanation is great. Would you mind including why we would want to construct a search like this? I am starting to understand the how of syntax searches but I don’t always understand the why. What practical conclusions can I draw from doing this search?

  3. Hi Ed.
    Regarding practical conclusions, I’ll try to get into that in posts in subsequent weeks. In this particular example, one could ask things like “What sorts of things are done ‘in word’ or ‘in the son’?”.
    So often we focus our study on a word, and this can be useful and profitable. But we should begin to pay more attention to the things *around* the word that we study, the sorts of things that modify it and provide immediate context and information as to how the word is used. With prepositional phrases like this one, we could ask “What verb is being modified by the adjunct that contains the prepositional phrase?” Then we could either look at other instances of that verb to see what other sorts of contexts it occurs in (so, are there other prepositional phrases or adverbs used with it?). Or we could look at the prepositional phrase itself to see what other verbs it is used with. Or we could look to see who the subject is, or what the complement is when this sort of prepositional phrase is used.
    Basically, having a syntax graph allows us to get a glimpse of the grammatical goings-on in the current passage, and it allows us to start asking questions not just about words, but also about relationships between our study word and other word level or syntax level components. Once we start asking these sorts of questions, we can gain insight into the current context-sensitive usage, and our exegesis will profit from it. Even if the conclusion is only to say, “yeah, that’s what it says.”
    In linguistic terms, examining syntax allows us to “explicate the implications”. That’s fancy-talk for saying it can help confirm or make explicit the things we implicitly (though perhaps not consciously) make note of when reading the translation. It can also, with work, help us make explicit the things that the original hearers might have heard that we would otherwise miss; making explicit those things that the original hearers would’ve just understood but we end up missing because we’re so far removed from their context.
    Hope it helps!

  4. Mike Hogue says:

    Hi, and let me start by saying I’m really enjoying these “syntax search” blogs.
    My own confidence in my ability to use them, however, is shaken. . .normally, when a problem is introduced I try to come up with the solution myself before reading the answer. I’m beginning to feel like it would be tough to have confidence the results for a syntax search would give me all cases I’d be looking for.
    If I were looking for this structure as in Heb 1:2, I’d probably open the opentext version and look at how it was laid out, and search for that construct. . .in this case, the Adjunct clause component. How would I gain the knowledge of the opentext.org to tell me that prepositions could be used in the 2 ways described above? These blogs are extremely helpful for things like that, but make it difficult for an average joe like me to get a search result and have confidence that all the cases of what I’m looking for would be covered. . .I’d think “what kind of clause component will this show up in that I’ll miss with this search”. Certainly, I’ll get some results I’d want, but will I get them all?
    Are there any rules of thumb or good things to read to see that “In OpenText.org word groups, prepositional phrases function two different ways. ” or other constructs that may not be obvious?
    Thanks for your help. I do really enjoy reading these syntax search blogs, and would like to see more so I can learn.
    Mike