[NB: The update at the bottom of the article is new; if you've found this article useful please review it. Thanks! — Rick]
The most recent issue of the SBL’s Journal of Biblical Literature (vol 126, no 3) has an article entitled “The Syntax of εν Χριστω in 1 Thessalonians 4:16″ (pp. 579-593). SBL members are able to download the article from the Society of Biblical Literature web site.
The article’s authors, David Konstan and Ilaria Ramelli, examine the question of whether or not the prepositional phrase εν Χριστω (“in Christ”) attaches to the clause subject (οι νεκροι, “the dead”) or to the clause verb (αναστησονται, “will rise”).
Why is this important? Basically the question the authors seek to answer is whether it is more appropriate to translate the clause “the dead in Christ will rise” or “the dead will rise in Christ”; important to the authors as they state:
The choice between the two versions is of considerable importance. On the first interpretation, only those who have died in Christ will be resurrected, whereas the second can be taken to signify that all the dead will be resurrected in Christ—the necessary premise for the thesis of universal salvation or apocatastasis defined by Origen and other patristics writers, including Gregory of Nyssa. (580)
At this point, I think it is worth stating that the way one answers the question may allow for an interpretation of universal salvation, but it surely doesn’t dictate it. I should also note that the authors don’t say that the way one answers the question dictates interpretation; I just thought I should make that clear.
I’m not going to interact directly with the article’s argument; I just thought it would be helpful to use this as a springboard to talk some more about (surprise!) syntax searching. Because examining questions like this really is syntax searching.
The authors of the article locate all instances of the prepositional phrase (there are 84 instances)* and then work through many of them looking to see what light they shed on how the prepositional phrase is attached. Of course, if you’ve used the OpenText.org Syntactically Analyzed Greek New Testament, you know that you can at least get their reading on questions like this. Here is how they organize 1Th 4.16:
As you can see, the OpenText.org SAGNT read the prepositional phrase (εν Χριστω, “in Christ”) as modifying the noun phrase, thus “the dead in Christ.”
Next we can search to find all instances of the prepositional phrase εν Χριστω. As you can see, The OpenText.org SAGNT does not specifically mark items as prepositional phrases, but it does have consistent encoding. There are two ways that prepositional phrases are annotated, and it depends on if they are adjectival (modifying a noun) or adverbial (modifying a verb). As can be seen in the above example, when the prepositional phrase is adverbial, one has a modifier that contains a modifier that is a specifier followed by a word that is the prepositional object. This query could be expressed as follows:
Adverbial instances are different; Romans 9.1 is a good example:
Inside of the word group (wg), the head term contains the exact same structure as the modifier in the adjectival version above. This can be expressed in the Syntax Search dialog as follows:
If you combine both searches with an OR, you can get a list of all of the instances of εν Χριστω to follow along and consult as you read the article.
This essentially gives you a second opinion to check out while you follow the authors’ argument. And for technical arguments like the sort made in this article; that can be helpful.
* The authors’ count is 84; however a syntax search returns 86 hits. There are two verses that have two hits apiece. First is 1Co 4.15, which has εν and Χριστω separated by a postpositive γαρ in the second hit of the verse. The other verse is Php 4.19, which has an ambiguous modification structure (εν δοξη εν Χριστω Ιησου) that causes searches to locate each εν as the basis of the hit. Therefore a Syntax Search provides evidence of 85 instances; as the authors of the article do not provide a comprehensive hit list, there is no way to tell where these lists differ. My guess is that their count is a count of verse instances (84) and not of hits (85), though they do phrase it as if the number 84 reflects instances and not number of verses in which instances are found—a subtle but important difference.
Update (2007-12-07): I’ve revisited my original syntax search and the hit count discrepancy (84 vs 85). I’ve determined that 84 is the proper number. In my original syntax search, I should have done two things differently. First, I should have stated morphological criteria for the lexical form χριστος; or I should have just searched for the inflected text Χριστω. Second, the anything objects were unnecessary. A screen shot of the revised query is below. This query returns 84 instances, and these are likely the same 84 instances cited by Konstan and Ramelli in their article.
Hopefully this clarification helps.