I think we’d do best to start with Robert J. Matthews’ framework that the JST contains four types of material:
(1) restoration of the text to the way that it originally read
(2) material that was not originally part of the biblical text
(3) Joseph Smith’s commentary
(4) material added for doctrinal harmonization
Phil Barlow offers a similar list. So the first thing that we have to do is to try to figure out what kind of material we are dealing with; sometimes this is easy and sometimes it is not.
For example, the JST for Mark 2:14 includes the phrase “as was customary in those days,” making it pretty clear that it does not represent something that was originally in the text because it is extremely unlikely that Mark would refer to events in Jesus’ lifetime as if they were ancient history and unfamiliar to his audience.
In another case which I discuss here, it seems that the JST adds language which does not contribute to the meaning of the text in any obvious way–it seems to just make the text repeat itself–but I think it does create a chiasmus. Now, it is always possible that these structures are mere coincidences, but I think it is also possible that this is deliberate, not intentional on the part of Joseph Smith (or even necessarily recognized by him), and thus could make a claim for being an earlier version of the text than the canonical version.
While placement of each JST into Matthews’ four categories can never be entirely certain, I believe it to be an important first step as we move toward using the JST. So the role that the JST should play in LDS study of the NT is going to depend, I think, on what kind of JST it was. Some of them will shed no light whatsoever on how the original text read because they solely contain information for later readers; others may represent a restoration of the text to an earlier iteration.
I think there are also instances where the JST is basically telling us something like this: Joseph Smith recognized a problem in the text and offered up a solution, although it might not be the only or even the best solution. For example, several JSTs change instances in the Hebrew Bible where God repents, sometimes so that a different person in the story repents. (See, e.g., Moses 8:25.) I agree with Joseph that there is a theological problem with God repenting. I doubt, however, that the story originally had Noah repenting. We might argue that the text could have been translated differently (such that God relented or was persuaded . . . assuming you don’t have theological problems with those options) or that a broader definition of repent (=changing one’s course) consistent with God’s actions was intended.
In a really interesting recent article by Heather Hardy, she argues that Jesus’ statement that none would taste death until the kingdom of God came in power (see Mark 9:1)–long a difficult verse to interpret, as well as other scriptures that imply a quick return of Jesus, are best understood to apply to Jesus’ appearance to the Nephites. She notes that the JST does not adopt this solution but rather softens many passages which seem to suggest a quick return of Jesus, apparently indicating that Joseph Smith (1) recognized that those verses were problematic, (2) did not recognize that the Book of Mormon explained them, and thus (3) posited his own (inferior) solution in the JST. Once again, my hunch is that the JST is sometimes better at identifying problems than it is at offering solutions. This language is going to make some of you uncomfortable, but I’m OK with the idea of a prophet being fallible and struggling to make sense of things and learning line upon line, etc.
Also, there’s an article that I’m having a hard time putting my hands on right now (where IS it?) that discusses the material that Joseph Smith translated twice. It shows, if I recall correctly, that the two efforts at translation are not identical, thus supporting the idea that he was sometimes identifying problems and then speculating about solutions. So I’m more comfortable reading some JSTs as indicators that something in the text triggered Joseph’s spidy sense. This is useful, however: it reveals something about his thought process. (One major thing it suggests to me is that he was not a passive consumer of scriptures, muttering “oh, OK” as he read but rather was reading critically and thus noticing problems in the text.
I’m also not convinced that the JST should get any more weight than anything else Joseph Smith taught. For example, the LDS idea that the priesthood was restored during the scene on the Mount of Transfiguration (I object to this reading for reasons not relevant to this discussion, so I won’t go into them here, but this reading suffices to make my point so I’m going to assume it) stems not from anything in the JST but rather from Joseph Smith’s teachings (TPJS, p158). I don’t think that we should weigh the idea that priesthood keys were given any less reliable just because the JST is silent about it. Unfortunately, simply because the JST (or, some of it, anyway) is printed on gilt-edged pages and between fake leather covers, we assume that the JST is “better” or more official than other things that Joseph taught. That’s an understandable human reaction, but I don’t think it is ultimately defensible.
I am concerned that the JST receives too much uncritical acceptance as “the original text” in LDS circles (Sunday Schools if not academic circles); I don’t think this reflects Joseph’s intention or our theology very well. I’d like to see the JST treated as another data point that needs to be weighed along with everything else: textual variants, translation issues, structural issues, theology, etc., not treated as a trump card.
One more thing: I’ll quote myself from an old post here: “I’m actually fairly amazed at the number of JST changes that make the text easier to read–there is a lot of changing “saith” to “said” and that sort of thing. It makes me think that Joseph Smith thought that it was important that the text be as easy to read as possible for the average person.” One function that the JST might therefore serve for us is as an advocate for the adoption of a new translation of the text that makes it easier for speakers of modern English to understand. Thus, sidelining the KJV in favor of, say, the NRSV, becomes not a betrayal of Mormonism but rather an outgrowth of the work of the JST.