After my recent presentation on using Semantic Web technologies for capturing information about New Testament Names, i had the the opportunity to talk with Tim Bulkeley. My only previous interactions with Tim were through the blogosphere, so i was happy to make the connection in person. Tim mentioned his interest in creating a freely-available hypertext Bible dictionary, and wondered aloud about whether NTN might provide some initial help. After all, he pointed out, most of the entries in a typical Bible dictionary are things with names.

This made a light go on in my head. He’s right of course: the people, places, and other things designated by proper names are the most concretely referential things in the Bible. Named things are important for organizing the narrative: they give us a handle for the characters whose repeated appearances allow us to form a composite view from the individual passages that mention them. When extra-biblical sources are available to enrich our understanding of the background, the names are the links (e.g. the Herod in this passage is the same Herod described in Josephus). The very fact that they’re named tends to help them persist into later church tradition (whether historically based or not).

The light bulb for me came against the backdrop of my own dim sense of the value of NTN: what good are a bunch of OWL statements (even if you make them accessible in a browser like Longwell)? What i realized is that providing the skeleton for a dictionary (albeit a very non-traditional one) was an important use case i’d totally missed.

Suppose you were going to create a dictionary for the New Testament: where would you begin? You’ve got to decide what kinds of things it will cover: a basic background dictionary (Smith and Easton are older, now public domain examples) is quite different from a dictionary of theological terms. So let’s say for starters, you want to talk about people and places: since i’ve already got those in NTN, these form the index. You need to target an audience: my choice here would be the novice reader who’s just trying to understand the basic context, not the advanced academic. Let’s take Barnabas as an example entry: you’d want to say a little about his background, his relationships with other people (especially Paul and Mark), his activities (including his missionary travels). You’d back these comments up with Scripture references for the facts.

Well, many of these things are already in the structure of NTN, but with some important differences:

  • I haven’t included the scripture references. This isn’t too hard for a list of names, given a concordance (though there are some challenges with underspecified names like John that can refer to multiple people). That doesn’t capture all the pronominal references, which (for frequently mentioned people) can be substantial. And of course, no decent Bible dictionary includes all the references except for very sparsely mentioned individuals: otherwise they would overwhelm the other text in the entry.
  • Relationships to others are links without explanatory text (but this could be added)

Suppose you wanted to create your own dictionary, covering the same range of items but with different emphases: perhaps archaeological information about the places, or sociological analysis about individuals, or commentary by later Church Fathers on specific incidents. You’d still want to build on the same foundation of basic facts (it’s hard to start a discussion of Barnabas without some background), and then add your own facts and commentary. This has a natural expression in the Semantic Web framework:

  • You expand the topics of discussion by adding subclasses to the ontology: if the upper ontology is appropriately broad, this is just an extension (rather than starting over with a new ontology)
  • You add relationships with additional specificity or relevance to your emphasis (again, an extension, rather than starting over)
  • You simply omit from your presentation relationships that aren’t relevant: this is easily and mechanically done

This seems to open up a whole range of possibilities for building new resources.