The strange aspects of alternative verse numbering

Well folks, this is a curious result for the alternative sort for the silly graph that I did a few days ago. When I sort by the English verse and group by the Hebrew Verse, of course I get a difference - because the Hebrew versification is different from the English versification! Above is the Hebrew count of verse by verse sorted by the English verse number. And below the verses counted and sorted by English verse number with the expected descending values from 42 verse 1's to 1 verse 41.

נשא and some of its objects

To the left are of the more common objects of נשא in this preliminary graph.

Here you can see that נשא (lift up, bear, carry, forgive) is used with ראש (head), with פנים (face or faces), with פשע and חטאת and עון (transgression and sin and iniquity) - עין (eye) especially in the plural and קול (voice).

There are several less common objects: e.g. various kinds of load - from spices to bodies, or lift up with hand meaning a promise, or bearing a name, taking a census, - in fact, bearing just about anything. I think the Psalms may provide a microcosm of the combinations. Let's see...

The nice thing about testing

is that you verify your data - I was perplexed that the high point on this image was only 8. Then I looked at my data and the redundant verse was missing for all chapters greater than 8. So I rewrote the query and bingo! More data - still not 100% comprehensible though.

There's a point at which there is a little dip - and for verses numbered over 42 chapters there will be an equal count until one of the chapters is shorter and then there will be a dip that will not rise again! I see the problem though - it has to do with the odd nature of the Hebrew verse sequence - and the fact that I cannot yet sort the x axis series data by a different field. It must be sorted by the Latin verse field, not the Hebrew one. Another feature will soon be in this product.

Verse count by character and chapter in Job

This is pretty silly - but here is the concentric structure of Job (of course it is concentric, it is a doughnut!)

The red shows the Narrator to whom I have also attributed chapter 28. The blue is Elihu, the purple יְהוָה and the alternating green and yellow the comforters dialogue with Job (green). The scribbles are unreadable verse counts. On this diagram I would prefer to suppress them, but that property will come later. Here is another view - hard to tell who's who here.

If I choose bars you might be able to get the circles.

And here's one just for laughs - I am not sure if it is anything but pastel.
It is the count of each verse number expressed as a percentage.

And the last one maybe is the count of H1N1 inoculations you can get from Job.

נשא in the OT

A beginning of analysis - this graph shows by book the occurrences of נשא in the Old Testament.

Ezekiel likes the word - Kings and Chronicles less so. Only two books do not use it at all - Obadiah and Zephaniah.

But this rich word that means so many things - all somewhat related - how is it used? That will take a better search than I currently have - I am asking around for multi word searches - fine if you have the text in a database, but I have only some parts of it such.

What do you use for searching?

Database technology and the Bible

John Hobbins has outlined his requirements for a research database for the Study of Ancient Hebrew Poetry. (see here)

The overview of recent research provided in the preceding posts illustrates the degree to which parallel structures characterize ancient Hebrew verse at every level of the textual hierarchy. A research database designed to facilitate the study of the phenomenon would ideally have the following features. Components of texts presumed to be poetry would be tagged at the macro-structural, prosodic, semantic, syntactic, morphological, and sonic levels. Each of the six levels, to be sure, is multidimensional.

My own specialty over the past 40 years has been in data analysis and database structures. There is a vast gulf between natural language and database. I have not had a chance yet to do much study of a tagged text such as is in the morphological parsing and tagging from the Westminster Theological Seminary.

Here is another undated statement of requirement:

At present there does not exist a freely available syntactic database and corresponding search engine for the Biblical Hebrew and Aramaic syntactician and discourse analyst. Further, there is no standard, universally available database whereby the scientific community can repeat and verify the results of such study. Such a database would permit the researcher to make comprehensive statements about the behaviour of Biblical Hebrew syntax and textgrammar. Since it is first and foremost a research tool, a fundamental requirement is that the data and analysis are completely accessible and configurable by the researcher to reflect varying theories and improved understanding of the text and theories used to investigate the text.

I am sure the techniques are legion and incompatible. The approach I would take would be to discover the objects that are implied and their relationships - has anyone done this type of analysis yet? Some years ago I did a bit of analysis for a text-based database linking verse with scholar. It would not do as a database for what I am doing now or for what John wants, though some of its entities might be extendable in that direction (see this entity-relationship diagram).

There are two related problems to database. The first is design - the 'right' and 'extendible' set of objects. We know there are such objects because people use books and verses to hit each other with - but this is not necessarily the right starting point. The second problem is loading the data from a verifiable source. Both these are almost intractable given the explosion in thinking about Bible texts that is evident in the noo-sphere today. But if we do find the right structure, people will understand what Fred Brooks wrote in the Mythical Man Month in the '60s: "show me your logic and I will be mystified, show me your data and I won't need your logic to understand you."

January in Seattle

The speaker's list at Bibletech 2008 is posted. I will be speaking there on - ha - guess what!

Visualizing Micro and Macro Structures in Scripture.

I will be presenting some selected diagrams of Hebrew poetry showing both internal and external structure. This presentation’s theme will not be getting things done 'faster', but the sustained learning process that a web-based software framework allows over long periods of time. The talk will also display new ways that the drawing surface expresses both aspects of structure. In doing so, I will explore visually what the ancients would have heard in an aural performance.

If you have any ideas you want to contribute to my talk, please let me know. I will certainly be crediting John Hobbins and his general rule of prosody - and my talk will explore some of the ways in which prosody confirms or is in tension with semantic structures. Right now I am thinking the following topics:

1. Coping with the volume of content -
remembering where we were, seeing where we are going.

2. focussing on detail -
examples of how to learn at the micro level
(see for instance the recent discussion of Psalm 1:1-3 here, or Psalm 2 - does concentric structure establish a frame for an uncertain term? I think you could pick almost any psalm for this process

3. Learning over time -
the tension between panic and hope - e.g. my most recent draft of psalm 37.

So two conferences coming up - SBL to meet people whose names I know and Bibletech to discover more of the potential for technology in learning and community.

I have some specific technological problems in Hebrew, some of which I hope for some light. I have experimented with root derivation, word-counting, and transcription of Hebrew. It is some help with searching but there are several limitations to my algorithms. I expect I will need to develop base forms in a dictionary as well as algorithmic helps for a fuller model. As a more traditional problem space, I would like to see some support for prosodic measurement, at least after the fact.

Technologia

As a programmer, I know that not all technology is nice or bug free. Sorry to have inflicted on you the snappy clicks of that preview thing. I have removed it since they started to advertise links not to my liking.

I never liked the noise - the computer room should be a place of inner quiet - right? :)

Grammar, structure, and form

I guess it should come as no surprise that our comprehension is improved if we perceive the grammatical, syntactic and various parallel structures of text. My marketing staff came across this research into online reading. In the absense of effective oral performance, or as an adjunct to it, the visual perception of structure can help our entry into ancient texts even though they were designed for hearing rather than seeing.

Here's some more about process in my Psalms project.

The application lives on a couple of servers. The properties of the diagram are in an Oracle database and are fully rendered on the web. To my knowledge this combination of diagram + web + database is unique in the world today. The tool supports the thought process as the user determines the presentation of material. Users can work with complexity that goes to the level of fully integrated systems – supporting projects that may occupy multiple authors and users over many years.

My first tests of mapping structure were with the Epistle to the Hebrews . I did this based on the work of Albert Vanhoye (La structure littéraire de l'épitre aux Hébreux). You can ‘see’ the concentric structures and the ‘logic’ of the epistle. Click on the 1:1-4 link on that diagram, there are three items of note:

1. the beautiful structure of the first 4 verses
2. the sentence in 2:1-3 - an eccentric set of circles highlighting the word 'salvation'
3. the shaded quotations from the Psalter.

The conversation recorded in the epistle between the Father and the Son is taken from the Psalms. So I decided to translate the Psalms. I had, after all, a reading of this poetry from the first century and the reading I grew up with from the 16th century, how could I discover the earlier reading without entering the world of translation? As a result, I have created 154 diagrams + a few more that are essentially a production environment for testing the product.

The translation of the Psalms is a long term project like a performance measurement project. It uses queries, charts, indicators and alerts just like a performance project would - only my queries and charts are to data that is ‘in the diagram’ rather than in a document or spreadsheet somewhere. (For example, the data displayed in indicators, alerts, and queries on the table of contents.) I can even get a node to alert me when a word is used twice in the same Psalm - showing me possible keywords in the Psalmist's intention for the poem.

One thing I lack at the moment is automation of Hebrew parsing - but I am experimenting with ways of doing it. The combination of Hebrew + English could help with recognition of the various aspects of the Hebrew affixes. I would love to hear from others who are approaching this problem in similar or other ways.