Tag: hebrew

Strongs Numbers & the Problem of a Universal Index

With the advent of masses of dig­i­tal resources inte­grated by Bible soft­ware user inter­faces, the prob­lem has arisen, How shall all these resources be inte­grated together? Typ­i­cally, the uni­fy­ing ele­ment has been the bib­li­cal text itself, espe­cially the text in the orig­i­nal Hebrew, Ara­maic or Greek.

Actu­ally, the prob­lem is not unique to the Infor­ma­tion Age, but has been addressed since at least the mid-​​19th cen­tury. James Strong (1822−1894) com­piled The Exhaus­tive Con­cor­dance of the Bible (1890). One of his goals was to align the Eng­lish trans­la­tion of the Bible with the orig­i­nal lan­guage. He chose to do this at the word-​​level, which fit one of his other objec­tives: an exhaus­tive con­cor­dance of the Eng­lish text, namely the King James ver­sion. It took him and more than 100 col­leagues 35 years to com­plete this mon­u­men­tal task — and with­out com­put­ers! His num­ber sys­tem rec­og­nizes 8,674 Hebrew and 5,523 Greek lem­mas. Let me focus on the Hebrew side of things.

The lex­i­cal schol­ar­ship upon which Strong’s Hebrew dic­tio­nary depends is that of Wil­helm Gese­nius. In 1833, Gese­nius pub­lished a Latin work, Lex­i­con Manuele Hebraicum et Chal­daicum in Vet­eris Tes­ta­menti Libros. There were suc­ces­sive edi­tions until the end of the cen­tury, when BDB (Brown Dri­ver Briggs, A Hebrew and Eng­lish Lex­i­con of the Old Tes­ta­ment, 1891–1905) became the new schol­arly stan­dard. Strong’s Con­cor­dance most cer­tainly used Gese­nius for the lex­i­cog­ra­phy of the Hebrew Bible.

BDB reflected the new dis­cov­er­ies in the Mid­dle East dur­ing the lat­ter half of the 19th cen­tury and, impor­tantly, the rise of new meth­ods of the study of lan­guage: struc­tural­ism (Saus­sure), descrip­tive (Bloom­field, et. al) and com­par­a­tive lin­guis­tics, that is, using other Semitic lan­guages to help puz­zle out the mean­ings of Hebrew words and expres­sions. But in this hey­day of archae­o­log­i­cal dis­cov­ery, even BDB was quickly superceded by dis­cov­er­ies (in 1929 and later) in Pales­tine, espe­cially at Ras Shamra, ancient Ugarit. There a huge repos­i­tory of clay tablets were dis­cov­ered, includ­ing those using an alpha­betic writ­ing sys­tem to record a lan­guage that is closely related to Hebrew.

With such a wealth of new mate­r­ial, Hebrew lex­i­cog­ra­phy changed dra­mat­i­cally, with new lem­mas pro­posed and old lem­mas dropped. Addi­tional schol­ar­ship reas­signed lem­mas to spe­cific occur­rences in the Hebrew text. If Strong com­piled his Hebrew and Greek dic­tio­nar­ies and asso­ci­ated list of lem­mas today, it would be quite a dif­fer­ent list, includ­ing the assign­ment of those lem­mas to words in the text. And Strong might not have cho­sen the King James Ver­sion to concord.

Con­nect­ing dig­i­tal resources

Strong’s Con­cor­dance indeed became a stan­dard and is widely used even today, being reprinted reg­u­larly. Other lex­i­cons, word books and study Bibles included Strong’s num­ber­ing, even when the text was no longer the King James Ver­sion. Hence, when soft­ware devel­op­ers began to write Bible study soft­ware, the need arose for an index between the Hebrew text and var­i­ous other resources, Strong’s Num­bers were a nat­ural choice. There was the con­sumer demand for a tool famil­iar to them from printed Bible study resources. Also, there was a prac­ti­cal eco­nomic con­cern on the part of devel­op­ers: the work was mostly done for many resources. Fur­ther, if one needs to cre­ate a uni­ver­sal index, how does one chose? Strong’s num­bers seemed to be a nat­ural choice, already “uni­ver­sal” in some sense. Finally, there was the require­ment that the index never change. If the index changes, the link­ing to other resources is bro­ken and it costs time and money to fix it. The impli­ca­tion of this is that we already cer­tain that Strong’s 160-​​year-​​old lemma­ti­za­tion of Greek and espe­cially Hebrew is com­plete, cor­rect and need not ever change.

And that is how I got involved with this ques­tion. At the Groves Cen­ter we main­tain a lin­guis­tic data­base known as the West­min­ster Hebrew Mor­phol­ogy (WHM), which, among other things, offers a lemma for each and every one of the approx­i­mately 480,000 mor­phemes found in the Hebrew Bible. The lemma assign­ment is based upon the lat­est schol­ar­ship that we have, but in the final analy­sis is a deci­sion based upon our own judg­ment. We never con­sulted Gense­nius’ Lex­i­con, never mind Strong’s sys­tem of lem­mas. Given its age, it never occurred to us.

Imag­ine our dis­may when we were asked why we didn’t have Strong’s num­bers assigned to our lem­mas. Such a map­ping is impossible:

  • Some of Strong’s lem­mas don’t exist in the WHM.
  • There are new lem­mas in WHM; what Strong’s num­ber should they have?
  • Some of Strong’s lem­mas have been split into dif­fer­ent mean­ings (homonyms, for exam­ple) in the WHM.
  • Many of the lemma assign­ments to indi­vid­ual words in Strong’s have changed in WHM.

These dif­fer­ences make using Strong’s num­bers as a uni­ver­sal index for inte­grat­ing dig­i­tal resources prob­lem­atic and just plain wrong. The demands of the con­sumer and prag­matic and eco­nomic con­cerns must be resisted; else, we are per­pet­u­ally stuck in the mid-​​19th cen­tury of bib­li­cal scholarship.

A fresh look at resource integration

Let’s step away from the ques­tion of the suit­abil­ity of Strong’s num­bers for resource inte­gra­tion, and look at the issue of inte­gra­tion afresh. Strong’s num­ber­ing of Hebrew and Greek lem­mas is only one pos­si­ble solution.

Two pos­si­ble — and prac­ti­cal! — solu­tions come imme­di­ately to mind: search engines and topic maps. These tech­nolo­gies were responses to the need to inte­grate resources that are dynam­i­cally chang­ing and are seman­ti­cally diverse. One is not lim­ited to lem­mas but can index any arbi­trary string. Topic maps allow for more than one way to iden­tify a sub­ject. They han­dle ontolo­gies (as under­stood by com­puter sci­en­tists; philo­soph­i­cal ontol­ogy is some­thing else) quite well. Con­sider that com­pet­ing lemma­ti­za­tions such as Strongs and WHM are com­pet­ing ontolo­gies for the vocab­u­lary of the Bible in the orig­i­nal lan­guages. In this sce­nario, it doesn’t have to be “either Strong’s or WHM”, but “both-​​and.” Then we allow the user to decide what is most valu­able or cor­rect. This solu­tion to the prob­lem is bet­ter because both the num­ber and inter­nal con­tent of resources one inte­grates can change freely as desired.

I am encour­aged to see signs that Bible soft­ware is grad­u­ally evolv­ing in this direc­tion. As I see it, as the num­ber of dig­i­tal resources increases, the above two solu­tions become ever more compelling.

Exemplary Exegesis

As some­one who focuses on inter­pret­ing the Hebrew Bible, I face the prob­lem of “insuf­fi­cient data.” So much of the his­tory, cul­ture and lit­er­a­ture of the Ancient Near East has been lost, mak­ing the prob­lem of inter­pret­ing the ancient text very prob­lem­atic. The temp­ta­tion is to fill in the blanks to a greater degree than is sup­ported by hard, empir­i­cal data. Bible schol­ars for some rea­son hate to admit that they sim­ply don’t know and that we can’t know the answers to some ques­tions until and unless new evi­dence comes to light. So there is a lot of spec­u­la­tion and out­right fic­tion being writ­ten about the Hebrew Bible.

It is, then, a plea­sure to run across exam­ples of hon­est grap­pling with the data as it actu­ally exists. The most recent exem­plar are two related blog posts, here and the ear­lier Part 1 here, on the vexed ques­tion of the iden­tity of “Daniel” in Ezekiel 14:14 & 20. Does it refer to Ezekiel’s con­tem­po­rary whom we know from his epony­mous book, Daniel? Does it refer to the Canaan­ite leg­endary hero, dn’il, as nar­rated in the Ugaritic texts? Or per­haps to some­one whose exis­tence and biog­ra­phy has since been lost to us?

I’ll leave it to you, dear reader, to peruse the details of the debate. What is exem­plary is that:

  • the extent and lim­its of the real-​​world evi­dence is described
  • the writer grap­ples with how far that evi­dence can be pushed
  • he refuses to go beyond what can be justified

The reader comes away with a clear under­stand­ing of the issues, the evi­dence, and the pos­si­ble con­clu­sions that may be drawn. Most of all, it’s hon­est. This is the way bib­li­cal schol­ar­ship (or any schol­ar­ship, really) should be done.

I’ve added this blog to the blog list at the bot­tom of the page (see Fur­ther Infor­ma­tion).

Web babeling

I had a request today about the best way to get Greek and Hebrew into web­pages. In doing the research on this request, I came up with some excel­lent resources I want to share with you.

The­o­ret­i­cally, there are three ways one can do this:

  1. Use HTML enti­ties for Greek and Hebrew
  2. Use a css style (here’s one way of doing it)
  3. Set your web­page encod­ing to utf-​​8 and sim­ply type or paste in Uni­code encoded text

Num­ber 3 is my pre­ferred way of doing it. And it’s sim­ple. In your web­page, force utf-​​8 encoding:

<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8 ">
...
<head>

And you sim­ply type or copy and paste your Uni­code text (of what­ever language).

Hebrew is more com­pli­cated because of the right-​​to-​​left issues, espe­cially when embed­ding Hebrew into Eng­lish text, or vice versa. A quick exam­ple. Sup­pose you want to write and display:

3 Hebrew let­ters (אבג) - aleph, beth, gimmel

The code you put into your web­page to get this would be:

<p>3 Hebrew let­ters (<b lang=“he”>אבג</b>) -<b> aleph, beth, gim­mel </b></p>

This is only one way to do it. There are oth­ers, some more appro­pri­ate to a par­tic­u­lar sit­u­a­tion. If you don’t get good results, it could be that you don’t have a Uni­code font installed for Greek or Hebrew; or else the text itself is not in Uni­code. Make sure that the text you are copy­ing is in utf-​​8 encod­ing. Or that the key­board you are using is pro­duc­ing utf-​​8 encoded characters.

There is a very, very cool web­site that I just dis­cov­ered dur­ing the research for this ques­tion: Tips for Devel­op­ing Non-​​English Web Sites, from the PennState Teach­ing and Learn­ing With Tech­nol­ogy web­site. There are generic tips and spe­cific instruc­tions for a wide range of lan­guages. Highly rec­om­mended. It’s going into my list of links. Pages for Greek and Hebrew are very helpful.