Monday, December 12, 2011

Searching for Greek Semantic Domains Using Louw-Nida in Accordance, BibleWorks, and Logos

In the previous post, I indicated how to go about searching for Greek roots in order to get a broader view of how a particular concept is presented by an author. While Accordance makes such a search quite simple (it's harder in Logos and would take even more work in BibleWorks), I also noted that what a person may really want is not a search based on word roots but one based on a particular concept or idea. What this means is that we really want to search on semantic domains. In my work, I want to do a domain search as often or more as I want to do a root search. For the Greek New Testament, this means that I want to use Louw-Nida's Greek-English Lexicon of the New Testament: Based on Semantic Domains. (This resource is standard in BibleWorks and is standard with any of the Greek NT packages in Accordance or Logos. Not too much has changed since an earlier review I did.) 

In the preface of this excellent lexicon it states its purpose:
This Greek New Testament lexicon based on semantic domains has been designed primarily for translators of the New Testament in various languages, but biblical scholars, pastors, and theological students will no doubt also find this lexicon of particular value, since it focuses on the related meanings of different words.
What Louw and Nida did was take every word in the NT and assign it to groupings of concepts (= semantic domains), so that one could compare words that are related conceptually but not lexically (i.e., by word roots). To continue the goal of the previous post where we were searching for roots related to κρινω in James, here is what a search for κρινω looks like in Accordance, BibleWorks, and Logos

Accordance: The search brings up the results (use the Show Paragraph), and the Browser can be expanded to show the organization.

BibleWorks: The search actually brings up the first instance of κρινω in the lexicon, but the middle pane on the left provides access to the other occurrences. 

Logos: The search actually just brings up the entry listings for κρινω in the lexicon, and from here you click on the entry that matches the concept you desire.

You will also see in the screenshots that I have located domain 56, "Courts and Legal Procedures," and its subdomain E, "Judge, Condemn, Acquit," which includes fifteen entries: 56.20-56.34. If it is the concept of "judging" (and not just the words whose roots are κρινω) we want to find in James, then what we want to do is search for all the words in this subdomain. 

So, those screenshots show me what group of words I'm looking for.
How do I go about conducting a search for words in a domain in the Bible software programs?

Accordance: Apparently it is not possible in Accordance to search the Greek NT by Louw-Nida domains. (Someone may have found a way...)

BibleWorks makes it quite easy to conduct Louw-Nida domain searches. First select a morphology Greek version (e.g., BGM, BNM, GNM...) and then right click in the command line. The popup offers an option of "Insert Louw-Nida Domain Code." Clicking on it opens a box with the LN domains, and it is easy to search for domains with a particular word or to browse through the domains to find the one you want.
Additionally, if you want to search multiple (sub)domains, you can use an OR search and simply type in, using angle brackets, the additional ones you want. As usual, BW is quite fast. A search for all the words in the NT in subdomains 56.20-56.34 took less than a second and returned 332 hits.

Logos isn't quite as straightforward as BW, and there are actually a couple ways of obtaining the desired results.
  • You can use a Syntax search and use the Lexham Syntactic Greek NT. When the search dialog opens and you start a new query, first add a Word. Among the options that are available, you can search for LN Domains (e.g., domain 56 on "Courts and Legal Procedures"), Subdomains (e.g., subdomain 56E on "Judge, Condemn, Acquit"), or Articles (e.g., 56.20 on "make legal decision). Unfortunately, Logos does not display any of those headings, so you need to first have consulted LN to find out which domain or subdomain or article you want. You can mix search categories using an OR command. As you can see above, a search for an article OR a subdomain in James took less than 5 seconds (a search on just subdomain 56E in James took less than a second), and Logos does return a nicely highlighted display of results in both the Greek and another version of your choice.
  • You can also use a Bible search using one of the Lexham Greek NTs and simply type in the command line in angle brackets the domain range you want (along with boolean operators for multiple ranges). As you can see below, a search for domains 56.20-34 in James took about a second. The advantage of conducting your search this was is that you have the option of displaying the results in a Grid, by Verses (with a Greek / English parallel if you wish), Aligned, or Analysis view as shown in the graphic.

A couple other related matters... 
BOTTOM LINE: BibleWorks has the most straightforward way of searching on Louw-Nida domains. Logos has more options in searching and especially in its display of results. Accordance is unable to conduct this kind of search.

Next: What about domain searches in Hebrew?

Thursday, December 8, 2011

Searching for Greek Roots in Accordance, BibleWorks, and Logos

This has turned out to be a somewhat lengthy adventure for me. It started with reading Rubén Gómez' posting on "Finding all the forms of a Greek word in Accordance" which was in turn a response to Michael Hite's posting on "Finding all forms of a Greek word with Logos 4." As Gomez noted, the method demonstrated for Logos 4 was rather convoluted. Furthermore, it's simply inadequate and misses a number of hits, because it fails to include words that have prefixes attached to the roots or instances where the root word has forms that are not spelled quite the same. In the example provided in both videos, the task was to find all the words sharing the root of κρινω = judge. The Logos demonstration found, by picking words alphabetically close to κρινω: κρινω, κριμα, κρισις, κριτηριον, κριτης, κριτικος. What it missed, however, were these words that also occur in James based on that root: διακρινω, αδιακριτος, ανυποκριτος. Gomez demonstrates the very simple and elegant way that Accordance found all those words which is accomplished merely by right-clicking on a word and choosing "Search for root." The results are very nicely displayed as well. 

How does Accordance accomplish this task? They have evidently compiled lists of cognate word groups. These are simply the collection of words based on the same root. Such collections of cognate groups have been regularly used as aids to vocabulary memorization, and you can find them in Metzger's Lexical Aids for Students of New Testament Greek, Robinson's Mastering NT Greek, Van Voorst's Building Your NT Greek Vocabulary,or Trenchard's Complete Vocabulary Guide to the Greek NT. (Incidentally, I am fairly certain that Accordance and Logos referred to Trenchard's listings, because, for example, he--inaccurately, I believe--includes νηστευ· forms with the εσθι· root and εραυν· forms with the  ερωτα· root. A real linguist, which I am not, may be able to confirm that or not.) By using such a listing, Accordance has made it possible to search for all the words in a group when you search for any one of them based on its root. (Searching by root was a feature added in version 7. Matters are similar but not the same with Hebrew and its system of 3 consonant roots.)

As far as I can tell, Logos is not able at this time to conduct such a root search directly. [UPDATE: 2012.11: It is now easily possible in Logos 5 by simply right-clicking on a Greek (or English) word and choosing "Root" in one of their appropriately tagged versions.] Fortunately, however, it is able to do so with a little work. What you need to use is the The Lexham Analytical Lexicon to the Greek New Testament. (It's included in their Silver library or higher or for purchase separately. $30 list) Each entry in this lexicon includes the word, a gloss, a list of cognate words, links to usages in Louw-Nida, and a listing of all the forms in the NT. 

Unfortunately, however, there is no simple way to search on that cognate list. (I'm thinking if Logos folks read this, it wouldn't be a particularly hard thing to implement.) What I did was to copy the list, paste it into a word processor, then find/replace every instance of ", " with " OR lemma:" and then touch up a bit so that ultimately I could use a Morph search and paste in the full list of κρινω cognates: 

That search will generate the results you want, and Logos returns them with the attractive options of viewing it as Verses (with a parallel English version if you wish), Aligned, or the Analysis view which provides all sorts of helpful information.

Can BibleWorks conduct such a root search? No, unless you manually typed in all the roots you would have had to gotten from elsewhere. You can get pretty close, however, by searching for *κρι* in one of the Greek morphological versions, but that really only works because the κρι stem is rather regular. (Note that in BW, the command line search would actually be: .*κρι* The initial asterisk allows for prefixes, and the final asterisk allows for various suffix-type endings. A similar sort of search could be done in Logos using a morph search with lemma:*κρι* OR lemma:*κρί* in the command line. I had to specify both an unaccented and accented iota to get all the results.) For other roots, the spelling is irregular, and so you can't count on a wildcard search. (E.g., γινομαι shares its root with forms of γιν· γεν· γενν· γον·) 

Bottom line: If you want to do a word root search, Accordance is the tool you want. Now why exactly would you want to do such a search? It does give you a broader picture of how an author is using words sharing the same root. In the case of James, it helps you gain a broader picture of the matter of 'judging' throughout the book. It also demonstrates a bit of James' eloquence in pairing αδιάκριτος (=impartial) and ανυπόκριτος (=not hypocritical) together. You wouldn't have caught that pairing in any English translation. But if our real concern is to find all the instances of a concept, then we are not necessarily looking for just a word root but for any words related to that concept. For that, we want to use semantic domains, and in the next posting I will describe the use of Louw-Nida's Lexicon of the NT Based on Semantic Domains.