Morphological Search Omissions

After being told by an OT professor that the Logos morphological search is inferior to Even-Shoshan's New Concordance of the Old Testament, I did a search both on Logos and with Even-Shoshan, for the word נוח. I searched BHS SESB under the SESB Morphology and BHS/WIVU with the WIVU Morphology and got the same results with both.

There were six instances of נוח found in Even-Shoshan that were not found by Logos. Four of them (Esther 9:16, 17, and 18, and 2 Chron 6:41) showed up under נוֹחַ. Although this is the form for the noun "rest," it is also the form for the infinitive absolute of the verb; it should show up in a morphological search, especially since the form in question is identical to the lemma ( נוח) apart from the pointing.

One (Isaiah 7:2) was parsed under the root נחה by Logos, even though some English translations parsed and translated it as "rest" (נוח). If a morpheme can be found under two lemmas, it should come up during a search for either lemma.

The most disturbing was Habakkuk 3:16, were Logos parses it under נוח but did not include it in the results of my morphological search.

I realize that a discrepancy of six passages out of 138 isn't the end of the world, but I would like to know how to prevent this from happening in future searches. I would prefer to use Logos instead of other options for searches like this.

Any advice on how to get more complete results on a search like this?

Find more posts tagged with

Comments

Vincent Setterholm

The disturbing Habakkuk 3 passage has a different homograph number, so it is analyzed as a different word in both the databases you searched. At some point we probably need to support wild-carding the homograph number for people who want to ignore that portion of the analysis in their searches. I know we did some work on that when Logos 4 first came out, but I don't think that feature was ever finished.

There are plans to add root searching data that would make it easy to find a verb and all the other forms derived from it, which would pick up the first four instances you mentioned. Stay tuned for more information.

The Isaiah 7:2 instance is interesting. There probably is some value in tagging things that are either ambiguous or that have a strong translation history of reading with an alternate analysis with the alternate views. With current technology, we'd have some issues with lexicon look-ups (our system does a nice job of ignoring differences in spelling and homograph numbers to get to the right lexicon articles, but changing things up so that each segment could theoretically go to multiple lexicon articles would be a significant project).

Both the WIVU and SESB Bibles make a lot of use of HALOT as a primary lexical tool. So I think you'll find that if you use these Bibles in conjunction with HALOT, you'll be able to have some confidence that you're finding everything. HALOT lists alternate analysis options for most of the places you'd need that, and it lists derivative words of roots making it easy to find things that newer lexicons tend to split which Evan Shoshan was inclined to join.

John Coatney

https://community.logos.com/discussion/comment/402208#Comment_402208

Thank you, Vincent; that's helpful (especially the HALOT part).

If you get a chance, could you tell me more about the homograph number (what is it? what does it do?)? If it's easier to link to another resource instead of explaining it, that's fine, too.

Thanks again!

Dave Hooton

As regards parsing this word:

5662 ...
נוח (nwḥ): see 5663

5663 I. נוּחַ (nûaḥ): v.; ≡ Str 5117; TWOT 1323—1. LN 23.78–23.87 (qal) rest, i.e., be in a state or process of restoring physical energy in the body (Ex 20:11); (hif) rest (Ecc 11:6); (hof) rest (La 5:5+); 2. LN 85.32–85.54 (qal) set, i.e., place or put an object in a place (Jos 3:13); (hif) put, place (Ge 2:15); (hof) be set (Zec 5:11+); 3. LN 85.65–85.66 (qal) rest, settle, remain, i.e., leave in a place for any period of time (Ge 8:4); (hif) leave, keep (Ge 42:33); 4. LN 22.42–22.47 (qal) rest, i.e., be in a state of favorable circumstances (Ne 9:28); (hif) give rest (Dt 3:20); 5. LN 34.22–34.30 (qal) be in alliance, formally, rest, i.e., be in an association, as a figurative extension of a favorable circumstance, and lack of hardship (Isa 7:2); 6. LN 13.1–13.47 (qal) wait, formally, rest, i.e., continue to be in a certain state until an expected event (1Sa 25:9; Hab 3:16); 7. LN 16 (hif) lower, formally, rest, i.e., place limbs in a lower position, and not raised high (Ex 17:11), note: for n.masc. in Est 9:17, 18, see 5665; note: in Isa 7:2, see 5698; Eze 41:9, 11, see 4965; note: for another parsing of Hab 3:16, see 5664; note: some parse Nu 23:7; 1Sa 22:4; 1Ki 10:26; 2Ki 18:11 as 5697

5664 II. נוּחַ (nûaḥ): v.; ≡ Str 5117; TWOT 1323—LN 25.135–25.145 (qal) groan, lament, wail, i.e., make noise in anticipation of a future welcome event, implying a current emotional suffering (Hab 3:16+), note: oth parse as 5663

5665 נוֹחַ (nôaḥ): n.masc.; ≡ Str 5118; TWOT 1323—LN 85.67–85.85 dwelling, abode, resting place, i.e., a place in which one lives or resides, with a focus on the place of dwelling (2Ch 6:41+), note: in context it refers to the Temple, see also 4957; note: for another parsing in Est 9:17, 18 as qal inf. abs., see 5663

Swanson, J. (1997). Dictionary of Biblical Languages with Semantic Domains : Hebrew (Old Testament)

As regards Logos use of Hebrew morphologies:

there are 3 main ones (BHS Westminster 4.2, BHS/WIVU, and Andersen-Forbes)
Lexham Hebrew-English Interlinear, The Hebrew Bible Andersen-Forbes & English Reverse Interlinears (OT) use Andersen-Forbes

For the word in question:-

A-F often has a different lemma to the other morphologies
the word is parsed as a verb by one and a noun by another in the passages you quote
I couldn't reliably get a lemma unless I knew some passages and could use the context menu. e.g. in Morph Search for נוח
- BHS/WIVU Hebrew suggested lemma:נֹוחַ for "rest" but I got zero results for the search
- Westminster Hebrew suggested lemma:נוח.1.v for "rest" and I got 140 results
- A-F Hebrew suggested lemma:נוחה for Nohah!

Whilst I know something about Searching, the Hebrew search often leaves me puzzled.

John Coatney

https://community.logos.com/discussion/comment/402218#Comment_402218

Thanks, Dave!

I take it that the homograph numbers are those set down in Swanson's Dictionary?

I have BHS/WIVU, Anderson-Forbes, and SESB morphologies. Hypothetically speaking, if I were to get Westminster as well, might that increase the number of results I get? Or is it more dependent on the homograph number?

My original lemma search began with Isaiah 23.12 (יָנ֥וּחַֽ); I right-clicked it and clicked lemma, then morph search, and this is what popped up in my search window: "lemma:נוח:1". Using either SESB or BHS/WIVU I get 138 results in 136 verses.

Vincent Setterholm

https://community.logos.com/discussion/comment/402212#Comment_402212

If you get a chance, could you tell me more about the homograph number

Sometimes you have two words that are spelled the same, but have different meanings. Look at בַּד, for example. It can mean 'part, portion', or a type of linen or some sort of carrying pole, some sort of boasting or idle talk, or a type of priest. For the most part, these are not different definitions of the same word, but rather different words that have different etymologies. The homograph distinctors split these words that sound the same into their different lexical entries. In many lexicons, this is just a number (roman numeral or otherwise). In some databases and lexicons, there is also a part-of-speech element.

The homograph numbers in each Hebrew database don't necessarily match the roman numerals in any particular lexicon. Though you'll find a fairly close match between SESB/WIVU and HALOT, it is not exact (though it should be an exact match to the glossary that came with the SESB).

Dave Hooton

https://community.logos.com/discussion/comment/402223#Comment_402223

I take it that the homograph numbers are those set down in Swanson's Dictionary?

The Swanson lexicon gives different meanings based on the Semantic Domains of Louw-Nida (the LN numbers) but the homograph identities depend on the morphology of the resource.

Hypothetically speaking, if I were to get Westminster as well, might that increase the number of results I get? Or is it more dependent on the homograph number?

It would increase the chances of getting a search result as described in my previous post! Ultimately, it will reflect a different parsing of the words. The homograph should be obtained from the Context menu of a word in the specific bible as the same homograph will not necessarily work with another morphology e.g. from the word in Is 23.12:-

Westminster: 140 results from lemma:נוח.1.v
WIVU: 138 results from lemma:נוח:1.V
A-F: 70 results from lemma:נח:1

The WIVU lemma will not work in Westminster (BHS/WHM 4.2), even though the difference is .V as against .v

Lee

After being told by an OT professor that the Logos morphological search is inferior to Even-Shoshan's New Concordance of the Old Testament,

Funny thing is, I was just told by a B.H. expert very much the same thing. And although I had the alacrity to tell him that computer searches are fast and accurate now, then and there I was not prepared to do a one-on-one against the venerable Even-Shoshan. I might have gone one-on-one using another tool that I have, but that's another story for another forum.

There are plans to add root searching data that would make it easy to find a verb and all the other forms derived from it, which would pick up the first four instances you mentioned. Stay tuned for more information.

I'm wondering if there are any updates?