Root searching in the Lexham Hebrew Bible: Possible Bug

I decided to apply Michael Hite's excellent video on root searching (where he used the New Testament) to the Lexham Hebrew Bible. I did a root search on sh-m-r in Genesis, using the analysis option, sorted by root and then lemma. I got the following results:

The root of the tetragrammaton (!?) comes up with two hits for which the lemma is shammah (!?), in Genesis 36:13 and 17. The root mishmar comes up with 7 results, 6 for the lemma mishmar and one for the lemma mishmeret. (These seem accurate hits for the lemma, but the root should be shamar, not mishmar). There are two results each for the roots shemaryahu, shemaryah, and sh-m-ayin, all with the same lemma (shammah) occuring Gen 36:13, 17. There is one result for the root shimron and the lemma shimron.

Finally, there is the almost accurate information at the bottom. Under the root sh-m-r, there is the lemma shamar with 15 results, the lemma mishmar with 6 results, the lemma shammah with 2 results (this is the not accurate part), the lemma mishmeret with 1 result, and the lemma shimron with 1 result.

Since beginning to write this post, I have discovered that the problem is that in the Lexham Hebrew Bible, the word shammah in Gen 36:13, 17 is tagged with lemma 2 shammah and 5 different roots. This is really weird.

I did another search on root:ch-l-m in the Lexham Hebrew Bible and got good results: 34 results on the noun chalom and 14 results on the verb chalam. One quick question, though, even on this: why does the lemma for the verb not have vowel points? We don't want the root to have vowel points, but surely we want verb lemmas to have them. They have them in BDB and HALOT.

Find more posts tagged with

Comments

MJ. Smith

One relevant piece of information: proper names are treated as their own root. I don't know why - merely that a Logos language guru gave me that explanation recently for what struck me as an inconsistency.

Vincent Setterholm

Hello, Tim! In reverse order:

The lemma spellings and homograph numbers in LHB are very similar to HALOT (we strove for consistency marking any accents that weren't on the final syllable, and we tried to correct any mistakes found in HALOT so that we weren't perpetuating typos, but taken as a whole, the lemma spellings are very HALOTish). HALOT does not put vowels on verb lemmas.

The roots analysis for Hebrew and Aramaic owes more to the way HALOT treats the topic than BDB. Rather than trying to put every word under some three letter root which may or may not be a verb in Hebrew or some theoretical proto-Semitic root, words are marked up for what they derive from. So for example, the verb 'to stone, throw stones' is probably derived from the noun 'stone', not the other way around. So in this analysis, 'stone' is the root word. (Yes, I wish we had a better word that 'root' for describing that without being too fussy, since 'root' often means something very specific to Semiticists.)

This approach is also not artificially flattened. If you look at Bethlehemite, it derives from Bethlehem, which derives from Beth and lehem, the latter of which likely derives from laham. So Bethlehemite is a child of all of those words. This opens up some interesting search options for those intermediate 'roots' instead of only having access to the top level.

So in your example with mishmar, since mishmeret derives from it, it is given a root tag, but you'll notice that both mishmar and mishmereth are also given the root tag of SH-M-R, alone with 8 other lexemes. So you can easily search for mishmar and its derivatives, or go one level higher and search for SH-M-R and all its derivatives.

The issue in the case with the tetragrammaton was that Shammah is probably short for Shemaryah/Shemaryahu or something like it which is a theophoric name - that is with and embedded divine element (yah/yahu). In this case just the heh from yah/yahu remains, but it's still there. Looking into this, there is still some clean-up that needs to be done on theophoric names. For example, חֵלֶק, is probably short for a theophoric name, but none of the theophoric element remains, so we probably shouldn't have yhwh tagged as a root there. We cleaned up some, but apparently not all of those types of instances.

Michael Hite

Since beginning to write this post, I have discovered that the problem is that in the Lexham Hebrew Bible, the word shammah in Gen 36:13, 17 is tagged with lemma 2 shammah and 5 different roots. This is really weird.

First, I am glad the video was helpful - at least for NT Greek. I too have been playing with this technique in the Hebrew and to be honest I am not having much luck yet either.

Tim Finlay

Thank you, Vincent, for spending the time to give me such a detailed reply. I may not entirely agree with your decisions but at least I now understand the rationale behind them and can work around the results to get the information I need. I had forgotten that HALOT didn't have the vowels on verbal lemmas (I gave my print copy to a minister in Africa and intend to purchase a Logos copy this coming year). The way you organize the roots is more diachronically oriented than what I was expecting, and that has its value.