missing two lemma hits

Kristin · 2025-03-25T07:07:01+00:00

There was an error rendering this rich post.

I have a word that I want to review (see all the lemma hits), and I decided to use the BHS to keep the familiar 2Chr ending.

Anyway, I know this word has 159 hits, but the BHS is only showing 157 hits. So I opened the BHW and it is also showing 157 hits.

Am I doing something incorrect? Or does anyone have an explanation?

Thank you.

Bildschirmfoto 2025-03-25 um 01.59.20.png

Find more posts tagged with

Accepted answers

All comments

MJ. Smith

What morphology is used on the top text; the bottom uses the SESB morphology.

Donovan R. Palmer

Hey @Kristin

I tried to investigate this with Accordance, but it keeps crashing… I am running the latest version, so I will not fight it.

I have the same question: what is the morphology of the text on each platform?

Anon

@Kristin

Or does anyone have an explanation?

Could it be the following: the HMT search includes two bracketed results and if excluded you have 157 hits matching the BHS and BHW count.

Francis

Sometimes different morphologies make different tagging choices where a form can be ambiguous. For instance, I was looking recently at Ephesians 4:6 where the form of πας could be neuter or masculine. The morphology has to pick one. This concerns gender not lemma but it can happen with lemma in Hebrew. This may or may be the explanation for your particular case, but it is at least a distinct possibility.

If you have ability to search for the root in Accordance, it'd be interesting to compare with the root count in Logos' BHS (or even more confusing). As for the bracketed hits in HMT, you could do a search excluding these two verses to test your hunch (or simply compare searches in HMT and BHS outside of Ezekiel). Either way, you can try to isolate with certainty at least one case where there is a difference between the two counts and then investigate whether the lemma is debated or if it is something else like Qere vs Ketiv, etc.

Kristin

What morphology is used on the top text; the bottom uses the SESB morphology.

I have the same question: what is the morphology of the text on each platform?

Hi @MJ. Smith and @Donovan R. Palmer, the HMT-W4 is using the Groves-Wheeler Westminster Hebrew Morphology and my text says v.4.22. I hope that answers the question alright.

I tried to investigate this with Accordance, but it keeps crashing… I am running the latest version, so I will not fight it.

Hi @Donovan R. Palmer, thank you for trying to check this, and I am sorry it keeps crashing. :(

Could it be the following: the HMT search includes two bracketed results and if excluded you have 157 hits matching the BHS and BHW count.

Hi @Anon, that is a really good idea, but this is apparently not the case since the "bracketed" words are higher in Logos, and I can attach a screenshot. (Please ignore the yellow highlighting, it is regarding something else with my work).

Bildschirmfoto 2025-03-25 um 09.20.55.png

Sometimes different morphologies make different tagging choices where a form can be ambiguous. For instance, I was looking recently at Ephesians 4:6 where the form of πας could be neuter or masculine. The morphology has to pick one. This concerns gender not lemma but it can happen with lemma in Hebrew. This may or may be the explanation for your particular case, but it is at least a distinct possibility.

Hi @Francis, ya, I agree it must be something like that, but I need to isolate the specific issues.

If you have ability to search for the root in Accordance, it'd be interesting to compare with the root count in Logos' BHS (or even more confusing).

So I did a root search in Accordance which produces 661 hits, but I can't figure out how to do it in Logos. If I try, it limits it to a feminine noun, which is literally missing the point of a root search, so I think I am not doing it correctly. I will post a screenshot of that as well.

Bildschirmfoto 2025-03-25 um 09.36.55.png

As for the bracketed hits in HMT, you could do a search excluding these two verses to test your hunch (or simply compare searches in HMT and BHS outside of Ezekiel). Either way, you can try to isolate with certaintyat least one case where there is a difference between the two counts and then investigate whether the lemma is debated or if it is something else like Qere vs Ketiv, etc.

Hopefully my screenshots have started this investigation, but I am honestly feeling sort of concerned. As we see from my screenshot, there are more bracketed words in Logos, so the count should be higher, not lower, which means two are added there, but then four are missing somewhere else. So now instead of tracking down two missing hits, there are four missing hits, unless I am misunderstanding something.

DMB

Just repeating advice (no offense) but you need to take bigger screen shots … hard to see what you're using. Smiling.

Unfortunately, BHW in Logos doesn't have roots. They use the 'root' symbol from mathematics, in the right-click menu.

Roots show up in the LHB, and LHI (and reverse interlinears but forget them, for counts). Personally, I'd do the morph search comparison by hebrew book-groups, and then by book, and then by chapter (to find it). Roots will go crazy.

Kristin

Just repeating advice (no offense) but you need to take bigger screen shots … hard to see what you're using. Smiling.

Hi @DMB, thanks for letting me know. Tbh, I thought I had been doing that. As an example, is there something specific lacking in the screenshots above? (If so, it would help me understand better what is off about them).

Unfortunately, BHW in Logos doesn't have roots

😐️

they use the 'root' symbol from mathematics, in the right-click menu. Roots show up in the LHB, and LHI (and reverse interlinears but forget them, for counts).

Well, apparently the BHS doesn't have the root thing either, as that is where my screenshot had been. So all this BHW vs BHS, and neither have basic root function? I remember someone had told me before that the Lexham should be avoided for scholarly research (I frankly don't remember which thread it was, or who said that), but whatever the case, I just ran the search in the Lexham and the root has 449 results? (I guess that is why it isn't used for scholarly work).

So then BHS and BHW don't have root functionality, and while Lexham does, it doesn't actually find the roots. So I am attaching a screenshot, and also of the root in Accordance (with a bigger screenshot).

Bildschirmfoto 2025-03-25 um 10.35.13.png

Personally, I'd do the morph search comparison by hebrew book-groups, and then by book, and then by chapter (to find it). Roots will go crazy.

Would you mind clarifying this a little? I am not entirely sure what you mean.

DMB

@Kristin if you look at the image above my post, it's not easy figuring out which hebrew Bible is being used. And you're right, BHS doesn't have roots … I should have known that without checking … GBS product.

I'm assuming you're on the trail of roots now (differences). My comment was back when you were on the trail of morphs … when comparing two document totals, and no easy way to find the difference, you use the old 'binary search method' … repeat the search comparison at successively smaller chunks.

Francis

@Kristin perhaps this could be another method:

Assuming that the list of verses with results can be copied or exported from Accordance, make a passage list in Logos by using the "copy from clipboard" option.
Export your Logos search results in BHS to another passage list.

You can then compare the two lists. I think there may be a way to isolate the two waldo's using the merge options but I'm not sure and don't have the time to look into it just now.

Kristin

@Kristin if you look at the image above my post, it's not easy figuring out which hebrew Bible is being used. And you're right, BHS doesn't have roots … I should have known that without checking … GBS product.

Hi @DMB, Thank you for clarifying this, and ya, I see the problem with my screenshot now. I will try to keep that in mind going forward.

I'm assuming you're on the trail of roots now (differences).

Actually I don't know if I would say that. While I do need to sometimes look at roots, I had actually just done so since @Francis had said it would be interesting to compare the root in Accordance to the BHS root. So I got the Accordance root, but then ran into issues with the BHS root, as we apparently can't even search for the root. :( Maybe @Francis can comment regarding the root issue, as it seems like a major problem for BHS to not even have the function, and for Lexham to give radically different root counts, enough that it seems like an error. So what do people here do if they need to find a root?

My comment was back when you were on the trail of morphs … when comparing two document totals, and no easy way to find the difference, you use the old 'binary search method' … repeat the search comparison at successively smaller chunks.

I see, thank you for clarifying this. Given how time consuming that sounds, I think for this particular word I better just get back to work in Accordance. The whole idea was that I was going to try to do my next word in Logos and see how it went, but the first word I tried it with is ironically giving a different lemma count from Accordance's equivalent, and since it seems like it might take considerable time to find the issue, I think it makes sense to do it in Accordance?

@Kristin perhaps an easier method would be: Assuming that the list of verses with results can be copied or exported from Accordance, make a passage list in Logos by using the "copy from clipboard" option. Export your Logos search results in BHS to another passage list. You can then have fun with the merge options to find out what the differences between the two lists are though I don't have the time right now to detail how. Perhaps DMB will.

Hi @Francis, I had written the above asking you about the root issue and I was just getting ready to press enter and saw this. Regarding this idea for the morph discrepancy, that is a great idea! Accordance has a verse compare function too, and a few months ago I had tried to compare the texts (I think it was an issue in the Greek), but I couldn't figure out how to get passages out of Logos. However that was before I knew of the Passage List function, so when I can (hopefully in a few hours) I will attempt to compare the texts again doing that. Thank you for the idea!

Francis

Maybe @Francis can comment regarding the root issue, as it seems like a major problem for BHS to not even have the function

Apologies: I assumed but did not realize that this text did not have roots. I don't really use them often. Looks like LHB is the only Hebrew Bible that proposes that feature in Logos.

Kristin

Apologies: I assumed but did not realize that this text did not have roots. I don't really use them often. Looks like LHB is the only Hebrew Bible that proposes that feature in Logos.

Hi @Francis, no worries for making that assumption, as I certainly did too. I don't use root functions often either, but it seems pretty basic for any text which will be used for scholarly purposes. I think I am actually more concerned with the Lexham text though, as not having a root function is better than having an inaccurate root function.

Francis

In this example, I ran a lemma search in BHW Genesis and then in BHW Genesis-Deuteromy (just to have a list and then a longer list that contains the first). Then from the longer list I used the merge option difference with the shorter list and got the 1 passage difference in the merged list:

I think it should work though it's end of day and my mind is tired :)
Unless of course, the extra hits are not in different verses.

Kristin

In this example, I ran a lemma search in BHW Genesis and then in BHW Genesis-Deuteromy (just to have a list and then a longer list that contains the first). Then from the longer list I used the merge option difference with the shorter list and got the 1 passage difference in the merged list:

Hi @Francis, Thank you very much for the screenshots, I appreciate it. I have some complicated work I need to focus on at the moment, but this evening I will do my best to replicate your screenshots. It looks challenging since I am not familiar with it, but the merge looks really useful. I will keep you posted if I can pull it off or not.

Just to be clear, these "lists" are passage lists? Or do you not need to actually take it that far and just running it is enough to create the list in the screenshot?

Francis

You can create the list directly from the search:

Then you can delete it from its own kebab (3 dots) menu when you're done with it if you just needed the list temporarily.

Kristin

Hi @Francis, ok, great. Thank you for the screenshot and clarifying this. :) I will do my best to accomplish this.

Kristin

Ok, so I am finding this so confusing that I am not even writing anything on my screenshots. I will just say that the Accordance "passage list" which I imported is called "Acc tdzq" and I had it find the verses in the BHS, then I did the passage list for real in the BHS, and did a few different things. I had it show verses which were only different and it came up with tons of verses, not just two.

Also, even though both the Acc list and BHS list were done in the BHS, all the merge things defaulted to the ESV.

Logos also is giving a strange count for the Accordance list since it seems to be crunching verses which are next to each other, I guess, so I will include a screenshot of Acc also for comparison. I am also finding this confusing, so I hope I am attaching what is needed.

Bildschirmfoto 2025-03-25 um 23.40.41.png

Bildschirmfoto 2025-03-25 um 23.32.39.png

Kristin

Btw, I was able to figure out how to get the Logos Passage List out of Logos, so I then I moved it to Accordance and created a Reference List (Accordance's version of a Passage List) and I compared it to find the differences from Logos and Accordance, and Accordance and Logos, and both versions say there are no differences.

I think this would make sense as it goes back to the idea above that the two extra hits are part of bracketed text, so within the same verses.

I am really confused by the Logos screenshots above, but here are the Accordance screenshots.

Bildschirmfoto 2025-03-25 um 23.54.21.png

Bildschirmfoto 2025-03-25 um 23.55.49.png

Kristin

Bildschirmfoto 2025-03-26 um 00.05.38.png

Sorry, I accidentally posted one version twice… here is the other one…

Keep Smiling 4 Jesus :)

My search of Lexham Hebrew Bible (LHB) for lemma.h:צְדָקָה found 157 results in 150 verses:

Precise Bible Search of LHB for root.h:צדק found 449 results in 422 verses:

Keep Smiling 😊

Kristin

My search of Lexham Hebrew Bible (LHB) for lemma.h:צְדָקָה found 157 results in 150 verses:

Hi @Keep Smiling 4 Jesus :)

Thank you for the screenshots. I appreciate it. Since I was having such a hard time comparing what was going on, I re-ran the BHS and also got those results, which is how it originally was, with it missing two hits compared to Accordance's 159 hits. However, the fact that it is in 150 verses like Accordance really seems to confirm that it is a bracketed issue in an unknown verse. I think to figure it out it really would need to be done comparing only one book at a time as @Francis was suggesting above.

Also, since I was so exhausted when I made those screenshots last night, here is an updated one of the search which matches yours.

Bildschirmfoto 2025-03-26 um 08.49.48.png

Precise Bible Search of LHB for root.h:צדק found 449 results in 422 verses:

Thank you for this as well, and this is concerning. I don't use root functions often, but I would hope to get an accurate result if I do. I am also concerned that the BHS and BHW don't have root functions. So again, since I was so exhausted, I am reposting the root from Accordance (the HMT), which is pulling 661 root hits. That is a pretty major discrepancy.

Is there a way to copy a the verse hits in a true copy format? What I would like to do is run the root in Logos' LHB and then find the differences in Accordance, but simply copying the verses from Logos in a usable format is a multi-step process. Is there a way to simply copy the verses to the clipboard to paste somewhere else? Or am I correct that multiple steps are needed?

Bildschirmfoto 2025-03-26 um 08.54.16.png

Francis

@Kristin at the broad level, I would caution against assuming that (1) the same assumptions are made about roots in both softwares. For instance, we frequently see differences between lexica in terms of how lemmas are categorized, sometimes as the same word with different senses, other times as different roots. This could account for some discrepancies. (2) It would also be useful to know how each software computes (what does the number actually show?). In Logos, you can see the lemmas included in a root when you do a BWS (Bible Word Study) on one of the related lemmas. The root section of the guide will show you the root and all the lemmas under it. If you wanted to take the time to do this, you could manually search for all the lemmas listed in accordance and add up the results to see how it compares with the root total in Logos. However, it may lead to more questions yet.

That being said, digital precision gives the impression that we should expect numbers to be 100% reliable. In reality, there are always human choices involved and that includes room for error and differing assumptions. Documentation is not always up to par either because it is too involved and costly to update. These are the ones that can often shed light of what is going on under the hood in each software. Such information tends to be scattered across forum discussions, wiki entries and sometimes documentation resources, help file sections and instructional videos.

Kristin

@Kristin at the broad level, I would caution against assuming that (1) the same assumptions are made about roots in both softwares. For instance, we frequently see differences between lexica in terms of how lemmas are categorized, sometimes as the same word with different senses, other times as different roots.

Hi @Francis,

Thank you for your detailed response, and I think you raised a lot of important points. I agree that the "issue" with the Lexham roots is a philosophical difference as opposed to an error, sort of like how Strongs and LN give different results for κύριος. I am not under the impression that LN is making an error, as much as they have different numbers because of a different philosophy.

I think this also really addresses the primary concern I have about starting a "lemma" in Accordance, and then moving to Logos, as I am worried about lemma philosophical differences. I would have the same concern if I started a word in Logos and moved to Accordance. I don't really care where the lexes are, I just want to make sure they are all accounted for and without overlap of lemmas.

(2) It would also be useful to know how each software computes (what does the number actually show?).

In case you or anyone finds it interesting, here is the lexical breakdown of the root in Accordance.

Bildschirmfoto 2025-03-26 um 10.23.00.png

Francis

@Kristin scroll down to "Technical details" on the Lexham product page of LHB for some details about how roots (and other aspects) of LHB work or are based on.

Kristin

Hi @Francis,

Thank you very much for the link, that was really helpful, and I appreciated the site not only clarifying their philosophy, but also how it differs from BHW and BHS. While I think I gravitate more to the philosophy of BHW, I think for what LHB has as a stated goal, that they do it very well. Thank you again for the link.

Keep Smiling 4 Jesus :)

Logos Concordance for LHB can show all the lemmas for a root:

Puzzling is precise Bible Search of LHB for root.h:צדק finding 449 results while Concordance shows 451.

Keep Smiling 😊

Kristin

Logos Concordance for LHB can show all the lemmas for a root:

Hi @Keep Smiling 4 Jesus :)

Thanks for the screenshot. To be honest, I don't really see the point of the LHB root, since on the extremely rare times I use the root function, the only reason I use it is because I am looking for words which MIGHT have a VERY distant relationship, yet because of the Lexham philosophy (which is valid, just different), they ignore potential words I would be looking for. So I would rather see Accordance's 601 hits and weed through unrelated words than LHB's 449 and miss something.

Puzzling is precise Bible Search of LHB for root.h:צדק finding 449 results while Concordance shows 451.

Ya, I find that unusual as well, but I think it really stresses the point that their root is not finding ALL words of a distant root. While I can't see using the root function, I did like the concept of the Concordance and so I tried it to look at the lex and I am kind of confused for a few reasons.

First, when I opened it, it defaulted and started generating all the words in the ESV (including ch. for chapter, which I would argue is not part of the ESV's text, per se). However, I was able to change it to the BHS and lemma, but I couldn't type in a lemma, like you seemed to do for your root.

Second, is the way to find a lemma simply to scroll down? I get this is user error and I am just messing up the alphabet, but shouldn't my lemma be near the middle of this list? I don't see it.

Bildschirmfoto 2025-03-26 um 14.47.36.png

Quick Links

All Categories
Recent Posts
Activity
Unanswered
Groups
Help
Best Of