BUG 10.3.3 - Tagging / Highlight Issue

[Matt Fredenburg]

If I hover my mouse over 'destroy' in 2 Cor. 10:5, no highlight appears in the parallel Greek text:

 

 

If I hover over καθαίρεσιν in 2 Cor 10:4, then 'destroy' is highlighted in 2 Cor. 10:4:

 

 

If I hover over καθαιροῦντες in 2 Cor. 10:4, then the same 'destroy' is highlighted in 2: Cor. 10:4:

 

 

It seems something is amiss.

 

Matt

[Helen Brown]

The word "destroy" in verse 5 does not have an equivalent in the Greek for this verse. It is supplied in the ESV to carry over the argument from the previous verse. You may want to use the MOUNCE-NT to see a more accurate phrase by phrase tagging.

[Matt Fredenburg]

The word "destroy" in verse 5 does not have an equivalent in the Greek for this verse. It is supplied in the ESV to carry over the argument from the previous verse. You may want to use the MOUNCE-NT to see a more accurate phrase by phrase tagging.

Hi Helen, please bear with me in this. The words in the form (word highlighted in the parallel text, Key Number shown in ID) represent what is displayed in the parallel Greek text and Instant Details when the prior English word is hovered over.

 

In the HCSB, 2 Cor. 10:4-5 is rendered as:

since the weapons of our warfare are not worldly, but are powerful through God for the demolition (καθαίρεσιν, G2506) of strongholds. We demolish (καθαιροῦντες, G2507) arguments and every high-minded thing that is raised up against the knowledge of God, taking every thought captive to obey Christ.

 

 

In the ESV, 2 Cor. 10:4-5 is rendered as:

For the weapons of our warfare are not of the flesh but have divine power to destroy (καθαίρεσιν, G2506) strongholds. We destroy (no highlight in the parallel text, G2507) arguments and every lofty opinion raised against the knowledge of God, and take every thought captive to obey Christ

 

 

In the HCSB, when I hover over the word 'demolish', Instant Details shows

demolish G2507 καθαιρέω kathaireō

[GNT-T] καθαιροῦντες kathairountes Verb pres act part masc plur nom to take down, destroy

 

 

In the ESV, when I hover over the word 'destroy', Instant Details shows

destroy G2507 καθαιρέω kathaireō

--

From this, I don't see how it could be the case that in the ESV 'destroy' does not have an equivalent in the Greek for the verse in question. From what I've demonstrated above and in my original post, then that would seem to indicate that the ESV committee decided to take both καθαίρεσιν and καθαιροῦντες and translate them as a single word 'destroy' and then manufacture destroy in verse 5 out of thin air.

Matt

 

[Steve King]

To be honest this is not a bug in 10.3.3 but a tagging issue. There is a 'report a correction' facility in Accordance to deal with text and tool errors which gets sent to the relevant team.

 

The verse indication in the ESV does not match that of the Greek text and that has confused Helen because there is no word for destroy in v5 both are in v4.

[Joel Brown]

Actually, I would argue its neither a tagging issue nor Accordance bug. The ESV committee decided to move the second 'destroy' to verse 5, where it no longer has a greek equivalent. Accordance's cross matching works on a verse level, so when the verse division moves, there isn't much we can do.

[Helen Brown]

It could have been tagged with the Greek word, as the Holman taggers did, but normally we tag with a word that is present in that verse in the original. Apparently the Holman scholars follow other guidelines. Tagging would give Instant details but not cross-highlighting.

[Matt Fredenburg]

OK, still confused, sorry about that! So, taking the Accordance staff's words at face value (i.e. neither a bug nor a tagging issue), would someone be able to explain why when I hover over 'demolition' in the HCSB, ID shows a Strong's number of 2506 and the Greek word καθαίρεσιν is highlighted in the parallel Greek text, and that when I hover over 'demolish' in the HCSB, ID shows a Strong's number of 2507 and the Greek word καθαιροῦντες is highlighted in the parallel Greek text, but, when I hover over either καθαίρεσιν or καθαιροῦντες in the Greek text, they both display a Strong's number of 2507? I checked in the Greek Strong's resource, and the lemma for each is different (2506, 2507).

 

Thanks,

 

Matt

[Joel Brown]

The reason the HCSB is different is that it keeps the second 'demolish' in verse 4, like the Greek. Because of this, it can match up with the Greek properly. Again, everything is tagged properly in all 3 texts (ESV, HCSB, GNT). However, cross highlighting and Instant Details cross-information is all based on a verse by verse matching. Since the ESV moved G2507 destroy to verse 5, it cannot match with any words in the GNT verse 5 or vice-versa. Since the HCSB kept it in verse 4, it can correctly match.

 

It may get a bit more confusing since Accordance will attempt a 'best-fit' match in certain cases, which is why καθαιροῦντες matches to G2506 in the ESV. This is necessary due to the many places where GBS's lemma is slightly different from the Strong's (or modified) lemma.

 

The reason you get G2506 in the ID of both places in the GNT is because you are using ESVS to display that information. Go to Preferences -> Compare Texts -> Interlinear Source Texts and change your Keyed Text to HCSB to get the proper keyed information for that verse.

[Helen Brown]

Sorry, I had missed that second part of the question. Upon further investigation:

 

HCSB,, NRSV, MOUNCE: all leave the phrase "We destroy arguments" or the equivalent, in verse 4, and the tagging and cross-highlighting work just fine with the GNT28-T or GNT-T.

KJV, NKJV, NAS, and ESV: all move that phrase to verse 5, following the Textus Receptus. The word is correctly tagged with 2507 which shows up on Instant details. It cross-highlights with the GNT-TR.

 

The problem arises when you try to cross-highlight between a version that has the phrase in verse 4 and another that has the phrase in verse 5. Accordance does not look across verse boundaries for this, so it just tries to find the best match for the word within that verse, hence the erroneous match in this case.

 

Remember, the Key number tagging is imperfect (a judgement call as well as a work in progress), and the cross-highlighting is done by algorithm. Both are just aids to the study of the Greek and should not be relied on for accurate details.

Looks like Joel and I both answered together, at least we pretty much agree.

[Timothy Jenney]

Hi, Matt!

 

I love the detailed investigation you are conducting. Before we go further, let me encourage you to take Greek (if you have not done so already). I doubt you will ever be satisfied with the sort of "rough approximation" the various key numbering systems provide. Strong's numbers give English-only users a rough idea of corresponding words in Greek/Hebrew, but it is only a rough approximation. Goodrick/Kholenberger is a better system, but even it has its flaws.

 

OK, that said, you will occasionally run into minor inconsistencies like this, especially when comparing texts that have been tagged by different people. Some of these are due to simple mistakes by a tagger. Others are due to slightly different understandings of Greek grammar among taggers: Are all nouns derived from verbal forms or only some? When a lexicon has lemmas for both the verb and the noun, which should be used? There is an honest disagreement among scholars for instance, whether the primary root form of agape is a noun or a verb. Different lexicons even handle some of these words differently.

 

In this specific case, the two numbers point to a verb form (S 2507) and a noun form (S 2506), both of which share the same common roots (so the issue doesn't make any practical difference). However, it is still fascinating!

 

Now, unless you are using the GNT-TRS (which uses both 2506 and 2507), your Greek text is not actually tagged with Strong's numbers—Accordance is reading these numbers from an English translation that is. That English translation should be indicated in square brackets just before the key number, like this: [HCSB] 2506. I'd like to know what translation you are using, as [ideally] it should be reading 2506 and 2507. [You can change it, by the way, in Preferences>Compare Texts>Interlinear Texts>Keyed Text.]

 

Note: I see Helen and Joel both answered more succinctly, while I was still composing my post. Now THAT'S customer service! :-)

Edited December 6, 2013 by Timothy Jenney

[Matt Fredenburg]

Hi, Matt!

 

...

 

Note: I see Helen and Joel both answered more succinctly, while I was still composing my post. Now THAT'S customer service! :-)

Indeed it is! Thanks so much to all of you to putting up with my imbecilic queries. I think half the reason I use Accordance is so I can be reminded of how much I don't know. But, all that being said, those last three responses got the point home loud and clear and I feel the better for it.

 

Matt

 

Note to self: this is neither a bug nor a tagging issue.

Edited December 7, 2013 by Matt Fredenburg

[Matt Fredenburg]

...

 

The reason you get G2506 in the ID of both places in the GNT is because you are using ESVS to display that information. Go to Preferences -> Compare Texts -> Interlinear Source Texts and change your Keyed Text to HCSB to get the proper keyed information for that verse.

Hi Joel, how does the 'Keyed Text' setting manifest itself in various uses of Accordance? I have tried changing and then examining what is displayed in Instant Details, and even with HCSB, it still shows the same number for both places in the GNT. I looked in the help, but I couldn't find anything on it other than "Defines the keyed text to use when displaying Key number information for an original word in the Instant Details and Dynamic Interlinear Display," which is probably crystal clear to most people, but I'm just not getting what it actually means and how it would be utilized.

 

I'm sorry to reanimate this dead thread, but I'd really like to understand the use case for this setting.

 

 

Thanks,

 

Matt

[Steve King]

Hi Matt,

 

Joel is describing what the instant details show when you mouse over the GNT. If you have ESVS set as the key text then the key shown for καθαίρεσιν and the key shown for καθαιροῦντες are both G2506 although the Greek parsing and gloss indicate correctly that one is a verb and the other a noun. If you set the key text to HCSBS then you will see that the key indicated in instant details when mousing over the relevant words in the GNT is G2506 for καθαίρεσιν but G2507 for καθαιροῦντες.

Edited December 9, 2013 by Steve King