Ignoring qere in [COUNT ?] searches using HMT-W4

[Juan D. Pinto]

I'm doing a word count search to find hapax legomena in Isaiah. When using HMT-W4 I would like to ignore all qere. Of course, I can set the search condition "Bracketed Words" to "Ignored." However, when using the [COUNT 1] command, Accordance takes ​qere into account in how it calculates word counts. Is there a way to make it completely ignore qere?

 

I am also using BHS-T, which doesn't have this problem at all, but it uses a more outdated version of the Westminster Morphology (4.14) than HMT-W4 (4.18), and I've already noticed a few differences in the way the morphologies tag certain words.

 

Any ideas? Thanks in advance.

[Ken Simpson]

He Juan, thanks for this question.

 

I am not having the issue however. If I set my range to Isaiah and the bracketed words to excluded, it ignores all the qere when I do  [COUNT 1] search ((I get 1111 hits)

 

If i set it to bracketed words exclusive, I get 4 hits and if I set it to included I get 1115 hits.

 

So here are the shots as an example

 

 

Have I misunderstood something?

[Juan D. Pinto]

Thank you for the response, Ken.

 

Sorry, I should have been clearer with my original post. Here are some of my results when I do a COUNT search using BHS-T:

 

Here is the same search using HMT-W4:

 

As you can see, there is an additional find in BHS-T that does not show up when using HMT-W4 (Isaiah 13:2). I then look into this by identifying the lemma of this word, searching for it, and I see that in BHS-T it is in fact the only time this lemma appears in the entire Hebrew Bible. When I do the same thing in HMT-W4, however, I get two hits:

 

This, then, is the reason that my [COUNT 1] search in HMT-W4 does not show this result. Yet only one of these two hits is the actual ketiv, whereas the other one is part of the qere.

 

My question is this: how can I perform a COUNT search where the COUNT entirely ignores qere words in every part of the equation?

 

 

Edited June 10, 2015 by Juan D. Pinto

[Λύχνις Δαν]

Hi Juan,

 

  I don't seem to have access to your photos. Even logged into to DB before clicking the links I get an error message. Now you can attach JPG, PNGs etc to your posts here. If you don't see an attach button click More Reply Options and you should see it. Then you can attach them directly here.

 

thx

D

[Juan D. Pinto]

Daniel, thank you for the tip! I've attached the photos now and replaced the links.

 

Juan

[Λύχνις Δαν]

I am perhaps not understanding the problem but it looks like if in your third example that you add the "ignore bracketed words" search option you should just get the one hit which sounds like what you want. Also in the first image.

 

thx

D

Edited June 10, 2015 by Daniel Semler

[Juan D. Pinto]

Daniel,

 

The third example is merely an attempt to understand why Isaiah 13:2 doesn't show up in my second example. It seems like the reason is that the [COUNT ?] command is including qere words in its searches. Even though I turned on "ignore bracketed words" in the second example, the search for [COUNT 1] does not turn up words that do in fact only show up once in the ketiv of their specific lemma but that additionally exist as qere. This is the case with the lemma שׁפה in Isaiah 13:2 (just using this as an example—this same thing occurs with many words throughout the Bible): my third example shows that in the HMT-W4 שׁפה occurs only once in the actual text (ketiv), but its additional appearance in the qere keeps it from showing up when doing a [COUNT 1] search.

 

Hopefully I'm making my problem clearer and not more confusing! 

[Λύχνις Δαν]

Ok so doing this query I get just one hit :

 

I have now re-read this thread. I have been a bit dim I think. The problem you are having is that the 13:2 hapax is not showing up in HMT when you ignore bracketed words. I thought they problem was too much showing up not too little. Oddly above it works in my query directly specifying the word itself.

 

I have no rerun your HMT query and got this :

 

 

As you can see I have Isa 13:2 here and a count of 1111 hits.

 

You are specifying [RANGE Isa] as a search query predicate not as a search option. These are not equivalent. If I run the query the way you do I get what you get - no Isa 13:2. The problem is that specifying [RANGE isa] in the search itself finds all the hapax in the search range which is the whole text, and then trims the results presented to the RANGE you specify. If you specify it as a search option then the search itself is restricted to Isaiah.

 

But, and this is perhaps the problem, this word is a hapax in both Isaiah and the whole HMT which I believe you said - sorry catching up. You search in your second image is this :

 

[COUNT 1]‏  @-  ‎[NOUN properName]‏ ‎[range isa]‏

 

Read that RTL.

 

 

COUNT 1 seems to be operating differently here depending upon how the range is specified. If you do this :

 

 

it works and you get 13:2.

 

Someone from Acc will have to comment on this. I don't understand yet how this is happening or whether it constitutes a bug or not.

 

Thx

D

 

 

 

 

[Helen Brown]

I agree that this is a bug. The Count is including the qere in the count, whatever the setting for bracketed words.

Try this search with different settings for the brackets:

 [COUNT 1] @שׁפה

 

The issue of how the Count and Range interact has been often explained, and it not relevant here. The Count relates to the words in the range set by the pop-up menu, not the Range command.

 

We'll look into the bug, but I cannot guess whether or when it might be fixed.

[Juan D. Pinto]

Thanks Helen, glad to know it'll be looked at!

 

 

Juan