8 replies to this topic

[Helen Brown]

The Hits command enables you to take all the Hit words (or tags, or keynumbers, etc.) from a search in one Search window and use them as criteria for a search in a new window. Here's a simple example:

Let's say I want to compare the vocabulary of the Greek Septuagint with that of the Greek New Testament. In one Search window (or tab in the workspace), I choose GNT-T as my search text, click the Search for Words radio button, type an asterisk to find every word, and click OK. In another Search window, I choose LXX1 as my search text, then type an asterisk (*), followed by the at (@) symbol, followed by a minus sign (-). I then choose HITS from the Enter Command submenu of the Search menu. Provided I only have two Search windows open, the HITS command will automatically link to my window containing the GNT-T. Otherwise, I'll have to select that window from a list.

If I haven't lost you yet, the search criteria in the window containing LXX1 should look like this: "*@- [HITS GNT-T]"

This search will find every word (*) in the LXX1 which (@) is not (-) contained in the GNT-T ([HITS GNT-T]). Pretty cool, huh? (Okay, so it's really pretty geeky! Work with me here!)

Now, to examine the words that were found, I click the Details button and view the Analysis Tab. This gives me an alphabetical listing of every word that was found. To see which words are used most frequently, I can choose "Set Analysis Display..." from the Display menu, and then choose "Count Down" from the Sort pop-up menu.

At this point, I notice that the vast majority of words which are unique to the LXX are proper names. To filter those out, I go back to my search, type another at symbol (@) and minus sign (-), and then choose "Noun" from the "Enter Grammatical Tag" submenu of the Search menu. In the dialog box, I choose "properName" from the Class pop-up menu and click OK. My search argument now looks like this:

"*@- [HITS GNT-T] @- [NOUN properName]"

Now, at this point, I'm not even sure Accordance will let me do a search containing two at (@) symbols, but if it works, this search should find:

every word (*) in the LXX1 which (@) is not (-) contained in the GNT-T ([HITS GNT-T) and which (@) is not (-) a proper noun ([NOUN properName]).

I cross my fingers and click OK, and lo and behold, it works! Well, at least I think it works. The search takes a long time on my G3 PowerBook, so while I wait, I check all the Mac rumor sites for any hint of an upcoming G5 PowerBook! ;-)

When the search is finished, I find that it has, in fact, worked the way I had hoped, and my Analysis now shows everything but the proper names.

I've still got a pretty long list, however (nearly 6000 words). So I'd like to filter my search further to find only those words which are used 50 or more times. To do this, I go back to my search, enter yet another at (@) symbol, and then choose "COUNT" from the "Enter Command" submenu of the Search menu. I type "50-6500", and my search now looks like this:

"*@- [HITS GNT-T] @- [NOUN properName] @ [COUNT 50-6500]"

I click OK, and while waiting for the search to finish, I start Googling for "Big Foot", which is apparently easier to locate than the fabled G5 PowerBook for which I long! ;-)

In addition to inflaming my desire for a faster computer, this search ultimately does what it's supposed to do, and my Analysis now lists every word in the Septuagint which does not appear in the Greek New Testament, which is not a proper noun, and which is used 50 or more times. Whew!

Okay, so if you're not interested in comparing New Testament and Septuagint Greek, what can you take away from this little odyssey of mine? Here are a few tips:

• Use the Hits command in conjunction with the at symbol and minus sign to compare vocabulary between different texts or parts of a text. You might compare the vocabulary of the New King James with that of the KJV, or compare the Aramaic of the Bible with the Aramaic of the Targums. Or you might compare 1 Peter with 2 Peter, or the Pastoral Epistles with the wider Pauline corpus. There's all kinds of possibilities here.
  
• Use the at (@) symbol to search for words which meet multiple criteria. Think of the at symbol as joining words with tags, key numbers, word counts (a l‡ COUNT command), and hit lists (a l‡ HITS command). And get in the habit of reading the at (@) symbol as meaning "which." For me, at least, this helps to reinforce what the at symbol is doing.
  
• Use the Details button to get an overview of what was found by a search. The big picture perspective which the Analysis, Plot, Table, and Concordance provide can help you zero in on ways your search might need to be refined.
  
• Get in the habit of building your searches incrementally. Try something, examine the results, and repeatedly refine your search until you get the results you are looking for.

There! If you've made it this far, you've been given a handful of tips that, hopefully, will prove helpful to you.

David Lang

[miqraot]

Thank you for your kind explanation, Helen.
It is really helpful!

Now, I nearly can understand the function of [HIT].
But, I have one question.
If I had the [GNT-T] module, I would follow what you explained, but I don't have the module.
Because of the reason, I couldn't understand totally.

So, I want to ask the function with a below picture.



Which type is correct?
A? or B?

I wish that you understand what I am asking.

[Helen Brown]

I am not sure I understand your question or your diagrams, but I think that A is correct. This search entry:
*@- [HITS GNT-T] @- [NOUN properName] @ [COUNT 50-6500]
simply means:
Find all words in the text except:

• words found in the GNT-T window
• or words which are Proper Names

but which occur 50 to 6500 times in the current text and range.

In other words, each @ qualifies the original * wildcard, and the order of the qualifiers makes no difference. In fact you no longer need the * once you add the [COUNT 50-6500], so you get the same result with:
[COUNT 50-6500] @- [NOUN properName] @- [HITS GNT-T]

@ simply means that the term after it applies to the same word as the term before it. However, the first term must be positive.

HITS simply means that you want to use the list of words found by the search in the other window.

I hope this helps.

[miqraot]

Wow, Now I understood clearly!!!

Edited by miqraot, 20 September 2005 - 05:08 PM.

[johnny_redeemed]

This is great.

How can I use this to compare the words used in different books of the Bible in the GNT. For example, how can I search and find all the words that are in the book of Hebrews that are not in the Gospel of Luke? Thus, I would know all the words that are in Hebrews that are not found in Luke's Gospel. Or what if I want to find all the words in Hebrews that are not found in Paul's letters (Romans, 1 and 2 Corinthians, etc.)? In other words, I would know, after the search, all the non-Pauline words in Hebrews.

[Helen Brown]

Let's take the first example:

Your first tab which I will assume you named GNT-T-Luke searches for words using a single * and the range of Luke (either via the pop-up menu or the command) so it highlights every word in Luke.

Your second tab with a range of Hebrews searches for words with this argument:
*@-[HITS GNT-T-Luke]

which being interpreted means: find every word that is not on the list of hits in my tab showing Luke.

Click Details and then Analysis to get the list of non-Lukan words in Hebrews.

[johnny_redeemed]

Let's take the first example:

Your first tab which I will assume you named GNT-T-Luke searches for words using a single * and the range of Luke (either via the pop-up menu or the command) so it highlights every word in Luke.

Your second tab with a range of Hebrews searches for words with this argument:
*@-[HITS GNT-T-Luke]

which being interpreted means: find every word that is not on the list of hits in my tab showing Luke.

Click Details and then Analysis to get the list of non-Lukan words in Hebrews.

I must be stupid, but I cannot get any of those commands to work.

[Jonna Schmidt]

Hi, I was catching up on an old blog, and trying to use the HITS command -- without success. I was reviewing David Lang's blog from Oct 2008 regarding the optative. Tried to see how often the optative was used in the GNT-T and the GNT-TR. Got 63 Hits in the GNT-T and 83 hits in the GNT-TR.

SO, I wanted to find all the instances where the optative is found in the GNT-TR but not the GNT-T. And that is just not working. I thought that I was supposed to start a THIRD window. I set up the search argument as follows: [HITS GNT-TR 2] @- [HITS GNT-T] -- and now I have entirely different HITS such as Matt 5:26 which were NOT in either of my original search windows.

So, since I don't know Greek, I wanted to try to compare instances where the NASB uses ADAM vs KJV. KJV has 30 hits for Adam; NASB 21. When I type in a search argument, [HITS KJV] <NOT> [HITS NAS95S] ,
I get an error message, that there are no hits. I ought to get about 9 verses. I have gone back over the podcasts, but I am not making progress

What am I doing wrong?

Thanks for any help that you can render.

[Helen Brown]

The HITS command uses the word list of HITS from the other windows, and is used to compare the vocabulary.

You need to use the CONTENTS command instead, as this compares the verses. In the second example of Adam your word list is Adam in both windows, but you are wanting to find the verses in KJV which use Adam where the NAS95S does not. To highlight Adam at the same time search KJV for:
Adam <NOT> [CONTENTS NAS95S]

To search the NAS95S for man where KJV uses Adam search NAS95S for:
man <AND> [CONTENTS KJV] <NOT> [CONTENTS NAS95S]

Similarly in the first example, your search was not helpful since you ended up searching for all the lemmas in the GNT-TR which are found in the optative in GNT-TR but not found with that tag in GNT-T. You could simply search GNT-TR for:
[VERB optative] <NOT> [CONTENTS GNT-T]
it gives 17 hits, but it potentially eliminates instances in the GNT-TR where the optative tag is used on a different verb in the same verse. This search:
[VERB optative] @-[HITS GNT-T]
gives the lemmas tagged as optative in GNT-TR which do not have optative forms in GNT-T.

I do not think there is a way to find exactly what you want, you will need to look at each instance of the optative in the GNT-TR and check the tagging in the parallel GNT to be sure of picking up each one.

I hope this is helpful.