Nouns and Adjectives and Graphical Queries, Oh My!

On Saturday morning I was studying the first part of 1Ti 4.6:

If you put these things before the brothers, you will be a good servant of Christ Jesus, (1Ti 4.6a, ESV)

When I was looking into the term “good servant”, I noticed that in the Greek it was an adjective and a noun that agreed in case and number. So, I wondered, what other things are called “good” in the Pastoral Epistles? This article explores ways to specify this sort of search with the Graphical Query Editor.

[I should note that I have been working through the Pastoral Epistles for some time. I blog about the Pastoral Epistles at and have some other information on my personal web site.]

This is the sort of thing that the Graphical Query Editor is designed to do without getting too bogged down in intricate search syntax. Sure, you could learn the syntax to specify it in a textual query, but it’s much more fun to make a pretty picture to specify your search query, like this:

But how did I do that, and what does it mean? Hopefully I’ll be able to explain the basics here. Open either the NA27 or NA27INT resources, head to 1Ti 4.6, and follow along. Please note that this query was done with an ‘alpha’ or in-development version of Logos Bible Software, but the dialogs and instructions should apply to the current (v2.1c) version of Logos Bible Software as well.

The first thing you need to do is open a new Graphical Query Document. So do that. You can use the new document icon in the standard toolbar, and then select “Graphical Query”. Then click OK. You should be staring at a blank graphical query canvas. This is where you paint the pretty picture that will actually specify a search. You’ll notice terminology running down the left side of the canvas. These are the things (“objects”) you can specify. You just drag and drop from the margin bar to the canvas.

The first thing we want to do is specify a search term, the word καλος. So use your mouse to grab the TERM box, and drag and drop it anywhere on the canvas. When you do this, a dialog will automatically open. This is where you specify information about the term in question.
Since this search will specify the term in Greek, the first thing we do is change the language to Greek in the drop-down. After that, the dialog will stop and think for a minute. That’s OK, it’s just preparing a Greek word list to help you type the right Greek word.

You’ll see the TERM box (in red) where you are to type in the Greek term. Click in there. Now you need to type in Greek. So you need to switch your keyboard to Greek. The easiest way to do this is (assuming your keyboard is currently in English) is to hit the F2 key twice. The first hit switches the keyboard to Hebrew, the second cycles it to Greek. Now you can type in Greek in the dialog. So, type in “καλο“. You should see a list of words generate in the left column.

Scan the list for καλος. You’ll notice that there are two instances. One has a grave accent over the omicron, the other an acute. Which one is the proper one to use? Well, it really doesn’t matter. If you look at the bottom of the dialog, you’ll notice a few boxes, one of which is “Match Marks”. This box is not selected. That means that when you search, you’ll only search for Greek words with the letters “καλος“.

So select either form of καλος (I selected the second one) and then click OK.

You’ve got the first part of your Graphical Query specified.
Next, we need to specify that this word (καλος) is an adjective. In Logos Bible Software, the morphological tagging is a reference. This means we grab the REF box from the margin bar and drag and drop it on the canvas (anywhere is fine, you can arrange the objects later).

Again, when we do this, a dialog pops up. This time the dialog allows us to specify properties of the reference. The first thing we need to do is make sure that the proper data type is selected. We want to use the Greek Morphology (GRAMCORD) data type, so make sure that is selected in the Data Type dropdown. When you do this, you should see the standard morphology panel (the same one from the Greek Morphological search dialog) appear.

At this point, we’re only worried about the part of speech of the word, so we want to click the Adjective checkbox in the Part of Speech section of the dialog. Don’t worry about case, number, gender or degree. Got it? Then click OK.

So you should have two boxes on your Graphical Query canvas. Now we need to specify that both of these things refer to the same term. If you hold your mouse by one of the boxes, you should see a gray arrow appear. Click down on that gray arrow, hold and drag to the other box. You should have a big, fat purple arrow connecting the two boxes now. When you do, double-click that purple arrow.

Now you should see the Proximity dialog appear. Since both of these pieces of data (the “surface form” and the “morphology”) refer to the same thing, we want the proximity to be “Same”. So click on the Same radio button. Then click OK.

You should see a large equals sign appear connecting the two objects, and the purple arrow should be gone.

If you’ve never used the Graphical Query Editor before, now is a good time to play. Click and drag one of the terms around, and see how the equals sign keeps the two terms connected. You’ve established a relationship of equality in place between these two things. If you did a search right now, you’d find all the places in the New Testament where KALOS is an adjective. Pretty cool, huh?

But that’s relatively easy to do with the normal Greek Morphological Search dialog. Now we’ll start to get into the interesting area of Agreement. But before we do, we need to create a new object on the Graphical Query canvas.

We need the noun object. So do the same basic process that we did for the adjective object. That is:

  1. Click the REF object in the margin bar.
  2. Drag it onto the canvas
  3. Make sure the proper data type is specified.
  4. Specify Noun as the Part of Speech.
  5. Click OK.

Got it? Good.

Now we get to specify agreement between the adjective and the noun. The term “agreement” simply means that we want to find adjectives and nouns that match each other for certain morphological critera. That is, we don’t want to simply search for nouns and adjectives that are in the nominative case. We want to search for nouns and adjectives in any case, provided they match each other. Same for number. On this search, we don’t really care about gender.

Have you got a guess by now of what the next step is?

That’s right. We grab the AGREE object from the margin bar and drag it onto the canvas. Then we need to connect the AGREE object to the things we are comparing — the noun and adjective references. Remember the little gray arrows? We do the same thing here. It may help to arrange the noun and adjective objects so that the AGREE object is between them, and there is a decent amount of space between all three objects. Hold your mouse outside of the adjective object until you see the gray arrow. Click and drag to the AGREE object. You should see the AGREE object pointing to the adjective object. Now do the same thing for the noun object.

Got it? Cool.

Next, we need to specify the criteria in which these two references agree. These are Case and Number. So double-click on the big yellow/orange agreement object. You should see the agreement dialog now.

Since searching for agreement of this sort is a common task, it is easy to specify this criteria. In the General section, you should see two options:

  • Case
  • Number

Make sure both of those boxes are checked. Then click OK. You should see the AGREE object now specify something like “Agree with respect to Case, Number”.

We’ve almost got the query specified. There’s just one more thing. Sometimes nouns and adjectives agree in case and number, but they aren’t necessarily related. We want to narrow down the number of words that occur between the adjective and noun so that we are only dealing with things that are likely in agreement; that is, where we can be relatively sure that the adjective really is modifying the noun, not that they just happen to occur in the same sentence.

So, we get to draw another arrow between the adjective object and the noun object. Go ahead and do that.

Now, double-click on the fat purple arrow that just appeared. Oh, by the way, you can arrange the arrow too. It doesn’t have to be straight, it will bend as well. So to make your query look better, you may want to bend the arrow around the agreement object.

When you’ve double-clicked the purple arrow, the proximity box will show again. We want to do a few things in this dialog:

  1. Make sure that Ignore order of terms is checked.
  2. Make sure that At most […] intervening is selected.
  3. Enter ‘3’ (or whatever other number you’d like) inside of the ‘intervening’ box.
  4. Make sure that Words is selected in the Unit: list.

When you’ve done that, click OK. You should see the text “0-3 words” appear inside the fat purple arrow, and each end of the arrow should have a pointer, indicating that order doesn’t matter for this query.

And now, we’re ready to search. It’s been a little bit of specification, but hopefully the explanation has been helpful in understanding what you are specifying. Once you get the basics down, future queries of this sort should be easier to specify.

But we’re not quite done yet. There are a few things to specify in the search dialog as well. So go ahead and click the Search button that is on the Graphical Query Editor toolbar (not the one on the main toolbar). You should see a dialog appear where we can specify stuff for this particular search.

The first thing we need to specify is the resource we are searching in. This is why I had you open the NA27 or NA27INT before we started. Now you can just click the Open Resources radio button, and then select NA27 or NA27INT from the drop-down that appears.

The next thing we need to do is select the unit we are searching. One thing that not too many people know is that you can actually search the NA27 or NA27INT by the unit of “Sentence”. So select Special in the By: section. See the drop-down? Hit the drop-down arrow and select Sentence. Now you’ve specified that your bounding search unit is the sentence, which makes much more sense than verse for these types of searches.

The next thing to do is to select the range. For this search, I’m immediately interested in things that are qualified as “good” in the Pastoral Epistles. That means I need a special search range. So, in the Range: section, I do the following:

  1. Click on Bible Text.
  2. Type “1Ti-Titus” in the box.

You’ll end up with something like this:

That’s it. I’m ready to go. All I have to do now is click the Search button. When I do that, the results are: 47 occurrences in 19 articles. The ‘articles’ are sentences, but the verse is listed right there too. You can now click on any of these to follow up, import them into a verse list, or whatever.

As a final note, I should say that this article ran quite long. I hope I’m not setting expectations too high; I don’t anticipate the majority of the articles I write for the Logos Bible Software Blog to be this detailed or long. But I am hoping that you have a better idea of the sorts of things that the Graphical Query Editor is designed to do, and how you can go about using it for meaningful searches.


  1. Great post and very well written!
    This is the kind of article that helps people use Logos in a powerful way. Very few techinical writers can write in such a clear way. Ask Bob for a raise. :>)

  2. Virgil Horner says

    Hey thanks for this article, I thought it was well done and it wasn’t too long or detailed, but clear and concise. I didn’t even know about the Graphical Query tool, but thought it was awesome!
    Virgil Horner

  3. Excellent – I still probably understand less than 5% of graphical queries, but at least now I have started.
    Just one thing: In the “Proximity” section, you say select “At least … terms intervening”. That doesn’t produce “0-3 words”, it produces “3+ words”. I think you need to select “At most … intervening” to get “0-3 words”.
    Murray Woolnough
    [NB: The article has been corrected as Murray suggests. Thanks! — RB]

  4. Tom Bailey says

    Mac bible software Accordance author has a comparison of this process with their own software. I’d like the author here to respond there. It might be informative to all who read it.
    Here’s the link (it’s post #6 in the thread):

  5. I’ve read David Lang’s article — I didn’t know about it before, thanks for the pointer — and I don’t think I should respond on the Accordance Forum. What I wrote was for Logos users to see how I use Logos Bible Software in a particular situation. But David does have a point about ‘J????’ and ‘N???’ — those really should be expanded to the language of the localized interface and not be reflective of the compact codes representing the parsings. I’ve suggested to Bob and our developers that we do exactly that in the Graphical Query Editor; hopefully they’ll be able to implement it.

  6. Rick … thanks this is very helpful, not only to open up the functionality of the GQE but also because this is often the kind of inquiry I find myelf embarking upon in my proviate devotions (as with your more recent post too)