Logos Syntax Searching – #2 & #3

Check out the Introduction and Part #I too!

This was originally going to be two separate posts, but it seemed better to keep them together.

The next step in the process of building, performing and analyzing a syntax search involves less technical building. Its much more about analysis of what we find.

Simply put, we expand out search beyond the small corpus of Ephesians from the original search to the entire New Testament.

With the small results from the initial Ephesians search, we also want to formulate some questions:

    1. How often does this construction occur in the New Testament?
    2. Does it occur more often with particular writers?
    3. Ephesians showed two prepositions can function in this manner. Are there any others?

As a reminder, our search looks like this:

In Entire Database:

Clause 1

-> Clause Component 1: Clause Category = Predicator

-> Word Group 1

-> Head Term 1

-> Modifier 1: Modifier Category = Specifier

-> Word 1: ((Part of Speech = preposition))

-> Word 2: ((Part of Speech = article))

-> Word 3: ((Part of Speech = verb AND Mood = infinitive))

Now let’s see if you can answer a few of our questions above.

The first is rather easy to answer since the results total is listed in the top right of the search window: 181 Occurrences.

The second question is rather easy to answer as well. Click on the “Graph Bible Search Results” link in the top right of the window. When the graph window opens up, select from the drop down box, “Number of hits in Book.” Also in the very top right, there are several buttons, be sure that “Hide Zero Items” is already selected, otherwise you’ll have every book of the Bible listed, including the OT, which wasn’t even searched. Once that is done the graph should look something like this:


This construction seems to be very common in Luke & Paul, but its somewhat deceiving in that Luke & Paul also wrote the most! So select the drop down box and choose, “Number of Hits in Book / 1000 words in Book.”


Now that’s interesting! If we average Paul’s numbers, we get 2.9838 / 1000 words. The average for Luke is 1.8154 / 1000 words.

The third question takes a bit more work. Unfortunately, Logos does not yet allow us to sort our results based on any usage or patterns, so we cannot simply have the data reorganized for us to see what prepositions are used (but I’m hoping such a feature will come in 4.0). Thus, we have to go through the results manually. Or do we?

A quick skim through the first dozen hits or so give us six prepositions:

πρὸς, πρὸ, ἐν, διὰ, εἰς, μετὰ.

Let’s go back to our search window. This is the beginning of the third step in the process. It is here that the second Narrow comes in. This last Narrow actually consists of several searches. Now that we have our results covering the entire corpus, we want to narrow in on different patterns that arise.

This time we’ll add a specific preposition to our search. Type in προς (Note: if you include the accents in your search, be sure to check the “Match Marks” box in the search window).


Let’s wait on examining the search hits until after we’ve organized all of our data. So open up the search window again. This time type in προ.


After the search is done, continue doing searches for the rest of our prepositions, εν, δια, εις, μετα.

The combined results comes out to 177 hits – four less than our initial total of 181.

Now, in order to find those last four, we must do one more search. Logos has a very helpful feature that helps save us quite a bit of time in digging out those last four occurrences. Enter all six prepositions in the Lexeme box one at a time so that your search window looks like mine. After that, click on the link that says, Match only the items in the word list. Click on the only and select anything except. Now your search box should look like this:


This search gives us our last four – and also four more prepositions:

ἕως, ἕνεκεν, ἐκ, ἀντὶ, all with one hit each.


This gives us a total of 10 different prepositions used in this construction. From this we could probably hypothesis that any preposition can be found in this construction. And by browsing the list, we can probably assume that any verb can be used as well.

In our conclusion (soon to be HERE) post, we’ll spend some time examining the hits themselves and formulate some thoughts about the meaning and significance of the construction.