Greek Syntax Search

This article is designed to help you use the Syntax Search capability in Verbum to find patterns, or identify particular structures, in Greek biblical texts. This is primarily for the New Testament, but this feature is also available for the deuterocanonical books of the Septuagint.

Other Verbum Search options provide rich capability in searching for words or phrases, for different forms of words, for different clausal elements. Syntax Search will do these things as well, but it really comes into its own when trying to find occurrences of particular structures or when different terms in a clause have specific relationships to each other.

For a general overview of what the Syntax Search is, and how to use it, please see the Syntax Search Document entry in the Verbum help file.

As an example of these types of searches, the Syntax Search below will show all clauses which have a verb and a noun in them that share the same Root.

Examples will be given through this article, but do take time to stop, reflect, and do your own study using the skills and knowledge you will have gained.

By the end of this tutorial, you will have a much greater understanding of how to use the Syntax Search capability in Verbum to find detailed patterns and structures within biblical texts. Specifically, you will:

  1. Have a better understanding of the capabilities provided by Syntax Search and how to use it
  2. Understand the different resources available to you that support syntax searching and when to use them
  3. Have worked through some examples to learn about some of the different options available
  4. Have thought through some of the things to be aware of when creating and using a Syntax Search

 

Skip ahead to:

What Resource Should I Use?

Constructing a Syntax Search

Some Examples in Action

Specifying Optional Search Elements

 

What Resource Should I Use?

Syntax Searches do not run against English Bibles nor the standard original language texts, such as NA28 or UBS5. Rather they run against original language versions of the Bible that have been specifically tagged to show the associated grammatical and syntactical structures. It can be helpful to think of this tagging as a tree structure with a number of branches coming off a central stem, feeding down to smaller branches, and eventually getting to the “leaves” at the end that correspond to the actual words.

In this section you will:

  • Find out which resources are available (and this will vary based on your library)
  • Get a basic understanding of what information these resources contain
  • Learn some of the key differences between them and which would be most helpful for different types of searches
  • Find out how to select the specific resource you are going to use

 

Understanding the Capabilities of the Different Resources

There are three different sets of resources available in Verbum that can be used for Syntax Searching:

They each have different types of information - and it is important to understand these so that you can choose the most appropriate resource for the particular search you want to carry out. To illustrate this, we’ll look at how each of these resources handles John 1:1.

 

OpenText Tagging of John 1:1

To open this resource from the library, enter the command Open opentext in your command box and select the Clause Analysis resource from the dropdown menu.

If you don’t have this resource, it is available here.

Then find the passage you are interested in by entering it in the reference box.

The OpenText resource analyzes John 1:1 as shown below. :

It shows there are three distinct clauses - indicated by the PC (which stands for Primary Clause) but does not give any indication of how these clauses are related. It is a relatively simple structure but uses terms and abbreviations which can take some getting used to:

  • S indicates the Subject of the clause.
  • P indicates the Predicator of the clause, which can be considered as the verbal component.
  • C indicates a Complement while A indicates an Adjunct, both of which are related to the Predicator, or verbal function. The complement “completes” the predicator so, in the third clause, it completes the statement about what the word was. The adjunct provides information about what was going on so, in the second clause, it shows where the word was.
  • cj Indicates a Conjunction that serves to link clauses together.
  • wg is a word group that, in general, contains a group of words. As can be seen from the example above, a word group often contains a single word.
  • hd is a head term, which is a way of indicating a word that does not modify any others words in its word group.
  • sp is a modifier that classifies or identifies the word it modifies. So in the third clause above, λόγος is the head term with the specifier making it clear that the word being spoken about is “the word.”

It is a simple, hierarchical structure with quite a limited set of terms. This can make it easy to use once you are familiar with the terminology but it is somewhat limited compared to the others. To see the definition of all terms used, consult the glossary.

One of the uses of these searches is to find where structures that appear in a particular verse appear elsewhere in the New Testament. So to find all occurrences in the New Testament with structures that match those referring to the λόγος in each clause of John 1:1, and that also include the word λόγος, you need a search similar to the one shown below (we will look at how to construct these searches later in the article):

Note how the same terms are used and how the structure matches what is shown in the Clause Analysis resource.

This shows that this construct, where the word λόγος is used along with this particular syntactic pattern, occurs 247 times in the Greek New Testament including three times in the verse we are looking at.

Note that the results for John 2:22 and 4:39 highlight much larger portions of text. Looking at the clause analysis shows why this is the case. For example, John 2:22 has:

The entire structure outlined in blue above matches the search definition which explains why more words are highlighted in the results.

 

Cascadia Tagging of John 1:1

To open this resource from the library, enter the command Open cascadia in your command box and select one of the Syntax Graphs resources from the dropdown menu. The different resources relate to different versions of the Greek New Testament. In this article, we will use the SBL edition (based on the text from the Society of Biblical Literature).

If you don’t have this resource, it is available here, in the Verbum 9 Full Feature Upgrade, and in the Verbum 9 Gold package.

Then find the passage you are interested in by entering it in the reference box.

The Cascadia resource analyzes John 1:1 as shown below:

Note that it includes the concept of a sentence indicated by Sn.

It shows that there are three separate clauses (indicated by CL) joined together by conjunctions indicated by conj.

Some of the terms used are:

  • S indicates the Subject Function of the clause
  • VC indicates a Verbal Copula Function - one of a number of verb indicators with this one indicating that there is a relationship between the Subject and the Verb as opposed to the Subject carrying out a particular action
  • P indicates a Predicate Function which states something about the Subject
  • np indicates a Nominal Phrase which is a phrase with a noun (or a word acting as a noun) at its head
  • pp indicates a Prepositional Phrase which contains a preposition and the nominal phrase it governs

It is much richer and more detailed than the OpenText analysis and enables the construction of more focused searches. To see the definition of all terms used, consult the glossary.

As before, to find all occurrences in the New Testament with structures that match those referring to the λόγος in each clause of John 1:1, and that also include the word λόγος, you need a search similar to the one shown below.

Note how the same terms are used and how the structure matches what is shown in the Clause Analysis resource. But it also includes a Terminal Node to indicate that it is introducing the word at the furthest end of the branch structure (it is not strictly necessary to use it but it does provide the most specific match - we will show later in the article how to avoid including this).

This shows that this construct, where the word λόγος is used along with this particular syntactic pattern, occurs 101 times in the Greek New Testament including three times in the verse we are looking at. The more detailed search parameters result in a more focused set of results.

Interestingly, this does not return results for John 2:22 or 4:39 which, in the case of the OpenText search, returned results with more of the text highlighted. Again, to understand why this is the case we need to look at the tagging of those verses. Taking John 2:22 as an example, the structure in this resource is:

The reason for this verse not being returned is the additional nominal phrase inserted in the structure (outlined in blue above). We could extend the search to include this phrase but then it would not return results for John 1:1.

If we want the exact structure as found in John 1:1, then the search as defined is the correct one. If, however, we want to include these different cases we can use the Matching skips levels option (available in the General section of the search definition). This instructs the search to ignore levels in the hierarchy if necessary.

The correct place to use it, in this instance, is on the Terminal Node definition as this is the element that this new nominal phrase is under.

Once specified, the arrow connecting the elements is shown as a dashed line to indicate this option has been set:

Running this new search includes the additional results expected:

 

Lexham Tagging of John 1:1

To open this resource from the library, enter the command Open Lexham Sentence in your command box and select one of the Syntax Graphs resources from the dropdown menu. The name of the resource suggests that its focus is on how sentences are structured and may be particularly useful to those with a sentence diagramming background. The different resources relate to different versions of the Greek New Testament.

In this article, we will use the SBL edition (based on the text from the Society of Biblical Literature). If you don’t have this resource, it is available in The Lexham Syntactic Greek New Testament: SBL Edition which is included in the Verbum 9 Silver.

Then find the passage you are interested in by entering it in the reference box.

The Lexham resource analyzes John 1:1 as shown below:

The Sentence boundaries are clearly marked with a number of clauses appearing within them.

It introduces the idea of a Segment Clause(indicated by Seg Cl) where multiple clauses are connected by a conjunction which, itself, is not actually labeled so is not searchable. The only additional term used in this verse is Prep Ph that indicates a phrase that contains a preposition and is governed by it. Other phrase types do appear elsewhere but, as can be seen, this analysis has quite a simple structure.

More information on terms used is available in The Lexham Syntactic Greek New Testament Glossary

As before, to find all occurrences in the New Testament with structures that match those referring to the λόγος in each clause of John 1:1, and that also include the word λόγος, you need a search similar to the one shown below.

This is straightforward to construct but doesn’t allow the level of specificity available in Cascadia. This shows that this construct, where the word λόγος is used along with this particular syntactic pattern, occurs 101 times in the Greek New Testament including three times in the verse we are looking at.

If you compare these results with those obtained from OpenText, you will see that the search there also returned Luke 24:44. Looking at how the Lexham resources structures that verse shows that it places the word λόγος in a Dialogic Frame which is why it is not returned in the search we specified.

The diagram below shows each of the three searches and their results (this is using the initial version of the Cascadia search without the matching skips levels feature applied):

This demonstrates the importance of understanding the sort of information you are searching for and selecting the appropriate visualization resource as the basis for the search.

 

Selecting a Resource to Search

Having three separate resources available raises the question as to which is better for different types of searches. There is no simple answer to this as they each provide different capabilities, some examples of which are shown below.

The first example looks at the phrase τοῖς πατράσιν in Hebrews 1:1.

The Lexham resource shows this within a Participial Clause:

This provides the ability to search for these words within clauses of this type:

The Opentext resource shows this within a Complement which, as it notes, includes the direct and indirect objects. But there is no mechanism to specify which of the two you’d like to search for

This provides the ability to search for these two words within a Complement:

The Cascadia resource shows the words within an Indirect Object (IO):

This provides the ability to search for these words within clauses containing these types of objects:

The associated search results are shown below:

These vary significantly, both in the number of results and their specificity. So it is important to choose the right resource for the search parameters you are interested in.

Another difference relates to the support for different types of clauses and levels of tagging. The Lexham resource provides a wide range of Clause Types to choose from:

Whereas the corresponding feature in Cascadia (the Clause Function) has a shorter list of options:

This provides greater search granularity for the Lexham resource. The Lexham resource also contains word-level tagging for Syntactic Force, which the others do not.

A search for a word tagged as infinitive of greeting finds just three results:

This type of search in the Lexham resource is not possible against the other resources.

 

Conclusion

As a general rule, balancing ease of use and functionality, a Cascadia-based search is probably the best starting point. But there are some features that are only possible using the Lexham tagging. OpenText would be a good choice if you are particularly familiar with the terminology it uses.

But it is a good idea to take the time to examine the capabilities of each of the resources, to understand the features they provide so you can make the most appropriate choice for any particular searches you need to carry out.

 

In this section, we will look at how we actually construct a Syntax Search with a focus on using the graphical design tool. The basic principles are the same irrespective of which resource you are using - for this section we will use the Cascadia resource.

 

Creating a Syntax Search

There are two ways of creating a new syntax search:

  1. From the Search panel
  2. From the Docs menu - this option is available as Syntax Searches are user-created documents accessible from this menu

 

Creating a Syntax Search from the Search Panel

To create a Syntax Search from the Search panel, select the Syntax option on the far right:

Now, open the dropdown menu to select the resource you want to use (as specified above, we will use the Cascadia resource):

Now open the Query dropdown:

At this point, there are three options:

  1. To select one of a number of predefined templates to search for particular constructs
  2. To reuse Recent searches by clicking them
  3. To create a New Syntax Search

For this article, we will create a new syntax search. Click that option to open the graphical editor used to create, modify and execute Syntax Searches as shown below.

 

Creating a Syntax Search from the Docs Menu

To create a Syntax Search from the Docs menu:

  1. Open the menu
  2. Click New to open the dropdown
  3. Click Syntax Search

This takes you directly to the same point that we reached when starting from the Search panel.

 

Elements of the Syntax Search Graphical Interface

The two buttons outlined in red are used to select the editing mode (via Show Details) or to execute the defined search (the Search button). You can give the search a name by editing the string outlined in blue - we will call it New Syntax Search.

On the right are options, again, to use predefined templates or to select previous searches. You can specify the passages you want to run the search against using the range selector:

And you can specify a different resource to use against which to execute the search:

The blank area (outlined in blue below) is where you construct your search, starting by clicking the + sign to add the first search term.

 

Basic Use of the Editor

Click the + sign to add a term to the search definition. The options on the left correspond to those we saw in the resource previously, the options on the right provide different ways of combining elements and we will come back to look at those later.

The search we used earlier was defined as shown below:

To start building this, we need to reproduce a section from the analysis resource, shown below (with the actual structures we will be searching for outlined in blue):

 

Step 1: Select the First Element

As all the constructs we are interested in start with a phrase, choose a Phrase as the first element

This will produce the first item in the construct.

Note: You can choose from a range of options on the right that correlate to the particular term. Click any element in the defined search to see the options for it.

 

Step 2: Refine the Definition of this Element

Examining the resource, and hovering your cursor over the specific element, shows it is a nominal phrase:

So, select the Nominal phrase type to match the pattern we are looking for.

This results in:

 

Step 3: Add the Next Element

Hover over an element to see its connection points:

Click the one on the right to add it as a subordinate entry - in this case a Terminal Node. Adding a Terminal Node here is not actually something we see directly in the resource, but it does indicate the lowest / furthest element in the structure and so needs to be included before adding in the final element.

 

Step 4: Continue to Add Subordinate Elements as Required

Then add a Word element and specify that it is an article using the Morphology section on the right.

This will complete that branch of the construct as shown below:

 

Step 5: Add Additional Branches to the Structure

The next phrase we need to add in is at the same level in the hierarchy as the Terminal Node. So click that element, then click its lower connection point and add in another Nominal Phrase.

You'll see something like this:

 

Step 6: Add Additional Elements to Complete this New Branch

Connect another Terminal Node and then complete the branch by adding in another Word.

 

Step 7: Specify Details of the Target Word

Specify the lemma we are searching for in the Text & Lemma: Lemma box.

Then, in the Morphology box, specify that we are looking for a noun. The Morphology information is required if we were just specifying a general search that included a noun. As we are searching for a specific noun here it is not strictly necessary but it is included here to demonstrate the principle.

This completes the search definition.

 

Step 8: Execute the Search

Execute the search by clicking the Search button at the top left to generate the results

 

Refining the Search

Our search currently returns 101 results, showing that this particular construct occurs 101 times in this resource. There are a number of ways of refining it, to result in more precise matches such as adding extra elements to the definition.

As shown previously in this article, the resource we are using has the pattern shown below:

So we could add elements to the “front” of the search to specify a Subject Function

and a Verbal Clause.

 

Step 1: Add Elements Further Up the Branch

To start adding higher elements, click the Phrase element on the left of the definition, and add in a Clause Function.

Specify that it is a Subject Function.

 

Step 2: Add in the Next Higher Element

Then, using the same technique, add in a Verbal Clause to the left of that.

 

Step 3: Execute the New Search

Execute the search again to get the refined set of results where this particular construct is present.

 

Some Examples in Action

In this section, we will look at a number of worked examples to show some of the things that are possible.

 

Thinking about “Faith in Christ”

One of the questions that has caused a lot of debate and discussion over recent years is what Paul meant in Galatians 2:16 where there is a phrase that is normally translated as “faith in Christ”. Some have argued that it should be translated as “the faith of Jesus Christ” or the “faithfulness of Jesus Christ”.

The associated structure is a set of nested Nominal Phrases containing three nouns.

Hovering your cursor over any of these nouns will show it is in the Genitive Case.

It is possible to build a Syntax Search to represent this structure using what has been outlined previously in this support article.

Have a go at building it before looking at the example below - and specify that the first word has the lemma πίστις:

Note the requirement to include the Terminal Node element, even though it doesn’t appear in the associated diagram, to introduce the lowest element in the branch. This is required unless using the Match skips levels option described above and demonstrated in the next worked example.

Run the search to find where else this construct appears.

This would suggest that when considering the meaning in Galatians 2:16, it is also important to consider its meaning in Romans 3:22 and Galatians 3:22.

While the search above finds genitive instances, the debate over whether the phrase should refer to of the noun to “faith in Christ” or “faith(fulness) of Christ” depends on which type of genitive is in use as outlined by Richard B. Hays in his book, The Faith of Jesus Christ:

With regard to the phrase Ἰησοῦ Χριστοῦ, the fundamental grammatical question is whether πίστις followed by a proper noun in the genitive case should be understood to mean “faith in X” (objective genitive) or “faith of X” (subjective genitive).

As discussed above, it is possible to use the Lexham resource to search for the syntactic force of a term and this could be useful here. Creating the basic search is a similar process to the process for Cascadia, resulting in a search as shown below:

Running the search produces the same set of results:

To add in the Syntactic Force tagging, click the first element and select the Syntactic Force section on the right:

Scroll down and select the two options so that the search will return a result if either tag has been used on this word:

Running the search, however, returns no results - indicating that this level of tagging has not been applied here:

This is an important reminder that these searches will only find instances in the text that have been appropriately tagged. And even if it had returned hits for one of these options, we would still have had to recognize that this was the opinion of the group tagging the text. So care must be taken when constructing these searches, and care must be taken when interpreting them as well.

 

Matching Verbs in a Sentence

The first two verses in Hebrews show that while God had spoken in many ways previously, he has now spoken through His Son. This is a case where the same verb, at least in English, is used in the same sentence, both in the past tense. In this example, we are going to look at how to find all similar occurrences.

It is a long and complex sentence with the two verbs outlined in blue below:

Hovering the cursor over the first one shows that its morphology is aorist, active, nominative.

Repeating this for the second shows that its morphology is aorist, active, indicative.

We’ll use this information to constrain the search to limit the results - for a wider set of results, fewer constraints could be used.

We are going to use a very simple structure, using the Matching skips levels option so we are not limiting ourselves to the specific structure in these verses. We also need to introduce another option, the Anything element, to indicate that we may be missing out on branches in the structure.

While the Matching skips levels feature allows matches where not all elements down the tree are shown, the Anything feature allows matches where not all branches are shown.

 

Step 1: Build the Basic Search Structure

So the search starts off very simply with a Sentence containing a Word (with Matching skips levels selected):

 

Step 2: Introduce the Anything Element

Then add in the Anything element beneath the Word:

This results in:

 

Step 3: Add in the Next Verb and Specify Its Morphology

Then add the second Verb underneath:

Modify the morphology of the first so it matches the passage in Hebrews:

 

Step 4: Add in the Next Verb

And then do the same for the next:

 

Step 5: Specify the Matching Criteria using the Agreement Feature

To search for matching verbs, we need to use the Agreement feature, by creating a Rule.

 

Step 6: Specify that the Results Will Be Highlighted

Finally, set the Highlight this term in results option for both verbs:

 

Step 7: Execute the Search

Run the search to find all occurrences of this type:

 

Exploring Other Matching Options

The example above makes use of lemma matching but, as is shown below, there are many options for matching available:

With the ability to apply this matching across multiple levels and elements within a search construct, this provides very powerful features to find repeated constructs and even variants.

For example, at the beginning of Romans 1, there is a phrase containing four nouns, the first two in the nominative and the last two in the genitive.

A search to find this type of construct, including a number of agreement rules is shown below:

This returns a number of results including the one in Romans 1:1:

We could now introduce a disagreement - in this case, specifying that the number of the last element is different from that of the third:

Running the search now returns a much-reduced set of results:

This provides very powerful capabilities for finding very specific combinations of terms and patterns.

 

Searching for Repeated Phrases

One of the repeating elements in the second and third chapters of Revelation is that of writing to the angel of the churches - such as:

To the angel of the church in Ephesus write: These are the words of him who holds the seven stars in his right hand and walks among the seven golden lampstands. (Revelation 2:1)

To the angel of the church in Smyrna write: These are the words of him who is the First and the Last, who died and came to life again. (Revelation 2:8)

To search for this type of construct, we can examine, and replicate, the structure in Revelation 2:1:

This involves building a search construct that:

  • Starts with a Sentence that contains
    • A Verbal Clause that contains
      • An Indirect Object Function
      • A Verbal Function

The search looks like this:

However, running the search does return a result for Revelation 2:1, it does not include the other similar references such as Revelation 2:8. This is a helpful reminder that it is important to recognize that what we see in the English text may not match exactly what is in the Greek.

Looking at the structure of Revelation 2:8, shows that there is a conjunction joining the sections together and that this conjunction (as well as the previous construct) is preceded by an additional clause:

This search structure would look like this:

Running this search does find the later references but not the one for Revelation 2:1

To combine these two, use the Or function that indicates that either one expression or the other is present. Starting from the initial construct, add an Or to the bottom:

This gives you:

Then add the second structure underneath that:

This search returns both sets of results and other places in the New Testament where these constructs occur:

This is an example where running the search against a different resource can result in a simpler search to construct, but less focused results.

Looking at the Lexham structure for Revelation 2:1 shows that it comprises a:

  • Segment Clause that contains
    • An Article
    • A Noun
    • A Relative Clause that contains
      • A Prepositional Phrase

Comparing this with Revelation 2:8 shows that the only structural difference in this resource is introducing the Conjunction καί before the first article. This allows us to use the same search to find both types - as it is not necessary to specify every term in the tree, so the term for καί does not need to be included . So this results in a search such as:

This results in:

So, as mentioned above, thinking about exactly what you are interested in and considering the structure of the different resources can help determine the best combination of resource and search construct.

 

Searching for Particular Constructs

Sometimes it is helpful to find occurrences of particular Greek constructions. In this example, we will look at finding periphrastic participles. These are where the word εἰμί is used with an associated participle to communicate a single idea.

An example of this is Mark 13:25 which says that the stars will fall with:

  • will translating the future tense of εἰμί
  • fall translating the participle of πίπτω

The corresponding construct looks like this:

To start building the search, we specify a verbal phrase that has, as one of its elements, the verb εἰμί.

Note: We have used the Matching skips levels (indicated by the dotted arrow) feature to avoid having to build the entire tree structure.

We now add in the second verbal phrase and indicate that it contains a participle:

Note: We have specified the morphology at the phrase level, using the Head Term Morphology option which removes the need to add in the extra levels of the structure and specify it there. We have also selected the option to highlight this participle in search results.

Running the search gives a set of results including Mark 13:25 as expected:

However, sometimes terms are ordered differently in Greek while retaining the same meaning which could result in the Greek participle appearing before the verb εἰμί. We can allow for this in a Syntax Search using the Unordered Group feature. Click the + button to the right of the first Phrase element and select Unordered Group.

This will extend the search definition to this:

Re-running the search returns 202 hits as opposed to the initial 171.

Note: This may not be find all occurrences of periphrastic participles, just those that are found in this particular construct. For example, it doesn’t find the participle in Mark 7:15 which has εἰμί in a Verbal Copula and where the whole clause is rather more complex.

 

Specifying Optional Search Elements

It is important to remember that the search matches precisely the construct defined, although there are ways of providing flexibility as well. Take, for example, the structure in John 3:16 that speaks about the one and only Son.

A search for this construct in John’s Gospel, where the Adjectival Phrase is highlighted in the results, would look like this:

It gives the following results:

But if you are interested in these constructs whether or not the Adjectival Phrase is present, you can specify that it is optional:

This gives a much wider set of results, while still highlighting the Adjectival Phrase where present:

 

Conclusion

Working through this tutorial you have found out about:

  • What the Syntax Search is, what it enables, and how to use it
  • The resources you have in your library that support Syntax Searching and when to use them
  • Some ways in which to construct Syntax Searches to find particular patterns
  • Some of the things to look out for when constructing Syntax Searches

A good understanding of this capability will enable you to find detailed grammatical patterns in the New Testament to aid you in your study. If you have questions or want to explore things further, please ask questions or discuss what you are trying to do on the Verbum community forums. There are other advanced Verbum users there who will engage with you and provide pointers and advice.

Was this article helpful?
Suggest an improvement or request a feature

We're sorry to hear that! Tell us why.

Have more questions? Ask the community.

Can't find what you need? Contact us!