QUESTION
In Matthew 9:27, how can the subject of the dependent clause (“Jesus”) be dative?
Καὶ παράγοντι ἐκεῖθεν τῷ Ἰησοῦ ἠκολούθησαν δύο τυφλοὶ κράζοντες καὶ λέγοντες ’Ελέησον ἡμᾶς, υἱὸς Δαυείδ. (Nestle 4th ed., 1937)
Καὶ παράγοντι ἐκεῖθεν τῷ Ἰησοῦ ἠκολούθησαν [αὐτῷ] δύο τυφλοὶ κράζοντες καὶ λέγοντες· ἐλέησον ἡμᾶς, υἱὸς Δαυίδ. (NA28)
DENNIS’ RESPONSE
παράγοντι is an oblique circumstantial participle taking its case from a dative object τῷ Ἰησοῦ, which takes its case from the dative-governing verb ἠκολούθησαν. That is the chain of syntactical case causality, even though it flows opposite the word order direction (the speaker was able to think ahead). Circumstantial participles have implicit subjects but lack syntactical grammatical subjects except for the genitive absolute. This is a useful general grammar rule in our project Greek New Testament – Grammatical Commentary (GNT-GC).
Dative τῷ Ἰησοῦ is also the implicit subject and reference source of παράγοντι. It is in the swing position between the two verbs. In this construction, it looks like τῷ Ἰησοῦ is the explicit subject of παράγοντι because they are concordant. It looks like τῷ Ἰησοῦ might be dative because παράγοντι is dative. And semantically it works for translators to think of it that way because the goal of the average GNT reader is to get the meaning over into English where thought begins. However, diagramming Greek as if it corresponded to the English translation patterns does not work; it is impossible. παράγοντι is not the verb head of the subject τῷ Ἰησοῦ and τῷ Ἰησοῦ is not the grammatical subject of παράγοντι.
Based on analogy, you previously ventured the opinion that this sort of construction was a dative absolute. The defect in that opinion is that “absolute” actually has a precise prior meaning. It means not case-determined by another word. παράγοντι is an oblique circumstantial participle because its case is determined by τῷ Ἰησοῦ instead of being absolute.
Matthew could have written the Greek sentence the way you think of it based on the English translation if he had wanted to. It would be Καὶ παράγοντος ἐκεῖθεν τοῦ Ἰησοῦ ἠκολούθησαν δύο τυφλοὶ. Matthew had the option to choose whether he wanted to make Jesus the subject of the participle or the main verb even though it is implicitly both. In Greek he must make a choice because of the case system in a way that uninflected English does not. He didn’t choose the way you think in English because oblique circumstantial is not an English construction. It is more idiomatic Greek than the genitive absolute. Nominative absolute is theoretically banned by the Chicago school, but everyone understands it anyway because it has been valid English for a thousand years, and they get its equivalence to Greek genitive absolute.
What the Matthew text actually says is “As he was going from there, two blind men followed Jesus,” not “As Jesus was going from there, two blind men follow him” although they are semantically equivalent. The goal of translators is not to preserve Greek syntax but to render idiomatic English; so they prefer the second sequence as better English, but we cannot read the English syntax back as a guide to the Greek syntax when we are labeling Greek grammatical relations. Theoretically, there is no such thing as a dative subject. That idea is probably a product of imposing English translation syntax on Greek.
However, human language is complex and issued by imperfect humans and may sometimes cross borders in exceptional fringe constructions and fuzz categories and rules even in the GNT. Consider Matthew 8:23,
Καὶ ἐμβάντι αὐτῷ εἰς τὸ πλοῖον ἠκολούθησαν αὐτῷ οἱ μαθηταὶ αὐτοῦ.
Here we have two different instances of αὐτῷ, one as the dative object and the first one as apparently the subject of ἐμβάντι. Since there is no other function for this first αὐτῷ, it may exceptionally be called a dative subject of the oblique participle. It is still not a dative absolute since its case is determined by the verb. Alternatively, one of the datives may be coded as a literary redundant repetition or resumption.
The model here is to make falsifiable definitions for grammatical relations and then test them (so that we start with clear concepts). If they are never violated over a large population, they are absolute rules. If there are a few exceptions, they are general rules (useful mnemonically) and we will provide lists, statistics and user interface access to the exceptions in the forthcoming GNT-GC app.
To illustrate how this works, I have coded 10 other participial subjects (out of over 4000 subjects) besides this one, often ἔκαστος, personalizing the plural to the individual. Probably the most dramatic one is Romans 1:27
ἄρσενες ἐν ἄρσεσιν τὴν ἀσχημοσύνην κατεργαζόμενοι
“men doing something shameful in men,”
although this one also might be coded as merely a literary redundant repetition of the main subject.