[1] As will be discussed below, not all verbs have both internal and external arguments; many verbs, i.e. intransitive verbs, have only one argument, and for these we would have to decide whether the single argument is internal or external.
[2] The three exceptions are Acts 20:3; 26:6; 1 Cor 10:10. The former two have the verb γίνομαι and refer to a plan or a promise coming into existence through someone. I suggest that this is an idiosyncrasy with this particular interpretation of γίνομαι with these kinds of nouns, since things like plans and promises are made by people. It is thus the nature of the noun that allows for the ὑπό-phrase. The only other exception is with the verb ἀπόλλυμι, which does not have a passive form in the NT-era suggesting that this verb has a 2-way contrast in the aorist paradigm and should be taken more like a medio-passive.
[3] In an earlier stage of Greek, πορεύω did exist, and it meant to ‘transport something.’ When the active form dropped out of use, the middle/passive forms came to mean ‘go’ rather than ‘be transported by something.’
[4] I am indebted to Nick Ellis for the background to this discussion, particularly the link between baptism and following the individual in whose name you would be baptized.
[5] Whether or not the Pharisees ate in a state of purity is a complex oft-debated topic in NT studies. See Furstenberg (2023) for a recent discussion. This verse is evidence that they did and even that the practice had spread to non-Pharisees.
[6] This also accords with the three instances of βαπτίζω in the middle in the LXX, all of which are clearly reflexive (4 Kgdms 5:14; Judith 12:7; Sirach 34:25).
[7] This is against other grammarians who consider this to be a “permissive middle” meaning something like ‘have yourself/get baptized.’ Wallace (1996:425) comments, “If βάπτισαι were a direct middle, the idea would be ‘baptize yourself’—a thoroughly unbiblical concept.” However, this statement is followed with a footnote that says, “this [self-immersion] was the method that Judaism practiced in the first century.” Köstenberger, Merkle, and Plummer (2020:197) agree with Wallace, again citing it as “unbiblical.” However, the fact that Paul is addressing a Jewish mob here that wants to kill him, who would have considered it perfectly natural for Paul to purify himself, is not mentioned by these authors, nor is the fact that Mark 7:4 clearly refers to ritual immersion, both suggesting that this should be understood as ritual immersion.
[8] I describe this as Paul’s presentation of himself, but other options are also possible. The text says that this speech was given in Hebrew (Acts 22:2). Assuming that Luke did not know Hebrew, he might not have known how to translate this part of Paul’s speech. Regardless of the explanation, all of the grammatical evidence points to a difference between the description of the event in Acts 9:18 and Acts 22:16.
[9] There is one example where δένδρον is the subject of περιπατέω, but in this case, a now partially-blind man describes people as looking like walking trees. It is not as if trees actually engaged in the event referred to by περιπατέω.
[10] There is a voluminous literature on this in linguistics. For a recent survey of motion verbs and how other verb classes are split along the manner vs. result divide, see Levin (2015)
[11] She also includes transitive verbs that have middle or passive morphology in her definition, but this is not important for our purposes.
[12] Dowty’s (1991) influential article on proto-agent and proto-patient characteristics lists five verbal entailments for each role which map onto external (proto-agent) and internal (proto-patient) arguments. His list includes:
Proto-agent characteristics:
a. Volitional involvement in the event or state
b. Sentence (and/or perception)
c. Causing an event or change of state in another participant
d. Movement (relative to the position of another participant)
e. (exists independently of the event named by the verb)
Proto-patient characteristics:
a. Undergoes change of state
b. Incremental theme
c. Causally affected by another participant
d. Stationary relative to movement of another participant
e. (does not exist independently of the event, or not at all)
Dowty’s basic idea is that prototypical agents (i.e. agents with all the listed characteristics) are always going to be subjects of transitive verbs, while prototypical patients (i.e. patients with all the listed characteristics) are going to be objects. Applying this to intransitive verbs, arguments that are more agentive are going to be more likely to surface as external arguments with a verb in active voice, while arguments that are more patient-like are more likely to be internal arguments and to go with a verb in middle or passive voice.
Alexiadou, Artemis. “A form-function mismatch?: The case of Greek deponents.” Of trees and birds. A Festschrift for Gisbert Fanselow (2019): 107-117.
Campbell, Constantine R. “Advances in the study of Greek: new insights for reading the New Testament.” (2015).
Doron, Edit. “Voice.” Syntax: Theory and Analysis (2015): 749-776.
Dowty, David. “Thematic proto-roles and argument selection.” language 67.3 (1991): 547-619.
Furstenberg, Yair. Purity and Identity in Ancient Judaism: From the Temple to the Mishnah. Indiana University Press, 2023.
Köstenberger, Andreas J., Benjamin L. Merkle, and Robert L. Plummer. Going Deeper with New Testament Greek, Revised Edition: An Intermediate Study of the Grammar and Syntax of the New Testament. B&H Publishing Group, 2020.
Levin, Beth. English verb classes and alternations: A preliminary investigation. University of Chicago press, 1993.
Levin, Beth. “39. Verb classes within and across languages.” Volume 2 (2015): 1627-1670.
Wallace, Daniel B. Greek grammar beyond the basics: An exegetical syntax of the New Testament. Harper Collins, 1996.
Williams, Edwin. “Argument structure and morphology.” (1981): 81-114.
