Mark 1:11 (and Matt 3:17; 17:5; Luke 3:22; 2 Pet 1:17)

The verb εὐδοκέω

Part of determining the best English equivalent of εὐδοκέω is carefully distinguishing between the available options. The states delight and be well pleased entail that the subject has a positive state of mind or emotion about something. Our first question is whether there are contexts where εὐδοκέω cannot refer to a positive emotion. If this can be shown, then the verb itself does not contribute a positive emotion like delight / be well pleased. Indeed, there are examples where εὐδοκέω takes an argument which would clearly elicit a negative emotion in which the subject is not delighting.

• (357) τότε ἐντραπήσεται ἡ καρδία αὐτῶν ἡ ἀπερίτμητος, καὶ τότε εὐδοκήσουσιν τὰς ἁμαρτίας αὐτῶν. (Lev 26:41)
• Then their uncircumcised hearts will be put to shame, and then they will accept their sins
• (358) καὶ πολλοὶ ἀπὸ Ἰσραὴλ εὐδόκησαν τῇ λατρείᾳ αὐτοῦ. (1 Macc 1:43)
• And many from Israel consented to serving him
• (359) Διὸ μηκέτι στέγοντες εὐδοκήσαμεν
• Therefore no longer being able to bear it, we accepted being left in Athens alone

In all these examples, the subject of the verb εὐδοκέω does not show positive emotion towards the argument of the verb. Leviticus 26:41 describes the people repenting, and when they εὐδοκήσουσιν their sins, they accept their guilt; they are not expressing a positive emotion in their sins as they repent. In fact, the context would demand the exact opposite emotional response – the people mourn over their sin. This example alone shows that εὐδοκέω is fundamentally distinct from a verb like delight, which would be impossible in this context. The context in 1 Macc 1:43 is similar in that the people who ‘approve’ serving the king are not delighting in his service but are doing so reluctantly. Finally, 1 Thess 3:1 explicitly says that Paul was feeling a negative emotion, but despite this, he still ‘accepted’ being left in Athens alone. It would be a contradiction to translate this as ‘delight.’

In other examples, neither a positive emotion nor a negative emotion is present, but something is agreed to or accepted. Sometimes this occurs where two options are presented before someone.

• (360) καὶ εὐδόκησαν
• And they agreed thus
• (361) καὶ εὐδόκησαν ἐν Ἀλεξάνδρῳ, ὅτι αὐτὸς ἐγένετο αὐτοῖς ἀρχηγὸς λόγων εἰρηνικῶν (1 Macc 10:47)
• And they approved Alexander, because he had become a leader to them of peaceful words

In Tob 5:17, Tobit is speaking to Azariah, and he proposes a deal to him, namely to pay him a certain wage. After Tobit finishes his proposal, both parties ‘agree’ to the terms (referred to by οὕτως ‘thus’). Again, no positive emotion is entailed. In 1 Macc 10:47, Jonathan and the people just finished hearing the letter of King Demetrius read to them, and 1 Macc 10:46 states that they did not believe or accept them. Contrary to how they responded to King Demetrius’s proposal to them, it says of the Judeans that they ‘approved’ Alexander. This does not necessarily mean they had positive emotions about him, though it does not contradict that. It simply means that they approved Alexander as a leader rather than Demetrius.

All these examples suggest that εὐδοκέω need not refer to a positive emotion like the predicates delight and be well pleased. Where, then, is the translation coming from? There are three issues here. First, I suppose that the biggest reason for the translation ‘be well pleased’ is the desire to be “morpho-syntactically transparent” between English and Greek. The verb εὐδοκέω is made up of the adverb εὐ ‘well’ and the verb δοκέω ‘seem.’ In the long line of “literal” English translations, there has been a general desire to match one morpheme in English to another morpheme in Greek. If we break down the word into its parts, the verb εὐδοκέω should be translated with ‘well’ in it. As we have seen with particle verbs in general (5.2.2), this does not work. In addition to the parallelism between εὐ and ‘well,’ the argument ἐν σοί ‘in you’ can also conveniently be translated “literally” when used with the predicates be well pleased or delight (more on this below). It appears that εὐδοκέω ἔν τινι would map well onto ‘be well pleased/delight in something’ – if the translations accurately reflected the meaning of the original. The second reason is that Bible translations in general are very conservative with iconic verses. It is no surprise that all major English Bible translations essentially follow the KJV on this verse (which is itself known for being very “literal”). The third reason for the “emotion” translation, however, is more robust. There are some examples that do seem to refer to a positive emotion.

• (362) ὅτι εὐδοκεῖ κύριος ἐν λαῷ αὐτοῦ καὶ ὑψώσει πραεῖς ἐν σωτηρίᾳ. (Ps 149:3–4)
• Because the lord delights in his people, and he will lift up the humble with salvation
• (363) ἐὰν ὑποστείληται, οὐκ εὐδοκεῖ ἡ ψυχή μου ἐν αὐτῷ (Hab 2:4)
• If he withdraws, my soul does not delight in him

Notice, all the unambiguous examples have a key feature in common – they are found in the present/Luo form. Indeed, this is consistent in both the LXX and NT. I say “unambiguous” because there are many cases where someone “approves” someone or something that they do and exhibits positive emotions. For example, BDAG lists 2 Kgdms 22:20 as the first example of ‘be well pleased, take delight’:

• (364) καὶ ἐξήγαγέν με εἰς πλατυσμὸν καὶ ἐξείλατό με, ὅτι εὐδόκησεν ἐν ἐμοί.
• And he led me out into a large space and he rescued me, because he approved me/delighted in me

This comes in the middle of David’s “song of deliverance.” The question is whether David’s statement is best understood as referring to God having a positive emotion about David or God having approved him to lead as king. We know that both are true, and neither fits the context better than the other. In contrast, a verse like Psalm 151:5 probably refers to David’s brothers not being approved/chosen by God in contrast to him:

• (365) εὐδόκησεν ἐν αὐτοῖς κύριος. (Ps 151:4b–5)
• And he anointed me with the oil of his anointing. My brothers were beautiful and big, and the lord did not approve them

The contrast between David and his brothers is very similar to what we saw in 1 Macc 10:47 where the people approved of Alexander rather than Demetrius. David says that God anointed him rather than his brothers. This is not to say that God did not delight in his brothers. In the context, the issue at hand is not God’s delight in one party over the other but God’s choice between David and his brothers.

Before moving onto performatives, we must address a potential counterargument to this data, namely that the argument ἔν τινι is mapping onto a direct object in English in ‘approve someone/something.’ This was part of the reason why we suggested English translations have opted for ‘delight/be well pleased in someone .’ However, English is notorious for having an extremely broad accusative case. This means that the same argument that surfaces as an accusative argument in English (like ‘approve him ’) might surface as a PP complement or an oblique argument (i.e. an argument in dative or genitive case) in another language like Greek. This seems to be at least part of what is happening here. The verb εὐδοκέω can take an accusative argument (357), a dative argument (358), an infinitive complement (359), or a PP complement (361). There does not seem to be a clear distinction in meaning when these various arguments are used, but the data do suggest a difference between εὐδοκέω used with the Luo form versus the Aorist. Given what we have said about the meaning of εὐδοκέω and the meaning of Luo and the Aorist, this difference is understandable. Based on all the data, the verb εὐδοκέω cannot entail a positive emotion like delight, but some contexts carry this inference. A positive emotion would be a state, and we have seen that the Luo form combines easily with states (6.3.3.2), whereas the Aorist does not (see summary in 44.1.3.2). It would make sense for the Luo form to normally select the positive emotion sense of εὐδοκέω. On the other hand, εὐδοκέω often does not refer to a state but refers to an event of approving or accepting something where no positive state is entailed (though it may be inferred at times). Because the Aorist would be the natural form to refer to an event with an endpoint, it is normally used for the approval sense of εὐδοκέω. Since the Aorist is found in Mark 1:11, it is this sense that is most relevant for our passage.

Performatives across languages and in Greek

The philosopher J. L. Austin was the first to describe performatives, and his description of them involved two essential characteristics: 1) “they do not ‘describe’… anything at all,” i.e. they “are not ‘true or false;’” and 2) “the uttering of the sentence is, or is a part of, the doing of an action” (Austin 1975:5). Austin then provides several examples of performatives:

• (366) a. ‘I do (sc. take this woman to be my lawful wedded wife)’ – as uttered in the course of the marriage ceremony.
• b. ‘I name this ship the Queen Elizabeth ’ – as uttered when smashing the bottle against the stem.
• c. ‘I give and bequeath my watch to my brother’ – as occurring in a will. d. ‘I bet you sixpence it will rain tomorrow.’

Most of these examples could be updated to be more intelligible to twenty-first century English speakers, but in each case, the uttering of the sentence is not being used to convey information. When a bride or groom utters “I do,” something happens, namely a commitment that changes the legal status of the individuals. By declaring the name of a previously unnamed object, the object’s name changes. By writing something in your will, ownership will change when the condition is met. By saying I bet you, an offer is extended. In every case, the world is changed in some way by virtue of the speaker uttering the sentence. This implies that the speaker must meet the prerequisites for performing the action, such as being in the middle of a legal wedding ceremony (a), having ownership over the ship (b) or one’s inheritance (c), or possessing the authority to make a bet (d). Austin discusses these “felicity conditions” in detail (1975:13–15), but we might summarize them as the requirement for the speaker to possess the authority to perform the action through uttering it.

Not only do these sentences have similar kinds of effects on the world, they also share several grammatical properties. First, each sentence starts with a pronoun in 1st person. If the speaker is the one doing something by virtue of uttering the sentence, this makes perfect sense. Only the speaker can declare what he or she is doing as the utterance is made. Second, the verbs are all things that can be done by the subject with words, i.e. they are verbal actions and are not states or physical actions. Performatives are not possible with stative verbs like I am hereby tired or purely physical actions like I hereby touch you. Third, at least in English the adverb hereby is possible with performatives and signals to the hearer that the action described in the statement is being done by virtue of it being uttered. Fourth and finally, the verb form is always in the simple present.

The first and second grammatical properties should remain constant across languages because they are essential to the definition of what it means to be a performative and are not dependent upon language-specific features. However, other languages will, of course, have their own adverbs that signal the performative use (or may not have an adverb at all to signal it), and the verb form may vary. Fortuin’s study on performatives across 106 languages finds that there are only tendencies for which verbal forms are used, and “it is not possible to formulate simple and straightforward universally valid rules which predict the TA [tense-aspect marking] of performatives in a particular language” (2019:4). For our purposes, the primary issue is whether past tense morphemes can be used for true performatives. Near the end of a lengthy discussion of tense in performatives, Fortuin explains that some verb forms do indeed appear to be true past tenses (2019:25, 29):

This implies that performatives, which are expressed by verbs referring to dynamic events, should have a past time reference. Similarly, in many languages which employ a perfective performative the same form is used for a recent or immediate past (Kirundi, Luganda, Lucazi, Mian, Totonac). In my view, a speaker using such a form may very well be said to express both the idea of totality and the idea of completion at the moment of speech, implying that when the sentence is uttered the action is completed (i.e. past). This is also suggested by Amharic (Manahlot 1988: 626), where the imperfective, instead of the perfective seems to be used if the action has relevance past the moment of speech (e.g. in the case of ‘beg’), whereas the perfective is used in contexts where in the mind of the speaker the action has already been completed before the sentence is uttered (e.g. in the case of ‘decide’)… I therefore conclude that (with the exception of Old Russian and possibly Latin) the occurrence of perfective (past tense) or perfect performatives can be explained in terms of their aspectual-temporal meaning, which expresses an event as complete and therefore completed (i.e. performed) the moment the sentence is finished.

The data from Fortuin’s study suggests that a past-tense form is not semantically incompatible with performatives, even if we would never say I hereby named this ship the ‘Queen Elizabeth’ in English. In fact, there may be certain kinds of performatives where a past-perfective form is more suitable if the language allows for multiple performative forms (as Greek does – the Luo form is also possible). To anticipate the below discussion, ‘approve’ (similar to decide) would fit Fortuin’s description of the kind of predicate that would be “completed before the sentence is uttered.”

Corien Bary’s study (2012) on the Ancient Greek tragic Aorist investigates the Greek data from a semantics perspective and attempts to explain how a past-perfective form could be used for a performative. She notes that performatives can be found in both the present/Luo form as well as the Aorist. Her reasoning is that performatives should semantically require perfective aspect, since the entire situation is included in the time when the utterance takes place, but they should also be present tense because the action happens in the speaker’s present as the utterance is taking place. Since Greek lacks a present-perfective form, it must accommodate by using either an imperfective-present form or a perfective-past form. Both strategies are employed, but the result is that there is a mismatch between the semantic value of the verb form and the meaning of the performative. There is certainly some merit to this argument. No language has a dedicated “performative” form, and languages will often “accommodate” to specific uses by using less than ideal forms instead of developing a form for each use (which would almost endlessly multiply forms). In this explanation, the past tense meaning of the Aorist would be suppressed, or in Bary’s words, it is “not interpreted” (2012:48). Likewise, the imperfective interpretation of the Luo form would not be interpreted when it is used as a performative.

Fortuin’s explanation differs in that a semantic explanation is given for past performatives. One thing that makes his analysis more attractive is that the cross-linguistic data would seem to require such an explanation even in the absence of the Greek data. Some languages use past-perfective forms for performatives, and this needs to be explained. Instead of revising our semantics of the past or not interpreting it (as Bary does), Fortuin suggests, with whom I agree, that we should revise our semantics of the performative. Performatives are not necessarily present, and they are not homogeneous as a class. They are used with a present tense form in English, but this does not mean that they are incompatible with past tense forms in other tense and aspect systems. Indeed, the data suggest they are not. In some languages, some performatives can be thought of as events that occur in the past in the sense that the event was already performed in the mind of the speaker and is only actualized when it is uttered. With these considerations in mind, let us return to God’s statement to Jesus at his baptism.

The interpretation of εὐδόκησα in Mark 1:11 (and Matt 3:17; 17:5; Luke 3:22; 2 Pet 1:17)

The first hint that ἐν σοὶ εὐδόκησα in Mark 1:11 should be understood as a performative is the fact that the verb is in first person. This meets the first grammatical condition for being a performative. Second, we need the correct verb. This is where every English translation excludes the possibility of understanding the clause as a performative. Since performatives require something to happen by virtue of the utterance being made, we have said that they are incompatible with states where nothing happens. The predicates delight and be well pleased are both states, so they would exclude the performative reading in English. However, our analysis of εὐδοκέω showed that the verb only rarely maps onto states with a positive emotion, which the predicates delight and be well pleased would entail. In fact, the data unambiguously show that εὐδοκέω cannot entail a positive emotion, and when it is used with the Aorist, it often refers to an action of acceptance or approval with no specific kind of emotion attached to it. While we can certainly infer that God has a positive emotion towards Jesus from the context (since he just called him ‘beloved’), it is not the verb εὐδόκησα that specifies this emotion.

Not only is ‘approve’ an event, but it is also the kind of event we need for a performative. In virtue of the utterance being made by the speaker, the argument of the verb would be ‘approved.’ The subject, God, also has the authority to make the pronouncement of Jesus’s approval and would affect something in the world by uttering such a statement. The surrounding crowd would recognize that Jesus is not only someone whom God loved, but he would also be the individual approved and commissioned by God, marking the start of Jesus’s public ministry. Finally, if Fortuin’s suggestion is correct that past-perfective performatives are used with predicates where a mental action has already taken place, the clause would mean that God is not newly approving of Jesus at his baptism, even if he is publicly expressing that for the first time. Since all performatives in English use the simple present tense, this would not be reflected in English, but the use of the Aorist would have signalled to the original readers that God’s approval of Jesus was made before the utterance and actualized publicly with the utterance.

For all these reasons, the clause ἐν σοὶ εὐδόκησα is best understood as a performative, and this alters the meaning of the verse, at least compared to the typical English translations. God is not declaring that he delights in his Son – he is announcing his approval of his son in front of an audience at the start of Jesus’s ministry. When the utterance is made, something happens in the world. Jesus becomes one approved by God to the crowd. Having God’s stamp of approval and receiving it in front of John the Baptist and his followers, Jesus then sets out to preach and usher in the Kingdom of God. We conclude that God’s statement to Jesus should be understood/translated as, “You are my beloved son. I approve you.”

Suggested exegetical exercise

Choose from one of the following passages and describe the meaning of the verb form(s) in context: the imperfective forms of ποιεῖ and ἁμαρτάνω in 1 John 3:4–10, the Aorist forms of δοξάζω in John 13:31, the perfect form of ἐγγίζω in Matt 3:2 (and similar passages), or the future form of ζᾷ in Luke 4:4. Your analysis should include discussion of the lexical meaning of the verb with which the verb form combines.