The Start Of Instruction To Wives And Husbands—Ephesians 5:21 Or 5:22?

By Benjamin L. Merkle

[Benjamin L. Merkle is Professor of New Testament and Greek at Southeastern Baptist Theological Seminary in Wake Forest, North Carolina.]

Abstract

In Ephesians 5:21 Paul exhorts his readers, “Submit to one another out of reverence for Christ” (NIV). It is well known, however, that the Greek word translated “submit” is a dependent adverbial participle (ὑποτασσόμενοι) that is better translated “submitting.” Consequently, verse 21 is closely associated with the preceding context. Even so, NA28, UBS5, and many major English versions add a sentence and even a paragraph break before verse 21. This article examines the strengths of dividing the text before verse 21 and after verse 21 and argues that the grammar and the syntax of the text support breaking after verse 21.

***

In Ephesians 5:21 Paul exhorts his readers, “Submit to one another out of reverence for Christ” (NIV). It is well known, however, that the Greek word translated “submit” is a dependent adverbial participle (ὑποτασσόμενοι) that is better translated “submitting.” Consequently, verse 21 is closely associated with the preceding context. Barth rightly notes, “For grammatical reasons 5:21 belongs to the preceding verses.”[1] Similarly, Roberts comments, “In the Greek text, [ὑποτασσόμενοι] is a participle, (‘submitting’) that depends grammatically on the imperative (‘be filled’). Thus, verse 21 completes the exhortation that began in verse 18.”[2]

Despite the grammatical and syntactical evidence,[3] NA28, UBS5, and many major English versions add a sentence and even a paragraph break before verse 21. Understanding of the structure of this passage can influence interpretation of the passage (especially the meaning and nature of submission in verses 22-33) and, vice versa, interpretation of the passage can influence how one sees the structure of the passage. As Klyne Snodgrass remarks, “Failure to understand the structure [of this passage] has made this section one of the most misappropriated texts in the Bible.”[4] This article will examine the strengths of dividing the text before and after verse 21, demonstrating that the latter position is more likely, and will look at implications for translations and commentaries.

Where Should The Text Be Divided?

Divisions of biblical texts are somewhat artificial, since the original manuscripts did not mark divisions. Nevertheless, meaning is bound up in the syntax and structure of a text. So, even without punctuation marks or capital letters, it is appropriate to speak of sentences and paragraphs or, better, discourse units. Levinsohn writes, “A ‘discourse unit’ consists of one or more propositions and ‘presents a new development in the discourse.’ ”[5] Thus, the goal is to find the most natural place to see a break in the text. In one sense, nearly everyone agrees that verse 21 is a transitional or hinge verse. Yet it is still appropriate to press the question and ask whether verse 21 is more closely associated with what precedes (vv. 18-20) or with what follows (vv. 22-33).

Dividing The Text Before Verse 21

Many Bible versions and commentators divide the text before verse 21, with verse 21 beginning a new sentence (or paragraph) related to wives and husbands.[6] If a new paragraph begins at verse 21, then this verse can be viewed as the heading for verses 22-33 and therefore as the interpretive grid through which the rest of the text is interpreted. In other words, verse 21—which calls for mutual submission (“submit to one another”)—is often seen as the governing verse of the entire passage. The main reasons for this position include (1) the independent (imperatival) function of the participle (ὑποτασσόμενοι) in verse 21, (2) the common themes of submission and fear in verses 21-33, and (3) the lack of a verb in verse 22.

The Independent (Imperatival) Use of the Participle. Although the majority of commentators and grammars do not consider ὑποτασσόμενοι to be an imperatival (and therefore an independent) participle, this is the position affirmed by a few grammars and commentators, as well as (implicitly) by many English Bible versions.[7] For example, those who affirm this position often quote Blass, Debrunner, and Funk, who refer to the imperatival function as a “peculiar use of a participle in place of a finite verb and without any connection to one, usually in a long series and in an imperatival sense; it is common in Paul and even more so in Peter.”[8] After referencing Ephesians 5:21, they maintain that the participle is “greatly detached from the finite verb and already approaching the imperatival usage.”[9] Thus, as an imperatival participle, the verb functions independently and is not formally attached to the imperative “be filled” (πληροῦσθε) in verse 18.[10]

The Common Themes of Submission and Fear. A second reason that some place a division before verse 21 is the similarity of the themes of submission and fear found in verse 21 and in verses 22-33. Verse 21 has “submitting/submit” (ὑποτασσόμενοι), which is implied in verse 22, and the indicative “submit” (ὑποτάσσεται) occurs in verse 24. Lincoln maintains that verse 21 is transitional because “it contains a new topic which has the same subject matter as the verses which follow—submission.”[11] Likewise, Thielman argues, “The final participial element . . . stands apart from the other three elements conceptually and begins a lengthy section of its own.”[12] He later adds, “The substance of the verse, with its focus on submission, means that it is best taken with what follows and should be understood as an introduction to it.”[13]

In addition to the theme of submission, the theme of fear is also present, forming an inclusio for the section.[14] In verse 21, Paul instructs his readers to submit to one another “in the fear” (φόβῳ) of Christ, and in verse 33 he instructs each wife to fear or respect (φοβῆται) her husband. Consequently, the theme of fear brackets the section, setting it off as a literary unit.

Furthermore, some have noted that the first four participles in verses 19-20 refer to corporate and liturgical behavior, whereas verse 21 regulates ethical behavior outside the worship service.[15]

The Lack of a Verb in Verse 22. The third reason for viewing verse 21 as closely associated with verses 22-33 instead of with what precedes is that verse 22 lacks its own verb, being dependent on the participle (ὑποτασσόμενοι) in verse 21.[16]

Therefore, to break the text after verse 21 leaves verse 22 without a verb. Thus, some argue that the type of submission in verse 21 is precisely the type of submission in verse 22. For example, Keener writes, “It is clear that the submission of verse 22 cannot be other than the submission of verse 21 from the simple fact that the word ‘submitting’ does not even appear in the Greek text of verse 22: it has to be borrowed from verse 21.”[17] Powell similarly argues that verse 21 “cannot be a different submission on the part of the wife since there is no new verb, and the only verb used is one which in fact also governs all other relationships. . . . Submission is not the duty of the wife alone any more than love is the duty of the husband alone!”[18]

In sum, verse 21 is closely linked with verse 22 (1) because verse 21 begins a new sentence with the use of an independent imperatival participle, (2) because of the common themes of submission and fear, and (3) because verse 22 lacks its own verb, being dependent on verse 21.

Dividing The Text After Verse 21

Others prefer to break the text after verse 21.[19] This is the view of Hoehner, who states, “This verse is not the beginning of a new section but a fitting conclusion . . . to the section which deals with the filling by the Spirit beginning with 5:18.”[20] Likewise, Best comments, “It is . . . better to associate [v. 21] with what precedes than with what follows.”[21] There are several reasons to prefer this interpretation, including (1) the dependent (adverbial) use of the participle, (2) the use of a hook word, (3) the use of a vocative to introduce a new section, and (4) the textual variants in verse 22.

The Dependent (Adverbial) Use of the Participle. The first reason to place the section break after verse 21 is that the participle “submitting” (ὑποτασσόμενοι) is almost certainly dependent on the imperative “be filled” (πληροῦσθε) in verse 18 and, as such, is not an independent verb (despite the claim of Blass, Debrunner, and Funk). If the break is made before verse 21, then the participle functions imperatively, but imperatival participles are rare.[22] Winer stated, “ Ὑποτασσόμενοι, like the other participles in vss. 19, 20, certainly belongs with the principal verb πληροῦσθε . . . and is not to be taken for an imperative.”[23] In addition, Larkin, who starts a new paragraph at v. 21, admits, “Save for the absence of a verb in 5:22 there are no syntactic signals that this participle caries imperatival force and thus initiates a new section.”[24] As a general rule, when there is doubt, it is better to view a participle as dependent on a main verb (adverbial) rather than as an independent imperatival participle.[25]

Some, however, argue that the verb still carries an imperatival force. That is, while acknowledging that the verb is not an independent imperatival form, because of its relation to the imperative πληροῦσθε, some maintain that it still conveys “imperatival implications.”[26] If the participle is not imperatival (and almost no one argues that it is), then it is dependent and should not be construed as beginning a new sentence. If the participle is dependent, for it to carry an imperatival force, it must be interpreted as a participle of attendant circumstance.

Two problems arise with such an interpretation. First, the grammar and syntax do not fit this category. Wallace lists five criteria that are found with ninety percent of all instances of attendant circumstance participles: (1) the tense of the participle is usually aorist; (2) the tense of the main verb is usually aorist; (3) the mood of the main verb is usually imperative or indicative; (4) the participle usually precedes the main verb; and (5) the verbs are found in narrative literature.[27] In this case, only one of these criteria is met (#3, the main verb is an imperative). A second problem is that even if it is granted that the participle is an attendant circumstance participle, it should not be treated as introducing a new sentence. Therefore, either the participle is an independent imperatival participle or it is an adverbial participle concluding the previous sentence.

It is best to interpret the participle as adverbial modifying πληροῦσθε (v. 18), communicating result.[28] Indeed, ὑποτασσόμενοι is the last of the string of five anarthrous (present tense) participles that all describe what it looks like when someone is filled by the Spirit. To break the text before the last participle seems somewhat arbitrary. As such, this participle should not be construed as imperatival and certainly not as beginning a new sentence. Wallace states, “The participle follows a string of present adverbial participles, all dependent on πληροῦσθε in v 18. ὑποτασσόμενοι is most naturally taken with them (in spite of its distance).”[29] This distance—thirty-five words removed from the main verb—is simply because it is the fifth in a series of dependent participles.

In fact, a clear majority of modern commentators favor interpreting the participle as dependent. For example, O’Brien comments that the participle ὑποτασσόμενοι “is best read with the four preceding participles—there are no indicators in the Greek that it should be taken otherwise.”[30] Snodgrass similarly writes, “Verses 18-21 are one sentence; with five participles explaining what it means to be filled with the Spirit, the last of which is ‘submitting to one another in the fear of Christ.’ ”[31] Although writing in the NIV Application Commentary, he willingly admits, “The NIV hides the structure of the Greek text.”[32] Many commentators who divide the text before verse 21 readily admit that the participle is dependent and that the grammar does not favor their position. For example, Thielman acknowledges, “Some commentators . . . attach this verse to the previous section, and the grammatical structure of the passage certainly supports this view.”[33]

The Use of a Hook Word. The second reason to view the break after verse 21 is that ὑποτάσσω serves as a “hook word.” Guthrie explains the function of hook words: “By use of a common word at the end of one section and at the beginning of the next the author generated a transition between two sections.”[34] He continues, “The use of this hookword at the end of one section and the beginning of the next makes a smooth transition between the two.”[35]

The connection between Hebrews 1:4 and 1:5 provides an illustration.[36] In verses 1-4, the author demonstrates how Jesus is superior to the Old Testament prophets. He ends verse 4 with the words “having become as much superior to angels as the name he has inherited is more excellent than theirs” (ESV).[37] He then starts the following section by repeating the word “angels.” “For to which of the angels did God ever say” (v. 5, ESV). Commenting on this literary feature, Ellingworth writes, “Mention of angels acts as a typical ‘hook-word’ . . . between vv. 1-4 and 5-18.”[38] Guthrie maintains that Hebrews 1:4 presents a clear example of a “Hooked Key Word,” or a transition effected by “a characteristic term used in the second unit and introduced in the conclusion of the first unit.”[39] Westfall also notes, “There is a cohesive link of ἀγγέλων/ἀγγέλους (angels) in vv. 5, 7, and 13 with ἀγγέλων in 1:4.”[40] Nearly all commentators agree that the proper break occurs after verse 4, even though verse 4 contains the theme of angels found in the subsequent section.[41] Similarly, in Ephesians 5 it is best to view the verb ὑποτάσσω as a hook word.

One difference with the Hebrews 1 example, however, is that ὑποτάσσω is not repeated in Ephesians 5:22 but is merely implied. Furthermore, verse 22 does not imply the same form of the verb (a participle) but a different form (most likely an imperative). But this type of ellipsis should not be too surprising to the careful student of Greek grammar. For instance, Paul sometimes employs zeugma, which is “a special type of ellipsis requiring a different verb to be supplied.”[42] In 1 Timothy 4:3 Paul states that certain false teachers “forbid marriage and require abstinence from foods” (ESV; κωλυόντων γαμεῖν, ἀπέχεσθαι βρωμάτων). The verb supplied by the ESV (“require”) is actually the opposite of the verb previously stated (“forbid”). In other words, Paul states that the false teachers “forbid marriage”; and they do not forbid abstinence from certain foods but rather require it. In 1 Corinthians 3:2 Paul states, “I gave you milk to drink, not solid food [to eat]” (HCSB; γάλα ὑμᾶς ἐπότισα, οὐ βρῶμα).[43] The missing verb (ἐπότισα, “to give drink to”) is really only appropriate for the first half (γάλα, “milk”) and not the second half (βρῶμα, “solid food”) since one can drink milk but must eat solid food. Again, this is not the construction that Paul uses in Ephesians 5, but it does demonstrate that it is common for verbs to be implied, even verbs that are different from those previously mentioned. If that is the case, then Paul could easily imply a different form of the same verb. In addition, in the greater context, the word is not merely implied since it is stated in verse 24 (ὑποτάσσεται).

Because verse 22 does not have its own verb, but borrows one from verse 21, the connection with the preceding verse is undeniable. Consequently, verse 21 functions as a “hinge verse” that not only looks back to the imperative in verse 18 (indicating that being filled with the Spirit involves submitting to others) but also introduces the household codes in 5:22-6:9.[44] That is, the following three paragraphs (5:22-33; 6:1-4; and 6:5-9) provide specific examples of what Spirit-filled submission looks like.[45]

As previously stated, some maintain that the text should be divided before verse 21 because the theme of fear is found in verses 21 and 33, forming an inclusio. But such a literary feature is secondary, not being as determinative of the structure as the syntax of the sentence. Because 6:9 also mentions fear (φόβου), Best suggests that “the two references to fear [in verses 21 and 33] may then be accidental.”[46] In addition, one could argue that breaking the text before verse 21 ignores the “one another” theme found in verse 19 (ἑαυτοῖς) and verse 21 (ἀλλήλοις). Furthermore, some have suggested that a chiasm exists with verses 19-21.[47]

5:19a	relation of believers to other believers
5:19b	relation to God
5:20	relation to God
5:21	relation of believers to other believers

The Use of a Vocative to Introduce a New Section. A third reason to place the break after verse 21 is that verse 22 begins with a vocative (αἱ γυναῖκες, technically a nominative used as a vocative or a nominative of address).[48] As is often the case, the vocative signals the start of a new section. Young notes, “Vocatives or nominatives functioning as vocatives are rarely found within paragraphs (Col. 3:18-4:1).”[49] This usage is confirmed by the use of vocatives to begin the following sections addressing “children” (τὰ τέκνα, 6:1) and “slaves” (οἱ δοῦλοι, 6:5).[50]

In addition, the lack of a conjunction (asyndeton) may also signify a break.[51] Levinsohn writes that asyndeton “is commonly found at the beginning of a new paragraph or section if that unit has its own nucleus.”[52] He adds that “the absence of a conjunction is significant only in connection with other potential boundary features such as vocatives.”[53] Thus the combination of asyndeton and vocative adds further support for a break after verse 21.

The Textual Variants in Verse 22. A final reason to place the break after verse 21 relates to the textual variants found in verse 22. The NA and UBS texts omit the imperatival form of ὑποτάσσω from verse 22 since it is not found in some early manuscripts. The imperative verb ὑποτάσσεσθε (“submit”) is lacking in P46 (c. AD 200), B (Vaticanus), and a few early fathers but is included after γυναῖκες in D, F, and G and after ἀνδράσιν in most Byzantine text-types.[54] The alternate form ὑποτασσέσθωσαν (“let them submit”) is found after γυναῖκες in Ψ and after ἀνδράσιν in א (Sinaiticus), A (Alexandrinus), and other significant manuscripts. The UBS committee preferred the shorter reading because it “accords with the succinct style of the author’s admonitions, and explained the other readings as expansions introduced for the sake of clarity, the main verb being required especially when the words Αἱ γυναῖκες stood at the beginning of a scripture lesson.”[55]

The point here is not to argue for or against the verb’s inclusion (though the original probably lacked the verb).[56] Rather, these variants demonstrate that earlier Christians often favored breaking the text after verse 21. The verb was probably added because the new section was thought to begin at verse 22 and so a verb was supplied. The additional verb would have been especially helpful for lectionary readings, which “could hardly begin with a verbless sentence.”[57] Not surprisingly, most manuscripts include an imperative to solve this apparent difficulty, since there are imperatives in 5:25 (ἀγαπᾶτε), 6:1 (ὑπακούετε), and 6:5 (ὑπακούετε), as well as in the parallel passage in Colossians 3:18 (ὑποτάσσεσθε; cf. 1 Peter 3:1). Thus, the addition of a verb in verse 22 in most manuscripts suggests that the text was often read independently of its preceding context.

The Lingering Discrepancy

Despite this evidence, most modern Greek New Testaments, many English Bible versions,[58] and some commentators still divide the text before verse 21. Both NA28 and UBS5 put a paragraph break before verse 21, and UBS5 even includes the paragraph heading “Wives and Husbands,” which further separates the verse from what precedes. But the NA and the UBS Greek New Testaments have not always put the break before verse 21. Indeed, previous Greek New Testaments consistently put a break after verse 21; placing any kind of break before verse 21 is a relatively recent phenomenon. For example, NA did not change the text until the 26th edition was released in 1979. UBS followed soon after in their third corrected edition in 1983. I was not able to find a single earlier critical edition that placed the break before verse 21.[59]

Another recent development is the change in location for the editorial heading in the NIV. There are no changes in the text and punctuation in the NIV 2011 relative to earlier editions: verse 21 begins both a new sentence and a new paragraph with the participle translated as an imperative (“Submit to one another out of reverence for Christ”); verse 22 likewise begins a new sentence and a new paragraph. However, the section heading “Wives and Husbands” was removed from after verse 21. Instead, there is a new heading before verse 21: “Instructions for Christian Households.”

On the one hand, this change is welcome because it demonstrates that verse 21 is the hinge verse for the entire section (5:22-6:9) and not simply for the section on wives and husbands (5:22-33). On the other hand, it further removes verse 21 from the previous context, which is not helpful.

Why does this discrepancy persist? It is possible that translations and commentators who choose to break the text before verse 21 have been influenced by the NA and UBS texts. If both leading critical Greek New Testament texts agree as to where the text should be divided, that could easily influence later decisions. Many commentators have admitted that the grammar and syntax do not support the division that they are using. For example, Fowl, Larkin, Lincoln, Neufeld, Roberts, Slater, and Thielman all acknowledge that the participle in verse 21 should not be taken as imperatival but is a dependent adverbial participle.

Conclusion

Ephesians 5:21 is certainly a transition or hinge verse that is closely associated with what precedes and with what follows. Consequently, the various positions are complex and cannot be reduced to an “either/or” dichotomy. Nevertheless, based on the grammar and syntax, verse 21 is more closely linked with the preceding context and should not be punctuated so that it is severed from that context. Editors of Greek New Testaments, Bible translators, and commentators need to let the grammar and the syntax of the text speak for itself.

Notes

1. Markus Barth, Ephesians 4-6, Anchor Bible (Garden City, NY: Doubleday, 1974), 608.
2. Mark D. Roberts, Ephesians, Story of God Bible Commentary (Grand Rapids: Zondervan, 2016), 193-94.
3. Ephesians 5:15-21 can be divided into three hortatory sections (5:15-16, 17, 18-21), with each section containing a negative and positive imperative or assertion (βλέπετε . . . πῶς περιπατεῖτε μὴ ὡς ἄσοφοι ἀλλ᾽ ὡς σοφοί [v. 15]; μὴ γίνεσθε ἄφρονες, ἀλλὰ συνίετε [v. 17]; μὴ μεθύσκεσθε . . . ἀλλὰ πληροῦσθε [v. 18]). The final imperative “be filled” (πληροῦσθε; v. 18b) is then followed by five participles: “speaking” (λαλοῦντες; v. 19a), “singing” (ᾄδοντες; v. 19b), “singing praise” (ψάλλοντες; v. 19b), “giving thanks” (εὐχαριστοῦντες; v. 20), and “submitting” (ὑποτασσόμενοι; v. 21).
4. Klyne Snodgrass, Ephesians, NIV Application Commentary (Grand Rapids: Zondervan, 1996), 286.
5. Stephen H. Levinsohn, “Some Constraints on Discourse Development in the Pastoral Epistles,” in Discourse Analysis and the New Testament: Approaches and Results, ed. Stanley E. Porter and Jeffrey T. Reed, Journal for the Study of the New Testament Supplement 170 (Sheffield: Sheffield Academic, 1999), 318.
6. So NA26 (1979), UBS3cor (1983), NA27 (1993), UBS4 (1993), NA28 (2012), UBS5 (2014); RSV (1952), NIV (1984, 2011), NRSV (1989), NLT (1996); T. K. Abbott, Epistles to the Ephesians and to the Colossians, International Critical Commentary (Edinburgh: T. & T. Clark, 1897), 164; Barth, Ephesians 4-6, 585; F. F. Bruce, The Epistles to the Colossians, to Philemon, and to the Ephesians, New International Commentary on the New Testament (Grand Rapids: Eerdmans, 1984), 378; Stephen E. Fowl, Ephesians (Louisville: Westminster/John Knox, 2012), 181; William J. Larkin, Ephesians: A Handbook on the Greek Text, Baylor Handbook on the Greek New Testament (Waco, TX: Baylor University Press, 2009), 120-21; A. T. Lincoln, Ephesians, Word Biblical Commentary (Dallas: Word, 1990), 338; Ralph P. Martin, Ephesians, Colossians, and Philemon, Interpretation (Atlanta: John Knox, 1991), 67; Heinrich A. W. Meyer, Der Brief an die Epheser, Kritisch-exegetischer Kommentar über das Neue Testament 8-9 (1880; reprint, Göttingen: Vandenhoeck & Ruprecht, 2008), 424; C. Leslie Mitton, Ephesians, New Century Bible (London: Marshall, Morgan & Scott, 1976), 194-95; Thomas R. Yoder Neufeld, Ephesians (Scottsdale, PA: Herald, 2002), 253; Heinrich Schlier, Der Brief an die Epheser (Düsseldorf: Patmos, 1957), 250; Rudolf Schnackenburg, Der Brief an die Epheser, Evangelisch-Katholischer Kommentar zum Neuen Testament 10 (Zürich: Benzinger & Neukirchener, 1982), 245; Thomas B. Slater, Ephesians (Macon, GA: Smyth & Helwys, 2012), 150-51; Frank Thielman, Ephesians, Baker Exegetical Commentary on the New Testament (Grand Rapids: Baker, 2010), 353-55; Philip B. Payne, Man and Woman, One in Christ: An Exegetical and Theological Study of Paul’s Letters (Grand Rapids: Zondervan, 2009), 278-79.
7. See NIV, NLT, NRSV, NJB, RSV; Friedrich Blass, Albert Debrunner, and Robert Walter Funk, A Greek Grammar of the New Testament and Other Early Christian Literature, ed. Robert W. Funk, 10th ed. (Chicago: University of Chicago Press, 1961), §468(2); Stanley E. Porter, Verbal Aspect in the Greek of the New Testament, with Reference to Tense and Mood, Studies in Biblical Greek 1 (New York: Peter Lang, 1989), 376-77 (Porter is tentative in this classification); Barth, Ephesians 4-6, 608; Bruce, The Epistles to the Colossians, to Philemon, and to the Ephesians, 383 n. 78; Mitton, Ephesians, 195. Fanning lists the participle as a dependent participle attached to an imperative (Buist M. Fanning, Verbal Aspect in New Testament Greek [Oxford: Oxford University Press, 1990], 386).
8. Blass, Debrunner, and Funk, A Greek Grammar of the New Testament and Other Christian Literature, §468(2).
9. Ibid.
10. Cf. 1 Peter 3:1 ([αἱ] γυναῖκες, ὑποτασσόμεναι τοῖς ἰδίοις ἀνδράσιν), where the participle ὑποτασσόμεναι clearly functions as an imperative.
11. Lincoln, Ephesians, 338; see also p. 352.
12. Thielman, Ephesians, 365.
13. Ibid., 372. Abbott states, “The connexion with the preceding context is, in fact, only in form, that with what follows is in substance” (Ephesians, 164).
14. Fowl, Ephesians, 186; Larkin, Ephesians, 129; Lincoln, Ephesians, 352; Martin, Ephesians, 68; Slater, Ephesians, 151.
15. E.g., Slater, Ephesians, 150; see also Best, Ephesians, 516 (though he does not affirm this position).
16. Bruce, The Epistles to the Colossians, to Philemon, and to the Ephesians, 383 n. 78; Lincoln, Ephesians, 338, 352; Thielman, Ephesians, 375.
17. Craig S. Keener, Paul, Women and Wives: Marriage and Women’s Ministry in the Letters of Paul (Peabody, MA: Hendrickson, 1992), 169.
18. C. Powell, “A Stalemate of Genders? Some Hermeneutical Reflections,” Themelios 17 (1992): 16.
19. So NA25 (1963), UBS1 (1966), UBS2 (1968), UBS3 (1975); KJV (1769), NKJV (1982), NASB (1995), ESV (2001), HCSB (2004), NET (2005), ISV (2011), CSB (2017); Wallace, Greek Grammar, 639, 651; Chrysostom, Ephesians5:22-24 [PG 62:135]; Arnold, Ephesians, 342; E. Best, Ephesians, International Critical Commentary (London: T. & T. Clark, 1998), 515-17; Charles J. Ellicott, A Critical and Grammatical Commentary on St. Paul’s Epistle to the Ephesians (Boston: Gould & Lincoln, 1865), 130; Harold W. Hoehner, Ephesians: An Exegetical Commentary (Grand Rapids: Baker, 2002), 689; Peter T. O’Brien, The Letter to the Ephesians, Pillar New Testament Commentary (Grand Rapids: Eerdmans, 1999), 378-79, 386-88; E. K. Simpson and F. F. Bruce, Commentary on the Epistles to the Ephesians [Simpson] and to the Colossians [Bruce] (Grand Rapids: Eerdmans, 1957), 128; Snodgrass, Ephesians, 307, 313; Ray Summers, Ephesians: Pattern for Christian Living (Nashville: Broadman, 1960), 117; Brooke Foss Westcott, Saint Paul’s Epistle to the Ephesians (Grand Rapids: Eerdmans, 1952), 82. The Latin texts also divide the text here. See Nova Vulgata: Bibliorum Sacrorum Editio (1913, 1986); Novi Testamenti Biblia Graeca et Latina (1959).
20. Hoehner, Ephesians, 716.

21Best, Ephesians, 517.

22Wallace, Greek Grammar, 659 n. 6 (see also p. 639).

23G. B. Winer, A Grammar of the Idiom of the New Testament, trans. and rev. W. F. Moulton, 7th ed. (Andover, MA: Draper, 1892), 351.

24Larkin, Ephesians, 130. See also Eadie, who states, “It is out of all rule . . . to take the participle as an imperative” (John Eadie, A Commentary on the Greek Text of the Epistle of Paul to the Ephesians, 2nd ed. [1861; reprint, Eugene, OR: Wipf & Stock, 1998], 406).

25A. T. Robertson observes, “In general it may be said that no participle should be explained in this way that can properly be connected with a finite verb” (A Grammar of the Greek New Testament in the Light of Historical Research, 4th ed. [Nashville: Broadman, 1934], 1133-34). Similarly, Brooks and Winbery note, “Certainly no participle should be explained as an independent participle if there is any other way to explain it” (James A. Brooks and Carlton L. Winbery, Syntax of New Testament Greek [Washington, DC: University Press of America, 1979], 138). See also Runge, Discourse Grammar, 266-67; Wallace, Greek Grammar, 651, 659 n. 6.

26Fowl, Ephesians, 186.

27Wallace, Greek Grammar, 640-42.

28Wallace, Greek Grammar, 639; Richard A. Young, Intermediate New Testament Greek: A Linguistic and Exegetical Approach (Nashville: B. & H., 1994), 157; Arnold, Ephesians, 344 (who identifies it as an instrumental participle); Eadie, Ephesians, 406; Ellicott, Ephesians, 130; Fowl, Ephesians, 182, 186; Larkin, Ephesians, 129; Lincoln, Ephesians, 338, 365; Neufeld, Ephesians, 255; Roberts, Ephesians, 193-95; Slater, Ephesians, 151; Snodgrass, Ephesians, 287; John R. W. Stott, The Message of Ephesians, The Bible Speaks Today (Downers Grove, IL: InterVarsity, 1979), 207-8; Thielman, Ephesians, 365; Keener, Paul, Women and Wives, 158.

29Wallace, Greek Grammar, 659 n. 6.

30O’Brien, Ephesians, 387 n. 107. Hurley likewise writes, “Verse 21 is thus grammatically a part of verses 18-20” (James B. Hurley, Man and Woman in Biblical Perspective [Grand Rapids: Zondervan, 1981], 139). Hoehner also notes, “Starting a new sentence with a participle with no finite verb and no conjunction would be odd” (Ephesians, 715 n. 6).

31Snodgrass, Ephesians, 286.

32Ibid.

33Thielman, Ephesians, 372. Earlier he writes, “To begin a new major section of the epistle at this point may seem odd, since from a strictly grammatical perspective 5:21 sits in the middle of a sentence that runs from 5:18 to 5:22” (ibid., 365). Bruce likewise admits, “It can be argued that the plea for . . . submissiveness (v. 21) is attached to the preceding paragraph, the participle ὑποτασσόμενοι being coordinate with λαλοῦντες . . . ᾄδοντες καὶ ψάλλοντες (v. 19) and εὐχαριστοῦντες (v. 20)” (The Epistles to the Colossians, to Philemon, and to the Ephesians, 383 n. 78). Additionally, Neufeld comments that v. 21 is “grammatically dependent on the command to be filled with the Spirit (5:18)” (Ephesians, 255 [emphasis original]). The addition of an assumed imperative in verse 22 by most translations (“Wives submit to your husbands”) is sometimes used as evidence for an imperative in verse 21.

34George H. Guthrie, The Structure of Hebrews: A Text-Linguistic Analysis (New York: Brill, 1994), 96.

35Ibid. Similar to this concept is “tail-head linkage.” Runge defines this as “the process of stating an action from the previous clause (the tail) at the beginning of the following clause (the head) in order to more closely link it to the preceding clause” (Runge, Discourse Grammar, 163). Levinsohn similarly states that tail-head linkage is “the repetition, in an adverbial or participial clause at the beginning (the head) of a new sentence, of the main verb and other information that occurred in the previous sentence (the tail)” (Discourse Features of New Testament Greek: A Coursebook on the Information Structure of New Testament Greek, 2nd ed. [Dallas: SIL International, 2000], 197). Although Runge seems to limit this feature to narratives, it also appears in the epistolary genre. Young broadens the application of this discourse feature in his definition. He explains, “A writer may anticipate what will be taken up next as he or she closes a section. Thus something will be mentioned at the close of one section and at the start of the next” (Young, Intermediate Grammar, 254). He cites James and 1 John as examples where such a feature is common.

36See David L. Allen, Hebrews, New American Commentary 35 (Nashville: B. & H., 2010), 170; Peter T. O’Brien, Hebrews, Pillar New Testament Commentary (Grand Rapids: Eerdmans, 2010), 66 n. 1.

37Emphasis added in this and the next citation.

38Paul Ellingworth, The Epistle to the Hebrews, New International Greek Testament Commentary (Grand Rapids: Eerdmans, 1993), 103; see also David Alan Black, “Hebrews 1:1-4: A Study in Discourse Analysis,” Westminster Theological Journal 49 (1987): 192; William L. Lane, Hebrews 1-8, Word Biblical Commentary 47A (Dallas: Word, 1991), 17; Donald Guthrie, Hebrews, Tyndale New Testament Commentary 15 (Grand Rapids: Eerdmans, 1983), 69.

39Guthrie, The Structure of Hebrews, 100. He explains that 1:4 “is both semantically and syntactically tied to the exaltation statement in 1:3. The term γενόμενος in verse 4 is a temporal participle which refers to the finite verb ἐκάθισεν. Thus structurally, verse 4 belongs to Heb. 1:1-4” (p. 101).

40Cynthia Long Westfall, A Discourse Analysis of the Letter to the Hebrews: The Relationship between Form and Meaning, Library of New Testament Studies 297 (London: T. & T. Clark, 2005), 92.

41A few exceptions are R. T. France, “Hebrews,” in Hebrews–Revelation, Expositor’s Bible Commentary 13, rev. ed. (Grand Rapids: Zondervan, 2006), 40; Philip Edgecumbe Hughes, A Commentary on the Epistle to the Hebrews (Grand Rapids: Eerdmans, 1977), 50; Arthur W. Pink, An Exposition of Hebrews (Grand Rapids: Eerdmans, 1954), 42.

42Blass, Debrunner, and Funk, A Greek Grammar of the New Testament and Other Christian Literature, §497(2). See also Robertson, Grammar, 1200-1, 1383; Winer, Grammar, 622; James Hope Moulton, A Grammar of New Testament Greek, vol. 1, 3rd ed. (Edinburgh: T. & T. Clark, 1949), 241.

43See also Luke 1:64 (ἀνεῴχθη δὲ τὸ στόμα αὐτοῦ παραχρῆμα καὶ ἡ γλῶσσα αὐτοῦ); Acts 1:21 (εἰσῆλθεν καὶ ἐξῆλθεν ἐφ᾽ ἡμᾶς = εἰσῆλθεν [ἐφ᾽ ἡμᾶς] καὶ ἐξῆλθεν [παρ᾽ ἡμῶν]); Rev. 1:12 (ἐπέστρεψα βλέπειν τὴν φωνήν).

44Wallace, Greek Grammar, 651, 659 n. 6; O’Brien, Ephesians, 399; George W. Knight, “Husbands and Wives as Analogues of Christ and the Church: Ephesians 5:21-33 and Colossians 3:18-19, ” in Recovering Biblical Manhood and Womanhood: A Response to Evangelical Feminism, ed. John Piper and Wayne Grudem (Wheaton: Crossway, 1991), 166; Hurley, Man and Woman in Biblical Perspective, 139-41; Snodgrass, Ephesians, 287; Larkin, Ephesians, 130; Christopher Ash, Marriage: Sex in the Service of God (Leicester: InterVarsity, 2003), 307. Hurley states, “The broader context demands that we should view verse 21 as a general heading for the section, applicable to the slave/master and parent/child relation as well” (Man and Woman in Biblical Perspective, 139).

45Two English versions attempt to demonstrate this connection: Lexham English Bible (LEB), “being subject to one another out of reverence for Christ—wives to their own husbands as to the Lord”; Common English Bible (CEB), “and submit to each other out of respect for Christ. For example, wives should submit to their husbands as if to the Lord.”

46Best, Ephesians, 516. He maintains that the inclusio should connect verse 21 with the teaching on slaves and masters (6:5-9).

47Ibid., 515-16; Larkin, Ephesians, 130.

48Blass, Debrunner, and Funk, A Greek Grammar of the New Testament and Other Christian Literature, §147(3); Wallace, Greek Grammar, 58.

49Young, Intermediate New Testament Greek, 253.

50See Runge, Discourse Grammar, 117-18. He states, “In Eph 6, the forms of address are needed to clearly identify Paul’s intended audiences as he moves from topic to topic” (p. 117).

51Runge states that “asyndeton is the default [i.e., the most unmarked] means of connecting clauses in the Epistles” (Discourse Grammar, 20). He adds, “Asyndeton is used at the beginning of a new thought, or simply where the relation between clauses is clear” (p. 21). He continues by noting that it may “be used in contexts of close connection, such as moving from generic to specific” (p. 23). See also Winer, Grammar, 537; Robertson, Grammar, 428, 440; Herbert Weir Smyth and Gordon M. Messing, Greek Grammar, rev. ed. (Cambridge: Harvard University Press, 1966), 484-85; Blass, Debrunner, and Funk, A Greek Grammar of the New Testament and Other Christian Literature, §458; Wallace, Greek Grammar, 658; Levinsohn, “Discourse Development,” 318, 330; Stephanie L. Black, Sentence Conjunctions in the Gospel of Matthew:καί, δέ, τότε, γάρ, οὖνand Asyndeton in Narrative Discourse, Journal for the Study of the New Testament Supplement 216 (Sheffield: Sheffield Academic Press, 2002), 182.

52Levinsohn, Discourse Features of New Testament Greek, 276 (see also pp. 118, 278).

53Ibid.

54The form ὑποτάσσεσθε is also found in Colossians 3:18, which makes it a bit suspicious here.

55Bruce M. Metzger, A Textual Commentary on the Greek New Testament, 2nd ed. (Stuttgart: UBS, 1994), 541.

56Most commentators favor its omission, but for an exception, see Hoehner, Ephesians, 730-31.

57Best, Ephesians, 531. See also J. Armitage Robinson, St. Paul’s Epistle to the Ephesians, 2nd ed. (London: MacMillan, 1914), 204.

58English versions that break the text before verse 21 include Moffatt (1926), RSV (1952), JB (1966), TEV (1976), New World Translation (1984), NJB (1985), NRSV (1989), New Berkeley Version (1993), CEV (1995), NLT (1996), the Inclusive Bible (2007), and the NIV (2011).

59See, e.g., Η Καινη Διαθηκη (London: British and Foreign Bible Society, 1923); The New Testament in the Original Greek, ed. Brooke Foss Westcott and Fenton John Anthony Hort (New York: Macmillan, 1948); Novum Testamentum Graece, ed. Alexander Souter (1956); Novum Testamentum Graece, ed. Augustus Hahn and Edvardo Robinson (1842). See also The New Testament in the Original Greek: Byzantine Textform, ed. Maurice A. Robinson and William G. Pierpont (Southborough, MA: Chilton, 2005). They make a new paragraph at verse 22 because they include the verb ὑποτάσσεσθε. One text that places the break before verse 21 is The Greek New Testament: Being the Text Translated in the New English Bible (1961), ed. R. V. G. Tasker (Oxford: Oxford University Press, 1964).