The linguistic and historical questions that surround Acts 20:28 are admittedly intricate, but none of them yield the verdict that the traditional rendering—“the church of God, which he purchased with his own blood”—is either impossible Greek or a late theological invention. The case can be set out in four coordinated observations.
First, the grammar. In Luke-Acts every instance of the double-attributive construction noun + article + ἴδιος is straightforwardly adjectival (“their own hands,” “his own life,” “his own disciples”). The collocation αἷμα + τοῦ ἰδίου therefore conforms to Luke’s regular usage and to the broader Koine pattern attested in papyri and in Hebrews 13:12: διὰ τοῦ ἰδίου αἵματος, “by his own blood.” The alternative, a possessive genitive suppressing an implied “Son,” cannot be ruled out a priori, but it must be judged the marked reading, for it would create in Luke a syntactic innovation without parallel elsewhere in his writings. It is this grammatical datum—not an a priori “Trinitarian lens”—that prompts most essentially literal translations (RSV 1952, NASB 2020, ESV, CSB, LSB, NET) to place “his own blood” in the text and leave the possessive paraphrase to a foot-note. Dynamic-equivalence versions that supply “Son” in the running text do so expressly to aid readability, not because ὁ ἴδιος forces the supplement.
Second, the manuscript evidence. Modern critical editions (NA 28, UBS 5, SBL-GNT, Tyndale House GNT) agree that ἐκκλησία τοῦ θεοῦ is the earlier reading; ἐκκλησία τοῦ κυρίου is a well-attested early change (𝔓74 A D Ψ 33) that plausibly arose precisely because scribes stumbled over the implication that “God” could be said to bleed. That impulse reveals how naturally second- and third-century copyists heard the Greek: if τοῦ ἰδίου is read attributively, to say that “God purchased the church in his own blood” is to identify the crucified Lord with the God whom Paul and Luke elsewhere call θεός.
Third, the first-century horizon. To claim that no reader in Luke’s generation could imagine divine blood is to overlook the apostolic idiom itself. Paul can speak of “the Lord of glory” being crucified (1 Cor 2:8); John can report Thomas hailing the risen Christ as ὁ θεός μου (John 20:28); the author of Hebrews can ascribe the Exodus to Christ (Heb 11:26). Within two decades of Acts, Ignatius of Antioch could exhort the Ephesians to “be renewed in the blood of God” (Eph 1:1) and the Smyrnaeans to “await the one who is above time—timeless, unseen—who for us became visible, impassible, yet for us suffered” (Smyrn 3). Ignatius’ diction is not a post-Nicene artifice; it is the natural outgrowth of the communicatio idiomatum already visible in the NT. Luke’s theology of God’s self-involvement in the saving mission of his Messiah (cf. Acts 2:36; 3:15; 20:28) therefore sits comfortably in its first-century milieu.
Fourth, the appeal to Raymond E. Brown or Murray J. Harris. Both scholars are careful, not skeptical. Brown’s caution—“we are by no means certain that this verse calls Jesus God”—registers precisely the point at issue: the verse is not the sole datum on which Christology stands, but neither can its most natural sense be suppressed. Harris’ own conclusion (“unlikely, although not impossible”) rests on the grammatical judgment that τοῦ ἰδίου might be a Christological title. Yet his prior 130 pages have already conceded multiple unambiguous uses of θεός for Christ; his refusal to add Acts 20:28 to that list hardly vindicates the NWT’s conjectural “Son.” On the contrary, Harris specifies that if one construes ἰδίου adjectivally (the default Greek), “God” refers to Jesus and the phrase enshrines the ancient confession that the Church was bought in divine blood.
The subsidiary assertions advanced in support of the NWT paraphrase fare no better. “Begotten” does not mean “created” in either Second-Temple Jewish discourse or Nicene theology; Proverbs 8 is read christologically because Wisdom there is said to be with God “before the ages,” not because the verb κτἰζω licenses Arianism; θεότης and θειότης are cognates but denote, respectively, the fullness of deity and divine quality; Athanasius rejected any formula that made the Son merely “like” the Father because only ὁμοούσιος safeguards the apostolic proclamation that all God is, the Son is. None of those points stands or falls with Acts 20:28, but each underscores why the church perceived that a merely subordinate Christ could not account for the language of Scripture.
A translation such as “the church of God which he purchased with the blood of his Own” is grammatically serviceable, yet it concedes the very logic the objection resists. If “his Own” is elliptical for the Son, then the verse still predicates the redemptive act to God and ascribes the cost—the cross—to the one who is God’s Own in a unique sense. Luke’s audience, steeped in a narrative where the exalted Jesus bears divine titles and exercises divine prerogatives, would hardly have balked at that communion of subject and predicate.
The bulk of the translation tradition, the earliest patristic witness, and the syntax of Luke-Acts therefore converge: Acts 20:28 most naturally speaks of the flock as God’s precisely because the shedding of Christ’s blood is the act by which God himself secures it. That reading neither invents a later metaphysic nor stretches weak evidence; it follows the Greek where it leads and lets the theology of the NT say what it has said from the first generation onward.