Did God die on the cross? Part 1

Read time: 7 minutes

This has been giving me a headache.

Luke has Paul say to the Ephesian elders: “Pay careful attention to yourselves and to all the flock, in which the Holy Spirit has made you overseers, to care for the church of God, which he obtained with his own blood” (Acts 20:28). In a sermon posted last week on The Gospel Coalition site Jason Helopolous insists that the last part means what it says—or, at least, what it appears to say: God saved or preserved the church “through his own blood” (dia tou haimatou tou idiou).

This has to be wrong, surely?

The New Testament nowhere (correct me if I’m wrong) speaks of God dying on the cross or of people being saved by the blood of God—or even of God suffering. The story is that God sent his Son, as he had earlier sent the prophets, to the vineyard of Israel; the Son suffered and died; God made that suffering the ground for the renewal of his people; and God raised his Son from the dead and exalted him to a position of the highest authority. Father and Son are two distinct actors.

When Jesus died on the cross, he was not God, he was not taking the place of God. He was Israel, he was taking the place of Israel. He was the despised but righteous Jew who suffered because of the transgressions of the many. He was the prototype for the suffering churches (cf. Rom. 8:17).

In Luke-Acts “blood” is part of a narrative about the persecution of the prophets and judgment on Israel. The “blood of all the prophets” will be “required of this generation” of Jews (Lk. 11:50-51). Pilate’s shedding of the blood of the Galileans in the temple prefigures the coming judgment on Jerusalem (Lk. 13:1-5). The high priest expresses concern that the blood of Jesus will be brought upon the Jews (Acts 5:28). When Paul is reviled by the Jews at Corinth, he declares, “Your blood be upon your own heads!” (Acts 18:6; cf. 20:26). Paul admits having stood by when the blood of Stephen was being shed (Acts 22:20).

There is no place in these narratives of suffering and destruction for the belief that God shed his own blood in order to redeem a people for himself. Paul’s soteriology states that God put forward Christ Jesus “as a propitiation by his blood” (Rom. 3:25)—again, two distinct actors—and even this idea is hard to find in Acts.

Following the healing of the lame man on the way to the temple Peter says: “But what God foretold by the mouth of all the prophets, that his Christ would suffer, he thus fulfilled” (Acts 3:18). Two distinct actors.

Hebrews twice has the expression “through his own blood” with reference to Jesus, though in a slightly different form. Jesus “entered once for all into the holy places… by means of his own blood (dia… tou idiou haimatos)”; he sanctified the people “through his own blood (dia tou idiou haimatos)” (Heb. 9:12; 13:12). If anything, the different construction suggests that we should look for a different interpretation for dia tou haimatou tou idiou in Acts 20:28. See point 4 below.

Nevertheless, attempts have been made to defend the “blood of God” interpretation.

Helopolous recognises the seriousness of the problem but resorts to a convoluted Chalcedonian solution, which I think can be reduced to: Jesus is fully God, therefore it can be said that God’s blood was shed on the cross. But whatever value this may have as a post-biblical theological rationalisation, it is worthless from a historical-critical point of view.

Fitzmyer rather lamely proposes that “Luke may be thinking of the action of God the Father and the Son as so closely related that his mode of speaking slips from one to the other; if so, it resembles the speech patterns of the Johannine Gospel”. But even John, as far as I can tell, maintains a careful distinction between the Son who suffers and the Father who consecrated him and sent him into the world (Jn. 3:17; 8:26; 10:36; 17:18, 21, 23, 25).

Keener also thinks that the language “might perhaps be intended to imply the unity of Jesus with his Father to the extent that it was as if the Father offered his own blood”.1 The critical phrase in that statement is “as if”: the “blood of God” becomes more or less a metaphor for the closeness of the Father to the Son in his suffering. Keener goes on to suggest that the expression is rhetorical: “Perhaps Luke deliberately provides a jarring statement to grip attention, functioning something like hyperbole.” But he makes no attempt to explain why such a jarring and unaccountable figure of speech should be introduced here.

But if the Trinitarian interpretation is unsatisfactory, what are the alternatives?

1. Some manuscripts have “church of the Lord” instead of “church of God”, in which case “through his own blood” could be taken to refer to the Lord Jesus rather than to God. But since “church of God” is the more difficult reading here, scholars tend to assume that it was original and that a puzzled scribe “amended” the text to remove the theological problem. Moreover, while the expression ekklēsia tou kyriou (“assembly of the Lord”) appears occasionally in the Greek Old Testament (Deut. 23:2–4; 1 Chr. 28:8; Mic. 2:5), it is found nowhere else in the New Testament.

2. It has been suggested that “blood” means “blood relationship, kin”, so we might perhaps translate: “which he obtained by his own kin”. In a papyrus text on adoption we find the expression “begotten to you from your own blood” (ex idiou haimatos gennēthenta soi), but “blood” here refers not to an individual relative but to the biological relationship. In any case, it’s difficult to see why Luke would want to stress the “blood” relationship of Jesus to his Father.

3. Wall thinks that Luke may have been grammatically careless “because of his lack of theological interest in the efficacy of Christ’s blood in saving people from their sins”.2 But the consistency of the “blood” themes in Luke-Acts (see above) make this unlikely.

4. It could perhaps be argued that verses 25-27 are a parenthetic remark about Paul’s apostolic responsibility towards the Ephesian elders. The main line of thought is that Paul is leaving, he doesn’t know what the future holds, he continues to pursue the ministry that he received from the Lord Jesus, so the elders must shepherd the “church of God”, because after his departure “fierce wolves will come in among you, not sparing the flock”. Into this argument he interjects the personal and rather startling testimony that he is “innocent of the blood of all”. In this case we might suppose that the clause “which he obtained with his own blood” looks back to “the Lord Jesus” in verse 24: they are to shepherd the “church of God”, which was obtained by the blood of the Lord who gave Paul his apostolic ministry. Perhaps. To my mind a fall-back position if the next one really doesn’t work.

5. The best alternative would seem to be to translate dia tou haimatou tou idiou not as “through his own blood” but as “through the blood of his own”. The absolute phrase “his own” would be a term of endearment referring in this case to “his own Son”. There is supposedly evidence for the usage in the Greek papyri. Bruce cites: “So-and-so to his own (ho deina tōi idiōi), greeting”.3 But a couple of examples in Acts also roughly illustrate the point: when the apostles were released, they “went to their own (tous idious)…” (Acts 4:23); Felix gives orders that Paul should be kept in custody but that “none of his own should be prevented from attending to him (tōn idiōn)” (Acts 24:23). Perhaps Luke had a liking for the idiom. The sense, then, would be close to “He who did not spare his own Son (tou idiou huiou), but gave him up for us all…” (Rom. 8:32).

Fitzmyer regards this as “a last-ditch solution for this text-critical problem”, but it seems to me a lot more plausible than the “God’s blood” interpretation.4

6. One further observation may help to account for the “through the blood of his own” reading. I noted above that “blood” in Luke-Acts belongs primarily to narratives of persecution and judgment, so that a reference to the redemption of the church even by the blood of Christ seems out of place. But Paul has just insisted that he is “innocent of the blood of all (apo tou haimatos pantōn)” (Acts 20:26). It may be, therefore, that “through the blood of his own” (dia tou haimatou tou idiou) has been constructed as a rhetorical counterpart to the previous expression. The elders are corporately accountable (their blood is now on their own heads, not on Paul’s) for a community that owes its existence to the “blood” of God’s own Son.

  • 1C.S. Keener, Acts: An Exegetical Commentary 3: 15:1-23:35 (Kindle locs. 22400-22401).
  • 2R.W. Wall, The Acts of the Apostles (The New Interpreters Bible, vol. 10, 2002), 284.
  • 3F.F. Bruce, The Acts of the Apostles: The Greek Text with Introduction and Commentary (1951), 381.
  • 4J.A. Fitzmyer, The Acts of the Apostles (1974), 680.
Chris Wooldridge | Wed, 08/16/2017 - 11:58 | Permalink

Point 1 should be “ekklesia tou kuriou”, surely?

Kermit Zarley | Thu, 08/17/2017 - 23:35 | Permalink

I was a Trinitarian for 22 years. Then I undertook a 28-year, in-depth study resulting in my change in Christology to God being one person and Jesus not being God. Yet I still affirm all else the church says about Jesus. I wrote a 600 pp. book about it entitled The Restitution of Jesus Christ, citing 400+ scholars. (available at my website kermitzarley.com). I think my treatment of Ac 20.28 in this book corresponds to what you say here, Andrew. I’m now taking the liberty to append it (your software didn’t include my caps, italics, and scholarly refs):

1. Does “the church of God” in Acts 20.28 indicate that Jesus is God?


Some traditionalists list Ac 20.28 to support that Jesus Christ is God. In this verse, Luke records that the Apostle Paul, while journeying to Jerusalem, met with the elders of the church at Ephesus and spoke to them, saying,

28 “Be on guard for yourselves and for all the flock, among which the Holy Spirit has made you overseers, to shepherd the church of God which He purchased with His own blood.”

Three problems about the identity of Jesus emerge from Ac 20.28. Two of them are textual and the other one is grammatical.

The Greek Text and English Versions

The first problem to consider in Ac 20.28 is whether the Greek text should have theou (Gr. for “God”) or kuriou (Gr. for “Lord”). English versions slightly favor theou:

“the church of God” (av, rvmg, rsv, nasb, tev, jb, nebmg, niv, esv)

“the church of the Lord” (rv, rsvmg, nasbmg, neb, nivmg).

But the ms evidence, as well as other external witnesses, is about evenly divided:

ten ekklesian tou theou (Codex Vaticanus, Codex Sinaiticus, Latin Vulgate)

ten ekklesian tou kuriou (Codex Alexandrinus, Codex Bezae, some minor versions).

On the whole, textual critics seem about evenly divided between which of these two, well-attested readings in the ms evidence is to be preferred as authentic. The Committee for the United Bible Societies’ Greek New Testament (and many scholars) tentatively concluded that theou is the authentic reading in Ac 20.28. However, they gave it a C-rating, meaning that they had “a considerable degree of doubt” about it.

Much of the discussion among textual critics as to whether theou or kuriou is the authentic word in Ac 20.28 regards reasons why copyists might have changed theou to kuriou or vice versa. These reasons are actually principles of textual criticism. It is quite likely that a scribe changed the original text of theou by substituting kuriou because he considered theou to be confusing, since “God” does not personally possess a physical body consisting of flesh and “blood” while the “Lord” Jesus did. On the other hand, it is plausible, but less likely, that a scribe did the opposite—substituting theou for kuriou—in opposition to Patripassianism.

One piece of internal evidence supports that Ac 20.28 should read “God.” It is that the phrase, “the church of God,” appears eleven times in Paul’s writings whereas “the church of the Lord” does not appear in the nt.

The other textual problem with Ac 20.28 regards whether the word idiou (“own”) is to be taken as an adjective or a noun. That is, is the correct text tou idiou haimatos (“his own blood”) or tou haimatos tou idiou (“the blood of his own;” some would also translate it “his own blood”)? The former reading lends itself to calling Jesus “God.” Indeed, this reading is usually found in mss that have theou.

The UBS’ Committee also rendered tou haimatos tou idiou (“blood of his own”) as the correct reading and gave it a B-rating, meaning that they had “some degree of doubt” about it. This UBS text of Ac 20.28 therefore translates, “the church of God which he purchased with the blood of his own (One/Son).”

Two Theou Translations

So, if theou is authentic in Ac 20.28 there are two primary translations of it and three ways to understand it. They are as follows, with commentary appended:

1. “To shepherd the church of God which he purchased with his own blood.” This translation suggests that “God” refers to Jesus since the Father does not have blood. Traditionalist A.W. Wainwright admits, “it is difficult to imagine that the divinity of Christ should have been stated in such a blunt and misleading fashion.” However, this translation can also be understood to mean that God the Father purchased the church with Jesus’ blood, which belongs to both the Father and Jesus (cf. Jn 17.10).

2. “To shepherd the church of God which he purchased with the blood of his own (One/Son).” This rendering of the UBS Greek text clearly makes “God” refer to the Father, “the blood” belongs to Jesus, and “his own” means the Father’s own Son, viz., Jesus Christ. This translation, which is preferred by a majority of scholars, does not call Jesus “God.” Its meaning is encapsulated in a new song sung to Jesus in heaven, “Thou wast slain, and didst purchase for God with Thy blood men” (Rev 5.9).

Conclusion of Commentators

A very large majority of biblical exegetes render Ac 20.28 as a text that does not identify Jesus as God. Even most contemporary traditionalist commentators, especially those who have written extensively that the nt calls Jesus “God,” conclude that Ac 20.28 does not do so. Here are some examples among them: R.E. Brown says, “we are by no means certain that this verse calls Jesus God;” M. Harris deems it “unlikely, although not impossible;” A.W. Wainwright admits, “this passage cannot be adduced as convincing evidence that Jesus was called God in New Testament times.”

@Kermit Zarley:

Thanks. Very clear. I tend to think of myself as a “narrative Trinitarian”. The phrase is ambiguous: i) a Trinitarianism that sums up the narrative relationship between Father, Son and Spirit in the New Testament; and ii) a Trinitarianism that is part of the historical and therefore narrative witness of the church—in this case, in the context of an emerging Hellenistic-Roman Christendom.

“The New Testament nowhere (correct me if I’m wrong) speaks of God dying on the cross.”

 — This premise is very much tied up in the term “God.” All would recognize that ὁ θεὸς in the NT is primarily, with only a few exceptions, a designation for the Father of Jesus. That being the case, orthodox Christians (who reject Patripassianism) would not expect the normative NT language to be of “God” dying on the cross. The normative language is of Jesus Christ dying on the cross, and the theological question, “Did God die on the cross?” therefore depends on whether, and in what sense, it is correct to call Jesus “God.” (I believe all would have to concede that the title ὁ θεὸς is used for Jesus at least in John 20:28 and probably in other texts, so the statement “Jesus, who is God, died on the cross” must be correct in some sense, even if one contests what that sense is.)

Having said that, there are a number of NT passages that arguably speak implicitly of God dying on the cross, or at least in ways that evoke this idea.

- The most direct of these is the passage discussed in this article, Acts 20:28 — granting the author’s point that the reading “church of God, which He obtained with His own blood” is disputable on exegetical and text-critical grounds.

 — Having discussed Christ’s death in the immediate context, Paul in 2 Cor. 5:18-19 makes God the main actor and Christ the intermediate agent in the drama of cross-centered reconciliation. This falls short of saying that God died on the cross or that God’s blood was shed, but it certainly makes “Christ dying on the cross” an act of God.

 — Similarly, in Rom. 5:6-7, Paul speaks of Christ dying for the ungodly, with the emphasis squarely on the will of the one who is willing to die. Yet in v. 8, “Christ dying for us” is interpreted as God demonstrating his love for us. So, again Christ’s death on the cross is God’s act of love. This raises a serious theological problem: how can God take credit for Christ’s willingness to die for sinners? The doctrine of the Trinity, of course, contains the Church’s answer to this conundrum.

 — There is a phenomenon in 1 John, widely recognized by scholars, whereby the author is intentionally ambiguous about whether certain singular personal pronouns refer to God or to Christ. One of the most interesting instances of this is in 1 John 3:16 — “this is how we have come to know love: he gave his life for us.” The nearest antecedent for “he” is “God,” mentioned five times in vv. 9-10; Christ (“the Son of God”) was last mentioned in v. 8. Were we forced to choose between God and Christ as the referent of v. 16, we should probably choose Christ, but the intentional ambiguity (again, something that occurs frequently in the epistle) suggests that the writer does not want to force us into an either/or choice but to reflect on how the statement might be true of both God and Christ. In that sense, the writer is arguably pressing us to consider God as having laid down his life for us in some sense.

 — In the Gospel of John, if we combine the idea of Jesus’s flesh as the tabernacle of the Logos that “was God” (John 1:1, 14) with the idea of Jesus’s flesh as the temple that would be destroyed and raised up (John 2:19-21), we again have a picture of an entity that “was God” dying.

@Thomas Farrar:

The nature of Thomas’ confession, the fact that it happened after the resurrection, and its possible relation to Domitian’s demand to be addressed as dominus et deusmake it difficult, I think, to draw the direct theological inference that God died on the cross. But it’s worth keeping in mind.

What Paul says in 2 Corinthians 5:18-19 is that God was reconciling the world and not counting their sins against them. It does not say that he suffered, shed blood, or died. Why shouldn’t we just take this at face value. Jesus did one thing, God did something else.

I don’t see how Romans 5:8 would entail the suffering of God. The basic thought, surely, is given in Romans 3:24-25: God provides a gracious gift, he puts Jesus forward as a propitiation by his death, he passes over former sins. There is a clear separation between the suffering of Jesus and gracious forgiveness of God. Jesus suffers, God forgives.

The argument from 1 John 3:16 seems very tenuous. There is no immediate antecedent for ekeinos. It seems to me likely that the emphatic pronoun makes reference to John 15:13 or a derivative tradition. And can we really imagine that John wold have said of God that he “laid down his psychēn for us”?

The line of thought traced in John 1-2 is more persuasive, but it remains speculative, and the logic is disrupted by the switch from “tabernacled” (eskēnōsen) to “temple” (naos).

@Andrew Perriman:

Andrew, Thanks for this reply. With regard to John 20:28, Domitian’s honorific title may indeed be important to the interpretation of Thomas’s confession, but I don’t think it comes near to fully explaining it. A natural Jewish polemical response might be “our only Lord and God is the Holy One,” and a natural Jewish Christian polemical response might be “we have only one God, the Father, and one Lord, Jesus Christ” (cf. 1 Cor. 8:6). So it is remarkable that John’s polemical response (if it is that) is to have Jesus addressed as Lord and God. Furthermore, a Jewish auditor of John 20:28 would hear it in terms of texts like Psalm 35:23 LXX, where God is addressed with the exact same phrase but with theos and kyrios inverted. Finally, theos in John 20:28 must be read in light of the prologue (John 1:1 and most likely 1:18), where the application of theos to the Logos/Monogenes is clearly much more than a political statement.

I don’t claim that in 2 Cor. 5 or Rom. 5 Paul directly conveyed the idea that God died on the cross. I am only pointing out that the way he expresses the cross as an act of God’s love is remarkable and opens the door for further reflection about just how intimately God was involved in Christ’s willing sacrifice. I don’t think that “Jesus did one thing, God did something else” adequately captures Paul’s  emphasis here. Rom. 3:25a is, in my view, better translated as “whom God publicly displayed as a mercy-seat…” This is regarded by most scholars as part of a pre-Pauline formula and it depicts the soteriological significance of the cross-event symbolically using Levitical cultic imagery.

In 1 John 3:16, the lack of antecedent for the demonstrative pronoun is exactly my point. Making a complete argument would be quite involved, but notice the ambiguity of referent in 1 John 3:1-3: contextually it seems to be about God, but “did not know him” (aorist) exactly parallels what John 1:10 says about the Logos. In v. 3 we have an ambiguous ekeinos that seems likely to refer to God, whereas ekeinos in v. 5 seems likely to refer to Christ. Ekeinos in v. 7 could easily refer to either God or Christ. V. 16 continues this pattern of ambiguity.

@Thomas Farrar:

We’re getting a bit beyond the “did God die on the cross?” debate, but I agree that the Johannine literature is more susceptible to an incarnational reading, largely on the basis of a wisdom theology. Whether μονογενὴς θεὸς is the correct reading of John 1:18 and whether Thomas is supposed to be referring back to the Prologue are moot considerations, but I wouldn’t rule them out. But the difference between GJohn and pretty much everything else is striking. For Paul God showed his love by “sending his own Son in the likeness of sinful flesh” (Rom. 8:3).

I don’t see how your translation of Romans 3:25a affects the distinction between Jesus dying and God forgiving.