God’s Own Blood

KJV Acts 20:28 reads as follows:

“Take heed therefore unto yourselves, and to all the flock, over the which the Holy Ghost hath made you overseers, to feed the church of God, which he hath purchased with his own blood.”

Translations seem to be about evenly divided between those that read at the end of the verse something like the KJV’s “with his own blood” and those that read something like “with the blood of his son.” Trying to determine the better reading is an interesting exercise that involves textual criticism, grammar and theology. First I will lay out the possibilities, then I will discuss them, finally indicating where I would come down on this question.

The critical part of the verse for our purposes is the end, which reads in the KJV “the church of God, which he hath purchased with his own blood.” There are two related variants in close succession here. First, we have the following choices:

1. of God

2. of the Lord

3. of the Lord, even God

4. of the Lord God

5. of the Lord Jesus

6. of the Lord Christ

7. of Christ

Readings three through seven are found mainly in the Byzantine manuscript tradition and certain Church Fathers, and are obvious conflations of the two main choices, readings one and two.

Second, we also have the following choices:

1. with his own blood or with the blood of his own

2. with his own blood

Reading 1 is ambiguous and can be interpreted either way; reading two is not ambiguous. The Greek for reading 1 is dia tou haimatos tou idiou. The word dia is the preposition “through, with,” which with this meaning takes a genitive. The word tou is the definite article “the” in the genitive case, and haimatos is the word for blood, also in the genitive case. The word idiou means “(one’s) own,” which is also in the genitive case. The pattern article + noun + article + adjective, where the noun and the adjective agree in number, case and gender, is one way of expressing the adjective attributively. Thus, these words could be translated “with his own blood,” where “own” is an attributive adjective modifying “blood.”

On the other hand, idiou with the article could be taken as a substantive, meaning that even though it is an adjective it is used as if it were a noun. The word “good” is an adjective in English, but in the expression “the good, the bad and the ugly” it is used substantively. This reading would result in the translation “with the blood of his own one.” This is bad English and is usually understood as a reference to his own son.

Reading 2, which is dia tou haimatos, lacks this ambiguity. The pattern article + adjective + noun is the other common way to express an attributive adjective, but with this wording the substantive reading is not possible, and the words would have to be translated something like “with his own blood.”

As for theology, the translation “the church of God, which he hath purchased with his own blood,” is awkward, because usually the word “God” [theos] in the New Testament is a reference to God the Father, not to Jesus Christ, but it was Jesus who bled for us. Does God have blood? On the other hand, the translation “the church of God, which he hath purchased with the blood of his own [son]” might seem to some to be at variance with orthodox trinitarian perceptions. (I do not think this is necessarily the case myself, but I have seen indications of some discomfort with it on this ground in the literature.)

With the problem laid out, let us return to the textual question. The textual evidence between “God” and “Lord” is virtually even, so the question will have to be addressed based on other factors. The basic argument of those who defend “church of the Lord” is that that is the rarer expression, appearing but seven times in the Septuagint and not at all in the New Testament, whereas “church of God” occurs 11 times in the Pauline epistles. Thus, a scribe might have seen “church of the Lord” and changed the rare expression to the more common “church of God.”

The more difficult reading, however, is “church of God,” because of the awkwardness of the following reference to “his own blood.” The suggestion that God has blood may have led scribes to replace “God” with the more generic “Lord,” which could more readily be interpreted as a reference to Jesus. (For a more detailed discussion and citations on both sides of the debate, see Metzger’s Textual Commentary on the Greek New Testament.) The United Bible Societies’ editorial committee selected “God” as the main reading for this reason, labeling this change a “C.” (The letters A through D were used to indicate the relative degree of certainty of their choices, the letter “A” being the most certain and “D” being the least.)

As for the second variant, the textual evidence here is more lopsided and clearly shows that reading 2 is secondary, supported mostly by the Byzantine witnesses that had conflated the earlier variant. But even if we choose reading one (as the UBS did, labeling this one a “B”), we are still left with the grammatical ambiguity of what it means. Either translation is possible. It is, however, unidiomatic if the intended meaning were “his own blood”; the more usual way of expressing this would be either dia tou heautou haimatos or dia tou haimatos tou heautou, or perhaps an instrumental dative in lieu of dia + genitive. The other reading is also a bit awkward; first, for having a second genitive phrase dependent on one to which it happens to agree in case, number and gender, and second, for lack of a clear antecedent to idiou. The first bit of awkwardness is what it is (such constructions certainly occur elsewhere in the New Testament); the second can be solved in one of two ways:

– Some have suggested that the word huiou “son” originally stood at the end of the text and dropped out. While this would be easy enough given the repetition of the –ou genitive ending, this is a minority view and really unnecessary.

– Others (more persuasively, in my view) argue that the word idios (its nominative form) had a distinct usage in early Christianity, similar to agapetos “beloved” or monogenes “only.” In favor of this understanding, according to Moulton, this usage is attested in Greek papyri, as in a letter beginning “[So-and-so] to his own [friend], greeting.” Therefore, you will sometimes see the word “Own” capitalized by itself, and you will sometimes see the word “son” added (either with or without brackets). That is, either “the church of God, which he purchased with the blood of his Own” or “with the blood of his own [son].”

I personally like this last solution. It seems likely to me that when this usage died out, the text began to be read as referring to God’s own blood, and so the word “God” was changed to “Lord” to ease the difficulty that reading caused. Therefore, I would understand the verse to be saying that the church belongs to God, because he has purchased or obtained it with the blood of his own son.

My impression based on the commentaries I have seen is that over time this reading has become something of a dominant view. A good illustration of this is the RSV. The first edition read “the church of the Lord which he obtained with his own blood,” with footnotes saying “Other ancient authorities read of God” and “Or with the blood of his Own.” But the second edition and the NRSV switch the text and notes around; both read “with the blood of his own Son.” The NRSV footnotes say “Or: with his own blood; Gk with the blood of his Own.” I believe the progression of readings from the RSV to the NRSV moves in the correct direction.


  1. Julie M. Smith says:

    Kevin, I don’t have anything substantive to contribute, but I think your posts like this one are some of the best things in the bloggernacle. I hope you are linking them at feastupontheword.org.

  2. Kevin Barney says:

    Thanks, Julie. Coming from you, whose work I admire so much, that is quite a compliment, and I appreciate it.

    (Somehow I don’t think this post is going to reach the 300 comment mark….)

  3. I am no exegete, so I have to admit to be on quite the learning curve her, but I appreciate your explainations on things such as grading translation choices. I imagine that I am not alone when it comes to such things. Thanks.

  4. Enjoyed this post.

  5. Ed Snow says:

    Kevin, nice post. Just goes to show that nothing about a sacred text is easy to understand, or rather we take a lot for granted when we read one.

    I recently read Bart Ehrman’s Misquoting Jesus: The Story Behind Who Changed the Bible and Why and find textual criticism to be, well dry, yet intriguing in many ways. I’m not in the camp that you can’t know anything about the NT “original text” because of all of the variant readings, but it’s nevertheless challenging.

    Here’s one example from Ehrman’s book–a dramatice example, for sure–that many people aren’t familiar with, and actually shows how important this “behind the scenes” textual criticism is. The story about the woman taken in adultery with Jesus writing on the ground (John 8: 3-11) is not present in any earliest manuscripts and appears to be, with 99.99% accuracy, a late interpolation. The KJV manuscripts contained it, so it showed up in the NT and has become a beloved narrative. The overly provocative back page of the Ehrman book’s jacket says about this passage: “[it] doesn’t belong in the bible.” Well, that’s a weighty conclusion that textual criticism can’t bear in my opinion. You can say it’s not in the earliest manuscripts, but can you say it does not belong in the bible? I’d be interested in how you deal with these issues in your upcoming NT book.

    I think textual criticism is like quantum physics–you can never say where the electron is while it’s hovering about the nucleus, but you kind of know it’s there, at least. By observing it you even affect where it is! (my apologies to those I’ve offended with my naive understanding of quantum mechanics)

    In any one place, one might have a variety of competing and confusing readings of a sacred text, but you kind of know where it is, yet the very act of reading a text is interpreting a text, ie, affecting a text. The text never just says what it says.

  6. Kevin Barney says:

    Funny you should mention Ehrman. I just today read an interesting article about him in The Washington Post:


    I only recently learned that he had lost his faith and become an agnostic. The article talks about this a little bit.

    I think he’s an excellent scholar, and I like his stuff.

  7. Kevin, great post. I added a link to this and your post on D&C 110 (which I just barely saw) at the Feast wiki’s blog threads page devoted to scripture-focused blog threads.

    So, if this is Paul speaking (viz. Luke accurately recording Paul’s words), and if it’s roughly the time Paul wrote the epistle to the Romans (two big if’s, but aren’t they reasonable guesses?), then Romans 5 (esp. vv. 8-11) seems interesting to study in connection with these verses in Acts:

    Verse 1 talks about “peace with God through our Lord”, and then v. 8 says “God commendeth his love toward us, in that . . . Christ died for us.” I think these verses seem less confusing if we accept your reading in Acts as opposed to the others. That is, it seems most consistent to read Romans 5 and Acts 28:20 keeping in mind a distinction between God (the Father) who manifests his love by allowing his Son’s blood to be shed for us. This distinction between God and Christ runs through v. 11, and v. 9 explicitly mentions justification through the blood of Christ.

    Of course this a very indirect and probably not very strong way to support your argument, but I found it very interesting to read Romans 8 in light of your analysis of Acts 20:28….

  8. (Of course I meant Romans 5 in the last line above, not Romans 8.)

  9. Kevin Barney says:

    Robert, if you go to the next page you’ll see a couple of other scripture related posts I put up, one on Nephi’s steel bow and another on Ezekiel’s sticks, which you may want to link at the Feast site.

  10. Elisabeth says:

    Hi, Kevin-

    I won’t pretend that my untrained brain can grasp the main thrust of your post, but this quote resonated with me:

    …the church belongs to God, because he has purchased or obtained it with the blood of his own son.

    I love this idea, that it’s God’s Church because Jesus Christ shed His blood to make it so.

  11. Nice post, Kevin. This is an apt and engaging summary of the issues surrounding this part of the New Testament text. I’ve read about this ambiguity before, but you add some interesting details here. Unfortunately, it’s been more than 10 years since I did any appreciable amount of work in Greek, so I don’t have much to offer.

    My understanding is that the reading of “the church of God” (ten ekkehsian tou qeou in Nestle-Aland 26–excuse the poor transliteration) has become dominant primarily due to the Granville Sharpe rule. (ten ekkehsian tou qeou literally means something like “the worshipping assembly of God,” and I take it to refer more to the assemblers themselves than to the institution sponsoring the assembly. Thus, it would be the assemblers that God has obtained with whomever’s blood, and not the church. So I’ve always wondered why something like like “to shepherd the assembled saints of God, that he obtained…” isn’t considered more appropriate, even if it’s less traditional.)

    With the blood issue, I’ve less to say. Out of the shoot, I’m partial to “his own blood” just because its simpler and that’s my knee jerk reaction to handling Koine Greek. But I think that it should be translated more literally so that it preserves the basic ambiguity of the construction, “with the blood of his own.” I don’t see any reason to translate indeterminate Greek into determinate English.

  12. Thanks for the heads up Kevin (#9), I’d already added the steel bow post, but hadn’t seen the Ezekiel’s sticks—great stuff!

    Someday maybe the bloggernacle will follow category conventions so that aggregators like MA and LDSelect can filter posts by, say, a “scipture topics” category, then maybe I wouldn’t miss as many good threads like these….

  13. Kevin Barney says:

    Hi, DKL,

    I’m not sure how the Granville Sharpe rule would apply to this verse. That rule basically states that when you have the pattern article + noun [or other substantive] + kai + noun [or other substantive], where the nouns or other substantives are singular, in the same case, personal and common (IE not proper names), the nouns are referring to the same person. An example would be “the friend and brother.” Since this passage doesn’t have a second noun or other substantive, I don’t think Granville Sharpe applies. Here is a paper by Daniel Wallace (who did his dissertation on the G-S rule) that goes into detail on this:


    I think you are right that “assembly” would be a better rendering than “church” for ekklesia. The rendering “church” actually derives from one of the “rules” the KJV translators were to follow, which was to keep the old ecclesiastical language like “church” (in fact, I think “church” was a specific example used in these instructions).

    And you are also right that the verb underlying KJV “to feed” is actually poimainein, which literally means “to shepherd,’ and thus “to watch over.” The word “overseers” earlier in the v. is from GR episkopos, which is often translated ecclesiastically as “bishop,” although here the more literal “overseers” or perhaps “shepherds, guardians” is definitely better.

    So I don’t have a problem with the way you’ve rendered part of the passage (except for the allusion to the G-S rule).

  14. Kevin,
    I am not at all familiar with NT scholarship. How certain is the God/Father vs. Lord/Son distinction? Are there really no other ambiguous usages?

  15. Kevin Barney says:

    John C., I don’t think the distinction is absolute. I seem to recall that there might be one or two places where the term God is predicated of Jesus. So it is possible that God could refer to Jesus here. But that usage is definitely rare.

    In fact, now that you mention it, I think I understand where DKL was going by invoking the G-S rule. As I said, that rule doesn’t apply to this specific passage, but there is a passage elsewhere where the rule would indicate that in that particular instance “God” is a reference to Jesus. So yes, that is a possible reading of “God” in this passage, although I still think that here “God” refers to the Father.

    BTW, while looking for something else, I stumbled upon a reference that addresses this same passage:

    C.F. Devine, “The Blood of God,” Catholic Biblical Quarterly 9 (1947): 381-408.

    I don’t suppose anyone has access to this article and could report on what it says, eh?

  16. a random John says:

    I am completely out of my depth here but could “blood” be a refence to “son” in the same way that “my own flesh and blood” does not refer to me but to my children?

  17. C.F. Devine, “The Blood of God,” Catholic Biblical Quarterly 9 (1947): 381-408

    I’ve got it. Would you like a copy? Or just a synopsis?

  18. Kevin Barney says:

    arJ, I believe I looked at your question once. IIRC, our usage of “blood” in that sense to refer to a relation is unattested in Greek literature.

    Mogget, wow, a copy would be great. If you can send it electronically, my work e-mail is

    kevin.barney at kutakrock.com

    My fax no. is 312-602-4101 (Please put my name on it if you fax it)

    My work address for snail mail is

    Kutak Rock
    One South Wacker Dr., Suite 2050
    Chicago, IL 60606

  19. It’s in the snail mail.

    If you need something similar in the future, stop over at FPR and I’ll see what I can do.

  20. Kevin, rereading my comment from last night, I’m surprised by how terse/cryptic my points are (it made so much sense to me at the time I wrote it!). My reference to the Granville Sharpe rule was confusing. The English rendering “of the Lord, even God” is disambiguated from the Byzantine formulation “of the Lord and God” using Granville Sharpe, making it more probably a combination of “of the Lord” and “of God” (your examples 2 & 1) than if we rendered “of the Lord and God” as “of the Lord (on the one hand) and of God (on the other).” The late majority formulation is tou kuriou kai qeou, which is the combination of two singular nouns (tou kuriou [Lord in singulare genitive] and “tou qeou [God in singular genitive]”) joined by “and” with only one definitive article.

    Honestly, I only began studying ancient Greek because I didn’t have to take a lab with it (less work, same credit–why doesn’t everyone study ancient languages?), and I’ve never been much of a schoolboy. I may well be wrong, but to the best of my recollection, the Byzantine formulation is a text-book example of Granville Sharpe. (Which, for my part, I think that there’s reason to be skeptical of to begin with.)

    Incidentally, I’ll be picking up Devine’s article from the Boston Public Library tomorrow. I’m interested in hearing what you have to say about it.

  21. Kevin Barney says:

    Ah, DKL, now I understand: you were talking about one of the variant readings. Yes, that makes sense. I couldn’t figure out where you were getting a second noun, but now I see clearly what you were arguing.

  22. Hm, well, I’m afraid the DeVine piece will disappoint. There’s a newer one in NTS — from like 1985 or so, I think. But there’s not been much new on it for a very long time, even by textual criticism standards. It’s a very old challenge, although it’s taught in every intro class.

    Anyway DeVine’s from the RC tradition and wrote in 1947, so there’s a tremendous amount of energy expended in pursuit of what the Fathers had to say. But in the end, they have to weighed, not counted, just like any other facet of textual criticism. He’s trying to establish a Catholic trajectory on the matter and he does, but that ‘s about it.

    In case anyone cares what a grad student thinks, I (respectfully) don’t draw the same conclusions Mr. Barney does, nor do I agree that it is the scholarly concensus. Moultan, Westcott, Hort, and von Harnack are pretty ancient. Raymond Brown does take this position in an article on the broader question of whether Jesus is said to be God in the NT, but I’m not sure what he’d do in a commentary where he was focused on one text. Fitzmyer (Anchor Bible, Acts), however, considers this position a “last-ditch” choice.

    In the end, it’s best to leave the argument from authority and deal with the text. When I teach this, I use it to:

    1) Point out that there are instances where, with our present knowledge, we simply cannot come to any decent conclusions;

    2)Discourage students from using formulations found rarely, and outside the NT and the environs of its creation, at that;

    3) Encourage students to examine the role their own confessional choices might be playing in their reaction to the text;

    4)Introduce the idea that the authors of the NT were not systematic theologians and that reading the NT as if they were can be quite counter-productive;

    5) Talk about how a NT author’s larger theological interests influence this sort of a decision.

    But I will say that this was a well-done presentation of the issues and I hope I have not detracted from the effort or offended by adding my thoughts.

  23. Mogget, what you mention brings me full circle with respect to my original comment and my bias against things like the Granville Sharpe rule. I agree with you that the scholarly consensus is pretty ancient. I tend to think that the Byzantine sources deserve more credit than has been the fashion to give them since Westcott and Hort succeeded in making the Alexandrians favored. (A lot of the kind of things that folks say about what scribes did and why strikes me as more aggressively psycho-biographical than anything Brodie ever attempted.)

    I think it’s appropriate to add in the context of Mormon scriptural interpretation that one should not read the New Testament (or even the Old) guided by the assumption that there is necessarily some kind of doctrinal unity or consistency among the books or even within them.

    Anyway, I’d be inclined to give the passage a fairly untraditional rendering:

    “…to shepherd the assembled saints of the Lord and of God, that he obtained with the blood of his own.”

  24. Hm, well, I’m afraid I’ll have to disappoint again. As far as the TC goes, the best reading is “church of God.” I’d just handle the rest of the verse a little differently.

  25. Kevin Barney says:

    Mogget, it is still unclear to me how you would handle the rest of the verse differently. You gave a list of fine cautionary comments you teach your students, with which I would agree, but you never really said how you would understand the rest of the verse.

    (And I am not at all offended that you have a different take on this intractable problem! I appreciate you sharing your insights.)

  26. Mogget, thanks for the feedback. No disappointment at all–remember, I’m the one who took ancient Greek because it doesn’t have a lab.

  27. Well, actually I was telling you how I handle it, but it did come out sounding kind of coy.

    I usually don’t resolve it. I let it lie. Of course, that’s a classroom technique, not something that a translation committee or book author can usually afford to do. But it works so very well. All but the most immature students get quite serious. I think it dawns on them that working with a mss that’s 2,000 years old and half the world away is not going to be as trivial as it looks, or to put it another way, whatever they’ve had for a Sunday School equivalent is now over.

    If I had to take a stand, I’d probably say that Luke seems not to be as worried about theological precision as we are. All three elements of the triune God are there, and that, I think, was Luke’s intent in the larger context of Paul’s farewell to the elders of Ephesus at Miletus.

    It’s a toughie, though, so there’s no condemning anybody else for their thoughts. Makes me glad I’m an exegete and not a theologian!

    And it’s been a good, thought-provoking discussion, for which I thank you gentlemen.

  28. Kevin Barney says:

    I thought maybe you simply took a “let it be” approach, but I wasn’t sure. Thanks for the clarification.

  29. Kevin Barney says:

    Mogget, in case you see this in this now dead thread, I just wanted to thank you for sending me the DeVine piece, which arrived on my desk today. You are right, it wasn’t what I had hoped for, but it was really wonderful of you to send it to me.