Does the Bible really say Jesus was God?

Sorry to keep you. Fortunately this post can be briefer than I expected, since we’ve already reached agreement on a number of issues.

  1. Although the common meaning of θεός is “god”, it has a wider range of meanings which do not mean "god ( and which include the adjectival sense “divine”). This usage is widespread in Greek literature, the Old Testament (LXX), and Second Temple Period literature, and is also found in the New Testament. Jesus himself uses one such meaning, in John’s gospel (the only reason why I cited this usage by Jesus was to demonstrate that θεός does not always mean God; it doesn’t have any other relevance to John 1:1).

  2. The use of θεός in John 1:1c is not necessarily identifying Jesus as God, contrary to what you originally thought. Not only can it be understood to mean “divine”, the grammar alone does not get you to “God”. As you have said, “he grammar, on its own, does not get you to ‘God’ and leaves room for ‘divine’”. Consequently, as you have noted, the numerous commentators who read it as “God”, do so as an interpretive decision based on their interpretation of other verses first (as they state explicitly). Even though they don’t believe the Synoptic gospels present Jesus as God, they interpret some passages of John’s gospel as presenting Jesus as God, and therefore conclude that the logos is being called God in John 1:1. This is an interpretive conclusion based on what they believe John’s theology to be.

  3. Benjamin Sommer’s “multiple bodies” and “Hindu avatar” theory of conceptions of God in ancient Israel and fourth century to early medieval Judaism, have no relevance to our understanding of God in the New Testament, much less our understanding of John 1:1.

We won’t need to revisit these issues. Now I’ll answer your firebreak question.

In what sense do you think John thought Jesus was? How high was John’s christology in your view? You will need to explain this as you have been ambiguous by the word “divine” so far, since this word can also be taken to mean “literally God”. What does John mean when he calls Jesus “divine”?

By “divine”, I mean Jesus was divine in origin; he was the son of God. This placed him in a unique relation to God. Additionally, he was given unique authority, unique titles, a unique role, and God worked through him uniquely, specifically because he was the son of God, and therefore God’s representative on earth. He was a human of divine origin, who was appointed as the unique agent of God.

Now to other matters.

Point one.

I gave you the Greek of John 1:1.

John 1:
1 Ἐν ἀρχῇ ἦν ὁ λόγος, καὶ ὁ λόγος ἦν πρὸς τὸν θεόν, καὶ θεὸς ἦν ὁ λόγος.

I asked this question.

Do you understand why translating “theos” as “God” the first time, and “divine” the second time, is based on the actual Greek grammar? If you do, please provide the grammatical explanation.

You said you can barely read the Greek “I certainly can’t answer your question about it”. So now I’ll explain.

  1. The first grammatical reason for reading theos as “God” the first time and “divine” the second time is based on the fact that the two instances are grammatically different. As you know, one is τὸν θεόν (using the article), and the other is θεὸς (anarthrous). This alone is warrant to translate them differently, and this is why many people appeal to the translation “God” for τὸν θεόν and “a god” for θεὸς.

  2. However, the grammatical information does not stop there, since θεὸς is not simply anarthrous, it is anarthrous qualitative predicate nominative. The qualitative sense of θεὸς here indicates that we are dealing with θεὸς as an adjective, not a noun. This is why the NET arrived at the translation “fully God”, to preserve the adjectival sense of θεὸς in the clause, while also preserving the theological conclusion that the logos is “God” in some (undefined), sense.The footnote makes it clear that they believe John was writing as a Trinitarian; “This points to unity of essence between the Father and the Son without equating the persons”. They assume that John actually believed that God was more than one person (which is certainly questionable).

  3. So the anarthrous qualitative predicate nominative argues strongly against “a god”, and that’s even before we get into the more awkward consideration of the fact that “a god” means we now have two gods (a theological position which would then have to be somehow harmonized with the Second Temple Period context, and the content of the New Testament itself). However, in addition to the fact that the second instance of θεὸς is anarthrous, and in addition to the fact that it is also qualitative, we also have lexical evidence of θεὸς (nominative), placed before a vowel, being used specifically to mean “divine” (as I have discussed extensively with “Bill Smith”).

So these are the three grammatical arguments for translating the second θεὸς differently to the first.

  • It’s anarthrous
  • It’s anarthrous qualitative predicate nominative
  • It’s anarthrous qualitative predicate nominative before a vowel

In addition, there are lexical reasons for not translating it as “God”, namely that theos in the New Testament virtually exclusively refers to the Father. Trinitarian scholar William Hasker notes this.

“First, “God” is used to designate Yahweh, the God of the Old Testament, who was known to Jesus as Father and whom he taught his followers to address as Father. This is the standard usage of “God” throughout the New Testament, as can be verified by a cursory reading of the texts.” [1]

Trinitarian scholar Murray Harris spends seven pages examining all the uses of theos in the Synoptics, John, Acts, Paul’s letters, Peter’s letters, Hebrews, James, Jude, and Revelation, and concludes thus.

“When (ὁ) θεός [God] is used, we are to assume that the NT writers have ὁ πατήρ [the Father] in mind unless the context makes this sense of (ὁ) θεός [God] impossible.” [2]

He notes that there are 83 instances of θεός in John, and only two of them cannot refer to the Father.

“Of these 83 uses of θεός, the only places where the word could not refer to the Father are 1:1 (second occurrence, referring to the Logos); 1:18 (second occurrence, referring to μονογενὴς ;-see chapter III §§B-C); 10:34-35 (both plurals); and 20:28 (addressed to Jesus).” [3]

Another lexical reason for it is that we have evidence from Second Temple Period Judaism that the anarthrous θεός was used to differentiate between “the one true God”, and anything else which could be called θεός but was not actually God. Philo describes it this way.

There is one true God only: but they who are called Gods, by an abuse of language, are numerous; on which account the holy scripture on the present occasion indicates that it is the true God that is meant by the use of the article, the expression being, “I am the God (I);” but when the word is used incorrectly, it is put without the article, the expression being, “He who was seen by thee in the place,” not of the God (t**ou Theou), but simply “of God” (Theou**);” [3]

This is even more significant given the fact that Philo used θεός to refer to God’s Word (which he of course wrote as Logos). Philo was perfectly happy calling the Logos θεός (without the article), because to Philo the Logos was divine; it was God’s reason, wisdom, thought, and word, the power with which God created the world. Philo spoke of the Logos being with God, and being θεός. But for Philo the Logos was not an independent being, nor a hypostasis of God, nor a person in a multiple-person God. So we have a first century witness to the fact that the anarthrous θεός was used in Second Temple Period Judaism (even in the first century), to refer to that which was not the one true God, but that which could be called “god” or “divine” in some sense. This lexical understanding was preserved within the early Christian commentators. I have already cited Origen saying exactly the same thing. No Origen didn’t believe Jesus was God, he believed Jesus was a divine being separate from and second to God, and who had a different nature to God.

However, there are also theological reasons for not translating it as “God”. Murray, whom I quoted earlier, accepts the translation “the Word was God” only with strong caveats.

“But in normal English usage “God” is a proper noun, referring to the person of the Father or corporately to the three persons of the Godhead. Moreover, “the Word was God” suggests that “the Word” and “God” are convertible terms, that the proposition is reciprocating. But the Word is neither the Father nor the Trinity. Therefore few will doubt that this time-honored translation needs careful exegesis, since it places a distinctive sense upon a common English word. The rendering cannot stand without explanation.” [4]

Trinitarian scholar Charles Irons explains it this way, preferring the translation “divine”.

“I hesitate to say “Jesus is God,” nor would I say “Jesus is not God”. Instead, I prefer to say, as the New Testament says, that “Jesus is the Son of God.” Although it is possible to construe it in a valid sense, I am cautious about the statement “Jesus is God,” because the name “God” (with the definite article, ho theos) most frequently and properly refers to the Father. “Jesus is God” could be taken to mean “Jesus is the Father,” which would be modalism. On the other hand, following John 1:1, we have strong precedent for saying that “Jesus is divine” (theos as anarthrous qualitative predicate nominative).” [5]

Note that last sentence; he takes John 1:1 as strong precedent for saying “Jesus is divine”. Thanks Charles! You will note that neither Murrary nor Irons makes the theological argument you do, that it’s ok to say “the Word was with God and the Word was God” if God has multiple persons. On the contrary, they rightly say that this phrase causes problems when read as referring to a God with multiple persons, precisely because it confuses the persons.

As I see it, arguing that you can get around contradiction A by positing an additional contradiction B, is just piling confusion on confusion. It’s these kinds of ad hoc arguments which really expose Trintarian doctrine to serious criticism, not the least of which is logical incoherence. This incoherence has always been recognized by Trinitarians, and is captured well in the phrase “The Father is God, the Son is God, and the Holy Spirit is God, yet there are not three Gods but one God”. The phrase leads us to a logical conclusion and then takes an abrupt u-turn shouting “No, don’t go there!”. It’s not a good look.

Now I’ll answer this question of yours.

At this point, you have not yet defined what you think “and the Word was divine” means. What do you think John would mean by such a statement? That the Word is eternal and the medium through which all creation happened?

I explained it previously, but that was a while ago. It’s the word of God; the davar in Hebrew. The word of God, which represents His wisdom, reason, and power. We are told in Genesis 1 that God created by speaking, and in Psalm 33 the word of God is described explicitly with a bold anthromophism as “the breath of His mouth”. We find that same language in Philo, and we find it right here in John.

Keener (who of course believes Jesus is God and that Jesus is called God here), puts it this way.

"The Old Testament had personified Wisdom (Prov 8), and ancient Judaism eventually identified personified Wisdom, the Word and the Law (the Torah). By calling Jesus “the Word,” John calls him the embodiment of all God’s revelation in the Scriptures and thus declares that only those who accept Jesus honor the law fully (1:17). Jewish people considered Wisdom/Word divine yet distinct from God the Father, so it was the closest available term John had to describe Jesus.

1:1–2. Beginning like Genesis 1:1, John alludes to the Old Testament and Jewish picture of God creating through his preexistent wisdom or word. According to standard Jewish doctrine in his day, this wisdom existed before the rest of creation but was itself created. By declaring that the Word “was” in the beginning and especially by calling the Word “God” (v. 1; also the most likely reading of 1:18), John goes beyond the common Jewish conception to imply that Jesus is not created (cf. Is 43:10–11)." [6]

Virtually all of this is what I would say. I part ways with Keener only when he says that John 1:1 is clearly saying that the logos was a pre-existent person, and that John is deliberately breaking with the Jewish tradition.

Köstenberger puts it this way.

"The term “the Word” conveys the notion of divine self-expression or speech (cf. Ps. 19:1–4). The Genesis creation account provides ample testimony to the effectiveness of God’s word: he speaks, and things come into being (Gen. 1:3, 9; cf. 1:11, 15, 24, 29–30). Both psalmists and prophets portray God’s word in close-to-personified terms (Ps. 33:6; 107:20; 147:15, 18; Isa. 55:10–11), but only John claims that this word has appeared in space-time history as an actual person, Jesus Christ (1:14, 17).

Most critical in this regard is Isaiah’s depiction of God’s word as going out from his mouth and not returning to him empty, but as accomplishing what he desires and achieving the purpose for which he sent it (Isa. 55:11; cf. 40:8). In this passage Isaiah provides the framework for John’s “sending” Christology, which presents Jesus as the Word sent by God the Father who pursues and accomplishes his mission in obedience to the one who sent him. This sender-sent relationship, in turn, provides the paradigm for Jesus’ relationship with his followers (cf. esp. 17:18; 20:21–23; see Köstenberger 1998b)." [7]

I can agree with every word of that. Arnold describes it thus.

"Was the Word (1:1). Echoes of the creation account continue here with allusion to the powerful and effective word of God (“And God said, ‘Let there be light,’ and there was light”; Gen. 1:3). The psalmists and prophets alike portray God’s word (logos) in almost personal terms (e.g., Ps. 33:6; 107:20; 147:15, 18; Isa. 55:10–11). Isaiah, for instance, describes God’s “word” as coming down from heaven and returning to him after achieving the purpose for which it was sent (Isa. 55:10–11). John takes the prophetic depiction of God’s word in the Old Testament one decisive step further. No longer is God’s word merely spoken of in personal terms; it now has appeared as a real person, the Lord Jesus Christ (cf. 1 John 1:1; Rev. 19:13).

While the primary source of John’s depiction of Jesus as the Word is the Old Testament, his opening lines would resonate with his Greek-speaking audience. In Stoic philosophy, for instance, logos was used to refer to the impersonal principle of Reason, which was thought to govern the universe. It is a mark of John’s considerable theological genius that he is able to find a term (“the Word”) that is at the same time thoroughly biblical—that is, rooted in Old Testament teaching—and highly relevant for his present audience." [7]

Again, I would say exactly the same. You can see that calling the logos “theos” does not in any way require the logos to be a person, nor does it require the logos to be a person in a God who is one being with multiple persons.

This point one of mine addresses the argument in your three points which concluded “These contextual considerations rule out anything less than the Word being, literally, God”. While I’m here I might as well ask you why you interpret ὁ λόγος as a person, since we both know that the Greek word ὁ λόγος does not mean a person anymore than the English word “word” means a person.

This addresses everything in your previous post. Now I’ll move to your latest post. There’s a lot less to deal with in this one.

Point two.

You conceded Sommer to me, which was good of you, so that leaves this.

The works of Segal and Boyarin collectively demonstrate that Judaic monotheism wasn’t as simple “one God one person one everything” – and that the multiplicity of the persons of God, not just the bodies, was known and not at all heretical in pre-Christian Judaism. In that sense, there is no contradiction at all in John 1:1 since it can be understood as referring to the multiplicity of God’s person. And even if there was absolutely zero multiplicity of God’s person before Christianity, which there surely was, then this could be explained as a new Christian interpretation and we can still understand John 1:1 as referring to God’s multiplicity and thus not have to deal with a contradiction.

I will have more to say on Segal and Boyarin later (for now, on Boyarin, I recommend you read this, which notes that his case is based only three texts). I have several of Boyarin’s works but I haven’t read them all (I am only in the middle of “The Jewish Gospels”). I’m more familiar with Segal. However, here are a few points to start with.

  • The New Testament uses none of the “multiplicity of persons” language which we see in texts which actually are referring to a multiplicity of persons

  • The New Testament shows clear evidence that many Jews considered Jesus’ claim to be the son of God, to be deeply heretical; where is the evidence that they understood he was referring to himself as one of the persons in a multi-personal God, but had no problems with that?

  • Claiming that John 1:1 can be understood as referring to a multiplicity of persons is one thing, but demonstrating that this is what he meant, is quite another

  • To claim that a multiplicity of person was a Christian innovation is of course an even more difficult challenge to meet, but either way at some point you need to address the fact that the overwhelming use of θεὸς in the New Testament refers to one person, the Father, and the fact that God in the entire Bible is never referred to as “they”

You quoted this.

It will also be important to quote Richard Hays again when he says “Daniel Boyarin is another scholar who has provacatively destabilized conventional beliefs about what first-century Jews could and could not have believed about the multiplicity within the divine identity.”

I suggest strongly that you read closely what Hays is and isn’t saying about Boyarin, because by the time I have finished reading Boyarin I think you’ll be disappointed in just what he does and doesn’t prove. I’ll give you a hint; the word “provocatively destabilized” is a polite way of saying Boyarin has made extremely bold claims which have convinced very few people at all, and only highly qualified assent has been given to any of his claims.

Point three.

I asked “Do you understand that the frequency with which a Greek word is translated in an English Bible has absolutely no relevance to whether or not the word has a particular meaning in a specific verse?” You responded thus.

Actually, it most certainly has relevance. If we know the standard usage of a word in a language, then we must further know that we should only abandon understanding it in its primary sense if there is reason to do so.

No. The frequency with which an English translation has translated a word, does not give us any information about how that word is used in a specific context. Think about this for a minute; we’re talking about an English translation of the word. What is more important is how the author uses it in the original source language text. However, even if an author uses a particular word 99 times with one meaning, this does not provide us with any information on how they are using it in another passage, unless they’re using it in exactly the same semantic and grammatical context. How many times can you find θεὸς in John, used as anarthrous qualitative predicate nominative before a vowel? You know that John uses θεὸς 81 times to refer to the Father, so are you seriously going to suggest that the other two times, we must assume is is also referring to the Father? Surely you can see what a mess your own rule will get you into.

Point four.

I quoted Daniel Wallace saying this.

“The list of passages which seem explicitly to identify Christ with God varies from scholar to scholar, but the number is almost never more than a half dozen or so. As is well known, almost all of the texts are disputed as to their affirmation—due to textual or grammatical glitches—John 1:1 and 20:28 being the only two which are usually conceded without discussion.” [8]

You were very excited by this.

Incredible. So the text I cited is literally not even debated amongst scholars to clearly identify Jesus with God. This isn’t even an overwhelming majority in scholarship, this is an absolute consensus.

I hate to break it to you, but you’re wrong on this. After the sentence which concludes “John 1:1 and 20:28 being the only two which are usually conceded without discussion”, Wallace has a footnote which flatly contradicts your enthusiastic reading of his words. He says this.

“Even here there is debate, however.”

So in fact there isn’t a single undebated verse, there are only two which everyone agrees apply θεὸς explicitly to Jesus, and even the meaning of those two is debated.

Conclusion

I suggest that you give some thought as to why Hurtado adamantly stops short of saying Jesus was understood to be God by his earliest followers, and why He says that theos is applied to Christ in John’s gospel but does not say this means Jesus is God. He was pressed firmly on these points by Ben Witherington, and his response is revealing. Here is the exchange.

BEN: In your second chapter you say at one juncture (p. 73) that Jesus is treated in ways that liken him to God. Wouldn’t be better just to say, as Bauckham does, that in ways we don’t full grasp Jesus was considered part of God’s very identity, meaning of course that God was complex, involving more than one personal entity. This latter view seems to me to do better justice to the fact that: 1) while the term theos in the NT almost always means the one Christians call God the Father, nevertheless, even in our earliest source, Paul; 2) Jesus is called God in Rom. 9.5, as Metzger long ago showed, and in a text like Phil. 2.5-11 as well (not to mention in five other places in the NT— see Silva’s book). In other words, even in the earliest part of the NT era, there was beginning to be a reformulation of the meaning of the word theos among even some Jewish Jesus followers. Jesus is not merely likened to God he has begun to be included in what we call the Godhead. How do you respond? I must add as an addendum, that I find the tap dance done by Jimmy Dunn on this very matter unconvincing when it comes to Paul’s beliefs and expressions.

LARRY: I understand the desire to explore the relation between first-century Christological claims and subsequent formulations, but I also think that we should be cautious in respecting the differences, and cautious also about reading earlier formulations too much with regard to how they line up with subsequent ones. In my book, God in New Testament Theology (Abingdon, 2010), I discuss how Jesus is programmatically included in early Christian discourse about “God”, and also in the worship directed to “God.” Sure, at least in John 1:1 and 20:28, “theos” is applied to Jesus. But Jesus’ divine status is rather consistently defined with reference to “God” (“the Father” in Pauline and Johannine terms). Later (as I judge it), Christians took up questions about “ontology”, using Greek categories of “being”, and this entailed controversies such as the “Arian crisis” of the fourth century. But I try to confine my discussion of NT discourse to the categories in that discourse, and I don’t think it helpful to impute the later (albeit, perfectly reasonable) terms such as “Godhead,” divine “persons,” divine “substance” etc., into these texts. I’d rest with simply saying that the NT makes Jesus integral to any adequate discourse about “God” and to any adequate worship of “God.” Christian theology must do more than simply echo NT formulations, of course. But that’s a theological task, whereas my work is more concerned with historical analysis.

This reply is longer than I wanted it to be (just over 10 pages), but there was a lot to get through. If you think I’ve missed anything, or you want me to go over anything in more detail, just let me know.


[1] William Hasker, Metaphysics and the Tri-Personal God (OUP Oxford, 2013), 246-347.

[2] Murray J. Harris, Jesus as God: The New Testament Use of Theos in Reference to Jesus (Wipf and Stock Publishers, 2008). 47.

[3] Philo, “On Dreams” 1.229, Charles Duke Yonge with Philo of Alexandria, The Works of Philo: Complete and Unabridged (Peabody, MA: Hendrickson, 1995), 385.

[4] Murray J. Harris, Jesus as God: The New Testament Use of Theos in Reference to Jesus (Wipf and Stock Publishers, 2008). 69.

[5] Charles Lee Irons, Danny Andre Dixon, and Dustin R. Smith, “A Trinitarian View: Jesus, the Divine Son of God (Irons),” in The Son of God: Three Views of the Identity of Jesus (Wipf and Stock Publishers, 2015), 20.

[6] Craig S. Keener, The IVP Bible Background Commentary: New Testament (Downers Grove, IL: InterVarsity Press, 1993), Jn 1:1–18.

[7] Andreas J. Köstenberger, “John,” in Commentary on the New Testament Use of the Old Testament (Grand Rapids, MI; Nottingham, UK: Baker Academic; Apollos, 2007), 421.

[8] Daniel B. Wallace, Granville Sharp’s Canon and Its Kin: Semantics and Significance (Peter Lang, 2009), 27.