The Inward Jew: Romans 2:28-29 and Biblical Greek Syntax

October 24, 2022 // Kevin Grasso


Share this Post


Romans 2:28-29 in Greek reads: 28 οὐ γὰρ ὁ ἐν τῷ φανερῷ Ἰουδαῖός ἐστιν οὐδὲ ἡ ἐν τῷ φανερῷ ἐν σαρκὶ περιτομή, 29 ἀλλʼ ὁ ἐν τῷ κρυπτῷ Ἰουδαῖος, καὶ περιτομὴ καρδίας ἐν πνεύματι οὐ γράμματι, οὗ ὁ ἔπαινος οὐκ ἐξ ἀνθρώπων ἀλλʼ ἐκ τοῦ θεοῦ.

This verse has troubled translators and exegetes. Matthew Thiessen, in his work "Paul and the Gentile Problem", discusses the translation of this verse at length, and he eventually argues that Hans K. Arneson's translation is the best available option. It reads "For it is not the external Jew, nor the external circumcision in the flesh, but the internal Jew, and the circumcision of the heart in spirit and not in letter, whose praise [is] not from humans but from God."

For Thiessen's argument, this verse is crucial, and he goes against the tide in his interpretation of the verse when he says that Paul does not redefine Jewishness here. Rather, he says that Jewishness is still based on circumcision (and for Thiessen's Paul, this is exclusively circumcision on the 8th day), but only those Jews who are also circumcised secretly, or in the heart, can actually please God. According to him, that is the point of Romans 2:28-29. We will not get into all the various implications of his argument here, but we will look more closely at the syntax of these verses to see if the Greek supports his case.

As a point of clarification, let's look more closely at Arneson's translation. There are actually two ways to interpret the translation, one of which is more natural but an incomplete sentence and the other of which is unnatural (and, we will argue, not reflective of the Greek).

A simplified version of the translation reads 'For it is not the external Jew, but the internal Jew...whose praise is not from man but from God'. In this kind of translation, the predicate is 'whose praise is not from man but from God' - meaning that the assertion is basically 'It is the internal Jew whose praise is from God.' While this sentence makes sense in and of itself, it is very difficult to read the full English sentence in this way.

The more natural way to read the English is to interpret the "whose" clause as a relative clause describing the Jew and not as the predicate, which would be something like 'The internal Jew, whose praise is from God, is...' The reason why this is the most natural interpretation of the English is because 'whose praise is not from man but from God' is so far away from the initial copula verb 'is' in Arneson's translation, which is the third word in verse 28.

Part of what we will argue, and what is the big grammatical issue at stake, is that this last phrase that starts with 'whose' cannot be the predicate. In other words, the verse cannot be saying 'It is the internal Jew whose praise is from God,' but it must be saying 'The internal Jew, whose praise is from God, is...' To get to this translation, though, we need to introduce a few concepts from syntax, so we can figure out what the Greek is really saying. At that point, we will be in a better position to put the verses into English.

So there are three main syntactic concepts that we need to discuss. The first is called ellipsis, the second is called constituent negation, and the third is contrastive focus. Each of these concepts is crucial for analyzing the verse, so we will go through each slowly and tie it all together at the end.


First up is ellipsis. A ton of work has been done on ellipsis in general linguistics. In Andrew Carnie's introductory syntax textbook, he informally defines ellipsis as the phenomenon whereby "a string that has already been uttered is omitted in subsequent structures where it would otherwise have to be repeated word for word" (2013: 457). He then gives the following example "Darin will eat a squid sandwich but Raiza won't" (2013:458).

In this case, the rest of the verb phrase 'eat a squid sandwich' is understood, but it isn't pronounced. This is the crucial point about ellipsis for us here. There is a mismatch between what we understand and what we hear. This is completely normal in the world's languages, and it is true of Greek as well. But more than that, it is particularly true of the conjunction οὐδέ, which we find in our verse.

Let's look at a very simple, uncontroversial example of this from 1 Thessalonians 5:5. It says Οὐκ ἐσμὲν νυκτὸς οὐδὲ σκότους 'we are not of the night, nor of the dark.' Just like in our English example, we understand something that we do not hear in this sentence. What we really understand is Οὐκ ἐσμὲν νυκτὸς οὐδέ ἐσμεν σκότους where the verb (and the implicit subject) is supplied in the second clause. Although this is the correct way to understand the sentence, it isn't what we have written. The second verb is elided, so this is a case of ellipsis with the copula, which, we will argue, is exactly what we have in our verse.

Constituent Negation

Our second important grammatical topic is constituent negation. In constituent negation, we are not negating the whole sentence but only a particular constituent. One way we can do this in English is by putting an initial "It is not" at the beginning of the sentence followed by what we are actually negating. For example, we can say "It is not the ball that was hit." In this case, we are not denying that a hitting event occurred, but we are denying that the hitting event involved a specific constituent, namely a ball.

Notice that in English we use a dummy subject and verb "It is" that goes with the negator "not" before the constituent. The main verb of the sentence is then given after 'that'. This will be important for us later. In English, we can also use 'not' to negate a constituent when we are contrasting it with something else, as in 'Tim swung and hit, not a ball, but a bird.'

Greek has this same kind of construction, but it doesn't have the dummy subject and verb construction. Constituent negation in Greek just involves putting the negator right next to whatever constituent you want to negate. Another example from 1 Thessalonians is instructive. Verse 4:8 says τοιγαροῦν ὁ ἀθετῶν οὐκ ἄνθρωπον ἀθετεῖ ἀλλὰ τὸν θεὸν 'therefore, the one who disregards this, disregards not a human, but God.' So there is a disregarding event that still takes place here, but it is not a human who is disregarded--it is God. This kind of 'not...but' construction also involves ellipsis. The verb is not repeated after ἀλλά, but it is understood to be the same verb.

Contrastive Focus

Our third topic is contrastive focus. We have already mentioned this kind of construction--we have it when we negate one constituent and then present an alternative for which the predicate holds. We saw this in 'the one who disregards this, disregards not a human, but God' where the alternative 'but God' is focused.

Focus in semantics and pragmatics is best defined as the negation of alternatives. For more on the meaning of focus, you can check out our video on basic pragmatics concepts. We can briefly illustrate the meaning of focus with an example. When we say something like 'Mary chased after FRED', we implicitly understand that there is someone else in the context that Mary could have chased after, but she chased Fred and not that other person.

This is exactly the kind of construction that is explicitly given in constituent focus under negation. Again, 1 Thessalonians 4:8 makes this contrast. While you might think that you are disregarding a mere human by disregarding the command Paul gives in the previous verse, you are actually wrong if you think this. If you disregard the commandment, it is God you are disregarding.

The Interpretation of Romans 2:28

Okay, let's finally return to our verses. In 28, we read οὐ γὰρ ὁ ἐν τῷ φανερῷ Ἰουδαῖός ἐστιν οὐδὲ ἡ ἐν τῷ φανερῷ ἐν σαρκὶ περιτομή. Here, we have constituent negation with οὐ which goes with ὁ ἐν τῷ φανερῷ. Importantly, it does not go with the verb ἐστιν. If it did, we would expect the negative element to immediately precede the verb, but it doesn't. After the verb itself, we have οὐδέ. As we said, this conjunction loves ellipsis, so it is no surprise that we don't have an explicit verb. The verb for this clause comes from the first clause, so it would also be ἐστίν.

Now let's go back and look at the translation proposed by Arneson and endorsed by Thiessen. It says, "For it is not the external Jew, nor the external circumcision in the flesh." There are two big problems with this translation, though both are related. First, the English suggests that there is constituent negation that is modifying 'the external Jew'. This is in the phrase 'it is not the external Jew.' This correctly translates οὐ γὰρ ὁ ἐν τῷ φανερῷ Ἰουδαῖός, but crucially, it leaves ἐστίν untranslated, despite the presence of the verb 'is' in English.

Arneson's translation suggests that ἐστίν is actually the initial 'it is', but this would involve translating ἐστίν as the initial 'it is' in Greek. Given the location of έστίν in the sentence, this isn't possible. We already said that the initial "it is" in English is just a dummy subject and verb marking the constituent negation, but the "it is" is not needed in Greek constituent negation. So even though we might add 'it is' to best render this sentence into English, the initial 'is' cannot reflect the meaning of ἐστίν, as Arneson's translation suggests.

To accurately reflect the Greek, we would need another verb 'is' in English, and we would need a predicate. As we said from the outset, Arneson's proposal requires us to understand the very last relative clause in verse 29 'whose praise is not from man but from God' as the predicate. However, the word οὗ in Greek meaning 'whose' is never used as a predicate in the New Testament. It is only a predicate with the meaning 'where.'

Possibly an even bigger problem is that even when οὗ is the predicate, it is always found before the verb. In our verses, οὗ is found 21 words after the verb, and this is exactly why it is most naturally read as a description of the Jew being discussed rather than the predicate. So Arneson has to assume that this is the only time where οὗ means 'whose' as the predicate in the New Testament, and it is the only time where it is found after the verb as a predicate. In other words, it is almost certainly not the case that 'whose' is the predicate, which makes Arneson's translation make the wrong assertion.

What, then, is the predicate? The only real option is to make Ἰουδαῖός before ἐστιν the predicate of the clause. It could be that a second Ἰουδαῖος is also elided and serves as the subject to prevent redundancy, and we shouldn't be scared of missing words at this point. It is a normal part of language. But we could also just say that the subject is ὁ ἐν τᾦ φανερᾦ where the subject is a generic individual. Because we are talking about Jews, it would be natural to supply this as the referent of the generic person anyway, so it really doesn't matter too much in this context. A better translation, then, is 'For it is not the outward one who is a Jew.'

This leads into the second problem with the translation. Because Arneson's translation does not have a predicative copula in the first clause, it leaves the next clause without one as well. His translation 'nor the external circumcision in the flesh' only negates the constituent "external circumcision of the flesh," but it does not account for the implicit ἐστίν we would expect when οὐδέ is used. A better translation of the whole verse, then, would be 'For it is not the outward one who is a Jew, nor is it the outward one in the flesh that is circumcision.' This translates both the constituent negation and the explicit ἐστιν in the first clause and the implicit one in the second clause.

The Interpretation of Romans 2:29

Now what about verse 29? This begins with ἀλλʼ ὁ ἐν τῷ κρυπτῷ Ἰουδαῖος. As we said, constituent negation is often followed by ἀλλά 'but' to give the positive of what is being said, though it may be contrary to expectations. We again need to supply the verb ἐστίν which we saw from the very first clause, as is normal in this kind of construction. That verb connected a predicate and a subject, so it should do the same here.

Arneson's translation, again, does not do this. It just says "but the internal Jew." If we merely gloss each individual word outside of any context, this translation is fine, but this clause has to have a predicate (which Arneson's translation lacks) because the verb ἐστίν has been elided. So we could say 'but the one in secret is a Jew.' The same argument applies to the next parallel clause καὶ περιτομὴ καρδίας ἐν πνεύματι οὐ γράμματι which Arneson translates 'and the circumcision of the heart in spirit and not in letter.' Again, we need the elided verb, and here the subject is also elided, which is, again, completely normal when the element is found in the context, as περιτομή is. So we have 'and [circumcision] is circumcision of the heart by the spirit.'

Finally, the last clause is οὗ ὁ ἔπαινος οὐκ ἐξ ἀνθρώπων ἀλλʼ ἐκ τοῦ θεοῦ which Arneson translates as 'whose praise is not from humans but from God.' As we have said, for numerous reasons this is best taken as a relative clause describing the internal Jew. The assertion of the clause is not 'It is the internal Jew whose praise is from God.' Rather, the assertion is 'The internal one (or Jew), whose praise is from God, is a Jew.' We can provide the following translation, then, for our verses:

'For it is not the outward one who is a Jew, nor is it the outward one in the flesh that is circumcision, but the one in secret is a Jew, and circumcision is circumcision of the heart by the spirit, not the letter, whose praise is not from humans, but from God.'


So is Paul redefining what it means to be a Jew here? There is no getting around the syntax of these verses. Paul must be saying that, contrary to our expectations, it is not those who are outwardly circumcised that are Jewish, but it is the one who is circumcised in the heart that is Jewish. Of course, those who are outwardly circumcised may still be circumcised in the heart, but the implication seems to be that those who are not outwardly circumcised (that is, gentiles) can still be circumcised in the heart and would, therefore, qualify as being a Jew.

Whether we consider this a "redefinition" depends on how we understand "Jew" to be defined in the Bible as a whole. Is being Jewish primarily about physical circumcision or circumcision of the heart? The best way to understand Paul, a Jew himself who was steeped in the Jewish scriptures, is to say that Paul is interpreting circumcision here, not redefining it. He was participating in an inner-Jewish dialogue about what it meant to be God's people. Jews are those people who adopt the ancestral customs and laws of the Jewish God.

Could that be done without circumcision? Certainly not. Circumcision was necessary, but according to Paul's argument here, the circumcision that makes one Jewish is just not physical. It is circumcision of the heart by the Spirit.