Interpretation of Rom 4:1

Glenn Blank glennblank at
Mon Jan 28 15:01:42 EST 2002

>Thanks for the interaction. I enjoy it very much.

I have enjoyed it as well.  Your interpretations are always

I hope I have snipped your post fairly to preserve the crux of your

(To the rest of the list: warning, this post has gotten long, and I fear is
beginning to border on exegesis.  That is not my intent:  I am trying to
work out in my mind why Moon's analysis feels awkward to me by trying to
dissect Paul's argument structure.  So if you find quibbles about logical
syllogisms tedious, spare yourself the bother of reading further -- the
upshot of everything I write below is that I think a "wh-" question in Rom
4.1 fits the discourse structure better than a "yes/no" question.)

I had asked
>>Can verbs of "saying"
>> in NTG take an infinitive without accusative "subject" as a complement?
Moon answered

>          [came  saying  also vision   of angels to see] 

OK.  That would seem to answer my question.  Languages subcategorize verbs
rather arbitrarily as to whether the infinitival complement of a particular
verb needs an explicit "subject" of the infinitive or not.  For example, in

 I decided _____{i} to go.
 I said _____{j} to go.
*I told _____{i or j} to go
 I told him to go      OR     I told myself to go.

(Where ____{i} denotes a null infinitival subject co-referential with the
subject of the matrix clause, and ____{j} denotes a null infinitival subject
with a different referent.)

I had written

>> as an example of the interrogative constituent being moved out of the
>> subordinate clause to the front of the matrix clause,  Mt. 16.13

> So, we can say that 
>Rom 4.1 is obtained by moving TI to the front from the underlying

Yes, exactly.

>But my analysis seems as plausible as the traditional one. 
>Any comments?

Yes, syntactically speaking, assuming that ERW subcategorizes for infinitive
complements the same way LEGW does.  But I still think the traditional
analysis fits the context better.  (See below)

>(3) Finally, about the "accusative and infinitive" construction, where
>the accusative functions as the subject of the infinitive clause, A. T. 
>Robertson thinks that it is a misconceived notion and hinders the reader
>rather than help. [snip]  So, it follows that we need not worry too much
about the lack of 
>"accusative" of the infinitive hEURHKENAI in Rom 4.1. 
Yes, I agree.  I mis-spoke.  What I was driving at was the nominative
subject of the matrix verb cannot also be the "subject" of the infinitive
(I'm not sure if I concur with Robertson's linguistic analysis, but that is
here nor there).  The referent of the subject of the main verb might also be
the referent of a null subject of the infinitive, but only for main verbs
that subcategorize for allowing null-subject infinitive complements, and you
answered that above with Lk 24.23.


>(a) In Romans, Paul uses "TI OUN" or "TI OUN EROUMEN" a number of times to
>start a new
>stage of argument, often to counter-attack potential false inferences the
>might have drawn from the previous discourse.

OK.  For example, in Ro 3.5, the preceding line of argument had been to show
that God is not unrighteous, and in fact our very righteousness demonstrates
His righteousness.  What might one naturally infer from this is the
proposition that good is coming from my unrighteousness so God is unjust in
inflicting wrath for what has resulted in good.  So Paul counters with "What
then?  Is God unjust for inflicting wrath?"  And in Ro 3.9, the proposition
"Their [people who deliberately sin so good may come] condemnation is just"
naturally leads to the inference that "we are better than they . . . *we*
don't deliberately sin so that good may come." 

But now what is the false inference that one might draw from the discourse
preceding 4.1? 


>only Jews are children of
>to whom the promise is made, and so Gentiles cannot be the people of God
>equal footing with Jews. Rom 4 can be taken to counter this objection. 

But I can't see how one could naturally draw this inference from the
preceding discourse.  In fact, Paul has been explicitly stating the exact
opposite -- that God is God of the Gentiles as well as the Jews (3.29).
>> >It seems that "What then? Shall we say that we have found Abraham to be
>> >our forefather according to flesh?", to which the expected answer is no,
>> >matches better the flow of Rom 3:27-31, where it is argued that God is
>> >God of Jews only, but of Gentiles as well
>> >and thus he justifies both by
>> >means of faith.

OK, if I follow you correctly here, the false inference presented by the
question in 4.1 is that we become Abraham's children by means of the flesh
[meaning by works, or by circumcision?].  But once again nothing in the
preceding discourse suggests that inference.  In fact Paul has already
explicitly that it is not by the flesh but by faith (3.27; 3.30).

But there is another way in which TI OUN? in 4.1 does not fit the pattern of
the other TI OUNs in Romans.  Paul's argumentation uses both "yes/no"
questions (in this case the expected answer to which is "no") and "what"
questions (to which the expected answer is a full proposition.)  Your
analysis makes 4.1 a "yes/no" question.  My analysis makes it a "what"
question.  In every other instance of a "y/n" question in Romans, it is
followed first by a denial of the inference expressed in the question and
then reasons why that inference cannot be true.  For example, 3.9 -- 

Statement -- "Their condemnation is just"
Inference -- "Our condemenation is not just:  we are better than they."
Question -- "What then?  Are we better than they?"
Denial -- "Certainly not."
Reason -- Quotes from the songs and Isaiah stating that no one is righteous.

The pattern for "wh" questions however is that the preceding discussion
suggests an inference which is in fact the proposition Paul is arguing for. 
Then there is a question, the answer to which is that proposition.  Then
Paul states that proposition in answer to the question and then continues
the preceding line of argument, further explaining it.
For example, 3.26-27

Statement -- God's righteousness is demonstrated in His grace, and we
receive that grace by faith [paraphrasing to summarize the preceding
Inference -- We don't have any room to boast then, since it is based not on
works but by faith.
Question -- "Where then is boasting?"
Answer -- "It is excluded"
Question -- "By what law?"  (or on what basis?)
Answer -- "Not by the law of works but by the law of faith."
Further explanation -- "Therefore, we conclude that a man is justified by
faith apart from the deeds of the law."

Another example, 8.31

Statement -- "All things work together for good to those that love God,
Inference -- Hey, God really is interested in my good.  He is for me, and if
He is for me, good things have to follow.
Question -- "What shall we say to these things"
Answer -- "If God be for us, who can be against us?"
Further explanation -- "If He delievered up his own Son for us, surely he
will give us all things."

In other words, Paul seems to use the "y/n" questions to track down a
tangent that a potential objection could take his audience down, but he uses
the "wh" questions to reinforce and further develop his own line of

ISTM Ro 4.1 fits this latter pattern better:

Statement -- The circumcised and the uncircumcised alike are justified by
faith.  (3.30)
Inference -- Then even the Jews were justified by faith, not by the deeds of
the law.
Question -- What, then, would we say Abraham, our father according to the
flesh, discovered?
Answer -- "Abraham believed God, and it was counted unto him for
righteousness." (4.3)
Further explanation -- (4.4-5) ". . . to him who does not work but believes
. . . his faith is accounted for righteousness." 
>According to [Michael Cranford], in Rom. 4, Paul explains further the
thesis of Romans
>3:27-31 that
>Gentiles can be the people of God by faith, not just the Jews.

But ISTM that Paul uses "wh" questions rather than "y/n" questions for that
kind of transition into further explanation.

>He does that by arguing that Gentiles can be the children of Abraham by
>faith to whom the promise is given that he will be the heir of the KOSMOS
>(rom 4.13) and he will the father of all nations (Rom 4.17).

Yes.  But the issue of him being the father of all nations is only an
intermediate step to Paul's main argument, which is not whether or not
Gentiles are the children of Abraham, but *how* they became the children of
Abraham -- that is, by faith.

>If Abraham is the forefather of the people of the law only, the
>only, Gentiles are out by definition. So, it is critically important
>to establish that Abraham is the father to Gentile believers 
>as well as to Jews. I also find related statements in Galatians:

>Know therefore that those who are of faith are children of Abraham (Gal
>If you are Christ's, you are Abraham's seed and heirs according to the
>promise (Gal 3:29).

The issue in Galatians certainly is related to the issue in Romans.  But the
argument structure ISTM is different, in fact almost opposite in approach. 
The verses you quote have the proposition "Those who are faith" and "You are
Christ's" as the protasis, and "are children/seed of Abraham" as the
apodosis.  In Romans, on the other hand, the argument seems to move in the
opposite direction -- from the assumption that Gentiles are included in the
Abrahamic covenant to the conclusion that that could only be possible if the
covenant were based on faith rather than the flesh.  It does indeed seem
critically important in Galatians to establish the fact of whether or not
the Galatians are children of Abraham.  In Romans, on the other hand, that
seems to be an intermediate step in the arugment development.

This is what I mean:

Protasis  (4.9)  "Does this blessedness then come upon the circumcised
only?" -- assumed answer, "upon the uncircumcised also."

Apodosis (4.10)  "How then was it accounted?  Not while circumcised, but

Conclusion (4.16)  "Therefore, it is of faith that it might be according to

(intervening between 4.10 and 4.16 is a restatement of the interaction
between protasis and apodosis, and then an argument "ad absurdum" -- i.e.,
Assume that we received the promise by law;  conclusion, then we are under
wrath and the promise to Abraham was of no effect . . . which contradicts
the opening assumption, that Abraham was to be the father of many nations
(restated in 4.17), and the supposition that we receive the promise by law
cannot be correct.

In fact, regardless of whether you translate Romans 4.1 as a "wh" question
or a "y/n" question, the issue of whether Gentiles are the children of
Abraham is not the point of the question but presupposed in the question:


The answer is "no, he did not get to be our forefather *according to the
flesh*."  But that presupposes an assumption common to both writer and
audience that Abraham *is* TON PROPATORA hHMWN.

"What then did Abraham our father according to the flesh discover"

presupposes that Abraham is the Gentiles father, but moves from there to the
issue of what he discovered, which is that believing God was imputed to him
for righteousness (and in the following verses) if we are heirs of his
promise, it will be on the same basis as he received the promise -- by
>these statements [Gal 3.7 and Gal 3.29] indicate that the issue at hand is
for Gentile believers
>to be children of Abraham.

In Galatians, yes, but in Romans I would say the point that Gentile
believers are to be the children of Abraham is a side issue, which serves to
lend support to the main issue at hand -- "of faith according to grace." 
For the sense that (4.16) is the main point rather than (4.10-14), notice
the distribution of EK PISTEWS and KATA CARIN throughout the book, while the
issues of Abraham as PATERA or PROPATORA and Gentiles as KLARONOMAI of
Abraham occur only in this chapter.

>if Abraham is  only a typical example showing how one is justified?

In order to help establish his main point, which is precisely to show "how
one is justified."  His arguments about Gentiles being children of Abraham,
being limited to this chapter, are merely one approach among many for
establishing the fact of justification by faith.

In fact, Paul explicitly states in (4.23-24) that he is treating Abraham as
a typical example of how we are justified (EGRAFH . . .DI HMAS OIS MELLEI

>It is granted
>that Rom 4:2-8 is difficult to interpret from this perspective.

Difficult to interpret in your analysis, perhaps.  But crystal clear with my
analysis of 4.1

> But the points of these verses are not as clear as the points of 4:9-12
>the rest of
>the chapter. So, we need to interpret Rom 4:2-8 in view of the rest of the
>and the last paragraph of chapter 3, rather than make a particular
>of a unclear passage Rom 4:2-8  affect the other parts. 

I am not sure how one decides 4.9-12 is more clear than 4.2-8, or which
should be interpreted in light of which.  But my interpretation of 4.9-12
does not depend on 4.2-8, but instead on the larger context of Romans.  And
my understanding of the larger context and of how the arguments in 3.27-31,
4.2-8; 4.9-12, et al, relate to each other informs my analysis of 4.1, and
not the other way around.

I think we come out in the same place, so how we analyze 4.1 won't change
the ultimate point we understand Paul to be making.  But these are the
reasons the TI OUN? analysis seem awkward to me.

Enjoying the chase  :)

glenn blank
Pensacola FL

More information about the B-Greek mailing list