Romans

Romans is the longest, the most carefully built, and the least occasional of Paul's letters, which is why it is so often read as a theological treatise. It is not one. It is a letter with an errand.

Paul had never been to Rome. He did not found that church and, by his own stated policy in 15:20, he does not build on someone else's foundation. He writes because he intends to go to Spain (15:24) and needs Rome as his base of support for a mission into a region where he has no networks, no synagogue connections, and probably no Greek. A church he has never met is being asked to fund and vouch for him. That alone explains why the letter lays out his gospel at length: it is, among other things, a statement of what the sponsor would be sponsoring.

There is a second errand. Paul is about to take the collection from the Gentile churches to Jerusalem (15:25-27), and he is afraid of how it will be received (15:30-31). The collection embodies the whole argument of the letter, that Gentile believers are debtors to Israel and genuine members of the same family. If Jerusalem refuses it, the unity he has spent his ministry arguing for is broken in public.

And there is a situation in Rome itself. Suetonius reports that Claudius expelled Jews from Rome over disturbances instigated by one Chrestus, an event usually dated to AD 49 and usually taken as a garbled reference to conflict over Christ. Acts 18:2 confirms an expulsion. Claudius died in AD 54 and the edict lapsed. That gives you a congregation that spent five years without its Jewish members and then received them back into a house church culture that had changed. Chapters 14 and 15, on the weak and the strong, on food and days, read like a pastoral response to exactly that.

Hold those three together and the shape of the letter makes sense. Paul is defending the proposition that God has been faithful to Israel precisely in including the nations, and he is defending it to a congregation where that question is not academic.

The argument begins by building a case that has no exits. From 1:18 Paul describes the Gentile world: God's power and nature are visible in what has been made, humanity exchanged the glory of the immortal God for images, and God gave them over to the consequences. It is a devastating passage and it works, rhetorically, as a trap. A Jewish reader nodding along through chapter 1 walks into 2:1: 'You, therefore, have no excuse, you who pass judgment on another.'

Chapter 2 turns the same standard on the person holding it, and chapter 3 closes the case with a chain of Old Testament quotations: 'There is no one righteous, not even one' (3:10). Then the summary verdict at 3:20: 'Therefore no one will be justified in His sight by works of the law. For the law merely brings awareness of sin.' Whatever you conclude about what works of the law means, this much is textual: Paul has just spent two and a half chapters showing that both groups stand under the same judgment.

3:21 is the hinge of the letter. 'But now, apart from the law, the righteousness of God has been revealed, as attested by the Law and the Prophets.' Then 3:24-25: 'justified freely by His grace through the redemption that is in Christ Jesus', with God presenting Christ as a hilasterion. That word is contested and worth knowing about. The BSB renders it 'atoning sacrifice', which does not choose between the readings that follow, while the ESV renders it propitiation. It can mean a propitiation, something that turns away wrath; it can mean an expiation, something that cleanses sin; and in the Greek Old Testament it is the standard term for the mercy seat, the lid of the ark where blood was sprinkled on the Day of Atonement. Leon Morris argued hard for propitiation and C. H. Dodd argued hard for expiation. The mercy-seat reading, favoured by many recent commentators, arguably contains both, since the point of the day was that God provided the place where wrath and cleansing met.

Chapter 4 then takes the one case that could sink the argument. If anyone was justified by his obedience it was Abraham. Paul answers with Genesis 15:6, quoted at 4:3, and with an observation about sequence: Abraham was counted righteous before he was circumcised (4:9-12), which makes him the father of uncircumcised and circumcised believers alike. Notice that the chapter is not only about how an individual is saved. It is about who gets to be in Abraham's family, which is the question the whole letter is answering.

These four chapters are the part of Romans most Christians could quote from, and they move from the verdict to the life the verdict makes possible.

Chapter 5 opens with the consequence: 'Therefore, since we have been justified through faith, we have peace with God through our Lord Jesus Christ.' Then it sets Adam against Christ. One trespass brought condemnation for all men; one act of righteousness brought justification and life for all men (5:18). The hinge phrase in 5:12 is two Greek words, eph' ho, usually translated 'because' in modern versions, so that death was passed on to all because all sinned. The BSB prints 'because all sinned' in its text, which is the majority rendering and not the only one defended. The old Latin version rendered it in quo, 'in whom', and Augustine built a great deal on that reading, in which all sinned in Adam. Defenders of the Latin reading point out that Paul's own argument keeps saying the many died through the trespass of the one man (5:15, 5:17); defenders of 'because' reply that the grammar of eph' ho is the issue, and that Paul says all sinned. Most Greek scholars now regard 'because' as correct, but Paul's argument still binds the race to Adam in a way that has to be accounted for.

Chapter 6 answers the obvious objection. If grace abounds where sin increased, then sin more. Paul's answer is not a rule but a death: you were baptised into Christ's death, and a dead person is not available for the old service. Chapter 7 is harder. It describes a self at war, wanting good and doing evil, and the identity of the 'I' has been argued since the second century. Is it Paul as a Christian, Paul before Damascus, Adam, Israel under the law, or a rhetorical voice Paul adopts? Stanley Stowers and others have argued it is prosopopoiia, speech in character, a recognised ancient device. Augustine read it of the believer, and much of the Reformed tradition followed. Given Philippians 3:6, reading it as straight autobiography of Paul's pre-Damascus conscience is difficult. Teach it as open.

Chapter 8 is where the letter's argument arrives. 'Therefore, there is now no condemnation for those who are in Christ Jesus.' Life in the Spirit, the Spirit of adoption to sonship with the cry 'Abba! Father!', creation itself groaning together in the pains of childbirth, and the closing sentence that nothing in all creation will be able to separate us from the love of God that is in Christ Jesus our Lord. It is worth noticing that this famous crescendo arrives in a passage about present suffering, not about escape from it.

These chapters are not a detour. They are the point at which Paul finally faces the objection he has been deflecting since 3:1. If the gospel is going out to the nations while most of Israel does not believe it, has the word of God failed?

He answers in three movements. Chapter 9 says the promise never ran through physical descent alone, as the choices of Isaac over Ishmael and Jacob over Esau show, and pushes the argument to its hardest edge with the potter and the clay. Chapter 10 says Israel's problem is a righteousness of their own rather than God's, and insists the message has genuinely gone out. Chapter 11 says the rejection is neither total, since there is a remnant and Paul is part of it, nor final: branches broken off can be grafted back in, and the Gentile branches grafted into a cultivated olive tree have no standing to boast.

Then 11:25-26, and the famous sentence: 'And so all Israel will be saved.' The English 'so' can mean 'in this way' or 'and then'; the Greek is usually read 'in this way', but that choice is part of the argument. Three main readings are defended by serious people. The first takes all Israel as ethnic Israel, with a large-scale turning to Christ at the end, and points to the rhetoric of the chapter and to 11:29: 'God's gifts and His call are irrevocable.' The second, going back to Calvin and defended in modern form by N. T. Wright among others, takes all Israel as the whole people of God, Jew and Gentile together, and leans on Paul's redefinitions earlier in the section. The third takes it as the total number of elect Jews gathered in across history, which fits the remnant language of 11:5. A fourth position, sometimes called the two-covenant or Sonderweg reading, held by Franz Mussner and in a qualified form by Krister Stendahl, argues Paul expects Israel to be saved without coming through explicit faith in Christ; most interpreters reject it, but you should know it exists and why its defenders think 11:26 supports it.

What is not in dispute is the emotional register. Paul opens chapter 9 saying he could wish himself cursed and cut off from Christ for the sake of his brothers, his own flesh and blood, and closes chapter 11 with a doxology rather than a resolution. Whoever teaches these chapters without that grief has misread them.

This is the live one, and the course does not take a side. Read this section slowly and notice that both cases are being put as their own defenders would put them.

Start with the word. Dikaioo is a Greek verb and the noun group around it, dikaiosune and dikaios, is one family in Greek but splits in English into righteousness, justice and justification, which is a permanent source of confusion. In the Greek Old Testament the verb sits overwhelmingly in legal settings. Deuteronomy 25:1 has judges acquitting the innocent and condemning the guilty. Proverbs 17:15 condemns whoever justifies the wicked. In those settings the verb plainly does not mean to make someone virtuous; it means to declare a verdict in someone's favour. That is the philological backbone of the forensic reading. Against it, some note that verbs of this Greek form can be causative, that Paul's usage is not bound by Septuagint usage, and that the Latin iustificare, which shaped the Western church, does mean to make righteous, which is one root of the medieval and Roman Catholic account of justification as including transformation. The Joint Declaration on the Doctrine of Justification signed by the Lutheran World Federation and the Roman Catholic Church in 1999 did not dissolve that difference but did establish that the two traditions were not condemning each other's actual positions.

The Reformation reading, at its strongest. Paul is answering the question of how a sinner, guilty before a holy God, can be accepted. The answer is that God pronounces a verdict of righteous over the ungodly on the basis of Christ's work received by faith alone, apart from anything the person does. The textual case is heavy: 3:20 and 3:28 say no one is justified by works of the law, 4:5 speaks of God justifying the ungodly, 5:1 makes peace with God the immediate consequence, and chapter 4's treatment of Abraham turns on believing rather than doing. The theological case is that this is what makes assurance possible. If my standing before God depends in any part on my performance, I cannot know where I stand. Modern defenders in this line include Stephen Westerholm, whose 'Perspectives Old and New on Paul' (2004) surveys the history of interpretation at length and concludes that the Reformers understood Paul substantially correctly, and Thomas Schreiner, Douglas Moo and John Piper, whose 'The Future of Justification' (2007) was written as a direct response to Wright.

The New Perspective, at its strongest, and this requires getting Sanders right. E. P. Sanders published 'Paul and Palestinian Judaism' in 1977. His main work in that book was not on Paul but on Jewish sources, and his conclusion was that the pattern of religion in Second Temple Judaism, which he called covenantal nomism, worked like this: God chose Israel by grace, the law was given as the means of maintaining that relationship rather than earning it, obedience was the response of those already in the covenant, and where obedience failed the law itself provided means of atonement and repentance. In his shorthand, one got in by God's grace and stayed in by obedience. If that is right, then the caricature of Judaism as a religion of earning salvation by works, which much Protestant preaching assumed, is simply false, and Paul cannot have been attacking it. Sanders's own account of Paul was that Paul reasoned backward from solution to plight: having met the risen Christ, he concluded that everyone must have needed him, and therefore that the law was not the way. Sanders wrote, bluntly, that what Paul finds wrong in Judaism is that it is not Christianity. Note that this is Sanders, and that it is not the same as what came after him.

James D. G. Dunn coined the phrase 'the New Perspective on Paul' as the title of the Manson Memorial Lecture, delivered in 1982 and published in 1983, and took Sanders's picture of Judaism but drew a different conclusion about Paul. If Judaism was not legalistic, then when Paul says works of the law he is not attacking moral effort. Dunn argued that the phrase refers to the observances that marked Jews off from Gentiles, above all circumcision, food laws and Sabbath, and that Paul's target is the use of the law as a boundary fence around the covenant. Justification, on this reading, is God's declaration about who belongs. N. T. Wright developed this with a covenantal and law-court frame: the righteousness of God means God's own covenant faithfulness, justification is the verdict that a person is a member of the covenant family, and that verdict is announced in the present on the basis of faith and will be confirmed at the last on the basis of the whole life lived by the Spirit. Wright insists this last point is Paul's own, citing Romans 2:6-13 and 14:10-12, and denies that it makes salvation a reward for effort.

Now the critics, given the same room. First, on Sanders's picture of Judaism. The two volumes of 'Justification and Variegated Nomism' (2001 and 2004), edited by D. A. Carson, Peter O'Brien and Mark Seifrid, gathered specialists to examine Second Temple texts one corpus at a time, and their finding was that the sources are more varied than one pattern allows: some texts fit covenantal nomism well, others appear to make final acceptance depend on merit or on attained righteousness. Friedrich Avemarie showed a genuine unresolved tension in rabbinic material between election and reward. Simon Gathercole's 'Where Is Boasting?' (2002) argued specifically that Jewish texts of the period do anticipate final vindication on the basis of obedience, and that Paul's polemic therefore cannot be confined to ethnic boundary markers. Second, on Dunn's reading of works of the law. Critics point out that Romans 3:20 grounds the claim in the fact that the law brings awareness of sin (the Greek says knowledge of sin; the BSB adds the word 'merely'), which is a moral observation and not an ethnic one, and that Romans 2 and 4 have been talking about deeds. Third, on Wright's account of justification. The objection is that making justification primarily a statement about covenant membership shifts the centre of gravity from how a guilty person is forgiven to who is in the group, and that future justification on the basis of a life lived, however carefully qualified, unsettles the assurance the doctrine was formulated to secure. Wright's reply, in 'Justification' (2009), is that his critics have read the categories of a later system back into Paul and that the whole life lived is itself the Spirit's work, not the believer's achievement.

Two further streams keep this from being a two-party fight, and leaving them out would be misleading. The apocalyptic reading, associated with J. Louis Martyn's Galatians commentary (1997), Martinus de Boer and Douglas Campbell, argues that both the old and the new perspectives are too focused on human religion and that Paul's real subject is God's invasion of an enslaved cosmos to liberate it. And John Barclay's 'Paul and the Gift' (2015) reframed the whole debate by distinguishing several different things people can mean by grace: Sanders was right that Judaism was a religion of grace, Barclay argues, but grace was perfected differently in different texts, and Paul's distinctive move is the incongruity of the gift, that it is given without regard to the worth of the recipient. That book has moved many people on both sides, which is a rare thing.

Finally, one exegetical fault line that runs under the whole argument. The phrase pistis Christou at Romans 3:22, 3:26, Galatians 2:16 and Philippians 3:9 can be read as faith in Christ or as the faithfulness of Christ. Richard Hays made the sustained case for the second in 'The Faith of Jesus Christ' (1983); Dunn and others have defended the first. The Greek is ambiguous and the decision affects how the whole letter reads. Note that the BSB's text chooses the first reading at each of these places ('faith in Jesus Christ' at 3:22, 'faith in Jesus Christ' at Galatians 2:16, and 'faith in Christ' at Philippians 3:9), so a student reading only the BSB will not see that there is a question; Bible Hub's interlinear shows the bare Greek, and the case for 'the faithfulness of Christ' deserves to be weighed at full strength.

So what is settled and what is not? Settled: Paul insists no one is justified by works of the law; he insists Gentiles come in on the same terms as Jews; the vocabulary is drawn from the courtroom; and the caricature of all Second Temple Judaism as crude works-righteousness is dead and will not be revived. Not settled: what works of the law denotes, whether the righteousness of God is God's attribute, God's activity or a status conferred, whether justification's centre is the forgiveness of a guilty individual or the definition of the covenant people, and whether the two must be alternatives at all. Anyone who tells you this argument is over is telling you which side they are on.