Hostname: page-component-745bb68f8f-5r2nc Total loading time: 0 Render date: 2025-02-11T02:26:14.786Z Has data issue: false hasContentIssue false

Prison, Penance or Purgatory: The Interpretation of Matthew 5.25–6 and Parallels

Published online by Cambridge University Press:  08 March 2018

Nathan Eubank*
Affiliation:
Keble College, University of Oxford, Oxford OX1 3PG, UK. Email: nathan.eubank@theology.ox.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

Scholarship on Matt 5.25–6 has focused on the question of whether the saying offers mundane wisdom or threatens divine judgement, with the majority concluding that it refers to eternal punishment in hell. This article examines debt-prison and related phenomena before turning to the illuminating history of ancient interpretation. The article concludes that the ‘eternal damnation’ gloss widely favoured today is an overinterpretation first inspired by the exigencies of fourth- and fifth-century doctrinal controversy. Instead of eternal perdition, Matt 5.25–6 and its parallels suggest a time of straits followed by possible release.

Type
Articles
Copyright
Copyright © Cambridge University Press 2018 

Introduction

In Matt 5.21–6, the first of the Sermon on the Mount's so-called ‘antitheses’, Jesus expands the Torah's prohibition of murder to include anger, various insults, and finally any sin against a brother (v.23). In verses 25–6 the imagery shifts to a legal dispute: ‘Make friends with your opponent quickly, while you are on the way with him, lest your opponent hand you over to the judge, and the judge to the assistant and you will be thrown into prison.’Footnote 1 The saying concludes with a direct warning to listeners: ‘Amen I say to you, you will not go out from there until you repay the last penny.’ Parallel warnings appear in Matthew's parable of the Unforgiving Servant (18.23–35), Did. 1.5 and Luke 12.57–9. Commentators have focused on the question of whether the saying offers mundane wisdom or threatens divine judgement, with the majority concluding that it refers to eternal confinement in hell.Footnote 2 Turning to the reception of this saying in antiquity, however, one encounters a very different debate. Interpreters in the first three centuries vigorously disputed the meaning of Matt 5.25–6 and parallels, but generally assumed that the ‘imprisonment’ – whether literal or figurative – was only temporary, a possibility scarcely considered in modern scholarship.Footnote 3

In this essay I examine debt-prison and related phenomena before turning to the illuminating history of ancient interpretations of this saying. I argue that the ‘eternal damnation’ gloss widely favoured today is an overinterpretation inspired by the exigencies of fourth- and fifth-century doctrinal controversy. Instead of eternal perdition, Matt 5.25–6 and its parallels suggest a time of straits followed by possible release.

1. Prison: Coercing Payment

How would the threat of incarceration terminable by payment of money resonate with a first-century audience?Footnote 4 The scope of debt-prison and related practices in the Roman period is contested.Footnote 5 There were a number of attempts to limit or stop the seizure of debtors in the ancient world. The Torah knows nothing of debt-prison and places severe constraints on debt-slavery.Footnote 6 Rome ostensibly outlawed debt-bondage in 326 bce, and another law seems to have been introduced in the early Principate which allowed debtors to avoid incarceration by ceding only their property.Footnote 7 A well-known edict of the Egyptian prefect Tiberius Julius Alexander in 68 ce states that it is the will of divine Augustus that no free person should be imprisoned for private debt.Footnote 8 The impact of these laws is debated, but one thing is clear: imprisonment for debts public and private continued.Footnote 9 One piece of evidence of the widespread threat of confinement is the praxis-clause, a ubiquitous feature of contracts well known from Egypt and beyond for centuries. The praxis-clause defines the right of execution, that is, what the creditor is able to seize should the debtor fail to pay on time. Sometimes it is stated that the creditor may seize the debtor's possessions, but very often contracts add that the debtor him or herself could be seized (e.g. ἔσται σοι ἡ πρᾶξις ἔκ τε ἐμοῦ καὶ ἐκ τῶν ὑπαρχόντων μοι πάντων).Footnote 10 A less common stock phrase in documentary papyri is the ἀγώγιμος-clause, which sometimes accompanies the threat of praxis and states that the debtor is ‘liable to be seized and confined until he pays in full’ (e.g. ἀγώγιμον καὶ συνέχεσθαι μέχρι τοῦ ἐκτεῖσαι).Footnote 11

The threat of imprisonment was a powerful motivator to pay one's creditor on time. Prisons were often dark, airless and filthy.Footnote 12 Torture, which is mentioned in connection with debt-prison in Matt 18.34, seems to have been common.Footnote 13 Family or friends often had to supply food, and starvation was a possibility.Footnote 14 All this provided good reason to remain solvent. Seizure of the debtor's person seems to have been the last resort.Footnote 15 Once the threat was exhausted, what was incarceration itself supposed to accomplish? Judging from instances we know about, the goal was not merely punitive, punishing the debtor for failing to do what is right. Prison stays tended to be short, coercive affairs used by private creditors and even more often by the state to ensure payment.Footnote 16 Like the seizure of possessions, the point was to recoup the lost money. By their nature contracts threatening incarceration do not always reveal the eventual outcome, but other evidence gives us an idea of possible scenarios. If the debtor had any remaining resources one would expect him to act quickly to free himself. Cicero discusses a debtor whom assessors found guilty of failing to repay a loan.Footnote 17 He was unable to offer restitution at his sentencing, so he was given to the creditor as an addictus (bondservant). Though he had just lost his freedom because of his debt, he still had assets in the form of slaves. So, shortly after being made a bondservant, he gave a few of his slaves to his creditor and regained his freedom. In third-century bce Egypt a Cyrenean man named Callidromus acquired a donkey unlawfully, and the owner had him imprisoned to compel him either to pay for it or return the donkey.Footnote 18 The quickest path to freedom in this case would be to return the animal or pay up.

Most insolvent debtors seem to have relied on the kindness of family or friends. Once imprisoned, debtors sometimes wrote letters asking for help, as for example one Aurelius Sarapion, who wrote to his father and brother begging them to send money because he was locked in the local account office.Footnote 19 Lucian's Timon speaks of ransoming a man (ἐλεήσας ἐλυσάμην αὐτόν) who had been locked up until he paid the city 16 talents.Footnote 20 Dio Chrysostom describes how the ancient Athenian Miltiades died in prison unable to pay a fine of 50 talents. His son was locked up to assume the debt in his place, but managed to convince a wealthy man to pay his ransom by offering his sister to him in marriage.Footnote 21 Some debtors escaped incarceration by a hair's breadth. Plutarch and Cicero describe condemned debtors being rescued en route to prison or bondage.Footnote 22 The rescuing of debtors was common enough for Seneca to discuss the relative merits of saving debtors who had fallen into servitude to their creditors from one's wealth versus borrowing and begging to raise the amount needed.Footnote 23 First Clement claims that early Christians did even better. ‘Many among us’, the author writes, ‘have handed themselves over into chains, so that they might ransom others’ (55.2). A compassionate third party might also go surety for a prisoner. One note of indebtedness from 23 ce reveals how a man named Theon went surety (ἐγγυάω) for someone who had been imprisoned by the local ὑπηρέτης on account of some gold.Footnote 24 Theon won the man's temporary freedom, but put himself on the hook, as he solemnly promised to produce the man in thirty days, pay the debt or be imprisoned himself.Footnote 25 As Ben Sira said, ‘Do not forget the kindness of your surety (ἔγγυος), for he has given his life for you’ (29.15).

It is sometimes suggested that Jews did not imprison debtors and that Matt 5.25–6 deliberately evokes a foreign, Gentile setting.Footnote 26 It is true that Jewish law knows nothing of debt-prison and in Matthew's own, possibly Syrian, context most local magistrates would have been Gentiles. It would exceed the evidence, however, to claim with any confidence that Jews refrained from imprisoning their debtors. Among extant contracts between Jews, some do threaten execution (πρᾶξις) on the person of those who default, and there is also evidence of debt-slavery and of dependents given up as pledges, presumably to acquire loans.Footnote 27 Josephus claims that records were burnt in Jerusalem at the outset of the revolt in order to forestall the collection of debts (τὰς εἰσπράξεις ἀποκόψαι τῶν χρεῶν) and win many poor Jerusalemites to the cause.Footnote 28 Thus, while a Jewish audience might tend to associate judges and prisons with hostile goyim, the evidence suggests that insolvent Jews sometimes lost their freedom to fellow Jews.

In sum, debt-prison seems to have been used to ensure payment, either by scaring debtors to pay on time, or, should that fail, by forcing them or their friends to pay the ransom. Death in prison was always a possibility, but it was not the goal. The time spent incarcerated could be very brief. If payment was made prisoners could expect to go free.

2. Ancient Interpreters of Matt 5.25–6 and Parallels

How did the earliest readers of the gospels understand Matt 5.25–6? With such an obscure statement one expects a variety of interpretations, and this is indeed what we find. Is the saying about this life or the next? What exactly is the prison? Who is the opponent? Early readers offer a range of answers to these and other questions. But until the late fourth century it is assumed that in this passage, the Lukan parallel, and in Matthew 18 Jesus was talking about temporary restitution followed by release. No one takes it to mean eternal damnation or any other kind of final stasis. It was obvious to the ancients that the debtor's confinement came to an end.

Irenaeus (late second century), Tertullian (ca. 160–220) and Epiphanius (ca. 315–403) all complain that the Carpocratians cite Matt 5.26 to show that after death the soul is imprisoned in a new body and that this cycle of reincarnation continues until the soul has committed every possible sin and then ascends to God.Footnote 29 The late third–early fourth-century text Pistis Sophia interprets our passage along similar lines: the soul is repeatedly imprisoned in the body until it is suitably enlightened and then it is released.Footnote 30 In his counter-readings of the saying, Tertullian also assumes that the debtor goes free, but he offers different explanations of the nature of the imprisonment and its end. First he suggests that Jesus was speaking of an actual earthly prison.Footnote 31 Then he proposes that the ‘opponent’ is the devil (cf. 1 Pet 5.8) and the prison is the infernum, the lower regions from where ‘you will not be released until even the smallest offences have been paid for in the time before the resurrection’.Footnote 32 That is, after death, both good and evil people experience a foretaste of their eternal fate. After this temporary period of ‘paying the last penny’ they will go to their final home. In his attack on the supposed Carpocratian reading, it apparently does not occur to Tertullian to deny that the prisoner finds release.Footnote 33 Around the same time Clement of Alexandria (d. before 215) says that the arrest and imprisonment describe the suffering which Christians will experience at the hands of the devil and evil humans.Footnote 34 This imprisonment comes to an end, Clement says, because no evil powers ‘are able to separate us from the love of God’.Footnote 35

Cyprian of Carthage (bishop 248–58) wrote a book of instruction for converts ignorant of the Scriptures. It consists of points of catechesis such as ‘That we must boast in nothing’, and ‘That even our enemies are to be loved’, followed by biblical proof-texts designed to be self-explanatory and clear to the uninitiated.Footnote 36 One such catechetical point is that God will purge the sins of the faithful, but will not destroy them (fidelem emendari et reservari). Cyprian lists four passages to demonstrate this point.

In Psalm 116 [Ps 118.18 MT]: ‘The Lord has corrected me strictly (emendans emendavit), and has not delivered me to death.’
In Psalm 87 [Ps 89.32–3]: ‘I will visit their crimes with a rod, and their offences with scourges, but my mercy I will not scatter away from them.’
In Malachi [3.3]: ‘And he sits refining and cleansing as gold and silver, and he purifies the sons of Levi.’
In the Gospel [Matt 5.26]: ‘You will not go out from there until you pay the last penny.’Footnote 37

Cyprian apparently thought this verse not only promised eventual release from chastisement, but that it did so clearly and authoritatively. Elsewhere Cyprian quotes our passage to defend the church's practice of reintegrating Christians who had fallen away during a time of persecution.Footnote 38 Cyprian's opponents thought that those who had sacrificed to idols under pain of death could never return to communion. The Carthaginian bishop countered by citing many passages of Scripture showing God's acceptance of penitent sinners and by pointing out that these former apostates will not be restored in a quick, facile way. They will do penance, which Cyprian describes variously as paying the last penny in prison or as being cleansed by fire. Then they will be fully reintegrated.Footnote 39

Cyprian's contemporary in the east, Origen (d. 253/4), interprets the saying in light of the other economic language in the Synoptic Gospels. The coming judgement will bring a settling of accounts.Footnote 40 When the reckoning begins some will receive wages, others will be forced to repay vast sums, and others still will owe only the ‘last penny’.Footnote 41 Debtors will be thrown into prison, where they will repay their debts ‘by labour and works, or by punishments and torments’.Footnote 42 Origen does not know how long this will last, but supposes that those who owe vast sums, like the man in the parable of the Unforgiving Servant, will be in prison for a long time indeed. Those who owe only a penny will be released more quickly.Footnote 43 For Origen, it is a sign of God's great generosity that he allows sinners to repay the last penny, because it is through this process that God sets them free from evil.Footnote 44 As he writes in his Romans commentary, it is by paying the last penny that life comes to reign in all through Jesus because through this restitution all will be made righteous, able not only to hear the word of God but to do it.Footnote 45 As will shortly become clear, Origen's hope that everyone would repay the last penny and so be joined to God would prove pivotal for the fate of Matt 5.25–6 in later centuries.

Fourth-century interpreters continue to assume that the saying promises eventual release. Ambrose (ca. 334–97) carefully analyses the differences between the Matthean and Lukan versions and concludes that Matthew is about restoring peace between brothers, whereas Luke concerns sin of any sort.Footnote 46 To atone for these sins Ambrose recommends paying the last penny through acts of charity, though chastisements will also do.Footnote 47 Hilary of Poitiers (ca. 315–67) interprets Ps 69.5 christologically to argue that Jesus, who paid for what he did not take (quae non rapui, tunc exsolvebam), will pay for all people the last penny of their debts, which is why Jesus taught his followers to pray for debt-forgiveness in the Lord's Prayer.Footnote 48 In 381 Jerome described himself as a sinner waiting to ‘pay the last penny’, trembling but confident that ‘the Lord frees the prisoners [Ps 146.7]’.Footnote 49 Like Hilary, Jerome expects Christ to end the imprisonment.

Gregory of Nyssa (ca. 335–94) interprets Matt 5.26 along with the Unforgiving Servant to show that God applies just punishments to all, but will eventually eliminate all debts:

The gospel teaching speaks of a certain debtor who owed ten thousand talents [Matt 18.24], one who owed five hundred denarii, another fifty [Luke 7.41], and one who owed a quadrans, which is the smallest of coins [Matt 5.26]. The just judgement of God extends to all, adjusting the intensity of the recompense depending on the weight of the debt, but without overlooking even the smallest fault … When debtors have laid aside all that is alien to them – which is sin – and have stripped off the shame of their debts, they pass into freedom and confidence.Footnote 50

The repayment of debt is for Gregory another way of describing God's purifying fire which causes pain, not simply to punish, but to cleanse and draw humans closer to God. Similar interpretations of Jesus’ warning can be found in the writings of other fourth-century figures who shared Gregory's admiration of Origen.Footnote 51 In the west, Chromatius Aquileiensis (d. 407) says Matt 5.26 teaches that ‘it is not possible to be sent out from the fire of punishment until one has paid for even the least sin by the washing of punishment (ablutione poenae)’, a cleansing that Chromatius links with the fire in 1 Cor 3.15.Footnote 52

During the final decades of the fourth century the theology of Origen and his admirers received increasing scrutiny, leading ultimately to the Origenist controversy, a series of church-wide disputes over Origen's legacy that went on for centuries.Footnote 53 The controversy was famously multifaceted, but one of the principal issues was the Origenist hope that all post-mortem punishment was salutary so that, by its correcting influence, all would eventually be saved. In the wake of this controversy the notion that the prisoners’ stay in Gehenna would end upon payment suddenly became very problematic. Indeed, it is only as the Origenist controversy heats up that we have clear evidence of someone suggesting the debtor's stay in Gehenna must be everlasting. This suggestion comes from Augustine.

Though Augustine taught that prayers for the dead had a certain efficacy, he was a staunch anti-Origenist. He avoided describing punishment for sin as medicinal and insisted that there cannot be any true repentance or change of status after death, criticising other Christians who wanted hell to be temporary.Footnote 54 In the 390s, shortly after being ordained a priest, Augustine puzzles over the words ‘until you have paid’ and tries to avoid the implication that the suffering could end, though he admits he is not certain:

As for the words ‘until you have paid’, I should wonder if this doesn't signify the punishment which is called eternal. For how can the debt be paid where no opportunity for repenting and amendment of life is given? Thus, the expression ‘until you have paid’ is perhaps used in the same manner in which it said ‘sit at my right hand, until I place all your enemies under your feet’ [Ps 110.1] …Thus here it is possible to interpret ‘you will not go out from there until you have paid the last penny’ to mean you will never leave, because such a person will always pay the last penny while he suffers the everlasting penalty for earthly sins. Still, I would not say this so as to seem to stifle more diligent treatment of the punishment of sins.Footnote 55

This, we may say in retrospect, marks a revolution in the reception of this saying. Prior to the late fourth century there was no reason to avoid the plain sense of the words. Crucially, Augustine's exegesis is based on his prior assumption that post-mortem rehabilitation is impossible. By the 420s, after years of anti-Origenist debate, Augustine had developed a more confident defence of never-ending punishment, simply citing the ‘until’ of Matt 1.25 as evidence that the prisoner will not go free.Footnote 56

Origenists and their opponents continued to debate the meaning of Matt 5.26 for centuries.Footnote 57 In the sixth century we find letters between monks on how to refute those who cite Matt 5.26 as proof that ‘there is an end to the punishment of hell’.Footnote 58 One distraught monk encountered this interpretation and wrote to the famous recluse Barsanuphius to ask for help. Barsanuphius assured him that no one makes progress after death.Footnote 59 In the centuries to follow, commentators remain on guard against the Origenist reading.Footnote 60 Thomas Aquinas labels verse 26 a difficultas – why does the Lord imply eventual release when release is impossible? – and offers a range of attempts to avoid implying release from captivity.Footnote 61 The influential fourteenth-century Franciscan biblical commentator Nicholas of Lyra (1270–1349) says that ‘you will not go out until you pay’ really means ‘you will never go out, because in the infernum there is never a possibility of return’.Footnote 62 Like Augustine before him, Nicholas uses a preformed theological conviction to get around the sense of the words.

For the first 300 years of its reception, Jesus’ warning on debt-prison was interpreted in a variety of ways: as imprisonment in and eventual release from the body, as a literal prison, as a place of post-mortem anticipation of one's eternal fate, as penance in this life, as post-mortem purification, and as a fate from which one is redeemed by Christ. I have found no evidence that it ever occurred to anyone that the stay in prison could be permanent, even when it would have been convenient to do so. This, I suggest, is because such an interpretation simply did not fit with the ancients’ experience of how real-life debt-prison worked. The interpretive revolution came when Augustine and others needed to quash the hope of universal restoration, and it is the Augustinian interpretation which has in modernity acquired the prestige of seeming ‘plain-sense’.

3. Analysis

Were the ancients correct to assume that Matt 5.25–6 envisages release? One might counter with another question: how likely is it that everyone was wrong for 300 years and the theologically motivated post-Augustinian assumption that prison must be eternal is right? Though many of the ancient interpretations are overwrought by today's standards, the widespread assumption that verse 26 expects release cannot be dismissed as the product of the ancient Christian penchant for allegorising. Verse 26 is not just another detail – it is the solemn conclusion of the saying, which, like many other similar concluding statements in Matthew, drives home the point of the passage.Footnote 63 Moreover, as we have seen, interpreters offered a variety of allegorical and literal interpretations of the details in this passage, but the image of debt-prison always suggested eventual freedom. This – I propose – just is the gist of the saying. Precisely what it signifies theologically in its Matthean context is more difficult to pin down. In many ways these ancient readers were playing a different game from the one modern scholars are playing. They were constructing theologies from Scripture, whereas contemporary scholars are interested in isolating and understanding a particular book or pericope. And so, the questions remain: what sort of release might be in view? How hard can we press the saying for an answer?

As noted in the introduction, several recent commentators have interpreted the saying as mundane wisdom – advice on how to avoid going to prison – but commentators generally argue that it refers to divine punishment, as Matt 18.23–35 clearly does. In Matthew, ‘Amen I say to you’ always prefaces statements of great import pertaining to Jesus’ mission or the eschaton. The largest category of these sayings concerns eschatological recompense or the conditions for entering the kingdom.Footnote 64 The context in Matthew 5 suggests the same: verses 21–2 extend the prohibition of murder to include anger and insults, claiming that those who call another ‘fool’ are headed for Gehenna. Verses 23–4 warn that sins against others must be taken care of before approaching God in the temple. Finally, verses 25–6 move from a cultic to a legal setting, stressing the urgency of resolving sins before facing judgement. Pragmatic wisdom seems even less likely in the Didache and Luke. There was no law against accepting alms without need, so Did. 1.5 cannot be warning against ordinary imprisonment.Footnote 65 Luke's version appears in a series of sayings warning the crowds to prepare for divine judgement by repenting (13.1–5), remaining alert (12.35–56) and doing good deeds (12.13–34; 13.6–9).Footnote 66

As noted above, the most common interpretation of verse 26 is that those who fail to make friends with an opponent will be punished in Gehenna, or, in the parlance of modern Christian eschatology, ‘hell’, and commentators often stress that this punishment is everlasting.Footnote 67 Thus, for example, Robert Gundry writes that failure to make things right with an opponent ‘lands a person in hell, the prison of eternally hopeless debtors’.Footnote 68 Post-mortem punishment is often described as prison in Jewish, Christian and pagan texts, and the threat of ‘the Gehenna of fire’ is mentioned in verse 22, but the presumption that the saying threatens everlasting punishment is curious for two reasons.Footnote 69 First, though Matthew speaks of punishment that is αἰώνιος in other passages, here it does not say, settle with your opponent or be locked up forever. It says that release is impossible until the last penny is repaid, with ἕως plus the subjunctive ἀποδίδωμι apparently making release contingent upon payment.Footnote 70 As in documentary sources which make similar threats, the unhappy person is to be confined until the debt is taken care of. At the very least, it is not clear that eternal confinement is in view. This leads to a second observation: the emphasis in 5.21–6 is on relatively small sins – a rude name, or any little ‘something’ (verse 23) between two people. Even if one allows the tendency of the Synoptic tradition to exaggerate for emphasis, Matthew would be saying that any unresolved issue between two people sends the offender to hell forever. Matthew is known for placing great demands on those wishing to enter the kingdom (e.g. 5.20; 7.13–14), but commentators have not taken seriously their own conviction that Matthew says disciples go to eternal hell for failing to make friends with an opponent.

Given how many real-life debt-prisoners relied on others for ransom it is striking that few ancient interpreters link this passage to Jesus’ claim to give himself as a ransom for many (20.28). Presumably this is because all the Synoptic debt-prison sayings stress the personal responsibility of the sinner. Sometimes Matthew emphasises the free forgiveness of sins (e.g. 1.21; 18.10–14) and fantastically generous repayment of righteous deeds (e.g. 19.20; 20.1–16). On other occasions, however, perfection is demanded (5.48; 7.21; 18.3) and forgiveness of sins is conditional on certain behaviour.Footnote 71 Verse 26 is one of the sayings in Matthew that stress exacting, precise settling of accounts and the urgency of making things right.Footnote 72 If this verse is an image of temporary confinement and restitution, perhaps it is best understood as threatening temporary but painful correction if one fails to make friends with one's opponent before it is too late.

Does Matthew envisage post-mortem or eschatological repayment for sin followed by release? As we have seen, many ancient Christians took it this way. Kurt Niederwimmer argues that the Didache's version of the saying refers to ‘an eschatological place of punishment or purification’.Footnote 73 Belief in temporary post-mortem punishment is well attested in Second Temple and Tannaitic literature. Many texts speak of restorative chastisements – refining with fire, disciplining with whips and so on – and there was a range of views on whether and how far this might continue after death or in the eschaton.Footnote 74 Belief in eternal punishment for some did not preclude simultaneous belief in temporary punishment for others. According to the Tosefta, the House of Shammai taught that some people are punished forever, but others go to Gehenna temporarily to be healed of their iniquity as silver is purified in fire (יורדין לגיהנם ומצטפצפין ועולין הימנה ומתרפאין).Footnote 75 In the Mishnah Rabbi Akiva is said to have taught that the judgment of the wicked in Gehenna lasts only twelve months, whereas Rabbi Yohanan ben Nur said it was only from Passover until Pentecost.Footnote 76 Gehenna could be conceived as a place of purification, of eventual annihilation, of unending punishment, or all three at once.Footnote 77 Centuries earlier, 2 Maccabees portrayed prayers and sacrifices on behalf of those who had died while committing idolatry (12.38–45). In the Apocalypse of Zephaniah souls are imprisoned and tortured in Hades while the patriarchs intercede for them. Zephaniah asks whether post-mortem repentance is possible and is told yes, but only until the final judgement when it will be too late (esp. 10.1–11.6).Footnote 78 The Testament of Abraham depicts post-mortem trials coming to an end due to Abraham's intercession and God's mercy (esp. 14.1–15.).Footnote 79 In a passage with obvious similarities to Matthew 5, the Testament of Isaac describes the punishments of those who die before reconciling with a neighbour. They are tortured a year for every hour of enmity, but eventually they and those who suffer in the river of fire are released (5.1–32). Paul seems to have believed in the possibility of chastisements that lead ultimately to salvation in the day of the Lord, in one scenario distinguishing between apostolic workers who receive payment from God and those who are punished and then saved, as it were, through the flames (1 Cor 3.10–15).Footnote 80 Belief in post-mortem repentance or change of status did not go unchallenged. Pseudo-Philo, for instance, claims that repentance and intercessory prayer are impossible after death.Footnote 81

Belief in post-mortem or eschatological atonement for sins was not incompatible with an overriding apocalyptic dualism that placed the greatest emphasis on who is in and who is out. For Paul, the key question was whether one belonged to Christ or not, but he still expected those in the former category to face judgement according to deeds.Footnote 82 The Similitudes of Enoch focuses on the difference between the righteous and the sinners, but still allows that some who repent just squeak by, being saved without honour (50.1–4).Footnote 83 According to 2 Baruch, in the last day when there is no more chance to repent and when no intercession can help (85.12), God will destroy those polluted with sin, but will purge the sins of those bound for salvation (85.15). In Luke 12 Jesus says that erring servants will receive varying degrees of punishment, depending on their knowledge of the Lord's will (12.47–8) – that is, variegated punishments rather than simple in or out verdicts. All of these texts feature a prominent dichotomy between the saved and the damned, while maintaining that some of the saved might need additional purification or chastisement, or that some of the damned might deserve a more severe punishment.Footnote 84 There is nothing conceptually clumsy or surprising about Matthew, a text that draws a very sharp line between insiders and outsiders, also expecting everyone to give an account for every careless word (12.36), and to repay the last penny for every sin against a brother. Indeed, apart from theological prejudice there is no reason to reject out of hand the idea that Matthew, the Didache or Luke would speak of eschatological repayment for sin, a reckoning roughly analogous to the one described in 1 Cor 3.10–15. The context in Matthew 5 does suggest divine retribution of some kind; release from prison would mean that the punishment is temporary. This would then mean that Matthew's Jesus demands perfection, but does not preach everlasting torment for a single unresolved issue.

Is there any direct evidence for eschatological or post-mortem atonement in Matthew apart from the passage under discussion? Davies and Allison, among others, argue that the coming baptism of fire promised by John (Matt 3.11/Luke 3.16) was supposed to purify the elect as it destroyed the wicked, a view already held by some ancient Christians.Footnote 85 In the saying on blasphemy against the Holy Spirit (12.31–2), Mark and Luke say it will not be forgiven (Mark 3.28–9; Luke 12.10), but Matthew specifies that this particular sin will not be forgiven in this age or in the age to come (οὐκ ἀφεθήσεται αὐτῷ οὔτε ἐν τούτῳ τῷ αἰῶνι οὔτε ἐν τῷ μέλλοντι). For some ancient Christians, this implied that in the age to come other sins will be forgiven.Footnote 86 If modern commentators mention this interpretation at all it is only to scoff, but it should not be dismissed so lightly.Footnote 87 There were disagreements in ancient Judaism and Christianity on the extent of atonement after the present life. Some denied it was possible. Others accepted it, but only up to a point, and indeed some rabbis distinguished between sin that could be forgiven after death and sin that could not.Footnote 88 Matthew adapts the Markan tradition to specify that one particularly heinous offense will not be forgiven ‘in this age nor in the age to come’. It is perfectly reasonable to suppose other sins might be.

That said, these passages do not provide clear proof that Matthew expected post-mortem or eschatological atonement. As for Matt 5.25–6 and its parallels, the precise nature of the prison and the required payment in view are not clear enough in this suggestive, parabolic little saying to make any firm conclusions, and this lack of clarity is reflected in the range of repayment scenarios proposed by early Christian interpreters. As John Henry Newman argued in his essay on development, it is only when interpreted in light of other biblical promises of purifying fire and in light of the fairly universal Christian practice of prayer for the dead that these ‘obscure and indeterminate’ words came to speak unambiguously of purgative suffering on the path to beatitude.Footnote 89

Conclusion

Matt 5.25–6 and its parallels should not be glossed as ‘you will go to hell forever.’ This is not what the words say. This is not how prison in antiquity worked. This is not how anyone read the passage until the Origenist controversy required Augustine and others to oppose any hint that the fires of perdition could be temporary. Interpreters agreed on only one thing for the first 300 years: the debtor eventually goes free. Instead, the saying is suggestive of a penitential principle. For Matthew, sin against another cannot be brushed aside. It should be dealt with quickly. If not, a time of painful repayment will follow. Of all the interpretations surveyed here, perhaps Cyprian's reading in defence of penance is the closest to the sense in the gospels. For him, the dominical warning showed the seriousness of sin, but – contrary to the rigorists he opposed – it also assumed there could be eventual restoration.

Footnotes

I am grateful for comments on draft versions of this article offered by members of the Oxford New Testament Seminar, the Cambridge Senior New Testament Seminar, and the Durham University New Testament Seminar. Thanks also to Michal Bar-Asher Siegal and T. J. Lang.

References

1 Translations are my own unless noted otherwise.

2 In favour of mundane wisdom, see e.g. Fitzmyer, J., The Gospel according to Luke x–xxiv: Introduction, Translation, and Commentary (AB 28; Garden City, NY: Doubleday, 1985) 1001–2Google Scholar; Frey, J., ‘The Character and Background of Matt 5.25–26: On the Value of Qumran Literature in New Testament Interpretation’, The Sermon on the Mount and its Jewish Setting (ed. Becker, H.-J. and Ruzer, S.; Paris: Gabalda, 2005) 339 Google Scholar. Among many arguing for divine judgement, see Davies, W. D. and Allison, D. C., A Critical and Exegetical Commentary on the Gospel according to Saint Matthew (3 vols.; ICC; Edinburgh: T&T Clark, 1988–97) i.521Google Scholar; Luz, U., Matthew 1–7 (Minneapolis: Fortress, 2007) 241 Google Scholar; Reiser, M., Die Gerichtspredigt Jesu: Eine Untersuchung zur eschatologischen Verkündigung Jesu und ihrem frühjüdischen Hintergrund (Münster: Aschendorff, 1990) 271 Google Scholar.

3 The possibility of temporary confinement is briefly discussed in Eubank, N., Wages of Cross-bearing and Debt of Sin (BZNW 196; Berlin/Boston: de Gruyter, 2013) 5763 Google Scholar; Kinman, B. (‘Debtor's Prison and the Future of Israel (Luke 12:57–59)’, JETS 42 (1999) 411–25)Google Scholar argues that the Lukan parallel hints at the eventual restoration of Israel.

4 It is often assumed that Matt 5.25–6 and Luke 12.58–9 describe unpaid debts, but it is possible that another sort of legal dispute could lead to this end. In any case, the prisoner must pay to go free.

5 Particularly judicious is Krause, J., Gefängnisse im Römischen Reich (Stuttgart: F. Steiner, 1996), esp. 152–69Google Scholar. On Roman philosophies of incarceration, see Hillner, J., Prison, Punishment and Penance in Late Antiquity (Cambridge: Cambridge University, 2015)Google Scholar. See also Mitteis, L., Reichsrecht und Volksrecht in den östlichen Provinzen des römischen Kaiserreichs: mit Beiträgen zur Kenntniss des griechischen Rechts und der spätrömischen Rechtsentwicklung (Leipzig: Teubner, 1891) esp. 444–58Google Scholar; de Franch, R. Sugranyes, Études sur le droit palestinien à l’époque évangélique: la contrainte par corps (Fribourg: Librairie de l'Université, 1946)Google Scholar; Taubenschlag, R., The Law of Greco-Roman Egypt in the Light of the Papyri 332 bc640 ad (Warszawa: Państwowe Wydawnictwo Naukowe, 1955)Google Scholar; Llewelyn, S. R., New Documents Illustrating Early Christianity: A Review of the Greek Inscriptions and Papyri Published in 1982–83, vol. vii (North Ryde, NSW: Ancient History Documentary Research Centre, Macquarie University, 1994) 197224 Google Scholar.

6 Exod 21.2; Lev 25.39–54; Deut 15.12–15. Cf. Jer 34.14.

7 On the former, see Livy, History of Rome 8; Cicero, Rep. 2.59; on the latter, the cessio bonorum, see Woeß, F., ‘Personalexekution und cessio bonorum im römischen Reichsrecht’, ZSS 43 (1922) 485529 Google Scholar. According to Cod. Justin. 7.71.4 cessio bonorum was extended to the provinces.

8 OGIS 669.15–18.

9 Indeed, some argue that debt was the most common cause of incarceration. Bauschatz, J. (‘Ptolemaic Prisons Reconsidered’, The Classical Bulletin 83 (2007) 347)Google Scholar summarises the situation beyond Egypt as follows: ‘For the most part, prisons were temporary holding cells in which offenders, and these mainly debtors, endured brief stays’ (5).

10 So for example, out of many, P.Mich. iii.191; P.Mert. i.14; P.Yadin i.21. See also the praxis-language in the aforementioned edict (OGIS 669.16). Origen (Or. 29.15) and Eusebius (Dem. ev. 9.8) describe the extraction of funds in Matt 5.26/Luke 12.59 as ἔκπραξις.

11 See the discussion in Llewelyn, New Documents Illustrating Early Christianity, 222.

12 E.g. Cicero (Verr. 2.10.24), who describes incarceration as being thrown in vincla atque in tenebras. See the description of prison conditions in Wansink, C. S., Chained in Christ: The Experience and Rhetoric of Paul's Imprisonments (JSNTSup 130; Sheffield: Sheffield Academic, 1996) 4495 Google Scholar.

13 E.g. P.Amh. ii.77; Acts 16.23–24. Wansink (Chained in Christ, 46) notes that chains were so common they functioned as a synecdoche for prison like the modern expression ‘behind bars’.

14 Seneca (De ira 32.2–3) laments those who suffer vinculis, carcere, fame for peccadillos, usually involving money. Cf. Libanius, Or. 45.9; Pliny, HN 7.36.121; Valerius Maximus 5.7 Prison-fare, where it existed, is routinely described as very poor. E.g. Dio Chrysostom, Charid. 13.

15 Krause, Gefängnisse im Römischen Reich, 160.

16 Bauschatz, ‘Ptolemaic Prisons Reconsidered’, 4–5. On the coercive effects of late ancient debt-prison see Hillner, Prison, Punishment and Penance in Late Antiquity, 145–6. Hillner notes that creditors sometimes imprisoned debtors’ dependents to ensure payment (146).

17 Flac. 46–9.

18 P.Hib. i.34, 73.

19 P.Tebt. ii.420.

20 Tim. 49.

21 Fid. 6; see also Plutarch, Cim. 4.3.

22 Plutarch, Flam. 12; Cicero, De or. 2.63 (in reference to a play).

23 Ben. 3.8.

24 P.Oxy. ii.259. Cf. the ὑπηρέτης in Matt 5.25. The dramatis personae of Matt 5.25–6 and Luke 12.57–9 are well attested in documentary sources.

25 For more examples of standing surety from the papyri see Taubenschlag, The Law of Greco-Roman Egypt, 539.

26 Luz, Matthew 1–7, 241. See also Jeremias, J., The Parables of Jesus (London: SCM, 1963) 180 Google Scholar, who traces this meaning back to Jesus.

27 P.Yadin i.21; P.Hever 65; Mekilta de-Rabbi Ishmael on Exod 21.1–3; m. ʿEd. 8.2. See Goodman, M., ‘The First Jewish Revolt: Social Conflict and the Problem of Debt’, JJS 33 (1982) 417–27CrossRefGoogle Scholar.

28 B.J. 2.427. Regardless of the accuracy of Josephus’ reporting here, his perception was that fear of the collection of debts was a major concern for the poor of Jerusalem.

29 Irenaeus, Haer. 1.25.4; Tertullian, An. 35.1; Epiphanius, Pan. 27.5.

30 Book iii.113. See also Sent. Sextus 39; Testim. Truth (NHC ix 3.29.30). cf. Plato, Phaed. 107c–108c; Ap. John (NHC ii 1.27); Apoc. Paul (NHC v 2.21.17).

31 An. 35.

32 Ibid. See also An. 58; Res. 42; Or. 7.1.

33 See the discussion in Daley, B. E., The Hope of the Early Church: A Handbook of Patristic Eschatology (Cambridge: Cambridge University Press, 1991) 37 Google Scholar.

34 Strom. 4.14.

35 Ibid., citing Rom 8.38–9.

36 Test. 3.

37 Test. 3.57. It is sometimes suggested that this work was compiled prior to Cyprian, but see Murphy, E., ‘“As Far as My Poor Memory Suggested”: Cyprian's Compilation of Ad Quirinum ’, VC 68 (2014) 533–50CrossRefGoogle Scholar.

38 Ep. 55.20.

39 According to Jay, P. (‘Saint Cyprien et la doctrine du purgatoire’, RTAM 27 (1960) 133–6)Google Scholar, Cyprian speaks here of this-worldly penance, not purgatory.

40 Comm. Matt. 14.8.

41 Ibid.

42 Hom. Luc. 35.14 (according to Jerome).

43 Hom. Luc. 35.15 (according to Jerome).

44 Or. 29.15. It would be far worse, Origen avers, to be delivered over to one's passions as in Rom 1.24 (ibid.).

45 Comm. Rom. 5.1.39 (according to Rufinus). Cf. Or. 29.15. Origen also describes this period as Christ cleansing his people with fire before they enter paradise, depending on which biblical text is in view. See Daley, The Hope of the Early Church, 57.

46 Exp. Luc. 7.1626–39.

47 Exp. Luc. 7.1697–1709.

48 Tractatus super psalmos 68.9. On other occasions Hilary speaks of the possibility of sinners paying their own debts (Commentarius in Matthaeum, 4.19; see also Tractatus super psalmos, 118.5).

49 Epist. 4.2. Like Hilary, Jerome also speaks of sinners paying their own debt. See Epist. 130.7, and also the criticism of various interpretations in Comm. Matt. 1.562–96.

50 De anima 15.75–6. Gregory is citing his sister Macrina.

51 Though the treatment of the evidence reflects a clear Tendenz, see the learned study of Ramelli, I. L. E., The Christian Doctrine of Apokatastasis: A Critical Assessment from the New Testament to Eriugena (Leiden: Brill, 2013), esp. 506–41Google Scholar.

52 Tractatus in Mathaeum 11.103–10. Unlike Gregory, Chromatius stops short of affirming apokatastasis here, as the following comments on Matt 3.11 make clear (11.111–18.).

53 See Clark, E. A., The Origenist Controversy: The Cultural Construction of an Early Christian Debate (Princeton: Princeton University Press, 1992)Google Scholar.

54 See e.g. Civ. 21.17; Enchir. 29.110. On Augustine's development on this issue, see Moreira, Isabel, Heaven's Purge: Purgatory in Late Antiquity (New York/Oxford: Oxford University Press, 2010) 33–6CrossRefGoogle Scholar. See also Daley, The Hope of the Early Church, 138–41.

55 Serm. Dom. 1.30.

56 Dulc. 1.14. This was in response to the tribune Dulcitius’ question about the ‘diverse opinions’ on the issue of whether suffering in Gehenna is everlasting. Dulcitius’ own opinion was that Matt 5.25 and 1 Cor 3.15 suggest it is not (1.1). Cf. Quaest. ev. 2.38 (ca. 400).

57 Others continued to comment on the passage without weighing in on the controversy directly. John Chrysostom (Hom. Matt. PG 57.252–3) rejected ‘allegorical’ interpretations in favour of an earthly prison. Eusebius (Dem. ev. 9.8) links Matt 5.26 to Isa 3.12 and Matt 4.12–25 to ambiguous effect. After affirming eternal hell for some, Caesarius of Arles (ca. 468–542) says that Matt 5.26 refers to the temporary punishment experienced by some who pass through fire en route to salvation (Sermo 167.5–7).

58 Quaestiones et responsiones ad coenobitas 607.23–6.

59 Quaestiones et responsiones ad coenobitas 607.162–8.

60 E.g Bede, In Lucae evangelium expositio 4.12.

61 Super evangelium S. Matthaei lectura, lectio 5.

62 Included in the Glossa ordinaria on Matthew 5.

63 Cf. other Matthean ‘punchlines’: 13.49–50; 18.14, 35; 20.16; 21.31–2; 21.43; 22.14; 24.44, 47, 51; 25.13, 29–30, 46.

64 Matt 6.2, 5, 16; 10.15, 42; 18.3; 19.23, 28; 21.31; 24.46; 25.12, 40, 45.

65 Church discipline is theoretically possible, but ecclesial prisons and torture are otherwise unattested in this period, and it would be difficult, if not impossible, to determine if alms had been accepted unnecessarily. Cf. Herm. Mand. 2.4–7 (27.4–7).

66 Less clear is whether the moment of judgement for Luke refers to Jesus’ entry in Jerusalem (so Johnson, L. T., The Gospel of Luke (Sacra Pagina; Collegeville, MN: Liturgical Press, 1991) 209)Google Scholar or a final judgement.

67 E.g. Keener, C., Commentary on the Gospel of Matthew (Grand Rapids: Eerdmans, 1999) 185 Google Scholar; Gundry, R. H., Matthew: A Commentary on his Handbook for a Mixed Church under Persecution (Grand Rapids: Eerdmans, 1994 2) 87 Google Scholar; Davies and Allison, Matthew, i.521.

68 Matthew, 87.

69 On other-worldly punishment as prison, see e.g. 1 En. 10.13; Josephus, A.J. 18.14; Plato, Gorg. 523b; Seneca, Herc. Ot. 1005; 1 Pet 3.19; Herm. Sim. 9.28 (105).

70 Smyth, H. W., Greek Grammar (rev. Messing, G. M.; Cambridge, MA: Harvard University Press, 1956) §2429Google Scholar.

71 6.12 and 18.23–35 tie forgiveness to the sinner's forgiveness of others. Cf. 7.1–2.

72 Cf. judgement as account-settling: 16.27; 18.23–35; 20.1–6; 19.27–20.16; 25.14–30.

73 The Didache: A Commentary (Minneapolis: Fortress, 1998) 83 Google Scholar. Cf. Did. 16.5.

74 E.g. Prov 3.11–12; Zech 13.9; Mal. 3.1–3; Jdt 8.27; 1QHa xiii.15–19; Heb 12.6; Rev 3.19; 2 Bar. 78.6; Pss. Sol. 13.6–11.

75 t. Sanh. 13.3. See the following quotation of Zech 13.9 and 1 Sam 2.6. The House of Hillel counters that God inclines to mercy. Cf. b. Roš Haš. 16b.

76 m. ʿEd. 2.10. cf. t. Sanh. 13.4; Pesiq. Rab. Kah. 10.4.

77 E.g. t. Sanh. 13.3–5.

78 Cf. b. ʿErub. 19a.

79 For the intercession of patriarchs or other notables, see e.g. Philo, Praem. 166; T. Isaac 6.9–23; b. ʿErub. 19a; b. Sotah 10b.

80 See also 1 Cor 11.27–31, possibly 5.5. For action taken on behalf of the dead, see 1 Cor 15.29. On 1 Cor 3.10–15, see Frayer-Griggs, D., ‘Neither Proof Text nor Proverb: The Instrumental Sense of διά and the Soteriological Function of Fire in 1 Corinthians 3.15’, NTS 59 (2013) 517–34Google Scholar; Kirk, A. N., ‘Building with the Corinthians: Human Persons as the Building Materials of 1 Corinthians 3.12 and the “Work” of 3.13–15’, NTS 58 (2012) 549–70Google Scholar.

81 LAB 33.3–5. See also 4 Ezra 7.82, 102–115; 9.12; 2 En. 53.1; possibly Heb 9.27.

82 Rom 2.16; 14.10; 1 Cor 3.10–15; 4.5; 2 Cor 5.10.

83 Herms, R., ‘“Being Saved without Honor”: A Conceptual Link between 1 Corinthians 3 and 1 Enoch 50?’, JSNT 29 (2006) 187210 Google Scholar.

84 Cf. C. Hayes on rabbinic descriptions of Gentiles, ‘The Complicated Goy in Classical Rabbinic Sources’, Perceiving the Other: Ancient Interactions with Others in Antiquity and Modern Scholarship (WUNT i; Tübingen: Mohr Siebeck, forthcoming).

85 Davies and Allison, Matthew, i.317; Meier, J. P., Matthew (Collegeville, MN: Liturgical Press, 1980) 25 Google Scholar; Origen, Hom. Luc. 24; 26.3. Cf. Mark 9.49.

86 Augustine, De civ. Dei 21.24. Gregory the Great, Dial. 4.41. Most ancient commentary on the saying focuses on the status of the Spirit relative to God.

87 E.g. Luz, U., Matthew 8–20 (Minneapolis: Fortress, 2001) 208 Google Scholar.

88 This could be a matter of the quantity of sin (t. Sanh. 13.1–3) or of particularly egregious sins (m. Sanh. 10; t. Sanh. 13.4–12; b. ʿErub. 19a). Cf. 1 John 5.14–17.

89 An Essay on the Development of Christian Doctrine (London: James Toovey, 1845) 420–1Google Scholar.