A Critical an Exegetical Commentary on the Gospel according to Saint Matthew, Volume II, pp. 623-641
18. Jesus continues, giving Simon a new name (Peter, the rock) and speaking for the first time of ‘my church’, which will be built upon the rock against which the gates of Hades will not prevail. The verse is among the most controversial in all of Scripture. The literature it has generated is immense, and not a little of it rather polemical. For a helpful survey of the history of the discussion see Burgess (v).
There is an OT background to v. 18 which only a few scholars seem to have appreciated.” Of the OT figures to receive a second name, the most memorable are Abram and Jacob. The former was given the new name Abraham to signify that he would be the father of a multitude (Gen 17.1—8). The latter was renamed Israel, by which the people that would spring from him would be known (Gen 32.22—32). We find particularly intriguing the parallels between Gen 17 and Mt 16. In both cases we are witnessing the birth of the people of God (the Jews in the one case, the church in the other). In both that birth is associated with one particular individual (Abraham, then Peter), and in both that individual has a name which symbolizes his crucial function (Abraham is taken to mean ‘father of a multitude’, Peter to mean the ‘rock’ on which the church is founded). Are we to conclude that the Peter of Mt 16 is the parallel to Abraham?
Strongly buttressing this inference is Isa 51.1—2: ‘Hearken to me, you who pursue deliverance, you who seek the Lord; look to the rock from which you were hewn, and to the quarry from which you were digged. Look to Abraham your father and to Sarah who bore you; for when he was but one I called him and I blessed him and made him many’: cf. LAB 23.4. Here Abraham, like Peter, is a rock (the MT has .sfir, the LXX nétpav). So just as the OT figure whose name was changed in order to signify the coming into being of the people of God was likened to a rock, so too in Matthew is the birth of the church accompanied by Simon gaining the new name, l'létpog. In this we detect design.
It is crucial in this connexion to note that John the Baptist, in a saying preserved by Matthew, took up Isa 51.1—2 in order to issue a warning to the Jews: ‘Do not presume to say to yourselves, “We have Abraham as our father”; for I tell you, God is able from these stones to raise up children to Abraham’ (3.9, q.v.). In this saying the Baptist holds forth the possibility that God can bring forth a new people, not from the old rock or quarry (Abraham and Sarah) but from elsewhere. What we seem to have in Mt 16.17—19 is the realization of John’s dire prophecy. Here the new people of God is brought into being, hewed not from the rock Abraham but instead founded on the rock Peter.70
Compare Jalqut Shimon, I § 766, on Num 23.9 (in SB 1, p. 733): when God ‘saw Abraham who was going to arise, He said, Behold, I have found a rock (pith?) upon which to build and establish the world. Therefore he named Abraham 3 rock (surf (cf. Isa 51.1—2). We suspect Christian influence because of the lateness of the text and the use of the loanword pz‘grd’. But if this supposition is correct, that only goes to show how the roles of Peter and Abraham lent themselves to comparison.
miyd) fié not . These words do not require that v. 18 was first attached by Matthew to v. 17, for our evangelist never uses Kayo to string together previously unconnected sayings.71 On the combination Kai . . . 5e see BDF § 447.9. an on) at Hétpog, Kai éni room If] nérpq. oixofiom’loo) you n‘lv éxxknoiav.” Compare 7.24; Eph 2.20; lTim 3.15; Rev 21.14; 0d. Sol. 22.12.
‘You are Peter’ matches ‘you are the Christ’ (of) st in both cases). And just as Peter spoke revelation, so now does Jesus. But does this mean that Peter only now gets his new name? This is the judgement of perhaps most commentators. But Augustine thought otherwise (De con. ev. 2.53(109)). Gundry, Commentary, p. 335, concurs: 16.17 ‘probably implies a play on the name as already given rather than a giving of the name at this time. For the latter, we should have expected the future tense, “You will be called”; and repeated earlier references to Simon as “Peter”, most of them Matthew’s insertions, take away the point of a name-giving here. Otherwise we would have to suppose, doubtfully, that Matthew expects his readers to take all the earlier references as anachronisms’. We are not convinced. Gundry’s comment on the future tense is scarcely decisive. Indeed, does not the present tense (of) at) point to an event taking place as Jesus speaks? Such an interpretation would not, pace Gundry, involve an anachronism. For while Simon is called Peter before 16.17, he is only called this by the narrator (cf. the situation with regard to ‘Christ’!) No one else in the story uses ‘Peter’ until 16.17. Furthermore, the first notice of Simon is followed by this: ‘who is called Peter’ (4.18). The note does not indicate when or why Simon received this name (cf. 10.4). Certainly he does not receive it in 4.18. It is most natural to think that 4.18 looks forward to 16.17, that 16.17 explains a fact 4.18 mentions but leaves unexplained. We are confirmed in this by the parallels between Mt 16.17—19 and Gen 17.1—8 (see above). These gain their full force only when one thinks of 16.17—19 as recounting the giving of a new name.
As we know from In 1.42 and the Pauline epistles,"3 behind Ilérpog (and also, probably, nétpa) lies the Aramaic kephd’, a word which is usually said to mean ‘rock’.7“ But it has been argued that the first meaning of the Aramic kephd’ was ‘stone’ (cf. M609 and the first meaning of nérpog). Lampe (v), with this as his starting point, claims that when Jesus gave Simon the nickname Kephd’, he could not have been thinking of a rock foundation. In his estimation, Jesus gave Peter the nickname ‘stone’ for reasons we can only guess, but which were probably not theological or religious. It was only in the postEaster period, and only after Kephfi’ had already been translated by Ilétpog, that a wordplay was conceived with nétpa, and the apostle made out to be the foundation of the church. All this falls far short of proof. For one thing, Lampe (who wrongly asserts that kephd’ was not a personal name in pre-Christian times; see 1, p. 396), has evidently failed to consider the possibility that Jesus on one occasion gave Simon the name Kephd’ and then, on a later occasion, used that name (with perhaps slightly different sense) to make a theological statement.7 5 Even if the first meaning of kephd’ was ‘stone’, it also, as the Dead Sea Scrolls and the targumim show, meant ‘rock’ (cf. Fitzmyer (v), p. 115); and this fact could have allowed Jesus to play upon kephd’ in the manner Mt 16.18 recounts. No less importantly, Lampe (v) seems to assume that it would be awkward to speak of building upon a stone instead of a rock; but the image behind Mt 16.18 is of a temple being constructed, and in Judaism the temple was founded not upon a rock but upon a (foundation) stone, the ’eben ie'tiyyd (cf. the use of ’eben/ MOOV in Isa 28.16)
As we have just observed, Simon may, notwithstanding Matthew’s point of view, have been given his nickname for purely secular purposes. This would line up with what we find in rabbinic sources, where certain teachers give their students nicknames that reflect character or appearance.“5 But it is no less probable that the name Kephd’ was given to make a theological point (cf. Mt 16.18). In Jewish tradition the giving of a new name could mark an important event or a change in someone’s status. Examples include the renaming of Abraham (Gen 17.5), Sarah (Gen 17.15), Jacob (Gen 32.28), Jerusalem (Isa 62.2—4; cf. Zech 8.3; Bar 5.4), the conquering saints (Rev 2.17; cf. Isa 65.15), and Aseneth (Jos. Asen. 15.7). Also not to be forgotten are the OT texts in which a prophet bestows a name in order to register publicly a theological conviction. See, for example, Isa 7.3; 8.34; Jer 20.1—6; and Hos 1.4—9. Prophets used names in parabolic fashion (note also Isa 7.14; 9.6; Jer 23.6; Zech 6.12). This matters because Jesus conceived of himself as, among other things, a prophet, and he not only uttered parables but performed parabolic acts in the prophetic vein.” One must wonder whether the naming of Peter or the symbolic application of his name should not therefore be considered a prophetic act. (It is also possible that ‘sons of Thunder’, the name given to the sons of Zebedee, originally had some theological or prophetic significance now lost. This is made more likely by the fact that Peter and James and John, the only three to have received nicknames as far as we know, appear to have formed an inner circle during the earthly ministry.)
Kai éni mom If] firm has been the object of much heated debate and much wasted ingenuity. ‘This rock’ has been identified variously with Peter’s faith or confession,78 with Peter’s preaching office,79 with the truth revealed to Peter,80 with the twelve apostles,81 with Jesus,82 with Jesus’ teaching,83 and even with God himself.84 All this is special pleading. The most natural interpretation is that of Roman Catholic tradition: the rock is Peter. ‘The word-play, and the whole structure of the passage, demands that this verse is every bit as much Jesus’ declaration about Peter as v. 16 was Peter’s declaration about Jesus’ (France, Matthew, p.254). The only sound cause for objecting to this is that one might have expected Hérpog nérpcp instead of l'Is’rpoc; nérpq. Why two different Greek words, one masculine, one feminine? An explanation probably lies in this, that kephd’, the Aramaic presumably behind both Hétpog and nérpqr, ‘was used with different nuances. When translated into Greek, the masculine form petros would lend itself as a more likely designation of a person (Simon), and a literary variant, the feminine petra, for an aspect of him that was to be played upon.’85
It is possible that there is a mythological background to the imagery of the rock. ‘My church’, interpreted in the light of 28am 7 (see p. 603), evokes the idea of a temple, and the conception of the people of God as a temple was well known in both Judaism and early Christianity.86 This is important because in Jewish tradition the rock at the base of the temple on Zion, the so-called ’eben s'étiyyfi, is at the centre of the world.”7 It links heaven and the underworld, being the gate to the former as well as the portal to Hades, the realm of the dead. Note that in 16.18c mention is made of the ‘gates of Hades’. Perhaps, then, the informed reader should imagine the church at the centre of the cosmos, sitting on top of the powers of evil.
oixoéouéw, associated elsewhere in the NT with éxxknoia only in lCor 14.4,88 is used again in both 26.61 (from Mk 14.58) and 27.40 (from Mk 15.29), where people claim that Jesus threatened to tear down the temple (vaég) and then rebuild it or raise it up (oixofiouéw). What is the relationship between the three verses? The key is the evangelist’s treatment of Mk 14.58, from which he has omitted ‘not made with hands’. This last is usually, and rightly, taken to refer to the Christian community.89 Why its omission by Matthew? One can scarcely maintain that he rejected the conception of the church as a temple (it is implicit in 16.18). Rather, Mk 14.58, interpreted of the church, might be thought to put its founding in the postEaster period (‘after three days’), which would create tension with Mt 16.18, which places the church’s birth before the resurrection.90 The difficulty was solved by modifying Mk 14.58 so that one thinks more readily of Jesus alone being raised up (cf. Jn 2.19—22) instead of the Christian community. (Perhaps Matthew felt justified in his editing by the circumstance that Mk 14.58 is attributed to false witnesses and could therefore reasonably be held to be a misunderstanding or misrepresentation of something Jesus said.)
For other texts in which a community or group of people is spoken of as being built see Jer 12.16; 18.9; 31.4 (‘I will build you, and you shall be built, 0 virgin Israel’); 33.7; 42.10; Amos 9.11 (‘In that day I will raise up the booth of David that is fallen and repair its breaches, and raise up its ruins, and rebuild it as in the days of old’); 4QpPs“ frags. 1—10, col. 3.15—16 (God chose the teacher of righteousness as a pillar and ‘established him to build for him a congregation . . .’); 4Qplsad (=4Q164); 1 Cor 3.9-15; Eph 2.19—20; SB 1, pp. 723-3. Such usage was natural given the likening ofthe people ofGod to both temple9 ‘and house (cf. the phrase, ‘house of Israel)"2
arachnoid” occurs only twice in the canonical gospels, here and in 18.18. It is disputed whether Jesus could have used some Semitic equivalent (on this see pp. 612—4). As the word is used in Matthew it has two slightly different meanings. In 16.18 the universal church is in View, in 18.18 the local assembly (cf. the apparent fluctuation in Acts and the Pauline epistles).94 In the LXX, from Deuteronomy on, éxxkncia is the equivalent of qdhdl, which means according to BDB, s.v., ‘assembly, convocation, congregation’. It never stands for ‘e'dd, which is usually rendered by cuvaymyi’]. It is doubted whether the background to early Christian usage is to be found in the LXX, and it must be conceded that the NT uses of éxxkncia sometimes lack LXX parallels. Still, most scholars would concur with Bultmann that ‘in content “Church (of God)” éxxknoia (rofi @805) corresponds with t (YHWH)’, and ‘in understanding themselves as Congregation or Church the disciples appropriate to themselves the title of the Old Testament Congregation of God, the qe'hal—YHWH’ (Theology 1, p. 38).95 In line with this, the early Christians— like the Qumran convenantors, who spoke of their community as the qéhal (’El)96—thought of themselves as a counterpart to the Sinai-Congregation (cf. 1Cor 10.1—5; Heb 12.18—24), which in Deuteronomy is called the éxxkncia (e.g. 4.10; 9.10; 18.16; 31.30; cf. Philo, Ebr. 213; Decal. 32). That Matthew himself perceived the connexion we need not doubt. The typology that runs through chapters 1—5 (see on 5.1—2) and the numerous parallels between Jesus and Moses entail that the church had its origin in a new exodus.”
Despite the previous paragraph, it should be observed that in Matthew the church is not characterized as being 106 9806. Jesus specks rather of ‘my church’. For Matthew the community belongs to God through Jesus.
where the cornerstone laid in Zion will withstand the assault of water while those who have made a covenant with Sheol and death will be swept away) and lQH 6.19—31 (here the speaker has journeyed to the gates of death but finds refuge in a city founded on a rock). The notion that the underworld or the realm of the dead was locked by gates was common in the Ancient Near East and appears already in the Epic of Gilgamesh. Note Homer, 0d. 14.156, and Diogenes Laertius 8.34—5 (quoting Aristotle).99
The spectrum of opinion on these words, which in the early church were so often used against heretics, and which later came to serve as an apology for tradition,100 is unusually broad. Among the various proposals we note the following:
(i) Harnack (v), holding ‘on this rock I will build my church’ to be an interpolation (see 11. 72), argued that the original text had 000 instead of afnfig and thus took the whole line to be a promise of immortality to Peter. In favour of this, such an interpretation makes for a good connexion with 16.28 (‘there are some standing here who will not taste death . . .’),101 and it would go some way towards explaining why Mark or Mark’s tradition dropped 1617—19: the prophecy had been shown false by Peter’s death (cf. Otto (v)); There is also the fact that Jn 21.20—3 reflects the conviction that one or more of Jesus’ disciples would live to the consummation (see the commentaries). There is, however, no textual justification for the suggested emendation, and one does not wish to ‘correct’ the text unless it is otherwise unintelligible.102
(ii) B. P. Robinson (v), although he does not accept Harnack’s textual reconstruction, follows Harnack in finding a promise of immortality: afitfig refers back to Peter (mimj nj nérpq), not the church. That the apostle was dead when Matthew composed his gospel is no objection to this interpretation but rather proof that Matthew thought in terms of successors to Peter. The promise, in other words, is not spoken to Peter as such but to Peter as holder of a particular office.”3
(iii) Bousset, p. 65, associated 16.18 with the tradition of Christ’s descent into hell: ‘The company of the righteous who have fallen asleep also belongs to the ecclesia triumphans. The gates of Hades are opened and they no longer hinder passage to freedom’.104 One must agree with Bultmann, History, p. 139, n. 2: ‘apart from the fact that the descent into Hades is nowhere hinted at in the text, it is not possible to see any connexion between the bursting of the gates of Hades by Christ’s descent into hell, and the phrase of) Kanoxficovotv: the Church had not been imprisoned in Hades!’
(iv) McNeile, in his commentary, ad loc., cites the first passion prediction (16.21) as well as Acts 2.24 and 31 and goes on to find an allusion to Jesus’ resurrection: ‘The ecclesia is built upon the Messiahship of her Master, and death, the gates of Hades, will not prevail against her by keeping Him imprisoned’. C. Brown (v) puts forward a similar View, stressing that Mt 16.19 is in truth a passion prediction. We wonder, however, whether the meaning can be so far genealt? the surface of the text, which does not directly speak of Jesus imse .
(V) Schlatter, pp. 509—10, contends that the general resurrection is in view. The gates which, in the past, let people in and never out will, in the end, not hold those Christ has saved (cf. Ep. Apost. 28). The main difficulty with this interpretation is that it does not do justice to Kaucxéoouow, which most naturally connotes an advancing force (see below).
(vi) Gero (v), following the textual tradition of certain Syriac mss., urges that our line was originally about the ‘bars’, that is, ‘levers’ of Hades. The verse simply promises that Peter, the rock, will not be dislodged from his place, even by the powers of Hades. The church’s foundation is immovable. The consensus of the Greek textual tradition, however, is not to be disregarded unless there are truly compelling reasons. This is why we also cannot follow Kobert (v). He accepts the Syriac evidence but equates mt‘tklé with prison bars.
(vii) Another suggested emendation has come from A. Pallis1°5 and Eppel (v). They believe ‘gatekeepers’ (s'6‘dré, mmcopoi; cf. LXX Job 38.17b; 2 En. 42.1; b. Hag. 15b) should be read instead of ‘gates’.
(viii) According to Gundry, Commentary, p. 335, ‘the gates of Hades’ particularly represent ‘death by martyrdom’. The promise is that the church will not be obliterated by persecution. But the equation of ‘the gates of Hades’ with ‘death by martyrdom’ is, pace Gundry, hardly established by the prominence of persecution in the First Gospel. (ix) According to L. E. Sullivan (v), our text pictures the church on the attack, reaching into Hades to draw up its members. Is it, however, natural to see the church—which in this context is depicted as a building—as the aggressor?
(x) In Allen’s words, 16.18c could mean that ‘the organised powers of evil shall not prevail against the organised society which represents My teaching’ (Matthew p. 176). This accords with the dominant trend in Roman Catholic interpretation and makes for a good connexion with v. 19, which can be taken to concern Peter’s teaching authority.
(xi) Perhaps most contemporary expositors would concur with Schweizer, Matthew, p. 342: 16.18 simply ‘states unequivocally that death with all its power cannot put an end to the Christian community’. The church will endure until the end of the world (cf. SB 1, p. 736). This interpretation has in its favour the OT equation of ‘gates of Sheol’ with ‘gates of death’ (the older expression)?“5 and it can appeal to Prov 1.12 and Isa 5.14, where the image of Hades swallowing victims is an image of death: this is rather close to Mt 16.18’s depiction of an active Hades. (Most exegetes holding this View remark that ‘gates’ is to be understood as an example of synecdoche, the part (the gates) standing for the whole (Hades). As to why then ‘the gates of Hades’ is used instead of ‘the powers of Hades’ or just ‘Hades’, the answer lies in v. 19, where the mention of ‘keys’ inevitably conjures up the image of gates or doors. In other words, both W. 18 and 19 have to do with doors or gates.)
(xii) In the judgement of Jeremias, TWNT 6, pp. 923—7, ‘the gates of Hades’ refers not to the realm of the dead but to the ungodly powers of the underworld which will assail the church107 in the latter days. Citing passages for comparison from Revelation (6. 8; 9.1ff.; 20.3, 7—8) and lQH 5.20ff., he takes the text to mean that church will emerge triumphant from the eschatological assaults of evil. Compare Bultmann, History, p. 139: ‘in the end, when the powers of the underworld overcome mankind, the Church will be saved’.”8
In the attempt to come to a conclusion of our own, the following points are to be borne in mind. To begin with, we have already stated objections to interpretations (i)—(ix). These may be eliminated. Next, while it might be argued that the structure of W. 17—19 requires that abrfig refers to nérpq and thus Peter, this is not a decisive observation: the rules about parallelism are nowhere carved in stone. Further, as Zahn, p. 548, noted long ago, the nearness of (161119 to éxxknoia favours connecting the two, as does the common sense observation that the ‘gates of Hades’ must direct themselves against the whole church, not just a part of it. The RSV supplies the correct translation: ‘prevai against it’ (sc. the church).109 In the third place, although ‘gates of Hades’ is a fixed expression in the OT one must beware of reading the OT meaning into Matthew’s text, for conceptions about Hades and Sheol changed over time; By the first century there was a tendency to think of Hades or certain sections of it as an underworld peopled not by the dead in general but by the ungodly dead, as well as by demons and evil spirits.“° The simple equation of Hades with death probably does not hold for Mt 16.18. Fourthly, nothing stands in the way of viewing ‘gates’ as an instance of synecdoche: the ‘gates’ stand for the city and its inhabitants (cf. above). Furthermore, there are ancient texts which blur the distinction between ‘gates’111 and ‘gate keepers’.1 12 Lastly, mum + mncxuml ‘3 is not a recognized idiom (which is another reason for doubting that one can simply appeal to the OT usage of ‘gates of Hades’). But in the LXX the verb is always active when followed by the genitive.114 This means that the gates of Hades should be understood as active: the church on the rock is suffering an onslaught (cf. 7.24—7).
Taking everything into consideration, we do not wish to rule out interpretations (x) and (xi). But interpretation (xii) seems to us to be the best choice. One should probably think of the endtime scenario, when the powers of the underworld will be unleashed from below, from the abyss, and rage against the saints (cf. 1 En. 56.8; Rev 6.8; 11.7; 17.8). The promise is that even the full fury of the underworld’s demonic forces will not overcome115 the church. One may compare Rev 9.1—11, where the demonic hosts, under their king Abaddon, come up from the bottomless pit to torment humanity. They prevail against all save those with the seal of God. Also worth comparing is lQH 6.22—9. In this the author faces the gates of death but is delivered by entering a fortified city founded on a rock: and the whole context is the great eschatological conflict.
If Mt 16.18 be traced back to Jesus, it may have concerned Peter’s role as eschatological missionary. By adding members one builds a community. Thus in making Peter the foundation of the emerging community, Jesus was announcing his pre-eminence as a ‘fisher of men’. Congruent with such a proposal is, first, the fact that Peter was, to judge from Acts, the evangelist par excellence in the primitive community and, secondly, the fact that Peter felt impelled to live his life as a missionary, moving on from Jerusalem to other places.117 As for the meaning of ‘and the gates of Hades shall not prevail against it’ (see above), these words harmonize with the other promises in which Jesus foresees at least some of his disciples survivng to the end, despite eschatological tribulation (see on 10.23 and 16.28).
19. Kai Mum) cot 1:819 10.3n tfig Buctlsiag td'w ofipavébvfi” Compare Isa 22.22 (on this see below); Rev 1.18 and 3.7 (Jesus has the keys of Death and Hades as well as the key of David); 3 Bar 11.2 (the angel Michael is the ‘holder of the keys of the kingdom of Heaven’; cf. Par. Jer. 9.5); 3 En. 18.18 (‘Anapi’el YHWH the prince keeps the keys of the palaces of the heaven of Arabot); 48 C 3 (Metatron has the keys to the treasure chamber of heaven); PGM 3.541 (‘the keys of the triangular paradise of earth, which is the kingdom’). Heaven was conceived of as having gates or doors (cf. 1 En. 9.2).“9 It is not obvious, however, that we have in Mt 16.19 that image.12° For in the synoptics ‘the kingdom of Heaven’ is not equated with the heavenly world. This is why Jeremias construes ‘the Baotkeia 16w oupavdw, whose key Peter holds, as the royal dominion of God in the last time’ and, in our judgement rightly, proposes tht ‘if a concrete image lies behind the phrase rag Kleiéag rfig Baotkeiag raw obpavrbv, we should think of the gates of the future city of God’ (TWNT 3, pp. 748—9 and n. 55; cf. Rev 21.12~13, 15, 21, 25). (If ‘kingdom of Heaven’ has replaced an original ‘temple’ (as perhaps in v.18), then it would be pertinent to remark that ‘keys of the temple’ or ‘sanctuary’ was a known expression: 2 Bar. 10.18; Par. Jer. 4.4.)
Does v. 19a have the same sense as W. 19b and c, so that to have the keys is simply to have the power to bind and loose, or are binding and loosing instead instances of the exercise of the wider authority indicated by 19a? If possession of the keys means the power to bind and loose, then one may urge that Peter is promised no more than the other disciples, for in 18.18 the power to bind and loose is clearly held by others. But if v. 19a is broader in scope, then one can make the case for Peter having a unique function (cf. Vatican II, Dogmatic Constitution on the Church, 3.22). In our estimation, it is most natural to think of v. 19a as being explicated by what follows: to have the keys is to have the power to bind and loose. Further, 19a and vv.19b—c probably have to do with teaching authority (for the connexion between keys and teaching see Lk 11. 52; Sipre on Deut 32.25; b. Sabb. 3la—b; for ‘binding’ and ‘loosing’ as metaphors for halakhic decisions see below). We do, however, still insist that Peter is not thereby put on the same level as his fellow disciples. It remains true that only he is explicitly said to have the keys. More significantly, v. 19 cannot be isolated from W. 17 and 18, and in these last Peter is spoken of in terms not applicable to anyone else. Also, it should not be overlooked that whereas 18.18 concerns the local community or assembly, 16.19 is about the church universal (cf. v. 17); hence the authority bestowed in 16.19 is implicitly wider than that given in 18.18. For these reasons, then, we are not persuaded that the existence of 18.18, with its more general promise of the authority to bind and loose, diminishes Peter’s prominence. If the power to bind and loose was also given to others, that does not entail that those others exercised their power in quite the same way as did Peter, or that they too held the keys of the kingdom. See further Excursus XIV, p. 650.
not 6 éc‘rv 81101]; End ""19 7f]; iamt fisfisuévov év wig obpavoig, mi 6 HIV Mung éni 1:l yfig Ectm Asluuévov év wig ofipavo'ig. Compare 18.18 and Jn 20.23; also Job 38.31 (‘Can you bind the chains of the Pleiades, or loose the cords of Orion?’); Josephus, Bell. 1.111 (‘the real administrators of the state, at liberty to banish and to recall, to loose and to bind’: Maw 18 mi Sacusiv).
What, for Matthew, is the meaning of the juridic 19b—c? The giving of keys manifestly means the bestowing of authority; to have keys means to have power, to be in control (cf. Rev 1.18; 2 En. 40.9—11; b. Sanh. 113a). Beyond that, exegetical disagreement reigns. As with V. 18dc, one can list a host of competing interpretations.
(1) Because ‘binding’ and ‘loosing’ are, in intertestamental literature, most frequently associated with exorcism,121 one might see 16.19 as assigning to Peter power over demons (cf. 10.8; Lk 10.17—19). This can only be a suggestion with regard to the historical Jesus or a preMatthean stage (cf. Hiers (v)). As compared with Mark, Matthew is much less interested in Jesus’ activity as exorcist.122 More importantly, the Matthean context does not suggest exorcism; and in any case, while one can conceive of Jesus giving Peter the power to bind demons, can we think of him giving Peter the power to loose them? Or are we to take the binding clause to refer to the binding of demons, the loosing clause to the loosing or freeing of human beings?
(ii) One could contend that Peter is being told that his judgements about people in the present will determine their, destiny at the last judgement (cf. Schweitzer, Quest, p. 371); or that he has the authority to determine who goes to heaven, who to hell (cf. Falk (v)). With this 19.28 might be held consistent. But where else is the terminology of binding and loosing used in connexion with the eschatological assize?
(iii) Some have taken 16.18 to give Peter the authority to forgive or not forgive sins (perhaps in a baptismal setting).123 In favour of this, appeal has been made not only to Jn 20.23 but also to the targumic expression, ‘loose and forgive’.124 Already in the LXX Ma) can mean—as it does so often in the church Fathers—‘forgive’ (e.g. Job 42.9—10). Particularly striking is a passage in Tg. Neof. Gen 4.7: ‘If you perform your deeds well in this world, it shall be loosed and forgiven (yes'téré wéyes’tébeq) you in the world to come. But if you do not perform your deeds well in this world, your sin shall be retained (nétt‘r) for the day of judgement’. We find no insuperable objection to interpreting Mt 16.19 in terms of forgiveness. If it were not for the fact that another interpretation has even more in its favour (see below), we should be moved to accept it. ,
(iv) Falk (v) has discussed the Talmudic evidence for sages binding and releasing vows, that is, deciding which vows must be kept and which could be absolved (the relevant terms are ’dsar and hittir or prim/1; note m. Naz. 5.1—4; b. Hag. 10a). But one is at a loss to fathom how 16.19 can have anything to do with vows.
(v) Many have thought in terms of the authority to release or enforce a ban, that is, exclusion from the community.125 In support of this, 18.18 is clearly susceptible of such an interpretation; and in b. Mo‘ed. Qat. 16a ’dsar and .v'éré are used of issurng a ban and releasing a ban: ‘A toot (from the horn blown at banning) binds and a toot releases’. According to SB 1, p. 739, however, only in b. Mo‘ed. Qat. 163 does ’fisar mean ‘impose an excommunication’.
(vi) In the view of G. Lambert (v), 16.19 simply refers in general to all the powers given by Jesus to Peter. Citing Hebrew, Egyptian, Greek, and Roman sources he claims that the opposites (binding and loosing) have no specific content but denote totality. This is simply too vague to satisfy and cannot cite any patristic support. It also leaves ‘the keys of the kingdom of Heaven’ unexplained.
(vii) ‘Binding’ and ‘loosing’ appear as technical terms in many magical texts from the ancient world. In b. Sabb. 81b, for example, there is this: ‘R. Hsda and Rabbah son of R. Huna were travelling in a boat, when a certain matron said to them, “Seat me near you,” but they did not seat her. Thereupon she uttered something and bound the boat. They uttered something and freed it’. The terms used are ’fisar and s'éra’. There are parallels to this language in the magical papyri (Smith, Magician, p. 203) and elsewhere. F. C. Conybeare long ago tried to link Mt 16.18 to» such magical usage.126 As with interpretation (i), this is possible only for some pre-Matthean stage. Matthew’s animus towards magical practices is established (Hull, pp. 116—41).
(viii) Allen, p. 177, affirmed that one must distinguish between the church and the kingdom: ‘The Church was to be built on the rock of the revealed truth that Jesus was the Messiah, the Divine Son. To S. Peter were to be given the keys of the kingdom. The kingdom is here, as elsewhere in this Gospel, the kingdom to be inaugurated when the Son of Man comes upon the clouds of heaven. If S. Peter was to hold supreme authority within it, the other apostles were also to have places of rank The éxxhnoia, on the other hand, was the society of Christ’s disciples . . Although Allen is correct in affirming that church and kingdom are not simply identified by Matthew,127 one can hardly see the sense of referring v. 19 to the consummation, to the time when God’s will is done on earth as it is in heaven; for the contrast between earth and heaven (what is bound here will be bound there, what is loosed here will be loosed there) presupposes that heaven and earth have not yet been united under God’s will, that the two spheres are still separate.
(ix) B. H. Streeter interpreted 16.19 by calling upon Lk 11.52 (‘Woe to you lawyers! for you have taken away the key of knowledge; you did not enter yourselves, and you hindered those who were entering’). He wrote: ‘To Peter, then, is given that true insight into the nature of the righteousness taught by Christ—a righteousness that will “exceed that of the scribes and Pharisees”—which is the indispensable qualification of one who is “to bind and to loose” (i.e. to expound the moral law) with such discrimination that what he shall “bind on earth, shall be bound in heaven.”’ Further, Peter had the authority to decide ‘how much or how little of the Law the members of the new dispensation shall be required to observe’.1 23
(x) Streeter’s solution is not far from the major opinion of modern exegetes, which has it that Peter, as a sort of supreme rabbi or prime minister of the kingdom, is in 16.19 given teaching authority, given that is the power to declare what is permitted (cf. the rabbinic ie'ré/se’rd’) and what is not permitted (cf. the rabbinic ’dsar/’dsar). Peter can decide by doctrinal decision what Christians must and must not do.”9 This is the traditional Roman Catholic understanding, with the proviso that Peter had successors. Compare already Epist. Clem. to Jas. 5, in which Simon Peter speaks to the brethren about Clement as follows: ‘Since, as I have been taught by the Lord and Teacher Jesus Christ, whose apostle I am, the day of my death is approaching, I lay my hands upon this Clement as your bishop; and to him I entrust my chair ofdiscourse, even to him who has journeyed with me from the beginning to the end, and thus has heard all my homilies Wherefore I communicate to him the power of binding and loosing, so that with respect to everything which he shall ordain in the earth, it shall be decreed in the heavens. For he shall bind what ought to be bound, and loose what ought to be loosed, as knowing the rule of the Church’. This interpretation of binding and loosing in terms of teaching authority seems to us to be correct (see further below).
(xi) A variant of the preceding interpretation holds that the teaching authority which Jesus grants Peter in 16.19 is granted to other disciples or church leaders in 18.18, so stress should not be laid on Peter’s uniqueness.130
(xii) Another variant has it that the perfect tenses (lit. ‘will have been bound’, ‘will have been loosed’) required that decisions made on earth follow decisions already made in heaven.131 On this view, the movement is not from below to above but from above to below. Heaven does not ratify decisions made by Peter; rather, Peter declares what heaven has already determined. We are not convinced, for at least four reasons. (a) In 18.18 boa ér‘zv Shanta . . . écrm aaaepéva xx)». is immediately followed by this: ‘Again, I say to you, if two of you agree on earth about anything they ask, it will be done for them by my Father in heaven’. Here God’s action is indisputably subsequent to human deicision. Is this not a clue to understanding 18.18, and near proof that the future perfects can be pressed too far? (b) The LXX, the koine, and the NT can use periphrastic future perfects as though they were simple future passives; see Cadbury (v) and BDF § 352. Examples include LXX 2 Bao 22.3; Isa 8.17; Heb 2.13. In Mt 16.19 the future passives are probably emphatic futures (cf. Aristophanes, PI. 1027). Or maybe we have here an Aramaism (Grelot, ‘Origene’ (v), pp. 96—7, citing Dan 5.19). (c) There are a good number of synoptic (including Matthean) sayings in which the first clause describes a human action, the second a correlative divine action subsequent in time, and in which the verb is in both clauses the same (e.g. Mt 5.7; 6.14—15; 10.32—3). 16.19 belongs to this class. (d) One recalls that the rabbis had no difficulty affirming that their halakhic decisions ‘could ‘bind’ God. See the many texts to this effect cited by SB 1, pp. 741—7. Further criticism in Porter (v).
(xiii) According to SB 1, p. 739, interpretations (v) and (x) are not mutually exclusive: both can be true at the same time. Does this not overload the terms?
We have already remarked on our preference for interpretation (x): Peter is the authoritative teacher without peer. This harmonizes with the dominant rabbinic usage”2 and, more importantly, with 23.13: ‘But woe to you, scribes and Pharisees, hypocrites! because you shut the kingdom of Heaven against men; for you neither enter yourselves, nor allow those who would enter to go in.’ Here, as the context proves, the scribes shut the door to the kingdom by issuing false doctrine. The image is closely related to 16.19, and the inference lies near to hand that just as the kingdom itself is taken from the Jewish leaders and given to the church (21.43), so are the keys of the kingdom taken from the scribes and Pharisees and given to Peter. Supportive of this is the broader context of Peter’s confession. In the immediately preceding 16.5—12 Jesus warns: ‘Beware of the leaven of the scribes and Pharisees’. Matthew takes this to be about the teaching of the scribes and Pharisees. It would make good sense for the evangelist, in the very next paragraph, to tell a story in which Jesus replaces the Jewish academy with his own ‘chief rabbi’.
Additional considerations confirm us in our judgement, one being that revelation and doctrinal content are otherwise the major themes of our pericope, another being that if 16.1% and c refer to teaching authority, there is a possible connexion with 19a, for keys are associated with knowledge or teaching in both the synoptic and rabbinic traditions (see above). Lastly, Peter is not only called the ‘first’ of the twelve disciples who in chapter 10 are commanded to preach to Israel, but he, along with others, is in 28.16—20 instructed by the risen Lord to make disciples of all nations, ‘teaching them to observe all that I have commanded you’. Thus the pre-eminent tasks our gospel envisages for Peter are those of proclamation and instruction. This too well suits our interpretation of 16.19.
There is a variant of 16.19b—c in 18.18 and Jn 20.23. With Bultmann, History, p. 141, we judge 18.18 to be secondary vis-d-vis 16.19.133 As for Jn 20.23, it does not depend upon the synoptics (Dodd (v)) and would appear to represent an independent translation from Aramaic (or perhaps Hebrew); which is to say, 16.19 and In 20.23 are translation variants of the same logion.
Emerton (v) has made the satisfying inference that the original saying drew upon Isa 22.22: ‘And I (the Lord) will place on his (Eliakim’s) shoulder the key of the house of David; he shall open (pdtah), and none shall shut (ségér); and he shall shut (sdgar), and none shall open (p6te'ah).’ The pre-Matthean tradition, probably seeing a reference to teaching authority, rendered the Aramaic equivalent ofpitah—perhaps pe'tal; (cf. Tg. Isa 22.22)—with Mail“ and the Aramaic equivalent of sfigar—perhaps ’dhad (cf. Tg. Isa 22.22)—with Sea)!” John’s tradition, on the other hand, could have translated pétah and ’dhad with dupim and Kparém respectively. Such a tradition—history, although certainly hypothetical, is supported by this, that keys are normally associated with opening and closing, not binding and loosing. (Only here in the Greek Bible is Kksig found with Mal”) In other words, ‘binding’ and ‘loosing’ (l9b~c) do not naturally go with keys (v. 19a), but they do go with ‘opening’ and ‘closing’. For this reason one might maintain that v. 19a was not originally joined to W. l9b—c. This is unnecessary. The close parallel in Isa 22.22, which refers to keys as well as to opening and closing, seems sufficient to indicate that we have in Matthew an interpretative translation which has resulted in an unusual juxtaposition of images. Another pointer in this direction is Mt 23.13 (quoted above). If, as we think, this, and not Lk 11.52, is the better witness to Q, we have here, in early tradition, the idea that people can shut (Klaus) others out of the kingdom. It is reasonable to suppose that the picture behind Mt 16.19 was originally the same—especially as this firms up the link with v. 18. Whereas the latter has to do with gates, things which open and close, v. 19, in its original form, would have been about things which open and close gates, namely, keys.‘3"
One final point. Rev 3.7 has this: ‘The words of the holy one, the true one, who has the key of David, who opens and no one shall shut, who shuts and no one shall close’. Here ‘the key of David’—which ‘must be the key that opens the door into the kingdom of God’133—is used for opening and closing, not binding and loosing. Once more then we are encouraged to conjecture that what must lie behind Mt 16.19 is a saying in which Jesus handed over the keys to the kingdom and with them the power to open and to close.
With regard to the original meaning on the lips of Jesus, Cullmann, Peter, pp. 210—12, has proposed that vv. 19a and l9b—c were spoken on different occasions and dealt with different subjects. V. 19a had to do with Peter opening the door to the kingdom through his preaching, 19b—c with the authority to forgive sins. We should like to suggest that consideration be given to the possibility that vv."19a and l9b—c— emended in accord with our previous discussion—belonged together from the first and that both had to do with Peter’s evangelistic mission. If certain Jewish teachers shut the door of the kingdom to others by what they said and did (Mt 23.13 par.), Peter was to open that door by his missionary activity. The apostle had the keys of the kingdom insofar as he proclaimed the good news. By his preaching did he not in effect open the kingdom for those who accepted the gospel, and by the same preaching did he not close the kingdom to those who rejected him and his message? One is reminded of Q’s missionary discourse. The apostolic missionaries bring blessings upon those who receive them but judgement upon those who do not. The one eschatological message effects salvation for some, condemnation for others. It ‘opens’ and ‘closes’ at the same time.
