Hostname: page-component-f554764f5-rvxtl Total loading time: 0 Render date: 2025-04-17T17:40:59.214Z Has data issue: false hasContentIssue false
  • English
  • Français

Angel, Apostle, or Apparition? An Angelomorphic Tradition in Acts 12.15c

Published online by Cambridge University Press:  14 April 2025

Daniel B. Glover Open the ORCID record for Daniel B. Glover [Opens in a new window]
Daniel B. Glover*
Affiliation:
School of Theology and Ministry, Lee University, Cleveland, Tennessee, USA
*
Email: [email protected]
Rights & Permissions [Opens in a new window]

Abstract

This article evaluates a potential angelomorphic tradition in Acts 12.15c. It demonstrates that the most common interpretation of the phrase ὁ ἄγγελός ἐστιν αὐτοῦ (‘it is his angel’, 12.15c) as referring to Peter’s guardian angel fails to account adequately for the evidence. After surveying several different proposals for interpreting the phrase, it argues that angelomorphy/angelisation is the most plausible option available to Luke and his earliest readers. Finally, this article demonstrates how an angelomorphic interpretation of Acts 12.15c is congruent with both the broader concern for angelomorphy/angelisation throughout Luke-Acts, as well as the use of humour and irony in Acts 12.12–15.1

Keywords

angelomorphy/angelisation/angelificationLuke-ActsPeterdaimonificationheroesghosts/phantomsguardian angelsangelic doubles
Type
Articles
Information
New Testament Studies , Volume 70 , Issue 4 , October 2024 , pp. 470 - 485
Check for updates
Copyright
© The Author(s), 2025. Published by Cambridge University Press.

1. Introduction

Luke’s fantastical story of Peter’s escape from prison in Acts 12.6–17 is a carefully crafted narrative, filled with irony and humour.Footnote 2 It is also fairly straightforward: an angel wakes Peter, frees him from his chains and helps him miraculously escape his guards unseen. Now freed, Peter heads toward a gathering of fellow believers, where Rhoda, in a rush of excitement, leaves him knocking at the gate. But this same straightforward narrative has left contemporary exegetes scratching their heads at one central point – how the believers explain what Rhoda heard or saw. After Rhoda delivers them the news that Peter has escaped, the believers rebuff her, declaring confidently in response: ὁ ἄγγελός ἐστιν αὐτοῦ (‘it is his angel’, 12.15c). Some scholars have sought an explanation by reading the response as mockery intended for comedic effect: ‘It’s only his angel, Rhoda. Of course, Peter is not really at the door.’Footnote 3 This interpretation is reasonable; after all, Luke reports that the believers regard Rhoda as ‘manic’ (12.15).Footnote 4 Indeed, the scene’s irony and humour are both pronounced and difficult to miss since these very believers were, until that very point, praying, perhaps even for Peter’s deliverance. Yet when Rhoda ‘announced’ (ἀπήγγειλεν, 12.14c) their prayers had been answered, they refused to believe her (cf. Luke 21.37).Footnote 5 But appeals to humour or irony scarcely answer the question I am posing here: what could the believers, Luke, and his earliest readers understand by ἄγγελος? Once we have answered that, we must also answer a second question: how should we take the believers’ retort since the comment was likely made in jest?

The vast majority of contemporary exegetes have answered the first question quickly by claiming that the angel referenced here is Peter’s guardian angel, and they have essentially ignored the second.Footnote 6 I want to suggest here that there are problems with too quickly assuming that ὁ ἄγγελος αὐτοῦ refers to Peter’s guardian angel. The evidence for guardian angels in this period is paltry at best, and there are several more likely explanations available. I shall argue that the majority position is built on a misreading of a crucial ancient source and that angelomorphy is a better reading of this passage, which coheres with the instances of angelomorphy elsewhere in Luke-Acts.Footnote 7

This article unfolds in three movements. First, I describe the current majority position on the meaning of ἄγγελος in Acts 12.15c, as well as the evidence that has been read to suggest that Peter’s ‘guardian angel’ is in view here. I critique this stance by noting that the qualities attributed to Peter’s angel are not found in discussions about guardian angels before or during Luke’s time. Moreover, the concept of the guardian angel was not fully developed during Luke’s time, nor is there any evidence for the idea of the ‘celestial double’ in this period. Second, I evaluate several other possible ways of interpreting the phrase ὁ ἄγγελός ἐστιν αὐτοῦ, at which point I propose that the best way – that is, the way most readily available to Luke’s earliest and most immediate circle of readers – is to understand the phrase as referencing a post-mortem, angelised Peter. The believers, in other words, believe Peter to be dead. Third and finally, I demonstrate the continuity between their perspective and Luke’s interest in angelification, and I answer how we might go about interpreting this passage in light of its humour and irony.

2. Is ὁ ἄγγελος αὐτοῦ Peter’s Guardian Angel or Celestial Double?

The interpretation of ὁ ἄγγελός ἐστιν αὐτοῦ as referring to Peter’s guardian angel boasts a long history in the interpretation of Acts. Origen is our earliest witness to this reading. Associating this passage with Matt 18.10, he reports that some readers argue that ‘as they had learned once for all that each of the believers had some definite angel, they knew that Peter also had one’.Footnote 8 Origen himself, however, rejects this interpretation – and the believer’s statement itself – as theologically misguided (Comm. Matt. 28). John Chrysostom adopts the interpretation Origen rejects. He writes: ‘But they said, It is his angel. This is true: every person has an angel. And why did it occur to them to think it was an angel? They suspected this from the time of night’ (Hom. Act. 26.3).Footnote 9 Similarly, Bede’s commentary, along with some more recent commentators, points to Matt 18.10 as well as the Shepherd of Hermas 36.2–3 (Man. 6.2.2–3), the latter of which states that two angels live with each person, an angel of righteousness and an angel of wickedness, which respectively impel humans to act righteously and wickedly.Footnote 10 Early on, then, there was a stream of interpretation associating Acts 12.17 with a guardian angel concept.

A text’s Wirkungsgeschichte can be, and often is, instructive, but it alone can never be determinative for historical-critical analysis. So even if the guardian angel interpretation is early relative to contemporary scholars and held some currency among early interpreters, we still must ask whether the concept available to Origen, John Chrysostom and Bede was also available to the author and earliest audience of Luke-Acts.Footnote 11

2.1 Greco-Roman Texts

Was the guardian angel concept present in Luke’s Greco-Roman milieu? No, but there were some relevant precursors. In the Socratic tradition, for example, there was already a concept of the personal daimōn. Plato claimed that Socrates had his own, personal daimōn, though he left it to his later followers to elaborate upon it.Footnote 12 Enter Plutarch and Apuleius. Plutarch (Gen. Socr. 10 [Mor. 580D]) reports that ‘exactly as Homer has represented Athena as “standing at” Odysseus’ “side in all his labors”, so Heaven seems to have attached to Socrates from his earliest years as his guide in life a vision of this kind, which alone showed him the way, illumining his path in matters dark and inscrutable to human wisdom, through the frequent concordance of the sign with his own decisions, to which it gave a divine sanction’.Footnote 13 This daimonic companion, who accompanied Socrates from birth, gave him signs – either a sneeze (Gen. Socr. 11 [Mor. 581B] or a voice/vision (Apuleius, De deo Socr. 20.4) – to discern the decision most in conformity with the divine will (cf. Plutarch, Gen. Socr. 10 [Mor. 580D–E]). Daimones, intermediaries between humanity and the gods (Plato, Symp. 23 [202D–E]), communicate the divine will to all but only ‘find an echo in those only whose character is untroubled and soul unruffled, the very men in fact we call holy and daemonic’ (Gen. Socr. 20 [Mor. 589D]). Such a man was Socrates.

Daimonological reflection reached an apex with the Middle Platonists, who elaborate ontologically upon Socrates’s daimōn.Footnote 14 But while we have in the Platonic tradition a belief in a personal daimōn accompanying Socrates the sage, this is not evidence of the belief in the general ‘guardian angel’ concept. Indeed, Socrates’ daimōn does not fit the guardian angel concept comfortably – at least, not yet.Footnote 15 Serious development is still necessary, and there is little evidence for it by Luke’s period.Footnote 16 For Plato and his Middle Platonist interpreters, only the exceptional have such a daimonic guide. Plato, Plutarch and Apuleius discuss it only in relation to Socrates, such that the concept appears not only exceptional but unique. Peter, it is true, is an exceptional figure in Acts, though he is not unique, and one justifiably may suspect that in light of Luke’s pneumatology – the divine πνεῦμα poured out on all flesh (e.g., Acts 2.16–18) – such a concept would prove superfluous and ultimately unhelpful. Furthermore, the Socratic parallels in Acts are between not Peter and Socrates but Paul and Socrates.Footnote 17 Perhaps most important to note, though: these daimones are never said to take the appearance of their wards.

2.2 Jewish Texts

A few Jewish sources predating Acts speak of angels whose duty it is to keep watch over or care for humanity (Job 5.1; 33.23; Heb 1.14; Philo, Gig. 12; cf. Deut 32.8). But these texts do not speak of a guardian angel who belongs permanently or even regularly to any given individual. Further, none of these angels bear the image of or sound like their mortal wards. The angelic overseer is, therefore, not sufficient to explain Acts 12.15c.Footnote 18

Some scholars direct our attention to the Book of Tobit, which recounts how the angel Raphael guides and protects Tobit during his journeys to Media (5.4–22). While Raphael does disguise himself as Tobit’s kinsman Azariah, this reveals nothing more than that angels can change appearance, and it is clearly in imitation of Athena’s polymorphism and attendance of Odysseus in the Odyssey.Footnote 19 Raphael in Tobit does not give any evidence towards a generalised concept of the guardian angel or an angelic doppelgänger that scholars hope to find in Acts 12.15c. Tobit is simply favoured by heaven, as was Odysseus.

Pseudo-Philo’s Liber Antiquarum Biblicarum is also frequently cited in support of the guardian angel interpretation. In LAB 59.4, David sings about how God has delivered him from his enemies by guarding him with angels, who are called ‘guardians’.Footnote 20 These ‘guardians’ appear also in 11.12 (‘You shall not be a false witness against your neighbour, speaking false testimony, lest your guardians speak false testimony against you’), and 15.5 refers to angels who watch over the people of Israel (‘I will command my angels who watch over them not to intercede for them’). However, angels guarding over the people of Israel is no evidence of the individuated or generalised guardian angel concept; rather we have here evidence of the national angel or patron deity concept, which appears in many Jewish texts.Footnote 21 Certainly, nothing here should remind us of Peter – who, in the estimation of the believers, has likely died (and, therefore, not been guarded) – or of ‘his angel’, who ought to look and/or sound like Peter.

In the absence of textual predecessors to Luke-Acts, many scholars point to a significant Amoraic text purportedly in support of the guardian angel interpretation, adding to it the belief that the angel was also a doppelgänger.Footnote 22 For instance, Darrell Bock remarks: ‘the later rabbinic text Gen. Rabbah 78 [50a] on Gen. 33.10 has a specific personal angel whose look matches that of the person he protects’.Footnote 23 (The text in question is Gen. Rab. 78.3, according to the Socino versification.Footnote 24 ) Roloff comments to the same effect:‘Vorausgesetzt ist dabei die Vorstellung, daß jeder Mensch einen Schutzengel hat, der ihm als sein himmlischer Doppelgänger auch äußerlich gleich’, citing the same rabbinic text.Footnote 25 Commenting upon Gen 32.29, the Midrash reads: ‘thou [Jacob] hast striven with אלהים and conquered them, and with mortals, and hast conquered them. אלהים alludes to the angel. R. Ḥama b. Ḥanina said: “It was Esau’s guardian angel. That is what Jacob meant when he said to him: to see your face is like seeing the face of אלהים (Gen 33.10)”’Footnote 26 and, earlier, ‘to see your face is like seeing the face of אלהים – since you have received me with such favor (Gen. 33.10)’ (77.3).Footnote 27 At first glance, these texts appear to support the guardian angel interpretation, for we would have an example of a guardian angel who also looks like its human ward (i.e., is also the human’s doppelgänger).

These passages are clear evidence for the emergence of the guardian angel concept by the fourth century ce.Footnote 28 Nevertheless, they do not mean what Lukan scholars have taken them to mean. The quotation from Gen 33.10 (‘to see your face is like seeing the face of אלהים’) does not say that the angel’s face shares Esau’s physical features. Instead, as with many biblical texts, it admits of several rabbinic interpretations, and none of them refers to an angelic doppelgänger. In the earlier discussion, b. Ḥanina gives the full quotation (‘since you have received me with such favor’) to demonstrate that, for Jacob, seeing the face of Esau’s guardian angel was like seeing the face of אלהים, because אלהים blessed Jacob (32.26–9). By contrast, in the later discussion, b. Ḥanina omits the line ‘since you have received me with such favor’ in order to offer another interpretation: ‘to see your face is like seeing the face of אלהים’ because the face denotes God’s judgment. In his words: ‘even as the face of אלהים denotes judgment, so does thy face denote judgment; even as with respect to the face of אלהים [it is written], And none shall appear before My face empty-handed (Ex. 23.15), so art thou; none may appear before thy face empty-handed’.Footnote 29 Esau’s face does not physically look like the angel’s; it resembles God’s face – metaphorically – because Jacob stands in judgement before both. To interpret either of these passages as referring to an angelic doppelgänger misses the exegetical point: Esau’s face manifests alternately grace and judgement to Jacob and is therefore like the face of אלהים. That is the totality and the extent of seeing in Esau’s face that of אלהים.Footnote 30 None of it concerns the literal physical appearance of either Esau or his guardian angel, neither here nor in its later reception in the Midrash.Footnote 31

2.3 Early ‘Christian’ Texts

Other scholars point to early ‘Christian’ texts as evidence for an early guardian-angel-as-doppelgänger concept. One text of note is Matt 18.10, which reads: ‘See that you do not despise one of these little ones. For I say to you that their angels are continually in the heavens seeing the face of my heavenly Father.’Footnote 32 What Jesus describes here is by no means clear, though the terminology (οἱ ἄγγελοι αὐτῶν) evokes Acts 12.15c (ὁ ἄγγελός αὐτοῦ). In spite of the many scholars who cite him in favour of the opinion, James Moulton insists that the concept of the angel presented in Acts 12.15c and Matt 18.10 ‘is not … the existence of guardian angels’.Footnote 33 Moulton, rather, proposes that these angels are ‘“representative angels” as spiritual counterparts of human individuals or communities, dwelling in heaven, but subject to changes depending on the good or evil behavior of their complementary beings on earth.’Footnote 34 That suits the context of Matthew well. But whatever Jesus may be describing in Matt 18.10, it is clearly different from Acts 12.15. Matthew’s angels are ‘continually in the heavens (ἐν οὐρανοῖς διὰ παντὸς)’, before the face of God. Not so with Peter’s angel in Acts, who visits the faithful believers, perhaps, some have speculated, indicating to them Peter’s death. Furthermore, Moulton marshals no evidence from the history of religions to demonstrate that either guardian angels or representative angels resembled their wards.

Others invoke the Shepherd of Hermas 25.4 (= Vision 5.4) as the background to Acts 12.15c and as evidence for a belief that (at least some) humans possessed not only a guardian angel but also a celestial double.Footnote 35 This text, however, does not speak to that concern.Footnote 36 Here Jesus transforms from a human shepherd into a celestial form. This passage, therefore, is as clear an instance of polymorphic Christology as one can find.Footnote 37 It refers neither to guardian angels nor celestial doubles but to Jesus’ (varied) celestial appearance.Footnote 38

Finally, there is a conceptual problem with invoking the guardian angel interpretation. Although various interpreters propose that the guardian angel could indicate one’s death,Footnote 39 the sources do not bear this out: there is not a single example of a guardian angel delivering news of its ward’s demise in any ancient source. As a result, Daniel Marguerat rightly juxtaposes the readings: either the believers think it is Peter’s guardian angel, or they believe he is dead.Footnote 40

None of the parallels marshalled in defence of the guardian angel concept, then, satisfactorily explains what we find in Acts 12.15c, though we have steps towards the concept through the Socratic and Platonic traditions.Footnote 41 Even if we could find the guardian angel concept before the late second century ce – which grants an assumption that is not evident – there is no text that describes the guardian angel whose likeness matches the human over whom it is charged, and certainly none whose voice matches its ward. It is not impossible, of course, that Acts 12.15c represents the earliest example of the guardian angel and the only example of the guardian angel as a celestial doppelgänger. However, given the problems of the date of comparanda, the necessary combination of concepts (guardian angel plus heavenly double), as well as the exegetical difficulty that the guardian angel concept does not really play the role of messenger for the human but for God, the guardian angel and angelic doppelgänger concepts are not ultimately sufficient to explain Acts 12.15c. We should, therefore, seek out an alternate interpretation.

3. Apparitions and Angelification

Most interpreters, as I have noted, understand the mention of Peter’s angel to refer to his guardian angel as a doppelgänger sharing news of his death. This reading is not borne out by the sources, since none describes a guardian angel visiting one’s friends or family after one’s death. At least two other options present themselves, however: apparitions and angelifications.

In ancient apparition/ghost stories, such visitations are a common occurrence. While the category ‘ghost story’ is a broad one in Mediterranean antiquity, since under its umbrella fall (a) stories of disembodied spirits haunting the grave, the sea, or the home as well as (b) stories of revenants waking to embodied existence for a short time, it will still prove heuristically useful.Footnote 42

There are sundry examples of disembodied spirits haunting beyond the grave. Already in the Iliad, we read how the ghost of Patroclus visits Achilles to request proper funerary rites (23.65–110). Likewise, the Odyssey includes several scenes involving disembodied spirits, including the ghosts of Elpenor (11.51), Anticleia (11.85), Teiresias (11.90) and Atreus (24.191–2).Footnote 43 To Patroclus, Achilles says (Il. 23.97–101): ‘“come closer; though it be but for a little time, let us clasp our arms about one another, and take our fill of dire lamenting.” So saying he reached out with his hands, yet clasped him not; but the spirit like smoke (ψυχὴ … ἠύτε καπνός) was gone’ (Murray and Wyatt, LCL). In the Phaedo, Plato uses Socrates as his mouthpiece to justify ghosts philosophically. In certain cases, Plato writes, a person’s soul (ψυχή) might remain visible after death, if in life that person were wicked (προτέρας τροφῆς κακῆς οὔσης) or had an undue fear of the invisible or Hades (τοῦ ἀιδοῦς τε καὶ Ἅιδου) and so lingered on after death, grasping futilely for the life she once had (Phaed. 81C–D). Clearly, though, this is not what we are dealing with in Acts 12.15c, since Peter is not ‘spirit like smoke’ but solid, visible and able to rap on the door.

But these are not the only ancient apparitions. Phlegon of Tralles, for instance, reports on a teenaged girl who died, but whose body kept company with men in the evening – until her father found out, of course (Mir. 1). Yet the woman’s revenant returned to the grave for eternal slumber once the sexual liaisons ended. The problem with suggesting Peter’s ‘angel’ refers to a ghost or revenant is the terminology: although there are several terms for ‘ghost’ and ‘revenant’ (εἰδῶλον, πνεῦμα, φάσμα, ψυχή, and so forth), as far as I can tell, ἄγγελος is never included among them. There are, nevertheless, three paths to a connection with spirits of the dead here: the genius, the daimōn and the hero.

First, the term may refer subtly to the Roman conception of the genius. Ancient Romans conceived of the genius as a person’s ‘life principle’ or ‘procreative force’, a ‘supernatural power accompanying an individual’.Footnote 44 The problem with this view is that, as far as I can find, the Roman genius is never glossed by the term ἄγγελος. Nevertheless, there is no Greek equivalent, and among both philosophers and grammarians, the term was assimilated with δαίμων, which Luke would certainly never have used to describe any power associated with Peter.Footnote 45 If Luke’s readers interpreted ὁ ἄγγελος … αὐτοῦ as Peter’s genius, they would need to associate the ἄγγελος with the Greek δαίμων, which sometimes glossed the Latin genius.Footnote 46 This initial association of ἄγγελος with δαίμων is quite plausible in view of the following observations about Philo, but given that Romans believed the genius died with its human counterpart, this option seems unlikely.

A second and, in my opinion, better suggestion is the association of the ἄγγελος with the δαίμων and ψυχή. The term ἄγγελος was sometimes associated with a disembodied soul when those souls were identified with the concept of the daimōn. For Philo, as well as for other Middle Platonists (e.g., Plutarch, Rom. 28.8), the soul of the wise and virtuous may experience a daimonification at death, and Philo, uniquely among his Middle Platonist peers, describes angels in precisely these terms.Footnote 47 John Dillon remarks, ‘Philo’s angelology (or demonology) is, then, essentially Middle Platonic.’Footnote 48 This much is true. But it is also Jewish Middle Platonism, since he filters the biblical tradition through the philosophical and vice versa. A Jewish reading of Acts 12.15c cannot, therefore, dismiss this following parallel out of hand unless working with a problematic Judaism/Hellenism distinction, which Philo above all deconstructs.Footnote 49

In his treatise On Giants, Philo argues that souls (ψύχαι), although substantively the same, come to various ends in the cosmos. Some never became enfleshed in human beings but remained celestial beings in service to God and ministers to humanity. In his words: ‘some souls have descended into bodies, but others have never deigned to be brought into union with any of the parts of earth. They are consecrated and devoted to the service of the Father and Creator customarily employs them as ministers and helpers, to have charge and care of mortals’ (Gig. 12; cf. 15).Footnote 50 Other souls who live out their days concerned for material pleasures are swallowed by them and, ‘retracing their steps’ (παλινδρομοῦσιν αὖθις, Somn. 1.140), are caught within a cycle of reincarnation.Footnote 51 But those ‘who have given themselves over to legitimate philosophy (τῶν ἀνόθως φιλοσοφησάντων)…shall partake of an unbodied and immortal life (τῆς ἀσωμάτου καὶ ἀφθάρτου ζωῆς) beside the Uncreate and Immortal (ἀφθάρτῳ) One’ (Gig. 14–15).Footnote 52 Because this exalted existence shares with God incorporeality and immortality, it is indeed divine (μειζόνων φρονημάτων καὶ θειοτέρων ἐπιλαχοῦσαι, Somn. 1.140). Hence, when humans devote themselves not to the pleasures of the material world but to divine truth, virtue and philosophy, they free themselves from corporeal existence, becoming pure souls, communing with God and other souls in the air forever (Somn. 1.139). These convictions, therefore, lead Philo to confess: ‘Therefore, while souls, daimones, and angels bear different names, yet they are but one underlying thing. If you bear this in mind, you will have fled from even the weightiest deisidaimonia’ (Gig. 16).Footnote 53 The liberated and ascended soul is one and the same thing as an angel or daimōn, only without the negative connotation.Footnote 54

Luke’s quip about Peter’s ‘angel’ in Acts 12.15c makes a great deal of sense when compared to Philo’s Jewish-Middle-Platonist description of the virtuous philosopher’s post-mortem state. The problem in this case, however, is that the Middle Platonists imagine the daimonified ψυχή as invisible and incorporeal.Footnote 55 Gratefully, another potential option presents itself: heroification narratives.

Plutarch and other philosophers often speak in the same breath of heroes and daimones. For instance, Plutarch writes about the ascent and eventual deification of the soul: ‘virtues and souls together are carried up from humans to heroes, and from heroes to daimones, and from daimones, once they are at the end as in the final cleansing rituals and should be made holy, to be free from all mortality and suffering, to the gods’ (Rom. 28.8).Footnote 56 Whereas Plutarch ranks daimones more highly than heroes, other texts conflate the two. Hence, John Dillon claims, ‘the distinction between them [heroes] and daemons in the Platonic period is not quite clear’.Footnote 57 Elsewhere, Philo provides some connective tissue between the terms ‘hero’ and ‘angel’. In his words (Plant. 14):

This is the host of the bodiless souls. Their array is made up of companies that differ in kind. We are told that some enter into mortal bodies, and quit them again at certain fixed periods, while others, endowed with a diviner constitution, have no regard for any earthly quarter, but exist on high nigh to the ethereal region itself. These are the purest spirits of all, whom Greek philosophers call heroes (ἥρωας), but whom Moses, employing a well-chosen name, entitles ‘angels’ (ἀγγέλους). (Colson and Whitaker, LCL)

Even here, however, Philo imagines that these figures are invisible, although, in most contexts, authors regularly depict heroes as visible, audible and corporeal.Footnote 58

Nevertheless, these last two options – both species of daimonification/angelification – seem to fit the Lukan context better. But if Luke’s ὁ ἄγγελος … αὐτοῦ is understood as an angelification by way of philosophical discussions of daimonification or heroisation, it adopts the language without embracing all of its convictions. And here perhaps some methodological clarity is in order.

Jonathan Z. Smith helpfully points out that the nature of comparison never involves only two elements.Footnote 59 That is to say, the comparative equation is never ‘x resembles y’, for, as Smith remarks, ‘what is being asserted is not a question of the classification of species x and y as a common genus, but rather a suppressed multi-term statement of analogy and difference.’ Instead, comparison necessarily involves (at least) three terms. That is, ‘x resembles y more closely than it resembles z.’ In our case, I am suggesting that Luke’s ὁ ἄγγελος…αὐτοῦ resembles daimonification/angelification more closely than it resembles the concept of the guardian angel or Greco-Roman apparitions. This statement of resemblance is not necessarily genealogical, but it is bolstered by the observations that there is no evidence for the guardian angel concept until centuries later and that there is no evidence for the guardian-angel-as-doppelgänger at all. In the following section, I will further bolster this interpretation of Acts 12.15c as angelification by pointing towards several key angelomorphic texts in Luke-Acts which would have primed readers to hear angelification in Acts 12.15c.

4. Angelification and Interpretation of Acts 12.1–17

4.1 Luke’s Interest in Angelification

While Luke certainly exhibits some traits of what Philo and Plutarch call deisidaimonia, he shares with them the concept of angelification or daimonification of the righteous after death. There are several key passages to consider in this regard. Luke 20.34–6 emerges as particularly important. These verses read:

καὶ εἶπεν αὐτοῖς ὁ Ἰησοῦς· οἱ υἱοὶ τοῦ αἰῶνος τούτου γαμοῦσιν καὶ γαμίσκονται, οἱ δὲ καταξιωθέντες τοῦ αἰῶνος ἐκείνου τυχεῖν καὶ τῆς ἀναστάσεως τῆς ἐκ νεκρῶν οὔτε γαμοῦσιν οὔτε γαμίζονται· οὐδὲ γὰρ ἀποθανεῖν ἔτι δύνανται, ἰσάγγελοι γάρ εἰσιν καὶ υἱοί εἰσιν θεοῦ τῆς ἀναστάσεως υἱοὶ ὄντες.

Jesus said to them, ‘The children [or “sons”] of this generation marry and are given to marry, but those deemed worthy to obtain to that age and the resurrection from the dead neither marry nor are given to marry. For they are no longer able to die, for they are equal to the angels and children [or “sons”] of God, being children [or “sons”] of the resurrection.’ (My trans.)

With these words, Jesus answers the Sadducee’s question about the marital status of a woman married seven times in her lifetime – a problem raised by the Torah’s levirate law.Footnote 60 His answer avoids the particularity of their question and addresses, instead, the nature of the resurrected body. Jesus describes this nature as ἰσάγγελοι. This term is often translated as ‘like the angels’, following the Matthean (22.30) and Markan (12.25) parallels, which use the phrase ὡς ἄγγελοι ἐν τῷ οὐρανῷ/ τοῖς οὐρανοῖς (‘like angels in heaven/the heavens’). But Luke lacks ὡς and replaces it with the prefix ἰσο- (‘equal to’) – a rather significant departure from its Markan source. In a reception-historical study, M. David Litwa has convincingly demonstrated that early Christians such as Tertullian (Cult. fem. 1.2.5; Marc. 3.9.7; On Prayer 3.3; Res. 36.2), Clement of Alexandria (e.g., Strom. 6.13.105.1; 7.10.57.5), Origen (Cels. 4.29; Princ. 4.4.2; Hom. Luc. 39.2; Comm. Matt. 17.30; Hom. Lev. 9.11.1; Comm. Jo. 2.140; 10.187) and Eusebius of Caesarea (Comm. Ps. [PG 23.132.1, 1269.15–21]; Vit. Const. 3.46.2) among others, interpreted the passage and its term ἰσάγγελοι as promoting eschatological, post-mortem angelification.Footnote 61

If so many early Christians, who were native Greek speakers to boot, interpreted the NT hapax legomenon ἰσάγγελοι to denote not only equality to angels in status but also physical transformation, then we would do well to take it seriously. Although some scholars have derided this interpretation as a ‘popular’ outworking of ‘folk religion’,Footnote 62 there is good evidence for widespread – but not to say ‘popular’ – Jewish belief in angelification or angelomorphy.Footnote 63 The Enochic literature contains several accounts of Enoch’s angelification (1 En. 71.14–17; 2 En. 22.10; 3 En. 15) and examples of the patriarchs’ angelification in Philo (e.g., Sacr. 5–7), and the Dead Sea Scrolls also obtain.Footnote 64 Since Luke writes that the scribes comment that Jesus has ‘spoken well’ (διδάσκαλε, καλῶς εἶπας, 20.39), perhaps we should think that they share this view of the resurrected body.

The explicit phrase ἰσάγγελοι γάρ εἰσιν is not all the evidence we have in this passage of post-mortem angelisation. According to Luke, this angelic equality consists in receiving both immortality and the title or status ‘children/sons of God’ (υἱοί θεοῦ),Footnote 65 a term regularly used to refer or interpreted so as to refer to angelic or subsidiary divine beings throughout the LXX (e.g., Gen 6.2, 4; Deut 32.43; Pss 28.1; 81.6, 7; Wis 5.5; cf. also Job 1.16; 2.1; 38.7; Dan 3.25, 28) as well as in the apocalyptic Jewish texts (or ‘sons of heaven’; e.g., 1 En. 6.2; 4Q418 fr. 2.1–4; 4Q418 fr. 212, 213.4; cf. Jub. 5.1).Footnote 66 This divine sonship signifies a new status (as in John 1.12; Rom 1.3–4; 1 John 3.2) and ontology (οὐδὲ γὰρ ἀποθανεῖν ἔτι δύνανται, Luke 20.36!). Angelification or angelomorphy is, therefore, a likely interpretation of the personal eschatology promoted in this verse.Footnote 67

Elsewhere Luke speaks of Stephen becoming angelic at his death (Acts 6.15). Luke writes that those sitting in the Sanhedrin ‘looked intensely’ (ἀτενίσαντες) upon Stephen’s face, which appeared to them ‘like an angel’s face’ (ὡσεὶ πρόσωπον ἀγγέλου). The ‘intense stare’ of the Sanhedrin initiates them into the knowledge of Stephen’s true role: he is a prophetic messenger of God.Footnote 68 But more than that, Stephen himself seems to begin the process of becoming an angel.Footnote 69 Knowing that the end of his natural life is imminent, we can read Acts 6.15 as portraying Stephen relinquishing his hold on tangible things, even to life. He not only embodies Jesus’s radical call to self- and material-abandonment (Luke 9.23; Luke 14.26; 18.29), but he also parallels Jesus’ death, mimicking his final words (Luke 24.34, 46; Acts 7.59–60), expecting his ‘spirit’ (πνεῦμα, Acts 7.59) to ascend to God.Footnote 70 The remark about Stephen’s face becoming as the face of an angel certainly prepares the audience for his coming demise,Footnote 71 but it also communicates that, in the face of death, Stephen begins the process of angelic exaltation.Footnote 72

These Lukan passages strengthen connections between the believers’ assumption of Peter’s death and angelic identity in Acts 12.15c.Footnote 73 Contextual and linguistic considerations support interpreting Peter’s ‘angel’ as a kind of angelic apparition. That is, the believers presume Peter has died and, in words better suited to Luke’s Greek-speaking milieu, experienced daimonification. But because Luke reserves the term daimōn in the deisidaimonic sense of an evil ‘demon’, he employs the exchangeable and preferable term ‘angel’, following the linguistic custom of Greek-speaking Jews like Philo.

4.2 Other Hints of Angelification in Acts 12.15c

In addition to this broader Lukan concern in angelification, within the text itself, Luke offers several cues indicating to the close reader that Paul’s death and angelification are in view. The believers’ claim that Rhoda is ‘mad’ (μαίνῃ) corresponds closely to Festus’ response to Paul’s preaching on the resurrection: μαίνῃ, Παῦλε (Acts 26.24). Also related to Paul’s preaching on the resurrection is Acts 23.8–9. This passage mentions resurrection right alongside of ‘angel’ (in the singular ἄγγελον) and ‘spirit’ (πνεῦμα – the very substance comprising resurrected humanity, according to the historical Paul’s proclamation in 1 Cor 15.44).Footnote 74 Luke’s μήτε suggests that he views these categories as discrete,Footnote 75 and he certainly does not seem to believe that resurrected existence will be purely pneumatic (Luke 24.39).Footnote 76 Nonetheless, we would do well to remember that Luke does not just incidentally list the beliefs of the Sadducees here, since Sadducees crop up earlier in Luke-Acts. Rather, Luke here combines what elsewhere in the Lukan corpus are three components of resurrection life – resurrection, angel, spirit – to characterise Paul’s resurrection belief in contrast to those of the Sadducees.Footnote 77 So here, too, we find angels associated with Luke’s image of the afterlife.Footnote 78

Our passage also echoes the disciples’ disbelief over the ‘idle words’ (ἐφάνησαν…λῆρος τὰ ῥήματα) of the women’s testimony to the resurrection in Luke 24.11. The believers’ disbelieving dismissal likely does not evince doubt that Peter could escape from prison, since these were neither impossible nor uncommon in literature contemporaneous with Luke-Acts. Instead, their dismissal attests to the believers’ expectation that Peter had already been put to death. In other words, Rhoda is ‘manic’ because she claims to have seen a man who was supposed to be dead. Indeed, μαίνομαι often refers to being driven mad by spirits or δαίμονες (Cf., e.g., Herodotus, Hist. 4.79: ὁ δαίμων…βασιλέα λελάβηκε, καὶ…μαίνεται; cf. Euripides, Bacch. 298–9), so Luke chose an appropriately precise word here: Rhoda has been driven made by Peter’s δαίμων – or, in more pious or deisidaimonic language, his ἄγγελος.Footnote 79

4.3 How to Take the Humour and Irony in Acts 12.15c

In light of my interpretation of Acts 12.15c as Peter’s post-mortem angelification, how should we take the believer’s explanation in view of the narrative, which is filled with irony and humour? I suggest that although the believers are wrong to dismiss Rhoda’s claims by retorting that it was Peter’s angel, they nevertheless evince a personal eschatology coherent with Luke’s own. While Rhoda proclaims that the believers’ prayers had been answered, they choose to believe that Peter has died rather than believe the positive report of Rhoda the slave – setting up the punchline, as it were. These unbelieving believers now expect that Herod had by now put Peter to death like James before him (12.2), and if Rhoda believed Peter was rapping on the gate, it is because his daimonified/angelised spirit – and not a guardian angel, angelic doppelgänger or ghost – had passed by to declare his death, thus inspiring Rhoda’s ‘mania’. (We would, therefore, capture the sense of ὁ ἄγγελός ἐστιν αὐτοῦ better if we translated it as ‘it is the angel of him’.)Footnote 80 But Peter’s continued knocking jolts the unbelieving believers into belief that Rhoda might not be mad after all, and there they find Peter delivered (and their prayers answered). This scene, then, would repudiate not belief in post-mortem angelomorphy but rather the believers’ refusal to believe in God’s power to deliver from any circumstance.

5. Conclusion

This article has shown that the standard interpretation of Acts 12.15c as referring to Peter’s guardian-angel-as-doppelgänger does not withstand scrutiny. A more likely interpretation is that the praying believers believe that Peter has perished and, subsequently, experienced angelification. Angelification is attested in Judaism in the time of Luke-Acts and would follow the tendencies in certain philosophical literature too. Furthermore, angelification aligns with Luke’s interest in the topic elsewhere in Luke-Acts. Not only does Jesus explicitly advocate angelification and transformation into isangelic ‘children/sons of God’ in Luke 20.34–6, but Luke also portrays Stephen becoming angelic immediately prior to his martyrdom (Acts 6.15). Furthermore, the claim that Rhoda has gone mad (μαίνῃ) both corresponds closely to Festus’ response to Paul’s preaching on the afterlife (μαίνῃ, Παῦλε, Acts 26.24) and makes excellent sense on the presumption of daimonification/angelomorphy since ‘mania’ was often inspired by spirits and daimones. We have in Acts 12.15c, therefore, another passage in Luke-Acts that describes angelisation as the (pen)ultimate end of the righteous.Footnote 81

Competing interests

The author declares none.

References

1 I would like to offer my thanks to Drs. Crispin H.T. Fletcher-Louis and Emily Gathergood for reading earlier drafts of this paper and for their comments, which were valuable and instructive.

2 ‘Luke’ is used throughout for convenience to signal the author of Luke-Acts, whomever he may be. I presume that our author is an educated cultural producer, who would have been familiar with the important philosophical theories and trends of the day, at least to a small extent. Finally, on the date of the composition of Luke-Acts, see below.

3 On Rhoda as a comedic figure, see Harrill, J. Albert, ‘The Dramatic Function of the Running Slave Rhoda (Acts 12.13–16): A Piece of Greco-Roman Comedy’, NTS 46 (2000) CrossRefGoogle Scholar.

4 Acts 12.15: οἱ δὲ πρὸς αὐτὴν εἶπαν· μαίνῃ (‘But they said to her: “You’ve lost your mind!”’). In this case, they conclude that she is unreliable in part because of her enslaved status, and slaves, it was thought, were less trustworthy. Compare Electra’s similar response and effect in Sophocles, El. 879–83, and see the bibliography cited in n. 5 below.

5 ἀπήγγειλεν in 12.14c is evidently a pun on ἄγγελος in 12.15c. On comedic elements in this scene, see Kathy Chambers, ‘“Knock, Knock—Who’s There?” Acts 12.6–17 as a Comedy of Errors’, A Feminist Companion to the Acts of the Apostles (ed. Amy-Jill Levine; London: T&T Clark, 2004) 89–97.

6 See Moulton, James H., ‘It Is His Angel’, JTS 3 (1902) Google Scholar; Hermann L. Strack and Paul Billerbeck, Kommentar zum Neuen Testament aus Talmud und Midrasch (6 vols.; Munich: Beck, 1922–1961) 1:781–3; 2:707; Dibelius, Martin, Der Hirt des Hermas (Tübingen: Mohr Siebeck, 1923) Google Scholar; Kirsopp Lake and H. J. Cadbury, The Acts of the Apostles: Translation and Commentary (ed. F. J. Foakes-Jackson and Kirsopp Lake; vol 4 of Beginnings of Christianity Part 1: The Acts of the Apostles; London: Macmillan, 1920–1933) 138; Bruce, F.F., The Acts of the Apostles: The Greek Text with Introduction and Commentary (Grand Rapids: Eerdmans, 1951) Google Scholar; Haenchen, Ernst, The Acts of the Apostles: A Commentary (Philadelphia: Westminster Press, 1971) Google Scholar; Howard Marshall, I., Acts (TNTC; Leicester: InterVarsity Press, 1980) Google Scholar; Pesch, Rudolf, Die Apostelgeschichte (2 vols.; EKKNT; Zurich: Neukirchener Verlag, 1987) 1: Google Scholar; Conzelmann, Hans, A Commentary on Acts of the Apostles (Hermeneia; Philadelphia: Fortress, 1987) Google Scholar (so also with the updated Hermeneia Acts commentary, whose author I choose not to name); Johnson, Luke Timothy, Acts of the Apostles (Sacra Pagina; Collegeville: Liturgical Press, 1992) Google Scholar; Barrett, C. K., The Acts of the Apostles (2 vols.; ICC; Edinburgh: T&T Clark, 1994) 1: Google Scholar; Jervell, Jacob, Die Apostelgeschichte (KEK; Göttingen: Vandenhoeck & Ruprecht, 1998) CrossRefGoogle Scholar; Witherington, Ben III, The Acts of the Apostles: A Socio-Rhetorical Commentary (Grand Rapids: Eerdmans, 1998) Google Scholar; Le Cornu, Hilary and Shulam, Joseph, A Commentary on the Jewish Roots of Acts (2 vols.; Jerusalem: Academon, 2003) Google Scholar; Talbert, Charles H., Reading Acts: A Literary and Theological Commentary (rev. edn; RNT; Macon: Smyth & Helwys, 2005) Google Scholar; Bock, Darrell L., Acts (BECNT; Grand Rapids: Baker Academic, 2007) Google Scholar; Marguerat, Daniel, Les Actes des Apôtres (2 vols.; CNT 5; Geneva: Labor et Fides, 2007–2015) 1:Google Scholar; Parsons, Mikeal C., Acts (Paideia; Grand Rapids: Baker Academic, 2008) Google Scholar; Malina, Bruce J. and Pilch, John J., Social-Science Commentary on the Book of Acts (Minneapolis: Fortress, 2008) CrossRefGoogle Scholar; Wall, Robert W., The Acts of the Apostles: Introduction, Commentary and Reflection (NIBC; Nashville: Abingdon, 2015) 9:1–292, here 140Google Scholar; Holladay, Carl R., Acts: A Commentary (NTL; Louisville: Westminster John Knox, 2016) Google Scholar. This being a guardian angel is assumed but not discussed by Roberts Gaventa, Beverly, Acts (ANTC; Nashville: Abingdon, 2003) Google Scholar. Two rare examples of overlooking this peculiar phrase are Cho, Youngmo and Park, Hyung Dae, Acts (2 vols; NCCS; Eugene: Cascade, 2019)Google Scholar – surprising given their interest in the other ‘angels’ in this chapter – and Dunn, James D.G., The Acts of the Apostles (Grand Rapids, Eerdmans, 2016)Google Scholar.

7 Throughout this article, I use ‘angelification’, ‘angelomorphy’, and ‘angelisation’ (and their variants) interchangeably. I am aware that some Jewish texts understand angelomorphic identity might belong to the righteous in this life, so I would like to make clear that I mean specifically post-mortem angelisation, angelomorphy, or angelification.

8 Origen, Comm. Matt. 27–8 (here 28). (ANF trans.) Cf. also Origen, Princ. 1.8.1.

9 My trans. The Greek text (PG 60:211) reads: Οἱ δὲ εἶπον πρὸς αὐτήν· οἱ δὲ εἶπον, ὅτι Ἄγγελος αὐτοῦ ἐστιν. Ἐκ τούτου ἀληθὲς, ὄτι ἕκαστος ἡμῶν ἄγγελον ἔκει. Καὶ πόθεν ἐπῆλθεν αὐτοῖς ἄγγελον εἶναι τότε ὑπονοῆσαι; Ἀπὸ τοῦ καιροῦ τοῦτο ὑπώπτευσαν.

10 ‘That each of us has angels is found in The Book of the Shepherd and in many places in holy scripture. The Lord said of little children, Their angels always behold the face of my father [Matt 18.10]. Also Jacob referring to himself spoke of the angel who saved me from all evils [Gen 48.16]. And here the disciples believed that the apostle Peter’s angel was coming.’ Bede, Venerable, The Commentary on the Acts of the Apostles (trans. Martin, Lawrence T.; Cistercian Studies Series 117; Collegeville: Liturgical Press, 1989) Google Scholar.

11 The following three subheadings are not meant to communicate that Hellenistic, Jewish, and ‘Christian’ texts are three hermetically distinct contexts but simply a heuristic division to promote clarity in this particular discussion. I am also aware that to label this material as ‘Christian’ is anachronistic, hence the consistent use of scare quotes.

12 E.g., Plato, Euthyphr. 3A; Apol. 31C–D, 40A.

13 This and the following quotations of Plutarch come from De Lacy and Einarson, LCL, with slight alterations.

14 According to the Middle Platonists, daimones are airy and invisible (Apuleius, De deo Socr. 6.2–3; 9.2–4; 11.3–4), located mediately between humanity and the divinity (Plato, Apol. 27C–E; Symp. 202D–203A; Apuleius, De deo Socr. 13.1), and perhaps the souls of the deceased (Plato, Apol. 15 [27B–E]; cf. also Pindar, Ol. 13.105; Hesiod, Works and Days 110–27; Euripides, Alc. 1002; Hipp. 141–50; PGM IV.1475–85) or perhaps not (Plutarch, Def. orac. and Apuleius, De deo Socr. 15–16; cf. also Tertullian, An. 57), with varying degrees of nuance. Andrei Timotin, La démonologie platonicienne. Histoire de la notion de daimōn de Platon aux derniers néoplatoniciens (Philosophia Antiqua 128; Leiden: Brill, 2011) is indispensable on this subject. On the Middle Platonists more generally but also with attention to daimonology, see John Dillon, The Middle Platonists (80 b.c.–a.d. 220) (rev. edn; Ithaca: Cornell University Press, 1996).

15 Already the Stoics conceived of a universal daimonic guardian, but this was much more akin to a conscience than anything external or independently motivated. See Epictetus, Diss. 1.14.12–14; Seneca, Ep. 94.55. I would also like to note that we actually see developments in an opposite direction by the fourth century. Heliodorus (Aeth. 1.26.4) attests to the concept of a personal, evil daimōn that attends to humans. Even so, not every person possesses such a companion.

16 In agreement with a burgeoning minority of scholars, I take Luke-Acts to be a set of second-century compositions. See, e.g., Shellard, Barbara, New Light on Luke: Its Purpose, Sources, and Literary Context (JSNTSup 215; Sheffield: Sheffield Academic, 2002) Google Scholar; Mason, Steve, Josephus and the New Testament (2nd ed.; Peabody: Hendrickson, 2003) Google Scholar; Tyson, Joseph B., Marcion and Luke-Acts: A Defining Struggle (Columbia: University of South Carolina Press, 2006) 14–15Google Scholar; Nasrallah, Laura, ‘The Acts of the Apostles, Greek Cities, and Hadrian’s Panhellenion’, JBL 127 (2008) Google Scholar; Backhaus, Knut, ‘Zur Datierung der Apostelgeschichte. Ein Ordnungsversuch im chronologischen Chaos’, ZNW 108 (2017) CrossRefGoogle Scholar; Matthews, Shelly, ‘Does Dating Luke-Acts into the Second Century Affect the Q Hypothesis?’, Gospel Interpretation and the Q-Hypothesis (ed. Müller, Mogens and Omerzu, Heike; LNTS 573; London: Bloomsbury T&T Clark, 2018) Google Scholar. But it is difficult to place Acts much later than the 130s ce, since Marcion evidently knew Luke’s Gospel, and certainly no later than 150 ce, since Justin almost certainly knew Acts. If we cannot locate an idea’s genesis sometime before, or in, the early second century, it is methodologically difficult to claim that Luke could have known it, if we are thinking genealogically.

17 On the Socratic parallels, see Alexander, Loveday C. A., ‘Acts and Ancient Intellectual Biography’, The Book of Acts in Its Ancient Literary Setting (ed. Winter, Bruce W. and Clarke, Andrew D.; vol. 1; Grand Rapids: Eerdmans, 1993) 31–63Google Scholar, here 57–63; Glover, Daniel B., Patterns of Deification in the Acts of the Apostles (WUNT 2/576; Tübingen: Mohr Siebeck, 2022) CrossRefGoogle Scholar.

18 This angel-as-overseer idea likely explains Revelation’s angels of the seven churches of God (1.20). See Bailey, Jeremiah N., ‘Spheres and Trajectories: The Angels of the Churches (Revelation 1–3) in Context’, Early Christianity in Pompeian Light: People, Texts, Situations (ed. Longenecker, Bruce W.; Minneapolis: Fortress, 2016) Google Scholar, although Bailey refers to the concept somewhat imprecisely as the ‘guardian angel’ interpretation. Relatedly, see the patron angel concept in the Ps.-Clementine Recognitions 2.42. Of course, the concept of the patron angel who stands in for, and represents, particular nations or communities is well-attested by this time (e.g., Deut 32.8 LXX; Ps 82; Isa 24.21; Dan 10.13; cf. Rev 1–3) and is reflected in the Greek and Roman patron deities.

19 On the Athena/Raphael parallels, see Toloni, Giancarlo, ‘Homeric Echoes in Tobit?’, The Story of Tobit: A Comparative Literary Analysis (trans. Hogan, Patrick; JSJSup 204; Leiden: Brill, 2012) Google Scholar.

20 See D. J. Harrington’s translation in OTP 2:297–377, at 372. Similarly, see also T. Jac. 1.10.

21 For a brief yet informative discussion, see Gleason, Randall C., ‘Angels and the Eschatology of Heb 1–2’, NTS (49) 90–107, esp. 97–107Google Scholar, who cites, among other texts, T.Levi 5.3–7; 1 En. 56.5–6; Jub. 15.31–2.

22 See the major, critical commentaries, which pick up this reference and interpretation from Strack and Billerbeck, Kommentar zum Neuen Testament, 2:707.

23 Bock, Acts, 429. So also Barrett, Acts, 1:585.

24 I cite the numeration and translation of H. Freedman, Midrash Rabbah (2 vols.; London: Soncino, 1939). Quotations of the biblical text have been aligned with the NRSV.

25 Roloff, Jürgen, Die Apostelgeschichte (NTD 5; Göttingen: Vandenhoeck & Ruprecht, 1981) Google Scholar. Cited with approval in Pesch, Apostelgeschichte, 1:366 and Jervell, Apostelgeschichte, 335 n. 331.

26 Freedman, Midrash Rabbah, 2:717 (slightly altered).

27 Freedman, Midrash Rabbah, 2:711.

28 Hannah, Darrell D., ‘Guardian Angels and Angelic National Patrons in Second Temple Judaism and Early Christianity’, Angels: The Concept of Celestial Beings – Origins, Development and Reception (ed. Reiterer, Friedrich et al.; Berlin: De Gruyter, 2007) Google Scholar claims that Christians adopted the concept of the guardian angel ‘with very little variation’ from Jewish precedents. Hannah, however, does not entertain the notion that the two concepts developed in tandem, and he fails to distinguish sufficiently between the guardian and patron angels. A Greek patron deity by no means implies a guardian deity, nor are the ideas necessarily connected. I think the Middle Platonic Socratic daimōn is a more compelling history-of-religions genealogy to both the Jewish and the Christian concept. Ahuvia, Mika, On My Right Michael, On My Left Gabriel: Angels in Ancient Jewish Culture (Berkeley: University of California Press, 2021) 88–98Google Scholar follows Hannah on this score.

29 Freedman, Midrash Rabbah, 2:717–18.

30 As for the אלהים whom Jacob wrestled, we are dealing here with an angel, according to the rabbis. But the face that Jacob sees in Esau is like God’s (and not the angel’s) since it stands in judgement against him.

31 Midrash Tanḥuma expands on this tradition: ‘And there wrestled a man with him [Gen 32.25]. It was Samael, Esau’s guardian angel, who wanted to kill him [Jacob].’ Midrash Tanhuma-Yelammedenu (trans. Samuel A. Berman; Hoboken: Ktav, 1996) 213. See the discussion in Novenson, Matthew V., ‘The Universal Polytheism and the Case of the Jews’, Monotheism and Christology in Greco-Roman Antiquity (ed. Novenson, Matthew; NovTSup 180; Leiden: Brill, 2020) CrossRefGoogle Scholar.

32 My trans. Ὁρᾶτε μὴ καταφρονήσητε ἑνὸς τῶν μικρῶν τούτων· λέγω γὰρ ὑμῖν ὅτι οἱ ἄγγελοι αὐτῶν ἐν οὐρανοῖς διὰ παντὸς βλέπουσιν τὸ πρόσωπον τοῦ πατρός μου τοῦ ἐν οὐρανοῖς.

33 Moulton, ‘It Is His Angel’, 516.

34 Moulton, ‘It Is His Angel’, 514.

35 E.g., Roloff, Die Apostelgeschichte, 190: ‘Vorausgesetzt ist dabei die Vorstellung, daß jeder Mensch einen Schutzengel hat, der ihm als sein himmlischer Doppelgänger auch äußerlich gleich.’ Cited with approval in Pesch, Apostelgeschichte, 1:366 and Jervell, Apostelgeschichte, 335 n. 331; see also Haacker, Klaus, Die Apostelgechichte (ThKNT 5; Stuttgart: Kohlhammer, 2019) CrossRefGoogle Scholar.

36 Apocalypse of Paul 49 is sometimes also invoked as evidence for a celestial double. It reads: ‘Are you Paul, the one who is glorified in heaven and on earth?’, The Apocryphal New Testament (trans. J. K. Elliott; Oxford: Clarendon, 1993). The question is very clearly about Paul’s renown and glorification (a similar idea is present in John 17.1–5) and not at all about a heavenly double.

37 So also Marguerat, Actes, 1:437

38 On polymorphism, see Foster, Paul, ‘Polymorphic Christology: Its Origins and Development in Early Christianity’, JTS 58 (2007) 66–99CrossRefGoogle Scholar; Cartlidge, David R., ‘Transfigurations of Metamorphosis Traditions in the Acts of John, Thomas, and Peter’, Semeia 38 (1986) 53–66Google Scholar, though unfortunately without reference to the Shepherd.

39 For such interpreters, see, e.g., Haenchen, Acts 391; Parsons, Acts, 176; Polhill, John B., Acts (NAC 26; Nashville, Broadman, 1992) n. 155Google Scholar.

40 Marguerat, Les Actes, 1:437 poses it as a question: ‘ou croyaient-ils Pierre mort?’

41 In addition to these, the Greek magical papyri include a spell about binding gods/angels to attend to one’s will (PGM I.164–194), but this is a far cry from the content of Acts 12.15c.

42 See esp. Thompson Prince, Deborah, ‘The “Ghost” of Jesus: Luke 24 in Light of Ancient Narratives of Post-Mortem Apparitions’, JSNT 29 (2007) 287–301Google Scholar. Also indispensable are Ogden, Daniel, Magic, Witchcraft, and Ghosts in the Greek and Roman Worlds: A Sourcebook (2nd ed.; New York: Oxford University Press, 2009)Google Scholar and Felton, Debbie, Haunted Greece and Rome: Ghost Stories from Classical Antiquity (Austin: University of Texas Press, 1999), esp. 22–37Google Scholar. Unfortunately, these scholars’ categories are still somewhat imprecise, since they estimate heroification narratives as ghost stories. While these categories probably meet our expectations for ghost stories suitably, a heroification in antiquity is a quite different thing. In addition, both Felton and Prince distinguish ‘translated mortals’ as a discrete category, but in the ancient Mediterranean theological imaginary, a translated mortal is simply a hero.

43 Cf. also Vergil, Aen. 2.770–5; Phlegon of Tralles, Mir. 2.

44 Everett Ferguson, Backgrounds of Early Christianity (3rd ed.; Grand Rapids: Eerdmans, 2003) 171.

45 OCD4, s.v.

46 At other times, genius was glossed by τύχη.

47 On daimonification among the Middle Platonists, see David Litwa, M., Posthuman Transformation in Ancient Mediterranean Thought: Becoming Angels and Demons (Cambridge: Cambridge University Press, 2021) 57–72Google Scholar.

48 Dillon, Middle Platonists, 174.

49 For the relevance of the Middle Platonist tradition for understanding the Book of Acts, see esp. Hubbard, Jeffrey M., ‘Paul the Middle Platonist? Exegetical Traditions on Timaeus 28c and the Characterization of Paul in Acts 17.16–31’, HTR 115 (2022) CrossRefGoogle Scholar.

50 The translation is that of Colson and Whitaker, LCL (slightly alt.).

51 Cf. the parallel discussion in Cher. 114 as well as Somn. 1.137–9; QE 2.29, 40. See now Yli-Karjanmaa, Sami, Reincarnation in Philo of Alexandria (SPhiloM; Atlanta: Society of Biblical Literature, 2015) esp. CrossRefGoogle Scholar.

52 My trans.

53 My trans. The Greek text (Colson and Whitaker, LCL) reads: Ψυχὰς οὖν καὶ δαίμονας καὶ ἀγγέλους ὀνόματα μὲν διαφέροντα, ἓν δὲ καὶ ταὐτὸν ὑποκείμενον διανοηθεὶς ἄχθος βαρύτατον ἀποθήσῃ δεισιδαιμονίαν.

54 On angelification in Philo, see Litwa, M. David, ‘The Deification of Moses in Philo of Alexandria’, SPhiloA 26 (2014) Google Scholar; idem, Posthuman Transformation, 73–94.

55 Demons were often visible in other texts, however, even philosophically informed ones. See, e.g., Philostratus, Vit. Apoll. 4.10.2.

56 My trans.

57 Dillon, Middle Platonists, 47.

58 As for instance in the memorial inscription CIL 6.3, which depicts a heroized child returning to comfort his uncle; Ovid, Fast. 2.499–512, which depicts Romulus appearing to Julius Proculus; and Philostratus, Her. 11.1–2, which reports how Protesilaus appears to, speaks with and even touches a vinedresser. Most (though crucially not all) heroizations are preceded by some form of heavenly translation, and we do not find that in Acts 12.15. It is not impossible to imagine, however, that the believers presumed Peter’s heavenly assumption, though the text does not say it explicitly.

59 Smith, Jonathan Z., Drudgery Divine: On the Comparisons of Early Christianities and the Religions of Late Antiquity (Chicago: University of Chicago Press, 1990) .Google Scholar

60 E.g., Gen 38.8–10; Deut 25.5–10; Ruth 3.9–4.2. See Fletcher-Louis, Crispin H. T., Luke-Acts: Angels, Christology and Soteriology (WUNT 2/94; Tübingen: Mohr Siebeck, 1997) Google Scholar.

61 David Litwa, M., ‘Equal to Angels: The Early Reception History of the Lukan ἰσάγγελοι (Luke 20.36)’, JBL 140 (2021) Google Scholar. His discussion of John Chrysostom is also important (619–20). Here, Wirkungsgeschichte is more helpful than in the case of Peter’s supposed guardian angel above. In that case, there was no evidence that such a guardian angel concept was available to Luke and his earliest readers, whereas angelisation, on the other hand, was a concept both available to and frequently discussed among Luke’s Jewish contemporaries. Because Luke’s very ambiguous (to us) term ἰσάγγελοι was associated with angelisation so early and so uniformly, and because Luke’s Jewish milieu frequently spoke of angelisation, we have good warrant to think that is the interpretation most readily available to Luke and his earliest readers.

62 See e.g., N. T. Wright, The Resurrection of the Son of God (Christian Origins and the Question of God 3; Minneapolis: Fortress, 2003) Google Scholar. In addition, Wright asserts that ‘Few supposed, so far as we know, that human beings would become angels in that future life’ (418, emphasis original). But see M. David Litwa, Posthuman Transformation, esp. 45–114.

63 Early Jewish beliefs about the state or quality of the body in the resurrection varied drastically. See Chilton, Bruce D., Resurrection Logic: How Jesus’ First Followers Believed God Raised Him from the Dead (Waco: Baylor University Press, 2019Google Scholar).

64 On the Enochic literature, see Mach, Michael, Entwicklungsstadien des jüdischen Engelglaubens in vorrabbinisher Zeit (Texts and Studies in Ancient Judaism 34; Tübingen: Mohr Siebeck, 1992) 163–208Google Scholar; Schäfer, Peter, Two Gods in Heaven: Jewish Concepts of God in Antiquity (trans. Brown, Allison; Princeton: Princeton University Press, 2020) esp. 99–133CrossRefGoogle Scholar. On the DSS, see Crispin H.T. Fletcher-Louis, All the Glory of Adam: Liturgical Anthropology in the Dead Sea Scrolls (STDJ 42; Leiden: Brill, 2002). For a more general survey, see Somov, Alexey, Representations of the Afterlife in Luke-Acts (LNTS 556; London: T&T Clark, 2017) Google Scholar.

65 On this passage as presenting Luke’s soteriological vision of theosis, see Kuecker, Aaron J., ‘“You Will Be Children of the Most High”: An Inquiry into Luke’s Narrative Account of Theosis’, JTI 8 (2014) Google Scholar.

66 We could also add Luke’s phrase ‘the sons of light’, whom he contrasts with the ‘sons of this age’ (Luke 16.8), two phrases redolent of the DSS. Light is also a quality Luke associates with angels (e.g., Luke 2.9) and with Jesus (Luke 2.32; 9.31–2; Acts 9.3–4), one of only two other figures named a son of God in Luke-Acts (e.g., Luke 1.35; 4.3–9, 41; 8.28; cf. also Adam in 3.38).

67 In keeping with Luke’s larger interest in realised eschatology, Karlsen Seim, Turid, The Double Message: Patterns of Gender in Luke-Acts (SNTW; London: T&T Clark, 1994) 185–248Google Scholar reads Luke as perceiving the angelic life breaking into the present, akin to the already-not-yet status of God’s kingdom.

68 On such stares in Acts and their significance, see Strelan, Rick, ‘Strange Stares: Atenizein in Acts’, NovT 41 (1999) Google Scholar.

69 Baumbach rightly perceives the connection with death/martyrdom, but he interprets Stephen’s angelic visage only as an indication of his coming demise and not the state achieved thereby. ‘In Apg. 6,15 wird zwar von Stephanus gesagt “sein Angesicht war wie das Antlitz eines Engels,” aber diese Charakterisierung gilt Stephanus als Märtyrer, der als solcher schon auf der Schwelle zum ewigen Leben steht.’ Baumbach, Günther, Das Verständnis des Bösen in der synoptischen Evangelien (Theologische Arbeiten 19; Berlin: Evangelischer Verlagsanstalt, 1963) Google Scholar.

70 The textual authenticity of Jesus’ last words in Luke 24. 34 is disputed. Parsons, Mikeal C. and Whitlark, Jason, ‘The “Seven” Last Words: A Numerical Motivation for the Insertion of Luke 23.34a’, NTS 52 (2006) 188–204Google Scholar argue they are a later scribal insertion for numerological reasons. I am not yet convinced by their argument, however, and most scholars have not followed them in removing them from the reconstructed Ausgangstext.

71 This reading does not necessarily preclude an intertextual one that sees in Stephen’s face an imitation of Moses’ descent from Sinai, but I believe such a reading is less cogent.

72 Fletcher-Louis’ (Luke-Acts, 102–3) reading is not wholly improbable. He argues that Stephen’s claim that ἐλάβετε τὸν νόμον εἰς διαταγὰς ἀγγέλων refers not to the Law’s ordination or command by angels, as scholars typically interpret the phrase. Instead, the ‘primary sense [of διαταγή] is quite clearly that of an “ordered arrangement” which can then be applied to an individual’s lifestyle…The natural sense of Acts 7.53 διαταγὰς ἀγγέλων is therefore “the constitution(s) of angels… In other words the law was given in order to create a community whose lifestyle(s) and order are, or conform to, that of the angels”’. This interpretation does not explain, however, why, on my reading, Stephen’s visage perceptibly changes (εἶδον τὸ πρόσωπον αὐτοῦ ὡσεὶ πρόσωπον ἀγγέλου, Acts 6.15), and Fletcher-Louis notes the absence of history-of-religions parallels to this interpretation (105). I believe something more than a manner of life is in view, though perhaps it is an ontological or physical transformation in consequence of Stephen’s way of life.

73 In addition to the texts discussed here, see also Luke 15.7, 10.

74 Since Luke’s knowledge of the Pauline epistles is disputed, I will not hang anything on this coincidence. Although it is debated, several scholars have persuaded me that Paul sees the resurrected body as pneumatic in the Corinthian correspondence. See Martin, Dale, The Corinthian Body (New Haven: Yale University Press, 1995) Google Scholar; Engberg-Pedersen, Troels, Cosmology and Self in the Apostle Paul: The Material Spirit (Oxford: Oxford University Press, 2010) 26–31CrossRefGoogle Scholar; David A. Burnett, ‘A Neglected Deuteronomic Scriptural Matrix for the Nature of the Resurrected Body in 1 Corinthians 15.39–42’, Scripture, Texts, and Tracings in 1 Corinthians (ed. Linda L. Belleville and B.J. Oropeza; Lanham: Lexington Books/Fortress Academic, 2019) 187–211.

75 On the other hand, Luke’s use of τὰ ἀμφότερα (usually meaning ‘both’) may suggest that he views at least two of these three elements – usually ‘angel’ and ‘spirit’ – as coterminous. In such a case, the Sadducees disbelieve two (related) concepts: resurrection and angel/spirit.

76 Luke seems not to believe angels are pure πνεῦμα either since he describes them appearing like men (Luke 24.4, 23; Acts 1.10).

77 Daube, ‘Acts 23’, 493–7 argues that angelification is a frequent characteristic of the interim state between death and resurrection in apocalyptic Judaism. Hence, the denial is not that the Sadducees disbelieve in angels, which populate the Pentateuch, but in the angelification of the righteous dead preceding the resurrection. So also Sullivan, Kevin, Wrestling with Angels: A Study of the Relationship between Angels and Humans in Ancient Jewish Literature and the New Testament (AGJU 55; Leiden: Brill, 2004) CrossRefGoogle Scholar; Segal, Alan F., Life After Death: A History of the Afterlife in Western Religion (New York: Doubleday, 2004) Google Scholar. Finally, see Fletcher-Louis, Luke-Acts, 58–9, who perceptively points out that, in context, Paul’s proclamation of his encounter with Jesus’ resurrection in 22.6–11 and his subsequent epiphany (22.17–21) could provide the basis for the Pharisees’ speculation that a ‘spirit or angel’ may have spoken to him (as Jesus does again in 23.11).

78 We may even have evidence here that Luke imagines a two-stage afterlife: the first consisting of angelisation/spiritualisation and the second of resurrection. On a similar two-stage afterlife in 2 Baruch, see Karlsen Seim, Turid, ‘The Resurrected Body in Luke-Acts: The Significance of Space’, Metamorphoses: Resurrection, Body and Transformative Practices in Early Christianity (eds. Seim, Turid Karlsen and Økland, Jorunn; Berlin: De Gruyter, 2009) 19–39CrossRefGoogle Scholar. This interpretation would bring these passages into greater conformity with Luke’s parable of Lazarus and the rich man (Luke 16.19–31), which refers to some spiritual post-mortem states – Lazarus is ‘carried away by the angels’ (16.22a) while the rich man descends to Hades (16.23a) – as well as resurrection (ἐκ νεκρῶν ἀναστῇ, 16.31). It also makes sense of Luke’s emphasis on isangelisation as well as Jesus’ resurrected flesh and bones, on which, see Matthews, Shelly, ‘Elijah, Ezekiel, and Romulus: Luke’s Flesh and Bones (Luke 24:39) in Light of Ancient Narratives of Ascent, Resurrection, and Apotheosis’, On Prophets, Warriors, and Kings: Former Prophets through the Eyes of Their Interpreters (eds. Brooke, George J. and Feldman, Ariel; BZAW 470; Berlin: De Gruyter, 2016) Google Scholar.

79 There is likely a Dionysian parallel here between Rhoda’s ‘mania’ and Peter’s escape from the jail’s ‘automatic’ gates, but it is beyond the gambit of this essay. See Moles, John, ‘Jesus and Dionysus in “The Acts of the Apostles” and Early Christianity’, Hermathena 180 (2006) 65–104Google Scholar.

80 Daube, ‘Acts 23’, 496.

81 See esp. Mart. Pol. 2.3 (cf. 12.1); cf. Acts Paul 3.3.