First Person Narrative in Acts 27-28

Tyndale Bulletin 36 (1985) 79-109.

FIRST PERSON NARRATIVE IN ACTS 27-28

By Colin J. Hemer

I. INTRODUCTION

The voyage--narrative of Acts 27-28 is a key passage

for the interpretation of the 'we-passages' in Acts. It

has traditionally been a strong buttress of the view

that the writer was a companion of Paul, or at the least

that he used as his source the diary of such a companion.

But recent historical studies of the voyage have been

sparse, and mostly directed to long-standing debates on

points of detail.1 The focus of much recent study, here

as elsewhere in Luke-Acts, has been on literary and theo-

logical interests.2

The object of this paper is not to get embroiled

further in this discussion, beyond a preliminary sampling

of the range of opinion and some necessary clearing of

the ground. In the course of preparing a larger study of

Acts I have found this passage a test-case of alternative

______

1. R. M. Ogilvie, 'Phoenix', JTS n.s. 9 (1958) 308-314;

A. Acworth, 'Where was St. Paul Shipwrecked? A

Re-examination of the Evidence', JTS n.s. 24 (1973)

190-193; O. F. A. Meinardus, 'Melita Illyrica or

Africana: An Examination of the Site of St. Paul's

Shipwreck', Ostkirchliche Studien 23 (1974) 21-36, and

'St. Paul Shipwrecked in Dalmatia', BA 39 (1976) 145-

147; C. J. Berner, 'Euraquilo and Melita', JTS n.s. 26

(1975) 100-111, in response to Acworth. Only the last

is much concerned with the interpretation of the pass-

age in its wider context (cf. BJRL 60 [1977-81 41-42).

Most recently N. Heutger,'"Paulus auf Malta" im Lichte

der maltesischen Topographies', BZ 28 (1984) 86-88 arg-

ues for a relocation in Malta.

2. E.g., P. Pokorny, 'Die Romfahrt des Paulus and der

antike Roman', ZNW 64 (1973) 233-244; V. K. Robbins,

'The We-Passages in Acts and Ancient Sea-Voyages',

BibRes 20 (1975) 5-18, and 'By Land and By Sea: The

We-Passages and AncientSea Voyages', Perspectives

on Luke-Acts, ed. C. H. Talbert (Edinburgh: T. & T.

plark, 1978) 215-242; G. B. Miles and G. Trompf,

'Luke and Antiphon: The Theology of Acts 27-28 in Light

of Pagan Beliefs about Divine Retribution, Pollution

and Shipwreck', HTR 69 (1976) 259-267; D. Ladouceur,

'Hellenistic Preconceptions of Shipwreck and Pollution

as a Context for Acts 27-28', HTR 73 (1980) 435-449;

Susan Marie Praeder, 'Acts 27:1-28:16: Sea Voyages

in Ancient Literature and the Theology of Luke-Acts',

CBQ 46 (1984) 683-708.
80 TYNDALE BULLETIN 36 (1985)

approaches. I intend therefore to make some interim

observations on the character of the passage, and to offer

some samples of the kinds of documentation which seem to

me of material weight in my plea to put the discussion on

a rather different footing.

Most of the recent studies will not detain us here.

P. Pokorný sets the passage against a mystery-romance

background; G. B. Miles and G. Trompf, followed in part

and modified by D. Ladouceur, draw on a background in the

Attic orators Antiphon and Andocides to link peril at

sea with divine vengeance, and to take the preservation ofthe

accused as affording a presumption of innocence admiss-

ible as an argument of probability in a secular Athenian

dicastery-court. This narrative, then, at a focal climax

of Luke's work, is to be seen as a vindication of Paul

before a higher court than that of Caesar. Ladouceur

himself concedes (p.441) that there is no evidence that

such an immunity would be taken seriously in a first-

century Roman court, and his invocation of a vindica-

tory significance in the mention of the Dioscuri (Acts

28:11)as the figurehead designation of the new ship

(discussed by him at length on pp. 443-448) does nothing

to ease my difficulty. The glimpse these studies afford

of ancient popular thought are themselves most interest-

ing, but they do not provide a convincing background here.

I shall offer below a very different kind of parallel for

the reference in Acts 28:11.

More crucial to our study are those approaches which

affect our understanding of the function of the 'we-

passages'. The theological reconsideration of the voyage

in its bearing on this question stems effectively from

M. Dibelius, who separates it from the other 'we-passages',

and treats it as a pre-existing, literary narrative into

which references to Paul have been inserted.3 The

commentaries of H. Conzelmann and E. Haenchen pursue the

trail blazed by Dibelius. In the detailed treatment by

Haenchen one is particularly aware of an ambivalence bet-

ween his insistence that the narrative is tailored to a

glorification of Paul and his tacit acceptance of many

______

3. M. Dibelius, Studies in the Acts of the Apostles,

ed. H. Greeven (London: SCM, 1956) 204-206; tr. by

M. Ling from the German Aufsätze zur Apostelgeschichte

(Göttingen: Vandenhoeck and Ruprecht, 1951).

HEMER: First Person Narrative in Acts 27-28 81

details which are hard to dissociate from an account of

personal experience. I am not concerned for the present

to interact in detail with these approaches, though they

are much in mind in the writing of these pages.

A very specific problem is raised by V. K. Robbins,

in his contention that first person plural narrative was

characteristic of an ancient sea-voyage genre. While his

central interest is focused, especially in his article in

Perspectives on Luke - Acts, on literary questions and

thence on the theological significance of orientating

early Christianity towards the sea that leads to Rome

(Perspectives 41), the implication is that this explan-

ation disposes of any lingering notion that the form

could indicate personal participation in the event.

Robbins' work has not been without influence,4but I

find no sufficient reason for accepting the existence of

this first-person voyage-genre. His examples are not

necessarily representative, nor are they always taken

correctly in context, nor are they subject to control,

nor do they prove the conclusions he draws from them.

The first person plural is used with the same kinds

of semantic variation in Greek as in English, and differ-

ent usages even in the same passage are not necessarily

excluded. Neither language possesses a distinction like

that of Malay/Indonesian between 'inclusive' and 'exclu-

sive' pronouns; so that if in Acts 16:17 and 21:18 the

phrase 'Paul and we' separates him conceptually from the

first person group, no difficulty need be found in the

fact.5 Again, there is the easily illustrated 'authorial'

first person, whether singular (sometimes less a sign of

egotism than a caution that subjective opinion is not

attested fact) or plural (possibly trying to involve the

______

4. Cf., e.g., Schuyler Brown, The Origins of Christianity.

A Historical Introduction to the New Testament (Oxford:

OUP, 1984) 27-28.

5. In each case he phrase marks a transition ending a

'we-section', before the continuing account of Paul

is narrated in the third person. The subsequent resum-

ption of 'we' in the one case at Philippi, in the

other in Palestine, is consistent with a companion's

residence in those regions while Paul himself was resp-

ectively travelling and imprisoned. Cf. F. F. Bruce

The Acts of the Apostles (London: Tyndale, 19522) 315,

391.

82 TYNDALE BULLETIN 36 (1985)

reader in a tacit bond of sympathy with the author).

These two categories, and indeed fluctuations between

them, I find constantly pervading my own unstudied

speech and writing no less than Greek or other texts.

There is also a very natural 'we' which reflects a

writer's solidarity with his own place, time or nation.

In the calculatedly detached, third-person narrative of

Caesar's Commentaries the Romans are nostri (our men).

'In our time' is freely used for 'contemporary', and

Mare Nostrum (our sea) is the Mediterranean.6 The

common factor in these variations is their inclusion of

reference to the speaker/writer himself. If that is not

the case in ancient voyage narratives, the onus surely

lies on the advocate to establish his argument under

rigorous controls. The first obvious instances I happen-

ed to check (Caes. BG 4 23-24, 28; 5. 8; Lucian, Navig.

7-9) were all in the third person, and Lucian actually

puts his voyage in third person reported speech in a

first person dialogue context. Of course such narratives

are often first person accounts, because they recall

personal experience, and plural because they recall

communal experience. The same tendency is as true of

colloquial English as of literary Greek (or Latin), but

it is no proof of the existence of a literary style

appropriate towhat was not personal experience.

Apart from his many examples of first and third

person narration, Robbins offers three more precise

parallels with Acts. (1) The Voyage of Hanno 1-3.7

The two opening sentences are in the third person, and

the remainder of the document in the first person plural.

But paragraph 1 is a formal heading, recording briefly

______

6. Cf. in nostro mari (Caesar BG 5.1); also nostrum litus

of the MediterraneanCoast of Syria (Plin. NH 6.30.126);

Sall. Jug. 17.4; etc. So in Greek Scylax, Periplus

Maris Interni 40, in Geographi Graeci Minores (GGM) ed.

Carolus Mullerus (Karl Mueller) (Paris: Firmin Didot,

1855) Vol. 1, p.39: πρός πὴν ἐπὶ ἡμῶν θάλασσαν

(?4th BC); Marcianus of Heraclea, Periplus Maris

Exteri 3: ἡ καθ’ ἡμᾶς αὕτη θάλασσα (GGM 1.519) (4th-

5th AD); etc. Also e.g. apud maiores nostros (Cic.

2 Verr. 2.47.118); for nos of time cf. Hor. Odes 306.

46-48; Tac. Agric. 2.3; etc.; nostri anni (Οv. Fast.

1.225).

7. See the text in K. Mueller, GGM I, pp. 1-14. It begins

ἕδοξε Καρχηδονίοις Ἅννωνα πλεῖν ἔξω Στηλῶν Ἡρακλείων

. . . καὶ ἔπλευσε. The first sentence of the actual

narrative has the verb ἐπλεύσαμεν.

HEMER: FirstΠerson Narrative in Acts 27-28 83

the explorer's commissioning. His report begins at para-

graph 2, and is all in the 'we'-form, not as a literary

device for a fiction, but because he reports on the act-

ual adventures of his party. Paragraph 1 should be

printed as a prefatory paragraph, as it is by K. Mueller,

not as part of a continuous undifferentiated narrative,

as it is by Robins. (2) A papyrus narration of some

incidents in the Third Syrian War.8 Robbins says that

column I line to II.11 contains third person narration,

which shifts to first person plural in II.12 as a sea

voyage is narrated. But there is a difficulty in assess-

ing the context: the first half of every line in the

first column is lost, and no continuous sense can be

reconstructed. Yet the surviving part of line 18 contains

a first plural (παρ’ ἡμῶν) comparable with καθ’ ἡμᾶς in

II.13 (not 12), to which Robbins attaches special signif-

icance. The real transition comes in II.16, where L.

Mitteis and U. Wilcken restore an emphatic ἡμεῖς δέ. The

point throughout is that this is a narrative of conflict

between 'us' and 'them', the Ptolemies and the Seleucids,

narrated by a participant on the Ptolemaic side. Where

the 'enemy' are at sea (II.2-3), their voyage is re-

counted in the third person, but Robbins' citation only

begins at II.5, and misses the interaction of first and

third persons which can be traced throughout the document,

so far as columns I, III and IV are preserved, alike in

land and sea episodes of the campaign. (3) The Antiochene

Acts of Ignatius.9 This is much the most difficult and

elusive case. There is certainly an abrupt and unmarked

shift to the first person plural in mid course. J. B.

Lightfoot (pp. 83-391) is severe on the evident histor-

ical flaws of this account which seems to be composite

and very late. But it is precisely the 'we-section',

allied to an eyewitness profession and to its intrinsic

plausibility an lack of the demonstrable blunders app-

arent elsewhere, which leads him to entertain the poss-

ibility that this part contains authentic tradition. In

any case the document as a whole does not further

______

8. First published by J. P. Mahaffy, The Flinders Petrie

Papyri (Dublin: Academy House, 1893), Part 2, No. 45,

pp. 145-149. There is an improved text in L. Mitteis

and U. Wilcke , Grundzüge und Chrestomathie der

Papyruskunde, Vol. 1, Part 2 (Leipzig: Teubner, 1912),

No. 1, pp. 1-7. The account deals with the events of

246 BC: the author is not identified.

9. J. B. Lightfoot, The Apostolic Fathers, Part 2, Vol. 2

(London: Macmillan, 18892) 477-495 (text and comment-

ary), 575-579 (translation), 383-391 (critical discu-

ssion).

84 TYNDALE BULLETIN 36 (1985)

Robbins' thesis. As it is probably both late and compos-

ite, it is at best uncertain material for arguing literary

intention. Moreover, as it stands, the preceding part of

the voyage (where this document contradicts the authentic

letters) is rendered in the third person, and the 'we-

passage' (which has better credentials) begins at sea but

is largely devoted to leave-taking in Rome. The martyr

is distinguished from those ostensibly present with him.

Robbins' many other examples are open to criticism

on various similar grounds. There is wide variation

throughout ancient literature in the use of the first

(and indeed the second)10 person, but it occurs within

the flexible confines of natural usage. The Odyssey and

the Aeneid certainly use a technique of flashback first-

person narration, but this is part of the larger struct-

ure of the poems, and not confined to the limits of a

voyage motif. The same is true of a more specific sample

from a very different genre, the Hellenistic romance,

where Achilles Tatius' hero Clitophon tells his story as

a first person narration within a first person framework.

In 2.31.6 and 3.1.1 'we' denotes Clitophon and Leucippe

and their companions, and continues the pronoun used of

the same party travelling by land in 2.31.4-5, which is

not cited by Robbins. 4.9.6 is in direct speech, part

of a lament in which Clitophon apostrophises his suppos-

edly dead love and recalls their shared experiences. In'

ostensibly autobiographical literature, whether fact

(Jos. Vita 3.15) or fantastic fiction (Lucian VH 1.5-6),

the whole is structured on a first person narrative,

which becomes plural not only at the outset of a voyage

______

10. The second person would give no less rich a harvest

of rhetorical and poetical turns, to which however

I should not attach significance. In Latin poetry,

for instance, especially in the extreme metrical

stringency of Ovidian elegiacs, it is commonplace

to work intractable words and names into the line

by apostrophising a god or a goat (Ov. Fast.1.354,

357, 360; etc.). A simple example may be offered

from among the minor geographers from whom Robbins

draws examples: the versifier Dionysius Periegetes

(GGM 2.103-176) begins with the authorial 'I' (line

3), has a second person invocation of the Muses (62),

addresses the reader (1053-4, 1080) from standpoint

of authorial 'I' in 1054, and in conclusion address-

es the continents and islands (1181-1183) before

ending again in the first person (1184-1186).

HEMER: First Person Narrative in Acts 27-28 85

but wherever the writer is identified with a group. The

same is true of the personal narrative in Dio Chrysostom

(Or.7.2, 10). In the former passage the writer sails

with some fishermen, and 'we' reverts to 'I' when his

companions leave him: in the latter the plural continues

while he travels by land with a companion (ἐβαδίζομεν).

The same objection applies to Petronius Sat.114-115;

again the full context is crucial. Some of the poetic

examples are even less satisfactory. The example from

Ovid's personal lament in exile (Trist. 102.31-34) depends

on one first plural verb form, but ignores the commonplace

of Latin verse by which ‘we’ stands freely for 'I' metri

gratia. A glance at the poem shows the use of the first

plural in lines 16. 38, 67 and 70, all in non-maritime

contexts, whereas the nautical imagery of 75-84 happens

to contain only the singular, and 17-18 mixes the numbers,

but the meaning in every case is 'I'.

It will only weary the reader to pursue this kind

of analysis. I have chosen here to treat mainly literary

examples offered by Robbins from dates near to that of

the New Testament. His extension of the argument into

the unpretentious geographical compilers of varied and

often very uncertain dates does not strengthen his case.

The like phenomena are illustrated in them, often in a

naively unliterary way. Scylax uses the 'authorial' 'I'

baldly, beginning his work ἄρξομαι, and signalling the

ends of digressions to describe islands with a repetitive

ἐπάνειμι. He speaks of 'our sea' (40; GGM 1.39), but

such uses are no more significant than his slipping from

his impersonal catalogue into addressing the reader (ἐὰν

προέλθῃς ἀπὸ θαλάττης ἀνώτερον, 100, GGM 1.74). We cannot

attach importance to the brief lapses into the first

plural in the Periplus Maris Erythra 20 (GGM 1.273), nor

in ibid. 57 (GGM 1.299), where κατὰ τὸν καιρὸν τῶν παρ’

ἡμῖν is again of a familiar type followed by μέχρι καὶ

νῦν, immediately eelow. Varied trivial phenomena of this

kind may be abundantly illustrated from literary and

other documents which Robbins does not mention.11

______

11. Thus the impersonal compilation of Agathemerus,

Geographiae Informatio (GGM 2.471-487) announces a

new section ith a sudden lapse into authorial 'we'

with reference to the islands of 'our' sea (λοιπὸν

δὲ ἐροῦμεν τῶν καθ’ ἡμᾶς νήσων τοὺς περιμέτρους, 5.

20, p. 481) too the anonymous Compendium of

Geography (GGM 2.494-511) slips into the first plural

at 4.5 (p.49 ).

86 TYNDALE BULLETIN 36 (1985)

Of course, nothing I have said disposes of the fact

that voyage-narratives are often couched in the 'we'-

form, but I contend that this is a natural tendency

dictated by the natural situation, not an artificial

literary device. If the narrative is fiction anyhow

(as in Lucian, Achilles Tatius or Heliodorus), the

'we' still functions naturally within the dramatic

dimension of the fiction. Indeed, the examples under

discussion are drawn from widely differing genres (in a

more usual sense of that word), and the notion that an

exclusively defined Gattung can be isolated by simple or

composite verbal or syntactical criteria across a wide

variety of prose and poetry of different types and

languages seems to me inherently suspect.12 The paradigm

does not work, and it cannot be used to draw larger con-

clusions about the narrative of Acts 27-28.

______

12. A notable instance is the development of the idea

of the 'diatribe' in the sense developed from the

doctoral thesis of Rudolf Bultmann, Der Stil der

paulinische Predigtund die kynisch-stoische Diatribe

(Göttingen, 1910). Bultmann's concept of the diatribe,

which differs from the understanding of the term

among classical scholars, looks like a composite

construct whose diagnostic characters are assembled

piecemeal from a complex of stylistic parallels

represented among a range of ancient writers of

ostensibly different genre, Greek and Latin, prose

and, poetry, philosophy, satire and rhetoric. H. D.

Jocelyn goes so far as to deny altogether the exist-

ence of the 'diatribe'. He claims to have traced

the origin of this fashion of talking: it first sur-

faced in H. Usener, Epicurea (Leipzig, 1887) lxix.

He concludes 'The term should disappear from scholar-

ly discourse along with all the other bogus antiquit-

ies which the moderns use to adorn their essays on