appointed to the episcopate cannot account for the Church being called a virgin. Valois cuts the knot by emending to μεxpì ToÚтov. Heinichen prefers to follow Stroth, and talks of the badness of Hegesippus' style, which is perhaps hardly fair.' But a solution lies close at hand, so obvious that I can scarcely imagine that no one has hitherto suggested it. What is to hinder us from supposing that Eusebius has omitted a passage, not relevant to his immediate purpose, between B and the words dià Touro? This, indeed, would not be in accordance with his usual method of citation. When he quotes two passages from an early writer which are not consecutive, he usually introduces the second with some such phrase as roúrOLÇ SE μεθ ̓ ἕτερα ἐπιφέρει λέγων. But this is not an overwhelming difficulty. Nor is it unreasonable to suppose that Eusebius was guilty of so quoting Hegesippus as to leave his expressions without meaning. He does the same thing with other writers.3 But our conjecture will be made much more probable if we can supply, from other parts of the History, the passage, or a considerable part of it, which we suppose Eusebius to have here omitted. This we shall now endeavour to do.

It is first necessary to rescue for Hegesippus two fragments which are not obviously his. The first is that which election of Symeon to

is paraphrased in H. E. iii. 11, on the the episcopate; the second, a passage similarly treated in

the following chapter. The eleventh chapter begins with the words: Μετὰ τὴν Ἰακώβου μαρτυρίαν καὶ τὴν αὐτίκα γενομένην ἅλωσιν τῆς Ἱερουσαλὴμ, λόγος κατέχει τῶν ἀποστόλων καὶ τῶν τοῦ κυρίου μαθητῶν τοὺς εἰσέτι τῷ βίῳ λειπομένους ἐπὶ ταὐτὸ πανταχόθεν συνελθεῖν. What is the λόγος here referred to ? According to the latest, and I believe the best, English translation of the History, it is no more than an oral tradition, for the phrase λóyoç karέɣɛ is rendered “it is said." But this is incorrect. As Lightfoot has observed, the expression λóyoç Karéxt "is not confined to oral tradiλόγος κατέχει tion, but may include contemporary written authorities, and implies authentic and trustworthy information." have myself collected a number of instances of the use of this and similar phrases from the History, which completely corroborate this remark. In the majority of cases. where Eusebius introduces a narrative with the words λόγος (κατέχει, the document on which he relies is either indicated in the immediate context, or may be discovered by a search through the passages from previous writers scattered through his pages.

1 That of Mr. M'Giffert, in Nicene and Post-Nicene Fathers, vol. i.

Ignatius, vol. i., pp. 58, 238.

3 H. E. ii. 7; 17, §§ 6, 19; 22, § 2 (see §§ 3−7); iii. 18, § 1 (see §§ 2, 3); 19; 32, § 1 (see § 2); 36, § 3 (see § 7: the passage cited does not, of course, prove that Ignatius was actually martyred, though, judging from H. E. iv. 16, it is not impossible that Eusebius thought otherwise); iv. 5, § 1 (see § 2: I regard the statement that the bishops were short-lived as an inference from the number of names in the written lists-the diadoxai of v. 12); iv. 28 (see 29, § 3); v. 5, §§ 1, 2 (see §§ 3-7); vi. 28 (?).

H. E. iii. 20, § 11 (see 23, § 6, and cf. 18, § 3; 23, § 3); 37, § I (see v. 17, § 3); v. 10, § 1 (see vi. 19, § 13, where,

I

Only a few instances of the

however, it is not stated that Pantaenus was a Stoic. Observe that λóyos exeɩ is here, as it seems, contrasted with paσí). In H. E. iii. 24, § 5 we have the statement, depending on λόγος κατέχει, that Matthew and John wrote their Gospels "of necessity" (éπávayкes). It is possible that Eusebius intended λóyos κατέχει το cover only his assertion about St. Matthew. For when in § 7 sqq. he recounts a story of the origin of St. John's Gospel, for which no earlier authority is known, he refers, and apparently with some emphasis, to common report as the evidence for what he tells (paσí, §§ 7 bis, 11). His assertion about St. Matthew is scarcely more than a fair inference from extracts which he gives elsewhere from Papias (iii. 39, § 16), Irenæus (v. 8, § 2),

phrase remain, in which it does not seem possible to name the document referred to,' and in none of these is the use of documentary evidence excluded, or improbable. It may be regarded therefore as much more likely than not that in H. E. iii. 11 Eusebius is not merely reporting a current tradition, but paraphrasing a document."

To the conclusion that this document was the Memoirs of Hegesippus several facts at once point. In the first place, Hegesippus is named in the context.

At the end

of the chapter we are told: τὸν γὰρ οὖν Κλωπᾶν ἀδελφὸν τοῦ Ιωσὴφ ὑπάρχειν Ηγήσιππος ἱστορεῖ. This is either a refer ence to a passage of Hegesippus not now extant, or, as seems less likely, a loose citation of words already quoted (B). But again the words μετὰ τὴν Ἰακώβου μαρτυρίαν are a paraphrase of the expression with which Hegesippus introduces his statement, that Symeon was elected bishop of Jerusalem-μετὰ τὸ μαρτυρῆσαι Ἰάκωβον. And lastly, the statement that the martyrdom of St. James was

and Origen (vi. 25, § 4). That it was made by Papias in so many words, in the passage of which no more than the two concluding sentences are now preserved (iii. 39, § 16), is far from incredible.

1 H. E. i. 12, § 3; ii. 17, § 1; vi. 4, § 3 (perhaps referring to a letter of Origen, as in the next sentence, & TOU Onoìv avtós); vii. 12; 32, § 6; viii. 6, § 6; App. § 1. In ii. 1, § 13 reference is, no doubt, made to Irenæus, Adv. Hær. iii. 12, § 8, while for the statement in v. 19, § there can be no question that a written list of bishops of Antioch was the voucher.

2 This becomes clearer if the connexion between chapters 11 and 12 is noticed. The more probable view of the construction of the latter is that it depends on Xoyos Kaтéxeι at the beginning of chapter 11. But the statements of the two chapters are quite

HERMATHENA-VOL. XI,

C

heterogeneous. It is almost impossible to suppose that they are referred to as different parts of a single tradition: there is no difficulty in imagining them to be taken from different sections of the same book.

* I find that Zahn (p. 238) uses this as an argument on the other side. He writes: Da Eus. den Heg. ausdrücklich nur für eine ergänzende Bemerkung als Gewährsmann anführt, kann nicht dieser allein die einzige Quelle sein.' The edge of this reasoning seems to be turned, if we suppose the statement in question to come from a different context from that which Eusebius paraphrases in the previous sentences. But Zahn admits that chaps. 11, 12 are probably ultimately derived from Hegesippus, which is all that we are concerned to maintain. See also Hort, Judaistic Christianity, p. 170.

immediately followed by the sack of Jerusalem is Hegesippean. Eusebius follows two main authorities for the martyrdom, Josephus and Hegesippus. The former dates it as having taken place between the death of Festus and the arrival of his successor Albinus (H. E. ii. 23, § 21 sqq. · Jos. Ant. xx. 9, § 1), i.e. A.D. 61. This date Eusebius adopts in his Chronicon, where he assigns the murder to An. Ab. 2077 = 7 Nero. But Hegesippus, in words quoted above (A), puts it immediately before either the Jewish war or the siege of Jerusalem.' This is the date given in the passage before us.

=

There is thus high probability that throughout the chapter Eusebius is engaged in reproducing, in the oratio obliqua, a fragment of Hegesippus. In what part of the Memoirs did it occur? I answer, with very slight hesitation, as the immediate sequel of our fragment B. It is almost impossible to imagine that that fragment gives the whole of what Hegesippus had to say about Symeon's election. The phrase πρʊélevтo mávreç at least needs some explanation, and iii. 11 gives exactly the explanation which is required. It defines άvres as the surviving Apostles and disciples of the Lord, it narrates how they all assembled together and deliberated, and arrived at a unanimous decision. And further, if this passage immediately followed B, μετὰ τὴν Ἰακώβου μαρτυρίαν is no longer a paraphrase of a chance expression of Hegesippus in a different context, but of one which occurred in a passage which actually lay before Eusebius as he wrote. The statement about the connexion between the martyrdom and the siege of Jerusalem, again, is taken from the end of A, and confirms the conclusion already reached, that if

1 It was plainly in the latter sense that Eusebius understood the puzzling words of Hegesippus, καὶ εὐθὺς Οὐεσα πασιανὸς πολιορκεῖ αὐτούς; for he im

mediately afterwards paraphrases them, τῆς παραχρῆμα μετὰ τὸ μαρτύριον αὐτοῦ πολιορκίας τῆς Ἱερουσαλήμ.

A and B were not consecutive, they were, at least, not far apart. Thus we may write out part of H. E. iii. 11 as the third in our sequence of fragments from the Hegesippean Memoirs, enclosing it in brackets, since it is only known to us in paraphrase. How much of the latter part of it is Hegesippus, and how much the comment of Eusebius, I do not venture to pronounce.

C. H. E. iii. 11:

[(λόγος κατέχει) τῶν ἀποστόλων καὶ τῶν τοῦ κυρίου μαθητῶν τοὺς εἰσέτι τῷ βίῳ λειπομένους ἐπὶ ταὐτὸ πανταχόθεν συνελθεῖν, ἅμα τοῖς πρὸς γένους κατὰ σάρκα τοῦ κυρίου· πλείους γὰρ καὶ τούτων περιῆσαν εἰσέτι τότε τῷ βίῳ· βουλήν τε ὁμοῦ τοὺς πάντας περὶ τοῦ τίνα χρὴ τῆς Ἰακώβου διαδοχῆς ἐπικρῖναι ἄξιον ποιήσασθαι. καὶ δὴ ἀπὸ μιᾶς γνώμης τοὺς πάντας Συμεῶνα τὸν τοῦ Κλωπᾶ, οὗ καὶ ἡ τοῦ εὐαγγελίου μνημονεύει γραφὴ, τοῦ τῆς αὐτόθι παροικίας θρόνου ἄξιον εἶναι δοκιμάσαι, ἀνεψιόν γε, ὡς φασὶ, γεγονότα τοῦ σωτῆρος. τὸν γὰρ οὖν Κλωπᾶν ἀδελφὸν τοῦ Ἰωσὴφ ὑπάρχειν (Ηγήσιππος ἱστορεῖ).]

The way is now opened for a consideration of the short chapter which follows-H. E. iii. 12. This chapter, consisting of a single sentence, is, like the one with which we have been dealing, in the oratio obliqua. It has been debated whether it depends on κατέχει λόγος at the beginning, or on 'Ηγήσιππος ἱστορεῖ at the end, of chap. 1. For us the question is purely syntactic. In either case, if our reasoning has been correct, it is an indirect quotation from Hegesippus. We must return to it hereafter. For the present I only remark that the opening phrase κai ἐπὶ τούτοις seems to mark it as not continuous with the passage paraphrased in chap. 11.2

We have now succeeded in replacing one part of the

1 The third alternative, suggested by Zahn (p. 238), that the construction depends on ὥς φασι, seems highly improbable.

2 Cf. H. E. i. 8, § 16; ii. 1, § 14'; iii. 32, § 7, &c. But see also iv. 15, § 15. Compare Heinichen ad loc.