Clement of Alexandria I: Protrepticus and Paedagogus

necessary. Thus, the article is inserted at 7, 30; 11, 19; 15, 1; 50, 15; 62, 3; 71, 30; 150, 9; 219, 4; 220, 16; 226, 1; 245, 16; 272, 24. A relative pronoun was inserted by Arethas at 17, 18; 24, 5; 28, 27; 36, 7; 38, 17; 54, 15; 69, 9 (twice); 77, 3; 157, 27; 166, 2; 220, 6; 221, 16. Through this, asyndetic lacking conjunctions coordinate clauses were intended to be made into subordinate clauses, and thus a better sentence structure created. Admittedly, through these additions, the exact opposite was achieved in some places and the sentence structure was disrupted; see, for example, 17, 18; 54, 15; 157, 27; 166, 2. And it is very unlikely that the author's spelling was restored in the remaining places. For Clement loves precisely this choppy style with many short sentences without conjunctions; one need only read the beginning of the Protrepticus! The insertion of conjunctions such as γὰρ for 43, 2; 167, 28; 169, 21; 192, 28; 195, 5; 210, 30; 286, 23; δὲ but/and 77, 31; 149, 18; 182, 19; 207, 18; 264, 24; οὖν or γοῦν therefore/at least 162, 15; 195, 28; 228, 16; 269, 15; καὶ and 19, 30; 151, 3; 172, 6; 188, 23; 226, 23; 290, 20; διὸ καὶ on account of which/therefore 154, 14; ὅτι that 8, 31; ὡς as 165, 17; τε and 172, 18; ἢ or 231, 7; μὲν indeed 238, 13 had the same purpose as the insertion of relative pronouns. If we consider 1) that such additions were also frequently made by scribes or readers of manuscripts (e.g., 175, 6 + γὰρ P^3; 185, 4f + γὰρ M^2; 196, 23 + ὅτι F^2), 2) that some of the additions are obviously wrong, e.g., δὲ 77, 31 and 149, 18; διὸ καὶ 154, 14; τε 172, 18; καὶ 188, 23 and 226, 23, 3) that otherwise a γὰρ, γοῦν or μὲν (for the latter cf. e.g., 276, 26; 284, 13) is also often missing in Clement where one would expect it, we will be inclined to see in these additions improvements by Arethas himself rather than corrections based on a manuscript.

Now, however, alongside these additions, we also find others of a different kind. One may also regard a portion of them as conjectures by Arethas—which are several times incorrect; to these I count ἄνδρα man 9, 31; ὅρα look/behold 33, 17; οὐκ not 51, 8; 66, 15; οὐδ’ nor 79, 15; οὐδὲ neither 50, 22; μὴ not 178, 21; δι’ ὃν εἴρηται through whom it has been said 195, 7; εἰς into 172, 3; διὰ through 163, 20 (the addition shows that the entire passage was not understood); διὰ through 189, 30; καὶ ὑμεῖς and you 284, 21; ἐν τούτῳ in this 284, 25. But with some variants and additions, one will hardly be able to manage without the assumption of the use of a manuscript; cf. e.g., σύμφωνος τῷ λόγῳ concordant with the Word/Reason 150, 28; ἀλλὰ καὶ ὁ ἐσθίων φασίν but also the one eating, they say 160, 28; τοῦτον εἰ καὶ him, even if and οὐκ εἰς ἐκμέλειαν καὶ ῥᾳθυμίαν μεταχωρεῖν δεῖ one must not pass over into lack of harmony and laziness 149, 21f (here the additions destroy the construction, but they are also not to be understood as an interpolation); οὐδὲ neither and εἰ μὴ if not 204, 28 (the passage is not really understandable either with or without the additions); καὶ ἄγειν μὲν ἐπίπροσθεν τὸ and to bring forward the »ὁ ἐλεῶν πτωχὸν δανίζει θεῷ« the one who has mercy on the poor lends to God »καὶ τὸ and the 254, 2 (the addition is dispensable, but precisely for this reason