12 Stephen W. Frary scribe? That is, which is most difficult as a scribal creation, for it is more likely to be original. In the case of 1 Tim 3:16a, this question represents the disputed ballot of textual criticism, for it requires the interpreter to anticipate the motive as well as the behavior of the copyist, and can yield various results. For example, a scribe whose theology rules his pen might easily modify the original text to Θ̅C̅ in anti-Arian fervor, elevating one of the other two variants to the status of lectior difficilior. On the other hand, a grammar-conscious scribe has two paths open. Normally, the gender of the antecedent and the pronoun agree; thus J.A.T. Robertson notes: “...the true text ὅς is changed in the Western class of documents to ὅ to agree with μυστήριον,”32 making ὅς the more difficult reading, hence original. Elliott, on the other hand, holds that the “mystery of godliness” is in fact Christ, a concept “not alien to N. T. theology (cf. Col 2:2 and possibly 1 Cor 2:1.).”33 This relegates 1 Tim 3:16a to the category identified by Robertson which contains phrases in which a pronoun (ὅς) represents “the real gender [of the antecedent] rather than the grammatical.”34 In this case ὅ would be more difficult since the tendency would be to make the pronoun masculine in agreement with Χριστὸς. Interestingly, though, this does not seem to happen elsewhere in the Pauline corpus. Col 1:27 refers to Christ as a mystery among the gentiles: τὸ πλοῦτος τῆς δόξης τοῦ μυστηρίου τούτου ἐν τοῖς ἔθνεσιν, ὅ ἐστιν Χριστὸς ἐν ὑμῖν, ἡ ἐλπὶς τῆς δόξης. Although the neuter “mystery” is the antecedent to the relative pronoun, the real subject is the masculine “Χριστὸς,” yet the pronoun is the neuter “ὅ” Col 4:3 reads: τὸ μυστήριον τοῦ Χριστοῦ, δι’ ὃ καὶ δέδεμαι. If Χριστοῦ is taken here both as an epexegetical genitive and as the nearest antecedent, again there is no gender agreement with the pro- noun in the relative clause following. If the pattern, then, is for the relative pronoun not to assimilate to the gender of the antecedent when Christ is described as a mystery, the harder reading of 1 Tim 3:16a becomes “ὅς.” Clearly, the judgment of which is the more difficult reading is in this case subjective. Elliott’s final question would have us consider which reading is con- sonant with the thought and style of the author and makes best sense in context. In the present passage this presents the immediate challenge of determining authorship and extends beyond the issue of Pauline author- J.A.T. Robertson, A Grammar of the Greek New Testament in the Light of Historical 32 Research (Nashville, TN 1934) 713. Elliott, The Greek Text of the Epistles, 59. 33 Robertson, Grammar, 713. 34