Jump To ECM Entry Online Commentary Introduction
GC: Reading f can be excluded because of its weak and incoherent attestation. It has probably developed several times independently out of reading a by homoioteleuton. This assessment is confirmed by considering further witnesses cited in TTMk.
There are early citations of Origen that may show variant f arose very early. On the other hand, we have to consider that the works cited stress the fact that the OT prophesy was fulfilled by Jesus, so the omission may be due to Origen himself.
Only readings a and b are under consideration here. An earlier range of A-related witnesses supports b, 037 and several A-related minuscules support a, the majority reading.
This variation unit is one of several places where 03 and 05 agree in the omission of a small word against the majority of witnesses. Traditionally, their agreement has been seen as ancient and valuable, but there are incoherencies in the attestation of b, υἱοῦ θεοῦ, in contrast to the good coherence of a, υἱοῦ τοῦ θεοῦ.
The opening lines of the Gospels were popular as amulets, but they do not contribute decisive evidence. T10 reads υἱοῦ θεοῦ (along with 03 and 05). T4 has a lengthier addition, υἱοῦ τοῦ θεοῦ υἱοῦ ᾽Αβραάμ. The omission (variant f) is found in T21 (with the genitive article before Χριστοῦ).
Most of the versions are witnesses for the longer readings (a, b, d, the Ethiopic for a, b, c, d since it cannot differentiate between κυρίου and θεοῦ). Only L:Ialt. K: Smss. CPA: CL attest the omission.
TP: It is usually considered more likely that the article before θεοῦ was added than omitted. This argument is not very strong, however, because the usual Greek rendering of “the son of God” would be ὁ υἱὸς τοῦ θεοῦ, although variant d is attested by only two late minuscules. In the NT the form with both articles prevails, while there are several more examples for the anarthrous form (Matt 14:33, 27:43, 27:54; Mark 15:39, Luke 1:35, John 19:7). For the form with an article with θεοῦ only there is only one further example in the NA28 text: John 10:36. In Mark 1:1 the MT has the form least common in the NT.
In favor of the shortest reading, the omission in variant f, Head puts forth several arguments, namely that it is “supported by earlier and more diverse patristic witnesses;” that there is a “wide spread of Greek witnesses to the short reading from the second century onwards” while there is a “total absence of Greek witnesses to the long reading until c. AD 400;” and that an “addition of ‘Son of God’ to the short text is more probable than its omission from the long text,” not least because “one does not ... expect errors due to tiredness in the first verse of a work.” Strutwolf, however, shows that the patristic evidence, including Head’s main witness Irenaeus, is at least ambiguous. In addition, both France and Williams make the case that “Son of God” is intrinsically probable, arguing that the confessions in 8:29 and 15:39 are based on these opening words. The first hand of 01 is the earliest Greek manuscript witness in a completely incoherent attestation. The incoherence is in fact a sign of accidental omission for which homoioteleuton, or simply the graphical similarity of subsequent nomina sacra, provide an obvious cause. Head’s reasoning against the probability of encountering this phenomenon at the beginning of a writing is disproved by variant e, τοῦ θεοῦ.
Other advocates of the shorter text are Ehrman, who interprets the presence of υἱοῦ (τοῦ) θεοῦ as an anti-adoptianist addition, and Collins. The conclusions of Collins underscore that two particular arguments are decisive: the secondary addition of υἱοῦ (τοῦ) θεοῦ appears more probable than deliberate omission, and that its accidental omission is unlikely. IrLat should be taken seriously as the earliest witness for the longer text, but caution must be exercised since the citations occur in a work against heresies. The context of at least one of these citations of the long text, however, shows no sign of Christological polemics.
 On the pre-genealogical coherence of the attestation based on the TTMk evidence see Wasserman, “Historical and Philological Correlations and the CBGM;” and Wasserman and Gurry, A New Approach, 43-50.
 Cf. Strutwolf, “Remarks on the Patristic Evidence.”
 de Bruyn and Dijkstra, “Greek Amulets,” 172.
 Cf. Greeven (in Textkritik, 42) referring to Globe (“The Caesarean Omission,” 217). Greeven himself opts for variant f: “die Auffüllung lag zu nahe” (in Textkritik, 43).
 Head, “A Text-Critical Study of Mark 1.1,” 624–25, 626, 627, 629.
 Strutwolf, “Remarks on the Patristic Evidence,” 76-77. Cf. Wasserman, “The ‘Son of God’,” 26-32, about the ambiguity of Greek patristic evidence for the shorter text.
 France, Mark, 49; Williams, Mark, 18–19.
 Ehrman, Orthodox Corruption of Scripture, 72–75; Collins, “Establishing the Text,” and Mark, 130–32.
 Collins, “Establishing the Text,” 125.
 Cf. Strutwolf,”Remarks on the Patristic Evidence,” 76-77.