Add to My Library 
Mark 1:1 reads as follows: jArchV tou' eujaggelivou jIhsou' Cristou' uiJou' qeou' (“The beginning of the gospel of Jesus Christ, the Son of God.”) The last two words in Greek are disputed because of their omission in some important manuscripts. These same manuscripts cannot be charged with being the products of heretics, however, because in 15:39 they all record the centurion as saying, “Certainly this was God’s Son.” The issues at stake must be put on a different plane.
a* Q (28) and a few others omit uiou qeou, while virtually all the rest of the witnesses have the words. Although normally the shorter reading is to be preferred by most textual critics, this rule cannot be applied mechanically. In this case, if a good reason for an accidental omission can be found—especially since the MSS lacking the words are very few, then the longer reading probably should be regarded as authentic. And although a is a major witness to the text of the New Testament, the first corrector of this MS added the words uiou qeou, suggesting the possibility that the omission was simply an oversight. Apart from this lone fourth century MS, the rest of the Greek testimony is quite late, coming approximately 500 and 700 years later. To be sure, the shorter reading is found in Origen and a few other early patristic authors, but the very fact that these writers seem to be using the verse for their own purposes, rather than commenting on the whole of the text, may imply that ‘God’s Son’ simply did not fit into their particular objective (see Cranfield’s commentary on Mark for some insights here). Further, the reading in question is a compound nomen sacrum following immediately after another compound nomen sacrum. That the words could have been omitted by accident is quite likely, since the last four words of v 1, in uncial script, would have looked like this: iucruuuqu. With all the successive upsilons an accidental deletion is likely. Further, the inclusion of uiou qeou here finds its complement in 15:39, where the centurion claims that Jesus was uio" qeou. This Christological inclusio that encompasses the whole of the Gospel finds parallels in both Matthew (“Immanuel… God with us” in 1:23/ ”I am with you” in 28:20) and John (“the Word was God” in 1:1/”My Lord and my God” in 20:28), probably reflecting nascent Christological development and articulation.
But even more can be said: tou' eujaggelivou jIhsou' Cristou' uiJou' qeou' is one of only eighty-three places in the NT in which four or more words in a row end in –ou; of these, only twenty texts have five or more words in a row (besides Mark 1:1, cf. Matt 7:5; 9:20; 14:36; Mark 6:56; Luke 8:44; Acts 6:13; 12:12; Rom 1:3; 1 Cor 1:9; Gal 2:20; Phil 3:8; Col 2:2; 1 John 3:23; 2 John 3; Rev 9:13; 14:10; 15:7). There are only two texts in which as many words end in –ou as in Mark 1:1 (1 Cor 1:9 has seven in a row, while Rev 14:10 has nine in a row). To be sure, there are other places in which a string of genitives occur (e.g., Matt 1:1 [7]; Rom 1:29 [5]; Heb 11:32 [5]; 1 Pet 1:1 [5]), but these do not all end in –ou. An examination of the multiple –ou texts reveals the following textual variation statistics:1 ten of the twenty quintuple –ou texts—exactly half!—show omissions, substitutions, etc. that break up the multiple –ou construction. And of the 83 quadruple or more –ou texts, Sinaiticus breaks up the sequence ten times (cf., e.g., Acts 28:31; Col 2:2; Heb 12:2; Rev 12:14; 15:7; 22:1)—or twelve percent of the time! There is thus a significantly higher possibility of accidental scribal omission due to homoioteleuton (similar ending words) in such a MS. This argument, however, depends for its strength (to some degree at least) on the supposition that a’s ancestry may reach back early enough that the nomina sacra would have been written in their uncontracted form. Thus, although we might charge the lineage of which Sinaiticus is a part with mild carelessness here, we would at the same time be affirming its roots very deep into the second century, perhaps even earlier. This is because all of our New Testament MSS have nomina sacra—even the ten Greek papyri that are now dated in or very close to the second century.
In conclusion, in light of the slim pedigree for the omission, coupled with the high probability of homoioteleuton here, as seen in other multiple —ou passages, as well as intrinsic evidence, it is most likely that Mark wrote ‘God’s Son’ in the opening verse of his gospel. But the strongest argument that Sinaiticus accidentally omitted these words actually becomes evidence for the great antiquity of its form of text. More study of course needs to be done, but this textual problem may help point to the antiquity of a text’s Vorlage by the kinds of errors found in that MS.
1 The evidence is preliminary and based on partial data to date: NA27, the text of Codex Sinaiticus, Hoskier’s text of the Apocalypse. More data can only make the case against the omission stronger.