100 Andrew Wilson fied his canon by excepting small omissions and cases of homoeoteleu- ton26. Similarly, Dirk Jongkind argued that “perhaps Royse comes close to misinterpreting Griesbach”27, in that most of the omissions noted in Royse’s study fell into the categories of Griesbach’s exceptions. However, Silva’s own study shows the problem with this objection. Silva’s results cast serious doubt on Griesbach’s rule that “the shorter reading … is to be preferred to the more verbose, for scribes were much more prone to add than to omit”28, a rule for which Griesbach offered no textual evidence. Further, one word omissions account for at least 55 (81%), and possibly more29, of Silva’s 68 omissions, meaning that we can hardly view Griesbach’s short omissions exception clause as valid either, for to classify 81% of omissions (the dominant error) as an excep- tion requires reinventing the meaning of the word “exception”. We might equally say that all Europeans are Spaniards (with a few exceptions) or that sheep are naturally carnivorous but in exceptional circumstances nibble grass. The fact that the majority of Royse’s evidence falls within Griesbach’s exception clauses is precisely why Royse is right: Griesbach’s canon is more honoured in its breach than in its observance. In the present study, as can be seen from the table below, when singu- lar readings were analysed according to the length of the text omitted, one and two word omissions constituted 67% and 14% of all omissions, respectively, amounting to 81% of omissions. Short omissions are there- fore not an exception, but, in large part, the rule itself. The attempt to defend the canon by appealing to the short omissions “exception clause” would not appear to be viable. Length (words) 1 2 3-7 8+ Totals Add 769 (71%) 167 (15%) 122 (11%) 30 (3%) 1088 (Dittography) (53) (20) (17) (8) (98) Omit 1143 (67%) 246 (14%) 230 (13%) 93 (5%) 1712 (Homoeo-teleuton) (237) (65) (87) (47) (436) (Homoeo-arcton) (68) (36) (44) (22) (170) Ht or Ha (%) 27 41 60 74 35 26 See B.M. Metzger’s translation of Griesbach’s first rule in The Text of the New Testa- ment: Its Transmission, Corruption, and Restoration (New York – Oxford 1992) 120. 27 D. Jongkind, Scribal Habits in Codex Sinaiticus (TS 3rd series, 5; Piscataway, NJ 2007) 139. 28 Quoted from Metzger’s translation of Griesbach’s Prolegomena to the second edition of his Greek New Testament (1796-1806) in The Text of the New Testament, 120. 29 Silva’s figures for addition and omission in the table on page 24 do not tally with the figures provided earlier in his study. This could possibly be because the category in the table marked Miscellaneous includes some additions or omissions. However, Silva provides three different figures for omissions in