New Testament textual critics are accustomed to accept it as a rule that the shorter of two competing readings is more likely to be the true one. This rests on the assumption that the change which produced the variance of reading in different manuscripts was editorial, not accidental. Accident may omit; an addition is usually the result of deliberate alteration. A classical philologist, however, like the Corpus Professor of Latin at Oxford who writes this book, is more familiar with accident than intention as the cause of corruption in ancient texts, and therefore brings a fresh eye to these New Testament questions.Taking accident as a more important element than Westcott and Hort, or most modern critics, have believed, Professor Clark holds that many of the readings of Codex Bezae and the "Western Text" (which are often longer than those of the Vatican and Sinaitic codices) are probably original. His argument is (1) that, as the papyri show, books in the early centuries were frequently written in narrow columns of ten to twelve letters; (2) that both the Sinaitic and the Vatican manuscripts show many undoubted omissions, where, as is made plain by counting the letters, it is probable that one or more short lines of an ancestor have been dropped; (3) that hence it is likely that many other cases of the same accident took place at an early time; and (4) that these can be detected by counting the number of letters in the disputed readings.It is indeed the fact that a large number of the additional phrases in the "Western Text" can be explained in this way, and it must be admitted that New Testament critics have too much neglected the significance of this form of accident as a cause of variation. Especially when similarity of ending of two words ("homoioteleuton") appears, the facts presented by Clark are important. It must be said, however, that in the case of the longer passages, like Mark 16 9-20, which he defends as genuine, the evidence from the coincidences between the numbers of letters contained in the passages is not at all convincing, and it will not do to assume that long and important verses and sections omitted by accident from one early copy would have continued absent from texts used in widely distant churches for long periods. In the Book of Acts the theory breaks down, for here the multiples of ten to twelve letters fail to appear consistently, and in any case a large degree of intentional editing (whether by way of enlargement or of diminution) has to be admitted.The essay collects valuable material, should stimulate to further study, and is a needed reminder of important considerations often overlooked. It illustrates that welcome contributions to textual criticism may be expected from many quarters; but also that no single line of observation, however brilliantly conceived and patiently carried out, will lead to a general solution of a problem which has been complicated by the intricate working of many forces.