For all that lies beyond his personal experience the historian is, of course, dependent on sources, documentary or otherwise. The modern writer recognises the duty of testing and verifying the accuracy of the sources he uses for his narrative, and in producing his own account of affairs he is expected, where desirable, to state the sources upon which he has relied. The ancient historian also made use of sources, but (1) he used them uncritically, with little or no anxiety concerning their accuracy, and (2) it was his custom simply to select from the available material any passages, long or short, even words or phrases, which served his purpose, and to incorporate these in his work, frequently without any indication of the borrowing. Only in certain instances was the source precisely referred to. Moreover (3) the utmost freedom was exercised in dealing with the passages thus chosen. Sometimes they were reproduced word for word; at other times they were partially or wholly transformed to suit the new context. This may seem an unwarrantable procedure to us, but one has only to examine the actual instances of these adaptations or transformations of unnamed sources to perceive that the ancient¹ writer has acted in perfect good faith, with no suspicion that the manipulation was in any way blameworthy. How indeed could it have been otherwise? The science of literary criticism was unknown, “notions of literary propriety and plagiarism had not been thought of, and writers who advanced no pretensions to originality for themselves were guilty of no imposture when they borrowed without acknowledgement from their predecessors” (Skinner, _Kings_, p. 7). ¹ Nor need one go back to antiquity for an instance. Most instructive examples of composite narrative compiled uncritically but quite innocently by mediaeval chroniclers from earlier sources may be found in Chapman’s _Introduction to the Pentateuch_ (in this series), pp. 260 ff. Compare also an illustration from Arabic historical writings given by A. A. Bevan in _Cambridge Biblical Essays_, pp. 12 ff.