Chapter 2

      That such a mass of literature, much of it of great interest and high merit, should be to a great extent anonymous is very remarkable, considering how strong the force of tradition was in Iceland. It is true that in the middle ages there was a carelessness as to the exact authorship of literary works, to an extent that seems surprising at the present day, but nowhere does it appear to have been so prevalent as in Iceland. The explanation presumably lies in the fact that so much of the written matter had its origin in the oral narratives which had been transmitted in a set form for several generations. The first writers of these traditions probably did not add much of their own to the story as they had received it, and therefore saw no good grounds for claiming the title of authors. In other words, the ink and parchment were at first little more than a substitute for the human memory, and the skill of the teller or reader was still of more importance than the art of the writer. Another reason may be found in the way in which Icelandic writings were freely altered and adapted by any one who wished to copy them or utilize them for his own purposes. Icelanders of the thirteenth and fourteenth centuries were constantly abridging or expanding, combining or interpolating, re-arranging or re-writing, the works of their predecessors, and it would often have been impossible to assign the name of any single author to the form which they finally assumed. In some cases we know the names of the men who produced the existing recensions of a work, while that of the original writer has been completely lost.
      In the case of sagas which have a purely fictitious basis, the subject-matter gives but little clue to the date of their composition, or to the part of Iceland in which they were written. The sagas relating to Iceland itself reveal a little more on both of these heads, but not infrequently they also present features which render a precise answer impossible. As will be seen later, the sagas dealing with the older period-the saga-age proper represent the west and north of Iceland in a far greater degree than they do the east and south; the proportion is actually something like ten to two. It is not difficult to infer from this that most of them were therefore written in the west and north, seeing that so much in the stories themselves has a strong local interest, and would naturally be most fully preserved in the district, and in the families, to which the leading persons belonged. A close examination of a saga often confirms this, to the extent of absolute certainty. The unknown author may plainly indicate not only his district, but even his own part of that district either by a minute knowledge of the locality, or by the adverbs of direction which be employs. It is often easy to perceive that outside of a certain area his knowledge of places, and of their relative positions and distances from each other, is vague, while within that area he is familiar with every foot of the ground. The same local knowledge is often displayed in the genealogies which occur in most of the sagas; the author is well informed in the history and relationships of certain families, while as to others his knowledge is limited or inaccurate. In some cases it is not difficult to conjecture to which family he himself belonged.
       These frequent genealogies are among the most useful indications which the sagas give as to the date of their composition; but in using them for this purpose some caution is necessary, as nothing was more likely to be inserted by the copier of a saga than a later name in the family tree. The relative dates of different sagas may also sometimes be inferred from the fact that one cites another as its authority for a special piece of information, or as giving a fuller account of some episode. Here, however, there are two complications to be considered. It is not always certain whether the references are to a written saga, or to the current oral version; and such citations might very easily be inserted by later copyists. Hence the most diverse views as to the date of a saga have sometimes been based on the very same passages, scholars holding them to be mere interpolations in an early text, while others contend that they are integral parts of the narrative and prove that it is late.
       To some extent also, the date of a saga may be inferred from its style, and from its manner of telling the story. The simpler and purer the style, and the more straightforward the narrative, the earlier it is likely to be; a more elaborate diction, and tendency to a romantic tone and colouring, are indications of a period when the influence of foreign romances had begun to be felt. Again, however, there is the difficulty of deciding whether the text in the latter case is the original one, or has been worked over by a later hand, in order to bring it more into accordance with the prevailing taste. In some cases it is quite obvious that this has happened, for both the earlier and later versions of the saga are preserved, in whole or in part, and thus the process of conversion to the more romantic and less thoroughly national form can be clearly traced.