The combination of narrative themes is so frequent a phenomenon in folk and formal literature that one almost forgets to wonder at it. Yet in point of fact the reason for it and the means by which it is accomplished are mysteries past our present comprehension. If we could learn how and where popular tales unite, if we could formulate any general principle of union or severance, we should be well on the way to an understanding of the riddle which has hitherto baffled all students of narrative, namely, the diffusion of stories. We have theories enough; our immediate need is for more studies of individual themes, careful and, if it must be, elaborate discussions of many well-known cycles. Happily, these are accumulating and give promise of much useful knowledge at no distant day.

One principle has become clear. Since motives are so frequently found in combination, it is essential that the complex types be analyzed and arranged, with an eye kept single nevertheless to the master-theme under discussion. Collectors, both primary and subsidiary, have done such valiant service that the treasures at our command are amply sufficient for such studies, so extensive, indeed, that the task of going through them thoroughly has become too great for the unassisted student. It cannot be too strongly urged that a single theme in its various types and compounds must be made predominant in any useful comparative study. This is true when the sources and analogues of any literary work are treated; it is even truer when the bare motive is discussed.

The Grateful Dead furnishes an apt illustration of the necessity of such handling. It appears in a variety of different combinations, almost never alone. Indeed, it is so widespread a tale, and its combinations are so various, that there is the utmost difficulty in determining just what may properly be regarded the original kernel of it, the simple theme to which other motives were joined. Various opinions, as we shall see, have been held with reference to this matter, most of them justified perhaps by the materials in the hands of the scholars holding them, but none quite adequate in view of later evidence. The true way to solve the riddle appears to be this: we must ask the question,—what is the residuum when the tale is stripped of elements not common to a very great majority of the versions belonging to the cycle? What is left amounts to the following,—the story reduced to its lowest terms, I take it.

A man finds a corpse lying unburied, and out of pure philanthropy procures interment for it at great personal inconvenience. Later he is met by the ghost of the dead man, who in many cases promises him help on condition of receiving, in return, half of whatever he gets. The hero obtains a wife (or some other reward), and, when called upon, is ready to fulfil his bargain as to sharing his possessions.

Nowhere does a version appear in quite this form; but from what follows it will be seen that the simple story must have proceeded along some such lines. The compounds in which it occurs show much variety. It will be necessary to study these in detail, not merely one or two of them but as many as can be found. Despite the bewildering complexities that may arise, I hope that this method of approach may throw some new light on the wanderings of the tale.

Of my debt to various friends and to many books, though indicated in the body of the work, I wish to make general and grateful acknowledgment here. My thanks, furthermore, are due to the librarians of Harvard University for their courteous hospitality; to Professor G. L. Kittredge for his generous encouragement to proceed with this study, though he himself, as I found after most of my material was collected, had undertaken it several years before I began; and to Professor R. K. Root for his help in reading the proofs.