led very much farther than that when we follow it systematically.

Once grant the conception of a composite document, framed by the accretion of earlier documents, selected and arranged by an editor, and it becomes obvious that we may expect to find some traces of the original divisions. If, further, we have three such composite documents, based on similar, and sometimes the very same, materials, it is obvious that their agreements and differences will indicate the various limits of the component parts. No other postulates than these are required. The processes of this structural criticism will be best understood by seeing its practical application to the actual problem of the synoptic gospels. The gospel of John is clearly supplementary to the others, and hence it cannot be treated with them in our methodic process. We here deal only with the three synoptic gospels.

6. The Classes of Episodes.

On comparing the synoptic gospels, we see that they are composed of many different classes of material. Quite irrespective of the

sense, the mere position alone, as compared with the position of the similar passage in another gospel, may vary in many different ways. To study the position and structure, we should regard the separate episodes solely as units labelled by chapter and verse, equivalent to other units in other gospels. We must only regard their comparative positions; and we must not consider their meaning until their purely structural relations are explored.

In considering the structure, there is not only the presence of any episode to be noted, but the sequence of it with other episodes. It may stand in the same relative order toward surrounding episodes, or toward large masses of separated episodes, or toward a scattered chain of episodes. And these relations may occur in any two or all three gospels.

The first and most obvious course is to remove from each gospel every episode which does not occur in the same order in both of the other gospels. Everything that is not in common to all three in a parallel text, should be set aside. Thus there remains a body of episodes which is identical in order in all three gospels. This must be the Nucleus or common

But

basis on which each gospel has been built. It may have been larger originally, if any gospel has omitted matter from the Nucleus. It cannot have been smaller, as it is very unlikely that additional episodes would have been inserted at the same places in the Nucleus by separate writers. The Nucleus very probably grew out of shorter documents; and in some cases we can judge that a passage breaks the thread of idea and has probably been inserted, or that a natural close can be detected. such personal estimates cannot be taken into account at this stage. Here we are only observing that a definite stage of documentary accretion was reached where the Nucleus passed as a whole, in unvaried order, into different lines of growth leading to several gospels. The utmost variation in sequence that may be tolerated is the inversion of two verses or short episodes by a literary author like Luke; their juxtaposition is fair proof that their places were already fixed and that we have not to deal with a random insertion.

The second class of documents after the Nucleus are the episodes, which only occur in two gospels in the same relation to the Nucleus.

Here we have apparently the work used in common by two writers, but unknown to the third, or used by him quite independently in another connection. In some cases such episodes may have been parts of the Nucleus, but have been deliberately omitted by a third writer. But considering that all the evangelists largely expanded the Nucleus, it is very unlikely that they should have omitted any part of it.

The third class of documents are those groups of episodes which occur in the same relation one to another, but are inserted at different parts of the preceding structure by the writers who use them. Their relative order together being the same, though in different gospels, proves that they belong to a single document, which was adopted and used by different writers independently.

The fourth class is a small one, of scattered episodes or short sayings, which may be detected in the same order in different gospels, although widely separated. They are apparently due to a document having been utilised by extracting sayings from it and inserting them one after another in appropriate places.

All of the preceding classes are probably from documentary sources, as the exact sequence of the portions of them is their distinctive mark.

The fifth class is of isolated episodes or sayings which are never found in the same immediate connection or in the same scattered order in any two gospels. From their positions they are more probably derived from oral tradition, or from single detached notes and memoranda which did not preserve any connection of sequence between different episodes. The sixth class is the residuum of the material which is only found in a single gospel, and on which comparative structure has nothing to say.

It will be seen that in such a classification of the material there is no room for personal opinion. The facts arrange themselves so soon as these successive stages of sorting are applied. We have an absolutely impersonal critical engine which will produce exactly the same result, whoever may be the operator.

Thus there is obtained a firm platform for subsequent historical discussion and for more personal judgment of detail.

B