that by such means the act of understanding performs itself, as it were, by the eye and the flash of apprehension runs automatically through the mind. A good example of this explanation through the eye is the diagram from Darwin (page 183), to aid in the explanation of what he calls "rather a perplexing subject." He is showing "how this principle of benefit being derived from divergence of character, combined with the principles of natural selection, tends to act;" and to do this he had to get into the mind of his reader something like a panoramic scheme of a very large body of varieties and species and genera: for the moment, however, all that he cared about was their common tendency to divergence. If he had named each of these classes, even with the minimum of description, before we came to the end, we should have forgotten what it was all about. If, on the other hand, he could cut each of them down to this single aspect of divergence, he could give us the present bearing of these facts with an immense saving of attention. For this purpose he naturally used a diagram; and the bare lines of his diagram are effective in a way that a detailed explanation never could have been; for by a single act of the mind, you see the lines and with them the facts they stand for, all together as a single fact. Explanation by diagram, then, wherever it is possible, is not only the simplest and the most effective way of explaining a subject, but it is the type which all explanations ought to follow. Though, as in this very case of the "Origin of Species," a diagram will explain only

small parts of the subject, the other devices of explanation, as I shall hope to show presently, constant summaries, transitions carefully marked, and connectives, all serve the same purpose of giving you this feeling that somehow you see the subject altogether and as a single thing.

13. Now in an explanation which gives you this feeling of a clear view of the whole subject, what are the essential features? In the first place, such an explanation will have unity. Just as a glance at a map or a diagram tells you what its limits are, so a good explanation will leave the subject rounded and complete in your mind. Green's explanation of Elizabeth as "at once the daughter of Henry and of Anne Boleyn" is as satisfying in its completeness as is the rigid definiteness of a portrait by Holbein. A rambling and desultory explanation would be like a map in which one side of the boundaries were complete, and the other side left ragged.

In the second place, a good explanation will be intelligibly arranged. To follow up the analogy, when you look at a map or a diagram you see, not blotches of black and white or of colors lying at random on the paper before you, but larger blotches which mean counties or states and round black marks which mean cities or villages; or on a meteorological map masses of different shading which mean different amounts of rainfall. Each division is signified by its own mark in a way which is at once apparent to the eye. So it is with a good explanation; its arrangement will be

immediately clear and intelligible. Not only will it make you understand the extent and the limit of the subject, but as Grote in his account of the natural features of Greece discusses the rivers, the lakes, the obstacles to land travel, the deeply cut shore line, each carefully in its own paragraph, so every good explanation will show this same perspicuous and careful separation and arrangement of the different parts of the subject.

Besides having unity and clear arrangement, any explanation which gives you this sense of immediate and luminous understanding will be couched in fixed terms. Just as on the map or diagram, each kind of line means the same thing all over the map, so a good explanation must have its terms defined - and the figure of speech, you will notice, is significant; the words will mean certain fixed things, which will always be the same. Professor James could not have explained so clearly and so succinctly why he holds that religious faith may be logical if he had not declared 1 so carefully and explicitly in the beginning just what he should mean by certain words.

1

And finally, besides having unity, intelligible arrangement, and definiteness of terms, a good explanation will explain real things, not merely empty abstractions and general terms. To follow out the figure, just as a good map is filled in with all sorts of facts which concern human life, rivers, hills, roads, towns, so a profitable explanation will arrange and make intelligible specific and ultimate facts. Macaulay

1 See pp. 283-284.

has set the standard for us here: even if his views in history or in criticism are sometimes discredited to-day, his method of setting them forth in rigidly concrete terms has set a permanent standard. Explanation is always at its best when it sets forth general principles by means of the specific cases.

It is safe to say, then, that a good explanation should be like a good map or diagram, in that it will have, obviously and unquestionably, unity, clear arrangement, definiteness of terms, and concreteness.

So far, then, it is clear that this analogy of a map or a diagram is serviceable in stating the theoretical necessities of an exposition. In practice, it is always a useful analogy to keep in mind: when you are at the actual work of writing out your explanation try to feel that you are so laying out your subject as to give an easy and comprehensive view of the whole. In this discussion, to which I now turn, of the principles and practical devices of explanation, I shall urge you to bear always in mind the ease of thought and the sensation of clarity, of a comfortable grasp on your knowledge, that you yourself have when you first understand a difficult subject, a sation which is often so physical that you might liken it to the actual clearing of space on your desk when you put the papers and books in order. Bear in mind this almost palpable simplification of the furnishing of your own mind; and remember that if your explanation is to be worth while it must produce this same comforting clarification in your reader's mind.

sen

14. I will turn now to a more direct search for principles and devices; and I shall try to keep this search as practical as possible by constant reference to successful explanations.

As I have already said, the essence of the act of understanding lies in the reducing of multiplicity to unity. The essential underlying unity may be something which for a long time baffles us, as in the case. of Darwin's search for his theory; and until many great thinkers had spent years of study in preparing the way, it was not possible for Tyndall to explain the innumerable phenomena of heat and light, which seemed to our ancestors so diverse and incomprehensible, by bringing them all under the single principle of motion. Nevertheless, an explanation which does not thus reduce multiplicity to unity is no explanation. A catalogue of plants, no matter how minute in description, contains no germ of explanation of the whole unless it sets forth some system of classification by which the plants are grouped under genera, the genera under families, the families under orders, the orders under classes, all dependent on characters more and more widespread. Such a bare list as Gray's "Manual of Botany" is in a sense an exposition, in that it contains the material for reducing the chaos of different plants to unity; without this marshalling of the infinity of species into ranks it would confuse rather than explain. Even in an explanation of a game, as of tennis or of baseball, unless you have some underlying conception of how the game is best played, or of what is the key to win