The thoughts which have been gathered in the modes pointed out in the last chapter are now to be arranged in the most effective order. It will not usually do to begin a speech with those things we happen to first think of, and proceed to others that are less obvious. This would lead to an anti-climax fatal to eloquence. A speaker who adopted this mode once complained that his speeches often seemed to taper to a very fine point, and that he lost all interest in them before finishing. The explanation was simple; he uttered first those thoughts which were familiar to himself and came afterward to those which had been sought out by more or less painful effort, and which seemed less certain and valuable. The remedy for this fault is found in careful arrangement. The most familiar thoughts will naturally be jotted down first, but it does not follow that they should occupy the same place in the finished plan of the speech.

The true mode of improving your plans is to bestow a great deal of time and thought upon them, and to make no disposition of any part for which you cannot give a satisfactory reason. This direction relates only to the beginner. In time the formation of plans will become so natural that any variation from the most effective arrangement will be felt as keenly as a discord in music is felt by a master in that art. From such carefully constructed plans, firm, coherent, and logical discourses will result.

There are certain general characteristics that each plan should possess. It must fully indicate the nature of the proposed discourse and mark out each of its successive steps with accuracy. Any want of definiteness in the outline is a fatal defect. You must feel that you can rely absolutely on it for guidance to the end of your discourse or be always in danger of embarrassment and confusion.

Each clause should express a distinct idea, and but one. This should be repeated in no other part of the discourse; otherwise, we fall into wearisome repetitions, the great vice, as it is often claimed, of extempore speakers.

A brief plan is better, other things being equal, than a long one. Often a single word will recall an idea as perfectly as many sentences, and it will burden the memory less. We do not expect the draft of a house to equal the house in size, but only to preserve a proportionate relation to it throughout. The plan cannot supply the thought, but, indicating what is in the mind, it shows how to bring it forth in regular succession. It is a pathway leading to a definite end, and, like all pathways, its crowning merits are directness and smoothness. Without these qualities it will perplex and hinder rather than aid. Each word in the plan should suggest an idea, and be so firmly bound to that idea that the two cannot become separated in any exigency of speech. You will find it sorely perplexing if, in the heat of discourse, some important note should lose the thought for which it previously stood and become an empty word. But with clear conceptions condensed into fitting words this cannot easily happen. A familiar idea can be expressed very briefly, while a strange or new conception may require more expansion. But all thoughts advanced by the speaker ought to be familiar to himself as the result of long meditation and thorough mastery, no matter how strange or startling they are to his hearers. Most skeletons may be brought within the compass of a hundred words, and every part be clearly indicated to the mind that conceived it, though perhaps not to any other.

There may be occasions when a speaker is justified in announcing his divisions and subdivisions, but such cases are exceptions. Hearers do not care how a discourse is constructed, so it comes to them warm and pulsating with life. To give the plan of a speech before the speech itself is contrary to the order of nature. We are not required first to look upon a grisly skeleton before we can see a graceful, living body. There is a skeleton inside each body, but during life it is well hidden, and there is no reason that the speaker should anticipate the work of the tomb. It is hardly less objectionable to name the parts of the discourse during the progress of the discussion, for—continuing the former illustration—bones that project through the skin are very unlovely. The only case, I presume to think, where it is justifiable to name the parts of a discourse, either before or during its delivery, is where the separate parts have an importance of their own, in addition to their office of contributing to the general object. Much of the proverbial “dryness” of sermons arises from the preacher telling what he is about to remark, firstly, before he actually makes the remark thus numbered. Whenever we hear a minister read his text, announce his theme, state the parts into which he means to divide it, and then warn us that the first head will be subdivided into a certain number of parts, each of which is also specified in advance, we prepare our endurance for a severe test.