1. The authority of one head may be absolutely extinguished in the field where conflict exists. Here we have legalized the state of things described above as existing with a combination of one spineless department-head and one very spiny one. It works, but at the expense of everything that tends to the efficiency of the extinguished authority, and I do not recommend it.

2. An attempt may be made, as noted above, to draw a line between the two spheres of authority and keep each in its place. This appeals to those who are fond of detail, for it can be done only by considering and ticketing details. A line, defined by some one clear principle, cannot be drawn in a field of this kind between two things both of which logically cover that field. It is logical that the children’s librarian in a branch should be wholly under the authority of the branch librarian, since she is a branch employee like the others. It is just as logical that she should be wholly under the authority of the supervisor, of whose department she is a part. If we are to define the things in which she is to obey the one and the other, they must be enumerated one by one. And then other things will turn up that have not been thus enumerated, and we are in trouble again. This plan, as I have said, appeals to those who revel in regulations and specifications, but I can recommend it no more than the other.

3. One department may formally and distinctly be set above the other. Or, what is the same thing, the librarian may resolve, when a conflict arises, always to decide the matter in favor of one particular department. This means, in the special case that we have been using as an illustration, either that the children’s department shall be allowed to do nothing in a branch library without the consent of the branch librarian, or of the supervisor of branches, if there is one; or that all questions involving the administration of a branch children’s room must depend ultimately on the chief of the children’s department.

This may seem to be the same as the plan by which the authority of one department is absolutely done away in the disputed sphere. It is of the same type, but not so drastic. In the other plan one has not authority to do anything; in this, one must ask permission—not the same thing by any means. This plan is practically in effect at some libraries; it would probably be regarded as equitable by most department heads—provided their own department were put ahead of the other. The trouble is that it involves an arbitrary subordination—one that does not exist in the nature of the classification. And this subordination is local and partial; it cannot hold good for the whole department. No one would think of placing the branch department, as a whole, under the children’s department, or vice versa. And the objections, although not so strong as those to the extinguishment plan, are of the same kind. The efficiency of one department or the other is bound to suffer, and for this reason I do not consider this the best plan.

4. All department heads in conflicting spheres, may be regarded simply as advisers of the librarian and not as possessing authority in themselves to give orders. A conflict is thus reduced to contradictory advice from two sources. The librarian then pursues whatever course seems good to him. This plan has attractive features, especially to administrators of the type that like to keep a finger in every pie. There is doubtless danger in aloofness. The librarian must know what is going on, but I see no advantage in requiring him to decide questions of trivial detail at frequent intervals, as he must do under this plan; for conflicts generally begin in questions of detail and it is at the beginning or even earlier, in anticipation, that they must be caught and adjusted. This plan works, but it reduces the department head to a consulting expert and burdens the librarian with detail. It does not appeal to me at all.

5. The two conflicting departments may co-operate, intelligently and courteously without sacrifice of authority or self-respect, under the advice and orders of the librarian.

This is the plan that I recommend. It is the most difficult of all, and no regulations or specifications can be formulated for carrying it out. For this reason it will never be widely in favor. A wicked and rebellious generation demands a sign, and in this plan there is neither sign nor formula except that general principle of helpfulness and willingness to place the common whole above the selfish part that is at the antipodes of both wickedness and rebellion. It is a personal matter and it adds one important qualification to those already necessary in department heads—the ability to do team work. This qualification, however, is so important, quite apart from its necessity in connection with this plan, that we may consider it an advantage, rather than otherwise, that the plan puts it forward and insists upon it. On the whole I think that a library with mediocre department heads having this qualification is better manned, and will do more satisfactory work than one with a staff of supremely able experts, cranky, self-centered and all pulling different ways. The efforts of members of a body like a library staff are not to be measured arithmetically—they are what mathematicians call “vectors”—directed quantities, like force, velocity or acceleration. To know where a man will bring up one must have not only his speed, but its direction. The sum of two equal forces may be anything from zero up to their double, depending on their relative directions, and if the sum is zero, no matter how large the components may be, the result is precisely the same as if those components are small, or as if neither existed. It is this sort of thing that an eminent employer of labor had in mind when he advised, “If two of your subordinates don’t get along together, discharge both of them, no matter how good they are.” In this man’s estimation the relative value of team work evidently stands pretty high. I should not follow his advice, however, without giving everyone a fair chance. I have known the opinions of one department head about another and their ability to work together to improve greatly on acquaintance.

The part necessarily played by the librarian in this scheme may be regarded by some as an objection. I have already referred to administrators who, like the late Czar of Russia, prefer to regulate all the details of the kingdom by personal supervision. There is also the precisely opposite type, who like to make a good machine, set it going, and then let it alone. The trouble is that machines will not run of themselves. They need oversight, oiling, cleaning and repairing. The best require a minimum of all this, but all must have some of it. And such machinery as there is in this plan requires a maximum of oversight. It is, however, not the control of details but rather the watching of general methods and results. Is everything running smoothly, without “lost motion” or “backlash,” and turning out a satisfactory finished product? If not can the trouble be located? Yes; these two cogs do not work smoothly together. Let us find out which is at fault and adjust or replace it; but if our investigation is fruitless, possibly the best plan is to discard both.

I trust I have misled no one by treating here specifically of two departments. I might have substituted the names of a dozen others. All through library administration, and especially in the administration of a system of branch libraries, these possibilities of conflict occur. In branches they are generally between the branch administration and the central departments—finance, supplies, cataloging, book-orders, reference and circulation.

The handling of this whole matter depends, of course, on the librarian. He it must be who is to decide on general policies or go to his Board for a decision in cases so important that he feels their action necessary. If the work of departments overlaps in some field where the library’s policy has not yet been decided upon and defined, he has no one to blame but himself if the adjustment is difficult. And if policies are defined in advance and pains taken to inform department heads thoroughly of their existence and import, the likelihood of serious disagreement will be considerably lessened.