Of course the desirability of drilling separate divisions of a fleet, ana separate ships, turret crews, fire-control parties, and what-not, in accordance with the requirements of fleet work does not prevent them from drilling by themselves as often as they wish—­any more than the necessity of drilling in the orchestra prevents a trombone player from practising on his instrument as much as the police will let him.

Thus the fact of keeping a fleet together does more than merely give opportunity for acquiring skill in handling the fleet itself, and in handling the various ships so that they will work together as parts of the fleet machine; because it shows each of the various smaller units within the ships themselves how to direct its training.

For this reason, the idea so often suggested of keeping the fleet normally broken up into smaller parts, those parts close enough together to unite before an enemy could strike, is most objectionable.  It is impossible to keep the fleet together all the time, because of needed repairs, needed relaxation, and the necessity for individual drills that enable a captain or division commander to strengthen his weak points; but nevertheless since the “mission” of training is to attain fighting efficiency in the fleet as a whole, rather than to attain fighting efficiency in the various parts; and since it can be attained only by drilling the fleet as a whole, the decision to keep the fleet united as much as practicable seems inevitably to follow.  Besides, the statement cannot be successfully controverted that difficult things are usually not so well done as easy things, that drills of large organizations are more difficult than drills of small organizations, and that in every fleet the drills that are done the worst are the drills of the fleet as a whole.  How could anything else be expected, when one considers how much more often, for instance, a turret crew is exercised at loading than the fleet is exercised at the difficult movement of changing the “line of bearing”?

The older officers remember that for many years we carried on drills at what we called “fleet tactics,” though we knew they were only tactical drills.  They were excellent in the same sense as that in which the drill of the manual of arms was excellent, or the squad exercises given to recruits.  They were necessary; but beyond the elementary purpose of training in ship handling in fleet movements, they had no “end in view”; they were planned with a limited horizon, they were planned from the bottom.

General Staff.—­In order to direct the drills of a fleet toward some worthy end, that end itself must be clearly seen; and in order that it may be clearly seen, it first must be discovered.  The end does not exist as a bright mark in the sky, but as the answer to a difficult problem; it cannot be found by guessing or by speculating or by groping in the dark.  Strategy says that the best way in which to find it is by the “estimate of the situation” method.