While it is true that the principles of gunnery training, long-range firing, and range-finding are well known and have been explained in publications within the past few years, it is also true that many officers are apparently not sufficiently acquainted with them. This is due partly to the inaccessibility of these publications, partly to the almost universal disinclination to read anything not contained in current numbers; but, doubtless, also to their being written in a form not always easy to comprehend, because of the assumed acquaintance on the part of the readers with certain of the essential principles involved.
It is for the above reasons that I have attempted the following explanation of this important subject; and I hope that it will be the means of changing certain opinions (that I believe are now impeding our progress), held by many whose present duties necessarily prevent them giving these matters close attention.
The conditions of record target practice, having been frequently published, are too well known to require a detailed description. The firing is conducted during fair weather, the speed of the ship is prescribed, the course is marked by a line of buoys, the limits of range are between 1400 and 1600 yards and the target is a rectangle, 17 feet high by 21 feet wide. Manifestly, these are not the conditions we may expect in battle; then we shall have to take the weather as it comes, there will be no line of buoys, the enemy will not be stationary, the ranges will vary continuously and sometimes rapidly, and will probably never be as small as 1600 yards. Therefore, certain critics declare that this is "trick shooting," and is of no practical value; that the range for record practice should be largely increased; that the ships should fire in a sea way; that there should be no line of buoys; that the target should be under way, that ammunition should not be wasted upon cut-and-dried shooting, and, in a word, that all firing should be at battle practices, in order that our gunnery training may always be of a "practical" nature.
There are many other opinions that closely resemble those just quoted; and besides there is the persistent traditional opinion that the famous man behind the gun (the gun-pointer) should be a real marksman (not simply the man that points the gun), and should be trained to estimate the distance of the enemy, to observe the fall of his shot and correct the sight-bar setting accordingly, to allow for the change of bearing due to the speed of his ship and that of the enemy, to estimate and allow for the force of the wind, etc., etc.
The object of this paper is to show that such opinions, though formed in all honesty, are founded some upon misconceptions of the requirements indispensable to success in teaching men to aim accurately, and some upon certain mistaken facts.
It is admitted, of course, that the ultimate object of all gunnery training is that when we meet the enemy upon the sea, we may be able, no matter what the weather conditions, range, etc., to hit him more times per minute than he can hit us. It follows that any stage or any feature of the training that does not contribute directly to this result, may be either unnecessary, and therefore simply a waste of ammunition, or it may be false training, and perhaps dangerously misleading.
Before we can discuss profitably the question of proper methods of training pointers, we must first settle the fundamental principle upon which our system is to be based; that is, we must decide whether the man that fires the gun is to be trained as a gun-pointer, whose sole duty is to aim as ordered, or as a marksman, that is, a man that estimates the range, directs the setting of the sight, aims, fires, observes the effect of his shot, etc.
It goes without saying that if this could be successfully accomplished, the advantages would be very great, and the whole question of gunnery training would be simplified; for if the men that fire the guns could be trained to expertness as marksmen, in the same sense that a hunter is a marksman, the commanding officer of a man-of-war would need only to designate the target and give the order: "You may fire when you're ready," and the duty of his divisional officers would consist in seeing their guns served with rapidity and safety. The advantages of such a system are so manifest that there could be no valid reason for not adopting it unless it could be shown to be quite impracticable. This therefore, is the first question to be settled, for, otherwise, the advocates of the different systems cannot advance beyond the stage of reflecting upon each others intelligence.
The whole question of the possibility of training a man as a successful great-gun marksman depends upon the possibility of training the human eye (either naked or using a telescope, and whether observing from a casemate, a turret, or an open gunport—all relatively near the water) to estimate distances with sufficient accuracy to insure hitting—or, what will answer the same purpose, to estimate with sufficient accuracy the amount by which his shots miss his target. If distances up to about the maximum battle ranges can be estimated with accuracy by the eye, then it might be possible to train the pointer as a marksman—though even then it would be exceedingly difficult, as it would require a very cool and nimble brain to apply all the necessary corrections with sufficient rapidity and accuracy. If, on the contrary, the errors in estimating the distance are in the majority of cases so great as to render the method useless—that is, so great as to render hitting very improbable, then the traditional marksman is a myth, at least as applied to the conditions of modern gun-fire.
Fortunately, it is no longer a question of opinion, but of fact; and, as usual, it is a case of the familiar shield with different colored sides. From the time that great-guns were first used on board fighting ships, up to a comparatively recent date, the guns were so weak that ranges were necessarily very short, the vertical angle subtended by the enemy was consequently large, and the accuracy required to make hits was not great. In those days the pointer habitually estimated the distance of the enemy with considerable success. In some cases he made his estimates by an unconscious method depending upon the quality of the so-called "nautical eye," which, we are informed, is no longer found in the theoretical heads of those who have never really acquired the "sea habit." Sometimes the method was more scientific, if not more successful. When all men on the enemy's decks looked alike, the range was moo yards, or over; at 800 yards you could distinguish a man from an officer; at 600 yards you could make out a pair of well-developed side whiskers; at 400 yards you could hear the men swearing; and so on down to the range at which you could see the whites of their eyes. The traditional pointer made his allowance for range by selecting his point of aim, the topsail yard being a favorite. He was the Captain of his gun, and he did not want advice from anybody.
Since the introduction of rifled guns of high velocity, and the consequent increase in fighting ranges, all this is changed—except the opinions above indicated. As a matter of fact, it has been demonstrated by exhaustive experiments that it is quite impossible for the eye to estimate distances greater than about 2000 yards with anything like sufficient accuracy to be of use in directing gun-fire. For example, the following conclusions derived from a test made by about 30 observers (one-third being experienced officers and the remainder principally gun-pointers) will show the futility of such a method.
"(1) The estimates are so wild as almost to preclude laying down any laws or drawing any conclusions. The averages of errors, however, compare favorably, increasing with an increase of range, as would be expected, and from these averages certain deductions may be drawn.
"(2) The probable error on a single estimate at any range is so uncertain as to be worthless for any gunnery purposes. At 1193 yards, when the vessels were broadside on, 19 out of 20 observers on the battleship, and 4 out of 6 observers on the torpedo boat estimated the range within one-half of the danger space of a 20 foot target, but even at this short range there was a variation of 1200 yards between extreme estimates; and later on, at the same range, with vessels end-on, the number of these same observers who estimated within one-half of the danger space of the target were reduced, respectively, to 8 and 3, and the extreme limits of variation of estimates rose to 2400 yards, showing the great effect of the difference of conditions of position alone on estimates made by the same men.
"(3) Beyond 2100 yards, in no case did more than 1 observer on each vessel estimate within one-half of the danger space of a 20-foot target. Between 2000 and 2100 the number who estimated with this degree of accuracy were, battle ship 5; torpedo- boat 2; but it must be noted that the observers who estimated with this degree of accuracy, say at 2100 yards, would not be the ones who did it at 2000 or 2095 yards, thus showing the mere chance of the estimates, and the futility of ever counting on this method of range-finding in action."
(The half danger spaces above referred to are those of the 6-inch 40-calibre gun, I. V. 2400 f. s., which are about as follows: 260 yards at 1200, 140 at 2000, 75 at 3000, 45 at 4000, 30 at 5000, and 20 at 6000. For comparison, the half danger space of the 12-inch 40-calibre gun, I. V. 2800 f. s., is as follows: 395 yards at 1200, 230 at 2000, 140 at 3000, 100 at 4000, 70 at 500o, and 55 at 6000.)
If anything further is needed to demonstrate the uselessness of attempting to train a modern pointer as a marksman, the results of similar and equally conclusive tests show that, even if the pointer were able to estimate the amount that his shots fall short, it would be, in the great majority of cases, impossible for him to identify his own shots, when a number of guns were firing at the same time, because the smoke and gases from his own and from adjacent guns very seldom permit him to observe his projectiles throughout their flight, though his opportunities for aiming and firing may be very frequent.
Assuming, therefore, that the pointer is to be trained simply to point his gun as directed, and that his sight will be set by direction of skilled range-officers advantageously placed for observing, it remains to discuss the means by which we can train pointers, in the shortest space of time, and with the least expenditure of ammunition, to such a degree of expertness that they may be depended upon always to point their guns quickly and with unfailing accuracy.
It is, of course, not proposed to give a detailed description of these methods, but simply to determine, if possible, the conditions that are indispensable to success in training a man to aim with accuracy, the requirements of a method that will test fairly his skill and his nerve, and the rational way of employing the pointers after they have achieved expertness.
Before proceeding with the exposition of the simple principles involved, I feel that almost an apology is due for the elementary nature of some of the following explanations and illustrations; but I believe that they may be useful in dislodging certain singularly tenacious opinions held by men of influence who have apparently not taken the pains to analyze the basis of their convictions.
It will at once be admitted that no real progress could be made in training a recruit unless you use a gun that is accurate enough to make a hit every time it is accurately aimed. This applies, of course, to sub-calibre practice of any kind. For example, suppose your sub-calibre tube is an old-fashioned blunderbuss that cannot be depended upon to make one hit out of 10 shots on a target of a certain size at a certain range, even if the gun and the target are solidly planted on shore, and the gun is accurately aimed by an expert. Manifestly, any attempt to train a recruit with such a machine would be a waste of ammunition, not to mention your time and his; for when the recruit fires and makes a miss, he learns nothing by the experience. If, however, you substitute a modern rifled tube that can be depended upon to hit every time when accurately aimed, then when the recruit fires and misses, he has learned something; he knows, in the first place, that the fault was his and not the gun's, and the position of the miss, with reference to his supposed point of aim, shows him where he made his error in aiming.
The principle involved is that, for efficiency in training pointers to aim, the conditions must be such that the gun will make a hit in the bull's-eye every time it is accurately aimed thereat; in other words, that the shot and the line of sight will always pierce the target in practically the same place. I say in "practically" the same place, because no gun is absolutely accurate, no matter how short the range. The point is that the range must be short enough and the target large enough to ensure a hit every time the gun is accurately aimed. This is the gist of the whole matter, and its very simplicity is possibly the reason of its being so persistently misunderstood. Therefore, at the risk of being tedious, I will further illustrate it as follows:
Suppose you secure a modern, high-powered, .30-calibre rifle in a heavy vise in such a manner that it will not be deranged in the least by the shock of discharge; and suppose the line of sight to be directed at the middle of a screen placed at a distance of, say, 300 yards. Fire a few shots, and you will probably find that no two of them have pierced the screen at the same point. The best cartridges differ slightly in power from each other, the bullets are not absolutely uniform in shape, weight and position of centre of gravity, there are slight differences in the way they "take" the rifling, the latter is not perfect, etc., etc. I assume all other causes of deflection to be eliminated, as, for example, the effect of wind, etc. If, therefore, you fire moo shots, you will find that the holes in the screen are distributed about uniformly over a certain area. Draw a line around these holes, including all of them, and you will find that the figure is nearly a circle (more nearly an ellipse, I believe) of a certain diameter, say 4 inches. If you fire a second string of moo shots, with the same quality of ammunition and under the same conditions, they will all fall within this circle.
Similarly, a "perfect shot" who aims exactly at the center of this circle will make a hit every time he fires. Of course a perfect shot does not exist; no man can train his muscles to such a degree of refinement as to hold a rifle perfectly still. Experience shows, however, that a man can be trained so that his line of sight will not vary from the point of aim more than a certain amount. Suppose this amount to be such that the line of sight remains within a circle 4 inches in diameter on the screen at 300 yards. Such a man we will call a "good shot." Then, if the diameter of the circle enclosing the shot holes be increased from 4 to 8 inches, it is evident that a "good shot" could hit this bull's-eye, 8 inches in diameter, every time. Therefore, in order to train men (using this gun and ammunition) to aim accurately, the relation between the size of the bull's-eye and the range at which it it used, should be as 8 inches is to 300 yards (for ranges not differing much from 300 yards). If the range is reduced to 150 yards, the diameter of the bull's-eye should be something less than 4 inches; similarly, if the range is increased to 600 yards, the bull's-eye should be somewhat more than twice as great as at 300 yards. As a matter of fact, it is made 2.5 times as great for 600 yards, or six times as great in area. (The diameters of the bull's-eyes of regulation army targets are as follows: 8 inches for 200 and 300 yards, 20 inches for 500 and 600 yards, and 36 inches for 900 and moo yards.) Neglecting for the sake of simplicity of explanation, all causes of error except those above mentioned, we see, therefore, that a "good shot" can hit with equal facility at any of the three ranges indicated, provided the diameter of the bull's-eye bears a certain necessary relation to the range.
For the remainder of this illustration, I assume the bull's-eye to be used as the target—that is, the bull's-eye alone, not surrounded by a screen that will show the amount by which shots miss it, as is the case with the regulation Army target. The reason for this assumption will be shown later when an essential feature of great-gun targets is explained.
If, then, such a bull's-eye target is made much too large (to put the extreme case, suppose it to be 10 feet in diameter), a man need not be a "good shot" to hit; therefore a green man cannot be trained to expertness by the use of it. If, on the contrary the target is much to small (say one inch in diameter), it is almost entirely useless to attempt to train him to expertness by firing at it. For example, suppose that the 8-inch bull's-eye is used as a target at 600 yards, instead of the 20-inch one. As the former is but onesixth of the area that it should be at this range (600 yards), it is evident that even the "good shot" cannot expect to hit it more than once in six shots, on an average. Therefore, to attempt to train a green man. by using the 8-inch bull's-eye as a target at 600 yards is largely a waste of time and ammunition.
The above, will, I believe, make it clear, that for purposes of training a pointer to aim accurately, or as a test of his acquired accuracy, the size of the target must bear a certain definite relation to the range—depending both upon the unavoidable errors of the gun used and upon the normal error in aiming of "good shots." It follows that if the target is reduced in the proper proportion, a very short range may be used with equal advantages, as far as concerns training a man in accuracy of aiming alone, which, of course, must first be attained before any further progress is possible. For example, a man can be trained to aim accurately by practicing at a range of less than 5o yards, provided the target is reduced in size to correspond therewith; or by placing a small, 22-calibre tube in a rifle, he may practice with advantage at 50 feet, or less. Most officers are familiar with this principle, and with its important application.
It may be replied that if the above is true, there would apparently be no reason for training riflemen at long ranges; and this would be true if the fire of riflemen could always be controlled in battle in the same manner and with the same efficiency as that of a number of heavy guns firing from the same position—grouped on the same ship. The frequent necessity for independent action on the part of the rifleman forbids this, to a large extent. Therefore the rifleman must be trained as a marksman. He must have the skill of the big-game hunter in estimating the distance, setting the sight, allowing for wind, etc., when firing at long ranges. In order to give him this skill, he must be trained by actually firing at long ranges—but each range must have a target of the proper size, otherwise the firing is not a training.
If the fire of riflemen could, upon all occasions, be controlled by their officers much more effectively than is now possible by the best riflemen acting independently, it would be advisable to train them as accurate pointers only, and not as marksmen. Battle firing at long unmarked ranges is, necessarily, not very accurate, even under favorable conditions. The average of shots fired to men killed is very great.
The principle of the proper relation between the range and the size of the target necessary for successful training applies to all calibre of guns from .30-calibre rifles to 13-inch turret guns; and, for certain calibres of naval guns, the dispersion of accurately aimed shots has been obtained by actual firing on shore against a screen target. Experience in firing guns of different types mounted on board ship has supplied similar, though less accurate, information. It is from a consideration of such data that the relation between the record-practice range and the size of the target has been determined. It would, of course, be desirable to have a different sized target for each calibre of gun that fires at the same range, but as this is impracticable, the target is made large enough to include all the unavoidable errors of any of the guns that use it. (For reasons that need not be stated here, we continue to fire some guns on the 1000 yard range when a consideration of their ballistic qualities alone would assign them to a longer range). Thus, the target for all guns above 3-inches in calibre is 17 feet high by 21 feet long, and the range is 1400 to 1600 yards.
This target is larger than would be necessary to catch all shots that are aimed with perfect accuracy and fired at a distance that is exactly known—which would be the conditions existing if the gun and target were on shore. But when firing at sea, perfect accuracy in aiming cannot often be attained, and there is always more or less error in the range; therefore the screen must be enlarged accordingly. It may be a trifle too small for some conditions of sea firing and a trifle too large for others, but this is not apparent from the results of firing under the condition of record practice. For convenience of illustration, it will therefore be assumed to be of exactly the correct size for the gun in question, when firing at 1600 yards.
Note that the entire area of this target (assuming it to be the correct size) corresponds to the bull's-eye of a rifle target, and the so-called "bull's-eye" (51 inches square) that is painted in the middle of the 17 by 21 rectangle, corresponds to the 4-inch circle, mentioned above in connection with the 8-inch bull's-eye of the small-arm target used at 300 yards. In one case, the rifleman (the "good shot") is supposed to keep his line of sight within a 4-inch circle, at 300 yards; and in the other case, the gun-pointer (also a "good shot") must not aim outside of the 51-inch square, if he wants to be sure of making a hit. Of course it would be a great convenience in spotting misses, if we could use a very large great-gun target, say, 60 feet square, with the present target (17 by 21) painted in the middle of it, and hits on the latter only to count; but such targets would be impracticable to handle, not to mention the expense of mounting them.
The requirement that the great-gun pointer shall aim within the 51-inch square (assuming this to be the proper size for the aiming spot) refers to firing under the conditions of record practice. These limits are necessary on account of the difficulty of holding the line of sight steadily within them during the "firing interval," while the ship has a certain amount of motion; of course with a naval gun and target on shore, an entirely unskilled person can place the wires on any indicated point, within a few inches.
Record "target practice" is a misnomer which we have inherited from the past, but which it would be inadvisable to change at this time. It is, in reality, principally a "firing test" to determine the actual skill of the pointers and gun-crews. Before it begins, a division officer knows almost exactly the relative accuracy and rapidity with which his various pointers can aim, but in the case of untried pointers, he does not know whether or not they will show the necessary nerve and steadiness under the excitement of firing. Sometimes he is disappointed to find that an otherwise excellent pointer gets "rattled"; but there is one thing that both he and all of his pointers understand thoroughly, namely, that a hit will be made every time that the gun is accurately aimed and no mistakes are made in ranges, etc.; that the target is large enough to include all of the errors of gun-fire afloat, that cannot be avoided by any precautions on their part, such as differences in the initial velocity of different charges, etc., etc. In other words, the relation between the range and the size of the target is such that all hands know that the result of the shooting is a fair comparative test of the skill of the personnel, and that success does not depend upon luck; hence the intensity of loyal competition, without which only indifferent success can be attained in gunnery training, or in any other training requiring skill and dexterity.
As far as concerns training alone, there would be no objection to doubling the present range, provided the target were made just large enough to catch all well-aimed shots. For most guns, this would probably require a screen somewhat more than twice as high and twice as wide, or more than four times as great in area as the present one. To hold up the present screen (17 by 21) in a moderate breeze, requires a target raft weighing 6 tons, and 700-lb anchors with 3 1/2-inch mooring lines; so it may be imagined what gear would be necessary to carry more than four times as much sail; and in many respects the result of the firing would be less satisfactory than under the present condition,—not to mention the increased difficulties of spotting and of handling heavier screens, the longer distances to be travelled by the ship and the repair boat, and the consequent loss of time, etc.
If the present target were used at 3200 yards, the best pointers could not make more than about one hit out of four shots, under the ordinary conditions of practice, and, consequently, the firing would not be a true test of skill; the element of luck would enter so largely into the result that the spirit of competition would disappear, and with it the efficiency of team-work and the skill of the pointer.
Firing under such conditions would be, in effect, the same as using the blunderbuss to test the skill of a rifleman. As an instrument for training pointers, the very best gun becomes a blunderbuss when the range is much too great or the target much too small. Many years ago, all of our target practice was exclusively blunderbuss training. We fired at a small triangular target (about one-seventh the area of the present one) that a perfect pointer could not hit once out of many shots. Consequently, we did not expect to hit it. We estimated the amount by which a pointer missed this target, and we now know these estimates to have been greatly in error. Unless the misses were very wide, we were quite satisfied. We had to be, for there was no standard with which the accuracy of a shot could be compared. The consequence was that the pointers learned almost nothing. To double the present range without doubling the size of the target, which is impracticable (as well as useless), would be to reproduce the blunderbuss training of the past in all the essential particulars that rendered it largely a waste of ammunition.
It is therefore apparent that in order to train the individual pointer and test the effectiveness of his training, we must fire at about the present range ( t600) as long as we use a target of the present size (17 x 21). in any event, we must always maintain a certain definite relation between the size of the target and the range at which it is used. In all cases where this relation pertains, it would be convenient to call it a "training range."
I am aware that, for this year's prize firing, the British have changed the range from 1600 to 2700 yards, but they have not thereby changed the principles involved. They have increased the area of the target somewhat, but if they have not given it the size necessary to insure a hit for every shot that is well aimed, the consequence will be that the result of their prize firing of 1904 will not be a true test of the skill of the pointers, and the efficiency of the training of the latter must inevitably suffer.
We all know that we can train gun-crews and gun-pointers to a high degree of skill—this skill being measured by firing under the conditions of record practice. "But," object the critics, "of what use are such pointers if they cannot make a large percentage of hits at long range?" "Why do we waste ammunition on fool shooting' at 1600 yards when we should be teaching our pointers to make 6070 of hits on an enemy at 5000 or 6000 yards?" "What is the sense in conducting practice always under favorable conditions when we may have to fight an enemy in an ocean swell?" etc., etc.
I beg that it will not be assumed that these questions represent straw men that I have set up in order to demolish them. They are actual questions propounded, not in the winning spirit of the seeker after useful information, but with a marked elevation of one side of the lip, a habit peculiar to the unthinking. The confusion of thought indicated by such questions may be shown by a few simple illustrations.
A perfect pointer, firing a perfect gun, with the sight-bar correctly set, could put every shot through a bull's-eye, equal to the diameter of the projectile, at any distance up to the maximum range of the gun. Even the critics know this, but they greatly underestimate the influence of the errors involved when they insist that a system of training is not a practical one unless the pointer can eventually make a large percentage of hits at a long range.
In principle the matter is so simple that it would seem that all differences of opinion should disappear with a careful consideration of the facts.
When a great gun is fired at a target at a considerable range, there are a dozen or so different errors that may contribute to decrease the possible percentage of hits. It will not be necessary to consider all of these individually, in order to illustrate the combined influence of certain errors in decreasing hits at all ranges that are greater than the "training range,"—that is, greater than the maximum range at which a good shot can make a clean string of hits on the target used. For the sake of simplicity, only errors in range will be considered.
Though the charges of smokeless powder are made up at a proving ground with all possible care, it is found that even those of the same index will vary in initial velocity from is to 25 foot-seconds from the desired velocity. Thus the extreme variation of one of these charges from another may be from 30 to 50 foot-seconds. The temperature at which they are fired may increase or decrease this variation. The powder sometimes becomes irregular with age. The weight of the shell or the diameter of the rotating band, may change the range appreciably. Usually the effect of such errors cannot be anticipated and allowed for by the personnel on board ship. They are commonly called the "unavoidable errors" of the gun. Suppose that, with a certain gun firing at 160o yards, their maximum effect, when they all act in the same direction, is to displace the shot four feet vertically, either up or down from the point of aim. If such a gun were mounted on shore, and each shot accurately aimed at the center of a screen, this screen would have to be eight feet high in order to catch all shots fired at its center. Therefore, if the same screen is placed at 3200 yards from the gun, it is evident that more than half of a large number of shots will miss it—going either too high or too low—and at 6000 yards, more than two-thirds will miss it; and these misses, being due to the "unavoidable errors" of the gun, cannot possibly be diminished, except by diminishing the errors. That is, if a pointer could be trained to absolute perfection, it would be entirely impossible for him to avoid making the percentages of misses indicated, while using the gun in question, at this range.
Now suppose the gun described above to be fired from a ship's deck, at the same range (1I600), under the usual conditions, i. e., aimed and fired rapidly in fair weather. Several additional errors are at once introduced. The pointer is not perfect. The best he can do is to keep his aim within a certain distance of the center of the target, the range is not very accurately determined, even when marked by buoys; the sights are inevitably somewhat out of adjustment; the jump of the gun varies sometimes according to the angle (from abeam) at which the gun is fired, etc., etc. Suppose that these additional errors, when acting in the same direction, amount to 4.5 feet, either up or down. The "unavoidable errors" of gun-fire (with this particular type of gun) on ship board are, therefore, 4.5 feet, either way, consequently the target must be 9 feet higher, or 17 feet in all to catch all shots. In other words, when a 17-foot target is used for this gun, the proper "training range" is 1600 yards. No attempt at exact accuracy is made in the figures used—they are assumed for purposes of illustration only.
It is evident, therefore, that with a ship anchored in fair weather, the range known within narrow limits, and the gun fired by a good pointer, it is possible to hit the target every time only when the range is 1600 yards or less (supposing the figures assumed to be correct). If the range is doubled, then less than half the number of hits can be made. It is simply impossible to make, on an average, more than so% of hits on this target, at this range, and with the unavoidable errors assumed. No amount of training can improve the results appreciably, as I have assumed the pointing to be done by one of the most expert pointers (whose error in aiming is assumed to be very small), and all the other errors are unavoidable with the gun in question. The percentage of hits at this range (with this gun) can be increased only by more uniform powder, more accurate sights, more accurate methods of determining range, etc.
This should dispose at once of the astonishing opinions one frequently hears expressed, to the effect that a large percentage of hits should be made at long ranges, when the fire-control system is sufficiently improved and the pointers become expert at those ranges—the assumption being that when we can determine the range accurately, the pointers should hit every time.
Thirty-two hundred yards is not a very long range. If the screen in question is placed at 6000 yards, it is evident that less than one-third of the shots will hit. As a matter of fact, very much less than one-third will hit, as the dispersion of the shots becomes greater with the range, which must be known within much narrower limits in order to make a hit, even with a gun free from error, because the danger space of the target at a long range is relatively very small. Knowing the maximum errors of pointing, the errors in range of the guns, etc., the probability of hitting could be calculated for this range, but as this paper concerns only the principles involved, this is not necessary. It may be assumed, however, that at 6000 yards the probability of hitting under the conditions in question (a ship at anchor, etc.), would not be greater than one hit in six shots. Therefore the maximum that the best pointers could do under the very best conditions, with this gun, is 100% at 1600; 40%, say, at 3200 yards, and 16% at 6000 yards; and this will be considerably diminished, especially at long ranges, by lateral misses.
Nor is this all; for when you come to practice shooting at a towing target, from a moving ship, the most fruitful source of error is in keeping the range. As an example, take the conditions at 6000 yards. The danger space of the target is, say, 50 yards, i.e, a shot that just misses the top of the target will strike 50 yards beyond. You must therefore know the actual distance (or correct sight-bar range) within 25 yards in order to hit every time with an errorless gun; and the best range-finders have an error of over too yards at this range. Therefore, with both an errorless gun and an errorless pointer the chances of hitting are very small. Add to this all the unavoidable errors of gun-fire and of aiming at sea, and it requires no calculation to see that the maximum possible hits on a target 17 feet high at 6000 yards would be exceedingly small.
Previous explanations have shown, I believe, that it would be absurd to attempt to train green men by firing at a 17 foot target at 3200 or 6000 yards, but that, with this target, the 1600 yard range fulfills approximately the requirements for training a man to aim accurately, and for testing the results achieved. With permanence in gun stations we can doubtless bring up the general average of our pointers and thereby increase our efficiency, but there is a limit to the degree of skill to which a man's muscles can be trained in handling a heavy gun. For purposes of illustration suppose that the best pointers can aim within a bull's-eye, or aiming spot, 4 feet in diameter at 1600 yards. That is to say, the angular variation of the line of sight from the center of the bull's-eye, while aiming, is not more than 2 feet, or 1 min. 35 sec. of arc. Improvements in gun-gear and sights may enable us to reduce this somewhat; but let us suppose this to be the least variation at present attainable under moderately favorable weather conditions.
As this angular variation of the line of sight will be the same at all ranges, it follows that the longer the range the greater the vertical error in the fall of the shot, due to the error in aiming. For example, under the conditions assumed, this error is 2 feet at 1600, 4 feet at 3200, and 7.5 feet at 6000 yards. That is to say, the vertical dispersion of shots, fired by the most skillful pointer will be 4, 8 and 15 feet, respectively, at 1600, 3200 and 6000 yards; and still those who will not take the trouble to reason out these simple propositions, have shown themselves ready to condemn all systems of training pointers, or of controlling gun-fire, that do not produce as great, or nearly as great, a percentage of hits at a long range as at a relatively short one.
When the pointers have been trained to such a degree of skill that they can make a very large percentage of hits at the training range, under the favorable conditions prescribed for record practice, they have demonstrated their fitness for further training under less favorable conditions. To exercise them at the latter before they have achieved reasonable expertness at the former, would be as unjustifiable a waste of ammunition as to exercise green riflemen at long ranges before they could hit the standard small-arm target at 200 yards.
When the conditions are unfavorable, the shooting will be neither so rapid nor so accurate, but when pointers have attained their maximum efficiency under unfavorable conditions, their education is finished. They are ready to render us the best possible service in hitting an enemy under any conditions when they have been so trained that we can depend upon them to aim always as directed, and always with the maximum accuracy to which men can be trained to aim under the various conditions in which shooting is possible.
Such pointers will make the maximum number of hits that can be made under the circumstances—the circumstances being (1) the unavoidable errors in aiming of the best pointer, under the weather conditions existing at the time; (2) the unavoidable errors of their guns (sights, powder, etc.); and (3) the accuracy with which we, their officers, can keep the range and set their sight-bars.
Better than this no pointer can do (until more accurate means are invented), and it necessarily follows from the above, that such pointers, trained at "fool shooting" on the training range, will render us better service, no matter what the conditions under which the enemy is met, than pointers trained by any other system in which the accuracy and reliability of the pointers is not the prime object.
If it were possible to rig mechanical appliances that would at all times automatically keep the line of sight exactly on the center of a target, we would then have ideal pointers, and could dispense with human ones. With such aiming appliances, and a perfectly accurate range finder, we would have no errors to contend with except those due to the gun (sights, powder, etc., etc.); but these are so considerable at long ranges, and the danger space is so small, that few hits could be made. Simple as this is, it is constantly being lost sight of; and it is important that it be clearly understood, otherwise the extent of the inevitable errors of long-range firing will be seriously misapprehended, and the consequence will be disappointment and dissatisfaction with the pointers, even though they be practically perfect, and distrust of the system of training. The following example will, I believe, make this point clear:
Suppose that at a certain long range, the danger space of a vertical screen target is 40 yards. Then, with a perfect pointer aiming at the bull's-eye, a perfect range finder and an errorless gun, all shots would pass through the center of the target, and would strike the water 20 yards beyond. Now suppose the gun's errors in range, due to powder, etc., to be such as to cause shots to fall anywhere between 100 yards short and 100 yards beyond the point of impact of a shot passing through the center of the target. Evidently, the chances of hitting would be as 40 is to 200, or one in five; and these chances cannot possibly be increased by any refinements of training or of range-finding, since we have assumed these to be perfect.
To many readers these principles will appear so self-evident that I am sure many will suspect me of amusing myself by demolishing straw men, but such is not the case. Actual statements are made, with evidently sincere conviction, substantially as follows: "I have known pointers that could estimate the range at 6000 yards within a few yards, and make a hit nearly every time; and that's what we should be doing now instead of wasting ammunition in this manner."
Many of the persons who made such statements enjoy such a high reputation for ability that those who suffer from chronic cerebral inactivity have accepted them without question; and such "opinions" have doubtless influenced writers of reputation to make similar statements in deliberate articles and sell them to responsible publications. For example, in the Outlook for July, 1904, Mr. James Barnes, in an article entitled: "The Battle off the Capes" ("part of a stirring romance with that title by Mr. Barnes—a book, as he expresses it, which deals with 'possible history'") describes an imaginary naval action in which occurs the following "possible history" (the italics and parenthesis are mine):
"The Washington swung in a little closer to the black line of the English fleet, but not one gun was fired at her. Both sides seemed paralyzed, shocked, half dismayed into forgetfulness of action (by the spontaneous explosion of a whole fleet of submarines— an entire impossibility, but very effective for dramatic purposes). Then with a crash, the Englishman broad off the beam of the Washington opened (fire). Black specks, whirled in the air from the grand old battleship's decks. She almost staggered under the blows that reached her, for the lines (of vessels) were now but little more than six miles apart."
"Naval experts had said that they could not have approached that near without losing ships on both sides." . . .
"A bugle rang out. The bridge beneath their feet, as they started for the ladder, leapt and trembled. Emery saw the projectile plainly as it soared out in its flight. He paused a second (it takes a 12-inch projectile 19 seconds and a 6-inch projectile 27 seconds to go 6 miles) to watch the projectile, then—'Good shot, he cried, straight into her!' The young lad in the top screamed out shrilly, repeating the words as in joy: 'Hurray, good shot!'"
This is "stirring romance" all right, but it certainly is not possible history." The projectiles of our most powerful 12-inch guns have an angle of fall of about 11 1/2 degrees at 6 miles, therefore the danger space of a battleship, say, 40 feet high, is about 66 yards. Therefore in order to make a hit, with a perfect gun and perfect aiming, the range must be known within one-half this amount. The best range finders are almost useless at such a distance. The error is certainly not less than 5 per cent, or 600 yards. The dispersion of the shot in range, due to difference in powder charges, etc., would be at least 200 yards either way. Therefore, assuming the aiming to be perfect, the chances of hitting would be about as 33 is to 1600, or one hit in 48 shots—and each 12-inch gun is supplied with not more than 60 shells. Manifestly, this is not sufficiently stirring for the purposes of romance. As the projectiles of the most powerful 6-inch guns have an angle of fall of over 27 degrees at 6 miles, the range must be known within 13 yards in order to hit, and the chances of hitting with this gun are less than half that of the 12-inch gun. Under these circumstances it is evident that anything like the traditional storm of projectiles that make shambles of a ship's decks, at a distance of 6 miles, is pure "romance," though doubtless "stirring" enough for those who do not know better.
As for the appliance by which Mr. Emery, of the U. S. S. Washington and the "young lad in the top," were enabled to inflow the flight of a 12-inch projectile for a distance of 6 miles, and assure their anxious Captain that it went "straight into her," I may say, confidentially, that such an appliance, securely protected, is worth millions.
SUMMARY OF CONCLUSIONS.
- Great-gun pointers cannot be trained to be efficient marksmen; and if this could be done, we could not utilize their skill on account of the impossibility of their being able to identify the splash of their own shots when many guns are firing, because this requires an uninterrupted view of the projectile during its entire flight; which can be obtained only from a comparatively elevated position, well above the smoke and hot gases from the guns.
- Given a certain gun, in order to train pointers to the highest possible expertness in aiming, or in order to test the expertness they have achieved, firing on a training range is essential; that is, a range which bears such a relation to the size of the target used that a "good pointer" can hit it every time, notwithstanding all unavoidable errors.
- When the same target is used at a much greater distance than the training range, as above defined, the shooting is not a training for the pointer—it is blunderbuss training.
- Having trained pointers to the highest expertness in aiming that experience shows to be possible at present, they can be utilized at any range to make the greatest possible percentage of hits that the range (angle of fall), the unavoidable errors of the gun, and the conditions of the firing will permit—provided, of course, the range-finding is efficient.
- The three principal causes of missing are (a) errors of pointing, (b) unavoidable errors of gun-fire at sea (due to powder, etc.), and (c) errors in establishing the range.
- Of these, the first (the error of pointing) is by far the least; and if it could be entirely eliminated, the difficulties of hitting at long ranges would be but slightly decreased.
- As the present skill of the best pointers is such that no very considerable improvement in aiming is possible, it follows that the effectiveness of gun-fire at sea can be materially increased only by more accurate methods of finding, and keeping, the sight-bar range, and by diminishing the present unavoidable errors of the gun.
- With all of the unavoidable errors just mentioned, it is possible to make, with a certain gun, at a certain range, only a certain maximum percentage of hits on a target of a given size.
- If we actually make this maximum possible percentage of hits, say 20 per cent, at a certain long range, it shows (a) that our pointers have faithfully applied their skill by aiming accurately, as directed, and (b) that the range party has established the correct range. If we make less than 20 per cent, either the aiming or the range-finding has been inaccurate—we cannot tell which, but the chances are very large that the range-finding is at fault.
- With reliable pointers at the guns, long-range firing becomes, therefore, principally a training in range-finding—whatever the method employed.
- With unreliable pointers at the guns, long-range firing is an inexcusable waste of ammunition, because, when you do not make the percentage of hits that you should (say, 20% in this case), it is impossible to tell where the fault is, therefore the firing is not training for anybody concerned.
- In order to train new pointers, and maintain and test the skill of old ones, we must always fire at training ranges; and this practice can never be abandoned without an immediate loss of accuracy in aiming, which is the first essential to the success of gun-fire at sea; for if the pointers are not both skillful and reliable, all firing at battle ranges is practically useless as a training of the 'ship" in the second essential requirement—accurate range-finding. For upon accuracy of pointing and accuracy of range-finding depend the success that it is possible to achieve with the guns you are given to shoot with.
You must first train your pointers always to aim accurately, then train your "ship" to get the maximum results from their skill. As long ranges, even perfect pointers would be useless on an untrained "ship," and vice versa.
The foregoing explanations refer only to the vertical errors due to errors of pointing, errors of the gun and errors of range-finding; but, while these are not so easy to avoid as lateral errors, still the latter present many difficulties. The most serious of these can be compensated for only by fitting the guns with properly designed sighting appliances. It follows, therefore, that long-range firing underway, carried out with guns not so fitted, is profitable only to the manufacturers of ammunition.