The naval officer has always been an applied scientist. Centuries ago he could navigate, which put him among an educated few; of course, he has always been an expert in rule of thumb mechanics. The industrial revolution brought a revolution in naval science, and the naval officer adapted readily. He mastered steam power, modern ordnance, and electricity; he even learned to exploit whole new worlds in the air and under the sea. Now the electronic revolution is in process, and the naval officer must become master of yet another major scientific field.
But this one is somewhat different. The electronic revolution is all-pervasive. It affects the propulsion of ships through automatic control systems (nuclear propulsion would not be possible without it). It provides stable platforms and computers for fire control, and more recently guidance systems for the projectiles themselves. It provides a whole new world of communications. Electronic senses—radar and sonar—far outstrip man’s own. And now, most recently, electronics is being given command responsibility. Electronic guidance systems fly some of our warplanes during almost their entire missions. In the F-106 interceptor, for example, the electronic guidance system literally flies the aircraft from soon after take-off to attack position. Having detected the target, it carries out the attack and fires the missiles at the proper time for the greatest kill probability. The guidance system then flies the aircraft back to its base (or a nearer one), and returns control to the pilot only for the landing. And this is not just a plan; the F-106 has been in quantity production since 1959. Similar developments have taken place in naval aircraft (the F4H and A2F, for example), and are also taking place afloat. The Navy Tactical Data System is designed to weld together an entire surface ship force into a defensive unit, making maximum use of all available weapons and detection devices in repelling an attack. In its initial form, the NTDS is not completely automatic, but, undoubtedly, it can become so. As the time available to react to a threat becomes less and less, the NTDS must become more and more automatic.
The naval officer can be pardoned if he views this with some dismay. In the first place, the new electronic systems are extremely complex; they require a “priesthood” of expert technicians, “programmers” and the like, who speak a strange language he does not understand. In the second place, he is disturbed at the loss of control during the crucial moment of combat—will he be reduced to the status of a helpless passenger? And where will it end? Will today’s officer live to see electronic monsters lurking in the sea without a man aboard, or with at most a handful of passive maintenance technicians? Or will there be some limit to the electronic revolution?
While it is dangerous to attempt to predict the future of a dynamic and revolutionary field, the future is not completely without roots in the past. Throughout history naval science has been influenced by certain factors. These factors determined the courses of earlier “revolutions” in naval science. They are still as influential as ever, and we may be justified in assuming that they will also determine the course of the “electronic revolution.” Let us try to identify these factors and, through considering their effects on the “electronic revolution,” to predict the outcome.
First, there are two basic factors which have always spurred on the search for improved technology. They are largely responsible for the “electronic revolution” and they can be counted on to continue spurring it. They are:
Accuracy. For centuries the technique of hitting a target with a thrown missile steadily improved. Weapons were built with greater precision, the projectiles were made more uniform so that their behavior would be predictable, and aiming techniques (later, fire control systems) got better and better. Now, technology has provided self-contained guidance systems which change the thrown missile into a target-seeking robot. Apparently this has solved the problem; the “missile thrower” need be aimed only roughly, and the missile itself will correct its course while in flight.
In reality, the self-contained guidance system has simply substituted one problem for another. Perhaps the old accuracy problem can be considered solved, but there is a new problem of building guidance systems having the required accuracy despite countermeasures. The science (art) of countermeasures against target-seeking robots is a new one and is shouded in secrecy, but we can expect to see it grow. The defense may use countermeasures such as jamming the missile’s electronic “senses,” setting up false targets, or attacking the missile with “anti-missile missiles.” The designer of the attacking missile, however, can anticipate these countermeasures and incorporate “counter-countermeasures.” The result is a new race between offense and defense much like that between armor and ordnance a generation ago; the pressure for increased accuracy, in this somewhat new and different form, is as great as ever.
Speed. There has always been pressure to increase both the speed of the missile seeking its target and the reaction speed of the defender in countering the attack. At the moment, the offense seems to have the speed advantage; the ballistic warhead, the Mach 3 jet, and the nuclear attack submarine are all too fast for conventional defensive reaction speeds. Great strides are being made by the defense, however. Electronic “senses”—radar and sonar—are being improved very rapidly and are constantly extending their reach in the air and under the sea. The human reaction time to note the information and take action has become impossibly slow, even with electronic senses, so electronic techniques are being applied here, too. The Navy Tactical Data System is a first step toward an automatic “conditioned reflex” for an attacked force; hopefully, by doing away with the laggard human, such systems can meet the challenge of the attacker’s speed. In general, the requirement for increased speed is one of the most critical existing today.
Until recently, this list of factors which spur on the search for improved technology would have included range, possibly at the top of the list. In the past, the most decisive technical advantage was always to have weapons that outranged the enemy’s. Modern missiles, however, can be built with essentially any range; they can far outreach the detection devices against which they must operate. Ballistic rockets of a thousand-mile range or more are now so small and light that they can be carried by any ship and by some aircraft. Except for a true intercontinental engagement, then, we conclude that weapons of any desired range can be-mounted on any ship—the problem of range seems to have been solved (at least largely), unlike those of accuracy and speed. The torpedo is an exception; its range is still severely limited by the characteristics of its medium. The SUBROC and ASROC systems have neatly solved the problem, however. In both, the torpedo is propelled through the air for most of its journey and drops into the water only near the end of its run.
In view of the continuing pressure of accuracy and speed, it appears that the technological wave should continue to race on indefinitely. But there are other factors acting in opposition to the wave which act to slow it down or temper it; in the opinion of the writer, they will eventually bring it to a virtual stop.
Complexity and Reliability. Rear Admiral R. K. James, former Chief of the Bureau of Ships, said: “The ultra-sophistication in our systems will soon defeat our operational forces—unless they have college degrees before they go to sea.” Electronic devices sometimes fail and, the more complicated they are, the more likely they are to fail.
In the early 1950s, a crisis was formally recognized in all the services; the electronic equipment was not working well enough to meet its operational requirements. Major efforts were started to improve the reliability of military electronic devices, and great improvements have been made. Yet the newest electronic systems are far more complex than their predecessors; the latest sonars have ten times as many electronic components as some older sonars still in use. The advanced electronic systems of the future are bound to be still more complex; systems have been proposed that involve another tenfold increase in electronic complexity.
It is optimistically assumed in many quarters that improvements in the reliability “art” will somehow solve not only the present problems, but also those of the far more complex systems that are proposed. Performance to date indicates that this assumption is not justified. It is more reasonable to assume that unreliability and difficulty of operation and maintenance, which go hand in hand, add up to a combination that will impose a very real limitation on the “electronic revolution.”
Flexibility. Modern electronic computers are amazingly flexible. They translate languages and run petroleum refineries; incorporated in “weapon systems,” they control whole complexes of military equipment. There is no limit to the flexibility of computers, either; experts have said they will be able to perform any operation that can be conceived by man. The key to this is the “stored program” of the computer, a series of thousands of elementary commands which, when executed in sequence, enable the machine to do complex and variable jobs. Since there can be infinite variation in the programs, the computer can presumably perform an infinite variety of jobs. It is tempting to assume that, because of this, the role of the computer in military operations can grow without limit.
But the source of the computer’s strength is also the source of its weakness. The computer will do anything its program tells it to, but it will not do anything that is not included in the program. The only intelligence the computer has is the intelligence of the men who designed the program; the only flexibility the computer has was built into the program in advance.
If we consider a computer-controlled, independent combat unit, we wonder what genius will be able to provide the machine with a program designed in advance to meet all conceivable combat conditions. Imagine, for example, our robot combat unit attacked by an unconventional weapon (say a human torpedo) after damage has already incapacitated part of the directing computer itself. The program would have to be flexible indeed to deal with the situation. Yet any human commanding officer would be able to take some sort of improvised action, in this or even less likely circumstances.
It has been suggested that this problem might be solved by “learning machines”; machines so programmed that they will derive their own instructions for the future on the basis of past experience. If such a machine could be incorporated in the combat robot, perhaps it would not need to be programmed in advance for all conceivable situations. The possibility of such machines is a matter of controversy. However, even if they are possible, they are unlikely to be useful in the combat robot—a commanding officer in combat is given little chance to learn from his mistakes. The combat robot must do the right thing every time (or nearly every time), so the “learning machine” does not seem to be a solution to the flexibility problem. Lack of flexibility, then, would appear to be a definite limitation on the use of electronics for combat control purposes.
Responsibility. The commander of a combat unit has, in war or peace, the responsibility to act in the best interests of his country. What constitutes the best interests of his country? How does he make decisions when these interests are affected? In fact, when does he even know that the interests of his country will be affected by a decision? The commanding officer has accumulated experience, knowledge, and skills during his career which fit him to bear responsibility, and he is held accountable if he fails to exercise it well. One finds it difficult to imagine a machine being held accountable for its actions. It is still more difficult to imagine a machine being entrusted with the making and executing of decisions involving the “best interests of the country.” It is probably safe to assume that this will never be permitted, at least as long as human nature is the way it is. No matter how capable the robot may be, or how flexible its programmed instructions may be, surely it will be restricted to the execution of a limited mission whose implications have previously been considered by a responsible human authority.
To summarize, five factors that influence the “electronic revolution” have been discussed. All arc affecting it now, and are almost sure to continue doing so indefinitely. The pressures for increased accuracy and speed act to increase the pace of the revolution, while the requirements for reliability, flexibility, and responsibility of command tend to slow or limit it. What is likely to be the result? Compromise, certainly, as always with irreconcilable requirements, but what compromise? It is, of course, speculation to predict the ultimate form the compromise will take. But this speculation is not entirely random, because the factors have already produced some of their effects.
The continuing demands for greater accuracy and speed cannot be denied. We will inevitably see defensive “conditioned reflexes” built into ships and entire forces. The defensive senses, having detected the attacker at the limit of their range, will consult the commander only for a brief “okay” before taking countermeasures (including counterattack) which are optimally planned according to principles of operations research. When combat is expected, indeed, the “okay” may be dispensed with. This development is not speculative; all of the necessary techniques exist, the pressure is present, and work on such systems is under way (the Tactical Data System is a first, very large step toward this). The same is true of the advances in accuracy of robot weapons; they are bound to occur. Not only will guidance systems become more precise, but as the enemy’s defenses become more able to deal with the weapons, we will have to provide them with “tactical sense,” the countermeasures referred to above. The weapons will take evasive action based on the enemy’s defensive actions rather than proceeding straight to the target. They will be able to mislead the enemy by jamming and even by carrying decoys—and by the same token, they will develop an ability to discriminate between the real enemy and the decoys he uses. The attacking weapons may even carry rudimentary defense of their own —small missiles to deal with the enemy’s interceptors. Again, the techniques exist to make all this possible, the pressure is present, and developments are now under way. During the next 20 years, there should be a fascinating race between electronic countermeasure and “counter-countermeasure.”
The offensive weapons will inevitably be highly complex because they must deal with the enemy’s defenses, and they will be unmanned because of the reaction speed requirement as well as for humanitarian reasons. Reliability limitations dictate that their missions must not last too long, however, and that the weapons must not be too complex. Their- flexibility will necessarily be limited, too. This will probably restrict the robot weapons to the performance of single, clear- cut missions; it will probably not be feasible to use these for more complex, less sharply defined assignments. All this means that the oceans and the skies will not be patrolled by robots, but rather by manned “weapon platforms” of some sort—as they are today.
Turning our attention to the manned “weapon platforms,” we divide them into three categories operating in the three media of naval action aircraft, surface ships, and submarines.
Aircraft. In its January 1961 issue, the Proceedings carried a description of the Navy’s F4H Phantom II fighter. An engineer in the Bureau of Naval Weapons was quoted as saying, “ ... It is my opinion that when we design an interceptor that will out-perform the F4H, it will not be an airplane at all. It will be a missile, or possibly a rocket ship, with built-in brains.”
The engineer may have been intentionally exaggerating, but we are probably justified in concluding that in the future Navy the high- performance fighting airplane will surely be a thing of the past; the weapons themselves will have taken over the task of actual combat. Pilots must already depend on electronic senses rather than on their own, and already the “guidance and control” systems of modern aircraft are being designed to take over the controls at the crucial moment of attack, or even for the entire mission as in the case of the F-106 referred to earlier. With the requirements for high combat performance removed, the aircraft will be able to emphasize other factors: endurance, electronic detection capability (of course), and perhaps altitude and cruising speed, if these should turn out to be important. The Navy’s proposed “missileer” aircraft would have been a forerunner of these. The helicopter will surely be important, carrying detection systems and robot weapons for many purposes. The “drone” helicopter, in particular, may have a bright future. High flying, long enduring craft hovering high over a force carrying defensive weapons also seem possible.
As time goes on, ballistic missiles seem to be taking over all types of pre-planned “strike” missions. However, the requirements for flexibility and responsibility indicate that one of the Navy’s specialties, tactical support, will remain. A long or medium range missile cannot be expected to have the accuracy required for close support, and—oddly—it may also be too slow. Even at ballistic missile speeds, it takes time to travel from a launch site 50 or 100 miles away. The Marine or soldier will always require close, responsive air support, and the only possible way to achieve this is by having manned aircraft in the immediate area of action.
The helicopter can surely fill some of the required roles, and it is conceivable that other forms of vertical take-off and landing aircraft can fill the rest. VTOL aircraft are unlikely to match the combat capability of conventional aircraft, but as was pointed out, this should not be necessary. VTOL aircraft may be able to match the endurance, range, and carrying capacity of conventional aircraft; if so, their obvious advantages may bring them into general use. Despite the apparent successes of the early trials, VTOL development programs are not being pursued very vigorously.
If the VTOL aircraft are able to displace conventional aircraft completely, there is an important corollary—they would cause important changes in the aircraft carrier as we know it.
Submarines. There is no question about the importance of submarines in the future Navy. The question that is being debated has, rather, to do with surface ships—can they retain any important combat role at all, in the face of nuclear attack submarines? With the performance and endurance that result from nuclear propulsion, with the already-proven capability to launch robot missiles through either sea or air while submerged, the submarine threatens to become supreme. There is basically only one reason for this, however, and it does not lie in the capabilities of the nuclear submarine, but rather in those of its opponents. Antisubmarine forces do not have a really satisfactory means of detecting nuclear submarines. Improved radar scours the air and the surface, and has enabled the surface ship with its robot defenders to survive the onset of supersonic aircraft. The surface ship depends on sonar to provide corresponding defense against submarine attack, and modern sonars are still no better than the sonic signals the sea returns; these are so far not good enough to provide the surface ship with a defense against the nuclear submarine. The issue of the submarine’s future, then, and that of the surface ship as well, both ride on the answer to this question: will electronics be able to produce underwater detection equipment that can provide early warning of nuclear submarine attacks?
This problem might be solved by a dramatic breakthrough. Some device may be discovered to which the sea is as transparent as the air is to radar. This is completely unpredictable, of course; it would be sheer speculation to build any predictions on such an assumption. There is another way, however, in which the problem could be solved. Two sciences, oceanography and electronics, are being harnessed as never before in a search for improvements in sonar. For the first time the nature of the sea as a sonic medium is really being studied, and at the same time new sonars are being developed that have much more power, sensitivity, and discrimination than those of the past. It is entirely possible that the conventional sonar may still solve the problem, through some combination of greater knowledge and better equipment.
There are two ways, then, in which the nuclear submarine may be stripped of its mantle of secrecy. There could be a “breakthrough” such as radar, or the efforts to achieve adequate improvements in sonar may be successful. In recent years, electronics has met challenges that seemed as difficult (missile guidance for one). It seems reasonable to assume that electronics can meet the challenge of the nuclear submarine in one way or another.
The submarines, then, will not have command of the sea but will have to contest it as in the past. They will have all the advantages of the present nuclear submarine, doubtless with improved performance. We can expect deeper dives, faster dash speeds, and probably smaller crews. They will fire robot missiles while submerged, either through the water or into the air. And, of course, they will take advantage of the same detection devices that render them vulnerable—a sword that cuts two ways. There will surely be attack submarines as now, and surely ballistic missile submarines. Some will probably be specialized to hunt their own kind, because even with new detection devices, the submerged position will probably still be best for submarine hunting. Some may take new forms: the secret attack transport for limited war, for example. As an extreme example, perhaps some could be “mother ships” performing picket duty against enemy submarines by sending out patrol robots carried aboard; small unmanned submarines with sensitive detection gear patrolling on preset courses. This is mentioned as an example, to show that the action of the three restrictive factors need not prevent dramatic electronic developments. The reliability problem could be met by overlapping the patrols so that failure of one unit would not open a gap in the coverage. The required flexibility is not extreme. The question of responsibility does not arise, because the patrol robot has a passive function and need take no offensive action. To convince the reader who still considers this a “science fiction idea,” we quote the New York World-Telegram and Sun of 29 September 1962: “ . . . Navy spokesmen say it is conceivable robots will be used in detonating and laying mines, locating submarines, sampling the ocean floor, salvaging operations and checking terrain too risky for deep sea divers.”
Surface Ships. Surface craft will naturally have many of the same functions in the future Navy that they have today. Passenger and cargo ships for non-combat uses; landing craft; minecraft; repair ships—all these and others will surely exist in much their present form. Some may be nuclear powered, if this is called for. Some of the smaller craft may either be hydrofoil or air cushion craft, where these techniques have advantages. A study of the future of these would be interesting, but we are concerned primarily with combat ships, the “weapon platforms” that carry out attack missions and counter the enemy’s.
There seems to be no doubt that the warships of the future will be nuclear-powered. Nuclear power does not give the surface ship the same dramatic performance boost it gives the submarine. Range is increased tenfold, however, and full speed can be maintained for long distances if necessary. One nuclear- powered ship can do the work of two or more conventionally powered ones; this fact alone seems to make nuclear power a must. Our latest nuclear powered, missile-armed surface ships should afford us clues to the future. These ships are USS Bainbridge, a frigate of 8,000 tons (heavier than some light cruisers) and the Long Beach, a heavy cruiser of 14,000 tons—lighter than some other heavy cruisers. The Enterprise is not included. As a carrier for conventional aircraft, she necessarily is the product of their requirements, and is therefore not as good a guide to the requirements of the future. The two ships have high cruising speeds, taking advantage of their nuclear power plants. They are unarmored; robot weapons carry powerful warheads and the armor a ship can carry would be of little use. They have centralized command and control stations. Allowing for their being entirely different types, they are not far apart in displacement. Taking these characteristics into consideration, we conclude that the surface “weapon platform” of the future should have something like the following characteristics.
The ships must be large enough to carry and maintain an adequate load of robot weapons, and to support heavy detection systems. At the same time they need not carry masses of metal in the form of armor or heavy guns and, being nuclear-powered, they need not carry quantities of fuel. Perhaps they will displace 8,000 to 12,000 tons. Each must be a multi-purpose ship, capable of fighting in the air, on the surface, or under it. Though all ships will, of course, be independent fighting units, there will probably be some ships designed for force command which will have computers and communication terminals for other ships to “plug into,” so that an entire force can become one automatic unit in battle. These command ships may be somewhat larger, because this equipment plus its added crew could be of substantial size.
The crews should be smaller, because much of the ship’s machinery will be automatic. The two largest groups should be for main, tenance and repair of the robot weapons and the ship’s machinery, and the command force for conning and directing the ship. To add a' touch of familiarity for today’s officer, there will be the deck force—with the same timeless responsibilities as ever, and probably of much the same size as today. Because of the automatic controls and electronic senses, the battle stations for all hands can probably be in a concentrated, protected area below decks that has some degree of radiation protection. While the ship will be conned from a bridge above decks in maneuvering and while entering and leaving port, in combat, human senses will be useless, so the crew is best placed below.
Summarizing all these considerations, we have examined the factors that press the electronic revolution forward—speed and accuracy—and concluded that naval combat of the future will depend on robot missiles in the water and in the air. We have discussed reliability, flexibility, and responsibility, factors which tend to moderate the electronic revolution. We have concluded that, because of these, the robot missiles will never be entrusted with lengthy or complex missions.
Rather, they will operate from manned weapon platforms in the air, on or under the sea. The weapon platforms will carry and maintain the robot fighting machines, will house the heavy detection gear, and will provide the command functions—both human (in the decisions about the use of the robot fighters) and electronic (in the direction of the robot fighters). The weapon platforms will be able to operate independently using their own tactical control equipment, or they will be able to form a single “defensive organism” reacting as a unit against attack. The weapon platforms, no longer themselves the instruments of combat, will be more standardized and will emphasize such characteristics as cruising speed, load carrying ability (in the case of aircraft), and endurance. The endurance of submarines and surface ships, multiplied already by nuclear power, should increase still further as crews become smaller—there are long cruises ahead.
The line officer may be startled by this picture of the future electronic Navy, even though somewhat reassured that he will still have a place in it. But the fact that is really startling is this—the Navy is nearly there already. With the exception of long-range submarine detection systems, not one development is predicted in this paper that does not exist already, either in service or nearly so. The “future electronic Navy,” as pictured above with its robot missiles and weapon platforms, will be realized when devices and systems that already exist are perfected and become general.
This paper has defined the limits of the “electronic revolution” through analysis of the factors behind it. It appears that these limits have nearly been reached: that the “electronic revolution,” in the sense of radical innovation and overthrow of existing methods, is nearly over. Electronic development will not stop; there will be the race between the guidance system and countermeasures referred to earlier, the continuing effort to extend the range of detection systems, and similar developments. But timeless naval requirements for reliability and flexibility, requirements dating back to the time of galley slaves, dictate that the role of electronics in the Navy has a limit—and that limit appears to be near at hand.