Of all the fallacies with which the conduct of war is haunted none seems to be more persistent than the idea that to convoy merchant shipping is a defensive move in terms of strategy, and therefore one to be avoided if possible. Those who hold such opinions almost invariably go on to canvass the superiority of the alternative—and allegedly offensive— approach to the problem of shipping protection—namely to “seek out and destroy” the enemy who intends to attack shipping, instead of waiting for him in the vicinity of the convoy.
The purposes of this article are, firstly, to discuss how this fallacy has arisen; secondly, to prove by argument and statistics that the convoy system achieves its purpose far more effectively than any alternative strategy and cannot therefore be described as defensive; and thirdly, to suggest how the fallacy can be eliminated for good and all.
Let us first consider academically the word “convoy.” It undoubtedly existed in the very early days of the English language and came to us from France. The earliest reference given in Murray’s great dictionary is dated 1375. Nor did it originally have anything to do with the sailing and escort of merchant ships. It meant “to accompany,” as in the poet Lowell’s line “May sunniest hours convoy you from this land of ours.” In the 17th century Milton used it in that sense, and Dryden seems to have been particularly fond of it, as in his phrase “Your convoy makes the dangerous way secure.” Two hundred years later we find it applied, by Nelson among others, to merchant ships. Macaulay in his History of England uses it in a similar sense several times, but he applies it to the men-of-war instead of the merchantmen. It will thus be seen that at the beginning of this century there was good authority for accepting that the word meant “to sail in company,” for applying it to groups of merchant ships with or without warship escorts, and for attributing to the latter the function of defending the former. Mahan, for example, repeatedly uses the word in that sense, as when he wrote in his Influence of Sea Power on The French Revolution and Empire, 1793-1812, “In fact . . . the result of the convoy system, in this and other instances, warrants the inference that, when properly systematized and applied, it will have more success as a defensive measure than hunting for individual marauders—a process which, even when most thoroughly planned, still resembles looking for a needle in a haystack.”
Before the turn of the present century there had, however, been no hint that in any influential quarters convoy was regarded as defensive in terms of strategy. That view seems to be a development of quite modern times and to have come into existence in spite of the proven experience of many centuries of maritime war, and in flagrant contradiction of all that experience. It thus came to pass that in World War I the introduction of convoy was most strenuously opposed until in 1917 the condition of Britain had become desperate. At the start of the Second World War it was again regarded with serious misgivings, and today similar doubts seem to be arising once more. We shall return later to the study of the effects of convoy in both wars of this century, for it is necessary first to consider how this outlook has come to pass.
Of contemporary statesmen none has had a greater influence on the analysis of war, as well as on its conduct, nor placed the free nations more deeply in his debt, than Sir Winston Churchill. To discover his views on our subject must therefore be an important requirement. Nor does he leave us in any doubt on the matter. In The Second World War Volume I, pages 362-3, he wrote:
“The privateer, the raiding cruiser, and above all the U-boat have . . . exacted a heavy toll upon the life-line of our commerce and food supply. A prime function of defense has therefore always been imposed on us. From this fact the danger arises of our being driven or subsiding into a defensive naval strategy and habit of mind. Modern developments have aggravated this tendency ... I always sought to rupture this defensive obsession by searching for forms of counteroffensive . . . I could not rest content with the policy of ‘convoy and blockade'.” (This writer’s italics) In that quotation the reader will find Sir Winston’s vivid awareness that wars can only be won by offensive blows; but he will also find that he stigmatises the convoy system as strategically defensive. Now, if the dictionary sense of the word “convoy” quoted earlier, namely “to accompany” and “to defend,” were the whole purpose and effect of such a strategy, Sir Winston’s dislike of it would be justified. But it is, in fact, quite certain that convoy has always brought consequences of a far wider nature than mere defense; and the reason is that it places in the hands of the convoy escorts an opportunity to counterattack which can hardly arise in any other circumstances. The enemy must, as Mahan and many others pointed out, approach the convoy to find his prey; and only by attacking him can the escorts frustrate his purpose. How then can convoy be described as defensive? Opportunities for offensive action of the type which every fighting man desires are produced by it—not once, but constantly.
Let us now illustrate this by discussing the achievements of convoy in both worlds wars. It has already been remarked how reluctant the British Government was to introduce convoy in 1917, and contemporary history leaves us no room for doubt either on that score or regarding the immediate success it achieved. For three years the Allies, but especially the British had suffered appalling shipping losses. Relief came only with the start of Atlantic convoy, and by the year following its introduction the relief had become so complete that it amounted to a major victory. Of 20,000 Allied ships which sailed in convoy in the Atlantic and British home waters between 1st August and 31st October, 1918, the U-boats sank only 25; and of the seven U-boats sunk by naval forces in that same period every one fell victim to the convoy escorts. Can such results possibly be classed as defensive warfare? If it is desired further to reinforce the case for convoy in the 1914-18 war, a brief glance at events in the Mediterranean in 1917-18 will do so very forcibly. On that station the British Commander in Chief (Vice Admiral Sir S. A. Gough-Calthorpe) viewed convoy with considerable distaste. In spite of heavy sinkings it was not introduced until November, 1917, and then only by Admiralty order. The Commander in Chief’s strong preference was for what he called “an increased and unceasing offensive,” and his main effort took the form of a partly-fixed and partly-mobile barrage across the principal U-boat exit route in the Straits of Otranto. Convoys were stripped of their escorts to meet the vast requirements of this “offensive” proposal; and its accomplishments were almost negligible. During the whole life of the barrage only one U-boat was sunk there, and little if any delay was caused to their passages to and from the Adriatic. In strong contrast to this, out of the twelve U-boats sunk in the Mediterranean during the last two years of the war nine fell to the convoy escorts—and that in spite of their attenuation to meet the needs of the Otranto barrage.
At the end of the first German war the case for convoy and against “offensive” patrols and hunting had thus again been proved to the hilt, and the British official histories made no bones about it. Unhappily, we did not then possess the complete German records of the U-boat campaign, and it was not until they came into British and American hands after the second war that a really comprehensive study of the results obtained in the first war became possible. Though the historians of the 1914-18 war were, we now know, absolutely correct, the fact that they lacked the data to prove their conclusions beyond any doubt may have been a contributory cause to their not being heeded in between the wars. However that may be, it is certain that Admiral Beatty, who had commanded the Grand Fleet in 1917-18 and subsequently became First Sea Lord, had the clearest understanding of the value of convoy. “Above all” wrote his biographer W. S. Chalmers, “he was a firm believer in the convoy system ... He argued that from an offensive point of view submarines would be attracted to the convoys, where it would be easier to find and destroy them”; an opinion which is entirely confirmed by reference to Beatty’s own correspondence with the Admiralty. Precisely similar opinions are to be found in the writings of Admiral Beatty’s American colleague, Admiral W. S. Sims. In his report of 29th June, 1917, to the Secretary of the Navy, we find the emphatic statement: “It therefore seems to go without question that the only course for us to pursue is to revert to the ancient practice of convoy. This will be purely an offensive measure (this writer’s italics), because if we concentrate our shipping into convoys and protect it with our naval forces we will there by force the enemy, in order to carry out his mission, to encounter naval forces which are not embarrassed with valuable cargoes, and which are a great danger to the submarine.” After the long -awaited introduction of convoy the same officer wrote, “Thus from the first day that this method . . . was adopted it was an unqualified success in defeating the submarine campaign.”
In spite of this formidable body of experience it was not long after the end of the 1914-18 war that doubts, and more than doubts, regarding the efficacy of convoy began to be heard in British naval circles; nor was the fallacy of the “offensive” school slow in re-appearing. Throughout the 1920’s the experience so hardly won, and the facts so laboriously collected by the historians, appear to have been consigned steadily to oblivion; and British opinion became virtually unanimous in the view that merchant shipping was best protected by endeavouring to seek out, pursue, hunt, and harass whatever enemy might threaten it. The fallacy spread to the Royal Air Force and seems also to have impregnated the U. S. Navy. In 1935, with a new challenge on the seas plainly approaching, the Admiralty’s spokesman was confidently assuring the House of Commons that, in the event of a new war, convoy would not be introduced unless it was forced on us. The old arguments about its alleged disadvantages were resurrected, and the Air Ministry reinforced such arguments with the opinion that ships would be in much graver danger from air attack if sailed in convoy than if allowed to proceed independently. Happily, these views underwent substantial modification in the few remaining years of peace, mainly because it was realized that the Germans would very probably again wage unrestricted submarine warfare. In that event it was admitted, though reluctantly, that convoy would be necessary; and arrangements for its introduction were therefore pressed ahead. The sinking of the liner Athenia on the day that war broke out was in a sense a fortunate occurrence for Britain, for it left no one in any doubt that the quicker convoy could be introduced the better. We had in fact returned to and re-affirmed the decision of 1917. Nonetheless, during the first two years of war, Britain dissipated a vast amount of her precious flotilla strength in hunting operations, just as had happened in the Mediterranean in 1917-18, and at the price of reducing her convoy escorts sometimes to vanishing point. The hunting operations, it is hardly necessary to add, were uniformly unproductive. After about eighteen months of war, and very heavy shipping losses, there were, however, few doubters left in British naval circles. The convoy escorts had achieved important successes in the Atlantic in March 1941, and on the Gibraltar route in the following December.1 Escorts were being scraped together from every possible source, and the Admiralty’s policy was to convoy all shipping, outwards as well as homewards, as soon as it could be arranged. Air escort was, however, still in its infancy, and there was the strongest reluctance in Royal Air Force circles to divert any fraction of their bomber strength from attacking German towns and industry to escort duty in the Atlantic. As late as 1942 we find the Air Ministry calling the use of aircraft at sea “a predominantly defensive role” compared with attacking industrial targets, and Air Marshal Lord Trenchard, stating that “two-dimensional (air) operations in the Atlantic . . . are purely defensive.” In fairness to the air authorities, it must, however, be stated that the Admiralty repeatedly pressed for the bombing of U-boat bases and building yards, and that a large proportion of our resources was devoted to those objects. Only since the war has it been realized how little was accomplished by such means. The British Bombing Survey and the United States Strategic Bombing Survey are unanimous in their conclusions that the results achieved were much less than was believed at the time. Once again the much-advertised “offensive” measure did not produce the results expected;2 and all the while shipping losses were continuing at a very high rate due, in no small measure, to shortage of convoy air escorts.
Such was the state of affairs at the time of Pearl Harbor. But British doctrine and experience had meanwhile been reaching the U. S. Navy in a steady stream and had been re-circulated very widely by the Navy Department to its own ships and establishments. It might, therefore, have been expected that preparations to convoy American shipping would have reached an advanced stage by the end of 1941 and that convoy would have been quickly introduced as soon as the United States found itself at war. This, however, did not occur. True a “Convoy and Routing Section” on the British model was established in June 1941, but it was May, 1942, before Admiral King made it a section of his headquarters as “Cominch,” and from January to April of that year shocking losses were suffered among ships sailing independently off the American eastern seaboard and in the Caribbean. Admiral S. E. Morison’s brilliant account of the measures taken to deal with that onslaught by the U-boats make it plain that the slender surface and air forces available were in the main used to hunt for the enemies and to patrol the inshore shipping routes— and, until April, 1942, not to convoy. It is not for a British historian to discuss how this came about; but he may justifiably state that his own country’s records make it appear that the U. S. naval authorities were not convinced of the value of convoy and preferred to hunt. Morison’s account lends support to that view where, for one example, he has written, “Every submarine spotted was hunted; but the hunters, few in number, had not yet acquired the necessary technique to kill.” The same historian also quotes a U. S. Naval officer’s opinion that “the Navy does not like convoys. It is a purely defensive form of warfare” (this writer’s italics); and a remark of his own, about “Optimists who predicted that convoys were the answer to submarine activity were themselves answered by what happened in the Caribbean during the second half of 1942,” makes one feel that, at any rate when those words were written, Admiral Morison himself had some doubts on the subject.3 In fact, the possibility that during those months the “hunters” would have accomplished better results had they been used as convoy escorts receives strong support from the immediate success of the “interlocking convoy system” when it was finally introduced.
To return now to the Eastern Atlantic, the year 1942 saw an expansion of another “offensive” in the form of air patrols by Coastal Command against U-boats crossing the Bay of Biscay. Had our convoys been adequately escorted at the time there would have been much to commend in these harassing operations, which certainly caused the enemy discomfort, delays, and a few losses; but the convoys still lacked long-range air escorts and were still suffering heavy losses. Only one squadron of “Very Long Range” aircraft, operating from Iceland, was available for escort duty; and the accomplishments of the Royal Air Force’s No. 120 Squadron of Liberators leave no room for doubt regarding what stronger air escorts could have accomplished had they been made available earlier. Not until the spring of 1943 was that desirable result achieved; and the consequences were then startling. After having suffered very heavy shipping losses in March of that year, a sweeping victory was gained in April and May; and of the 56 U-boats sunk in those two months 36 were destroyed by ships and aircraft operating as convoy escorts or in support of convoys. Doenitz thereupon abandoned the battle of the convoy routes. The reason was, so he said, that his losses had increased to about one-third of all the submarines at sea— losses much too high. Could that possibly have been achieved by a defensive strategy? Finally, when Doenitz renewed the battle later in that same year, he suffered a second defeat as crushing as the first. Between September, 1943, and May, 1944, only twelve merchantmen were sunk in convoy, and in retaliation for each of these losses the escorts sank six U-boats.
The events of the last war, as of the previous one, do therefore seem to show beyond dispute that every convoy is a compact, offensive task force, that the final victory at sea owed almost everything to the successes achieved by counter-attacks made around the convoys, and that there is not, and never was, any ground for stigmatising the convoy strategy as defensive. Perhaps the most telling evidence against that fallacy is to be found in the enemy’s records. Again and again did Doenitz state in his War Diary, and also stress at Hitler’s conferences, the dire consequences to his U-boats produced by Allied surface and air convoy escorts. To quote but one example, in July, 1942, he publicly warned the German people that U-boat casualties were bound to rise once they had to grapple with escorted convoys. Constantly, throughout the whole long Atlantic struggle, did he try to find unescorted targets; and the only two periods which the U-boat commanders described as their “happy times” occurred when they accomplished precisely that—in the northwest approaches to the British Isles between June, 1940, and March, 1941, and in American coastal waters during the first four months of 1942.
Though we are here dealing mainly with the U-boat campaign, because it was the most serious threat to Allied maritime control, it may also be mentioned that if operations against surface raiders are analyzed the same conclusion is inevitable—that hunting for them was wasteful of effort and rarely succeeded in finding “the needle in the haystack”; whereas convoy and the careful disposition of forces in strategic areas utterly frustrated the enemy’s purpose.
To reinforce the foregoing arguments, it is now intended to quote two sets of statistics which have been compiled with the greatest care and with complete impartiality.4 They show in tabular form the results achieved from the beginning of World War II up to the end of May, 1943, when the U-boats suffered decisive defeat, and the results achieved firstly by “offensive” measures and secondly by convoy.
These two tables together show how the convoy system inflicted 65% of the enemy’s losses as against 23% attributable to “offensive” measures, while merchant ship losses in convoy amounted to only 28% as against 72% among “independents,” and stragglers.
Bearing in mind the weight of the evidence available, from the enemy’s sources as well as from our own, by 1956 one might have expected the issue to be beyond dispute. Yet signs are not lacking that the old fallacy is again rearing its ugly head. Quite recently First Lord of the Admiralty Thomas was reported as saying in a speech on an important public occasion “I can at least give some answers to those who prefer to think that the inventions of today have relegated navies to the defensive role of which the convoy system was the expression in the last two wars.”
Table I |
||
U-Boats Destroyed by “Offensive” Measure and “Convoys”* |
||
|
Number |
Percentage of total U-boats sunk |
1. U-boats destroyed by surface ship and aircraft “hunting” patrols |
21 |
9.4 |
2. U-boats destroyed by air patrols in transit areas |
24 |
10.6 |
3. U-boats destroyed by mines laid for offensive purposes |
6 |
3.0 |
4. U-boats destroyed by bombing of bases, yards, etc. |
Nil |
Nil |
Total accomplishments of “Offensive” measures |
51 |
23 |
Total U-boats destroyed by surface and air convoy escorts |
150 |
65 |
*The total U-boat sinkings from which these percentages are calculated exclude 14 U-boats destroyed by various agencies other than those here tabulated, and 14 U-boats lost “cause unknown.” |
Table II |
||
Ships Sunk by U-Boats When Independently Sailing and in Convoy |
||
|
|
Percentage of total sinkings |
1. Ships sunk sailing independently |
2,363 |
— |
2. Ships sunk after straggling from convoys |
208 |
— |
Total |
1,571 |
72 |
3. Ships sunk in convoy with surface escorts only |
604 |
— |
4. Ships sunk in convoy with air and surface escort |
16 |
— |
Total |
620 |
28 |
Is it possible, then, to take any step which might, once and for all, inhibit the fallacious thinking about the purpose of the convoy system? This writer fears that the “defensive” description is so deep-rooted that only by adopting a complete re-definition of the strategy can that desirable purpose be accomplished. The adjective “defensive” has for so long been married to the noun “convoy” that only by a divorce can we end the unhappy connection. The new definition must place full emphasis on the offensive function of the convoy escorts; it must at the same time accept that their operational conduct aims to protect the ships placed in their charge; it must include the suggestion that shipping control and routing form a part of the same strategy; and in these days it will be convenient if the new definition can be expressed by a terse, easily-remembered “short title.” After much thought this writer has concluded that the best alternative he can offer, which meets all these needs, is that which heads this article. If Convoy is buried, though not without shedding a tear of regret on account of its great accomplishments in the past, and Capros is introduced then it may quite quickly come to be accepted that by far the most effective attack on the enemy’s commerce destroyers is Counter Attack Protection and Pouting Of Shipping (CAPROS).
1. Editor’s Note: See Captain Roskill’s War at Sea, Vo). I, pp. 364-5, and 478-9 for the British side; and “The Scharnhorst-Gneisenau Team at Its Peak,” page 852, August, 1956 Proceedings.
2. Editor’s Note: See “Experiment in Total War,” page 837, August, 1956 Proceedings.
3. Editor’s Note: For a detailed account see S. E. Morison’s The Battle of the Atlantic, 1939-43 (1947) and The Atlantic Battle Won, 1943-45 (1956). Both are in The History of U. S. Naval Operations in World War II.
4. I am indebted to Commander F. Barley, R.N., and Lieutenant Commander D. W. Waters, R.N., for the research which has enabled these tables to be produced.