The assault upon Japanese-held Attu Island in May, 1943, was a Joint (Army-Navy) Operation that involved for the first time a complete exposure to problems presented by cold weather assault operations. That a faulty analysis was made in some particulars by the planners and executors will be evident at once when a study is made of the techniques employed. At the same time, however, perhaps by trial and error, much was learned that can be used to full advantage in cold weather operations of the future.
Some brief mention has been made in the press concerning cold weather exercises conducted in February, 1948, in the Kodiak Area. This should remind us that there is a tremendous fund of information already at hand with regard to sea and land force operations in frigid climates. It is doubted if anything new was uncovered in the little publicized Kodiak tests. Extant information concerning equipment, clothing, combat conditioning, and the like, together with special procedures and techniques required by low temperature considerations, may be found in the war-inspired reports emanating from our World War II efforts in Iceland, Newfoundland, Attu, Cold Bay, Amchitka, Shemya, and Kiska. The testing of new equipment, developed since the war, presents, of course, fresh material for consideration and must be pursued to the fullest extent.
For the United States, Attu was the only real cold weather objective in World War II which featured a determined amphibious assault against tenaciously defending enemy forces ashore. The amphibious phase of the assault lasted from May 11, 1943, to May 18, 1943, inclusive; the operations on shore continued until May 31. Thus it may be seen that the Battle for Attu was, in fact, a prolonged affair and included actual cold weather fighting by sea and land units which transpired under the disturbing presence of an armed and determined enemy. This particular Japanese force was heavily reinforced by a strong ally—weather.
The concept of the operation to capture Attu was really an offshoot from an original joint decision between the Commander in Chief, U. S. Pacific Fleet, and the Commanding General, Western Defense Command, to take Kiska as the first retaliatory blow against the Japanese who had stolen into the Aleutians in June of 1942. In December of that year directives were issued to Commander Amphibious Force, Pacific Fleet (ComPhibPac), to get busy on plans to seize Kiska. His headquarters were located at this time on the Broadway Pier, San Diego, a spot chosen for security reasons, within the confines of the Headquarters, Eleventh Naval District.
Situated in close proximity to the ComPhibPac headquarters was a Marine command, styled the Amphibious Corps, Pacific Fleet, and located at Camp Elliott, near San Diego. This Marine commander was called in by ComPhibPac and requested to produce an estimate and plan for Kiska. During this guessing phase, the Commander North Pacific Force (ComNorPac) in Adak, Alaska, also worked on a plan for Kiska.
In early January of 1943 a joint plan to capture Kiska was issued by the Commander in Chief, Pacific Fleet, and the Commanding General, Western Defense Command. Assigned troops were to be elements of the 7th Infantry Division (Army) with naval forces to be allocated, as needed, by CinCPac.
Up in the far north, about this time, the Commanding General of the Army Alaskan Defense Command, whose headquarters were in Adak, received news of the impending operation with undisguised satisfaction. This joy was shared by those of his sub- , ordinates who were acquainted with the plans in the making. At long last the Aleutian forces would have a chance to strike back and repay the Japs! The Commanding General immediately assumed that his troops would be well represented in any landing force assigned this job. (His assumption turned out to be erroneous.)
Commencing in December, 1942, the 7th Division had been readying itself for combat in Africa. The training ground was the dusty desert of California. Days and nights of tireless practice in desert warfare were spent in the consuming heat and enervating dust clouds of the wasteland. Tanks, weapons, and the men themselves went through every conceivable test to qualify them for duty on the African continent.
For reasons unknown to the writer, but probably dictated by its ready availability in California, the 7th Division was suddenly earmarked for the Aleutian job. The choice was a surprise to our forces already in the Aleutians and somewhat of a shock to the few 7th Division planners that were informed. The secrecy surrounding such a momentous decision in those days was such that only a small number could be informed of the radical change to a cold weather objective.
The Commanding General of the Seventh Army Division instituted keen and intelligent preparations to capture Kiska, while occupying himself personally with all the details required of a troop commander for such an adventure. In early January, 1943, he reported, with his command, to ComPhibPac for duty. The latter by now was also called Command Task Force Three.
The Army division housed itself at Fort Ord, near Monterey, California, several hundred miles north of San Diego. This intervening distance between the headquarters of the various units most intimately connected with the operation in itself introduced difficulties which contributed to some of the eventually disappointing features of the operation as a whole. It was found expedient and necessary to send teams of Naval and Marine officer instructors to Fort Ord to deliver the know-how of amphibious operations to the uninitiated Army troops. This requirement was made very complicated by reasons of security restrictions. Most of the soldiers, and the majority of their officers, could not be told that they were destined for Aleutian duty.
A series of training exercises were conducted for the purpose of thoroughly indoctrinating the green troops into the then somewhat mysterious amphibious game. All the weeks of intensive training for desert warfare had to be discounted and a new approach undertaken.
Late in January a Joint Staff was set up in San Diego, under the direction of ComPhibPac, to work out all details of the projected attack upon Kiska. This staff included representatives from all commands that were to be concerned, including those based in the Aleutians. In assigning staff duties and missions, wherever possible and when considered advisable, Navy or Marine Corps officers were paired with Army officers (flying and ground). The working staff was constituted as follows:
Assignment |
Officers |
Chief of Staff |
Cmdr., USN |
Deputy Chiefs of Staff |
Col., USMDC, USA |
Intelligence |
Lt. Col., USMC, Capt., USA |
Operations |
Cmdr., USN, Lt. Col., USA |
Air |
Col., USMC, Lt., USN, Col., USAAF |
Communications |
Cmdr., USN, Lt. Col., USA |
Naval Gunfire |
Lt. Cmdr., USN |
Medical |
Capt. (MC), USN |
Areological |
Lt. Cmdr., USN |
Secretary |
Lt., USN |
In addition to agents of the involved commanders that were actually present on the Joint Staff, the principal echelons involved despatched certain staff officers back and forth by air between the headquarters at Adak, Fort Ord, and San Diego. This permitted a continual interchange of essential information.
The comments and recommendations of the “cold-weather-wise” staff officers from the far north were considered in detail by the Joint Staff and the respective commanders. Certain implementation of the recommendations was made in the form of special equipment that was made available.
It was finally decided to employ a considerable number of troops, as a reserve, from the Alaskan Defense Command. The idea of taking the 7th Division to Alaska for training and combat conditioning was considered. This thought, finally discarded, intended an entirely realistic approach to the problems promised by Kiska. All ship-to- shore training was finally conducted in the southern California Area. Key personnel were sent down from Alaska to convey some cold weather lore to the higher planners and the top commanders, and these reinforced the positions of Alaskan members of the Joint Staff when it came to “selling” certain cold weather procedures (so as to insure inclusion in the final planning).
A considerable amount of realistic training ensued at San Clemente Island, California. This practice, with complete full-dress rehearsals for combat, included the landing of troops under live air and naval gunfire support. During the training phase, most of the fire support ships scheduled to participate in the Kiska venture were on hand at San Clemente. The aircraft were, unfortunately, only on a loan basis and were not destined to go north with the attack force. This meant that the planes to assist in the Kiska attack would receive no preliminary training in air support or even in amphibious operations. The aircraft assigned to the operation remained in Alaska (11th Army Air Force).
In early March, ComNorPac suddenly conceived the idea of attacking Attu rather than Kiska. This fortuitous action, like the one to attack Leyte rather than Mindanao in late 1944, inspired by reports from Admiral Halsey’s planes that Leyte was lightly held by the enemy, was timely and sound. The basis for the decision to attack Attu was that such a move would not only by-pass Kiska but would at the same time be less costly for us because of the comparatively few Japanese thought to be on Attu. Intelligence again demonstrated its vital role in helping to decide wars.
The Commander in Chief, Pacific Fleet, reacted quickly and acted favorably on the suggestion of ComNorPac. Work upon the Kiska planning was discontinued forthwith and intense preparations to take Attu were commenced. The approval of CinCPac for the new plan was undoubtedly influenced by the attractive reduction in forces that would be required for Attu. (He was heavily committed in other areas at this time.)
In late March, ComNorPac became Commander Task Force Sixteen (CTF 16) and ComPhibPac was designated Commander Task Force Fifty-One (CTF 51). These titles thereupon prevailed for the remainder of the operation. The Pennsylvania became the Task Force Fifty-One flagship, in the absence of a more appropriate Amphibious Force Headquarters Ship (AGC), and elaborate arrangements were made on board the “Pennsy” to handle the huge operations and communications load characteristic of amphibious operations. For the actual attack, CTF 16 was to be in supreme command from his shore headquarters in Adak while CTF 51 would command the assault force, on the spot, at Attu.
The nebulous nature of the untrained air support prospect manifested itself early in the conferences. Finally, the CVE Nassau with her valuable aircraft was assigned to the Assault Force. But the Nassau was given a primary mission of providing air cover for the sea elements since it was entirely possible that the Japanese would launch air attacks from their bases in the Kuriles. (The CVE planes actually delivered a considerable amount of air support in the assault, due to the absence of Japanese aircraft.)
The air support aspects of the Battle for Attu were most unusual. Combat inexperienced Army Air Force pilots were to be depended upon to support the troops from the air. An Army Air Force Colonel, a member of the Joint Staff, was to coordinate the air support from a B-24 flying over Attu! Quite a far cry from air support as we know it now.
One of the unique features of the operation was the plan to land certain scouts from a destroyer, an APD, and two submarines. These scouts, with reconnaissance elements attached, were to be landed at widely separated points in order to feel out the Jap defenses and support the attacks upon the beaches. The troops chosen for this task organized along Commando lines. The planners were unable to forecast the rugged trials and tribulations in store for these hardy fellows.
A somewhat strange but vivid highlight of the planning was the fact that the Assault Force Commander (CTF 51) and the Landing Force Commander were to embark in different ships and were to perform their respective functions in two widely separated areas at the objective. While CTF 51 was to be in the Pennsylvania, the Landing Force Commander would be found in the APA Zeilin. There were many worthwhile considerations that determined this disposition of the commanders. The overpowering reasons for this procedure were that CTF 51 desired personally to supervise and influence the struggle at Holtz Bay (secondary landing area), while the Landing Force Commander felt that he should be in the vicinity of Massacre Bay (main landing area).
The entire task force left San Diego-San Francisco in late April and set course for Cold Bay, Alaska—the “jump-off” point.
Prior to departure from the west coast of the United States, cold weather gear, clothing, and equipment were loaded on board various ships, in camouflaged containers, in such utter secrecy that it was possible to completely safeguard the fact that the force was bound for the frigid zones. Orders provided that none of the packages could be opened until the ships were well at sea. Unfortunately, some few vessels failed to make connections with the “man who distributed the boxes,” and thereby had to depart without any heavy weather equipment. (The equipment was later received at Cold Bay.)
The secrecy attendant to the operation was so strictly observed that many Commanding Officers of ships, especially transports, could not be given their missions until they were well at sea and enroute to the objective. As a matter of certainty, many vessels got underway on April 24, for Cold Bay, Alaska, thinking that they were about to participate in another routine training exercise. Some thought that the expedition would eventually be headed for a jungle objective. This sort of secrecy turned out to be a boomerang and most detrimental to proper appreciation of the problems to be faced squarely in cold weather warfare.
A humorous sideline on the stringent security regulations for Attu came when a destroyer squadron commander reported to CTF 51 for duty. This officer was fresh out from duty on the staff of the Commander in Chief, U. S. Fleet, where he had been entrusted with secrets vital to the entire Allied Cause. He may have even been the one who originated or inspired the Cominch phrase, “classified information can be divulged only to those who need to know.” At any rate, he was shocked to find upon his arrival in San Diego that he was now among those “who did not need to know.” Another not so humorous item concerned an officer decoder on the PhibPac staff who accidentally told his wife that the force was headed for the cold country. This individual’s subsequent contacts with the FBI, OSS, and other erstwhile cloak-and-dagger bodies left him in somewhat of a sweat but assuredly a wiser and sadder man.
Cold Bay was the scene of final conferences and last minute plan adjustments, and it was here that local Alaskan ship pilots made their first important donation to the effort about to take place. They painted a gloomy picture, however, of uncharted shoals, navigational hazards, and other stumbling blocks. One of these pilots, entering the amphibious business for the very first time, was later to serve well as beachmaster for Admiral Turner in the Central Pacific’s more famous and more grandiose campaigns.
The information received at Cold Bay relative to navigational hazards in Massacre Bay was completely unknown up to this time. Important changes had to be made, as a result, in the transport and fire support areas for the main landings. (This should be a lesson to future planners. In this case the Assault Force Commander was not in supreme command in the theatre of operations and was required to depend upon higher authority for intelligence of the target area. Complete knowledge of a projected attack area must surely be in the hands of the commander responsible for executing the attack plan if he is expected to operate to best advantage. A lack of such knowledge in the future may result in disaster.)
As soon as the conferences were over and final plans distributed, the task force prepared to sortie from Cold Bay. After a bad weather delay of twenty-four hours the ships finally got back into the open sea and set course for Attu via Amukta Pass.
Immediately after passage through Amukta, the Bering Sea became almost glassy and the ensuing fueling operations were completed in California-like sunny weather. This phenomenon and paradox of the cantankerous Aleutian weather was completely confusing to those embarked in the ships. It was not for long, however, since the characteristic cold and fog soon returned with a bang, the usual aftermath in that part of the world. Along with the shift to usual weather came a despatch from the “Gods on Mount Olympus,” changing D-day ahead from May 7 to May 9. (ComNorPac headquarters, together with that of the Commanding General, Alaskan Defense Command, was located on the summit of a promontory on the island of Adak. This elevated sanctum was dignified with the title “Mount Olympus” since direction of the over-all Aleutian war emanated from here. The humorous allusion to the seat of the all powerful Gods of mythology was not entirely facetious. Immense power was wielded from this site.) A subsequent postponement, again because of weather, set the day ahead to May 11.
On May 9 and 10, after having received word that a Japanese reinforcement group was approaching Attu from the Kuriles, CTF 51 got orders to sally forth with three battleships and four destroyers to do battle if contact could be made. The transports scurried to the north and east to safety and to keep out of the fight.
The search for the Japs did not materialize so the Assault Force Commander and his ships rejoined the transport group, in a dense fog, near midnight, on May 10. The fog was so thick that many near collisions resulted. Two destroyers actually collided and received such extensive damage that both had to be withdrawn from the operation. One of these, the Sicard, was to have been a landing craft control vessel—an all- important job—and her removal at this late hour caused considerable worry in the flagship Pennsylvania.
D-Day, May 11, 1943, found all forces converging upon Attu as planned. The dense fog persisted throughout the approach and through most of D-Day. The scout and reconnaissance elements were landed by the submarines with little difficulty, but there was a considerable delay in getting ashore the units that were embarked in the destroyer and APD. (All these parties got ashore by noon. The scouts that were landed by submarine were destined to pass some harrowing days. Before finally being rescued they suffered all kinds of privations, including running out of food and getting lost in the snow. They had a considerable numbered casualties from frostbitten feet.)
The reconnaissance men and scouts, landed north of Beach Red near Holtz Bay by the destroyer Phelps, reported no enemy and that conditions were favorable for landing.
The troop landings were made in two areas. Red Beach was scheduled to receive a battalion as a secondary effort, while the main landing, in force, was to be made in Massacre Bay on beaches Yellow and Blue.
Red Beach was one of the most treacherous and difficult landing points selected during the entire war. The narrow water approach to this beach was filled with large rocks and restricted the number of entering boats to not more than one at a time. After clearing the approach the remaining run to the beach itself was so short that the boats could not build up enough speed to beach properly. Once ashore, the attackers faced terrain that was precipitous to men and equipment. They had actually to scale a bluff before they could establish themselves. It turned out that this particular beach soon came under fire of Japanese batteries and the movement inland really became tough. The surprise approach in the fog completely confused the Japs, however, and a good foothold was soon gained at this point. Beaches inside Holtz Bay could not be used, since entering boats would have been under enveloping and continuous fire from ashore.
At Massacre Bay, scene of the main landings, things were a bit more favorable. All boats reached the beach without difficulty and the troops got ashore without too much trouble.
The landings at Holtz and Massacre Bays were made without opposition except for light and sporadic enemy fire. This was made possible mainly because of the thick fog but also because the Japanese were emplaced well inland from our selected landing beaches. (They had manned the beaches for a few days before May 11, but had withdrawn to their inland strong points when the Jap commanding officer decided that we were not going to attack Attu.) The enemy headquarters did not discover that our troops were ashore until about 1600 in the afternoon of D-Day. They heard the preliminary bombardment but assumed that it was just another hit-and-run affair and decided that the safest place for them was in their holes in the ground. (Their quarters were literally holes in the ground and resembled rabbit warrens more than human habitations.) These preliminary bombardments had been delivered by North Pacific Force battleships, cruisers, and destroyers at regular intervals, prior to the arrival of our assault forces, and they undoubtedly contributed to the ability of the troops to land undetected.
Beach barrage rockets were on hand, installed in LCM landing craft under supervision of scientists from the California Institute of Technology rocket project. When the rocket carrying craft made their approach to the firing positions the fog was so thick that the land proper could not be located. The boat officers then, on their own initiative, dumped the rockets overboard and used their boats for their primary purposes of unloading the transports. Rockets were actually used later, in some instances, against shore targets. These rockets came from a stock supplied by the Navy to the land forces and were actually fired by the latter.
No air support was delivered on D-Day because of the poor visibility. The gunfire support consisted of fire upon targets considerably (but safely) removed from the front lines of our troops. For instance, the supporting battleships maintained fire all day upon the Chichagof Harbor Area. During this day it became necessary to airdrop supplies to the scouts who had landed on Scarlet Beach. (These troops continued to be a liability until rescued.)
On D-plus-1 Day the visibility improved to such an extent that assistance from all supporting arms was afforded the troops, much to their advantage, in their push forward. The Japanese exhibited a decided reaction and went so far as to fire upon some of our ships with their artillery.
During the late afternoon of D-plus-1 Day, the Pennsylvania was attacked by a Japanese submarine. A torpedo passed close aboard, but was avoided by the quick thinking of the captain of the flagship. The submarine was pounced upon and destroyed by the destroyer Edwards, assisted by the Farragut. This submarine was believed to be on her way out of Holtz Bay after delivering supplies and perhaps evacuating key personnel (pilots and the like).
An interesting feature of the operation manifested itself about this time. Battleship observation planes, hampered in their usual work by poor visibility, enlivened their somewhat boring spotting and antisubmarine work by dropping a few depth charges here and there, upon Japanese shore installations. Considerable cheering damage was inflicted upon the enemy in this manner.
It was not until late on D-plus-2 Day that our troop artillery was able to open fire after having landed on Red Beach. It took that long to get the pieces emplaced and made ready for firing.
On D-plus-3 Day a considerable amount of close land fighting between our forces and the Japanese took place. There were sufficient breaks in the fog to enable offshore watchers to observe the hand-to-hand combat. The writer saw one soldier move up a steep hill, under the protecting fire of his fellows and the devastating fire of the enemy. Time and time again he could be seen tossing hand grenades into Jap-occupied holes in the side of the hill. This individual later received the Medal of Honor for this exploit. He captured this commanding position practically single-handed.
A considerable amount of close air and naval gunfire support was delivered to assist the Holtz Bay troops on this critical day. (D-plus-3 Day.)
The first few nights ashore were worse than uncomfortable for the troops. Many had landed without sleeping bags and spent a miserable first night. Feet became wet because of the wrong kind of footgear, and very soon news began to drift back to the flagship that there were several hundred cases of severely frostbitten feet among the troops. This alarming news was sadly confirmed, and at the end of the operation it was determined that many men were so affected; foot amputations were required in the worst cases. The shoes worn by the troops were the regular GI leather ones. It was later determined that shu-packs, consisting of bootlike, rubber footed, leather topped construction were the type that should have been provided.
During the night of D-plus-3 Day, the Landing Force Commander decided that he was not making enough progress. He attributed his slow pace to the difficult terrain coupled with the elusively concealed positions of the Japs. (They seemed to just sit in the hills, out of sight in the fog, and take pot shots at our soldiers.) The General also began to suspect that the Japanese forces were more numerous than our intelligence had calculated.
About this time several of our CVE Nassau planes were lost in the turbulent Aleutian air currents over the island of Attu. These air currents build up to tremendous force in a sudden and violent fashion. The local name given to such a disturbance was “williwaw.” The characteristic air agitations -increased the difficulty of our aircraft operations.
On D-plus-4 Day, the situation ashore appeared to be worsening and it was thought that fresh reinforcements would shortly be required. The Commanding General of the troops ashore had his mind on the reserve troops that were waiting for further orders in Adak, several hundred miles away.
When the reinforcements, did arrive they were provided with proper footgear and equipment according to their long experience and service in the Alaskan Defense Command. As far as is known, these units experienced no frostbitten feet or other exposure difficulties traceable to improper clothing.
The amphibious phase of the operation was declared complete on May 18, 1943, and the Assault Force Commander, in the Pennsylvania, departed for Adak that date. Some transports and supporting ships were left in the area, but to all intents and purposes the remainder of the campaign proceeded as a purely land affair.
A few days after departure of the Force Commander, the Japanese made their one air attack, from the Kuriles, with their “Hetty” type aircraft. Our destroyers and the gunboat Charleston repulsed the attack so successfully that most of the enemy planes were destroyed with no damage to our ships. This was the sole Jap reaction to the entire operation except for the submarine attack previously mentioned.
The Jap troops fought on until May 31, when organized resistance was overcome. The final fanatical banzai attack, in the manner later to become commonplace, was of frenzied nature and was vividly described by Robert Sherrod in his Time magazine columns. The Commanding Officer of the Jap defenders, Colonel Yamasaki, met his fate on May 29 in one of these savage, suicide attacks. Many accounts of the enemy killing their wounded and committing suicide by holding hand grenades to their chests were received with foreboding by the American public as a token of the type of the long war still to be fought in the Pacific.
The types of Jap soldiers encountered at Attu were of the highest order. They were huge specimens and had been thoroughly trained and acclimated in Manchuria prior to use at Attu. The Japanese expected these troops to give a good account of themselves, and they did so. Among the bodies identified there were also found some few aviation pilots. Their planes had been destroyed by strafing or bombing and they reverted to defense duty. It was believed that some pilots were evacuated by submarine, and several were sunk in the submarine that tried to hit the Pennsylvania with torpedoes. Of the total of about 2,300 defenders of Attu, only 29 were captured. The rest perished.
The landings upon Attu by our amphibious forces were efficiently executed. Over 11,000 troops were landed without the loss of a man! Complete surprise was achieved, and the weather was a major factor in our favor, in this respect. As far as the beach assaults were concerned, they were made under conditions exactly duplicating nighttime operations. Full use had to be made of radar since the thick fog was equivalent to complete darkness. The assault could not possibly have been executed without electronic gear. Much of the naval gunfire was fired in full radar control with but few glimpses of the island itself—much less the designated target areas. The lack of accurate charts and the presence of unknown shoals and navigational hazards were of continual concern to the commanders of all units. The battleship Nevada, in particular, steamed back and forth in Massacre Bay with little more than a prayer to guide her.
Much of the unloading, particularly in the Holtz Bay area, was done while the ships were under enemy gunfire. This caused but little delay in the final result—disaster for the Japs.
Shore Fire Control Parties, serving as links between the troop battalions in assault and the fire support ships, performed continuously after they set up ashore. Their employment at Attu was probably as extensive as in any subsequent Pacific operation. The exposure to snow and cold played havoc with these personnel, as with the assault troops. The naval officer in charge of each Shore Fire Control Party lived in a foxhole and was, to all intents and purposes, a soldier.
Operating in frigid climates, especially where poor visibility is the rule, demands that due consideration be given to the distinct possibility that ceiling at or near zero will deny air support for assaulting troops as well as possibly affect the gunfire support. In other words, weather, and all its implications, must be carefully considered. Good weather should of course be chosen, where possible, in order to take advantage of all supporting weapons. It must be realized that, in many cases, troops can advance in snow or ice when properly supported by air and naval gunfire. Without such assistance they may get pinned down and stay down for unacceptably long periods.
The lessons from Attu are many. The quality of the planning and execution of the amphibious attack deserves the careful study and full appreciation of those who expect to fight in cold weather in the future. The mandates that should stick into our minds evolve from the following considerations:
(a) Common sense security measures.
(b) Properly acclimated and cold weather trained sea and land elements.
(c) Adequate and proper clothing and equipment.
(d) Effect of weather upon supporting operations.
(e) Adequate intelligence, including reconnaissance, where possible, of objectives.
(f) Full appreciation of medical problems involved in cold weather.
It is the bounden duty of all of us to become intimately familiar with all that was learned during World War II with regard to fighting sea and land forces in cold weather. If a conscientious approach to this problem is not made, we shall surely find ourselves, in the event of war, making a frantic, last- minute effort to gird for fighting in snow and ice—with only a hasty glance at information available. A leisurely, detailed study, during peacetime, should pay dividends.