Society entrusts the dam safety community with a critical responsibility. Failures of dams can inflict greater damage to life and property than failure of any other civil engineering works. Besides the direct consequences of failure the loss of drinking and irrigation water supply, and of power generation capability on which society has become so reliant can be equally as serious. Dam safety is typically paid little attention by the public, largely because the dam safety community does a good job in maintaining the safety of these critical structures. As such, there is some degree of complacency in the public’s attitudes towards dams, the safety of which is taken for granted.

The dam engineer, however, cannot afford complacency, recognising that dams are dynamic structures restraining the forces of nature, and requiring the highest vigilance and attention throughout their construction and operational life. The dam engineer further recognises that just because a dam has performed safely for many years, this condition will not necessarily continue. In the struggle between a dam and the forces of nature it must restrain, a long period of apparent peace, and the sense of complacency into which it is only natural to be lulled, can come to a gradual or abrupt end. It is the dam engineer’s responsibility to gather intelligence data through a surveillance programme that will give ample warnings of escalation of the struggle, and to have in place plans to react to unfavourable developments. A nation’s military is faced with an even greater public trust: the protection of the nation’s security from external aggression.

The public’s perception of the military is that it exists only to fight, and in time of peace the military is largely taken for granted, consuming resources many would prefer to see expended in different ways. However, it is during peacetime when the military performs perhaps its most important function: the gathering and dissemination of intelligence data as to the real or potential intentions and capabilities of its possible opponents. Based on these data, the military must have in place contingency plans, just as the dam engineer must have emergency response plans in place. The military is no more immune from an extended peacetime-induced state of complacency than are the public or dam engineers. This state of complacency can have dire consequences.

History’s most famous example of this occurred on the morning of 7 December 1941, when the US Pacific Fleet, at its anchorage in Pearl Harbor, was taken completely by surprise in an attack by carrier-based planes of the Japanese Imperial Navy. The Japanese attack was successful despite a wealth of intelligence data gathered by the US that pointed clearly to a major Japanese attack in the Pacific. The exhaustive post-war inquiries concluded that had the intelligence data gathered been appropriately disseminated, communicated, interpreted and acted upon, the debacle suffered by the US could have been prevented.

These problems can also afflict those responsible for dam safety. Dam engineer Ralph Peck could in another context have been speaking about Pearl Harbor when he said: ‘I would venture that nine out of ten recent failures occurred not because of inadequacies in the state of the art, but because of oversights that could and should have been avoided because of lack of communication among parties to the design and construction of the dams, or because of over-optimistic interpretations of geological conditions. The necessary knowledge existed: it was not used.’ Such unfortunately was the case at Pearl Harbor.

The battle begins

At 6:00 am on the morning of 7 December 1941, the first attack wave of 183 Japanese planes were heading for the US Pacific Fleet base in Pearl Harbor and surrounding airfields. Two Army operators at Oahu’s northern shore radar station detected the Japanese air attack approaching and contacted a junior officer who disregarded their report, assuming they were American B-17s due in from the US west coast.

Pearl Harbor was not in a state of high alert, nor in fact was it in any state of alert whatsoever. Based on available intelligence, Pearl Harbor concluded there was no reason to expect an imminent air attack. Aircraft were left parked wingtip to wingtip on airfields, and anti-aircraft guns were unmanned with many ammunition boxes kept locked in accordance with peacetime regulations. No torpedo nets were protecting the fleet anchorage, despite the US Navy’s recognition that air-launched torpedoes could run in the shallow waters of Pearl Harbor.

At 7:53 am, the first Japanese assault wave, consisting of torpedo planes, dive bombers and high level bombers with fighter support commenced their attack. The raid lasted until 9:45 a.m. US losses were five battleships sunk, three damaged, three light cruisers sunk, three destroyers sunk, and 188 aircraft nearly all destroyed on the ground. Casualties were 2335 servicemen and 68 civilians killed, and 1178 wounded.

The US Pacific Fleet had been moved from San Diego to Pearl Harbor after continuing Japanese aggressions in China. This move was intended to represent a ‘deterrent’ against continued Japanese aggression in Asia and the Pacific. The Japanese response to the presence of the Pacific Fleet at Pearl Harbor was to welcome the opportunity to rid itself of a potential threat. Pearl Harbor proved that ‘deterrence is an elusive doctrine that works only against those willing to be deterred’ (Prange et al, 1986). The forces of nature acting against dams can be restrained but, like the Japanese in 1941, are unwilling to be deterred. As such, those responsible for dam safety must constantly be on alert, and must know all that is knowable about how these forces are interacting with the dams within their charge. US failure in this regard cost them dearly at Pearl Harbor.

Despite its shortcomings, the US military was very capable and powerful. How could it have been caught so completely unaware? To place this surprise in context, it is necessary to first understand the intelligence they had amassed which, when revealed after the war, came as a shock to the Japanese and most Americans alike.

Magical communication

Well before 1941 the US had devised a decoding system named Magic – a form of cryptanalysis that had allowed translation of encoded transmissions. The US had been intercepting and decoding Japanese transmissions before the Japanese attack on Pearl Harbor had even been contemplated. For its day, Magic was a technological triumph, yet superior technology alone did not save the US from disaster. Similarly, the dam safety community today has at is disposal a remarkable array of instrumentation and technology that provides remote monitoring, automatic data acquisition, even the triggering of alarms. Pearl Harbor served as a warning that this advanced technology will no more save the dam safety community from calamity due to human error than was the case for America’s military in 1941.

  By late November, the Americans had amassed and decoded dozens of intercepts that, had they been pieced together, led to but one conclusion: a major Japanese task force, including the majority of the Japanese aircraft carriers, was at sea and about to strike. War was inevitable, and these intercepts alone should have been enough to put all Pacific commands at the highest level of alert. The US command in Hawaii was aware but waited for the command from Washington to go to high alert. Washington, however, knowing Hawaii had received as least the general gist of these intercepts, assumed that Hawaii would take appropriate action without waiting for orders from Washington.

On 6 December the Magic operators intercepted a 14-part message from Tokyo to Japan’s ambassador in Washington. President Roosevelt had just sent an eleventh hour appeal for peace directly to the Emperor of Japan, underscoring how tenuous Washington considered the situation. The intercepts were decoded rapidly and communicated immediately to the President and the Secretary of State. The intercepts confirmed that, precisely at 1:00 pm in Washington, 30 minutes before the Japanese strike force was to appear over Pearl Harbor, the Japanese ambassador was to inform the US Secretary of State that Japan was severing diplomatic relations. The Americans knew of this before the Japanese ambassador did. This was, in effect, a declaration of war. The news was communicated to General Marshall, who then ordered that all Pacific commands be alerted.

An alert was sent out by the War Department but via commercial telegraph. The urgency of the message was not communicated to the telegraph operator, who saw fit not to transmit it immediately. In fact, the alert did not arrive in Pearl Harbor until after the Japanese attack had concluded. General Marshall did not follow up to confirm that his Pacific commands had received the message and responded appropriately – he assumed’ that they had. For their part, Marshall’s commanders in Hawaii, Admiral Kimmel and General Short, had been receiving, in dribs and drabs, similar messages for weeks. The warning from the War Department may not have been enough to spur them to high alert even had it arrived in time.

The US had amassed an impressive array of intelligence data, but failed completely in making effective use of it. Had they fully appreciated its implications, and communicated that information to those who needed to know, would it have made much difference on the morning of 7 December? Collection, interpretation, and communication of intelligence data are but part of the equation. The other essential ingredient is having appropriate defences in place, together with emergency action plans for potential scenarios.

An inexpensive matter

The waters of Pearl Harbor are very shallow, and the US Navy believed, until several months prior to the Pearl Harbor attack, that the waters were too shallow for plane-launched torpedoes to successfully run. Once the Navy did recognise the possibility, however, it should have been a simple, and inexpensive matter to install torpedo nets to protect the ships from torpedo attack. The nets were not in place because the US did not believe the Japanese intended to carry out such an attack, even if it was accepted they were capable of mounting one. However, history has repeatedly shown that if an enemy can launch a certain kind of attack, in all probability he will.

Those responsible for dam safety would do well to heed this lesson of military history. For its part, the US Navy would have done well, in the context of the torpedo nets issue, to have subscribed to the ‘belt and suspenders’ approach to dam design advocated by Arthur Casagrande. A significant improvement in defensive capability could have been achieved quickly and at minimal expense.

Another deficiency was the lack of aerial reconnaissance to detect an enemy striking force prior to the enemy being able to launch an attack. A report prepared in August 1941 recommended as an urgent priority that such reconnaissance be established, but the report was not acted upon until after 7 December. The report stated that the cost of such reconnaissance would be less than that of one modern battleship, ironic give that five battleships were sunk in the attack.

The approaching waves of Japanese planes were detected by radar 50 minutes before they appeared over Pearl Harbor. This was plenty of time to at least scramble Hawaii’s air defences, which were in fact formidable. Yet a junior officer, believing the planes to be American B-17 bombers, neglected to inform a superior.

Here was another examples of superior technology being insufficient to protect against human error. Without appropriate checks and balances, and established procedures in place, no technology is so robust as to compensate for the inadequacies of the human element. This is another lesson to be heeded by those responsible for dam safety.

Valuing personnel

In the congressional investigations into the Pearl Harbor debacle that started in 1945, much criticism was directed at the US intelligent practices. A recurring topic in that criticism was the unfortunate lack of respect afforded those in the military specialising in intelligence. Such specialisation was looked upon as a dead end and an impediment to an officer’s career advancement.

In dam engineering, the glamour and glory of design and construction quickly fades to the droll drone of surveillance and maintenance activities. Once that happens, the safety of the dam is out of the designer’s and constructor’s hands, and is entrusted to dam safety personnel, who must monitor and judge the performance of the dam against nature’s relentless onslaught. Their role in protection of the public is no less significant than that of the designer, and is perhaps greater.

Dam owners would be well advised to recognise the value of those personnel in their charge who carry out dam surveillance and maintenance. Neglect of those personnel can, as happened in the case of Pearl Harbor, lead to faulty intelligence practice. This neglect carried with it a terrible cost at Pearl Harbor, in the aftermath of which the congressional committee recommended that ‘intelligence must not be regarded as just another tour of duty, but as a profession’. So it should be with dam safety in general, and those personnel responsible for dam surveillance in particular.

Relevance to dam safety

Experiences gained at Pearl Harbor are relevant to the dam safety community today. Those responsible for dam safety are very much in the intelligence business. They are required to devise a surveillance programme, gather and analyse data, interpret and report on it, and take appropriate action. They are also required to have in place defensive systems, either as design features or as contingency plans, for dealing with all conceivable unfavourable developments.

Examples of further ways in which a dam surveillance programme can fail are listed below:

• Some failure modes are not amenable to monitoring.

• Failure to anticipate and monitor for the most unfavourable conceivable conditions, not just expected conditions.

• Failure to quantify threshold levels at which certain actions must automatically be triggered.

• Incorrect interpretation of the data.

• Complacency.

• Failure to consider temporal trends rather than just ‘spot’ readings or observations.

• Lack of redundancy in the surveillance system.

• Failure to communicate effectively.

• Failure to obtain periodic independent, objective review that looks at the forest rather than the trees.

Perhaps the most important lesson of all is that the Pearl Harbor debacle happened to an organisation that despite its shortcomings was extremely powerful and capable. Those in the dam safety community who consider themselves immune from fatal errors of parallel consequence should consider the failure of Teton dam in the US on 5 June 1976. In the words of Peck (1980), Teton dam ‘had been designed and constructed under the supervision of an organisation considered by engineers and the public alike to be among the most authoritative and experienced in the world’, the US Bureau of Reclamation.

The lesson is an obvious one but bears repeating: no individual, organisation or system is infallible.
Related Articles
Failure of the Teton dam: geotechnical aspects