Post 9u: Collision at Sea-What to do? Pt 4a Commanding Officers
Introduction
This post seeks to derive important High Reliability Organizing (HRO) insight by focusing on the senior officer involved in the collision between the ALNIC MC and the USS JOHN S MCCAIN, the MCCAIN’s Commanding Officer (CO). It goes beyond the two official reports illustrate what can be learned. This involves no mind reading or second guessing. The approach here deals only with the decisions he made and what they reveal about principles of HRO Risk Management and Watchstanding that many people, even HRO practitioners, claim to know, but seldom articulate until after a serious problem.
Collision Recap: Two CO Decisions
In previous posts, I noted that many problems existed on the JSM prior to the collision. These were latent errors in James Reason’s terminology. They included: routine steering system operating mode, insufficient system technical documentation, lack of formal procedures for making system changes, lack of watch stander proficiency, etc. All influenced the outcome, but none of the latent errors caused the collision that resulted in 10 dead U.S. Navy Sailors, 48 more injured, over $100 million in damage, and ruined careers. Two CO decisions are worthy of deeper analysis: planning to station the Sea and Anchor Detail (SAD) after entering the Singapore Strait Traffic Separation Scheme (TSS) and changing the configuration of the steering system immediately prior to entering the TSS.
The day before the collision, the CO decided not to station the Sea and Anchor Detail before entering the Singapore Strait TSS. The investigation reports provide no information on his motivation or reasoning. This was CO Decision 1. The Navy investigation report only notes that three of his officers tried unsuccessfully to change his mind (Sec 7.3, p.60).
Five minutes before the collision, the CO made CO Decision 2, modifying the ship’s rudder and engine control configuration while the JSM was overtaking the ALNIC MC, in the presence of many other surface contacts, traveling at 18 knots, and approaching the TSS in near-total darkness (NTSB Report, p.46).
I focus on CO Decision 1 in this post and will address CO Decision 2 in the next one.
Learning from CO Decision 1
We know nothing of the CO’s perception of the risk of entering a zone of high traffic density in darkness with an inexperienced watch team. The experience levels of the watchstanders are noted in the NTSB report (helmsman, lee helmsman, and BMOW p.34; the officers pp.17-18). No additional watchstanders were stationed. Stationing the Sea and Anchor detail was only one option for mitigating low experience. We don’t know why he rejected requests by three different officers to change his mind. It doesn’t matter. What he did matters more than what he thought. What can we learn from this?
First, start by recognizing that the CO has overall responsibility for the command’s safety, mission accomplishment, and everything else as identified in Navy Regulations. Good to know, but not exactly a guide for what to do. “[W]hat makes it tough to lead [is that leaders] know they need to listen, tell, structure and trust, but in what sequence? With what blend?” (Weick).
* Weick, K. Leadership When Events Don't Play By the Rules, retrieved from https://positiveorgs.bus.umich.edu/wp-content/uploads/Leadership-When-Events-dont-Play-by-the-Rules-Weick.pdf
What does the CO actually do to exercise his “overall responsibility”? How is this enacted by his or her behavior? One perspective is that he exercises his judgment to keep the ship safe by permitting personnel with specific responsibilities to operate safely to accomplish the mission. He trains and mentors personnel, sets priorities, and provides his experienced feedback for the decisions of his senior leadership team that impact safety and mission accomplishment. In this view, he only intervenes in situations when he deems it necessary to protect the ship and crew from harm. Then he steps back out of the way.
A consequence of this perspective is that the CO doesn’t fix problems or relieve personnel of their problem ownership, but rather ensures problems don’t persist by closely monitoring the performance of his leadership team. He is most effective when he doesn’t make many decisions. It is his obligation to allow his senior leaders to make decisions and only seek to change them if he judges that they are likely to produce unacceptable risk of harm or mission failure. This should be considered an obligation for three reasons. First, no one learns if someone else makes all the decisions. Second, people only feel responsible for the decisions they make. Third, the CO might be wrong.
It is the CO’s prerogative to be wrong, but being wrong in certain contexts can have terrible consequences. Being wrong about setting the SAD has much greater consequences for ship safety than decisions about the timing of breakfast. CO Decision 1 illustrates what I call HRO Risk Management Principles. In priority order they are:
HRO Risk Management Principle 1 is don’t accept more risk unless mission failure is at stake. Otherwise, what is the point of making the tradeoff? You must start with Principle 1 as a CO for any of the other principles to help.
Principle 2 is to empower people to question your risk perception and decisions. This goes far beyond telling them to do it. You have to make it safe. No yelling when you disagree or don’t like how they do it. You have to tell them that you *expect* them to do it, regularly ask for their opinion about risks, and you have to seriously engage with their concerns. You can use questions like “What do you think?”, “What am I missing?”, and “Do you see the risk differently? Why?” Practicing Principle 2 is not easy because egos tend to inflate with seniority and many senior people are not comfortable having others question them, even in private. Like improving any skill, you have to want to get better at it and practice it regularly.
Principle 3 is don’t force subordinates to accept more risk without additional controls. It doesn’t matter if the CO thinks the risk is acceptable if their subordinates don’t. THEY are the ones managing most of the operational risk. The CO’s role in risk management is to step in and assist subordinates with a better plan or emergency intervention if things go wrong. He can back his leaders up, but they can’t easily back up the CO because stepping in to reduce the risk is SOLELY the CO’s prerogative. As a form of professional development, a CO should seek an explanation of his subordinate’s risk management plan when he suspects it is too conservative. “What is the risk you perceive and why have you chosen this way to manage it?” is a good start. A good follow-up is “What will you do if …?” In reviewing a subordinate’s risk management plan, the CO has to judge whether they have the competence and skill to execute it. If not, the CO should state so and insist on a modified plan. This is a risk management dialogue that does not relieve the subordinate of the responsibility of designing the risk plan and executing it. Unless failure of the mission is at stake, the CO should let the subordinate execute their risk management plan, possibly putting his reservations in writing before the event to be shared for mutual learning afterward.
Principle 4 is that COs exercise their responsibility for safety best when not in the “operations loop.” While the CO may be the most skilled operator on the ship (if only in his mind), being an operator IS NOT HIS ROLE. When, not if the CO makes a bad decision, his responsibility “is absolute” (Navy Report, 7. Findings, p. 59). No one has the authority to overrule the CO and Navy culture is such that no one is going to try. This means the CO can be the backstop for decisions by his officers, but they can only back him up BEFORE he makes his decision. This responsibility asymmetry is the consequence of the CO’s authority and responsibility.
When the CO stationed himself on the bridge at 0115 for the approach to the TSS and the greater shipping traffic expected, he inserted himself into the operational risk management loop of the watchteam. The CO’s rationale noted in the NTSB report was revealing. He “expected traffic to be heavy and that he would be called often to come to the bridge” (NTSB report, p.9). This is just one sentence in the report, but is very meaningful if you have ever been an OOD. Normally the OOD makes reports to the CO, but he only calls the CO to the Bridge when he or she is confused or needs help to manage serious risk to the ship (e.g. collision, grounding). All CO standing orders state this. Stationing himself on the Bridge made the CO the de facto ship control senior supervisor, placing him directly in the operational loop for risk management. He was no longer backing up the OOD, he was managing the risk WITH the OOD just like he would if he were called to the Bridge in an emergency.
A CO that empowered his XO or Senior Watch Officer to question his risk decisions would think hard if one of them were to ask, “Captain, why do YOU have to assist the Bridge watch team?” It is easier than you think to fall into the trap of thinking that you are the best operator or the best risk manager on the ship. After all, excellent operational skills are what got you to the position of CO. If you fall into that trap, the only way to get out is to listen when the XO says, “Captain, we have other officers that should do that. You can still station yourself on the Bridge, but you shouldn’t be the safety monitor for the OOD.”
If three officers try to change your mind about a decision and you refuse, there are only two possible conclusions you can reach about the situation: a) you are the smartest person onboard, surrounded by idiots, or b) they see a risk that you do not. Each has profound consequences for risk management.
Not addressed in either report was an important risk consequence of CO Decision 1: the transition from underway watches to the Sea and Anchor Detail HAD to take place within the high-traffic, restricted maneuvering area of the Traffic Separation Scheme. This illustrates the danger of ignoring Principles of High Reliability (HR) Watchstanding. Like the Principles of HR Risk Management, these aren’t on a list, are not explicitly taught, and are seldom discussed until something bad happens. Then senior leaders state them as if they were obvious and personnel involved in the bad outcome should have known them. There are many Principles of HR Watchstanding, but I will only discuss those that are illustrated by the two CO decisions.
CO Decision 1 illustrates Principle A of HR Watchstanding: only conduct watch relief when conditions are stable and will remain stable until you take some specific action. The hundreds of changes involved in replacing watchstanders, adding new watchstanders, changing equipment lineups, and changing communications between control stations that result from setting SAD are second only to General Quarters (“Battle Stations” for non-Navy types) in generating operational chaos and disruption. A Risk Management Plan that requires you to violate Principle A is, by definition, a bad plan.
This concludes my analysis of the learning opportunities associated with CO Decision 1. In my next post, I will identify what can be learned from CO Decision 2.
