No question this is embarrassing for the military. These ships are supposed to be able to detect and shoot down missiles arriving at mach 3. But they didn't detect a 40000 ton cargo ship traveling at 15 mph? A cargo ship that was broadcasting its GPS location constantly?
That said, this article really adds nothing to that which is already known about the incident, besides that someone tweeted "people weren't paying attention".
Even discounting anything high tech, there's more than one person on watch, and the destroyer is much more nimble than a cargo ship. I imagine the postmortem will show a crazy breakdown across many processes.
What word do you suggest he use? If anything, I feel like 'postmortem' most commonly implies an after-action report - with the specific word 'autopsy' being used to describe, well, autopsies - at least in US english.
This is presumably a US English v English thing then, as post-mortem wouldn't be the term used here usually, and certainly not if someone had died (unless if was being performed on them). To someone from the US it would perhaps seem a little distasteful to describe the investigation as an 'autopsy' - hence my comment.
Oxford "British and World English" dictionary, post-mortem: "An analysis or discussion of an event held soon after it has occurred, especially in order to determine why it was a failure." https://en.oxforddictionaries.com/definition/post-mortem. Maybe not common in the UK, but I'd say it's perfectly correct word usage.
I grew up in USA Midwest, though I have been accused of a "southern" accent. "Postmortem" is completely idiomatic in this situation, but if you'd care to suggest a better term we'd all learn something.
The problem is, I think, that someone, something, somewhere, did detect the Crystal, but didn't act on it. The technology almost certainly worked right. It was likely the people that screwed everything up. The Crystal was probably on autopilot with no one (awake?) in the bridge and it would appear that the Fitzgerald was in the same state.
That said, for the US this is a pretty serious embarrassment, because it screams "not combat ready".
I am pretty sure that in a combat situation this collision would not have happened. It's the routine that is dangerous plus the mentioned "I am bigger" vs. "I am military" both assuming they have the right of passage.
It's extreme clickbait, it induced me to read the story thinking it would detail evidence of the cause, instead of just more wonderings why, without even any reasoned speculations as to causes.
That commander should be relieved of his duties at once. There is absolutely no reason a modern warship should have an "accident" like this. How could there not be automated systems screaming "collision course ahead".
This is the same thing as always, people are lulled into complacency and then accidents happen.
What did the commander have to do with this accident? He was asleep in his stateroom.
Edit: The captain is responsible for his crews performance and the navy way means he'll never be given command again, regardless of the results of the investigation.
But my response was to someone who apparently thought the captain was gunning the engine while playing chicken with the cargo vessel. For one specific point, the destroyer wasn't on collision course, the cargo vessel was.
The commander bears at least some responsibility for everything that happens on his command. The Navy will do a thorough investigation and probably lots of things will be to blame, but "he was asleep at the time" is not a get out of jail free card.
True but he should not have allowed a climate/attitude where this could happen. It was his responsibility to keep his sailors at the ready and never allow them to become complacent.
Not sarcasm. I was replying to someone who assumed some arrogant naval captain was playing chicken with a merchant vessel and I was just pointing out he wasn't involved in the decision making.
And we do know he will lose his command and never get another one, whether he's shown to have culpability for his crews failures or not. It's the navy way.
* I was replying to someone who assumed some arrogant naval captain was playing chicken with a merchant vessel*
Odd, because when I go back to the OP, I don't see any indication that suggests he thinks the captain made an active decision to be reckless.
Maybe they edited after you replied?
Or is it the quotes around "accident" in this sentence that lead you to interpret their comments this way?
There is absolutely no reason a modern warship should have an "accident" like this.
If so, I read this differently. In my view, an accident is something that occurs due to unexpected changes in a situation. For example, if a car crash occurred because a deer jumped across the road and lead a vehicle to swerve, I'd use the word "accident".
But in cases of negligence, I find the word "accident" makes it far too easy to blame the situation rather than the people. In this case, in my view these two ships colliding was no "accident" in that there must have been clear, conscious human decisions that lead to the event, even if those decisions were simply "I'm going to have a nap now (and thus be negligent in my duties)".
Incidentally, I think we're far too willing to refer to car crashes as "accidents", and for this same reason.
What happens if the CIC (where the radar operators would sit), lookouts, and bridge team were all properly posted with trained watchstanders, who collectively dropped the ball?
Don't get me wrong, the CO is still "accountable" here, but while the buck may stop with him, there is going to be plenty of room for blame by the time this is all settled out. I don't care who the CO is, there's no excuse for a Navy watchteam on a modern destroyer to miss on oncoming container ship so badly.
It was his responsibility to ensure someone competent to do so was on watch, so while it is a failure of personnel management rather than one of ship-handling, it is still the Captain's responsibility.
Was he asleep when he chose who to be on watch? Why are you fighting the captain's corner so hard in the face of good reasons for someone in a position of authority to take responsibility for a failure? It's refreshing, frankly - non-military life doesn't usually work that way.
We don't know what specifically caused the accident yet. We do know the Captains career is over regardless of the cause. It's still possible the captain did his job well, and still one of his crew members screwed up.
And comparing a ship's captain to a babysitter? When a babysitter comes to my house I expect them to be attentive for 4 hours and then to go home. I don't expect captains to never sleep during long voyages. There is a whole process around allowing the captain to sleep, and waking them for any possible problem, which obviously didn't occur her.
But by all means some posters can compare his job to babysitting.
They don't do a random draw to see who gets to be the captain. You're supposed to be able to operate under extreme conditions, including sleep deprivation, if it's necessary.
Collisions can happen anytime a ship is underway, which in the U.S. Navy can happen for months on end. Navy ships can be refueled and resupplied without having to visit port.
So yes, provision has to be made for a CO to sleep at some point. And the Navy accordingly has provisions for this -- provisions which failed here. The CO will be held accountable for maintaining -- or not -- an environment conducive to safe navigation, but in any event the blame will not attach solely to the CO here.
I never served in any navy but I've learned a lot of USN through my readings of history and other topics.
In USN, the captain is ultimately responsible for everything that happens to the ship and on the ship. I admire this trait in USN.
1) I really do feel bad for the poor captain as this is most likely not his fault directly but someone else's.
You will probably see one of these on the final report for placing the blame on the captain:
- Not keeping crew trained/motivated enough to be alert to avoid the accident.
- Not giving clear instruction to the crew to wake him up when another ship is near by.
- Having a hostile environment where the subordinates avoid waking him up or dealing with him directly.
- Not keeping equipment in proper order
I'm not saying any of it may be true or not, but usually that seems to be the common thread.
2) Check the case of 2009 USS Port Royal grounding. A USN missile cruiser grounded near Honolulu Airport. Airline Passengers coming in for landing at Honolulu Airport could see the grounded ship.
The captain was relived of his duty within days. He didn't even get to stay long enough to command recovery effort of the ship.
The final report states following as the reasons of the grounding.
combination of a misread navigation system, a sleep-deprived commanding officer, broken equipment, and an inexperienced and dysfunctional bridge team.
The ship had just finished a major overhaul and they were busy testing the system, resulting in a captain with not enough sleep.
The direct reason for actual grounding?
Military contractors (engineer/technician) had flown into Hawaii to work on the ship during the overhaul and were on the ship. They were on the ship during the testing out in the sea which lasted longer than planned that day. And the contractors really really wanted to get back to their hotel that evening to play a round of golf the next day before returning to US mainland. To allow the military contractors to leave the ship on the small crafts that evening safely, someone ordered the USN cruiser move closer to the port. And it was during this movement, the cruiser grounded. This is not in the wiki but I remember reading about this in the local newspaper. Because some dudes wanted to play golf before leaving Hawaii, they ended up being one of the direct causes of causing multi million dollar damage to a USN cruiser.
3) Grounding a USN ship usually gets the captain fired, almost 100% of the time, like within days of it happening. Considering bureaucratic delays usually involved in an institution like USN, disciplining captain for accidental damage of a ship is lightening fast.
HOWEVER, Admiral Nimitz did ground a destroyer when he was a young captain sometime in 1920s. He famously went to sleep quite peacefully on the deck of the ship that night because he knew there was nothing that could be done that night. Ultimately, he was not blamed for the grounding because the chart he was following was later proven to be faulty. He went on to lead the USN brilliantly in defeating the Japanese Navy in WW2. Considering how revered Admiral Nimitz is by USN, it's quite an exception.
"Because some dudes wanted to play golf before leaving Hawaii, they ended up being one of the causes of causing multi million dollar damage to a USN cruise"
That's such a bullshit thing to say. Obviously the civilian dudes didn't asked the ship crew to run aground. And even if they would have with those exact words, it's the job of the person in charge of the vessel to say no. Maybe one can say such things when there is an adverse power dynamics (such as it were with the Polish Air Force Tu-154 crash in 2010), but I don't think that would apply in this case.
Notice USS Port Royal wouldn't have grounded even with all those failures/issues of inexperienced crew, sleep deprived captain, misred equipment, if they they just didn't get close to the shore at evening time.
And the USS Port Royal got close to the shore ONLY because they wanted to let those contractors return to their hotel.
I think it's been done. The Fitzgerald's captain is facing (EDIT: the prospect of) court martial, and had to be airlifted off the ship with serious injuries.
Not sure that I'd agree, though it depends on the outcome of the investigation as to how much it'd apply to this situation. One line of thinking is that the US is now about to pay tens or hundreds of millions of dollars, not to mention the loss of life; that's a pretty expensive lesson that was learned, why would you throw away that valuable knowledge/experience?
Of course it remains to be seen if there was indeed dereliction/incompetence/any other more-malicious issues at play, which would be a different story -- not advocating one way or the other for this scenario. I however do think that, in the majority of organizations, firing people for honest mistakes tends to not be the best course of action.
Have this image of the Navy, strict protocols, attentive crew all responding quickly to the commands of a hardened captain. For this accident to happen, nothing close to that was going on. It's hard to comprehend just how many people had to drop the ball for this accident to happen.
An attentive crew responding quickly to orders is by no means incompatible to what happened here.
The question is, who was in a position to know about the incoming container ship? And who was in a position to act on that knowledge? The helmsman would have promptly responded to an order to take evasive action, I'm sure, and the engineering crew would have very quickly "answered the bell" -- but the order was never given.
In fact, the more there was a central point of focus within the watchteam (in this case the Officer of the Deck, not the Captain), the more likely it is for things like this to happen. If the OOD were to have been mistakenly convinced that the ACX Crystal was not a threat (whether due to its old course or because it was an overtaking vessel) then the OOD alone could have prevented a proper watchteam response. We've seen similar things with airline disasters, where a pilot loses their situational awareness and the crew failed to speak up in time.
Sadly, I could easily come up with plausible and workable theories of how a Navy watchteam would have this slip through the cracks. They involve complacency, or contact density, or misapplication of the navigation "Rules of the Road", or a mixture of all of those. The hardest part to explain is how a lookout didn't spot the Crystal in the lead-up to the collision -- dark or not, she would have been lit up to some extent by her nav lighting alone, and can't have been overtaking so fast that there wouldn't have been minutes of warning.
I guess there was a deal in place to reduce the look-out crew and radar-crew- which where allowed to nap.
This would reduce the multiple layers of security to a single point of failure- which then happened.
For those who are interested in a relatively accurate (but dramatized) portrayal of what the operations on a destroyer look like you should watch the first and second season of The Last Ship. It does fudge a few details and its obviously a TV show that can't show anything close to classified material but it does give you a general sense of what its like.
The Crystal Bay is 294m and crewed by 20–25. It isn't very maneuverable and was most likely on autopilot. Yeah, the Crystal was for all intents and purposes a drone ship and the crew was probably asleep.
The USS Fitzgerald is 153m and crewed by 200. It is extremely maneuverable and is professionally navigated. While its AIS transponder was most likely turned off for security reasons, it has radar and an AIS receiver and deck watches.
I get that part, but if the whole lot of them went to sleep, they must have been very confident that nothing happens or that they will be awoken before anything happens.
There are only one or two people awake on lookout in a civilian ship.
Air-conditioned bridge, slowly rolling waves, practically nothing to see or do... If they're on a typical watch pattern, 0230 means they got out of bed an hour ago and had just settled into the watch routine.
...for 3-12 minutes. Then BNWAS flashes the light, then it sounds an alarm, then it sounds a different alarm both on the bridge and in the captain's cabin, then it sounds an alarm that the entire crew can hear...
However, yes, even if BNWAS were operational the watch could doze off for long enough to miss the fact that they were being passed, but not by enough, by a smaller faster vessel with no lights and AIS turned off.
The easiest way to track vessels at sea is using the AIS system[1], which is based on VHF transponders that broadcast position, course, speed, and other useful information. This data is overlaid on the chart by navigation software which also provides configurable alarms. In fact, thanks to AIS you can even track ships online on sites like MarineTraffic[2].
AIS transponders are required on merchant ships above a certain length. Like with all technology, this can make people complacent and rely on it too much. As others mentioned, in this case the destroyer probably switched their AIS transponder off (but should still be able to receive).
Even then, the navy ship would still appear on the radar screen (not to mention its navigation lights), but radar is much more noisy since it can be affected by the environment (e.g. rainy squalls or waves).
Nope. Autopilot, like cruise control, relieves you of the burden of driving but not of the responsibility of watch [1]. The Crystal Bay also must have had radar and an alarm should have gone off. Yeah, the Crystal Bay was a drone ship but the Navy is the freaking Navy; moreover they were sailing dark in a crowded sea lane. It's impossible not to fault the crew starting with the captain.
ColRegs Rule 5
Every vessel shall at all times maintain a proper look-out by sight and hearing as well as by all available means appropriate in the prevailing circumstances and conditions so as to make a full appraisal of the situation and of the risk of collision.
This accident is actually fairly well understood and is a result of a particular aspect of naval and shipping culture that most of us never experience.
It is described in detail, with laugh-out-loud illustrations of boat paths, in an entire chapter devoted to these kind of accidents, in the excellent book _Normal Accidents_ by Charles Perrow.[1]
In short, very, very large ships are captained by egomaniacal captains who play chicken with one another - each of whom is certain that they have the right of way (due to ship size, military vs. civilian, etc.) These kind of accidents are nothing more than: "you get out of the way", "No, YOU, get out of the way". A very large cargo ship interacting with a US Navy destroyer is a perfect recipe for this kind of interaction.
Again, the illustrations of the boat paths in the book are literally laugh out loud funny as these huge boats slowly point directly at each other and crash or are on-course for near-misses and at the very last second veer into one another as they mistake who is turning where.
I highly, highly recommend this book - not just for the funny (but not that fascinating) chapter on the shipping lanes, but for the chapters on nuclear plant accidents and the challenger accident.
The captain of the US ship was asleep. The article suggests the cargo ship was completely computer-controlled at the time of the incident and for a little while afterwards, until the crew realized what happened. I don't think the two-egos-colliding story works here.
Additionally, the destroyer was hit around the middle of the right side while the cargo ship hit head-on.
The egos work also on lower levels, e.g. someone looking at a radar and assuming they have the right of passage b/c they are bigger vs. military. I can not image that no radar alarm was heard and reacted upon on both ships.
But it takes a lot of things to wrong for this to happen.
I don't think this was the result of egomania. The destroyer's captain was asleep --it's speculated everyone on the freighter was asleep.
The problem was complacency --that may be somewhat forgivable on the freighter's side, but not on the destroyer's side --it's a warship. It should always be on alert and at the ready. They should never be complacent. The captain allowed his sailors to become complacent.
I am hopeful the navy will not only examine this crew but will also examine the rest of the navy for laxness and failure to perform/dereliction of duties, etc.
How is there not some kind of "crew is asleep detecting autopilot" on these ships? If not on the military vessel, then on the cargo transport? Seems like it would be easy enough if they have radar. Just detect when you're on a collision course and adjust 3 degrees to port / starboard if there isn't human input?
That's a good question. I would think that modern ships should scream "collision course ahead" with light, sounds, etc. However, psychologically, maybe it's believed that this would make people even _more_ complacent? I don't know, just speculating. Never the less, the captain should never allow his sailors to think, "we're almost there, we're in friendly waters, we can relax". No, always on alert. It's unforgivable.
Of course the cargo ship has BNWAS installed. The captain might have turned it off? If the bridge personnel didn't know about the (no-AIS, running dark, much faster) navy boat, however, their wakefulness wouldn't have prevented the end result.
I don't think it has to be the actual captain of the ship who could not, of course, be at the helm 24 hours per day. If this was indeed a manifestation of what is described in the book I refer to it could be anyone in command of the ship who has been led to believe that their ships size and/or status gives them universal right of way.
Like everyone else said, captain was asleep. And the cargo ship hit the destroyer obliquely, from behind. The destroyer clearly had right of way and no one was playing chicken because everyone was asleep on the cargo ship.
> no one was playing chicken because everyone was asleep on the cargo ship.
Unless everyone on both ships was asleep, this statement doesn't have to be true. If anything, it seems eminently plausible the destroyer's crew was playing chicken with the freighter thinking they were playing against a human opponent, when in fact they were dealing with an automated system.
This is more realistic. But it begs the question, why didn't the destroyer out-run it? The freighter was approaching from it's back side, and should have given way, but once you realize it's not going to you have 100,000 horsepower to accelerate away with.
And standing orders were to wake the captain whenever another ship gets within a minimum distance. So the watch had to not only decide to play chicken with the freighter, but not wake the captain when it breached minimum safe distance.
"Unless everyone on both ships was asleep, this statement doesn't have to be true. If anything, it seems eminently plausible the destroyer's crew was playing chicken with the freighter thinking they were playing against a human opponent, when in fact they were dealing with an automated system."
Interesting you should mention this because, IIRC, the book has an aside about ships that (unknowingly) play chicken with lighthouses.
Everybody's responsible for avoiding an accident. Thinking in terms of right of way is exactly how this kind of accident happens. That's what makes you think you're right to proceed towards what looks like a certain collision.
(It's the same on the road; you have to take it as an article of faith that other people will respect your right of way, but at the same time, it makes sense not to go past junctions with waiting cars too quickly, just in case they don't see you. If you get hit, it might not be your fault, but that's little consolation if you're injured or killed.)
Well, the international regulations that govern maritime traffic do think in terms like right of way. They define for various scenarios of two ships approaching the other which one is the "stand on" vessel, that must maintain course and speed, and which one is the "give way" vessel, which must maneuver around the other.
Of course, the rules do also say that both ships are responsible for avoiding collisions with the other, and that both are free to maneuver to avoid collision if they are put in extremis by failure to follow the navigation rules.
Both captains were asleep. The question is why the American one was when, given it's a military vessel, they should be concerned about being approached by other ships. Aren't they held to a higher standard than some random container ship crew?
Captains have to sleep, which is why warships have a watch team who are awake. They failed here, which is unfathomable, but that is not, per se, the CO's fault. The CO has to sleep sometime -- I've stood many watches as the senior-most officer awake on a nuclear submarine while my CO slept, and I had to avoid collisions just as much as I would have had to do if my CO had been awake.
It is true that many COs will try to time their periods of sleep for when the operational tempo is lower as a risk management measure. I can only assume the Fitzgerald's CO had made his risk assessment and figured that it was better to sleep at night than to sleep during the day when the contact density would be even higher.
But for a Navy warship being "approached by other ships" is a fact of life for every hour of every day, you can't always plan your sleep schedule for when you're the only boat on the pond.
Yes, and every Captain out there puts the minimum "closest point of approach" for any contact that they're willing to accept into their standing orders, precisely so that they will be woken up by the watch team with enough time to make it to the bridge and assess the situation.
But this still relies on the watch team detecting the contact and realizing that it will pass near to the ship, and so again the ship's captain is back to relying on his or her watch team to at least get to that minimal level of competence -- because if they can't rely on the officers on watch to be able to wake them up properly, we're back to captains who can never sleep.
It sounds like an interesting book. Is there a more up to date canonical text on failures of complex systems?
The Wikipedia page includes "almost all serious nuclear accidents have occurred with what was at the time the most recent technology". Chernobyl was trying out a new cooling method, but it wasn't a very new reactor. Fukushima wasn't very new either. However, the book was written before both nuclear disasters.
I do think there is a social sciences aspect to the risk of nuclear energy. People who know enough to minimize risks are afraid enough of others' over-reactions to transparency about flaws and mistakes that there is an incentive to cover things up when they could be fixed. The other problem is that Chernobyl and Fukushima are talked about as the worst a nuclear disaster can be, when their impact was greatly contained by heroic efforts of individuals
Not the parent who mentioned _Normal Accidents_, but this is one of my hobby-horses, so I'm glad you asked!
Prof. Nancy Leveson of MIT has done extensive research over the last forty years on the safety engineering of complex systems, and her 2012 book _Engineering a Safer World_[0] (open-access PDF at that link) is partly a response to Perrow's _Normal Accidents_.
In particular, and unlike Perrow, she thinks we really can design complex sociotechnical systems which aren't disasters waiting to happen, and her book provides a theoretical framework for doing so, plus a fully-worked design analysis technique and accident investigation technique using that framework. These have been and are being used under real-world conditions in a variety of industries including transportation, the military, and even software.
There's a conference every year at MIT which I've been going to every year for about four years now where folks present. Every year there are more industries and disciplines represented, and there have also been presentations of analysis techniques applying the theoretical framework in practical contexts like cybersecurity and privacy.
You're also very right that there are huge social and human factors components to safety, and her book talks a lot about getting that right and analyzing what went wrong at various levels of abstraction, including cultural and regulatory. I highly recommend it, if you're interested in this subject.
Based on what I know of the US Navy, I'm certain that the accident investigators working on this collision are doing so informed by this methodology.
Somehow I don't think even the Captain Queeg would lay asleep in his bunk while playing 'chicken' with another large ship.
As the article points out, the one explanation that matches the tracks and observations is that everyone was asleep and the two ships were on autopilot.
And it is considered "impossible" by many that discipline on a Navy ship would have broken down so much that the entire watch conspired to ignore and not report their fellow sailors dereliction of duty.
As a result 'well understood' is not a good description of this event :-)
So it's a hugely clickbaity title. The article provides no new evidence or even offer any possible causes as to the accident, it just details what happened and wonders why.
We already have a good idea what happened.
1) It was a very dark night, the moon had not yet risen. Even a large cargo ship can be hard to see against a black horizon at night.
2) The destroyer was running without transponder on, as per typical Navy behavior so it couldn't be tracked by unfriendly powers. This meant the automated navigation system on the cargo ship didn't see it.
3) The cargo ship made a significant course change a few miles before the collision, which meant if the destroyer crew had plotted its course prior that change, they might have not realized they were suddenly on a collision course.
4) The cargo ship hit the destroyer from an acute angle, almost from behind. On that heading rules of navigation say it was on the cargo ship to give way to the destroyer, it's possible the crew thought they would.
5) if any of the cargo ship crew were on watch, they could not have seen the destroyer once they were close, given how tall their ships deck was.
6) The track after the collision indicates the cargo ship crew was asleep and the automated navigation system turned at the collision then resumed course, and it took 30 minutes for the crew to take control, figure out what had happened and returned.
7) Allegedly AEGIS radar systems are often tuned to filter out surface traffic so they can be more accurate at picking up air traffic.
8) The fact the captain was asleep in cabin is evidence no alarm was given nor threat detected before impact. Every time the captain goes to bed they leave standing orders to wake them if any ship gets within a certain safety margin.
What is still a mystery is why the crew didn't detect the cargo ship before the collision. Certainly a combination of radar and deck watch should have been able to see the ship. It should have been well lit enough to see once it was close. It should have been signaling it's location with a transponder. Even if the crew incorrectly thought it would pass a minimum safe distance the captain should have been woke and brought to the bridge.
>>4) The cargo ship hit the destroyer from an acute angle, almost from behind. On that heading rules of navigation say it was on the cargo ship to give way to the destroyer, it's possible the crew thought they would.
how do you come to this conclusion? The destroyer's damage is on the starboard side- it's not on or near the stern.
looks like the destroyer thought that the cargo ship is overtaking, and thus the cargo ship is responsible to yield and later - when it became dangerously close - to take evasive action. While the cargo ship thought that him/her being on the starboard side of the destroyer makes the destroyer responsible to yield/to take the action. Or may be the cargo ship was completely asleep. Anyway, the destroyer waiting for the cargo ship to take that action... that bow coming closer and closer to the destroyer's bridge ... deer in the headlights comes to mind. A junior officer mistake of insisting till the very end on your right of way against a 4x times heavier civilian bathtub of a ship.
It seems the warship was being overtaken here, so rule 13 applies in this case "an overtaking vessel must keep out of the way of the vessel being overtaken".
Since the navy boat is much faster, we can't assume it was being overtaken. How often do such boats bob along at 10 knots in shipping lanes? IMHO it seems more likely that the navy were overtaking, but unfortunately their course cut too close to the course of the cargo ship. Imagine a car in the leftmost lane racing to get ahead of slower traffic, then turning or merging right. Only, instead of doing so directly ahead of a cyclist as such cars often do, this one merged right in front of a 29,000-ton cargo ship.
A destroyer can certainly move fast, but they don't have to, and in fact it would generally be safer in a crowded area to try to maintain the same speed as the civilian traffic, on top of being fuel efficient.
I don't think "safety" was the primary objective on the navy boat, but even if it had been, an accident with this damage (starboard deckhouse of navy, port bow of cargo) when both boats are going similar speeds means that the navy boat was the "give-way" yet stupidly crossed the "stand-on". There are red and green nav lights installed on every motorized boat ever manufactured, to help idiots not make that mistake. The assumption that the speeds are the same makes this incident entirely the navy's fault.
Not necessarily true. A ship that is overtaking another ship, even if coming from the right side, is the "give-way" vessel until she is well clear of the ship she had been overtaking.
This is true even if the course the ship is using to overtake the other transiently puts them on the starboard side of the other ship.
To determine "crossing situation" vs "overtaking situation" you have to know the courses of both ships, and that's data we don't have yet as far as I've seen.
It was your hypothesis that they were going the same speed. I just pointed out the only possible implication of that. I think that the navy overtaking from port is more likely to have been the case. You seem to be suggesting now that the navy boat was going much slower than the cargo ship's maximum 10-12 knots, which is just goofy.
Besides we've already heard the testimony of the cargo ship's captain, which testimony is 100% corroborated by the AIS track. As the stand-on, they did everything they could to hail and signal to the navy boat. The navy boat kept coming across their bow. At the last moment, they turned hard to starboard. The navy will never release any of their own records, because such would only make the obvious, more obvious. It's not as though there haven't been other navy boats doing similar stupid shit over the last few years: Porter, Port Royal, etc., not even to mention the problems that their subs keep having. Ask a merchant mariner about navy safety. Their motto is, "if it's gray, stay away!"
In most of those cases they've been able to claim extenuating circumstances. That won't happen here: the cargo ship was operating exactly as one would expect, in a busy shipping land, in clear weather. The captain and OOD have the deaths of sailors on their hands. Those who imagine that they lead this navy, should ponder what sort of response would best prevent a repeat of this tragedy.
You're right, I have a hypothesis, nothing more or less.
Never mind that at least my hypothesis is consistent with the location of the damage on both ACX Crystal and Fitzgerald, since it's possible that the location of the damage was a result of last-minute maneuvering and not the course of each ship in the minutes before the crisis.
But let's look at your hypothesis:
> As the stand-on, they did everything they could to hail and signal to the navy boat. The navy boat kept coming across their bow. At the last moment, they turned hard to starboard.
You say this is "100% corroborated by AIS", but working to try to hail a Navy warship and then choosing to make a last-second attempt to turn is completely inconsistent with their AIS track that shows them not turning back to investigate until 30 minutes after the collision, and only once they spotted the ship they hit, did they turn back to resume course and report the collision to the Japanese Coast Guard.
If they were that busy trying to hail turn to avoid the Navy warship, then there was no need to double-back to figure out why their bow sustained damage, nor a reason to delay an hour to make the report to shore. Their AIS track is consistent with a cargo ship on auto-pilot (which to be fair, is the normal and expected mode of operation), with the computer auto-pilot encountering an unplanned navigational issue that it then piloted through to get back on track, with a human crew following up minutes after and then turning around to see just WTF had happened.
On top of that, the captain of the ACX Crystal claimed to use "flashing lights" to signal the Fitzgerald that she was standing into danger. But the COLREGS demand the use of an audible danger signal (5 or more short blasts), and no other signal. The captain also claimed to start a hard turn to starboard, but then says that it took 10 minutes to hit Fitzgerald from then.
But even for a container ship, if you were in the middle of a hard turn for 10 minutes you'd have turned all the way around. In fact if you look at the AIS track https://twitter.com/imgregorous/status/875905972924948480/ph... they make a sharper turn after the collision (presumably when the human crew turned off the auto-pilot and made a hard turn to quickly get back down the original track and figure out what happened).
Fact is, if the ACX Crystal crew were really so in tune with the traffic situation as to be able to watch the Fitzgerald make a turn, spend minutes on bridge-to-bridge VHF trying to warn the Fitzgerald, and spend "minutes turning" to avoid a ship traveling on a single course and speed, they're going to be found liable as well. If they really had so much awareness of the ongoing situation to watch it unfold in the way their captain is said to have relayed it to investigators, they also had enough time to unilaterally avoid the collision as they were required to do by COLREGS.
> The navy will never release any of their own records, because such would only make the obvious, more obvious. It's not as though there haven't been other navy boats doing similar stupid shit over the last few years: Porter, Port Royal, etc., not even to mention the problems that their subs keep having. Ask a merchant mariner about navy safety. Their motto is, "if it's gray, stay away!"
Definitely not an axe to grind here, is there? I guess it's a good thing we already know what happened then, I guess we can tell Japan to call off the investigations?
P.S. the Navy has released many of the records for "similar stupid shit" you mention (e.g. audio from the bridge of USS Porter in the most recent destroyer-involved collision), so I do not understand why you say the Navy would not here.
> In most of those cases they've been able to claim extenuating circ...
Perhaps I shouldn't have been so certain about that point.
But look at the angle the Crystal hit at. If the Crystal had hit at a 90 degree angle the destroyer would have been split in two. The damage shows it was clearly hit at less than a 45 degree angle. If the Crystal was approaching with a smaller angle (I think it's 25 degrees) it loses it's right of way. Essentially it's running down the destroyer.
Now we know the Crystal should have been cruising at a more leisurely pace than the destroyer, so in reality that might mean the Crystal was ahead of the destroyer before the collision. But I strongly suspect something happened where the Crystal was the faster vessel and overtook the Fitsgerald, the crash itself is evidence of it.
> 2) The destroyer was running without transponder on, as per typical Navy behavior so it couldn't be tracked by unfriendly powers. This meant the automated navigation system on the cargo ship didn't see it.
I hope this behavior is changed during peace to prevent unnecessary human fatalities.
Turning on transponders across the entire fleet is not a proportional response here. There are lots of military vessels out there and 99.99% of the time they don't crash into cargo ships. A lot of things went completely sideways for this to happen, and there are reasonable failsafes that can be enacted to prevent it in the future.
Although the AIS transponder was turned off - the vessel would still have appeared on radar. The cargo vessel's system wouldn't have had the same amount of information for anticollision calculations, but still would have been able to detect it as an obstacle and would have flagged it to anybody on the bridge who was paying attention.
Plenty of ships go to sea without transponders on, such as large fishing vessels that don't want to give away where the good spots are.
Large fishing vessels turn on their AIS when crossing shipping lanes. Also, the whole point of modern military naval architecture is to reduce radar cross-section. Sure these boats can be seen from up close, but we can't assume that decades-old cargo ships are outfitted with the latest radar tech either.
Even a consumer grade radar would be able to spot a 9,000 ton vessel from a thousand yards away no matter how "stealthy" it is, and they got closer than that.
Whereas, on the current footing, all it will take is a bunch of 12-knot cargo ships on autopilot to take them out. Either the captain and OOD should both be court-martialed out, or USA Navy needs to go back to the drawing board.
A single submarine loss (the Thresher accident) was enough to prompt the US Navy to institute the SUBSAFE program, possibly the single most successful safety program in history.
Considering the size of the US navy, around 430 ships, that's extremely unlikely. There's not enough submarines in the world to sneak attack them all at once.
No one would use submarines for the first strike in a major war unless it was to launch nuclear warheads from an unpredictable location. It would take months for them to get into position, even if the attacker had that many subs and it would reveal sub locations at the start of the conflict.
Hitting a ship with a missile is hard because they're usually out of land based radar range but if they all had their transponders on, it would be so much easier, especially since medium and long range missiles are getting increasingly more advanced. I wouldn't be surprised if state of the art recon satellites could pick up a cluster of transponders in a carrier battle group and use that data for targeting. If all an enemy did was take out our aircraft carriers, it would cripple our force projection outside of stationary military bases (which would all be hit in a first strike) and turn the navy into glorified mobile silos. The element of surprise is often the only deciding factor in skirmishes and battles between equal forces so it is paramount that giant, slow, and vulnerable fortresses don't broadcast their position.
Nukes don't work very well on ships much to the dismay of military planners.
There's a new generation of ballistic missile which might work, but they're extremely expensive, have limited range, and are rare. This idea of sinking the entire US fleet in one strike is utterly absurd.
> So it's a hugely clickbaity title. The article provides no new evidence or even offer any possible causes as to the accident, it just details what happened and wonders why.
That's how news works when something is still being investigated. And your comment proceeds to do the exact same thing: offer details about when happened, and then raise the same unanswered question that everyone has arrived at (including the article). So... what's your point, exactly?
The clickbaity part is the title indicates the article is going to provide an answer.
And I at least provided detailed possible issues that contributed to why the collision occurred, and why mysterious behavior occurred (the Crystal taking 30 minutes to turn around).
My point is that the NY Times should offer me a high paid job to make their articles better.
> The destroyer was running without transponder on, as per typical Navy behavior so it couldn't be tracked by unfriendly powers. This meant the automated navigation system on the cargo ship didn't see it.
It sounds like there may be some design flaws in their navigation protocols or automated systems.
When in busy international shipping lanes you are going to be spotted visually and on radar by other ships, including ones from countries that cooperate with those unfriendly powers you want to avoid. Your general position is going to get back to those unfriendly powers.
So is there any point in turning off your transponder when in such waters? It seems from my inexpert viewpoint that if you are running with your transponder off, you should also be keeping out of visual and radar range of other ships.
Sure, if the transponder is off so that the unfriendly power is only able to track via reports from friendly (to them) ships, there will be some delay in their tracking. The unfriendly power won't be able to launch a long range cruise missile attack on you based on such tracking, but that's only relevant when you are in an active conflict with someone who has that capability. The US currently is not. The unfriendly powers currently interested in US navy ship positions only need to know the general area of the ship to figure out the things they are trying to figure out.
The AIS transponders are now being read from satellites and there are multi-year histories of ships movements. It's understandable that a Navy wouldn't want this information available. What I don't understand is why they didn't detect the cargo ships transponder and sound an alarm.
> 1) It was a very dark night, the moon had not yet risen. Even a large cargo ship can be hard to see against a black horizon at night.
Negative, skipper. It's easier to spot a cargo ship in cloudless pitch dark. There are red, green and white nav lights. That's minimum and there's usually a Christmas tree of other lights. As a vessel greater than 50m, more lights. Constrained by draft, more lights. Cargo ships are not hard to spot, especially with radar and an AIS receiver.
Whether the Fitzgerald itself was visible at all is a good question. It had its AIS transponder off and Arleigh Burke-class destroyer has a semi-stealth design. But marine radar still should have spotted it.
Even if it was in the middle of a hurricane with twenty foot swells on a moonless night you'd still expect a destroyer to be able to spot 29,000 tons of metal coming at it.
I know how dark it can be outside of the city. I also don't carry around a radar system worth tens of millions of dollars that's operated by a crew that's supposed to be paying attention to where their $1.7B vessel is going.
Surely a ship of that sort has a pretty robust night vision system, if not multiple methods of detecting things including thermal or lidar.
The cargo ship made a significant course change a few miles before the collision, which meant if the destroyer crew had plotted its course prior that change, they might have not realized they were suddenly on a collision course.
This "change" would have been in keeping with the local TSS [0]. It's like when you're driving and there is a bend in the road, you steer so as to keep the car in its proper lane.
Click bait doesn't get much more severe than this. Someone should write that article and tell us the answer, when we know it - but this ain't that article. Grab a little integrity, NYTimes.
That's useful. Clearly the destroyer didn't detect the container ship before the collision; there was no collision alarm and no maneuvering to avoid the collision. But why? That hasn't been reported publicly yet. The U.S. Navy probably knows in general terms; the bridge and CIC crew members will have been interrogated by now.
No real point in speculating in advance of evidence.
"As someone with a few years experience at sea, I've never experienced this culture."
I encourage you to read the book - it makes the case that among very large ships there is indeed such a culture.
My personal experience as a sailor in the US Navy neither bolsters nor diminishes what is described in the book but I do believe that a culture of machismo and egomania can permeate the ranks.
Just because it was the officer in charge and not the actual (sleeping) captain might not make much difference.
I spent years at sea as an Officer of the Watch in the Royal Navy - during that time I spoke to quite a few other watchkeepers from around the world. The attitude I encountered was always one of terror of getting it wrong rather than machismo.
I can see how in ambiguous scenarios with large cargo vessels this could lead to a collision. In a relatively nimble destroyer I'm very confident this wasn't a simple case of "I'm the stand on vessel, I won't move."
146 comments
[ 3.1 ms ] story [ 297 ms ] threadThat said, this article really adds nothing to that which is already known about the incident, besides that someone tweeted "people weren't paying attention".
Very lucky there weren't more deaths.
That said, for the US this is a pretty serious embarrassment, because it screams "not combat ready".
This is the same thing as always, people are lulled into complacency and then accidents happen.
Edit: The captain is responsible for his crews performance and the navy way means he'll never be given command again, regardless of the results of the investigation.
But my response was to someone who apparently thought the captain was gunning the engine while playing chicken with the cargo vessel. For one specific point, the destroyer wasn't on collision course, the cargo vessel was.
If lookouts were not posted or the radar was not properly manned, the commander is ultimately responsible.
Heads will proverbially roll on this one, and his will absolutely be one of the first.
And we do know he will lose his command and never get another one, whether he's shown to have culpability for his crews failures or not. It's the navy way.
Odd, because when I go back to the OP, I don't see any indication that suggests he thinks the captain made an active decision to be reckless.
Maybe they edited after you replied?
Or is it the quotes around "accident" in this sentence that lead you to interpret their comments this way?
There is absolutely no reason a modern warship should have an "accident" like this.
If so, I read this differently. In my view, an accident is something that occurs due to unexpected changes in a situation. For example, if a car crash occurred because a deer jumped across the road and lead a vehicle to swerve, I'd use the word "accident".
But in cases of negligence, I find the word "accident" makes it far too easy to blame the situation rather than the people. In this case, in my view these two ships colliding was no "accident" in that there must have been clear, conscious human decisions that lead to the event, even if those decisions were simply "I'm going to have a nap now (and thus be negligent in my duties)".
Incidentally, I think we're far too willing to refer to car crashes as "accidents", and for this same reason.
Don't get me wrong, the CO is still "accountable" here, but while the buck may stop with him, there is going to be plenty of room for blame by the time this is all settled out. I don't care who the CO is, there's no excuse for a Navy watchteam on a modern destroyer to miss on oncoming container ship so badly.
And comparing a ship's captain to a babysitter? When a babysitter comes to my house I expect them to be attentive for 4 hours and then to go home. I don't expect captains to never sleep during long voyages. There is a whole process around allowing the captain to sleep, and waking them for any possible problem, which obviously didn't occur her.
But by all means some posters can compare his job to babysitting.
So yes, provision has to be made for a CO to sleep at some point. And the Navy accordingly has provisions for this -- provisions which failed here. The CO will be held accountable for maintaining -- or not -- an environment conducive to safe navigation, but in any event the blame will not attach solely to the CO here.
That's the proper analogy.
In USN, the captain is ultimately responsible for everything that happens to the ship and on the ship. I admire this trait in USN.
1) I really do feel bad for the poor captain as this is most likely not his fault directly but someone else's.
You will probably see one of these on the final report for placing the blame on the captain:
- Not keeping crew trained/motivated enough to be alert to avoid the accident.
- Not giving clear instruction to the crew to wake him up when another ship is near by.
- Having a hostile environment where the subordinates avoid waking him up or dealing with him directly.
- Not keeping equipment in proper order
I'm not saying any of it may be true or not, but usually that seems to be the common thread.
2) Check the case of 2009 USS Port Royal grounding. A USN missile cruiser grounded near Honolulu Airport. Airline Passengers coming in for landing at Honolulu Airport could see the grounded ship.
https://en.wikipedia.org/wiki/2009_USS_Port_Royal_grounding
The captain was relived of his duty within days. He didn't even get to stay long enough to command recovery effort of the ship.
The final report states following as the reasons of the grounding.
combination of a misread navigation system, a sleep-deprived commanding officer, broken equipment, and an inexperienced and dysfunctional bridge team.
The ship had just finished a major overhaul and they were busy testing the system, resulting in a captain with not enough sleep.
The direct reason for actual grounding?
Military contractors (engineer/technician) had flown into Hawaii to work on the ship during the overhaul and were on the ship. They were on the ship during the testing out in the sea which lasted longer than planned that day. And the contractors really really wanted to get back to their hotel that evening to play a round of golf the next day before returning to US mainland. To allow the military contractors to leave the ship on the small crafts that evening safely, someone ordered the USN cruiser move closer to the port. And it was during this movement, the cruiser grounded. This is not in the wiki but I remember reading about this in the local newspaper. Because some dudes wanted to play golf before leaving Hawaii, they ended up being one of the direct causes of causing multi million dollar damage to a USN cruiser.
3) Grounding a USN ship usually gets the captain fired, almost 100% of the time, like within days of it happening. Considering bureaucratic delays usually involved in an institution like USN, disciplining captain for accidental damage of a ship is lightening fast.
HOWEVER, Admiral Nimitz did ground a destroyer when he was a young captain sometime in 1920s. He famously went to sleep quite peacefully on the deck of the ship that night because he knew there was nothing that could be done that night. Ultimately, he was not blamed for the grounding because the chart he was following was later proven to be faulty. He went on to lead the USN brilliantly in defeating the Japanese Navy in WW2. Considering how revered Admiral Nimitz is by USN, it's quite an exception.
That's such a bullshit thing to say. Obviously the civilian dudes didn't asked the ship crew to run aground. And even if they would have with those exact words, it's the job of the person in charge of the vessel to say no. Maybe one can say such things when there is an adverse power dynamics (such as it were with the Polish Air Force Tu-154 crash in 2010), but I don't think that would apply in this case.
And the USS Port Royal got close to the shore ONLY because they wanted to let those contractors return to their hotel.
http://www.nbcnews.com/id/29065902/ns/us_news-military/t/nav...
The article doesn't say anything about contractors wanting to play golf but I"m telling you I read it in the local press.
And yes, the captain simply could've have said no and told the contractors to stuff it. But he didn't for whatever reason.
He will never command again.
Of course it remains to be seen if there was indeed dereliction/incompetence/any other more-malicious issues at play, which would be a different story -- not advocating one way or the other for this scenario. I however do think that, in the majority of organizations, firing people for honest mistakes tends to not be the best course of action.
If only they could experience bIj!
He will never command again is a given, regardless of the outcome of the investigation. The Navy way.
The question is, who was in a position to know about the incoming container ship? And who was in a position to act on that knowledge? The helmsman would have promptly responded to an order to take evasive action, I'm sure, and the engineering crew would have very quickly "answered the bell" -- but the order was never given.
In fact, the more there was a central point of focus within the watchteam (in this case the Officer of the Deck, not the Captain), the more likely it is for things like this to happen. If the OOD were to have been mistakenly convinced that the ACX Crystal was not a threat (whether due to its old course or because it was an overtaking vessel) then the OOD alone could have prevented a proper watchteam response. We've seen similar things with airline disasters, where a pilot loses their situational awareness and the crew failed to speak up in time.
Sadly, I could easily come up with plausible and workable theories of how a Navy watchteam would have this slip through the cracks. They involve complacency, or contact density, or misapplication of the navigation "Rules of the Road", or a mixture of all of those. The hardest part to explain is how a lookout didn't spot the Crystal in the lead-up to the collision -- dark or not, she would have been lit up to some extent by her nav lighting alone, and can't have been overtaking so fast that there wouldn't have been minutes of warning.
The USS Fitzgerald is 153m and crewed by 200. It is extremely maneuverable and is professionally navigated. While its AIS transponder was most likely turned off for security reasons, it has radar and an AIS receiver and deck watches.
I'm wondering why they would be that confident.
Air-conditioned bridge, slowly rolling waves, practically nothing to see or do... If they're on a typical watch pattern, 0230 means they got out of bed an hour ago and had just settled into the watch routine.
Pretty easy to drift off.
However, yes, even if BNWAS were operational the watch could doze off for long enough to miss the fact that they were being passed, but not by enough, by a smaller faster vessel with no lights and AIS turned off.
AIS transponders are required on merchant ships above a certain length. Like with all technology, this can make people complacent and rely on it too much. As others mentioned, in this case the destroyer probably switched their AIS transponder off (but should still be able to receive).
Even then, the navy ship would still appear on the radar screen (not to mention its navigation lights), but radar is much more noisy since it can be affected by the environment (e.g. rainy squalls or waves).
The above is based on my experience as a sailor.
[1] https://en.wikipedia.org/wiki/Automatic_identification_syste...
[2] https://www.marinetraffic.com/
ColRegs Rule 5
[1] https://www.navcen.uscg.gov/pdf/navrules/navrules.pdfYou aren't kidding...found this video of the same class of destroyer doing a 180 degree turn. Surprisingly nimble. https://www.youtube.com/watch?v=5Vih4tGmqjs
It is described in detail, with laugh-out-loud illustrations of boat paths, in an entire chapter devoted to these kind of accidents, in the excellent book _Normal Accidents_ by Charles Perrow.[1]
In short, very, very large ships are captained by egomaniacal captains who play chicken with one another - each of whom is certain that they have the right of way (due to ship size, military vs. civilian, etc.) These kind of accidents are nothing more than: "you get out of the way", "No, YOU, get out of the way". A very large cargo ship interacting with a US Navy destroyer is a perfect recipe for this kind of interaction.
Again, the illustrations of the boat paths in the book are literally laugh out loud funny as these huge boats slowly point directly at each other and crash or are on-course for near-misses and at the very last second veer into one another as they mistake who is turning where.
I highly, highly recommend this book - not just for the funny (but not that fascinating) chapter on the shipping lanes, but for the chapters on nuclear plant accidents and the challenger accident.
[1] https://en.wikipedia.org/wiki/Normal_Accidents
Additionally, the destroyer was hit around the middle of the right side while the cargo ship hit head-on.
But it takes a lot of things to wrong for this to happen.
The problem was complacency --that may be somewhat forgivable on the freighter's side, but not on the destroyer's side --it's a warship. It should always be on alert and at the ready. They should never be complacent. The captain allowed his sailors to become complacent.
I am hopeful the navy will not only examine this crew but will also examine the rest of the navy for laxness and failure to perform/dereliction of duties, etc.
I don't think it has to be the actual captain of the ship who could not, of course, be at the helm 24 hours per day. If this was indeed a manifestation of what is described in the book I refer to it could be anyone in command of the ship who has been led to believe that their ships size and/or status gives them universal right of way.
Unless everyone on both ships was asleep, this statement doesn't have to be true. If anything, it seems eminently plausible the destroyer's crew was playing chicken with the freighter thinking they were playing against a human opponent, when in fact they were dealing with an automated system.
And standing orders were to wake the captain whenever another ship gets within a minimum distance. So the watch had to not only decide to play chicken with the freighter, but not wake the captain when it breached minimum safe distance.
Interesting you should mention this because, IIRC, the book has an aside about ships that (unknowingly) play chicken with lighthouses.
https://en.wikipedia.org/wiki/Lighthouse_and_naval_vessel_ur...
(It's the same on the road; you have to take it as an article of faith that other people will respect your right of way, but at the same time, it makes sense not to go past junctions with waiting cars too quickly, just in case they don't see you. If you get hit, it might not be your fault, but that's little consolation if you're injured or killed.)
Of course, the rules do also say that both ships are responsible for avoiding collisions with the other, and that both are free to maneuver to avoid collision if they are put in extremis by failure to follow the navigation rules.
It is true that many COs will try to time their periods of sleep for when the operational tempo is lower as a risk management measure. I can only assume the Fitzgerald's CO had made his risk assessment and figured that it was better to sleep at night than to sleep during the day when the contact density would be even higher.
But for a Navy warship being "approached by other ships" is a fact of life for every hour of every day, you can't always plan your sleep schedule for when you're the only boat on the pond.
But this still relies on the watch team detecting the contact and realizing that it will pass near to the ship, and so again the ship's captain is back to relying on his or her watch team to at least get to that minimal level of competence -- because if they can't rely on the officers on watch to be able to wake them up properly, we're back to captains who can never sleep.
That's pure speculation on your part, not an actual account or understanding of what caused this particular accident.
The Wikipedia page includes "almost all serious nuclear accidents have occurred with what was at the time the most recent technology". Chernobyl was trying out a new cooling method, but it wasn't a very new reactor. Fukushima wasn't very new either. However, the book was written before both nuclear disasters.
I do think there is a social sciences aspect to the risk of nuclear energy. People who know enough to minimize risks are afraid enough of others' over-reactions to transparency about flaws and mistakes that there is an incentive to cover things up when they could be fixed. The other problem is that Chernobyl and Fukushima are talked about as the worst a nuclear disaster can be, when their impact was greatly contained by heroic efforts of individuals
Prof. Nancy Leveson of MIT has done extensive research over the last forty years on the safety engineering of complex systems, and her 2012 book _Engineering a Safer World_[0] (open-access PDF at that link) is partly a response to Perrow's _Normal Accidents_.
[0]: https://mitpress.mit.edu/books/engineering-safer-world
In particular, and unlike Perrow, she thinks we really can design complex sociotechnical systems which aren't disasters waiting to happen, and her book provides a theoretical framework for doing so, plus a fully-worked design analysis technique and accident investigation technique using that framework. These have been and are being used under real-world conditions in a variety of industries including transportation, the military, and even software.
There's a conference every year at MIT which I've been going to every year for about four years now where folks present. Every year there are more industries and disciplines represented, and there have also been presentations of analysis techniques applying the theoretical framework in practical contexts like cybersecurity and privacy.
You're also very right that there are huge social and human factors components to safety, and her book talks a lot about getting that right and analyzing what went wrong at various levels of abstraction, including cultural and regulatory. I highly recommend it, if you're interested in this subject.
Based on what I know of the US Navy, I'm certain that the accident investigators working on this collision are doing so informed by this methodology.
As the article points out, the one explanation that matches the tracks and observations is that everyone was asleep and the two ships were on autopilot.
And it is considered "impossible" by many that discipline on a Navy ship would have broken down so much that the entire watch conspired to ignore and not report their fellow sailors dereliction of duty.
As a result 'well understood' is not a good description of this event :-)
We already have a good idea what happened.
1) It was a very dark night, the moon had not yet risen. Even a large cargo ship can be hard to see against a black horizon at night.
2) The destroyer was running without transponder on, as per typical Navy behavior so it couldn't be tracked by unfriendly powers. This meant the automated navigation system on the cargo ship didn't see it.
3) The cargo ship made a significant course change a few miles before the collision, which meant if the destroyer crew had plotted its course prior that change, they might have not realized they were suddenly on a collision course.
4) The cargo ship hit the destroyer from an acute angle, almost from behind. On that heading rules of navigation say it was on the cargo ship to give way to the destroyer, it's possible the crew thought they would.
5) if any of the cargo ship crew were on watch, they could not have seen the destroyer once they were close, given how tall their ships deck was.
6) The track after the collision indicates the cargo ship crew was asleep and the automated navigation system turned at the collision then resumed course, and it took 30 minutes for the crew to take control, figure out what had happened and returned.
7) Allegedly AEGIS radar systems are often tuned to filter out surface traffic so they can be more accurate at picking up air traffic.
8) The fact the captain was asleep in cabin is evidence no alarm was given nor threat detected before impact. Every time the captain goes to bed they leave standing orders to wake them if any ship gets within a certain safety margin.
What is still a mystery is why the crew didn't detect the cargo ship before the collision. Certainly a combination of radar and deck watch should have been able to see the ship. It should have been well lit enough to see once it was close. It should have been signaling it's location with a transponder. Even if the crew incorrectly thought it would pass a minimum safe distance the captain should have been woke and brought to the bridge.
They have separate surface search / navigation radars installed. According to Wikipedia, this particular ship had two.
how do you come to this conclusion? The destroyer's damage is on the starboard side- it's not on or near the stern.
https://static01.nyt.com/newsgraphics/2017/06/18/fitzgerald-...
The right of way rule is: "A boat approaching from your starboard (right) side has right of way."
Damage on the right side of the destroyer means the cargo ship had right of way. It suggests the crew manning the ~$1bn destroyer failed to yield.
|
| \",
Very often. Shipping lanes are often the most direct route from A to B and 10kts is about the most economical speed for many different ships.
This is true even if the course the ship is using to overtake the other transiently puts them on the starboard side of the other ship.
To determine "crossing situation" vs "overtaking situation" you have to know the courses of both ships, and that's data we don't have yet as far as I've seen.
Besides we've already heard the testimony of the cargo ship's captain, which testimony is 100% corroborated by the AIS track. As the stand-on, they did everything they could to hail and signal to the navy boat. The navy boat kept coming across their bow. At the last moment, they turned hard to starboard. The navy will never release any of their own records, because such would only make the obvious, more obvious. It's not as though there haven't been other navy boats doing similar stupid shit over the last few years: Porter, Port Royal, etc., not even to mention the problems that their subs keep having. Ask a merchant mariner about navy safety. Their motto is, "if it's gray, stay away!"
In most of those cases they've been able to claim extenuating circumstances. That won't happen here: the cargo ship was operating exactly as one would expect, in a busy shipping land, in clear weather. The captain and OOD have the deaths of sailors on their hands. Those who imagine that they lead this navy, should ponder what sort of response would best prevent a repeat of this tragedy.
Never mind that at least my hypothesis is consistent with the location of the damage on both ACX Crystal and Fitzgerald, since it's possible that the location of the damage was a result of last-minute maneuvering and not the course of each ship in the minutes before the crisis.
But let's look at your hypothesis:
> As the stand-on, they did everything they could to hail and signal to the navy boat. The navy boat kept coming across their bow. At the last moment, they turned hard to starboard.
You say this is "100% corroborated by AIS", but working to try to hail a Navy warship and then choosing to make a last-second attempt to turn is completely inconsistent with their AIS track that shows them not turning back to investigate until 30 minutes after the collision, and only once they spotted the ship they hit, did they turn back to resume course and report the collision to the Japanese Coast Guard.
If they were that busy trying to hail turn to avoid the Navy warship, then there was no need to double-back to figure out why their bow sustained damage, nor a reason to delay an hour to make the report to shore. Their AIS track is consistent with a cargo ship on auto-pilot (which to be fair, is the normal and expected mode of operation), with the computer auto-pilot encountering an unplanned navigational issue that it then piloted through to get back on track, with a human crew following up minutes after and then turning around to see just WTF had happened.
On top of that, the captain of the ACX Crystal claimed to use "flashing lights" to signal the Fitzgerald that she was standing into danger. But the COLREGS demand the use of an audible danger signal (5 or more short blasts), and no other signal. The captain also claimed to start a hard turn to starboard, but then says that it took 10 minutes to hit Fitzgerald from then.
But even for a container ship, if you were in the middle of a hard turn for 10 minutes you'd have turned all the way around. In fact if you look at the AIS track https://twitter.com/imgregorous/status/875905972924948480/ph... they make a sharper turn after the collision (presumably when the human crew turned off the auto-pilot and made a hard turn to quickly get back down the original track and figure out what happened).
Fact is, if the ACX Crystal crew were really so in tune with the traffic situation as to be able to watch the Fitzgerald make a turn, spend minutes on bridge-to-bridge VHF trying to warn the Fitzgerald, and spend "minutes turning" to avoid a ship traveling on a single course and speed, they're going to be found liable as well. If they really had so much awareness of the ongoing situation to watch it unfold in the way their captain is said to have relayed it to investigators, they also had enough time to unilaterally avoid the collision as they were required to do by COLREGS.
> The navy will never release any of their own records, because such would only make the obvious, more obvious. It's not as though there haven't been other navy boats doing similar stupid shit over the last few years: Porter, Port Royal, etc., not even to mention the problems that their subs keep having. Ask a merchant mariner about navy safety. Their motto is, "if it's gray, stay away!"
Definitely not an axe to grind here, is there? I guess it's a good thing we already know what happened then, I guess we can tell Japan to call off the investigations?
P.S. the Navy has released many of the records for "similar stupid shit" you mention (e.g. audio from the bridge of USS Porter in the most recent destroyer-involved collision), so I do not understand why you say the Navy would not here.
> In most of those cases they've been able to claim extenuating circ...
But look at the angle the Crystal hit at. If the Crystal had hit at a 90 degree angle the destroyer would have been split in two. The damage shows it was clearly hit at less than a 45 degree angle. If the Crystal was approaching with a smaller angle (I think it's 25 degrees) it loses it's right of way. Essentially it's running down the destroyer.
Now we know the Crystal should have been cruising at a more leisurely pace than the destroyer, so in reality that might mean the Crystal was ahead of the destroyer before the collision. But I strongly suspect something happened where the Crystal was the faster vessel and overtook the Fitsgerald, the crash itself is evidence of it.
I hope this behavior is changed during peace to prevent unnecessary human fatalities.
Plenty of ships go to sea without transponders on, such as large fishing vessels that don't want to give away where the good spots are.
The Royal Navy advertise their 8500ton Destroyer as having the radar cross section of a small fishing vessel: https://navynews.co.uk/archive/news/item/3951
I'm not sure what the weather conditions were like, but even in marginal conditions, they'll have been picked up on radar.
But even then they certainly don't want to make it simple to trace their complete deployments.
https://blog.usni.org/posts/2017/06/21/three-changes-the-nav...
Also, there are 274 ships in USN.
Also, this cargo ship did not intend to collide with the navy ship. In wartime, we can't make that assumption.
Hitting a ship with a missile is hard because they're usually out of land based radar range but if they all had their transponders on, it would be so much easier, especially since medium and long range missiles are getting increasingly more advanced. I wouldn't be surprised if state of the art recon satellites could pick up a cluster of transponders in a carrier battle group and use that data for targeting. If all an enemy did was take out our aircraft carriers, it would cripple our force projection outside of stationary military bases (which would all be hit in a first strike) and turn the navy into glorified mobile silos. The element of surprise is often the only deciding factor in skirmishes and battles between equal forces so it is paramount that giant, slow, and vulnerable fortresses don't broadcast their position.
There's a new generation of ballistic missile which might work, but they're extremely expensive, have limited range, and are rare. This idea of sinking the entire US fleet in one strike is utterly absurd.
That's how news works when something is still being investigated. And your comment proceeds to do the exact same thing: offer details about when happened, and then raise the same unanswered question that everyone has arrived at (including the article). So... what's your point, exactly?
And I at least provided detailed possible issues that contributed to why the collision occurred, and why mysterious behavior occurred (the Crystal taking 30 minutes to turn around).
My point is that the NY Times should offer me a high paid job to make their articles better.
It sounds like there may be some design flaws in their navigation protocols or automated systems.
When in busy international shipping lanes you are going to be spotted visually and on radar by other ships, including ones from countries that cooperate with those unfriendly powers you want to avoid. Your general position is going to get back to those unfriendly powers.
So is there any point in turning off your transponder when in such waters? It seems from my inexpert viewpoint that if you are running with your transponder off, you should also be keeping out of visual and radar range of other ships.
Sure, if the transponder is off so that the unfriendly power is only able to track via reports from friendly (to them) ships, there will be some delay in their tracking. The unfriendly power won't be able to launch a long range cruise missile attack on you based on such tracking, but that's only relevant when you are in an active conflict with someone who has that capability. The US currently is not. The unfriendly powers currently interested in US navy ship positions only need to know the general area of the ship to figure out the things they are trying to figure out.
Negative, skipper. It's easier to spot a cargo ship in cloudless pitch dark. There are red, green and white nav lights. That's minimum and there's usually a Christmas tree of other lights. As a vessel greater than 50m, more lights. Constrained by draft, more lights. Cargo ships are not hard to spot, especially with radar and an AIS receiver.
Whether the Fitzgerald itself was visible at all is a good question. It had its AIS transponder off and Arleigh Burke-class destroyer has a semi-stealth design. But marine radar still should have spotted it.
Get out of the city to realize just how dark it can be.
Surely a ship of that sort has a pretty robust night vision system, if not multiple methods of detecting things including thermal or lidar.
This "change" would have been in keeping with the local TSS [0]. It's like when you're driving and there is a bend in the road, you steer so as to keep the car in its proper lane.
[0] https://cdn-business.discourse.org/uploads/gcaptain_maritime...
No real point in speculating in advance of evidence.
I encourage you to read the book - it makes the case that among very large ships there is indeed such a culture.
My personal experience as a sailor in the US Navy neither bolsters nor diminishes what is described in the book but I do believe that a culture of machismo and egomania can permeate the ranks.
Just because it was the officer in charge and not the actual (sleeping) captain might not make much difference.
I spent years at sea as an Officer of the Watch in the Royal Navy - during that time I spoke to quite a few other watchkeepers from around the world. The attitude I encountered was always one of terror of getting it wrong rather than machismo.
I can see how in ambiguous scenarios with large cargo vessels this could lead to a collision. In a relatively nimble destroyer I'm very confident this wasn't a simple case of "I'm the stand on vessel, I won't move."
http://gcaptain.com/
http://gcaptain.com/uss-fitzgerald-fault/
http://gcaptain.com/uss-fitzgerald-fault-part-2-questions-an...