James Cameron: The Lessons of Titanic and other Reflections

Excerpt from:

“Risk and Exploration: Earth, Sea and Sky”

NASA Administrator’s Symposium

September 26-29

Naval Postgraduate School

Monterey, California

I am also honored to be part of this august panel, which includes two of my heroes from the undersea world, and some of the people I’m just meeting today. We live in an age when the land area of our planet has been explored, mapped, imaged, settled and exploited for whatever it has to offer. It’s definitional that what remains to be explored are the most remote, inaccessible and inhospitable parts of our world, or places that are not a part of our world at all. This basically means that the easy stuff has been done, if you want to consider polar exploration and all the great pioneering work in the ocean the easy stuff. The hard stuff is in front of us, and it means we are now confronting even more hostile and extreme conditions and requiring more sophisticated technology and support systems in order to do our exploration. Correspondingly, we are facing more complex and subtle forms of risk than ever before.

I have lived with risk for my entire professional career as an action film director. I regularly asked people with a completely straight face to set themselves on fire, to flip their car over, to leap out of an exploding building, to ride on top of a tractor-trailer truck that’s on its side skidding, to fly a helicopter underneath an overpass with two feet of clearance on either side of the rotor tip, and even to ride a sinking ship down underwater.

In twenty years of directing stunts and action and pyrotechnic effects, I’ve never actually had a serious injury on the set. That is because of application of a fairly rigorous and disciplined process. It is not as institutionalized as it is with NASA, but it has its own special rigor. Before any major stunt, or gag as we call them, I would walk the set myself, looking at every piece of rigging and turning over every possibility in my head. At that moment on a shoot when all the lights and the cameras are set up, it is the culmination of months of planning, engineering, testing and rigging. It has all been done by the industry’s leading experts up to that point. (I would just like to point out that the failure of the Genesis spacecraft was not due to the Hollywood stunt pilot. Of course, now we know that the science has been recovered, so it’s all good.)

These experts have decades and decades of experience doing stunts, explosions, car gags, fire, and whatever it is that we might be doing. But still, even after every single one of these people has signed off, I walk the set. I just call a complete hold. I walk the set. I look at the rigging. I ask questions. I think about it: what if this happens? What if that happens? Even though we have been over it and over it, I call that last minute hold, and I walk the set. I’m looking for something which is something that I’ve over the years come to call the x-factor, some previously unseen detail or some exotic combination of variables which could cause the stunt to go horribly wrong.

I guess my point here is that the personal touch is critical, and taking individual responsibility is critical for everybody in the chain. Systems protocols and institutional checks and balances are important, and they add great robustness to risky operations. However, those very checks and balances can often inhibit individuals from speaking up or taking action because they make the assumption that someone else has approved it. Someone else is going to catch it. Someone else has responsibility, and they don’t catch it before it’s too late.

We made the movie, “Titanic”. We began that production in a very unusual way. We actually dove to the wreck site of Titanic twelve times. It’s in 12,500 feet of water in the North Atlantic. We set ourselves some pretty ambitious goals. We were going to build a new camera system so that we could operate a 35mm movie camera outside the submersible, seeing ambient pressure at 5,500 psi. We were going to build new lighting equipment. We were even going to build our own remotely-operated vehicles so we could explore the Titanic wreck internally. I had some experience as a project manager developing new technology for underwater filming on the movie “The Abyss,” and that prepared me, to a certain extent, for the difficulty of engineering this new equipment. Nothing prepared me for the chaos introduced when we took that whole circus to sea on a research ship. We weathered three hurricanes and multiple equipment failures, but we managed to prevail and get the images of the wreck. In that process I got bitten by the deep ocean exploration bug.

After the success of “Titanic”, the movie, I found myself less interested in Hollywood filmmaking and more interested in the challenges of ocean photography and exploration. So over the next few years we developed new images and robotic exploration technology. Then I had to go out and raise the money by making films in order to pay for it. So we wound up returning to the Titanic wreck site in 2001 because I figured if I couldn’t raise money to go to the Titanic wreck I couldn’t raise money to do anything. We took our spanking new 3-D digital imaging system to capture the coolest stereo images of the wreck that we could before it disintegrates, and we made a film called “Ghosts of the Abyss” which was for the IMAX 3-D theaters.

We also created two very tiny and advanced ROVs [Remotely Operated Vehicles] which could fly untethered inside the wreck. They were untethered in terms of a power umbilical but they had a data tether which was a spool of fiber optic, kind of like a wire-guided torpedo. We were able to explore the wreck, room-by-room and deck-by-deck. These were launched from the submersibles after we landed on the Titanic wreck and they were flown inside the wreck by myself and two other pilots. With these bots we were able to capture some pretty amazing images inside the wreck in spaces which could never have been seen by human eyes and which probably will never be seen directly by human eyes. We were able to reveal in the lights and video cameras of these tiny robots a kind of lost grandeur of Titanic, which still exists deep inside that wreck.

For me that was the greatest adventure imaginable. If I wasn’t hooked before, I was certainly hooked then. Of course all the time I was very cognizant of the risks and as the person heading the team, the expedition leader so to speak, it was my responsibility. The buck stopped with me, so I was continuing to apply my lessons learned from my underwater motion picture film making experience, to this new realm. Of course we had a lot of problems and we had equipment failures and we got hit by another three hurricanes. Then the September 11th attacks cut short our expedition. It was certainly a bizarre and ironic experience to be literally down at the bottom of the ocean at the site of the defining disaster of the first part of the 20th century, while probably the defining disaster of the first part of the 21st century was taking place over our heads without our knowledge.

Having made 24 dives at this point by the end of the second expedition to explore the Titanic, I am now pretty continuously mindful of the lessons of Titanic as I continue with other exploration projects and any projects involving risk of any kind. The lessons learned from the sinking of Titanic caused sweeping reform of the maritime safety code in its time. But in the abstract, Titanic has a value as a kind of parable. The lessons learned are still valuable for us in our continuing exploration of the sea and of space.

Titanic was sunk primarily by institutional momentum. Just as the inertia of the ship was too great for the crew to be able to turn it in time to avoid hitting the iceberg, the inertia of their methodology was at least equally responsible for the collision. It was the policy of sea captains at that time to maintain full speed until they’d spotted the ice and then slow down only when it became absolutely necessary. This was for economic reasons, reasons of straight commerce. This was simply how it was done.

The Titanic’s captain was due to retire after this one last prestigious voyage after a long and unblemished career. He was captaining on the maiden voyage of the largest vessel ever created. His lifetime of experience taught him that on a crystal clear night, in a flat calm ocean, he was safe maintaining full speed despite the Marconi-gram sitting in his pocket warning of a huge ice field ahead. With a warning to the officer of the watch to be extra vigilant, he went to sleep as the ship barreled on toward its fate. Now was this arrogance or hubris as many have said? I don’t think so, not really. It was simply business as usual. These new ships didn’t handle like the previous ones. They took longer to stop or to turn. So everything he knew was actually wrong in that exact circumstance. The old operating methods didn’t really apply. The conditions had changed but the methods hadn’t kept up. It also required an unlikely combination of elements to create the disaster. It was a typical cascade failure where you had a number of things in series, all of which had to happen in that unique combination. The flat calm of the ocean meant that no swells were breaking against the icebergs, which reduced the ability of the lookouts to see the icebergs in the dark. The general mistake made by the crew was to underestimate the perversity of the ocean, even when it seemed at its most benign.

There are a few interesting parallels between the sinking of the Titanic and the loss of the Columbia Space Shuttle and her crew. In both cases, there were unheeded warnings. In both cases the warnings were dismissed, not out of negligence, but for reasons that made sense based on the experience and institutional memory at that moment. In the case of Titanic, the crew was well aware, because of wireless messages, that the ice lay ahead, but it was because it was the way it was always done that they proceeded at full speed toward the ice field.

With Columbia it was known from many past missions that the foam could separate from the external tank and possibly strike the orbiter, but that problem had been analyzed twenty years earlier and dismissed as a serious threat to mission safety. When foam was observed possibly striking Columbia during the launch, some engineers were concerned. But because this was the way we’ve always done it, the warnings didn’t propagate up the chain of command with enough force to change the outcome. So cultural momentum and institutional memory had worked against Columbia just as they had worked against Titanic.

Another parallel is that in both accidents an unlikely series of events were required to cause catastrophe. With Titanic it was the unlikely event of the very first iceberg that they spotted, the very first one out of a huge field of ice, happening to be exactly in the track of the ship. This was occurring on a night without the slightest swell activity to assist in spotting the berg in time. And all of this was happening to a new large class of ship whose crew was inexperienced in managing it in fast turns and sudden stops. With Columbia, it took the foam strike incident but then compounded it by the fact that this was one of the very few missions in recent years that did not go to the ISS. Had it been a mission to the Space Station it is likely that the Station crew would have seen the large hole in the leading edge of the wing during the operations. Then the station could have provided safe haven for the Columbia crew while everybody scrambled to launch a second orbiter to bring them all home safely.

So the vanishingly small possibility of a foam strike event actually damaging a flight-critical component was coupled with the statistically low probability of a non-ISS mission to create a disastrous outcome. These low-probability, high consequence events are the hardest to plan for and prevent, especially when it requires a number of low-probability events in combination in order to create a threatening scenario. Titanic teaches us to be constantly vigilant, to assume nothing about our methodology, to constantly ask the question “what are we doing wrong right now?”

I’ve lived with the lessons of Titanic and they’ve informed my judgment on subsequent expedition projects. After our second expedition to Titanic, we looked for other projects with more and greater challenges, of course. The following spring we imaged the wreck of the Bismarck, which is 16,000 feet down in the North Atlantic. Then we followed that up with stereo imaging at five hydrothermal vent sites along the mid-Atlantic ridge. We were pretty excited by the imaging results from that, and I decided to make a second IMAX 3-D film about the life surrounding that hydrothermal vents. It was my intention with this film to draw a kind of sea/space connection, on the basis of a kind of ocean analog, where we would bring NASA experts in analog missions and let them draw the parallels between undersea operations with multiple vehicles, deploying robotics. The submersibles would be like a Mars surface rover on a traverse being deployed from a habitat or a base camp for which the support ships or the surface ships were the analogs for that. But we were also drawing a connection between the types of life that existed in these chemosynthesis-based environments down at these hydrothermal vents, with the kind of life that we might encounter in extraterrestrial hydrospheres; if we were to find sub-surface water on Mars, if we were to find evidence of ancient life on Mars it might have originated in hydrothermal communities. If Mars was once covered with water or had a lot more water, that water may very well have been under ice, it may have been denied the ability to photosynthesize, it may have had to live within a chemosynthesis-based environment.

When we go to Europa, we may find evidence of life there, again, probably subsisting on a non-photosynthesis basis. So we draw the sea/space connection in that film. I just thought I’d point that out since it’s a room full of space buffs and we’re an ocean panel — that there’s a message there.

In the process of making this film I’ve formed a partnership to buy and operate two deep rover submersibles, which are actually co-designed by Sylvia Earle. They have a depth rating of 1000 meters. They are wonderful subs. You sit inside an acrylic sphere and you feel like you’re in a kind of vacuole within the ocean where you have unlimited visibility. You feel much better contact with the environment than you do looking through the small port windows of a typical deep submersible.

Now, previously we’d been working with the Russian Academy of Sciences and with their Mir submersible operation, which is a two-sub operation. I had a pretty good understanding of the working systems of those subs and of how the submersibles were operated and how two subs are operated in tandem with each other in diving ops. But I was certainly in for a very rude awakening by just how difficult it is to operate a manned submersible system when you’re starting from scratch and when you’re the one in front of whom the buck stops.

Now we began by assembling a new team to operate and maintain the Rovers and these were gathered from established submersible operations around the U.S. and Canada. The first task was to tear the subs down to their frames for ABS certification. Then we had to make all the modifications to adapt our 3-D digital technology with the [pan and fill] systems and our special lighting and all of that to the submersibles. That was certainly a daunting task. It took about six months and we were barely re-certified in time for our filming operations.

So we took our beautiful new subs to sea and met with the Russians out in the middle of the ocean, literally rendezvoused two ships in the middle of the ocean at the mid-Atlantic ridge. We dove them together in a joint diving operation with the Mir submersibles. This culminated with one dive where we actually had four subs rendezvous at the bottom in 870 meters of water at a site called “Lost City” which is a low temperature hydrothermal vent structure, a very interesting place.

This operation was very complex logistically because it involved the creation of new operational protocols for the launch and recovery of four subs in the same theater of operations at the same time. Tracking, communications, surface ship operations, the number of submersibles in the water made all these significantly more complicated. This had ramifications through every kind of contingency you could imagine. It wasn’t just twice as complicated as operating two submersibles — it was some multiple. There was some square law at work there. Also, we were dealing with underwater communications, which if any of you have done this you know that UPC communications can be spotty at best, and we were dealing with them in Russian and English between four vehicles at the same time on the same frequency.

So we had to go through a pretty rigorous process of defining our comms protocols before the fact. It was only because we had a good, long, healthy working relationship with the Russians that made that possible. I found that the principles of risk management and safety assurance that I learned as a film director were actually transferable to these new situations, at least at an abstract level, and certainly at a motivational level for myself in terms of applying the same kind of energy and passion to the safety of the operation as to the aesthetic results of the film making. Now, obviously there’s a very extensive body of established procedure for submersible operations, and we studied that pretty rigorously and we selected our team members accordingly on the basis of their experience with manned submersible ops. But it seemed like almost everything that we were doing was unprecedented, and it was often difficult to find any kind of existing guidelines in the literature. Often we were making up our own protocols in terms of what the safe procedures were for the launch of multiple subs or the manner in which we could descend them together for imaging purposes — sometimes only a couple of meters apart, how we could operate them on the bottom (proximity operations), how our acoustic comms would work during the dive, how we would work on the bottom with four subs together and a deployed ROV in the same area — a tethered vehicle.

We were able to pull on our experience from past dives, and we were able to anticipate and talk through in advance most of the contingencies that might arise on the dive. Because of the complexity of our dive ops, we always preceded each dive with a joint dive ops meeting between the Russian group and the American group. I call it the American group, but it was really a mixed group of people from Australia, Canada and everything else. The Russians called us the “American group”.

We used models of the submersibles to talk through the maneuvers. The process there was very simple. Here is a model of your sub. You are the pilot of the sub. You move it. I will give you a voice command. You move that model the way you think what I am telling you to do should result in action. This worked very well. So, literally it would be, Hands off. Okay, I want you to do this. I want you to do that. If they couldn’t visualize it on dry land where we could hear each other perfectly, then it certainly wasn’t going to be any better down at the bottom. Until we figured out what message for voice communications was going to foul us up on the bottom, we wouldn’t go into the water. That was one technique.

We had perfected that in doing our wreck dives where we had the wreck as the central focusing element for what we were doing. It got more complicated when we went to these vent sites, and we were unable to physically model the vent sites. We had, in some cases, good microbathymetry, and at the very least we had some decent site maps. We would use those as guides, and people would fly their models. We always knew in advance what we were trying to accomplish. This briefing would then get synthesized into a dive plan document, which was distributed to all of the various crew members. You have to appreciate that we had two observers and a pilot in each Mir, so that’s six. We would have an observer and a pilot in each of the deep rovers, so an additional four. Ten people were all going in the water, all having to know exactly what they were doing on a daily basis. An interesting lesson here was that the task loading from a planning standpoint became greater than the task load on the actual dive. In fact, I wound up getting most of my sleep during descent and ascent because I was spending the night working through the documentation for the dive the following day. The pace of operations was inappropriate to the scale of the logistics of what we were doing. That was the thing that emerged.

Each crew member got a dive plan which was individually tailored to their vehicle in terms of the timeline and their activities — the individual objects for each crew and the science activities as well. The science activities required a separate pre-dive meeting by the science group who would bring us their requests and recommendations for modifications to the sampling equipment on the front of the subs.

One of the things I would like to express here today is the idea that, regardless of how much you plan, you have to be willing to accept the idea of failure. I think that we are enthusiastic fans of exploration, probably everybody in the room, but failure is a part of exploration. It is absolutely woven into the fabric of the act of exploration. By definition, exploration means you’re doing something that has never been done before. It is absurd to assume that activities without precedent can be done in complete safety. If only the remote and hostile environments are yet to be explored, then we are inherently pushing the limits of human endurance and technical adaptation every time we advance the boundary of what is known.

It is absolutely important to use all of our accumulated knowledge to be as safe as possible. However, safety is not the most important thing. I know this sounds like heresy, but it is a truth that must be embraced in order to do exploration. The most important thing is to actually go. Because if safety were the most important criterion, we would not go to Mars for 10,000 years, because only then could we assure absolute, 100 percent success. Historically the success of cultures and nations has been the result of their ability to balance risk and reward — to put it another way, caution and boldness.

The problem with exploration is not the individual’s perception of risk; it’s the institutional, national and political perception of risk. Astronauts are smart people — I know a few of them. Most of them are Ph.D.s in one thing or another — engineering, physics, medicine. They know that riding a pointy end of a metal object that is screaming through the atmosphere at 20 times the speed of a rifle bullet, being propelled by one long continuous explosion is not quite the same as sitting at home in your Barcalounger.

They understand the dangers. They get it. They have assessed the risk. But their personal dream, their vision — not for themselves, but for the entire human race — dwarfs that risk. They know the importance of what they are doing because in their souls they are explorers. It’s not the astronauts who are going to hold up the progress of exploration. It’s the government that funds them and the people that empower that government to act who will set the limitations. Institutions gravitate inexorably toward a value system in which any risk becomes unacceptable at which point exploration ceases.

Now, we are lucky right now to be on a cusp with history where a presidential mandate has put NASA back on track with a renewed vision for exploration. NASA has reorganized around the guiding principle of exploration beyond Earth’s orbit. This is all very exciting, it is all very new, and it is definitely happening. I believe it is a wise plan, and an affordable and achievable plan.

But there is one huge challenge that still needs to be overcome, even if we deal with all of these short-term reorganization issues. We must overcome the fear of failure that may inhibit future leaders from allowing these missions to proceed. The challenge will be this: the only way to fail in landing humans on Mars is to actually go. If we study the problem, we build tools and systems and so on for the next 50 years, we can kind of jolly ourselves along that we are really honest-to-God going to do it someday, that we’re still those clever Americans who put a man on the moon back — when was that again?

That way we don’t put our self-image at risk. But the second the button gets pushed and we are really going, then we enter a much higher realm of risk. “Failure is not an option” was a good credo for getting the Apollo XIII astronauts back home safely, but as a driving principle it doesn’t really work. Failure must always be an option, or we stop being an exploring species.

When I started our most recent expedition project, I called a big summit meeting of all the department heads. I stood in front of a white board and put up on the white board three slogans. The slogans were there: “Luck is not a factor”, “Hope is not a strategy”, “Fear is not an option”. Now the first two were meant to convey my philosophy that to succeed in any complex task, it is essential to leave nothing to chance. You need to make your own luck by rigorous application of a robust process. You test everything in a very disciplined fashion, you don’t guess, you know the answer, you anticipate every negative condition that might possibly prevail. You assume it is going to happen. You have an A plan, a B plan, a C plan, and you assume that you’re going to be on the C plan by your second cup on coffee on morning one of the expedition, because that’s how it goes when you’re at sea.

I wanted to scare them, and I wanted them to respect their adversary — not the ocean, but the real adversary: entropy, which, as you know, is the tendency of things to go from a state of organization to a state of chaos.

The third slogan, “Fear is not an option”, was meant to inspire the boldness that actually sees you through these endeavors. It was the yin and the yang of the healthy paranoia which the first two slogans represented, because without a kind of faith which is not in luck and not in passive hope but in yourself and your team and in the greater meaning of what you’re setting out to do, you won’t find the strength to go through with it.

So my message is in whichever realm, be it going into space or going into the deep sea, you have to balance the yin and yang of caution and boldness, risk aversion and risk taking, fear and fearlessness. No great accomplishment takes place, whether it be a movie or a deep ocean expedition, or a space mission, without a kind of dynamic equipoise between the two. Luck is not a factor. Hope is not a strategy. Fear is not an option.