Skip to main content

How did Titanic really break up?

This article was published in Scientific American’s former blog network and reflects the views of the author, not necessarily those of Scientific American


Hello, everyone!

I'm a member of the Marine Forensics Committee, and author or co-author of three peer-reviewed papers on the “Titanic”. My most recent paper, “The Breakup Of Titanic – A Re-Examination of Survivor Accounts”, was presented at the First International Marine Forensics Symposium on April 4.

Working with Roy Mengot (with whom I co-authored one paper), I've been gathering evidence to support a reconstruction of the breakup of the “Titanic” that differs somewhat from the one you may have seen in movies or in other publications. The most important stages in our reconstruction are illustrated here:


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


In our reconstruction, the failure began in the ship's bottom structure, when the ship was at an angle of about 17 degrees. The failure spread across the breadth of the ship, then upward; it also spread forward and aft, probably along lightly riveted longitudinal seams, forming two separate pieces of the double bottom.

When the ship's stern section came back down into the water, buoyancy forces took over the job of holding up that part of the ship, so the breakup stopped for a little while. At that point, the two halves of the ship were held together by the uppermost strength decks, and by the double – thickness side shell plating. The outlines of the two double bottom pieces that broke away from the ship are indicated. The deckhouse is shown opening up at the aft expansion joint – but the split in the deckhouse was an effect, not a cause, of the hull break.

As the bow continued to flood, and as water poured into the engine room (which would have been opened to the sea when the bottom gave way), the bow section pulled down on the forward end of the stern section. The angle between the two pieces of the hull decreased – as if the ship were a giant paper clip that was first bent, then straightened out. At some point in this process, the two sections broke apart completely.

The breakup was not symmetrical. The “Big Piece” (a large piece of the ship's side, recovered in 1998) came from the starboard side – it had no counterpart on the port side. The after part of the ship is known to have listed to port fairly suddenly – so the starboard side may have failed first, leaving the flooded bow section to pull down on the port side of the stern section until the two parts separated completely. This left the stern section to float on its own, eventually standing nearly vertical in the water, then finally disappearing below the waves.

That's our reconstruction. Now, let's look at how it was developed, and why we believe that it's closer to what actually happened than any other model proposed to date.

These ideas began to take shape when I first saw the photos of what were then called the “Missing Pieces”. Looking at the way the keel bent at the ends of those pieces, I was sure that it was shaped by compression. Also, the remaining edges looked too clean to have been formed by multiple twists and bends – those pieces looked like they failed first, not last.

The “Big Piece” seemed to be the other important piece of the puzzle. It seems to have failed along three distinct edges. After years of puzzlement, I realized that the “Big Piece” must have formed part of the temporary hinge about which the stern section bent, first downward, then upward. Such a sequence of loads and ship motions could have created the piece as we see it today (Any attempt to say how each edge broke would be pure speculation – but if the ship broke from the top down, we would be able to explain how only one of the edges formed. Once the ship split in two, there would be no forces remaining that could produce the other edges.).

Any understanding of the sinking has to take into account the “Titanic”'s riveted construction. My first paper on the “Titanic”, published in 2003, gave me some insights into the behavior of the riveted joints. I realized that because of the way the hull was riveted together, the bottom was not as strong as most investigators believed. And because of the way the uppermost strakes (strips) of plating were constructed, they had much more strength than most investigators have given them credit for.

The high stresses around the deckhouse expansion joints – previously believed, even by the Marine Forensics Committee, to have been the starting point for the breakup – turned out to be a “red herring”. The deckhouse was made of lightweight plating, and carefully constructed so as NOT to share in the structural loads on the ship. (I recently rediscovered an old printout of a photo from the Discovery Channel website – no longer available online – which clearly showed that neither edge of the “Big Piece” was aligned with the deckhouse expansion joint.).

Roy and I did our groundwork. We made calculations. We created a computer model of a portion of the bottom structure, and identified potential weak spots. We compared our reconstruction with survivor testimony and the condition of the wreck.

We had help from other members of the Marine Forensics Committee (formerly the Marine Forensics Panel) – especially from its chairman, Bill Garzke. As I mentioned earlier, this is the same body that was once a leading advocate of the theory that the break started in the upper portion of the hull – but they gave us their full support.

We first presented our work informally at a session of the 2007 SNAME Annual Meeting. We developed it into a formal paper, which was presented at a local section meeting in New York in 2009. It was published in 2010 as “The Breakup Of Titanic – A Progress Report From The Marine Forensics Panel (SD – 7)” in SNAME's journal, “Marine Technology”. (The reconstruction given here includes a slight change in the later stages of the breakup, to be presented in more detail in my new paper at the Symposium, but the main features have not changed.)

In professional circles, our work has not been challenged. I was expecting all kinds of objections to be raised during our presentations – we did not get even one. But in some circles, the “top down” breakup model remains in vogue.

According to an article on the History Channel website, our work will be challenged in a documentary to be aired on the anniversary of the sinking.

To get some insight into the views of those who continue to believe that the breakup started at the upper edges, I would recommend a visit to the website of Parks Stephenson, perhaps the most influential advocate of a top – down break. The URL is: http://marconigraph.com/titanic/breakup/mgy_breakup.html.

Does it matter which version goes into the next generation's history books? Well, it matters to me as a professional. It may be a bit of a stretch to call this a paradigm shift, but it does have many of the characteristics of one – and engineers, like scientists, must always be prepared to deal with new evidence that may force us to discard our previous concepts and embrace new ones.

But we also hope that in the absence of such evidence, our work will be respected. If there is to be a debate over the validity of our work, we hope that it will take place in open scientific and professional forums, and that the outcome of the debate will be determined in the best tradition of scientific inquiry.

There is even more at stake in this particular case. The legacy of the survivors is, in a sense, on the line. If our reconstruction is substantially correct, then the survivors' testimony (with a few exceptions) turns out to have been quite accurate. There will, of course, still be some conflicts to resolve, but those conflicts will be easier to understand.

On the other hand, if a “top down” breakup goes into the history books, the survivors will be forever remembered as people who let the stress of the situation blind them to what was happening before their eyes.

Some of my colleagues have suggested that there may be some sort of middle ground – some way to combine elements of both breakup models. If there is such a middle ground, I would gladly stand on it – but at the moment, I don't see any more middle ground here than there was between, say, Kepler and Copernicus.

I've had my say. Now it's your turn. You are the jury. What part of the evidence would you like to see in more detail? What will your verdict be?

Richard Woytowich is a Professor in the Computer Engineering Technology department at New York City College of Technology. Previously, he was a Senior Engineer with the American Bureau of Shipping. Other previous employers include Gibbs & Cox, Inc., a naval architecture and marine engineering firm. He holds a Bachelor's degree in mechanical engineering from The Cooper Union, a Master's in mechanical engineering from Columbia University, and a Master's in science education from New York University. He is a licensed professional engineer in New York State, and has been a member of the Marine Forensics Committee (formerly the Marine Forensics Panel) for over 10 years.

More by Richard Woytowich