I think there's a lot of good discussion here (and some of it has been covered, but worth continuing to discuss).
Someone above I believe mentioned what's known as the swiss-cheese theory of accidents. Most accidents are NOT random events. They're a series of events lining up so all the holes in the slices of cheese line up. Change one thing and no accident.
Consider for example the recent Amtrak-CSX accident. If the train hadn't been routed onto that track... if they had been proceeding at restricted speed... If the CSX train had put out torpedoes (are those even a thing still?) 1 mile ahead... if this. that etc. Many ifs.
And you can break accidents down into mechanical failure, or I'll broaden that into broader "non-human" failures. Sometimes, even these aren't clear. But let's call the Acela break-up that. (I'll come back to that). But it appears the pin holding the drawbar broke/fell out or something.
And finally the human factors. This gets complex. This can be training, knowledge, skill, random factors, morale, or cultural.
One of the great reads on this topic is
The Challenger Launch Decision. It's a great analysis of the factors leading up to the disaster. It combines a bit of everything. Mechanical issues (O-ring burn thru), management issues ("prove to me we're NOT safe" vs. "prove we're safe"), and cultural issues, "we think we understand the problem" and more.
So.. with Amtrak... I think there's a lot of all the above going on.
As Tadman suggested, a drawbar like that simply should NOT fail. Especially after this many years and miles when wear and tear should be well understood. So is it something mechanical? Or a false faith in "we understand the wear and tear and stresses" or a training "we failed to inspect according to our given rules" or what?
CSX - We've become inured to accidents like this. We really have. They happen far too often.
And personally, in my opinion we're FAR to reliant on "Oh, PTC would solve this!"
I think it's in Charles Perrow's
"Normal Accident" (might one of the other books on my shelf) he discusses how to ships, equipped with radar collided into each other. Why, because both skippers could see where each other were.. so got closer and closer, assuming the OTHER would move. i.e. the use of technology led to bad judgement.
We see that with car features, like ABS brakes. They DO help. There's no doubt. But many safety features for cars never live up to the estimated number of lives saved.. why? Because it appears that folks see the safety features as allowing them to drive a bit closer to "the edge". To be truly safer, one has to understand the value of extra safety features and NOT use it as an excuse move closer to the edge...
i.e. you need the right culture.
And yes, so I think PTC is a must... but we still need to look at the culture around Amtrak and railroading in general.
There's a saying as I understand it (correct me if I'm wrong) that every rule in the rulebook is written in blood. So they can't be tossed aside lightly. But some things may need to be changed. Part of that includes stopping saying, "we can't do that because..." or "we have to do that because..." and go back and look at the ultimate goal and if there's now better ways of doing that. (better technology, better management styles, etc)
When we start to look at countries like Japan where they run their Shinkansen trains w/o fatalities for decades, we see differences. Some of it is non-human (closed corridors and the like). Some of it is cultural, calling out every signal (which even with one person in the nose has shown to have a positive impact). Heck, in programming some of us use
Rubber Duck Programming because it makes sure to focus and change how we think.
So... long winded soapbox...
I think a discussion IS important on this topic. And yes, some feelings may be hurt. Often culture and "that's how we do things" has to be reexamined at and that can bruise egos and cause hurt feelings. And it's a topic close to my heart (safety in general. I love studying disasters, how we approach them and how we change our thinking process to avoid them. It's why I wrote a
bookand have a
blog where I occasionally discuss it.
So I'm all for discussion. As long as we can keep polite and professionals (or the moderators can make sure we are).
Check out
QuiCR, Quick, Crowdsourced Responses for businesses.