Contained and uncontained engines failures, hydraulics, cables, ...
Just a few remarks.
Modern turbofan engines need to show, as part of the certification, that their casing can contain the separation of a fan blade during a maximum power run. There are impressive videos of such "fan-blade-off tests" on Youtube.
What cannot usually be contained is the failure of a compressor or turbine disc. These will exit the case at unpredictable angles (though more or less perpendicular to the rotational axis) and at high momentums and energies. At least since the Sioux City accident, where fragments from the tail-mounted #2 engine of a DC-10 ruptured all three hydraulic lines, one of the design factors mitigating the effects of such occurrences is to route redundant systems through different physical locations, if possible.
Great care is taken that such failures do not occur. But they still do, very rarely.
The DC-9's control systems, including those of the -80 series (otherwise known as "MD80"), are mostly cable-mechanical. The cables, connected to the control wheels, and control columns, move tabs at the ailerons and the elevator, which in turn move the control surfaces themselves by aerodynamic forces. Roll-control is supported by hydraulically activated spoilers, and nose-down pitch control in a stall situation is also hydraulically assisted, if necessary.
Normal rudder operation is hydraulical, but in case of a hydraulic failure, or when decativated manually, rudder control is also mechanical via a tab.
I have no idea about what happened in Madrid, and I would also not offer an opinion as to whether hydraulics are more or less vulnerable to shrapnel than control cables. Both seems possible.
Bernd