Quite right Firefrog - there are many volumns written on human behaviour & the causes of incidents. Even down to the design of control panels, layout of equipment, logical sequences etc. For example we had a double fatal the other night (it started as a search). The cause will be human error, however they were foreign students (who drive on the wrong side of the road), brand new car(familiarity with equipment), poorly lit windy road with sharp corners & a large unprotected irrigation channel. The root cause will be driver error (speed+alcohol), however there are a number of contributing factors which if removed may have prevented their deaths or at least reduced the severity - including the channel being full of water to celibrate the 50th birthday of the weir! Normally it would have been empty at this time of year. As an aside the nuclear power industry did some land mark research on this subject (for obvious reasons) in the 60's, petrochemical & aviation has followed as well as the US navy, to try and reduce the likelyhood of human error. And finally if you think about the control of a major incident, a look at the amount of information coming in & going out (inputs/outputs)& decisions that have to be made (& timeframes)people wonder why there are mistakes/ or their radio call isn't answer immediately! Some thing to consider!
cheers