Several points .. regardless of whether we are talking cars, planes, or whatever..
(a) the vehicle structure provides for a survival volume of sorts.
(b) the principal intent of the seatbelt assembly is to keep the occupant within that volume so that the hard bits don't do nasty things.
(c) the intent of upper body restraint (typically a sash belt installation) is not so much to prevent injury but, rather, to reduce it.
If the occupant is wearing only a lap belt, the crash dynamics have the occupant initially moving forward and then articulating (folding) about the now restraining lap belt. The main injury risk is associated with the impact velocity of the arcing head as it impacts vehicle structure .. hence the emphasis on head injury criteria (HIC) assessments/measurements in sled crash tests.
Upper body restraint seeks to reduce this impact velocity to a value low enough to avoid serious brain injury. From a practical point of view, the installation geometry and stretch characteristics of belt webbing mean that the sash is not going to preclude head impact but, with any luck, might reduce the impact velocity (under the design crash case situation) down to something in the vicinity of half the unrestrained impact velocity.
If you are of the school which advocates not wearing upper body restraint components, then you ought to take a look at a representative sled crash test of a lap belt restrained dummy. The slowed down high speed film record is likely to change your attitude very, very quickly ... Similarly, the film record will show the greatly improved situation in the case where upper body restraint is worn.
(d) all of the various design standards relate to specific vehicle impact velocities and characteristics. If the actual crash is much outside the relevant envelope, then the end result usually is that the accident becomes non-survivable.
Even then, there are numerous anecdotal accounts of passenger survival in quite high energy (or highly crushed structure) impacts where casual consideration of the resulting wreckage would suggest that the accident was unsurvivable. One well-reported example of the latter situation was the Sioux City DC10 accident where the flight deck crew survived with comparatively minimal injuries ... but the nose/cockpit area was rolled into a little ball sufficient that rescue workers didn't expect survivors there.
The message is ... seatbelts work pretty well most of the time .. but .. like crash helmets ... they don't work at all if they are not worn by the occupants ...
(e) aircraft design standards require that the pilot be able to do all the bits that a pilot has to do with the seat belt installation appropriately done up. If not, the upper body restraint components are required to be installed in association with an inertia reel assembly to avoid the problem. The practical situation is that after market installations are cheaper without the reel so, as often occurs, the necessary reel assembly is omitted. If your specific aircraft is in this category, perhaps a talk with the owner, operator, flying school, etc., about liability might be appropriate. If that achieves nothing maybe the relevant NAA might be approached via the appropriate reporting procedures ....
(f) the question of airline passenger seats fitted with lap belts is a little more complicated as, for current seat design standards, the seat in front becomes part of the design restraint system.
The best rules to avoid or minimise injury ? ...
(a) rule 1 ... don't crash ...
(b) rule 2 ... don't crash ...
.... and so on ...
[ 23 December 2001: Message edited by: john_tullamarine ]</p>