The simple answer is: relatively thin material (by comparison to other blocks) and a lot of outward force.
For the more complex answer, I'm going to type as I think. So any of you physics gurus that care to chime in or correct any of my misconceptions, be my guest. I enjoy deepening my understanding of the physics involved.
The net force acting on the engine block itself is not zero. Within the block (aside from gravity) there is both a net rotational force, and a net upward force. The gross forces acting on the upper part of the block include a force perpendicular to the deck, which is to say up and out, and an opposing force pushing down and in. The critical forces to consider in regard to your question are those that oppose each other at the top and bottom of the cylinder, the piston and head. Effectively these forces are acting along the 2 axes (plural of "axis") of the engine's V. One force acts by pushing "up and out" against each head while the opposing force pushes "down and in" against each piston crown. There is also a force pushing outward against the cylinder walls. Now I'm going to break each of these three forces down into more detail (skip to #3 for the answer to your question):
1. Cylinder wall forces due to cylinder pressure are the easiest: within the cylinder, the pressure on one side of the cylinder wall is cancelled out by the pressure on the opposite side of the cylinder. If the structural integrity of the lining and the block material is strong enough for this, distortion is minimal and the elasticity of the material allows the cylinder to retain its shape (elasticity, and metal fatigue are tangent to this subject). Although cylinder wall distortion obviously does happen at extreme power levels, this is not a huge concern in our blocks.
2. You know what happens to the "down and in" force. Some of it is transferred into the "rotational force," or torque that accelerates the car, and some of it pushes the crankshaft downwards against the bearing and/or the oil coating it. Since some of the downward force converts into rotation and none of the upward force does, these forces do not completely cancel each other out within the block. That means there's an actual upward force that acts against gravity and possibly the motor mounts, too. Also, the torque output at the crankshaft is equal to the torque applied to the engine block in the opposite direction, which is also resisted at the motor mounts. It took me a bit to think through this, but the torque applied to the engine block is actually a result of the net force that the pistons apply to the cylinder side walls. The 5.0 engine produces torque in a clockwise manner as you look at the engine from the front of the car. That's why the engine spins clockwise, and the opposing torque acts counterclockwise at the engine block, which explains why the driver's side wheel and tire are always the first off of the ground in a non-reinforced chassis.
3. Ok, finally! The force "up and out" applied to the heads is resisted by the head-bolts and then through block tension. The upward force is opposed through block tension partially to the main bearings/caps where the crank is applying the previously discussed "downward" force, and partially by gravity (the weight of the block), or if the upward force overcomes the engine's weight then the motor mounts also resist it. The outward force on one side of the engine is opposed through block tension by the outward force on the other side. When the tension through the block overcomes the block's strength at its weakest point in the lifter valley, it literally pulls the block apart. This can happen in one instant, or it can happen over time as the block is pulled beyond its elastic limits and cannot retain its shape.
I'm sure that the forces applied by the rotating assembly and the counterweights contribute, but by and large these forces are resisted at the main bearings, and I do not believe they are a significant contributer to the block splitting at the lifter valley. However, at exceedingly high RPM, they are extremely damaging to the crankshaft due to the flexing they cause and also to the main bearings as the caps "walk."
Blown engines go earlier because the internal stresses on the crank are increased, and because the blown cars have to overcome the parasitic loss of the blower, which can take some serious hp just to turn the belt. You don't see this power at the wheels or even at the crank, but the block "feels" it because the cylinder pressure is still there, and thus the "outward" force is there as well. Though a turbo car is easier on the block for this reason, you're still talking about a huge amount of force that's being exerted on the block.
Chris
