Note though that the whole issue of rotation causing a stress arises precisely because linear motion is somehow "more natural".
The stress exists because the portions of the body "want" to travel in a straight line, and a force is needed to keep them travelling in such a line.
The point of the OP is to wonder why travel in a straight line is so natural in our universe, so saying (in effect), "well given that things want to travel in a straight line, rotational motion requires forces and stress and so is so much less natural than travel in a straight line, therefore things want to travel in a straight line" (as you have basically done) doesn't really answer anything.
In fact all the notions touched on in this thread from Newtonian (and Einsteinian) physics can be packaged up in the (poorly named) "Principle of Least Action" (actually stationary action).
So, why do objects "want" to travel the path of stationary action?
Well we can simplify things a bit and assume that there is going to be some functional we can use, that will give us via a variational principle the observed motion in the universe.
Let's assume this.
Then the question becomes:
Why does the functional we use (namely the action) have the form it has and not another (the other, presumably, giving us other paths)?
The answer is going to go back to whether the universe has preferred directions.
The beauty of the "straight lines" is that they allow the universe to have no preferred directions. Other types of motion will require some directions of space to have different effects on motion than others.
If we assume all directions are as good as any other, we get straight line motion.
In more physicsy terms, it gets back to symmetries.
Hmm, thought I mentioned this, but I guess I left it out of my first post. Sorry about that.
Yes, ultimately it has to do with symmetries. I'll see if I can't explain it in another way:
The reason that objects like to move in straight lines is that there exists some translational symmetry: when it is true that movement from place to place in some region changes nothing, then this means that linear momentum will be conserved. If linear momentum is conserved, then one has to have some sort of interaction to change the direction or speed of anything's motion: you have to have some sort of force.
The next question is, what about angular momentum? If linear momentum being conserved leads to linear motion being natural, won't conservation of angular momentum lead to circular motion being natural? In a word, no. Not in the least: movement in a straight line
also conserves angular momentum, and what's more it conserves angular momentum about any point you might wish to measure it: remember that angular momentum requires one to define a point around which things rotate.
So, one might try to think of a situation where circular motion is more "natural" by proposing a system that is rotationally invariant, but not translationally: it's the same if rotated about some point, but different places are different. Our own solar system is a perfect example of this: rotate about the Sun and everything looks much the same (ignoring the planets and other objects for the time being). But move around from place to place, and things look quite different: the Sun will be either in a different direction or at a different distance from you. The first thing to notice here is that this is
not nearly as natural as the system that conserved
both linear and angular momentum, the system that led to straight line motion.
And yet, even here circular motion is by no means "natural" at all: orbits around the Sun can be any conic section you desire. You can have parabolic orbits, elliptical orbits, hyperbolic orbits. Circular orbits are an extreme special case. So there really is no way to get a system where circular motion is the "natural" thing.