The big problem with attempts to reduce libertarianism to a simplistic absolute principle is that the resulting political philosophy is inevitably under-specified. The list at http://libertarianmajority.net/free-variables-in-libertarian-theory now stands at over two dozen.
Even if we somehow agreed on axioms to fix all those variables, there is still the fundamental question of whether our opposition to force-initiation is about pledging abstinence from it or about seeking to minimize its net incidence. When non-aggression fundamentalists can actually be made to confront this distinction, their most common response is to hand-wave that the former is the only way to accomplish the latter. They apparently have never heard of things like http://en.wikipedia.org/wiki/Hill-climbing#Local_maxima or the http://en.wikipedia.org/wiki/Trolley_problem.
What's worse, the artificial feeling of axiomatic certitude is an intoxicant that throws off one's natural moral equilibrium, and leads to (among other things) intra-movement inquisitions. (For example, consider all the Angrytarians who vilify the Cato Institute and Reason Foundation.) The most dangerous force in human affairs has always been the absolute certainty of moral rectitude. I'm of course not saying that all morality is equal. I'm just saying that many Libertarians think that they personally have the most comprehensively correct set of political ethics ever assembled in human history, and yet none of these Libertarians seem to agree comprehensively with each other. Shockingly few of them are able to make the inference that is screamingly suggested by these two considerations.
The fundamental speed-of-light-style limitation we face here is the Is-Ought Problem, which is so tough it has very nearly been codified as the Naturalistic Fallacy. We should by now be extremely skeptical that there can be an argument that starts out with something like "A is A" (or any other set of "is" statements) and ends up with most people slapping their forehead and exclaiming "wow, I ought to be a libertarian!" (Is that how you became a libertarian? We need a systematic study of what turns people into libertarians.) If I had to bet on an argument that could solve the Is-Ought Problem, I'd put my money on Fred Foldvary's Universal Ethic.
The true measure of robustness of a theory of political ethics is not its axiomatic simplicity. Two other inter-related considerations are even more important.
First, how general is the theory across time and space and possible circumstances? Does it work for agrarian village life and for a 21st-century globalized society and for desert-island/lifeboat scenarios and for future human colonies remote from the civilization on Earth's surface -- and for transitions among all the above? Does it work when there are interactions between individuals and societies -- and even species -- that enjoy quite different levels of development and capability?
Second, does the theory have good potential to be considered morally persuasive to persons of varying backgrounds -- gender, age, intelligence, learnedness, wealth, ability, fortune, reproductive outlook, culture, religion, epoch, etc.?
When you think in these terms about the challenge of spreading the libertarianism meme, you start to realize that bumper-sticker-sized axiomatic simplicity might not be the only meta-theoretic virtue -- no matter how handy it is in making you think you've accomplished the great feat of winning a debate with somebody who already calls himself a libertarian.
(OK, cue one set of Angrytarians to complain that this posting opposes the Non-Aggression Principle, while yet another set of Angrytarians will complain that talking about ideas makes us a "debate society" incapable of working to advance those ideas.)