I do not usually shoot of little posts re-tweeting stuff I found, but the total destruction that is David Albert’s New York Times book review “On the Origin of Everything” reviewing “A Universe From Nothing: Why There Is Something Rather Than Nothing” by Lawrence M. Krauss is not only totally spot on correct and well written but moreover makes an important point toward the end, one that I often also try to hammer into science cheerleaders' brains, for example with Mix Science and God correctly or Don't: If you do not really know extremely well what the hell you are talking about, if you do not know the involved science very well AND simultaneously are also aware of the philosophical intricacies, DO NOT USE SCIENCE TO ARGUE AGAINST GOD! Why? Because YOU WILL FAIL and IT WILL BACKFIRE helping only one side: The religious nuts.
OK, David does not exactly say it like this, but one should stress the important point that he does also point out that proper and effective criticism of religion is entirely different and at most suffers from adding silly science.
Krauss has written a book whose main impact, apart from making him money and feeding confirmation bias in the target audience, is that religious people have yet more evidence proving that scientists and science writers simply do not understand the issues properly and have reduced themselves to selling mediocre books to a semi-intellectual new “third culture” mob of know it all loudmouths ridiculing all other religions but their own: naïve scientism.
But all there is to say about this, as far as I can see, is that Krauss is dead wrong and his religious and philosophical critics are absolutely right.
Thank you David Albert for having the courage to be outspoken in a discussion that is way too often tolerant of nonsense because of ill understood “professionalism”. [This is for example why the “Institute for Ethics of Emergent Technologies” (see my IEET profile here) refuses to publish anymore stuff from me: my take down of an academic pseudo-scientist was O-tone “unprofessional”! It is unprofessional for scientists to unrelentingly argue against pseudo-science; go figure. David seems to not care; of course, he also does no longer need to in his position.]
There are many now that try to jump onto the “Look I know why there is something instead of nothing” bandwagon, partially because it is funded well recently, thus developing into a burgeoning field; an also pretty bad attempt was just now delivered by Ethan from “Starts with a Bang”. The more I read of these, the more disappointed I become and the more confident I also become in simply promoting my own insight into the question (and yes, it is about the question first of all, not the answer), especially because the way I do it is the only one I have seen out there that is atheist in a mature and scientific way: No, I do not hype the newest bit of stringtheory in order to hide god down one more level and then rant about why criticism of such should be ignored because it comes from republicans who deny global warming and do not like condoms. In a scientific approach, pseudo-questions and magic ghosts simply do not arise during the use of sharpened terminology.
Anyway, go over there to the NYT and have a read. David Albert uses some nice images like that of the arrangement of fingers that make a fist and how fists do not pop out from nothing. It is well written in a way that speaks to a wide audience without trivializing the issue too much.
As far as I can see, there's two possible ways out of this dilemma. One would be to argue that some physical law is mandatory, and that this law then realizes the (naive) ex-nihilo creation propagated by Krauss and co. One possible candidate for such a thing might be the second law of thermodynamics: anything at all that can be in distinct states obeys it automatically, since it effectively just states that the most likely of these states are attained most often, which many surely view as tautologous. But there, we face some Münchhausen problem: the second law needs something on which to act, or for which to hold; but something, in order to exist, as per the above, would need the second law (or some other law) to hold.
Maybe then one can look to quantum mechanics: view it, not so much as a law of physics, but instead simply as a calculus of non-commutative probability, or, as per Zeilinger's principle, as a probability calculus appropriate for systems of fundamentally limited information, hold it thus to be mandatory, and then mumble something about quantum fluctuations etc.
Another way to look at this might be to observe that, at least, law can come from non-law: a given string of random digits is completely lawless, but still, its gross properties (relative frequency of digits, amount of deviation from this to be expected, etc.) are at least probabilistically predictable, so anything operating on the level of these gross properties would tend to follow fixed laws as the length of the string grows to infinity. One can then point to the similarity to quantum mechanics: the laws of mechanics (for instance) are not something put into the theory from the beginning, rather, they emerge in a similar way, through interference of all possible paths, each of which (or almost all of which) do not themselves follow these laws. F = ma (or more generally the principle of stationary action, etc.) then becomes a law similar to 'there are as many 0s as there are 1s in a random bit string'.
This already comes close to the second option, which I think is also the one you favor (though correct me if I misread you): observe that 'nothing' and 'everything' aren't really all that different; both, for instance, contain zero information (as measured, for instance, algorithmically: a program producing no output and a program producing, say, every possible bit-string are both of asymptotically zero complexity). Laws then get the character of constrains, selecting out the specific something we observe from all the other possible somethings: this just says that the constructive definition of (a set of) something(s) is of the same complexity and information content -- is a dual description of -- its complement within some totality -- i.e. getting something from nothing is the same as starting with everything and pruning it down to some specific something.
Well, or something like that; I've probably rambled on more than enough...