"Life" is negentropic organization, which only seemingly "violates" the 2nd law of thermodynamics, "closed system physics", since the existence of negentropic organizations must be offset by increased positive entropy in the environment. But, by the same token, "life" does not just simply emerge and evolve in relation to an environment: rather "life" emerges and evolves in relation to other "life" in its environment, feeding off of and recycling more or less complex organic materials. In other words, "life" can emerge and evolve only in the form of the development of a biosphere that sustains "life", which biosphere is itself subject to "laws", (or, more properly, constraints), of its own complexity, which are different than the general laws of physics. This yields a de facto 4th law of thermodynamics: a biosphere can emerge and develop only under long-run, quasi-permanent conditions of thermodynamic disequilibrium, (even as the biosphere must "work" to, in some measure, stabilize that disequilibrium).

Now a biosphere, once up and running, tends toward the development of increased diversity and novelty and hence toward increased aggregate complexity, (as different organisms provide niches for each other or evolve responses to competitive threats). But there is no general law for progress in evolutionary processes, but rather a radical lack of teleology. Most evolutionary change is a matter of running-in-place, the "Red Queen problem", in the face of shifting environmental conditions,- (most terrestrial animal evolution concerns insects, not just in quantity of species, but, I would guess, in terms of sheer biomass),- such that there is no general "law" tending toward increased complexity in organisms. (Insofar as certain threshold developments have occurred, such as the evolution of multi-cellular differentiated macro-organisms out of symbiotically functionning colonies of micro-organisms, that would only be due to gains in "constructed" thermodynamic efficiencies outweighing increases in "costs" of reproduction, always under biosheric/environmental constraints). The most that can be said is that the longer the evolution of a biosphere persists, the greater the probability of the emergence of more complex organisms, though only in niches and subject to the preponderance and persistence of a relative stasis of less complex organisms.

Outside of its precise definition in thermodynamics, it's difficult to define "information" in biological processes: what is causality and what is "information" at what level and whether information and causality might not trade places when considered from different perspectives or levels. Equally, biospheric constraints form a level of determination, which can not be reduced to general laws, but is no less efficacious for all that. But the main point is that the covariance between organisms and environments, "adaption", which characterizes and drives evolutionary processes, occurs through the generation of "information" by means of the evolutionary process itself, without which such "information" would not exist. But, needless to say, there is no general tendency to generate "intelligent life", and "life" always outweighs "intelligence".

I got into a long blog-comment debate over at "Pharyngula" with a C.S. guy, when I said that Daniel Dennett's claim that evolution as a whole could be characterized as an "algorithmic process" was a serious distortion and ultimately nonsensical. In fact, it substituted a "instructionist" explanation, involving predefined functions and transparent or pre-compatibilized information, based on "machine intelligence" for a proper selectionist/population account of evolution. I was constantly being put on the defensive, but, to my mind at least, never actually refuted. And one of the biggest drawbacks to Dennett's distortion is its occuclusion of the crucial ecosystemic dimension of "Darwinian" evolution, which is one of the most valuable matters it informs us about. The irony of Dennett's physicalist reductionism is that it ends up backhandedly, through attributing the teleonomy to be found in analysing individual organisms, to evolution as a whole, "finding" more purposiveness, covert teleology, in "Darwinian" evolution by natural selection than is at all warranted.

The upshot here is that "life", as explained by natural evolution, is neither reducible to mechanistic/physicalistic terms, nor is it a source of "metaphysical" insights.