Sean Carroll 2016. The Big Picture. New York: Dutton, 470 pp.

Sean Carroll sets out to explain what we know about the nature of our physical world and, more importantly, how (what scientific and philosophical methods and investigations) have led us to that understanding. The scope is therefore similar, for example, to David Deutsch’s The Beginning of Infinity but lacking that book’s exciting and daunting vision of what an infinite universe really implies. To my mind, Sean Carroll’s treatment is too heavy on philosophy at the expense of the foundations of mathematics and physics for a book with this ambition. His outlook is a philosophy he defines as “poetic naturalism”. Carroll’s naturalism is just what anyone else would call empiricism, that there is nothing in our experience (consciousness, free will, the origin and nature of life, etc) that cannot be explained from what we know of the physical and quantum world. The poetic part I never got and doesn’t seem useful; he is at pains to emphasise that there is no “purpose” or “why” to the empirical questions under investigation, which is fair enough, but it is a mystery to me why he feels “poetic” conveys that. Nevertheless, plenty to think about and he regularly uses Bayesian reasoning as a way of assessing competing explanations.

The first sections of the book are about the physical world and forces and our understanding of it: physics, the quantum world, cosmology and the universe, also Bayesian methods (which are explained early on and then applied often throught the remainder of the discussion).

The next sections are about entropy, complexity, origins of life, genetics and evolution. And computer simulations of some of these things. Although a physicist, Carroll is a good guide to biology, genetics and evolution but these topics have been covered in many in other places so skipped through this.

The penultimate section is titled Thinking and it is about consciousness, artificial intelligence, memory, free will and related topics. Carroll again tours well-trodden paths here but with the goal of showing that none of the phenomena of interest are beyond understanding using an adequate knowledge of the physical world (which we have now) and of the brain (which we are far from having). The contrary positions of many philosophers are dismissed as unnecessary and unsupported hypotheses.

In the final section Carroll declares his personal philosophy: how he lives a short, finite life as meaningfully as possible in a universe without God, afterlife and without human-directed purpose. It reads like a support session for recent converts to empiricism and atheism and I didn’t need it nor did I read much of it.

A quick summary of things that were interesting, or that I disagreed with, or just want to learn more about:

  • I found Nick Bostrom’s simulation argument was dismissed way too glibly (p. 92).
  • I now need to read about Robert Aumann’s proof as stated on p. 121: “two people, both acting rationally, who start with the same Bayesian prior credences for their beliefs, and who have access to the same information, including knowing what the other knows, cannot disagree about the updated credences for those beliefs”. Sounds intuitive but codifying as a formal mathematical proof is another thing. I am now aiming to read more on this, perhaps starting with this interview with Aumann.
  • I didn’t know that the psychedelic effects of LSD and other drugs are not caused by raising neural activity levels but, counter-intuitively, because the drugs suppress regions of the brain which normally filter from our consciousness incessant neural activity occurring in other regions of the brain (p. 137).
  • Carroll’s summary on pp. 166-171 of Hugh Everett’s Many-Worlds Interpretation is the clearest I’ve read. And all the more so for the close juxtaposition with a similar description of Bayesian reasoning a few pages earlier. Everett proposed in essence that the universe is a quantum wave function which does not “collapse” when an observer intervenes to make a measurement. Instead the wave function (if I understand it) can be thought of as having many solutions, each of which is a solution to the Schrödinger equation. Most importantly, the solutions have probabilities (of course) and thus can be investigated with Bayesian methods, as many have done. This lead me to speculate that perhaps Bayesian reasoning is not just an empirical tool to better approximate probabilities. Just perhaps Bayes’ formula is so so spookily successful across so many fields because it is a local approximation to the quantum wave function? Maybe it is either circular (or just plain invalid and loopy) to think that to Bayesian reasoning might be a more fundamental component of Everett’s solutions to the wave function but many physicists seem to think that Everett’s Many-Worlds Interpretation itself is loopy. Perhaps Everett himself thought this way (he was an early Bayesian). Probably this is beyond anyone who can’t work with Schrödinger’s mathematics but anyway I intend to read a bit more on this, starting with The Many Worlds of Hugh Everett in Scientific American. (An Appendix contains a presentation of the Schrödinger equation in the form of Richard Feynman’s path integral formulation of quantum mechanics.)
  • Niels Bohr: “It is wrong to think that the task of physics is to find out how nature is. Physics concerns what we can say about nature.”
  • Perhaps what is most glaringly absent from any book such as this is a discussion of whether in principle it is possible to understand a system that one is a part of. (I suppose Gödel, for one, would have said no, it is not possible?) This is an issue for seeking a complete understanding both of the universe, and of our brain and consciousness. I’m not sure that anyone has addressed this fully.