I have struggled with writing a review of Stephen Meyer’s book, Return of the God Hypothesis, since I finished it a few weeks ago. Every time I pick it up to reread portions of it I find myself wanting to approach the work from a different perspective. The book is neither a straight popularization of science nor an attempt to frame a clear scientific argument. Rather, it’s a well-crafted work of reporting and speculation at the frothy margins of scientific theory that, combined with a few leaps of logic, is harnessed in support of a foreordained conclusion.

I suspect that the science in this book – and there’s quite a lot of it – will, despite being well-presented by an eloquent and talented author, largely elude most readers. Perhaps more importantly, the context from which the science is drawn will likely be unfamiliar to most readers, who will have little familiarity with physics and cosmology beyond what is presented in this book. If this book were merely a popularization of the science of cosmology, that would be fine: people would gain a feel for the state of the field, for its complexity and nuance, and for the remarkable accomplishments that have been made in recent years. But that’s not what this book is. Rather, it’s an attempt to support a metaphysical argument by portraying science as inadequate both in practice and in principle, and so leave no plausible alternative but the eponymous God Hypothesis. To frame that argument responsibly would require considerably more scope and rigor than this already science-heavy book offers. To do it convincingly, on the other hand, requires much less effort, particularly if the reader is inclined to be generous and knows little of physics.

It has been said of Stephen Hawking’s bestselling book A Brief History of Time that it was purchased by many and read by few. I suspect the same is likely true of Return of the God Hypothesis: for many, it will be a tough read. Yet it is an impressive book, and it has lent a great deal of talk-circuit credibility to its author and his premise. The fact that Mr. Meyer is an eloquent speaker and a clever and charming guest undoubtedly adds to that credibility, and it’s understandable why he and his book have received as much praise as they have. Nonetheless, as I will attempt to explain in this review, I think his arguments are weak and his conclusions unsupported.

The book begins with a review of the relationship faith and science have enjoyed throughout history. Meyer is on solid ground when documenting the history of science, and his recounting of man’s march of discovery is readable, detailed, and entertaining. It isn’t relevant to his argument, but it is well-written and informative.

Then we get to the science. Meyer asserts, based on three “scientific discoveries,” these key ideas underlying his argument:

1. the universe had a beginning;
2. from the beginning (or shortly thereafter), various physical constants have had values that are unlikely to have arisen by chance – that the universe appears to be “fine-tuned”; and
3. the genetic coding in DNA represents a kind of “functional” information that is unlikely to have arisen by chance.

I find Meyer’s defense of each of these claims wanting, but, before I critique them, let me make two brief comments, one about the nature of Meyer’s three claims, and the other about the overall thesis of his book.

Meyer’s core argument is that our universe and the life in it are improbable – so statistically improbable as to defy any explanation other than that God designed and created it. It’s difficult to overstate the importance of this: Meyer’s thesis hinges entirely on that alleged improbability.

We can see the probability argument arising from the second and third claims, that of the universe being improbably fine-tuned, and of genetic material having an improbable amount of structure and function. But the first claim is different. It isn’t a claim about probability, but rather part of a necessary precondition to all of Meyer’s arguments. It is essential to his arguments that the universe be finite. It must have had a single beginning; it must eventually end; and there can only be one of them.

Why? Because in order for his statistical arguments of improbability to carry any weight, it’s necessary that the sample space not include an infinite number of instances. This is true because in an infinite number of universes everything that is statistically possible, however statistically improbable, will still happen – in fact, will happen an infinite number of times. And in an infinite number of those instances, the physical constants will have the seemingly improbable values we observe, the seemingly improbable chemistry will have arisen to bring about life such as us, and we will, as improbable as it may seem, be sitting here discussing his book.

Regarding the thesis of his book, I have a problem but I’m not quite sure how to state it. Science, including the science Meyer attempts to disprove in his book, has set itself upon the task of answering the “how, what, and when” questions: how does the world work, what laws govern it, when did or will various events occur? Meyer offers an answer to a question science doesn’t ask: “who?” Meyer wants to tell us who created the universe. He doesn’t attempt to present or defend an answer for any of the questions science asks and seeks to answer.

This seems important to me because it suggests that, contrary to Meyer’s oft-repeated claims, the God Hypothesis actually has no true explanatory power. Rather, it merely claims to name an actor – and an ill-defined actor at that. I wonder, how is Meyer’s claim stronger than this one:

“Some non-sentient but unknown natural mechanism, of which we are as yet completely and utterly unaware, established the conditions under which our observable universe exists and the life within it flourishes.”

That wildly ambiguous claim would at least be rooted in something that is consistent with our universal and repeated experience (as Meyer might put it), that of physical reality and the laws that govern it.

In any case, the fact that Meyer’s hypothesis doesn’t actually answer the questions science asks, and that it opens up a universe of new questions (where did God come from, how does God do what God does, what does the mathematics of God look like, etc.) in the process of not answering them, should give us reason to pause, at least.

Claim: The Universe Had a Beginning

Meyer is a science historian, and his account of the evolution of scientific theory regarding the origin of our universe is readable, detailed, and interesting. Most of what we think of as modern cosmology is quite modern, much of it less than a hundred years old, and some of it only a few decades old. It’s sobering to realize how much of what we know we figured out in just the last 50 years.

Yes, we’re pretty sure that everything in our universe was contained in a microscopic pinpoint about 14 billion years ago, and that that pinpoint expanded with unimaginable speed – and continues to expand today. That idea comports with our observations, and the theory supporting it seems robust. Meyer’s account of how we reached that understanding makes for good reading.

But no, we aren’t sure that the universe had a beginning. We admit that things – matter, energy, physical laws, the nature of space and time itself – were likely very different when the stuff of a billion trillion stars occupied a volume vastly smaller than a pinhead. (How many stars can dance on the head of a pin? All of them, it seems.) But we don’t know how they were different. Nor do we know what came before, nor what prompted the expansion, nor whether it happened exactly once or infinitely many times, or indeed whether or not it’s happening right now elsewhere in our own universe. We speak informally of the Big Bang as the beginning of our universe, but all we really know with confidence is that it was a moment in an evolving series of physical states. We don’t know what states came before, nor what states will follow our own.

Meyer is, in my opinion, too casual in his use of the word “beginning.” In Chapter 6: The Curvature of Space and the Beginning of the Universe, he quotes Stephen Hawking and G.F.R. Ellis as writing (in The Large Scale Structure of the Universe) that the general theory of relativity implies “that there is a singularity in the past that constitutes, in some sense, a beginning of the universe.” (emphasis mine)

What did Hawking et al mean by “in some sense?” I don’t know, and Meyer doesn’t pursue it. But it’s hard to conclude that a beginning “in some sense” is the same as, simply, “a beginning.” And in fact, later Meyer quotes Ellis as observing that some cosmologists now see, in Meyer’s words, “singularity theorems as an interesting piece of pure mathematics, but not as proofs of the beginning of our actual universe.” (again, emphasis mine)

In the same chapter, Meyer quotes Paul Davies, in reference to conditions in the very early universe, as saying: “If we follow this prediction to its extreme….” But must we follow mathematical predictions to their extremes? In particular, when it is widely acknowledged that we don’t know which of our physical laws pertain in the extraordinary conditions in the very early universe, how much stock should we place in predictions followed to that extreme?

It is worth remembering how little we understand of the conditions immediately prior to the expansion of the singularity – assuming there was a singularity. We don’t even know if the view conventionally held, that we can know nothing of the universe prior to the expansion of that initial singularity, is actually correct. We thought it was, but then the late Stephen Hawking made the case that black holes might evaporate through quantum processes, and Roger Penrose theorized that we might find echoes of that evaporation in the cosmic background – echoes of black holes that existed before the singularity itself. We now think that, just a couple of years ago, we may have identified one of these so-called “Hawking points,” these shadows of long-gone black holes of a prior universe, in the cosmic microwave background.

If we did – and it’s still too early to be sure – then the idea that the universe began with the Big Bang will have to be reworked a bit. Indeed, the entire idea of there being only a single universe would be effectively discredited.

Speaking of Roger Penrose, I find the omission of his Conformal Cyclic Cosmology theory odd. Meyer cites Penrose dozens of times in his book, but I’ve found only two references to Penrose’s own recent (2010) theory of how the universe might recur endlessly, both in the footnotes and neither actually engaging the theory. Meyer spends time critiquing less mainstream theories, including that of Max Tegmark whose theory, as Meyer describes it, claims that “every possible mathematical structure imaginable has a physical expression in some possible universe” or, quoting Tegmark now, “All structures that exist mathematically exist also physically.” This seems to me to be a peculiar prioritization on Meyer’s part, and makes me wonder if he is being perhaps too selective in the theories he chooses to present to his readers.

I believe Meyer does his readers a disservice by not accurately portraying the range of multiple-universe theories currently proposed, given that rejecting all of them is critical to his thesis. Meyer requires that there not be an infinite number of universes, either one following another throughout eternity or any number existing simultaneously in parallel. This is perhaps the strongest challenge to his argument from improbability, and it deserves to be treated with more rigor.

Claim: The Universe is Fine-Tuned for Life

Meyer’s second claim is, I think, his strongest, and its defense constitutes the largest portion of his book.

There is a widely held belief that our universe is a very improbable place, and that if any one of a few physical constants differed from its current value by an almost unimaginably small amount, the result would be a universe that could not contain us as observers. Some argue that such apparent precision is unlikely to occur naturally, and so is evidence of intelligence: that these constants were “fine-tuned” to be precisely what is necessary to allow the universe to develop as it has.

This is the core argument Meyer makes, and it can be compelling: accounts of extraordinarily improbable-seeming things can be powerfully persuasive.

But it’s worth considering what is implied by the claim Meyer is making. First, it requires that it be meaningful to speak of “different values for a physical constant,” and it isn’t immediately obvious that that’s the case. After all, we don’t know why physical constants have the values they do, and we don’t know how the various constants might be related to each other through some aspect of physical reality of which we’re still unaware.

Consider Einstein’s theories of special and general relativity. Until barely a hundred years ago we didn’t know that space, time, mass, and energy were mathematically related. We didn’t know that these aspects of the physical world were intertwined in mathematically determined and measurable ways, and that a value in one domain couldn’t be “changed” without influencing the other domains as well.

Now we think we know that most of the handful of (what we think are) fundamental forces and particles are mathematically conjoined, not truly independent of each other. It is no longer safe to assume that any of these things is truly independent of the others. And, just as it is nonsensical to speak of, say, the value of pi changing in relation to the circumference and diameter of a circle – because it is inherently linked to and constrained by both of those things – it may be nonsensical to speak of any given physical constant changing as well.

In particular, to the extent that the apparent fine-tuning of constants related to gravitational attraction is prominent in these examples – and it is – we should be particularly careful, as our best quantum theories still struggle to incorporate gravity, and especially under the exotic conditions of the early universe.

Perhaps there’s some meaning to the idea of certain physical constants “having different values.” It’s certainly essential to Meyer’s claim of so-called “fine tuning.” It is also certainly debatable, and debated.

But, for the sake of discussion, let’s assume for a moment that it’s meaningful to discuss the possibility of physical constants having different values than they do. Meyer’s contention is that, in most of these configurations, intelligent life could not form in the universe that unfolded from these different preconditions.

That strikes me as a very bold claim. To make it, one has to believe both of the following: first, that we can accurately predict the nature of a universe that follows laws other than the laws that govern our own universe; and, secondly, that we have a reasonable understanding of the range of conditions under which intelligent life might arise, and the nature of that life, in universes both like and unlike our own.

It’s worth noting that there is a great deal we still don’t know about the only universe of which we’re aware, the one we can actually observe. And it’s worth noting that we really don’t understand the mechanisms of intelligence, nor the mechanisms by which life emerged, nor whether there is life anywhere else in our universe including on those planets and satellites within reach of our own small blue orb.

Given how new and incomplete our own knowledge is of the universe we inhabit and the rules that govern it, we should be skeptical that we’re capable of anticipating the infinite range of alternative universes that might arise through the modification of various physical constants. Certainly, we have not invested thousands of cosmologist-years in studying these hypothetical alternatives.

Similarly, given that we have exactly one example of life from which to generalize in a universe likely containing literally trillions of planets, it seems prudent to hesitate before speaking with authority regarding which possible universes can and can’t support life.

Claim: The Genetic Code is Evidence of an Intelligent Designer

Put simply, Meyer’s argument here is based on the observation that the genetic code – the encoding of information in the DNA of living things – represents a particular kind of “functional” information storage mechanism that is unlikely to have arisen through purely natural processes. I find this the most unsatisfying of Meyer’s claims.

Meyer argues in his book and in his numerous public appearances that, in our consistent and repeated experience, every instance of such functional information storage is the result of a guiding intelligence. It follows, he argues, that the storage of functional information in DNA must also be the product of a guiding intelligence.

This seems to be such an obviously poor and illogical argument that I find myself wondering if I am missing something profound. But let’s break it down.

1. We are aware of numerous examples of the encoding of “functional” information in a structured form, from computer programs to grammars to all sorts of artificial symbolic schemes.
2. Our experience with all of these is that they are the product of intelligence. Specifically, they are the product of human intelligence.
3. It is, therefore, our uniform and repeated experience that such encoding is the product of intelligence.
4. But we are also aware of the encoding of “functional” information in a structured form in the DNA that is found in each of our cells. It follows, therefore, that this information too must be the product of intelligence, since it is our universal and repeated experience that all such information is the product of intelligence.

But wait. That is – at best – a circular argument. If we include DNA in our initial inventory of “functional” information, then it’s no longer our uniform and repeated experience that such information is the product of intelligence. Rather, it’s our uniform and repeated experience that man-made encoding of information is man-made. That says nothing about not-man-made encoding of information.

(On the other hand, it does seem to me that Meyer would be more consistent if he argued that, since every instance of encoded information of which we’re aware is actually man-made, DNA must also be man-made. But that would be an even more absurd argument.)

Instances of functional information storage in DNA both predate and outnumber every form which we can trace to an intelligent source – that is, every form which was created by man. Our actual experience is that every cell in every organism contains a vast amount of structured, functional information for which we can identify no creating intelligence. There is no basis, therefore, for his oft-repeated claim that, in our consistent experience, such storage is an artifact of intelligence, and the fact that he continues to repeat the claim strikes me as peculiar.

Note that this is subtly different from a probability argument. The argument is that it is the consistency of our experience regarding the origin of artificially encoded information that compels us to accept an intelligent origin of apparently naturally occurring encoded information.

The error seems too obvious to be overlooked, too often emphasized by Meyer to be accidental, and, frankly, too flagrant to be wholly innocent. Again, perhaps I am misunderstanding his argument in some way which will be immediately evident when it’s explained to me.

God of the Gaps?

Meyer doesn’t like this phrase, and I can understand why. We humans have a long tradition of invoking deities to fill the gaps in our understanding of the material universe. We have probably done it since our earliest moments of awareness – indeed, the utility of having that comforting and ready answer might, one can easily believe, be why we are inclined to believe in the supernatural.

Meyer has written a book that could have been written at any time during our long quest for understanding. The details would change, the sophistication would vary, but the product would be similar: a man standing on the edge of the unknown surveys the wisest men around him and concludes that, since they have no wholly satisfactory answers, one or another god is the most plausible explanation.

Though Meyer objects to the phrase and argues that he is not engaging in an argument from ignorance, here is how he describes, in Chapter 20, the argument he is making:

Premise One: Despite a thorough search, no materialistic causes have been discovered with the power to produce large amounts of specified information required to produce the first cell.

Premise Two: Intelligent causes have demonstrated the power to produce large amounts of specified information.

Premise Three: Intelligent design constitutes the best, most causally adequate explanation for the origin of the specified information in the cell.

Take a look at that Premise One: “Despite a thorough search….”

What does “thorough” mean, in this context? How does “Despite a thorough search” differ from “Thus far?” What aspect of knowledge does Meyer believe we have exhausted, in our thorough but failed search?

There will always be things we haven’t yet figured out. There will always be a precipice, beyond which is something mysterious and seemingly impenetrable. And there will always be those who stand on the edge and give up on the process and think, I guess God did it.

Of course, they could be right. But they don’t have a very good track record, and I think both science and religion suffer when people engage in this kind of end-run around the humble scientific method, or try to co-opt it to make a theological point. (Similarly, science suffers when scientists try to impugn God with their science. But that’s the mirror image of what’s going on here, and a topic for another day.)

The universe is unimaginably vast, at least 90 billion light-years across and perhaps orders of magnitude larger. It contains perhaps trillions of galaxies, each containing hundreds of billions of stars. It is old, several times older than our own sun, and is full of mysteries.

A hundred years ago our best and brightest argued over whether our galaxy was the only one, and whether the universe was older than we now know our own planet to be. We knew a lot – and, it turns out, we knew almost nothing.

Our best understanding, currently represented by quantum field theory, is bizarre and wildly unintuitive – and yet has wonderful predictive power and astounding mathematical rigor: in some ways, it is the most comprehensive and successful scientific theory ever devised.

It’s too early to throw in the towel.