1. Cosmology Must Include Astrobiology

Any thorough cosmology (i.e. a logical account of what we may find when we explore the universe) must include biology. At the simplest level, any explanation for the universe would be incomplete if it failed to reference the intelligence doing all this explaining (Joseph 2009; Stapp 2009). For example, if the physical world is created by the process of observing, measuring and knowing (Heisenberg, 1955) then the consciousness that generates cosmology affects the universe that it explains. Taking another approach, we may argue that biology is an important component of cosmology simply because life has been present on Earth for around one third of the existence of the universe (Mojzsis, 2005; Nemchin et al. 2008; O'Neil et al. 2008). But life’s cosmic significance extends far beyond these attributes of sentience and long-term persistence.

A thorough cosmology must incorporate an explanation of where everything came from (i.e. cosmogeny) and through this lens, biology has a deep and complex relevance to our understanding of the universe (Nitschke and Russell 2010). For example, scientists working at the interface of planetary and life sciences are finding that the one example of life we know has extensively reshaped the planet on which it occurs: it has transformed our atmosphere (e.g. see Joseph 2010; Kasting, 2001), it is responsible for the existence of more than half of the mineral species around us (Hazen et al., 2008). Biology has pulled our entire planet’s entropic equilibrium far away from those of our lifeless planetary neighbors (Lovelock, 1965). Life is an intrinsic feature of this planet. In other words, "Earth is no more a planet-sized chunk of rock inhabited with life than your body is a skeleton infested with cells" (Margulis and Sagan, 2000). More flippantly, we can say that Shakespeare was wrong to characterize life as "a poor player, that struts and frets his hour upon the stage and then is heard no more." Since life transforms the planet, it would be more accurate to say that life re-sets the stage as it acts, leaving behind a profound legacy by redesigning the environment to be inherited by future life. If life on our planet is a representative example of something broader, then biology is reshaping the cosmos. Thus a comprehensive cosmology must take account of how commonly life occurs and how predictably it influences the environments in which it resides.

The challenge of placing biology within a truly universal context (i.e. the study of the origin(s), evolution, distribution and future of life in the universe) is one that falls under the nascent, interdisciplinary science of astrobiology. Astrobiology exceeds the boundaries of traditional evolutionary biology which are limited, by definition, to describing the process of change in genetic material that connects today’s species back to the earliest life on our planet. Astrobiology seeks to connect this evolution into a bigger picture that extends back through life’s origin(s) and beyond, into chemistry and physics (desMarais 2008). Only from this big picture can we understand biology’s full impact upon the universe, its true relevance to cosmology. One day the questions of astrobiology may be answered empirically, through the results of space exploration (e.g. see Chela-Flores, 2007), but for the immediate future they require ingenious induction and inference from the one example of life that we know into the complementary insights of earth and space science.

2. Life’s "Fit" to the Physical Universe is Not Understood by Science

If life reshapes planets, then life’s impact upon the cosmos is tied to its ubiquity and thus -- barring a truly universal Panspermia (Gibson and Wickramasinghe, 2010; Joseph and Schild 2010) -- life’s cosmic significance is linked to the frequency with which it originates. This amounts to a question about the ease with which physics flows, via chemistry, into biology: the closer and more inescapable these connections, the more life-friendly (biocentric) our universe.

One early attempt to address this question concluded that "the fitness of the [physical] environment far precedes the existence of the living organisms" (Henderson, 1913). In other words, the physics of our universe appears finely-tuned to produce life. An intervening century of subsequent exploration for this important idea has occurred mostly within the physical sciences (e.g. see McMullin, 2008). In particular, proponents of the Anthropic Principle (reviewed in Barrow, 1986) have focused on the observation that due to the interaction of a small set of fundamental physical constants, our universe exhibits such life-friendly attributes as diverse chemical elements (including abundant carbon and water), and their accretion into stars that bathe their orbiting planets with electromagnetic radiation. Specifically, interest surrounds the observation that these constants appear to be independent of one another, yet each must take a value within a narrow range in order to define a universe in which life can exist. Thus, given that each of the constants does indeed fall within the appropriate range (such that we do indeed exist), some researchers have been led to the view: "The laws that characterize our actual universe, as opposed to an infinite number of alternative possible universes, seem almost contrived … so that life and consciousness may emerge" (Davies, 1995). If this is correct, then a sizeable fraction of the universe may be co-evolving cosmochemically, astrochemically, geochemically and biochemically.

However, many remain skeptical. For example, some assert that we observe a finely-tuned match between physics and biology only because our universe is part of a larger multiverse in which many different versions of physics co-exist (see Journal of Cosmology, 2010, Volume 4 for recent progress in this idea). Accordingly, our life may simply exist within the particular slice of physics particularly well-suited to contain us (e.g. see Coleman, 1988) and estimating the total cosmic frequency of life would require extensive knowledge of physical realities beyond that in which we find ourselves. Such understanding seems like a distant goal, particularly when placed alongside other criticisms of the Anthropic Principle that simply deny any fine-tuning exists within the physics we know. Specifically, there are those who maintain that from what we can perceive of the cosmos "the final laws of nature, the book of rules that govern all natural phenomena … are utterly impersonal and quite without any special role for life" (Weinberg, 1999). While further research within physics (e.g. Livio 2008) and chemistry (e.g. Barrow 2008; Lane 2010; Russell and Kanik 2010) can increase the scientific content of this particular debate, comparing these two critiques of the biocentric universe suggests a deeper underlying problem: that physics alone cannot measure the relationship between physics and biology. A clue to why this should be so lies in the very phrase "Anthropic Principle," which refers to the existence of humanity but was coined by physicists to discuss the ease of abiogenesis. To biologists, an enormous gap separates the emergence of life from the emergence of our sentient species - and this gap is filled by biological evolution. The potential importance of evolution for understanding life’s place within the cosmos is illustrated by yet another critique of the Anthropic Principle: that fundamental biology really does match closely the precise "settings" of the physical universe, but this fit is uninformative because life would have evolved itself to match any physics (Erwin, 2003).

This view usually comes from evolutionary biologists who are underwhelmed to hear that fundamental properties of biology match the physics of our universe, given experience in studying the exquisite adaptations by which organisms on Earth have evolved to fit their diverse environments. Now this assertion (that perceptions of fine-tuning within physics are a mere post-hoc fallacy) may oversimplify the findings of physics as much as any careless invocation of “anthropocentrism” oversimplifies the findings of biology. No clear empirical evidence demonstrates that life could evolve to fit a universe without carbon or water any more than it supports the assertion that this chemistry will produce sentient beings (though see Morris, 2008, Naganuma, and Sekine, 2010; Rampelotto, 2010; Schulze-Makuch 2010, for interesting ideas on each of these points). But taken together these ideas illustrate the true challenge for understanding the role and impact of abiogenesis within the cosmos as a whole: we need to better understand what life can look like.

Certainly, the past few decades have revolutionized biologists’ concepts of what constitutes a habitable environment (e.g. Pikuta et al. 2007) and one of NASA’s major goals for astrobiology is to ascertain the (as yet unknown) limits on the range of conditions to which life on our own planet has adapted itself (des Marais, 2008).

Meanwhile, exciting developments in abiogenesis are showing that life on Earth may have emerged with ease, perhaps even as an inescapable solution (necessity) to easing a tension of thermodynamics and entropy (e.g. Lane 2010; Russell and Kanik 2010; Nitschke and Russell, 2010). But in between this information about why life evolves, and the insights presented in this volume and elsewhere as to how the origin of life took place, a third and equally important challenge is to understand the relative contributions of chance versus necessity (or at least predictability) in shaping what properties we then expect to emerge through evolution. For example, even if we accept at face value that life will appear wherever rocky planets host groundwater containing carbon dioxide (Russell and Kanik, 2010), forced into existence to resolve the entropic disequilibrium (Nitschke and Russell, 2010), we remain far from knowing whether an independent origin of life would evolve into something similar to the particular version of biology that we know so well from our own planet.

3. What Features of Life Offer Clues to Chance Versus Predictability In Its Origin?

At first sight, it may seem strange to talk about "the particular life" of our planet: an obvious characteristic of biology to the casual observer is its impressive diversity. Some have gone so far as to assert that simply comparing organisms as different as "algae, beetles, sponges, jellyfish, snakes, condors, and giant sequoias" leaves the observer "hard-pressed to believe that they all came from the same universe, much less the same planet" (deGrasse Tyson, 2003). But since the middle of the 20th century, life scientists have come to see that this appearance is deceptive. Beneath the surface, all life on Earth shares a remarkably simple and unvarying nature. To be specific, the discovery of an inseparably elegant structure and function for DNA paved the way for Crick’s (1958) declaration of a "central dogma" to life’s molecular basis: information stored as genetic material (genotype) carries the complete and precise instructions for building all the components of an active metabolism (phenotype). This metabolism interacts with the environment to extract raw materials, energy and perform all functions necessary for successfully copying the genetic material into a new generation.

The simplicity and uniformity of the underlying biochemistry is notable: between the origin of life and the time of the “Last Universal Common Ancestor” for all of today’s biodiversity, a handful of biological constants evolved. For example, all organisms came to use the nucleic acid DNA to safely store genetic information in the form of a linear sequence of chemical letters. All came to spell out these genetic messages using an alphabet of exactly 4 such chemical letters (each letter comprises a ribose sugar joined to a phosphate and either a purine or a pyrimidine base). All came to use the similar nucleic acid, RNA, to transmit this genetic information for processing (by information-reading molecular machines) into networks of interacting proteins that constitute metabolism. All came to construct individual proteins by linking together amino acids drawn from an alphabet of exactly 20 different options. In order to transform genetic information into metabolism, all came to use an identical genetic code (the system of rules by which “letters” of a nucleic acid gene are translated into corresponding “letters” of a protein, just as the dots and dashes of Morse Code are translated into the letters of English). This fundamental biochemistry, this central dogma of molecular biology, became the framework for billions of years of subsequent evolution, from the emergence of E. coli to that of H. sapiens.

Researchers have now discovered some instances where these universal, biochemical constants have subsequently loosened a little. For example, some viruses have re-evolved an RNA genome, a disparate handful of lineages has evolved minor changes to the rules of their genetic code, and some have even added one of two amino acids to their protein alphabet. However, these observations merely emphasize that fundamental biochemistry is an evolved and evolvable phenomenon. This in turn highlights the question: which of these evolutionary moves were predictable and inevitable (i.e. might be expected to occur elsewhere), and which are part of a unique and idiosyncratic trajectory for Earth-biology?

Understanding whether the emergence of life on our planet represents a haphazard collection of accidents or the rapid evolution of predictable characteristics that we might anticipate elsewhere rivals the importance of physics' fundamental constants when it comes to interpreting our place within the cosmos.

4. Replacing Reasonable Debate with Science

Just as for the broader topic of a life-friendly universe, scientists have offered no shortage of confidently diverse views on the related question of what, if anything, is predictable about biological evolution. Some argue that "man knows at last that he is alone in the universe's unfeeling immensity, out of which he emerged only by chance" (Monod, 1972); or that "any replay … would lead evolution down a pathway radically different from the road actually taken" (Gould, 1989). Others assert more optimistic assessments: that "If the history of life on Earth could be rewound and replayed, many of the same innovations would reappear" (Than, 2006); or that "far from its myriad of products being fortuitous and accidental, evolution is remarkably predictable" (Conway Morris, 2010). When the question is reduced to the predictability of Earth’s fundamental biochemistry, optimism seems to predominate: we read that "the building blocks of life anywhere will be similar to our own" (Chela-Flores, 2007). These assertions can be applied generally, e.g. the only "Differences in evolutionary systems likely will lie at the higher-order levels" (Pace, 2001), or specifically, e.g. "If life were to arise on another planet, we would expect that … about 75% of the amino acids would be the same as on the earth" (Weber and Miller, 1981).

However, every quote presented in this essay up until this point has been carefully selected to illustrate an opinion of a scientist rather than science. In stating that, we do not mean to assert that these opinions are wrong, or even that they lack good reasoning. Indeed most represent the interpretation of an expert, offered within a formal or informal review of scientific data, and often consciously presented as an opinion). However, these assertions exist outside the framework of what constitutes science precisely because they present no clear test of what they assert. To be sure, some qualify as scientific hypotheses in the sense that they make explicit predictions (Weber and Miller, 1981; Chela-Flores, 2007) but until these tests are performed, our view is they may not be considered as fully-fledged science.

Previously and elsewhere we have used a rather less well-informed opinion to point out the problems that can emerge from muddling rational opinions with science (Freeland, 2010). Here, our message is more positive: each detail of life’s fundamental biochemistry suggests one or more adaptive hypotheses that are worthy of scientific investigation. Diverse opinions about the universality of biochemistry can be gathered and assessed objectively by simple application of appropriate scientific tests to yield a robust picture of the relative roles of chance and predictability in shaping life on Earth.

5. Current Scientific Knowledge for Chance Versus Necessity in Establishing Biochemistry

When considering life’s fundamental biochemistry, the key phenomena are of interest precisely because they evolved just once, early on. One possible explanation is that they represent an optimal solution to one or more fundamental challenges of life (replication, metabolism, adaptive evolution etc.), and thus we might expect them to occur wherever life emerges. However, the persistence through billions of years of evolution of a trait such as “a DNA genome using 4 nucleotides” is an unreliable indicator of its inevitability because characteristics that evolve early can become "locked in" by the emergence of new innovations that make adaptive sense only in relation to them.

However, Optimality Theory (Maynard Smith, 1978) is a well-established evolutionary research tool suitable for investigating just such situations. A researcher defines boundaries on what alternative outcomes were plausible, and then asks whether the particular outcome that became reality is one that distinguishes itself by virtue of being especially advantageous (i.e. optimal) such that it represents a predictable outcome of natural selection. The challenge here is to identify objectively measurable criteria that define the evolutionary fitness (value) of each possible outcome.

This approach has been widely used to study evolutionary outcomes that range from animal behavior (Charnov, 1976) to genome structure (Xia, 1995), and many such investigations have tackled aspects of life’s biochemistry. For example, when it comes to the nucleic acids in which genetic information is stored, we know that many nonbiological nucleotides are possible (Piccirilli et al., 1990; Bergstrom et al., 1997; Delaney et al., 2003). We also know that many are used outside of genetics within the metabolism of today’s biosphere (e.g. see Grosjean and Benne, 1998). From here, several focused optimality studies have offered detailed explanations for the size and composition of the genetic alphabet used by life (Gardner et al., 2003; Mukhopadhyay et al., 2003; Szathmary, 2003). Even the choice of ribose as the sugar component of nucleic acid has been proposed to be optimal for base-pairing strength relative to other prebiotically plausible alternatives (Egli et al., 2006). Assuming that some sort of nucleic acid available, then the details of life’s genetic biochemistry appear to be more or less predictable. Yet this assumption has a problem. The very existence of nucleic acids within early biochemistry remains one of the larger mysteries of life’s origin (e.g. see Shapiro, 2007).

Certainly, nucleic acid presents some highly attractive features as both genetic and catalytic material (Szathmary and Maynard Smith, 1995), and if life got under way using a less complex genetic system, then it is easier to understand how the "useful" properties of nucleotides could have been selected for (Bartel and Unrau, 1999; Eschenmoser, 1999; Pace, 2001; Gardner, et al., 2003; Szathmary, 2003). Indeed, there has been no shortage of suggestions as to possible forerunners of RNA. These include peptide-nucleic acid (Nelson et al., 2000), triose nucleic acid (Schoning et al., 2000), polyglyceric acid (Weber, 1989), and even inorganic crystals such as clay (Cairns-Smith, 1982). But so far, no one has proposed an unambiguous test to discriminate which of these suggestions is most relevant, and how they fit with broader insights into abiogenesis. Thus the current frontier for such investigations seems to be understanding exactly how nucleic acids entered early evolution. In effect, we are blocked from executing a broader optimality study because no clear model for the appropriate "possibility space" of alternatives is widely accepted. Indeed, even the simple suggestion that perhaps the primitive genetic system comprised fewer than four nucleotides has produced such varied and inconsistent ideas that the different hypotheses combine to undermine one another’s credibility (see Knight et al., 2004).

Meanwhile, when it comes to the amino acids, the situation described for nucleotides is almost precisely reversed: the preponderance of scientific effort thus far has gone to defining which ones were available to the origin of life, both on our planet and elsewhere in the galaxy (Martins and Sephton, 2009; Cleaves, 2010), and to exploring the historical process by which some of them became incorporated into biology (Higgs and Pudritz, 2009). All this work clearly indicates that biology has chosen only one small subset of what was chemically and biochemically possible (Lu and Freeland, 2006), but what has been largely lacking to date is an appropriate use of optimality theory to ask whether this sub-set shows any clear signs of being a predictable (high fitness) solution that we might expect in an independent abiogenesis. The first detailed attempt at such an optimality study asked whether life’s choice of amino acids showed unusually high statistical variance in size, charge and hydrophobicity: the three important properties that connect amino acids to the functional roles that they play within proteins (Lu and Freeland, 2008). Complex results brought us to the conclusion that "the question of whether early life selected a non-randomly diverse alphabet of amino acids remains far from clear … It is possible that this is because there is no clear pattern to be found … but … our purpose here is not so much to make a strong claim for or against the role of natural selection in choosing an amino acid alphabet; rather, it is to highlight the potential and demonstrate the plausibility of using a quantitative framework to think about the origin of proteinaceous amino acids". In other words, the point of replacing reasoned arguments with science is not that the reasoning of experts is inherently bad, or that science gives better answers immediately. Rather it is that science provides a clear and objective process for progressing knowledge. The repeatability of a given test allows a new round of research to choose exactly which assumption(s) to vary in order to see their effect on outcomes and interpretation. To take the specific example of the amino acids, one point of departure would be to question whether a better measurement of amino acid functional diversity would provide a different picture?

Meanwhile, other aspects of the amino acid alphabet remain almost entirely unexplored. For example the only attempts to address the related issue of whether the amino acid alphabet size of 20 is in some way optimal have been tucked in as afterthoughts within publications pursuing a fundamentally different theme (Wong, 1976; Szathmary, 1991). Here, as above, the challenge is to think through a logically defensible fitness function. One interesting insight is that the speed and efficiency of evolution is influenced by the pattern in which genetic code-words are assigned to the 20 standard amino acids (Zhu and Freeland, 2006). It thus seems likely that altering the size of the amino acid alphabet would have a similar or stronger effect: perhaps here we might find that the size of the amino acid alphabet complements that of the nucleotide alphabet as some sort of optimal match? Neither of these tests is likely to provide a final answer to the universality of life’s choice of amino acids, but by their very nature, both tests would sit in defined relationship to previous studies, sketching a coherent picture to the question of what we might expect from an independent origin for life, and thus what we might expect to find as a role for biology within the cosmos.

Acknowledgments: This material is based upon work supported by the National Aeronautics and Space Administration through the NASA Astrobiology Institute under Cooperative Agreement No. NNA09DA77A issued through the Office of Space Science.