In a recent New York Times op-ed entitled “Why Nothing Is Truly Alive,” Scientific American editor Ferris Jabr argued that “life” is a poorly defined, and consequently to some degree ephemeral, concept. Ultimately, he says, “nothing is truly alive.” This is an interesting and provocative claim, which is justified given the lack of scientific consensus among biologists about what kinds of things are alive and what aren’t (viruses being a notoriously thorny example). My claim, however, is that a clear understanding of the results of molecular biology and chemistry over the past century make it possible to give a simple and unambiguous definition of life. To understand this definition, one must first realize that life is a verb. It is not a property that one ascribes to structures per se, but rather to the activity of a structure. Once one makes this simple paradigm shift in thinking, many other apparently mysterious aspects of “matter in the living state” become understandable.
For instance, the list of properties of life that Jabr cites—“organization, growth, reproduction, and evolution”—no longer appear as a disconnected list of independent features, but as essentially correlated aspects of what it means to be a dynamic chemical process. In what follows, I attempt to explain the essential features of our best, current theoretical understanding of life and how they address persistent difficulties in answering the question “What is life?”
A living process is an organism. That means that it is a specific kind of organization of matter that is capable of carrying out a process of growth, responsive action, and (possibly) reproduction. If we understand how each of these processes becomes chemically possible, then we also understand the physical basis of life. There are two ideas from 20th century science that one must appreciate to understand how life works as a chemical organization. The first is that metabolism is circuitry, and the second is that information content is a measure of randomness.
The premises of Jabr’s argument are all essentially correct. He rightly stresses that there is no difference whatsoever between atoms in an organism and atoms in an inert substance. “All observable matter is, at its most fundamental level, an arrangement of atoms and their constituent particles,” says Jabr. “These associations range in complexity from something as simple as, say, a single molecule of water to something as astonishingly intricate as an ant colony.” He continues: “All the proposed features of life—metabolism, reproduction, evolution—are in fact processes that appear at many different regions of this great spectrum of matter.” Ultimately, Jabr concludes that for life, “there is no precise threshold.”
But, in fact, all that follows from Jabr’s premises is that one will not locate the difference between life and non-life by looking only at the atomic composition of something. Instead, we must look at what these atoms do, which is the chemistry of life, called “metabolism.” Metabolism is about the movement of energy through an organism, and its theoretical explanation relies on a very specialized field of theoretical chemistry and physics called non-equilibrium thermodynamics.
Over the last century, non-equilibrium thermodynamics has become a mature field capable of explaining the basic physical forces that would have been necessary to “force” life into existence on the early Earth.1 A fundamental result of this view of life is to invert the customary relationship between life and energy. It is universally recognized that life requires a source of thermodynamic potential and a sink that can receive entropic waste (or in simpler terms, that you have to eat). But in the thermodynamic picture, life is a chemical channel that provides a relaxation pathway for free energy stresses due to unbalanced geochemical cycles. In other words, biochemistry first began to occur not because some proto-organization was trying to “gain energy,” but because the earth was building up intolerable amounts of chemical potential energy, and the chemistry which became life occurred as way to allow energy to flow “downhill,” or alternately to allow the atmosphere to chemically “relax.” The best candidate theory for how this process occurred is termed the “metabolism-first” hypothesis, and it posits that what we call the “reverse citric acid cycle” or “reverse Krebs cycle” emerged spontaneously on mineral surfaces near hydrothermal vents spewing carbon dioxide up from the Earth’s core.2
However it occurred, it is a fact that the citric acid cycle is the hub of what is called “core metabolism,” a network of biochemical reactions that is essentially invariant and universal to every organism on the planet, without exception. It is also true that the citric acid cycle is thermodynamically equivalent to a class of processes termed “dissipative structures,” which include things like lightning bolts, hurricanes, and stars. Each of these phenomena is a channel state that enables a “current” of energy to travel much, much more efficiently from high potential to low potential than would have been possible through random atomic motion. You are familiar with these processes, you just don’t realize that they are the result of a breakdown from energetic stress. Lightning is, of course, electricity; it occurs when the charge difference between the air and the ground gets high enough to ionize the atmosphere and make a plasma channel, which then conducts electricity so as to make the disequilibrium in the distribution of electrical charge go away. A hurricane forms when the temperature difference between one part of the atmosphere and another gets too great, and the swirling air conducts heat from the ground to upper atmosphere. Both heat flow and electricity flows are just flows of energy.
Chemistry is no different, and the chemical cycles which represent the hub of activity inside every one of your cells is simply a patterned flow of electron movements designed to conduct chemical energy from high potential to low potential. The origin of life is a chemical hurricane. Note that this idea doesn’t separate life from the material world; it weds life to its environment in a fundamental and indissociable manner. In short, biochemistry is a manifestation of geochemistry.
The consequences of this outlook are shattering. Everyone realizes life needs energy, but the perspective that geochemistry is an integral catalyst and communal conspirator in life processes is a stark reversal of the notion of a “fortunate chemical accident” that somehow perseveres against immense odds to build some hospitable foothold for itself on an indifferent rock.
More controversially, the thermodynamic concept of life means that patterns of energy flow are the fundamental “units of selection” in the biosphere, which differs from the genetic-centrism that dominates evolutionary thinking today.
Information “Flows,” Not Just Genes
Which brings me to the next part of the definition of the living process: information storage and transmission. The reason that hurricanes, lightning bolts, stars, and other “metabolic” processes are not alive is because the flows of potential energy they help to conduct are not coupled to flows of information, which is the role played by genetic circuits for life. Notice that I said “flows” of information, and genetic “circuits,” not just “genes.” First, some biology everyone should know. Biomolecues are heteropolymers: they are long chains of different kinds of repeating units, like beads of different shapes on a string or written language. The “genetic code” is a set of relationships that allow molecular machines inside cells to use the sequence of chemicals contained in DNA as instructions for making proteins, which in turn do almost all of the chemistry that goes on in your body.
To understand this part of the definition of life, we must first know a little about information theory, specifically “Shannon entropy.” The Shannon entropy is a quantity that measures the number of bits of information required to encode a sequence of symbols. It is a counterintuitive notion of information because it does not reflect the value or use of a message, but only the probability of having randomly arrived at a given sequence by selecting from the set of all possible combinations of symbols. Thus, a completely random sequence, which has no correlations between one part and another, takes the most information to encode per symbol, but a completely redundant sequence (AAAAAA...) can be compressed to the single symbol and the command to repeat it indefinitely. Unsurprisingly, information theoretic analysis suggests that natural proteins are basically “slightly edited” random strings of the chemicals that make them up, called amino acids.3 This, in turn, implies that the DNA sequences that code for these proteins are more or less just random strings of the chemicals A, T, C, and G.
This point is significant for understanding the role of information in living organization. An evolving chemical organization could not have begun with a group of complex molecules highly specified to perform particular chemistry, as exists today, but rather from some random assortment of whatever was around doing whatever it could do. Think about what your muscles feel like, what fruits feel like. Living things exist in this condensed watery-gel phase because in such a crowded, sticky, snotty environment, full of oily chemicals next to watery chemicals, chemistry happens spontaneously and readily. So proteins can be random strings because any random string of amino acids is very likely to be capable of doing some chemistry.
Now return to early Earth and imagine that some chemical cycle, operating for no other reason than to reduce the free energy of the atmosphere, was in the process randomly spewing out ever more complex assortments of random strings of molecules, each of which has some probability of itself participating in a chemical reaction. Eventually, another loop of chemical reactions will lead out of and back into the original cycle, thus extending the metabolic network. Any such loop of reactions will have the concentrations of its members amplified by aiding the flow of energy through the central cycle. This is the thermodynamic rationale behind evolution at the molecular scale, and the explanation for why the structure of genetic control has to be “circuits” that receive inputs from the environment and ”respond” by producing outputs designed to maintain the stability of chemical flux through the system. It is because all of biochemistry is an elaboration of cycles of chemical reactions built around a central cycle, and chemical cycles are circuits.
Electrical engineers are today very familiar with how to design circuit response elements out of feedback loops and amplification cascades. As it turns out, life uses much of the exact same logic, and certain simple “network motifs” involving genetic regulation appear over and over again throughout all of life.4 The origin of responsive action is in the selection of mechanisms that allowed stabilization of energy flux through the earth, which is the fundamental rationale for life’s existence.
DNA is Not the Secret of Life
If you take nothing else from this essay, please know that life is a manifestation of chemistry and the only reason chemistry happens is to reduce the thermodynamic potential of a system. This view helps to overturn the idea that DNA is the “secret of life.” Gene frequencies certainly reflect the operation of selection in a population, but in truth DNA is mostly just random assortments of nucleic acids. Genes do not drive selection, they are not fundamentally the objects being selected, and they most certainly are not the “point” of living organizations.
No structure is alive simply by virtue of being a structure. A structure is alive because of what it does, and I would be perfectly comfortable calling any self-elaborating process that grew, drew inputs from its environment, and participated in any means of seeding the conditions for a similar organizational type to grow, alive. But I would not say that any structure, especially not one in equilibrium with its surroundings, is alive. Certain times, in certain places, under certain conditions, matter comes to life and engages in a living process. This is a sublime fact, and the reverence and moral obligation we feel toward other living things is deserved and real. It is not a mystery how and why matter performs this dance.
1. Harold Morowitz and D. Eric Smith. “Energy Flow and the Organization of Life,” SFI Working Paper, Santa Fe Institute (Aug. 29, 2006): http://www.santafe.edu/media/workingpapers/06-08-029.pdf.
2. I should stress that this is far from a consensus view among biologists, although the experimental evidence in its favor is growing steadily. For a review of this theory and its implications, I highly recommend this hour long talk by Dr. Eric Smith of the Santa Fe institute: http://www.youtube.com/watch?v=ElMqwgkXguw.
3. O. Weiss, MA Jimenez-Montano, and H. Herzel. “Information Content of Protein Sequences,” Journal of Theoretical Biology, Institute for Theoretical Biology, Humboldt University Berlin (Oct. 7, 2000): http://www.ncbi.nlm.nih.gov/pubmed/10988023.
4. Uri Alon Lab. “An Introduction to Systems Biology – Design Principles Of,” Weizmann Institute of Science: http://www.weizmann.ac.il/mcb/UriAlon/introduction-systems-biology-design-principles-biological-circuits.