The Problem of Defining Life

The question "what is life?" seems at first to be a strange one with clear answers. You typically know it when you see it, and everyday examples are easy to list: people, dogs, grass, trees, ants, and so on. All of those things appear to have something special that a rock or a screwdriver does not. However, it takes more than a few examples and a vague special something to form a rigorous universal definition for life. In practice, we have many working definitions of life which define it in terms of its properties and behavior, such as whether something undergoes Darwinian evolution or is capable of self-replication. For most of these definitions however, there exist counterexamples, like fire, which has the property of self-replication, but is not generally considered to be alive. At present, we still do not have a good, satisfying definition of what constitutes life.

The following discussion quickly descends into complex and deeply philosophical arguments if allowed to do so, but why have such a discussion in the first place? It turns out that several recently developed branches of science depend on having a definitive, practical answer to this question. Scientists who study the origins of life on Earth suspect that life arose gradually from a slow evolution of geochemical processes into ever more complex living systems. Without an ironclad definition of what constitutes life, deciding where in such a procession to draw the line between geochemistry and biochemistry is impossible. In the field of astrobiology, these problems are compounded, since we must decide on a universal definition for life using a sample size of exactly one known life-bearing planet. Any such definition will certainly need to be revised if humanity ever discovers a type of extraterrestrial life, since we likely cannot imagine every alternate form of life that may exist in the universe.

A Possible False Dichotomy

In logic, a false dichotomy is a fallacy where a situation is presented as having only two extreme options, when in reality other intermediate options exist. For example, the old expression "if you aren’t part of the solution, you’re part of the problem" pigeonholes people into one of two polar categories, leaving no room for bystanders and creating a false dichotomy. Likewise, classifying something as being either alive or not alive also presents dichotomy of a sort: a cat is alive, a bicycle is not, a virus is… let’s come back to that one. When considering the geochemical origins of life on Earth, it becomes evident that this dichotomy may also be a false one. Could it be possible that there exist intermediates on this scale, which are neither fully alive nor fully inanimate?

The synthesis of life is thought to have begun with inorganic molecules and eventually worked its way up to basic single-celled organisms, but is there actually a sharp line somewhere in the middle that divides life from non-life? It is possible that there exist intermediate steps which were somehow both more alive than a CO2 molecule, and yet less so than an early prokaryote? If the origin of life truly was so gradual a sequence of events, then the existence of such a gray area between the two extremes feels reasonable enough, not to mention that it would explain the difficulty of defining life. Controversial entities like viruses could have a home somewhere in the middle of the spectrum, since they have unique properties and behavior that make their life status difficult to agree upon. These considerations help us to realize that the constructions of life and non-life may not be as rigid as we make them out to be, and they may in fact be purely human constructions imposed on a universe with no such true categorization.

Viruses of a Different Kind

Another fascinating inhabitant of this gray area between life and non-life may be the modern computer virus. Many scientists, including the well-known astrophysicist Stephen Hawking, have argued the case for considering these malicious pieces of software as being alive. They display features like replication and evolution that we would normally associate with life, despite being entirely non-biological and entirely dependent on a parasitism. In many ways, they resemble biological viruses, which also display Darwinian evolution and reproduction, but depend on host cells for their reproduction and have no metabolism of their own. Are these few behaviors sufficient to conclude that a computer program can really be alive, or is there something important that is missing? Perhaps humankind has already managed to accidentally create a new form of life with little to no associated fanfare.

But something about that idea seems out of place, and many would say that life needs to have a physical, biological component. A computer virus may constitute life in some extremely abstract sense of the word, but it seems to fail to do so in a satisfying manner. Most people have a sense of life as being, for lack of a better word, squishy. It needs to be made of chemistry and a computer virus, despite its many analogous behaviors, seems like a cheap simulation at best. This is one place where the idea of a gray area can be important to consider. It allows us to fairly acknowledge that computer viruses have a few tempting behaviors similar to life without counting as full-fledged life.

Life as a Natural Kind

The gray area that becomes apparent in defining life could also be indicative of a larger issue. In science, we work to understand the universe we inhabit in a way that is as unbiased as possible. The push to define life stems from the human observation that there seem to be roughly two types of things on Earth: life and non-life. However, it is also possible that this observation is actually an illusion, and the idea of a gray area of life could be hinting towards this end. Perhaps the distinction between bits of matter being alive or inanimate is purely a human construction, and life is not what one might call a natural kind. The universe may in fact not distinguish between the two in any meaningful way, in which case the scientific search for a definition of life may be doomed to fail or to reach an unsatisfying, arbitrary conclusion. As with many endeavors in science, the hope is that this is not the case and that our search will be fruitful, but it is equally as important to acknowledge with honesty that the universe simply might not work that way.

Views: 172


You need to be a member of SAGANet to add comments!

Join SAGANet

Comment by Kelly C. Smith on May 2, 2016 at 10:20am

I really like the general approach where you try to suggest that perhaps life is not an either-or, zero sum category.  This is basically right, though I would add that this does not require us to believe that there aren't some important discontinuities between what we typically consider purely physical systems and ones we consider alive (see Sorites paradoxes).  This connects to your final thoughts on natural kinds.  We tend to think of these as clearly delineated categories like gold or water (though one can debate the boundaries here more than would make most scientists comfortable).  But maybe some natural kinds are just not like this.  What would a natural kind be if it were produced by an inherently stochastic process driven by variation (e.g., evolution)?  Answer: probably more like an attractor in possibility space than a category with sharp boundaries.  Philosophers will want to ask if it exists independent of human observers in order to say it's objective, but I prefer to ask this question:  would alien biologists, if they exist, probably also have a similar concept of life?  If so, then life is at least intersubjective in a critical way.  And I can't see the argument that they wouldn't since they would be produced by the same kinds of evoluitonary forces that produced us, with many of the same cognitive biases.

Comment by Alicia Gonzalez on April 28, 2016 at 12:59pm

This article is a great introduction to the philosophy of life!

The difficulty of creating a definition for life that satisfies scientists from all disciplines is explained in a rigorous but easy to understand way. This explanation is powerful because it touches several of the numerous approaches to the study of life: biology, chemistry, geology, astronomy and philosophy, including examples that everyone can understand.

The possibility of having intermediate steps between living and non-living systems is a really important concept that is not discussed as often as it should, perhaps due to our limitations when it comes to sample size. Including the example of virus, which many scientists don't think are alive, perfectly exemplifies why this question should be asked! 
I think this article would have been a great introduction for our Origins of Life class, it would have been a great "conversation starter" for the first day of class!

Comment by Cole Mathis on April 28, 2016 at 11:57am

A lot of deep thinkers have pondered this question. The great mathematician von Neumann actually came to a very different conclusion. He thought there should be a clear line separating the living world from the non-living world. His arguments were convincing to some but they still have not be tested directly. I think a lot of astrobiologists tend to agree with your stance though. The data we have doesn’t give us any reason to think that “living” is a well defined concept in the natural world. It’s easy to draw a box around a proton and say that it is a single unit, even though we know it’s composed of quarks. Drawing a box around a cell doesn’t give you the same kind of predictive power or certainty, it seems like life might be more sensitive to our observations than other ‘natural kinds. ’ You might want to check out some of the work by Addy Pross or more recently Jeremy England. They both have been trying to extend the framework of Darwinian evolution into the inanimate world. If they succeed it might prove your point, that the most general feature of biology is just a general feature of the natural world. 

© 2018   Blue Marble Space, a non-profit organization committed to science and science outreach.   Powered by

Badges  |  Report an Issue  |  Terms of Service