Over at the New York Times, science writer Ferris Jabr wrestles with the difficulty of differentiating living things from non-living things—viruses can reproduce themselves and evolve, but need host cells to do it; inorganic crystals can grow and (sort of) reproduce. He concludes that although “life” as we know it is a useful concept, it’s just that—a concept: “We find it useful to think of some things as alive and others as inanimate, but this division exists only in our heads.”
From there, Jabr goes on to a conclusion that (judging from my Twitter stream) has made a lot of people very angry and been widely regarded as a bad idea:
Not only is defining life futile, but it is also unnecessary to understanding how living things work. All observable matter is, at its most fundamental level, an arrangement of atoms and their constituent particles. 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. 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. There is no precise threshold.
Some things we regard as inanimate are capable of some of the processes we want to make exclusive to life. And some things we say are alive get along just fine without some of those processes. Yet we have insisted that all matter naturally segregates into two categories — life and nonlife — and have searched in vain for the dividing line.
It’s not there. We must accept that the concept of life sometimes has its pragmatic value for our particular human purposes, but it does not reflect the reality of the universe outside the mind.
I’ve joined in the discussion on Twitter, a little bit, but I felt like I needed more space for my reaction to this thinking, so here goes: The fact that “life” is a human concept doesn’t mean it has no basis in physical reality.
My initial reaction to the whole essay was to think about biologists’ debates over species concepts—the rules by which we decide that one population of critters is a different collective biological entity than another. It’s easy to say that species are populations of “actually or potentially interbreeding individuals,” per the biological species concept—but there are plenty of cases where things that seem to be different species turn out to interbreed all the time, or populations that look very similar turn out to be completely reproductively isolated. Most biologists would, I think, agree that “species” are, indeed, concepts—ultimately just ways for us to ensure that we’re talking about more or less the same critters when we use a particular Linnaean binomial.
But that doesn’t mean that species have no basis in biology! The genetic diversity of living things is not a uniform mush; it’s possible to trace lineages of ancestry that do not intersect going back millions of years, and members of such different lineages generally are, in fact, very different in appearance and in lifestyle.
Similarly, although there are cases where it’s truly difficult to decide whether a particular system is “alive” or not, the concepts of “life” and “non-life” capture real differences which exist independently of human observation and description. Living things generally engage in processes—conversion of unlike matter into copies of themselves, social interaction with other living things, reshaping their environments to make them more useful—that are not accessible, or even comprehensible, to nonliving things. The existence of twilight doesn’t mean that there aren’t real differences between day and night. (Even if, yes, those are all just concepts for our internal human experiences of ambient light levels!)
I’m not necessarily sure that Jabr is holding that there is no such thing as life at all—although his essay seems to get pretty close to saying that, and a lot of folks (including whatever editor wrote the essay’s headline) seem to have taken that to be his meaning. But regardless, I think it’s pretty clear that the fact that reality isn’t divided by sharp, binary boundaries does not mean that binary concepts can’t describe it pretty well—we just have to be willing to allow some margin for error.