In grad school I wound up hanging around with John Pepper (yeah, Dr. Pepper) a good bit. I think I disagreed with him more than I agreed with him, sometimes to the point of exasperation, but conversations with him were never boring.

One of John’s most annoying refrains was “is it an organism?” I was studying (and still study) a group of algae for which this question can be genuinely confusing. Most people would say a Chlamydomonas cell is a single-celled organism, and most would agree that Volvox is a multicellular organism, but what about the four-celled species Tetrabaena? A four-celled organism or a collection of four single-celled organisms? What about an undifferentiated colony of 32 cells, such as Eudorina? Or Pleodorina, which is around the same size but with two cell types? Somewhere between a unicellular ancestor and Volvox, a new kind of individual emerged. Among the extant species*, where do we draw the line between organisms and groups of organisms, or can we (or should we) draw a line at all?

My usual response was “It doesn’t matter.” Cells exist; some of them are connected to other cells. They live, some of them reproduce, and they die. All of this goes on without the slightest regard for what we call them. Give me, I challenged John, one biological question the answer to which depends on how we define organisms. He came up with a few, and we decided to write a paper about them. Neither one of us had much time to spare, so we agreed to devote a weekend to it. We spent both days writing on John’s back porch, stopping only for lunch, and by the end of it we had a reasonable rough draft of what would eventually become our Biological Reviews paper.

The literature review for that paper was my first real exposure to a very busy niche within the philosophy of biology: the discussion of how best to define a biological individual (or organism). If everything were a mammal, we probably wouldn’t worry too much about this. Individual mammals are ‘paradigm individuals‘: each is functionally integrated, physiologically autonomous, (usually) genetically unique, and (mostly) genetically homogeneous. But most living things aren’t mammals, and most of them don’t have all of these properties, at least not in the same units. Because of asexual reproduction, the physiologically autonomous units of bacteria, many protists, algae, aphids and dandelions are not genetically unique. Cellular slime molds, myxobacteria, and some ciliates form multicellular units out of genetically distinct cells. Colonial invertebrates, eusocial insects, and human societies are functionally integrated across the physiologically autonomous units.

Dozens of definitions have been proposed that prioritize one or more of these and other properties (Clarke 2010 has a pretty good summary). Recent proposals, such as those of Folse & Roughgarden, Godfrey-Smith, and Queller & Strassmann, tend to focus on properties relevant to evolution. A recent paper by Beckett Sterner compares two such concepts of individuality, those of Peter Godfrey-Smith and Ellen Clarke.

From the start, Sterner sets up Godfrey-Smith’s account as pluralistic and Clarke’s as monistic. I think this is a fair characterization, as Clarke’s account is in some ways a more extreme distillation of Godfrey-Smith’s. Clarke even claims to “…settle the matter once and for all,” at least for plants. Clarke’s views have evolved over time, but in their most recent iteration, individuality boils down to the possession of ‘individuating mechanisms’: traits that either increase the capacity for among-unit selection or decrease the capacity for within-unit selection. The capacity to undergo selection is governed by Lewontin’s criteria of phenotypic variation, differential fitness, and heritability of fitness.

Godfrey-Smith’s view focuses on the same criteria, but he considers these sufficient only for a minimal degree of individuality. ‘Paradigm’ individuals also possess (or belong to populations that possess) traits that allow them to undergo complex adaptations. These criteria play out in different ways for different kinds of reproducers. For multicellular organisms (‘collective reproducers’), the criteria for individuality are realized through the mechanisms of a single-cell bottleneck, separation of reproductive and somatic functions, and functional integration.

Both views are explicitly evolutionary, and both explicitly acknowledge that individuality is a matter of degree rather than a dichotomy. Next time, I’ll discuss Beckett Sterner’s comparison of Clarke’s and Godfrey-Smith’s views.

*We tend to think of these as (at least) analogues of ancestral intermediates, but volvocine evolution didn’t proceed from Chlamydomonas reinhardtii –> Gonium pectorale –> Pandorina morum –> Eudorina elegans –> Pleodorina californica –> Volvox carteri. Extant species, by (cladistic) definitions, can’t be ancestral to other extant species. We tend to assume that the ancestors of Volvox passed through a Gonium-like stage, a Pandorina-like stage, and so on, but these are inferences, not observations.