Thursday, March 21, 2013

Species (introduction)

We encounter species all the time when reading about biology. Darwin titled his most important work The Origin of Species. Articles about groups of plants or animals boast how many species they contain compared to other groups at the same rank of a classification ("this is the largest order of insects", etc.). Likewise, you can find information on how many species occur in this country as opposed to that. Some species are considered threatened and protected by law, and conservationists design species recovery plans to protect species from extinction. Evolutionary biologists calculate speciation and extinction rates for different groups of organisms, i.e. rates at which species appeared and went extinct in evolutionary history.

Given all that, it may come as a surprise that a good number of biologists would not consider species to actually, you know, exist.

Some time ago a colleague asked that question in our journal club over at the university: everybody here, do you think that species are real? Meaning, are species entities that actually exist out there in nature and that we can reproducibly and objectively discover, or are they merely arbitrary human constructs? And quite a few colleagues answered no, species don't have any (special) reality. My own reaction was to stall. I think it depends on the group of organisms you are talking about, was what I said.

That was then, and since then I have given the topic a lot more thought, which is why I want to write about it in the next few posts (exempting the usual botany pictures and whatever else may come up). I am certainly not an expert on the theory but as a taxonomist I need to deal with species all the time. I also do this in part to order my thoughts in preparation for some teaching I expect to be involved in later in the year.

First of all, as in so many other cases like the perennial favorites "does free will exist?", "does god exist", "is this activity science?" or "can science address the supernatural?", we cannot even begin to answer the question whether species are real without defining the term in question. First one has to define what species might be, then we can answer whether they are, under that definition, a discoverable reality or an arbitrary human construct.

Over the years, biologists have defined species in many different ways, sometimes formally and officially, sometimes only implicitly and indirectly. The various possible definitions are known as species concepts, and there is a rich literature on the subject, with people arguing for this and against that species concept. Here is a very concise list of 26 species concepts from the blog of a philosopher of biology who has dealt with the issue. The list gives a good overview and at the same time a good impression of the magnitude of the problem. Surely the impulse to throw up one's hands and give up is understandable if even professional biologists have not been able to agree on a definition or whether species are real at all.

But we should take even another step back before we give up so easily. Not all of these concepts do actually compete with each other.  For starters, I am not a philosopher and must admit that I do not, on a practical level, see the differences between some of the 26 as distinguished on that list. I try to apply concepts to a mental model of what we can assume to happen in the course of evolution to individuals, populations and clades, and so the attempts to differentiate between a concept where species are nodes in the phylogenetic tree and one where they are internodes in the phylogenetic tree or between the biological species concept and the mate recognition species concept appear a bit twee to me. More importantly, there are concepts that have a historical dimension and apply to phylogenetic trees, and there are others that do not have that dimension and are entirely concerned with delimiting species in the here and now. There are concepts for purely theoretical discussions and concepts for use in empirical taxonomic studies. So what we could do is pick this or that concept (or in some cases a family of similar concepts), and see what they are actually good for, and whether under that concept species are real and discoverable.

But there is even more. Me being a cladist, it is clear that I definitely consider clades to have a biological reality and to be discoverable. Their reality, and what makes them natural groups, is that they contain all descendants of one ancestral species (there is the word again, but more on that another time). They are discoverable by inferring a phylogeny and because the entire clade inherited some characteristics from its common ancestor, the so-called synapomorphies.

So clades have a reality, and they are discoverable, but they are nested within each other: larger clades contain smaller subclades. That means that there is no level of clades that is special: the vertebrates are a clade, the mammals are a clade inside the vertebrate clade, the bats are a clade inside the mammal clade inside the vertebrate clade, and so on all the way up to the clade "all of life" and down to where there are no clades any more because phylogeny turns into tokogeny. But that is not what most people mean with species: the giraffes would commonly be considered a species, but the mammals a clade of many species, and (to most of us) there is a fundamental difference between those two levels. Most importantly, people do not generally assume that there can be species inside of species, as clades can be inside of other clades. Species are generally assumed to be a special level of biological diversity.

That shows that when evaluating species concepts, we need to consider not only whether they provide a reason to think of species as having a biological reality (like clades) but also whether they provide a reason to think of species as having a special kind of reality that higher and lower levels of biological diversity do not have. In other words, we will have to be on the lookout for two different criteria: what is the grouping criterion that decides whether individual organisms belong to the same group? And what is the ranking criterion that decides whether the group of individuals is a species as opposed to another type of group that is not a species (e.g. a population or a supraspecific taxon)?

So, to summarize: over the next few posts, I will look at different species concepts. In each case, we should have an eye on (1) its grouping criterion and its (2) ranking criterion to be able to decide (3) what, if anything, the concept is good for and (4) if, under this concept, species are (a) real biological entities that can be reproducibly and objectively discovered by science and have a special reality different from higher and/or lower levels of biological diversity, or (b) real biological entities that can be reproducibly and objectively discovered by science but do not have a special reality different from higher and/or lower levels of biological diversity, or (c) arbitrary human constructs. Finally, I will most likely end this small series with a few general remarks to summarize my own position on the species problem.


  1. I'm a fish taxonomist/systematist. I look forward to your thoughts. From what I have seen, from a hobby interest, of botanical taxonomy, I expect you to see things somewhat differently than I do. It doesn't bother me that there are many different species concepts, because I think there are different kinds of species; many of them really real.

    I think, when I describe a new species, or make a dot map showing distribution of a species, that I am dealing with a coherent, bounded entity that exists in nature. On the other hand when I erect a genus, I am stating my opinion on how best to understand a relationship. That the first species I described is now in its fourth (and, one hopes, final) genus, may affect my thinking here.

  2. i hope this introduction will give more useful information