Ontology is that branch of philosophy that asks what exists. Traditionally, this has been understood to mean what kinds of things exist in general, but in recent times, it has also been applied to mean what objects a scientific theory requires to actually exist if it is to explain the phenomena. We must therefore ask what things evolutionary theory requires to exist. This is, of course, distinct from the question of what things we can observe or measure, which is a matter of epistemology, not ontology.
In metaphysics, a distinction is sometimes made between types of things, and tokens of the types. A similar and related issue is whether things are classes that can be defined or individuals that can only be described or ostensively defined (i.e., pointed at). Most evolutionary objects have been interpreted to be either types and classes or tokens and individuals. Part of the problem is that since evolution by definition involves a lack of stability in the objects it explains and covers, it is hard to clearly define the types of objects.
Evolution is usually understood as a biological process, but following Richard Dawkins, and with antecedents well before him, attempts have been made to formulate what is called by Dawkins “universal Darwinism.” This is a generalized model of evolution that is independent of the physical substrate—a kind of general theory of any possible evolutionary process that might be called “Darwinian.” Philosopher David Hull has applied it most extensively to the evolution of science, for example. We will consider biological evolution, but with an eye to the generalized ontological implications.
When we ask what “evolutionary theory” requires, of course we must take some representation of it, as that theory is itself evolving over time. The most widely discussed form is, of course, the conception of Richard Dawkins, sometimes called the “selfish-gene perspective,” after his seminal book, or the “received view.” Out of this and later works, a distinction was refined by David Hull, which has become known as the “Hull-Dawkins distinction,” between replicators (genes and anything that is copied accurately like genes, including cultural items, called memes) and the economic systems they are part of and are reproduced when they are copied, which Hull calls interactors (organisms and anything like them that are differentially successful at getting the resources needed to replicate), although Dawkins prefers a less voluntaristic term, vehicles. Replicators are defined in Dawkins’s The Selfish Gene as being those entities that are fecund (make more of themselves), faithful (have a high fidelity of replication), and are long-lived (persist over evolutionary time frames). Organisms do not persist over intergenerational time frames and so are not significant in evolutionary terms. Replicators are divided into “dead-end” replicators (for example, somatic genes), and germ line replicators (gamete genes).
The received view has been opposed or modified by critics, particularly Ernst Mayr (1988), who asserts that the objects subject to selection are organisms, not genes, which are “hidden” from evolution. Another critical stream is called “developmental systems theory,” and its adherents hold that the evolutionary objects subjected to selection are developmental systems and life cycles. It is worth noting that Dawkins’s view is a reworking and development of George C. Williams’s notion of an evolutionary gene, which is, as he called it, a “cybernetic abstraction,” defined as “any inherited information for which there is a favorable or unfavorable selection bias equal to several or many times its rate of endogenous change.” A primary feature of evolutionary theory is that what is hereditable, whether they are replicators or not, must have variant forms (in genetics, alleles) that can be in competition for resources in order for selection to occur.
Sewall Wright suggested a metaphor of an adaptive landscape, which is the fitness value of each possible allele of each locus in the genome. This is sometimes taken literally, as a description of the values of genes in an environment. The adaptive landscape is itself constituted by the states of other organisms and their genes, as well as by the physical environment in which all these organisms exist. The sum total of these features is called the ecosystem, which is arranged into classes called ecotypes. A particular role in an ecotype is referred to as a niche. These properties of evolving things are primarily economic; they refer to the physical resources required by interactors to persist until reproduction. William Wimsatt has influentially defined reproducers as the key object in evolution viewed from an economic aspect, instead of replicators. This has given rise to the distinction made by Stanley Salthe and Niles Eldredge between the genealogical hierarchy and the economic hierarchy. Genes apply only to the genealogical hierarchy.
A major set of disputes has arisen over the phyletic (that is, the evolutionary-tree level) entities in evolution. These include questions about the ontological standing of species and other taxa, whether lower than species or higher. Mayr asserts the ontological primacy in evolution of the population, which is a group of interbreeding organisms, or a Mendelian gene pool. Populations are understood to be primarily statistical in nature, with properties distributed over modal curves. It is expected that most populations will be polymodal (in biology, polytypic). On this view, a species is the set of all populations (the metapopulation) that exchange genes. Asexual (nonreplicator exchanging) species have to be defined in another way, usually based on their genome clustering about a “wild-type” mode that is at an adaptive peak.
A somewhat different approach derives from the taxonomic discipline of cladistics. This approach of classification is by genealogy (clades, meaning branches) rather than by grades or overall similarities. The fundamental object of cladistics is the lineage, which is variously understood as a sequence of populations or as the parent-child relationship of any organic entity. Lineages that split (cladogenesis) form clades, which are the ancestor and all descendent lineages of a particular branch. Such groups of taxa are called monophyletic and are the only natural group allowed in cladistic methodology. However, while clades are taxonomic groups based around species, cladistic methods also apply to genetic lineages, and the resulting trees of genes are considered to support phylogenetic trees if they agree. If they do not, a consensus of gene trees is held to give the taxonomic, or phylogenetic, tree.
Part of the problem with an evolutionary perspective for taxonomy is that the entities are not stable, and they are not all commensurate. While in the older Linnaean system, there were absolute ranks such as genus, family, order, class, kingdom, and so on, in the cladistic view, these are arbitrary and relative, and the only “natural rank” is the species, in part because the assignment of groups of species to these ranks is a subjective one. Cladistics is therefore a rankless system.
Beneath the level of species, there are smaller entities. In biology, there are subspecific groups known as geographical races, although the term race is now often avoided in favor of some other term, such as variety. Geographical groups are specified by a correlation of traits with distinct ranges. However, varieties can also be found across ranges, and in this case these are referred to as subtypes, varieties, or trait groups.
While the level of the individual organism (the inter-actor) is considered to be the primary focus of selection, another level is the kin group, in which the genetic fitness of a gene is the average fitness of all copies of the gene in all interactors (inclusive fitness). This allows selection to make individuals altruistic toward their kin, without making their genes less fit. However, it does not allow for altruism (in which a replicator makes its vehicle behave in ways that benefit other vehicles and not itself) to evolve toward nonkin, that is, those who do not share that gene.
Some have proposed group selection, in which the larger scale groups behave as the “fitness bearers.” Many think this is inconsistent with the replicator-interactor distinction or its precursors, while others seek to find analogues at higher levels to the canonical replicators and interactors. Group selection has been proposed for trait groups, temporary populations competing against each other, for species, and even for larger-scale clades. Even the classical argument of Williams against group selection allowed it could occur.
Some have also argued for the reality of body plans, or generalized structural arrangements of organisms, in evolution acting as constraints on the possible evolutionary trajectories in a “space” of morphologies (morphospace), although others consider this to be an artifact of the modeling methodology.
Ontologies and the Metaphysics of Evolution
Overall, evolutionary ontologies have been divided into those that are historical (process based) and those that are ahistorical (pattern based). Things that are defined ahistorically, such as some definitions of species, tend to rely on the characters or propositions used to define them. These things form classes, and they do not change, although a lineage might change into a class from another class or grade. A more recent approach is to see evolutionary objects as individuals, meaning not individual organisms or cohesive systems, but historical objects with a boundary in space and time. Michael Ghiselin and David Hull defined species as individuals in this way (although they have also defined species as cohesive individuals as well). Similar considerations apply to other evolutionary objects, such as genes; they can be classes on one account or individuals on another, but not both in the same account.
- Dawkins, R. (1989). The selfish gene. Oxford and New York: Oxford University Press.
- Gee, H. (2001). Deep time: The revolution in evolution. London: Fourth Estate.
- Gould, S. J. (2002). The structure of evolutionary theory. Cambridge, MA: Belknap/Harvard University Press.
- Hull, D. L. (1988). Science as a process: An evolutionary account of the social and conceptual development of science. Chicago: University of Chicago Press.
- Oyama, S. (2000). The ontogeny of information: Developmental systems and evolution. Durham, NC: Duke University Press.
- Sober, E., & Wilson, D. S. (1998). Unto others: The evolution and psychology of unselfish behavior. Cambridge, MA: Harvard University Press.
- Sterelny, K., & Griffiths, P. E. (1999). Sex and death: An introduction to philosophy of biology. Chicago: University of Chicago Press.
- Williams, G. C. (1966). Adaptation and natural selection: A critique of some current evolutionary thought. Princeton, NJ: Princeton University Press.
- Wilson, R. A. (Ed.). (1999). Species: New interdisciplinary essays. Cambridge: Bradford/MIT Press.