Construals of causation in the biological sciences can be usefully examined in terms of the logico-semantic relation of cause — reason, result and purpose — on the one hand, and in terms of the ergative function of agency — participant as cause — on the other.
(1) Cause: Reason, Result And Purpose
One source of ideational inconsistency in biology is the failure to distinguish different types of causal relations: reason, result and purpose.[1] Reason and result can be generalised as because reason P so result Q, whereas purpose can be generalised as because intention P so action Q.
A frequent example of reason and result being construed as purpose can be generalised as phenotypic trait X evolved in order to Y; for instance, eyes evolved in order to improve animals survival prospects. An alternative construal, one that is ideationally consistent with evolutionary theory, would be phenotypic trait X evolved because Y: eyes evolved because they improved animals survival prospects.
A similar case is the construal of evolutionary reason and result as behavioural purpose, which can be schematised as animal X does Y in order to pass on its genes. An alternative construal, one that is ideationally consistent with evolutionary theory, would be animal X doing Y can result in its genes being passed on. Animals do not act in order to propagate their genes. Rather, animals act in order to achieve multifarious ends across a range of multifarious contexts, and some of these actions can result in the propagation of their genes.
The distinction between reason and result and purpose is not a trivial one, epistemologically, and deep ideational consistencies can follow from the failure to make it. These can range from misunderstandings of the respective rôles of variation and selection, and the relation between the two, to the belief that “the purpose” of organisms is the passing on of genes. Note, incidentally, that this is ideationally distinct from organisms being a means by which genes replicate themselves, or genes being the means by which organisms reproduce themselves, both of which are instances of manner rather than cause.
(2) Cause As Participant: Agency
In terms of ergativity, an agent is a participant that is the external cause of a process, while a medium is a participant through which a process is actualised. As participants in evolutionary processes, genes are the medium of variation and selection processes, not the agent of variation and selection processes. Genes are the participants through which variation and selection are actualised; they are not the external causes of variation and selection. Construing genes as agents of evolution is ideationally inconsistent with biological theory.[2]
Incidentally, it might be added here that construing biological taxa as the agents of their own evolution, as in clauses like amphibians evolved eardrums or humans evolved big brains is similarly ideationally inconsistent with biological theory.[3] Biological taxa are the media through which evolutionary change is actualised.
As participants in developmental processes, genes function as both agent and medium in cycles of expressing and activating. That is, genes are the external cause of amino acid formation, for example, but the chemical environment is the external cause of the selective activation of genes in that process.
Similarly, to construe genes as self-replicators[4] is to construe them as the agents of their own replication, whereas they are the medium through which the replication is actualised. The external cause, the agent of gene replication, is the chemical environment that either initiates the process or maintains the status quo.
Footnotes:
[1] It is also useful to distinguish ‘purpose’ from the more general notion of ‘telos’, or ‘ends’. Previously established processes, such as those of biological development, have ends that the system in question moves towards. This is a more general meaning than ‘purpose’, which might be best rstricted to describing behaviours directed at satisfying value disequilibria. That is, development has ends, behaviour has ends that are purposes, and evolution, being open and contingent, has neither purpose nor ends.
[2] Construing selection metaphorically as the agent of evolution conceals the congruent agent causing the selecting.
[3] Typical, not only in science journalism, is the following, which combines ‘species as the agent of evolution’ with ‘evolution as purposeful’: ‘Dr Brennan believes [mallard] females evolved convoluted oviducts to foil the male rapists.’ [Maggie Fox, Reuters, SMH 2/5/07]
[4] Note also the Agentive suffix –or.