Embodying imagination: fictional species in real science

Jessica Bolker, Department of Zoology, University of New Hampshire

Abstract
A fictional species is a population of organisms whose biological properties are carefully and thoughtfully described – even though the organisms don’t exist. We often recognize such species in literature, especially science fiction and fantasy; indeed, they are a highly appealing element of those genres. But fictional species also inhabit real science, where they play an essential role as embodiments of theories, representations of ideas, and focal points for hypotheses. Identifying “model” and hypothetical species as fictions is a first step toward exploring their pivotal, though largely unrecognized, role in the scientific process.

Biologists invent and work with fictional species all the time – a proposition virtually all of them would deny if it were put in those terms. “Fiction” is commonly understood to be made-up, not real, divorced from the “facts” to which we are supposed to be wedded. But fiction can also be defined as a reference to something that does not exist in the real world. Genuine fictions are rule-based inferences beyond that which is directly known, and often serve as a powerful tool for thinking about how the real world works. Thus, biologists’ fictional organisms fall entirely within the tradition of the scientific method: proposing and exploring the properties of fictional species is one form of scientific hypothesis-building and -testing.

I will examine two uses of fictional species in scientific research:

1. Reconstructions

Fictional reconstructions, particularly hypothetical ancestors of real organisms, are most commonly found in evolutionary biology and paleontology. One reason we need to construct fictional versions of long-extinct organisms is that data about real species and lineages are often irredeemably incomplete. Second, the kinds of mechanisms and processes we can observe directly are separated by an essentially unbridgeable epistemological gap from the macroevolutionary patterns we can trace over geological time (Kemp, 1999).

2. Model organisms in developmental biology

By definition, what matters about a model species are its generalizable properties, not its unique characteristics (Bolker, 1995). Species chosen as models are intended either to represent a universal case (zebrafish as “the vertebrate”), or to act as proxies for a particular target organism that is difficult or unethical to study directly (e.g. humans, which are represented in biomedical research by everything from yeast to rhesus monkeys).

I claim that developmentalists’ “model” organisms are fictional species. The utility of a given model for illuminating fundamental biological mechanisms or for making medically relevant predictions depends on the strength and process of inference from a particular organism to a general case or to a different species. Either way, what matters is not the individual characteristics of the actual model, but the extent to which the species serving as a model allows us to study more abstract principles by exploiting the fiction that it is something beyond a unique real species.


Works cited:

Bolker, J.A. 1995. Model systems in developmental biology. BioEssays 17(5):451-455.

Kemp, T.S. 1999. Fossils and evolution. Oxford University Press.

Not available

Back to Abstract