Innovation in biology and technology: exaptation precedes adaptation.
Last modified: March 17, 2006
Innovation in biology and technology: exaptation precedes adaptation.
Durham Business School, Mill Hill Lane, Durham, DH1 3LB, United Kingdom
Tel. +44 191 334 5385, fax +44 191 334 5201, firstname.lastname@example.org
D.Sc., F.I.Biol., Mathematics Institute, University of Warwick, COVENTRY CV4 7AL UK
We are accustomed to adaptation, which assumes evolution of structure towards better function. Initially, however, at the root of any adaptive trajectory it is usual for a structure to have been subverted – perverted – to a new function: what Gould and Vrba called “exaptation”. Generally this has been regarded as contingent, serendipitous. We believe that “complexity” rather than reductionist models, particularly plotting the geography of exaptational phase space, suggest several regularities that could be exploited to explain puzzling biological innovations and perhaps to develop a new strategy for technological/industrial innovation. We give historical examples, including radio tubes (valves), feathers and teeth [not in text] as well as cases where biological understanding has enabled technical exaptation. We`conclude that the driver of exaptation is not just function and form but context.
Nearly all biological adaptations, and many commercial products now suited to particular markets and functions, began life as something different. Our challenge is to understand how the leap to the new function – prior to the adaptational trajectory that we think we usually understand – can be conceptualised other than as accident, contingency, serendipity. We have not taught innovation-by-exaptation because the new connection creates the new phase space; it is not that there was an island of possibility waiting to be exploited (Cohen et al., 1994) but that a new domain appeared in the universe of possibilities. Can we find the parameters for a phase space of exaptations?
Several classical concepts have changed their usage in ways that help our conceptualisation here. Biological/ecological “niche” used to refer to some pre-formed - or perhaps Platonic ideal – function waiting for an organism to evolve into it: as in “The scavenger niche has been emptied now we’ve killed off the vultures in India; what will replace them in that niche?”. In contrast to this lock-and-key image, since Lewontin, Laland and Odling Smee (Odling-Smee et al., 2003) we have seen a niche as a complicity between a lineage of organisms and the environments it is inhabiting and changing – a creative process, not a habitation-for-rent. This is just the difference between exaptation and adaptation. Because new niches are often repeated during evolution and the parallel innovations are quite detailed, they do seem to be rule-constrained. This is in contrast to the usual belief that they are simply contingent.
Cichlid fishes in Lakes Victoria, Malawi and Tanganyika invented parallel arrays of unlikely specialisations: in the three lakes the different riverine originators diversified into “species flocks” that included adapted plankton feeders, rock-scrapers and specialised piscivore predators. But all three lakes also have cichlids that have exapted to eating the babies from the mouths of other (mouth-brooding) species, others that specialise in eating only the eyes of other fishes, and some have exapted hinged teeth to scrape off the scales around the tail of a caught-but-not-swallowed prey. All of this requires a jump to a new function and cannot adapt smoothly (find Gould the flamingoes smile and the aerodynamic of wings). But there have been temporal repeats too, as in the sagas of acquisition of sabre-teeth in creodonts and horns in titanotheres. Each drives the other to extinction along an arms-race adaptational trajectory, then the exaptation repeatedly re-arises from the smaller-bodied parent stocks without sabers or horns. These examples strongly suggest contingent but rule-driven innovations. There are parallel innovations, such as the phenomenon of “steam-engine time” when everyone invents steam engines, in biology “natural-selection time” for Spencer, Darwin, Wallace. All these cases suggest that the opportunities for exaptation are at least as much to do with context as with the nature of the tool or the nature of the task. Can we distinguish between rule-driven exaptations and the Conway-Morris adaptational trajectories (Conway Morris, 1999). We believe his model to believe the intelligent design promoters who maintain the same function throughout an evolutionary event.
So we have three players in the game, not just the two of form and function. There must be permissive contexts, perhaps driving contexts, contexts that constrain toward contiguity of form and function, so that exaptation is promoted. Riverine cichlids in a filling lake basin, and newly arrived finches to the Galapagos are like radios and tape players that first moved into the transistor age, or early automobiles before the emergence of motorways and gas stations. The ‘invaders’ find environmental gradients that drive the emergence of new exapted functions. In order for the “adjacent possible” to include more possibilities, in order for the chances of exaptational contingency to be high, diversity-generators should be in operation. The most familiar of these are the symmetry-breaking bifurcations that result from chains of complicity between processes: early automobiles, rare but with rich drivers, made gas stations economically viable, and better roads, which in turn promoted diverse and economically viable automobiles for nearly all in that culture. As the cichlids, or the finches, multiplied and diversified, new opportunities for sympatric speciation arose (Stewart et al., 2003). These are ‘horizontal’ exaptations, widening the phase space. Another diversity-generator is architecture reconfiguration. Organisms, technologies and eco-systems are self-referential architectures of modular parts and functions. If the architecture collapses, as during a biological or industrial mass extinction, then the modules search for new architectures. Functions and tools may then be ‘vertically’ exapted in new architectures, or bodyplans. The potential for exaptation is not to be sought in properties of a tool or its uses but in the facilities, in that developing context, for new utilities to be engendered.
What properties of the context are permissive for exaptations? Clearly, a sparse, almost uninhabited region of phase space has few opportunities for the contiguities necessary to establish functional bridges among tools, technologies or indeed species. When sound-recording began, wax cylinders, discs then early wire recorders as electricity came into technical use, there were few contiguities. Equally, a very crowded region probably fosters high competition, low “rents”, and unacceptable costs for the initial stages of an exaptation, before progression on the adaptational trajectory. The early lake with few fishes gave few if any opportunities; equally, when the ecosystem had matured into hundreds of specialised niches, there was no physiological “room” to experiment. However, as the era of electrical-recording was paralleled by the digital revolution in electronics, the subversion of digital compact-disc technology to music, and then to sound-and-picture DVD’s was inevitable. Now the music-recording has leapt over into solid memory, and adaptational trajectories involve compression techniques and display/play development. Silicon Valley, however, has exploded away from the sparse/dense constraint. It is a multidimensional network of reconfiguration experiments. Exposure to new contexts (projects) favours translation of tools into a new function and perhaps forms (‘horizontal’ exaptation), whereas cooptation of tool modules into new architectures (‘vertical’ exaptation) generates new technological families. The system has became autocatalytic, almost ‘mesobiotic’ (Lancet ..).
In conclusion, enhanced models of evolution, based on a multilevel, hierarchical understanding of biological and technological world, can help the innovation process. In particular, the concepts of horizontal and vertical (or recombinant) exaptatation have the potential to provide managers (and policy-makers) with an additional tool to increase the speed of innovation processes.
Cohen J, Stewart I. 1994. The Collapse of Chaos: Discovering Simplicity in a Complex World. Viking: London
Conway Morris S. 1999. The crucible of creation: the Burgess Shale and the rise of animals. Oxford University Press: Oxford
Odling-Smee FJ, Laland KN, Feldman MW. 2003. Niche Construction: The Neglected Process in Evolution. Princeton University Press: Princeton, N.J.
Stewart I, Elmhirst T, Cohen J. 2003. Symmetry-breaking as an origin of species. Trends in Mathematics, Conference on Bifurcations, Symmetry, Patterns. Birkhauser Verlag: Basel, Switzerland