In my last article, I summarized a lecture presented at CELS (Conference on the Engineering of Living Systems) that presented an adaptation model based on the engineering principles employed in human-engineered tracking systems. I will now discuss the connection between these principles and design inference.
By way of examination, biological adaptation is often driven by systems that employ three subsystems:
- Sensors that monitor specific environmental conditions.
- Logic analyzers such as switches that trigger responses when certain conditions occur.
- Mechanisms that result in targeted output responses.
Complexity and irreducible delays
It is almost obvious to say that such monitoring systems could not have evolved gradually. Many examples of NGEs do not even directly help an individual organization but only a the whole population acting in concert. For example, increasing the mutation rate to quickly generate targeted genetic variation will often only help a few lucky individuals survive extreme threats such as an antibiotic.
More generally, not only are all traceability systems irreducibly complex, but they require careful integration of the subsystems. And integration components, such as switches (here, here), correspond to quantities of information much greater than those which could have been generated in the time available. This challenge is evidenced by the fact that timescales (waiting times) grow exponentially with the amount of new information required (here, here).
The design connection
The presence of highly controlled adaptive mechanisms is directly correlated with life using a top-down design that must meet many strict technical constraints. If organizations were the result of random, non-directed processes, their design constraints would be few and very flexible. Changing anatomy and / or physiology should then be relatively easy, and the same undirected processes could potentially lead to the changes. On the other hand, the presence of many strong constraints is correlated with a much more difficult modification of the system. Significant changes would generally require highly specified and coordinated changes.
Szallasi et al. in Modeling of systems in cell biology tacitly arrived at the same conclusion:
An oft-noted caveat against the type of analogies between biological systems and engineered systems that we have put forward is that these two types of complex systems arise in fundamentally different ways, namely through evolution versus top-down goal-oriented design (see, for example, Bosl and Li (2005)). Obviously, scalability is of paramount importance for living systems (Kirschner and Gerhart, 1998). Here, we think of scalability simply (perhaps naively) in the sense of controlled and structured change in lineages, rather than cells, over long time scales in response to possibly large variations in the environment. At the population level (of all engineering systems of the same type), it is evident that advances in engineering meet similar criteria. [Emphasis added.]
Note how the authors do not describe evolution using traditional terms such as “random” and “undirected”. Instead, they describe the change as “controlled” and “structured”. Their description of scalability looks less like a neo-Darwinian evolution than a technological innovation.