Redundancy has a bad name. Already in antiquity redundantia implied excess: literally, the superabundance of a resource (such as water); figuratively, an overflowing stream of words as in the Ciceronian “illa pro Roscio juvenilis redundantia” (Or. 30: 108). In the era of industry and post-industry, the word retains a ring of inefficiency. If something is redundant, by definition, it is something to be trimmed, something that is misaligned with history’s headlong rush into the future. The issue is not just efficiency, it’s also cost (duplication), balance (to have redundant elements is to overweight one facet of an argument at the expense of others), distraction (the overly weighted attracts disproportionate attention).

But whether in the domain of discourse, education, science or engineering, there is no safety, success, or sustainability without redundancy. Reliance upon a single system, line of argumentation, or approach amounts to putting your eggs in a single basket. This may work well enough in simple, highly constrained universes, whether social or mechanical, but not when it comes to grappling with complexity, noise, interference, or uncertainties of the world, not to mention nature.

Redundancy implies the presence of a backup, an additional channel, a fallback option, a support, embedded and intertwined structures. When an alpha option fails, a beta fallback quickly loses any taint of extraneousness. It ensures resilience. And resilience, understood both as elasticity and the ability to recover, is essential to the success of all social systems and life forms. Redundancy informs the fractal structures of the natural world, allowing for flexibility and adaptability. It drives processes of teaching and learning.

Late in her unfairly maligned Dark Age Ahead (2004), the great urban historian Jane Jacobs makes this very point with respect to education:

Nurturing and instructing human beings in a complex culture demands redundancy of mentors and examples. Redundancy is expensive but indispensable. Perhaps this is merely to point out that life is expensive. Just to keep itself going, life makes demands on energy, supplied from inside and outside the human being, that are voracious compared with the undemanding thriftiness of death and decay. A culture, just to keep itself going, makes voracious demands on the energies of many people for hands-on mentoring.” (159)

And what is true of education and life systems (from genes to cells), extends to domains such as AI, machine learning, and robotics where machines are called upon to interact with both familiar and unfamiliar, anticipated and unexpected tasks. There is a technical solution to the many of the resulting challenges; it involves respect for expense and the indispensability of redundancy in design.

- January 12, 2019