Natural systems have evolved to become incredibly efficient because of their self-balancing properties. This is significant because long-term sustainability depends on our ability to understand – and replicate – the properties of self-balancing and “scale-free” networks.

“[Technology] encounters problems common to all complex systems – problems already solved in the evolution of living systems.” – evolutionofcomputing.org

More introductory remarks: show

In nature, we consistently see two things happen: (1) Incredible efficiency, and (2) Emergent phenomena like consciousness and the incredible meta-awareness of living cells. In many natural systems, the parts are continuously aware of the whole, and this is responsible not only for great efficiency but also for incredible levels of resilience and coordination. This is made possible through a unique form of virutalization: in natural systems, each part carries an image of the whole (encoded as genetic information). This is the real significance of “scale-free” or “holographic” networks — even isolates cells are still networked.

Nature has already produced optimal solutions to many of the design challenges we face today and will face in the years to come. To offer just one example, living cells are essentially what nanotechnology should become given a few more years. Generally speaking, most of what we covet to be ‘futuristic’ technology (such as incredibly tiny computers) already exist in the natural world (consider living cells). In a sense, as technology evolves and continues its exponential growth, it only becomes more intertwined with the natural world.

What is flow?

Flow is the process of energy moving through a system. Most unattended flows travel from high to low energy density (the Second Law of Thermodynamics); it is consciousness that is negentropic. All natural systems can be modeled in terms of flow – that is, inputs and outputs.

“Flow systems have two basic features. There is the current that is flowing (for example, fluid, heat, mass, or information) and the [network] through which it flows.” – Adrian Bejan, J. Peder Zane

Flow is not a thing – it is nonpoint – but we can often find its fingerprint. As flows propagate, they form curious patterns – visual byproducts of optimization.

Flow is demonstrated by electricity as it permeates through a metal sheet, or water as it moves through a river delta, or electrical signals as they jump between neurons. Flow is a universal aspect of systems – living and nonliving alike – and it is an important new science because it is capable of describing both manmade and natural flow systems in a common language.

Example: A River Delta. show

Rivers share the same branching patterns as lightning bolts, cardiovascular systems, neural networks, and even cities (where roads and highways act like the circulatory systems for large urban centers). Without knowing it, manmade networks have already begun to replicate natural flow, precisely because it is most efficient.

Network Architectures and Flow Optimization Strategies

The brain’s neurons, the branches of a tree, the arteries of a heart, and street traffic through Lower Manhattan — are all flow systems. Flow describes the movement of energy and matter in universal terms, and by studying its dynamics, we can predict, optimize, and understand the behavior of both animate and inanimate systems. Flow unites the micro and the macro around the idea of optimization.

Because networks we see in nature are the result of incredibly long spans of time, they demonstrate great efficiency. Conversely, manmade networks tend to slow down over time and experience bottlenecks because they are structured differently. Whereas manmade networks typically follow a star (centralized) structure, natural networks are often more flexible and decentralized, despite remaining integrated and functional as a whole. This matters primarily in terms of scalability and efficiency.

Pictured on the far left, the so-called Legrand Star is a centralized network structure found in “rigid systems and institutions, where ‘all the intelligence [is] at the center of the network.’ These structures tend to be less resilient and adaptable to change, often embracing top-down, noninclusive decision-making, with closed, not open, information flows.” These networks are not commonly found in nature and may grow unstable beyond a certain size. Whereas most cities and beehives are shaped like Legrand Stars, the vast majority of structures in nature are best described as “peer networks” – much more dispersed, more flexible, and incredibly resilient. This model is discussed at greater length in Future Perfect: The Case for Progress in a Networked Age by Steven Johnson, which I would highly recommend (review).

If sustainability is our goal, then we should seek to emulate nature’s dynamism by building less Legrand Stars. Instead, we should focus on self-optimizing, distributed networks like the ones we see in nature. There are trade-offs, but distributed architectures are unquestionably more stable than their centralized counterparts. To adopt nature’s “scale-free” or “mesh” network architecture would mean implementing DVFS-like (“neuromorphic”) optimization wherever possible — to route inputs and outputs more intelligently, and to scale response more appropriately to need. This idea of dynamic scaling is extremely important from an efficiency perspective, especially when dealing with large networks. As networks grow larger, they must also stabilize their growth and perpetually become more efficient (to theoretical limits).

“Dynamic voltage [and frequency] scaling is a power management technique in computer architecture, where the voltage used in a component is increased or decreased, depending upon circumstances.” – Wikipedia

DVFS is very useful as a metaphor as well as a practical model to optimize networks of all kinds because it keeps networks stable. It is by definition self-balancing, so generally speaking, DVFS is how natural flow systems work. The core idea here is sustainability: flows that are efficient grow; flows that are less efficient merge or regress.

“For a finite-size flow system to persist in time (to survive) its configuration must evolve (morph) in time in such a way that it provides easier flow access.” – Prof. Adrian Bejan, Duke Univ.

This is really just another way of saying that “flows must flow better with time.” It is an evolutionary pressure and necessary consequence of there being time.

Nature’s Constructal Pattern: The Ultimate Technology

Nature seems to evolve to facilitate better flow, which adheres to the Principle of Least Effort. Interestingly, these patterns are self-optimizing – they ‘flow better’ with time and are capable of incredible transformations.

“The designs we see in nature are not the result of chance. They rise naturally, spontaneously, because they enhance access to flow…” – Adrian Bejan, J. Peder Zane

In other words, these recurring patterns we see in nature are not mere coincidence. They are the result of billions of years of natural selection, and they represent incredible thermodynamic efficiency.

If there is a “Grand Design” as Stephen Hawking puts it, the Constructal Pattern is a likely candidate. Its treelike branching pattern of neurons responsible for biological consciousness also seems to be – quite remarkably – also a structural feature of the universe itself. Indeed, Dark Matter superstructures in the Milky Way are visually identical to brain cells.

Order from Chaos

The Constructal Pattern is self optimizing. To me, it’s also incredible how matter, on all scales, is “inclined” to form this pattern – the same pattern related to consciousness.

Matter presumably organizes itself into the Constructal Pattern because it is the “optimal answer” to the laws of physics we experience — i.e. natural constants such as the force of gravity, electron spin and charge, and so on. It is stable. So in a sense, this means consciousness is the inevitable product of space and time. The universe “wants“ to make consciousness, because matter itself has an affinity to its structure. It also raises the idea of there being a greater multiverse of universes, with varying constants.

The Constructal Pattern is widespread – in fact, its prevalence suggests that it is a structural feature of the universe — maybe even written into the very fabric of spacetime itself or encoded in the Compact Dimension. In any case, this pattern can be found everywhere, from the subatomic (e.g. diffusion of electrons) to the super-massive (e.g. the Cosmic Web, on the scale of galaxies).

“Electrons follow the same natural laws that govern all systems that flow—electricity snaking its way from a storm cloud to Earth, rivers branching into ever smaller creeks and streams, or the spidery web of veins that distributes blood throughout your body.” – Symmetry Magazine

“[The Constructal Pattern] sweeps the entire mosaic of nature from inanimate rivers to animate designs, such as vascular tissues, locomotion, and social organization.” – Adrian Bejan, J. Peder Zane

The Internet shares a similar architecture, and interestingly, so does the distribution of dark matter in the Milky Way (see the Bolshoi Simulation and links in the footer).

It seems the Constructal Pattern serves as a holographic ordering principle for matter. Is the Constructal Pattern what shapes otherwise non-living matter into meaningful, living configurations? Another question to ask might be — is this pattern alive and sentient? We already know that the dense network of neurons in our brains is directly related to consciousness — if the same patterns persists on much greater scales than that, what might it imply about our universe?^[1]

Technology Becoming Nature?

There are signs that what we think of as nature and what we think of as technology are actually bound to converge – evidenced by how some manmade networks like the Internet are beginning to mimic the structures we see in nature.^[2] If the arrow of time always points towards greater efficiency, then the only design schematic we need may be the Constructal Pattern, because of its unique, self-improving structure. As described earlier, “[our] networks have begun to replicate natural flow, precisely because it is most efficient.”

“The natural world has found ways to work and live in harmony for a long time. If we want to also survive and create sustainability systems, biomimicry may be the key.” – Creating Resilience by Following Nature’s Lead

The core idea I seek to convey is that we can accelerate this process of co-evolution; we can learn from nature to make better technologies (because these ‘evolutionary forces’ of growth point in the same direction: better flow, improved efficiency, iterative refinement, so on). This is known as the field of biomimetics. Here’s a good TED talk on the subject:

The Significance of Recurring Patterns in Nature

A common trend of technology is to seek maximum thermodynamic efficiency: minimize entropy, and maximize work per unit energy. Scientists and engineers go through many design iterations to improve flow (and decrease waste heat) in new technologies. Sometimes we can take a huge shortcut by observing and emulating the designs of preexisting natural systems. Photosynthesis is just one example:

“Through photosynthesis, green plants and cyanobacteria transfer sunlight energy… into chemical energy with nearly 100-percent efficiency.” – Berkeley Lab, also see ScienceDaily, Wired.

Because natural networks are self-balancing, they grow to become highly resilient and adaptable with time. The ability of natural systems to self-configure for optimal flow is a trait we should seek to emulate in new technologies. Hardware that can be “programmed” like software is a very promising start (see Reprogrammable Chips Could Allow You to Update Your Hardware Just Like Software). This model can also be applied to cognitive computing and psychology, because we can treat the human brain as a flow system that configures itself for tasks it does repeatedly.^[4]

If we want to design more efficient technologies, we already have a great template to work from, and it’s all around us (and inside our own brains). The evolution of flow maximizes energy distribution while simultaneously minimizing heat loss.

Mass-energy equivalence and the speciation of flows in nature: show

Nature is Technology

By (1) modeling natural networks and (2) comparing the structure of these networks to existing manmade networks, we can find new optimizations to make and bridge the ‘efficiency gap’ between between what nature is today and what technology can be tomorrow.

This, I think, is the real meaning of evolution: The symbiotic co-evolution of nature and technology — the magic that happens at the intersection of life, the natural environment, and the tools we use to alter that environment (e.g. technologies). We can even start to think of “nature” and “technology” as the same thing — a unified direction, a force of evolution and flow optimization. Seen this way, there is no real difference between what we classify as “nature” or “technology” – they are just two halves of the same process.

For instance, we might think of the power of fusion as a supreme technology, but the stars we see in the night sky formed naturally. Similarly, the body, a truly incredible machine with tremendous information storage capacity, is biological.

Nature is a technology, and a surprisingly good one at that. In this sense, this means that evolution itself can be understood as the ongoing (and theoretically endless) cycle of nature and technology inspiring reciprocal designs (or ‘flow systems’). Jason Silva discusses similar ideas in the following video.

The Idea of Functionalism

Nature and technology are not opposed. It may not seem like that today, but we need to correct the notion that technology must always grow at nature’s expense. In truth, technological growth does not directly equate to environmental consequence. Like all new technologies, the first made are often the least efficient.

The ontological forces of nature and technology are indeed synergistic and parts of the same cycle – one of my core beliefs both as a scientist and as a person.

“The major problems in the world are the result of the difference between how nature works and the way people think.” — Gregory Bateson

Natural systems are optimally efficient because they’ve had the most time to evolve. Compared to the universe (a natural system evolving for roughly 14.6 billion years) human technology is relatively brand new – hence the frequently mistaken belief that nature and technology are somehow opposed. In any case, technologies that seem unnatural may simply ones that haven’t yet had enough time to to mimic nature’s flow, i.e. answer the flow optimization problem using nature’s constructal blueprint, thus becoming natural.

Nature and technology: an endless cycle? show

The ideas I’ve mentioned here are discussed in greater detail in Design in Nature: How the Constructal Law Governs Evolution in Biology, Physics, Technology, and Social Organization.