Evolution & Involution

24 Flares Twitter 19 Google+ 0 Facebook 5 LinkedIn 0 Email -- 24 Flares ×

This post will introduce a theme that runs throughout many of the ideas I have been working with recently – the relationship between evolution and involution.

We are all familiar with evolution, the idea that micro-scale changes accumulated over time – via consistent selection pressures – can generate macro-scale change.  The evolutionary concept has been applied successfully to a variety of domains – from biology to economics (firm survival), language and industrial design as just a few examples.  Nonetheless, the common conception of evolution neglects a critical component of the process, without which evolutionary “progress” would not be possible.

That missing piece will easily seen in a timely case study – Apple’s iPhone vis a vis the smartphone market.  The success of the iPhone cannot be explained by the accumulation of small changes alone.  If that is our conception of evolution then the iPhone is decidedly not a product of evolutionary processes.  It looks more like a product of intelligent design (which, from a certain perspective, it is).

Apple’s competitive advantage does not come from accumulating innovations more rapidly than its competitors.  Instead Apple competes by packaging innovations (often preexisting) into a more cohesively integrated whole.  In so doing Apple shifted selection pressures away from the feature set and onto the phone itself.  People buy the iPhone because it works, not because of gaudy processor speeds.  [That integration can be seen as an innovation itself.  This topic is rife with tricky figure/ground oscillations.]

Apple exemplifies the part of the process I am describing as involution.  If evolution is the accrual of micro-scale changes then involution is the integration of those micro changes into macro-scale concepts.  In other words, it is the packaging of smaller parts into larger entities that become operational units in and of themselves.

The dictionary offers several definitions which taken as a whole ( :-) ) convey my intended meaning:

  1. an act or instance of involving or entangling; involvement
  2. the act of being involved
  3. something complicated
  4. Botany/Zoology
    1. a rolling up or folding in upon itself
    2. a part so formed
The picture you get from the definitions above is one of variation being rolled or folded up into a compressed package.  The package entity can then be described as “involved” – in the same way that we commonly use that term – its interior constitution being dense or complex.

Involution in Practice

Involution is the conceptual counterpoint to evolution.  Evolution is unsustainable without involution.  The two are necessary yin and yang.  In the process, a more involved entity is produced at the cost of some loss of potential (i.e. information).

If you don’t understand involution then you end up with a skewed understanding of evolution.  To see why, let’s return to the Apple example.  In a recent techcrunch article MG Siegler writes that Apple’s Magic Is In The Turn Not The Prestige:

Look at the main video being displayed on Apple’s homepage. It’s several Apple executives talking about just what went into pulling off turning the ordinary smartphone into something extraordinary. Yes, again.

They’re the only ones photographing their assembly process with 29 megapixel cameras to ensure that a machine picks the exact inlet from 725 unique cuts. They’re the only ones who spend three years working on earphones. They’re the only ones who would go out of their way to try to re-design a device to look and act similar even though the bulk of it has largely changed.

To paraphrase Siegler’s argument, the iPhone’s competitors merely accumulate features.  The iPhone may accumulate similar features, but it is differentiated by the way Apple packages those features into a cohesive whole.  Everyone else evolves smartphone design while Apple involves iPhone design.  In so doing, Apple defines the category.  To use a cliche: the field produces only evolutionary change; the (original) iPhone was revolutionary.

This phenomenon of category creation is the domain of involution.  It is the side of the process that distills horizontal variation and generates the step-wise changes that we perceive as upward progress.

What would happen if no one played the involutionary role?

We need only look back a few years to the cell phone market before the iPhone.  It was a market that had, through successive product generations, accumulated evolutionary features without compressing those features into new categories.  The result was a chaotic mess – countless haphazardly designed products offering arbitrary feature sets.  The same could be said of the desktop pc market, where undifferentiated products could be found in an endless variety of configurations.

Eventually differentiation becomes necessary.  Unconstrained evolutionary variation produces diminishing returns and increasing involutionary pressures.  The same will be true any time expansion proceeds perpetually without structuring…such system eventually degenerate into chaos.  Involution is the structuring process.  It creates negative space by condensing evolutionary variations.  You might say that involved structures become the platforms for subsequent rounds of evolution to be built upon.

Natural Involution

If the example above sounds too convenient then consider what biological evolution would look like if it consisted only of selection and accumulation of adaptations.  It would degenerate into chaos of micro mutations working at odds with each other…add a couple hydrogen atoms here, drop a couple carbons there…no systemic drift.  Even if some micro-mutations were selected over others, accumulation would result only in optimization of a particular solution space to a particular environment.  It would not produce step-wise change.

When we discuss evolution casually we often gloss over the matter of what is being evolved…

What is the unit undergoing variation?
What is the unit acted upon by selection pressures?

Is it the individual nucleotide pair?  The gene?  Gene sequence?  The individual?  The group?

To some degree the answer is all of the above, but before any could act as an evolutionary substrate they each first had to develop into cohesive entities.  Each organizational unit had to first exhibit consistent patterns of expression before it could be systematically favored by the environment.  The very notion of “variation” implies a persistent form as the background for that variation.  We can recognize variations in DNA only because there is a persistent DNA-ness throughout all the variants.

To be clear, I am not suggesting that anything mysterious or supernatural is occurring.  I am suggesting that evolution moves in more than one direction.  Some environments favor novel solutions.  Other environments favor novel integrations of existing solutions.


Consider the same question with regard to language…what if language were flat?  What if we never grouped particulars into higher order categories but instead required a 1:1 correspondence between words and their object of reference?

If this hypothetical is difficult for you to fathom, there is a good reason for that.  The thought experiment leads to an infinite regress.

What exactly do I mean by an “object of reference”?
Is a computer an object of reference?

A computer is made up of numerous identifiable parts, each with their own names, so a flat language couldn’t treat a “computer” as an object in itself.  It would have to describe a computer in terms of its individual parts.  I’m sure you see where I’m going with this.  No matter how far you descend down the rabbit hole, you will always find a still lower level of constituent parts.  Our language (any language) is hierarchically involved because the reality it references is intrinsically involved.  

How Deep Is The Rabbit Hole?

The skeptic will contend that categories like “gene” merely represent our subjective understanding DNA…and likewise, that human language merely reflects our imperfect efforts to make sense of the objective world.  Ultimately (the skeptic will claim) these concepts only exist in our heads.

This brand of skepticism suffers from several problems.  The first is the infinite regress described above.  If higher level concepts are somehow unreal (subjective), then what is real?  Is it the molecules? The atoms? The subatomic particles?  Some more fundamental particles still yet to be discovered?

The unstated premise underlying the skeptic’s position is that beneath everything – beneath all the atoms and protons and quarks – there must be something akin to a “god particle”, a fundamental physical unit explainable only in terms of itself, which acts as the substrate for everything else.

On that matter you may label me the skeptic.  Every natural entity (or process) that humans have yet encountered has demanded description in terms of something else.  Any fundamental physical particle will beg for an explanation of its intrinsic behavior, which will necessarily implicate still lower level physical entities.  If we deny the objective existence of involved structures then the rabbit hole becomes a bottomless pit.  The alternative is to accept that there is no ground…there are only platforms.  The rabbit hole may be bottomless but the stepping stones along the way down are real.

[aside: I am reminded of a movie I watched dozens of times as a kid – The Never Ending Story – in which a storybook world is swallowed up by an indescribable void known as The Nothing, all because the boy reading the story refuses to believe it’s real.  Dogmatic reductionism is kinda like that…]

The second problem for the skeptic is that we aren’t the only ones who recognize structural patterns.  Animals exhibit consistent behaviors in response to some globs of hydrogen and carbon while reacting quite differently towards other such globs.  They instinctively recognize that some meat bags are food while others are threats.  Tropical storm fronts respond in one way to warm ocean waters and respond different to cold ocean waters.  We observe these patterns not just in their interactions with us (humans) but also in their interactions with each other.

If we inherently observe the world in terms of complex structural hierarchies, and if we naturally construct them ourselves, then there are really only two options:

  • The skeptical position implies a kind of dualism – that we humans are inherently artificial (whatever that means) and that we can’t help but impose our artificial conceptions upon an objectively featureless world.
  • The other possibility is that involution is a natural process, just as real as evolution, and that we conceptualize the world in the way that we do because our brains are a product of that same evolutionary/involutionary process, adapted to perceive the world that formed them.
I’ll take the second…

Drawing Out The Theme

If you have gotten this far you may be wondering what this contributes to our understanding of evolution.  While my aim at this point is only to introduce the theme, a few additional examples may help illustrate potential areas of fruitful cross-pollination…


It is often observed with some curiosity that in the US there is no philosophically consistent political party.  There is no party combining fiscal conservatism with social liberalism (an anti-government bias in both domains).  There is also no party combining fiscal liberalism with social conservatism (a pro-government bias in both domains).

Evolutionary/Involutionary dynamics can help us understand why this is the case.  Both positions end up being pragmatically untenable.  The former taken to its logical conclusion would recognize no role for government.  The latter taken to its logical conclusion would recognize no role for anything but government.  Pragmatically both are necessary.

The difference between the two major parties is in how they combine the two positions:

The republicans adopt involved attitudes towards government, grounded in religion, the constitution, Ayn Rand, or whatever else.  Those principles are accepted as an integrated package and provide the platform for an evolutionary bias in the private sphere.

Conversely, the democrats believe that the people need to be shielded from the Darwinian excesses of an unregulated private sphere.  They treat markets as involved entities chartered by government to serve specific purposes.  That requires them to adopt an evolutionary attitude towards the role of government.

Politics seems to make for strange bedfellows until you realize how both need each other.  Lest you think this a peculiarly modern phenomenon, the same tension can be found in the Declaration of Independence:

We hold these truths to be self-evident, that all men are created equal, that they are endowed by their Creator with certain unalienable Rights, that among these are Life, Liberty and the pursuit of Happiness.

If all men are created equal and all men are endowed with certain unalienable rights, then who is this “we” presuming to speak for everyone?

Interpersonal Relations

The internet has become a breeding ground for advice that follows the form – “do as I’ve done”.  Advice of this form assumes from the get go that the reader is similar in some relevant way to the pontificater.  Such advice rarely comes with any qualification like “this might not work for you if…”  It is simply assumed that what has worked for one person will work for everyone.

Even the best of us sometimes judge people who disregard advice as either ignorant or actively aversive.  We forget the possibility that other people, at some fundamental level, might have different experiences of truth, efficacy, salience, or right and wrong.  In such situations it may be enlightening to consider what evolutionary/involutionary biases might be in play.

Here are a few common scenarios that quickly lead to communication paralysis:

  • A couple: One partner consistently tries to pinpoint the root of their disagreements and to address that central concern.  The other partner translates every central idea back into specific examples that have yet to be addressed.  Both leave the argument feeling like their grievances weren’t really heard and frustrated that they merely talked in circles.
  • Business associates: One is thrilled because they managed to deliver their sales presentation flawlessly during a recent meeting.  The other is disappointed that the meeting didn’t result in a sale.  Both feel like the other is unrealistic.
  • Coworkers: One has a tendency to respond to fire drills by treating them as exceptions and devising workarounds that resolve the situation directly.  Her coworker prefers to pinpoint the cause of the crisis in existing work processes and to address the failure at its source.  Both accuse the other of behaving counter-productively and sabotaging their department’s performance.
  • Friends engaged in a debate: The first introduces one hypothetical situation after another in an attempt to illustrate a key point.  His interlocutor calmly bats away each proposition, explaining why the scientific evidence renders it implausible.  Both leave the conversation believing that the other is closed-minded.
In some of these examples it may be obvious who is playing each role.  Others may be more ambiguous.  As noted above, interpretations of evolution/involution dynamics depend on your perspective.  We know from thermodynamics that any effort to produce order also produces entropy.  The same is true here.  A person who creates order from one perspective only adds to the chaos from another perspective.  A person who proactively devises new solutions also creates clutter within that solution space.

Business Strategy

Which corporate strategy is superior, horizontal integration or vertical integration?

If you answered then you answered wrong.  Neither strategy is intrinsically superior.  Strategic decisions must take into account the comparative advantages of the firm relative to its competitive environment.  Notions like best practices are misnomers.  There are no “best” practices, only more appropriate practices and less appropriate practices.

Industries go through cycles of horizontal and vertical integration because both approaches produce diminishing returns once overextended.  Industries that have become too disjointed benefit from consolidation.  Industries that have become stagnant benefit from competition.

If the analogy is not already clear, horizontal integration represents evolution while vertical integration represent involution.  Horizontal integration expands the breadth of the industry, allowing new ideas to enter the market and supporting competition between those ideas.  Vertical integration packages the pieces of that evolutionary process into new cohesive entities, thereby creating new categories and changing the basis of competition.  Horizontal integration promotes competition between comparable parts whereas vertical integration shifts the competitive focus to the sum of the parts.

Neither structure is better than the other.  Both are necessary to undo the excesses of the other.  Both create the negative space that allows for the other.  Vertical consolidation reduces competition, creating space for subsequent waves of horizontal expansion.  Horizontal competition reduces end product differentiation, creating opportunity for subsequent waves of vertical consolidation.

To be continued…

A few months back I wrote a piece entitled The Denial of Art in Science.  While writing that piece I was circling around this same theme – the distinction between involution and evolution.  I just hadn’t found the right words yet.  What I described as art is an involutionary process.  Science (as described in the linked post) is an evolutionary process.  Empirical science introduces new information into the world.  Theoretical science (the art of science) compresses information into operational concepts.  Theoretical explanation provides context and meaning to otherwise flat information.

There are many more parallels to be explored but I am rapidly approaching 3000 words.  In the mean time perhaps you will think of a few of your own  :roll:  To be continued…


Image credit: NASA

Save & Bookmark

  • Neil LaChapelle

    You know Greg, I’m not seeing anything in involution other than an intensification of selection pressure. Apple’s evolution of the iPhone amounts to the imposition of a much stricter selection regime, a tightly constrained niche where only very few designs would survive.

    Flipping into the cybernetic frame, you seem to suggest that evolutionary processes will necessarily cause systems to violate the twin “Laws” of Requisite and Limited variety. In other words, they will proliferate traits that are inefficient to maintain without evolution’s complementary process of involution kicking in.
    However, variation under selection statistically leads to the Limited Variety constraint being upheld over time. Unnecessary variety falls away, because systems which abandon that variety experience no selection disadvantage, while reducing their energetic and information burden.

    These all seem like different selection regimes or ecologies within Darwinian evolution to me. Variation and selection remains the driver, it’s just a question of how ruthlessly a selection regime prunes the variation under local conditions.

    • http://OnTheSpiral.com/ GregoryJRader

      We’ve already had most of this conversation behind the scenes so just closing the loop on this comment.  I didn’t mean to imply that involution somehow circumvents the mechanisms of Darwinian evolution.  I would agree that involution is not a distinct process but is instead the result of a particular selection regime.  Different environmental conditions will bias the selection process towards different selection criteria.  

      Some environments will favor rapid proliferation and variation.  Other environments will favor organisms that are individually more adaptable, the organisms that are omnivorous both literally and figuratively.  I am calling the former evolution and the latter involution.  Technically both fall under the umbrella of Evolution (capital E), but I find the distinction useful because in non-technical conversation I think we overemphasize the former and lose sight of the latter.     

24 Flares Twitter 19 Google+ 0 Facebook 5 LinkedIn 0 Email -- 24 Flares ×