Evo Devo of Hierarchical Substrates (EDHS): A Starter List
Recall the Five Universal Hierarchies mentioned earlier in this chapter, and all the replicating complex systems that exist inside each hierarchy (physics, chemistry, biology, society, technology). Those subsystems that are on the direct path toward greater local intelligence, with all of its subvariables, can now called hierarchical substrates (HS). Putting them together with the way that we are presuming they are made, evolutionary development (ED), gives us a term describing both the process and the outcome: the evolutionary development of hierarchical substrates (EDHS). Whenever we use the term or acronym EDHS in this guide, we are referring to both the process of evo devo and the outcome of accelerating intelligence.
Fortunately, evolutionary developmental hierarchies are beginning to be discussed seriously by some evolutionary biologists, folks who have typically denied the idea of progressive, irreversible adaptive change. One widely-cited treatment of hierarchies for biological life is theoretical biologists Smith and Szathmáry’s The Major Transitions in Evolution (1995). They chart life’s progressive rise from replicating molecules to RNA to prokaryotes to sexual eukaryotes to protists to multicellular organisms to societies with language (memes). They specifically propose eight major transitions in the evolutionary developmental emergence of biological complexity in universal history to date:
- Replicating molecules to compartmentalized replicator populations (microspheres, etc.)
- Independent nucleic acid replicators to chromosomes (linked replicators)
- RNA as replicator template and enzymes to DNA as template and protein as enzymes
- Prokaryotes to eukaryotes (with captured mitochondria and chloroplasts)
- Asexual eukaryotic clones to sexual populations
- Single-celled sexual protists to multicellular organisms
- Solitary multicellular individuals to tribes
- Primate societies to human societies, with their endlessly-improving oral and behavioral language (“memes”)
Other scholars have proposed additional, increasingly “postbiological” transitions. Here are four more, bringing our starter list to twelve:
- Language-enabled ‘memes’ (replicating ideas and behavioral algorithms, in brains) generating ‘temes’ (replicating symbols and technological algorithms, in machines) (Blackmore 1999).
- Biologically-inspired machines leading to a technological singularity (autonomous machine intelligence) (Kurzweil 1999, Clark 2003)
- Postsingularity human-machine symbiosis leading to some kind of global superorganism (Miller 1978, Stock 1993)
- Postbiological intelligence eventually replicating its universe (or the univ. replicates itself, with a little guidance from intelligence) (Smolin 1992, Harrison 1995, Gardner 2000)
- Blackmore, Susan (1999) The Meme Machine.
- Clarke, Andy (2003) Natural-Born Cyborgs: Minds, Technologies, and the Future of Human Intelligence.
- Harrison, Edward R. (1995)The Natural Selection of Universes Containing Intelligent Life.A.S. Quarterly Journal 36(3):193-203.
- Gardner, James N. (2000) The selfish biocosm: Compexity as cosmology. Complexity5(3):34-45.
- Kurzweil, Ray (1999) The Age of Spiritual Machines: When Computers Exceed Human Intelligence.
- Miller, James Grier (1978) Living Systems.
- Stock, Gregory (1993)Metaman: The Merging of Humans and Machines into a Global Superorganism.
- Smolin, Lee (2004)Cosmological natural selection as the explanation for the complexity of the universe.Physica A 340(4):705-713.
Additional candidate stages can be added before replicating molecules (Transition 1 above), including replicating quantum events and matter production in the early universe and replicating stars and their complex planets. We leave those out of this treatment (see “Major Transitions in Evolutionary Development” at EvoDevoUniverse.com for more possible stages).
Note the similarities and differences between this list of Twelve Transitions (Hierarchical Substrates), Hawkin’s Five “Mindsteps” in cosmic understanding, and the Eight Great Migrations listed in Chapter 2. We are likely not nearly smart enough at present to build a definitive set of such lists. Our physics and information theory not yet precise enough to know how to best define where one evo devo substrate ends and another different and more intelligent one begins, as a simulation-predictable outcome of universal evolutionary development.
Nevertheless, simply offering a few candidate lists, as we will do throughout this Guide, gives us a rough idea of how hierarchies have emerged to date, our current place in them, and what may come next. Discovering such hierarchies in our psychological, social, economic, and technological spheres are of course very helpful for personal, organizational, and global foresight as well.
When we think about hierarchical substrates, we should also ask, when does “more” become “different”? The ancient Greek Sorites Paradox asks, when does a grain of sand become a heap? Change anything long enough, and the accumulation becomes something more. More eventually becomes different, as described in the famous essay by P.W. Anderson, “More is Different,” Science 1.4.1972. This is the essence of all emergence of new complex systems, laws, hierarchies, and environments associated with cumulative change.
This change can be either evolutionary (divergent, diverse, contingent) or developmental (convergent, integrative, predictable). For hierarchical substrates, the small subset of emergent systems on the leading edge of adaptive intelligence growth, accelerating developmental change is always involved, either before or after the substrate’s emergence, or often, both. But just because a particular process is accelerating doesn’t mean that we should expect disruptive new substrates to emerge from it anytime soon.
For example, consider the implications if today’s 77% growth in DNA Sequencing Cost Efficiency holds for just four more years. A $1000 (consumer cost) genome with strong redundancy (30X) was reached in 2014. If these trends hold a few years longer, a $100 genome will be reached in 2018. That sounds exciting, as it will get us a lot more personal genomic sequencing, but all that sequencing is likely to lead to very few major medical advances over the coming decade, as we are still very far from understanding how genes work, as physical and computational systems in living bodies. A $100 genome in 2018 will give many more of us the potential to do scans that we still can’t do very much with.
So we must recognize that some accelerating emergences are only small steps toward our desired end, and we’re going to need to know a lot more about how genes work before we can use the maps that these scans will give us. We must spend billions more on understanding genetic, cellular, and organismic biology. That process of basic research and experimentation is presently far less accelerative however. The manipulation of biotechnology for human benefit moves on a far slower and more ethically regulated timescale than our information and nonbiological nanotechnologies, and the health care industry is much less competitive and more plutocratic than the computer and machine automation industries. So the fields of biotech and medicine really are far less accelerative, in the critical areas of conducting experiments and competition, than the field of intelligent machines. These observations give us a clue that we won’t be seeing a new human anytime soon, but rather, a new humanlike machine intelligence instead.