(This is a loose adaptation of a talk I sometimes give on the Cambrian Explosion, smoothed a bit for popular consumption. That talk, in turn, draws heavily from a 2006 paper by Charles Marshall, titled “Explaining the Cambrian ‘Explosion’ of Animals”. It can be read in full here. I’m mostly just trying to test out the new site with an essay I had lying around, be moderately entertaining, and maybe try to suggest this event as a topic of interest for those who care about intelligence explosions and cognitive emergence. If I succeed in all three, then I encourage you to start with Marshall for the more technical, thorough, and correct analysis.)
The Cambrian Explosion is our name for an event that took place about 540 million years ago, one that accounts for the sudden appearance of advanced animal life. Emphasis on “sudden”; it’s like somebody flipped a light switch. What follows is just a story (fragments of a story, really), but it’s a pretty good one, and I don’t think it’ll lead you too far astray.
Let’s start 540 million years ago. The planet is still recovering from a series of catastrophic “snowball Earth” phases, ice ages so severe that the oceans had frozen right across the equator; the tropical glaciers wound their way through loose sediment to leave an enduring footprint. Now that the last of these has thawed, it’s finally starting to warm up in a more permanent way and get a little more hospitable. There are two major continents at the moment, but one of them is starting to break up into smaller and smaller chunks. As with our rising temperatures, this helps create a more habitable planet, since it gets you more high-nutrient coastlines per unit land area. Importantly, the last hundred million years have provided us with oxygen concentrations in the atmosphere at basically 21st-century levels. To the unprepared organisms that dominate the first half of Earth’s history, oxygen can act as a hazardous toxin, but if you harness it correctly you can really kick your metabolism into a higher gear. A good time to be alive, really.
The land is mostly barren. A little bit of pond scum holding on in shallow pools and other moist areas, maybe, but seeds and placentas won’t exist for a long age yet, and so far every reproductive system requires the presence of standing water. The ocean, though, the ocean is a different story, full of life and living. If you go by the rock record, the most common form of life is the stromatolite, a colony of green plant-microbes that secrete stone in distinctive whorling patterns. (As always, history most remembers those who write history). They flourish in shallow and sunlight-touched waters, gradually constructing ranges of gently rolling hills that are each sometimes meters across. Larger (but still microscopic) Eukaryotes wander through these hills like grazing deer. Or maybe like wolves. Unlike the helpful stromatolites themselves, most of these creatures rarely leave informative fossils, so it’s hard to say how complicated the ecosystem actually is. Certainly it’s a rich and dynamic biological scene, but to my mind it always feels just a little bit like Hobbiton. Gentle with a hidden strength, simple enough and flexible enough to endure. The official name for this time period is the Proterozoic, but when geologists think nobody is listening, we sometimes call it the Boring Billion.
But here in the closing years of that era, as temperatures thaw and the coastlines unwind and the atmosphere fills with oxygen, it’s starting to get a bit more interesting. Big. Some of the Eukaryotes are behaving oddly, bunching up into colonies that are a bit better at finding nutrients or sunlight. Often, they give up roaming and gather together in big fronds, anchoring themselves to the seafloor and spreading the sail out broadside to catch as much nutrient-rich ocean water as possible as it flows by. Sporifera, the sponges, use a different strategy, pulsing flagella inwards to create little artificial currents that channel nutrients towards themselves. Cnidaria, the jellyfish, actually lift off from the ocean floor altogether and drift through the water looking for scraps among the drifting plankton. The cells in these colonies are starting to take on specialized roles, this one binding the group to the soil, that one a tentacle, but their overall simplicity falls well short of what we usually mean by words like ‘animal.’
Despite these omens, it’s all rather sedate. Not much seems to change day to day, millenium to millenium. You’d be forgiven for checking your watch a few times every epoch. You might even take a nap, stay in bed for a few million years and relax. Maybe twenty million if you hit the snooze button. But when you wake up, you’ll be in for a shock.
Somebody broke it. They scourged the Shire. In just a few million years, no more than a wink of geological time, everything has changed. Many of those ever-green stromatolite hills have been stripped bare, and the ones that remain are going fast. Giant monsters prowl through the wreckage, hunting for anything made out of complex organics, hunting each other. There is a bewildering array of new lifeforms. Some of them are covered in spikes, others with odd numbers of eyes and strange probing tentacle-mouths. All of them have elaborate organ systems, strange tissue masses that express themselves in radically different, interlocking ways, despite having the same genome. The rise in diversity, and in disparity, is unequaled by any other moment in Earth’s history. Something like half of all 21st century animal phyla trace their origins back to that brief moment of generation. Take all the creative power of the last five hundred million years of animal evolution, compress it down to a fraction of a geological instant- that’s the power of the Cambrian Explosion. In less than twenty million years, there are molluscs squirming like modern sea urchins, echinoderms clinging to rocks like modern starfish. Even the trilobite, that ancient symbol of ancient life, suddenly appears here fully-formed. And then, swimming through the open waters, you’d see the most surprising thing of all: one of them has a brain. Animalia Bilateria Chordata, the chordate.
This is it, you see. The moment of encephalization, the moment that the cosmos wakes up.
There was the Earth, a giant rock drifting quietly through space. And one day it just spontaneously grows a brain, for no damn reason at all. What kind of rock does that, exactly? What kind of universe?
No, really, why did this happen? What was the mechanism, the bridge that takes us from the boring billion to the era of minds and monsters?
I don’t know.
Really, I don’t. There are hypotheses, sure. A few scraps of almost-understanding. And plenty of guesses, some of them really good. But the stone can only tell us so much, and it was such a very long time ago. The all-important middle of our story, the one that gives us the first moments of emerging biological consciousness, has not yet been recovered.
Can it be recovered, even in principle? Harder question.
Darwin, poor Charles Darwin, floundered on these shoals with the rest of us. For him, the only explanation was that we must be missing some huge fraction of the rock, so that it only seems like they all show up at once:
I cannot doubt that all the Silurian trilobites have descended from some one crustacean, which must have lived long before the Silurian age....Consequently, if my theory be true, it is indisputable that before the lowest Silurian strata was deposited, long periods elapsed, as long as, or probably longer than, the whole interval from the Silurian to the present day.....The case must at present remain inexplicable; and may be truely urged as a valid argument against the views here entertained.
He was wrong. Isotopic dating methods have since confirmed that there is no gap in the rock between the Proterozoic and the Cambrian, no space of hundreds of millions of years for us to move gradually from one extreme to the other. (In this case, we’re also early enough in the history of geology that precise chronology was still ambiguous- that’s why Charles refers here to the Silurian, a later era.)
And so, as the man says, the Cambrian Explosion may be truly urged as a valid argument against the views which Mr. Darwin entertained. What’s at stake here is not just the question of how this process of encephalization first occurred, but also why our most foundational biological theories fail so spectacularly to anticipate it. Are we even asking the right questions?
There’s a slightly more modern version of Darwin’s first tentative attempts at a patch, one that might explain why so many fossils would be missing from the pre-Cambrian record. If it’s true, there never was a Cambrian Explosion, the paragraphs above don’t correspond to the world as it was, and the Earth’s encephalization took place very gradually, over eons. A relief in some ways (our theory of evolution remains sound), but a tragedy in others (since we’d probably never be able to peer at what might be the most important moment in the history of life).
Here’s a different story:
By the end of the Proterozoic, there has been a thriving multicellular ecosystem for hundreds of millions of years, full of complex animals in a thriving dance of grazing and predation, reproduction and survival, gradually expanding and exploring the space of possible forms. But these organisms are all soft-bodied, with no bones or shells or rigid frameworks of any kind. When they die, their bodies rot immediately, leaving no trace for paleontologists to find. But remember when I mentioned that one of the continents is breaking up? All this new coastline, and new weathering, bring new minerals into the oceans. Suddenly, the system is flooded with dissolved calcium salts, ready and able to be incorporated into bones and other biological machinery. Naturally, a number of different well-established species take advantage of this. Skeletons, spines, and shells become very popular. And when they die, they’re preserved, some for a long enough for a poor foolish scientist to stumble across. And from our perspective, there’s a bright line with animals on only one side of it.
This is a possibility that we should take very seriously. A good chunk of the scientific community certainly does.
One of the things that swayed these scientists is a method of using ‘molecular clocks’ to learn from living genetic sequences as if they were a sort of fossil. The trick is this: look for genetic sequences of a very specific sort, those that change randomly, protected from the directionality of evolutionary selection pressures, and which have been doing so at a slow and steady rate for hundreds of millions of years. (In practice, certain structural elements of hemoglobin work well.) Sequence this area in two very different animal species for which we have already discovered the age of their last common ancestor- and by measuring the difference, we can precisely calibrate the rate of change.
Then, all you have to do is apply this same procedure to, say, a mollusc and a chordate. If the Cambrian Explosion happened like we say it happened, you should get about 500 million years of genetic drift.
Actual answer? 800 million. This gives us a full 300 million years to go from primitive sponges to trilobites, an eminently reasonable wait.
Still, I feel a little squidgy about this line of reasoning. Molecular clocks are a dangerously fragile tool, for one. And remember also that the 800 million year figure would have complex animals surviving the snowball Earth periods, where a frozen surface layer prevented gas exchange between the oceans and atmosphere; those oceans were anoxic, lacking any oxygen for the animals to breathe. But my main objection is actually a bit simpler: no ichnofossils.
An ichnofossil, or ‘trace fossil’ is just any biological remnant that doesn’t involve the actual biological thing itself. Footprints, burrows, etcetera. They’re kind of a pain to find- trace fossil hunting famously takes place at dawn so that the shadows throw your field area into sharper relief- but preserve a lot of valuable information that actual bones do not. Consider the famous archaeopteryx fossil that preserved, not just the skeleton of a dinosaur like so many others, but also the clear imprint of its feathers. That is the weight of an ichnofossil.
As you might imagine, we’ve really scoured the strata around the Cambrian transition as best we can. For the simpler organisms, we do find a number of trace fossils. That’s why I can tell you about the sponges and jellyfish; we have the impressions they left in the ground as they died. We even have enough detail to learn astonishing things like: “Modern jellyfish are quadrilaterally symmetrical, a circle with for identical quadrants. But when they first emerged, at least some jellyfish were trilaterally symmetrical, a circle with three identical parts, instead.”
But aside from the jellyfish and those weird asymmetrical fronds, we genuinely don’t see much. No burrows, no tracks, no trilobyte-shaped impressions in soft mud. So if there was this huge dynamic ecosystem, why didn’t it leave any footprints, even though conditions were favorable enough for such things that we can find the final resting place of a 600 million year old jellyfish? The most provocative traces we find are narrow (millimeter) horizontal trails or burrows, just 2-3 million years before the Cambrian. That’s the only prelude we’ve found so far, and it’s a brief one.
I don’t think this is strong enough to break the missing-calcium theory outright; if nothing else, it’s an argument from absence. But I do think it’s a pretty important nail in the coffin, and a reason for skepticism. And so- we still have at least a pretty good reason to think that the Cambrian Explosion is a real thing, and that our search for the middle part of that story is not foolish.
I can think of a few places to look.
Perhaps the answer we’re looking for is, ‘oxygen’. As we observed earlier, this is a period in Earth’s history in which oxygen had increased to near-modern levels somewhat recently.
Oxygen isn’t quite an absolute prerequisite for complex food webs, but it’s pretty close. Animal metabolism derives energy from a flow of electrons as they move from one molecule to another, almost but not entirely unlike a water wheel powered by a flowing river. The faster that water flows, the more energetic a machine you can power. To do that, you need a source of electrons, preferably in some dense high-energy package like sugar. But you also need to provide a grounding, a place for them to go, something that pulls in electrons as hard as possible. That is, our power system also needs an electron sink. Oxygen does this job remarkably well- thus, as animals, we survive by eating and breathing.
Without oxygen, you need to rely on a less energetic sink. There are species of iron that work alright, and there are some weird hacks available with hydrogen, but the machines that you can make with these power sources are only so impressive. If all you have is iron as an electron sink, you can probably manage grazing pretty well, and you can buy some wiggle room as long as you’re microscopic. But something high-energy like multicellular predation is a big ask. At least today, whenever we see low-oxygen environments in the deep ocean and so on, we always see a corresponding reduction in food web complexity and species diversity.
So could a quantitative increase in oxygen levels have produced a qualitative change in the structure of animal life? Possibly. But there was a significant lag between the rise of oxygen and the rise of animal complexity, almost a hundred million years. It begs the question, why not sooner? Oxygen probably had something to do with this, but it seems more like a prerequisite than a cause- and we have yet to account for the morphological changes. Why not just use the energy to make bigger fronds? Fronds for miles, fronds to span mountains, fronds beyond the wildest dreams of frondkind!
Think back to those horizontal burrows that show up in the very last days of the Proterozoic. There are a couple ways that this is really, really exciting. First, it implies that sensation probably was starting to get concentrated on one end of the animal, or at least one side- taste, smell, even vision clustered around a single area. That feels suspiciously like we’re starting to get nerve clusters that you can almost call a brain.
But also, significantly for our purposes, this means that animals might be starting to develop sophisticated hox genes.
The hox genes are, more or less, the standard library for the structural assembly of animal bodies. It’s a DNA-modifying type of gene, one that activates specific other regions of DNA during early development. Need a leg? Activate the ‘leg’ hox gene, and that will start a huge cascade of related processes that make a torso segment that contains a couple legs plus all the necessary hip-joints and such in a somewhat standardized way.** This makes it easier than you’d think to adapt animal body plans on the fly; rather than reinventing legs from the ground up every time you want to adapt from quadruped to hexaped, you can just have a mutation that calls the hox gene two more times. It is also a primary mechanism of directionality, in which animal bodies have a clear orientation with a front and back.
As you might imagine, animals almost never survive a mutation to the hox genes proper, let alone thrive and speciate. It tends to mean that you get born without a head or something. So, both humans and house flies tend to have a very similar set. That holds true across the entire animal kingdom, with only a few exceptions- hox genes are frozen in time, proportionate with their importance. Care to guess which types of animal lack fully formed hox genes?
That’s right: sponges and jellyfish. Even those have a rough proto-hox thing going on, but it’s a far cry short of ours.
It goes without saying that this will have implications for species diversity in the animal kingdom. Adaptation isn’t as easy as building an animal out of legos, but it has at least gotten a lot easier. Animals can experiment with body shapes in more radical ways, and be successful more often when they do so- they’re now exploring a much larger space of possibilities.
We’re starting to piece together a rough framework here, in which the rise of oxygen and the slow development of meta-genomic advantages work together to provide space for a more dynamic ecosystem, but is that enough to make sense of something as dramatic and surprising as the Cambrain Explosion? It’s a start! But let’s see if I can’t make it a little more complicated.
**I am lying harder than usual right now. This is complicated and I am not a geneticist. Please never believe me about anything.
We’ve got the oxygen levels needed for large-scale predation and multilayered food webs, and we’ve got the genetic toolkit to move quickly through different animal shapes as evolutionary pressures come down on us, sure. But again, why so many, and why all at once? What splintered the animal kingdom so thoroughly, and spread the shards of it so widely? Before the Explosion, we apparently had two or three general body plans, each with an accompanying niche. Afterwards- everything else.
Personally, my favorite answer to that question is, ‘eyeballs’.
Maybe also noses and ears, I guess. But those are a little less directional, and we’re pretty sure that eyes did in fact develop around this time.
Either way, what I really mean is, ‘long distance detection of nutrients’. This synchronizes nicely with the directional motion that shows up around this time, the ‘front and back’ and ‘top and bottom’ innovations that allow an animal to see food and then move towards it.
What I really really mean is, ‘predation.’
On the one hand, this makes it a lot more viable to be one step up in the food web. A blind predator is going to have some trouble; the ability to see and pursue opens up a new niche. But only the one new niche, so that alone still doesn’t explain the riotous diversity we new encounter.
Consider the criteria you must satisfy to be a successful precambrian animal. You’re going to need to absorb as many complex carbon molecules as possible. You’re going to have to solve the reproduction problem somehow. And you’re going to have to be structurally sound, rather than collapsing under your own weight or something. It’s fairly simple, mostly revolving around being able to access as much seawater as possible so you can filter organics out of it.
And in fact, the three major solutions we see in the fossil record are, “have a large broadside and catch water as it moves by”, “actively pump water towards yourself”, and “move quickly through the water.” When you think about it, that’s a fairly exhaustive list. Every one of these forms is clearly exploring ways to have physical contact with as much seawater as possible, and almost nothing else. And there are only so many ways to be the best at that job.
But now let’s add another criterion: “Don’t get eaten.”
It’s not just that this is a fairly flexible and ambiguous constraint. It’s that we’ve lost our chance to monomaniacally focus on the one strategy of ‘contact lots of water’. And when you have to balance different needs that are in competition, there are a few different trends that tend to emerge from that process. First, none of your strategies will be quite as successful as they were when you didn’t have to make any tradeoffs. And second, there will tend to be a number of different ways to balance those needs, each of which is roughly as effective as the others.
This is a really important but kind of abstract point, so I’m gonna hammer at it for a little bit. As the number of constraints increases, the number of different ‘best’ solutions will tend to increase combinatorically. If you’re a proto-sponge, and you’re suddenly under all this pressure from being hunted, you might try covering yourself in spikes, or borrowing into the mud, or growing in nasty inhospitable places where predators don’t want to go, or just picking up a certain amount of mobility. But remember, you still have to worry about filter-feeding yourself. And the proto-jellyfish has a similar number of different options, and the weird frond-looking things too. The net result is that we end up with a really huge number of equally viable body plans, rather than the two or three that are the winners of a simpler problem.
Basically, the idea here is that predation was such a radical change to the environment that it fractured a small number of deep ecological niches in to a large number of somewhat shallower niches. The radical speciation of the Cambrian Explosion is simply the natural result of those niches being explored and filled.
And somehow, from that crucible, the brain.
We aren’t there yet. We know that long-distance senses like vision probably would have given rise to directional motion and directional anatomy, and that the combined needs of directional motion and of sensation managed to concentrate nerves in one particular region of the animal. Maybe we can find out more with detailed genetic analyses, or maybe we’ll hit the motherload with some hugely important fossil discovery. Hard to say, but I doubt we’re done yet.
Analogies are always perilous, and I will personally get very annoyed at anyone who tries to make a political metaphor out of this or something. But in more general terms, it’s worth taking a step back and thinking about what the Cambrain Explosion says about our universe. The run up to this apocalypse seem to have included at least a few generalizable events:
The first is a deep well of underexploited potential energy. Oxygen, in our case, meant that the ratio between the possible and the actual was somewhat larger than usual. Finding a metabolic pathway to exploit oxygen was difficult, but once the blueprint existed, the machinery itself wasn’t particularly difficult.
The second is that new layers of useful abstraction emerged, which allowed innovation and conceptual mobility on larger scales than had previously existed. We accelerated faster through our search space.
These do not themselves lead directly to an abundance of new forms. Rather, the fits and starts of the early successes present new challenges to existing institutions. Those organizations that successfully adapt to the new environment are still constrained by the additional complexity, and windows open for entirely novel forms.
And that’s about as much of a metaphor as I’m willing to make. Still, it probably influences a lot of the way I think about, e.g., Silicon Valley. And it’s probably important to remember that any Singularity would be the second intelligence explosion that the Earth has gone through.
The stromatolites are still around, by the way. They never conquered the world again, but they still build their little hills in hypersaline and hyperthermal waters, places where animal life can’t survive. There’s a nice cluster of them in the Bahamas, which makes for some nice field expeditions for geobiologists. It’s not a bad retirement.