Life Without Hard Steps?

Many kinds of things could have happened in history. More precisely, life can take many kinds of steps in its long path from simple dead matter to our level, or to even higher “grabby” levels of being very visible from far away:

• Enable - Some key context had to switch to an enabling condition. 

• Delay - Some process had to play out for a roughly fixed time duration.

• Try-Try - The system searched in a vast space for a rare solution, with a constant rate per time of success.

• Try-Once - The system fell into some state and stayed there forever. There’s only a chance it was the right state.

Some noteworthy facts about this history on Earth, now ~4.5Gyr old

1. Life appeared on Earth relatively soon (~0.4Gyr) after it was possible here.

2. In ~1Gyr ordinary life will no longer be possible on Earth.

3. The durations between big revolutionary changes in the nature of Earth life are roughly similar to each other and to the durations of facts #1,2.

4. Our empty universe says the chance of grabby life per planet must be crazy low. (I estimate it appears roughly once per million galaxies.)

5. If we get grabby, we’ll probably do so in far less time (<<1Myr) than the duration between prior revolutionary life changes.

The hard steps model of Earth history, first described by Brandon Carter in ‘83, (and similar to a standard model of cancer from the 50’s) tries to explain Earth history up to now as a sequence of only hard (= long expected duration) try-try steps. This model, with roughly 3-9 steps, does a reasonable job of explaining facts #1,2,3,4. (It also works with fewer hard steps if a fraction of the time was taken up by enable or delay time.) However fact 5 seems best explained by our now facing only short delay and try-once steps; a hard try-try step would just take too long, and we don’t seem to need context enabled.

A new paper, “A reassessment of the "hard-steps" model for the evolution of intelligent life”, by Daniel Mills, Jennifer Macalady, Adam Frank, and Jason Wright, suggests instead explaining Earth life history in terms of enabling and delay steps: 

We advance a potential alternative model where there are no hard steps, and evolutionary novelties (or singularities) required for human origins can be explained via mechanisms outside of intrinsic improbability. … the “delay” in the appearance of humans can be best explained through the sequential opening of new global environmental windows of habitability over Earth history, with humanity arising relatively quickly once the right conditions were established.

They go into detail on many of the revolutionary steps in Earth history, saying how they might have plausibly waited for context such as rising oxygen levels. And I grant that each known revolutionary step might plausibly have arisen from enabling or delay steps; we just don’t know enough to say otherwise.

However, these authors don’t even discuss how to explain facts #1-4 without hard try-try steps. (A sequence of easy try-try steps looks a lot like a delay step.) To explain fact #4 without any hard try-try steps, they will need to invoke some very large try-once steps. That might be possible, but they should at least discuss plausible candidates.

Furthermore, they should also try to explain fact 3, that the revolutionary steps seem roughly equally distributed in Earth history. And it seems especially hard to explain, without many hard steps, that the typical duration between these steps is also roughly the same as the duration now left in Earth’s life window. I don’t see how that fact could be anything other than a remarkable coincidence, without hard try-try steps. (And if there was just one hard try-try step, then Carter’s hard step model applies to Earth’s life window minus the times needed for enabling and delay steps.)

By the way, these authors err in summarizing me:

Hanson proposed that there exists a “Great Filter”: some single, very unlikely step on the road to the development of Galaxy-spanning spacefaring life that prevents it from arising, despite optimistic estimates one might calculate from the Drake Equation.

No, I proposed that a set of steps add up to a large filter, but don’t assume the entire filter is concentred into a single step.

They also make this dig:

The popular appeal of the Great Filter framework is that it allows one to consider whether the hypothetical (and dubiously justified) Great Filter is “ahead of us” or “behind us.”

I don’t understand what they see as “dubious” here.