Work in Progress: Whither the building blocks of Life?

Friday 12 August 2022

I am now finished with the NoRCEL conference, which ran for three days, August 9, 10, and 11. The conference was at the University of St. Andrews in St. Andrews, Scotland. St. Andrews is a charming little town, with collegiate gothic buildings, a small harbor, and a surprisingly large swimming beach. After the conference wrapped up on its third day it was quite sunny, so I walked to the beach and swam in the North Sea. It was cold. While there were a lot of people at the beach, I noticed that no one stayed in the water for very long.

The conference was well managed, and the food was the best of any conference I have attended. Lunch was provided, and on the second day there was a dinner at a local restaurant. Speaking slots were set up for 20 minutes per speaker, with 10 minutes for discussion. This worked out quite well, since if someone went over their time for five minutes it didn’t impact the following speaker. There was the inevitable last minute swapping around of speakers for those who didn’t show or weren’t ready, but I don’t want to give an impression that it was at all chaotic. On the contrary, it was well run, and the schedule mostly came off like clockwork.

When I left the US my presentation was not yet complete. I worked on it while I stayed in Edinburgh and again at St. Andrews, and finished up my slides the evening prior to my presentation. Given that kind of last minute completion, it could have appeared hurried and somewhat incoherent, but I was reasonably pleased with the outcome. Of course, I could do better. As soon as I finish with a presentation I think of all the ways that I could improve it, and mostly I haven’t had a chance to repeat a presentation so I don’t get the opportunity to see if I actually could do better. However, this November I will have a chance to do my last year’s NoRCEL presentation again, “How many branches are there on the tree of life?” This will be another new experience, and there is always something to be learned from new experiences.

Also, the research program that I outlined in my current presentation was essentially aspirational: we don’t yet have the technology to do what I have suggested — the seriation of exoplanet chemospheres — but in order to be able to do this in the future, there is a significant amount of theoretical work that needs to be done. I commented in my presentation that currently recognized types of planets are little more than folk planetology, being both anthropocentric and geocentric. We speak of “hot Jupiters” and “super Earths” and “mini-Neptunes,” which makes it abundantly clear that we take our own solar system as normative, and extrapolate the planets with which we are familiar in order to account for recently discovered planets. (We don’t have a word to name prejudices that arise from the peculiarities of our own planetary system, which is why I called this folk planetology “anthropocentric” and “geocentric” although those terms aren’t quite right; and, for obvious reasons, “heliocentric” won’t work for this purpose either.)

So the first order of business is to produce a classification of planets, hopefully based on quantitative concepts. There is an immediate tension here, because ultimately the goal is astrobiological and origins of life research, so we want to keep this in mind, but our initial taxonomy of planets is simply to give us enough categories that we can compare planets of different ages in different planetary systems. The tension is that there is a temptation to bring into an initial classification of planets many or most of the factors assumed to be implicated in the origins of life, but this would pre-judge the outcome. We need a planetary taxonomy first, with just a few parameters, such as we have for stars, which I mentioned in my talk. We can easily classify the sun (G2V) according to taxonomic standards that apply to every other star in the universe. We don’t have anything like this for planets, but we need something like this if planetary science to be truly scientific and not rely on folk taxonomies.

Another way in which my research program is aspirational is that we cannot yet date stars, planets, and planetary systems with the degree of precision necessary. Given the rapid progress of both technology and technique in observation, I am confident that we can begin to approach the requisite degree of dating precision when we begin to come into the technology of exoplanet atmospheric spectroscopy. However, even at our present level of precision, we can easily distinguish between proto-planets still forming in a protoplanetary disk, a fully formed planet (a mature planet, in a sense), and necroplanets breaking up and impacting on the white dwarf that was once their sun.

The more precisely we can identify any given star (which is a function of the established stellar classification system), and the better our models are for predicting the evolution of stars, the more precisely we can date a star, and if we can date a star, we can date its planets. There are probably other ways to approach the question. If we could resolve smaller bodies in a given planetary solar system, we could judge the extent that planets have cleared their orbits of other bodies (which is the definition of a planet that excludes Pluto as a planet), assuming that the older a planetary system is, the more clearing has taken place. No doubt there are other methods that I simply haven’t thought of, but which others will develop (or already have developed).

One of the questions asked after my presentation is what level of resolution we would need in time in order to carry out my research program. I responded that tens of millions of years (10 Ma.) would probably be sufficient. At this scale, we would have 100 demarcations in each Ga. Of course, we could still learn a lot at a resolution of 100 Ma., which would give us ten demarcations in each Ga. Given the stable life span of a G class star is about 10–12 Ga., a resolution of 100 Ma. would give us 100 demarcations of development over the lifetime of a planet, and that might well be a workable level of analysis. More data than that might be too much. Scientific abstractions must strive for a “just right” degree of analysis that is still possible to generalize across multiple instantiations, but is sufficiently fine-grained to meaningfully differentiate instances that should be differentiated.

The more I thought about this, and also the more I worked on my presentation, the more I realized that temporal resolution is one of the most important aspects of origins of life and astrobiological research. Since we get all of our evidence of life from Earth, the developmental trajectory of life on Earth is, for us, the norm, but we do not know the extent to which the development of life on Earth is typical or atypical. According to the principle of mediocrity, life on Earth and its development should be typical, but, depending on the parameters adopted, variations across typical developmental patterns could still be significant. The more typical that life on Earth is, the more Earth exemplifies the principle of mediocrity, the finer resolution in time scales would be required to differentiate developments across multiple living worlds.

At present there is no consensus on even the most basic temporal parameters: how long does it take for life to begin? How should we identify the proto-life stage? (How we identify proto-life will determine how long this period endures.) How long does it take until the eukaryotic cell evolves? How long to multi-cellularity? An example I used in the Q&A after my presentation was that we don’t know if the “boring billion” was a necessary developmental stage in life on Earth — perhaps life requires a long, shallow curve of development before complexity begins to increase exponentially — or whether this was a kind of pause in development not experienced on other living worlds. Or it could be the case that, on other living worlds, there is a boring two billion, or three billion.

These billions of years really count for something. Think of it in a SETI context. Let’s say that life appears on three (or more) distinct planets at about the same time. One world has no “boring billion,” one has a “boring billion” like on Earth, and another has a “boring two billion.” Suppose these are all worlds that orbit a G class star, so that 10–12 Ga. of stable main sequence development can be expected. If civilizations eventually appeared on all three worlds and began transmitting SETI beacons, these civilizations would be separated from each other by a billion years — a separation in time probably sufficient to prevent their communication, or even an awareness of each other — despite life on these worlds getting started at about the same time. Understood in this way, even a “boring million” would make a big difference. For civilizations separated from each other by a million years to communicate, or simply to be aware of each other, the earlier civilization would have to work out how to build a beacon that could endure for a million years. Given that most of our technologies degrade within a few decades, this would be a remarkable achievement. Some kind of automated maintenance would be necessary, but who would maintain the maintainers?

Now, if these civilizations were separated by a million light years, then the earlier civilization could build a beacon that was then received by the civilization appearing a million years later, and in terms of one-directional communication, these civilizations would be “contemporaries” in a sense. We can easily imagine something like this in terms of our local cluster, since the Andromeda galaxy is about two million light years away. A civilization in Andromeda could build a beacon and transmit a signal in the direction of the Milky Way, and two million years later, any civilizations that happened to be “online” and listening at that time would be able to receive that signal.

In any case, returning to my earlier point about reconstructing the developmental sequences of exoplanets, this process may itself furnish some of the timing. Once we have the capacity to do exoplanet atmospheric spectroscopy, the seriation of chemospheric signatures should be relatively straight-forward, albeit rather complex. We should be able to note (on different worlds, in different parts of the sky, in distinct stages of development) the appearance of an atmospheric gas, or a suite of gases, the growth in abundance of gases and changes in the composition, and the eventual tailing off of a given gas or gases. We may be able to estimate the time it takes for an entire planetary atmosphere to cycle through various chemical compositions, and as we are able to reference an increasing number of particular examples, we may be able to refine this chronology and use it to date the development of newly discovered worlds.

Ward and Brownlee in their well-known book Rare Earth not only argued that life at the degree of complexity found on Earth is rare, they also argued that simple life may be very common in the universe. There may be countless biospheres that are the equivalent of archaea and bacteria, with little development beyond microbial mats. As such simple biospheres cycle through changes on billion year time scales, they may betray characteristic biosignatures that develop over time, with their own mass extinction events and recoveries from these events. Even worlds without “advanced” life will have much to teach us about origins of life; perhaps the ultimate question will be not the origins of life, but the origins of complex life.

In our focus on the “building blocks” of life (that persistent metaphor perhaps deserves more attention), we tend to gloss over the possibility that the building blocks of life present on Earth, which resulted in our terrestrial biosphere, on another world may be the building blocks of a life-peer that is only barely recognizable as life. It is entirely possible that life (in the most generic sense of the term) is a cosmological imperative, but that terrestrial life, life as we know it, is an outlier in the grand scheme of things. While this violates the principle of mediocrity, it must be acknowledged as one possibility, and if the most common forms of life in the universe are not life as we know it, we should not expect that these other forms of life converge on the same post-biotic emergent complexities as those we know on Earth.

It is (again) entirely possible that there is convergent evolution of forms of emergent complexity consequent upon life, so that even if the biochemistry of life is different among living worlds, and even if the evolutionary process differed from that of Earth (and there are so many radically contingent events that the evolutionary history of life must be distinctive to each living world), that sensation, consciousness, and intelligence, and social institutions predicated upon intelligence, remain a possibility supervening upon other forms of life. Indeed, we could identify this as one of the most important unstated postulates of SETI. I need to come up with a good name for this, something better than “convergence on higher forms of emergent complexity postulate.” Perhaps “SETI Convergence Postulate” (SCP) would do the trick, but I’ll have to think about this.

One thing that struck me time and again during the conference is that origins of life research forces those in this research to think in astrobiological terms, even if this was not the intention or the interest. Reaching the necessary level of generality and scientific abstraction to study the origins of life becomes an extrapolation that converges upon astrobiology. This is an interesting observation about how scientific abstraction not only allows for logically powerful formulations that can be used for prediction, but also how abstractions play a role in directing scientific thought along particular pathways. And if a pathway has already been blazed by others — a trail cleared, the basic concept formation has been taken care of, and a direction through the wilderness has been shown — then it is all the easier to borrow from this stock of concepts to fill out one’s own work.

All scientific abstraction falsifies the context from which it is drawn, but we accept this falsification as the price of tractable analysis and effective prediction. That from which we abstract when we abstract may be taken up by other sciences (which in turn abstract from that which we have taken up as our research focus), but it also may be set aside, as a kind of casualty of science. I often point out how science needs to expand, not only in the sense of taking up previously neglected phenomena, but more in the sense of making scientific method more comprehensive, which is as much as saying that we need to revisit the casualties of science and see if we can incorporate neglected aspects of the world into scientific thought.