Deconvoluting the past

[PB666]

I moved this out of the life thread, I think its a topic that deserves consideration in its own regard, considering the discussion in the New Horizons and Life thread.

So basically any field that has paleo in front of it studies past, events. One of the key features of paleo studies is that scientist and lay people alike confuse markers for past events for the past events. For example we here that a particular fossil was an ancestor of all birds. That may be true but,

-fossils are not ancestors of anything, and they represent transformed remains of durable remains

-the likelihood that any individual plucked is the significant common ancestor of an entire taxa is unlikely, particularly given that most species that have existed are extinct, meaning specifically most lineages have gone extinct including many small branches off of lineages that are extant.

There have been numerous examples where something that was claimed to be a missing link ends up being a side branch in evolution, possibly through leaky species barriers contributed some inheritable material, but for the purposes of what the purporting authors stated, a largely falsified argument.

So with this out of the way, we have to say that absolute declarations about past events with no historic corrobaration are troubling and best treated in a probabilisitic framework.

OK, so lets do a thought experiment.

Suppose we are alien spectators of life as it evolved on earth between 4.0 and 3.3 billion years ago, before the cell that appears to be the last common ancestor of all life, lived.

So we are going to make one assumption that once this appears all other life is at a distinct disadvantage, through a process of drift and selection, over time their numbers dwindle and all but dissappear (delegated to the deep ocean sedimentary rock and the like).

What might we see, surprising we would see life, but the properties of this life however inefficient would be highly diverse. So now I am going to re-introduce the game.

In this game there are 250 segments that need to be passed, and the pretext to the game is that when something cross the 250 segment barrier you take a screenshot

and the create a collage of screenshots (the floor is mutable, and I generally switch between jupiters and moon gravity to create variation). You keep playing a round until something falls off the map and repeat. Fiddle with the conditions and eventually you'll have one that flys off the end.

http://www.whiletrue.it/genetic_3-wheelers/

What you can see is that in each set of runs you get a different structure running off the end. On one mutable floor one structure may work and another may not but if the floor changes another structure may have success. This is actually emulating the conditions of early earth, but you could imagine how different an exoplanet might be, of for example life living in some hot vent on a Saturn moon or pluto (if such life exists). The form that eventually evolves is dependent on mineral content of the water (assuming the substrate is water), bioavailable organics, and forms of energy (heat, chemical, light, ultraviolet light).

So what is the answer to the question, for example what might past life looked like on earth.

The answer is that with experimentation and examination of other evidence of organic activity we can come up with a range of possibilities (bioneogenesis experiments in a laboratory), and try to correct that range with more information, such as evidence from other worlds. But one has to be careful with absolute statements like life has to use DNA, 20 amino acids, or ATP. Thirdly I should point out, that we should not be absolutely convinced that all bioneogenesis occurred here on earth, these seeds of life may have occurred elsewhere and serendipitously ended up here on Earth, and if that is also true then we have a major missing variable on the forces that acted on precellular life.

The problem of deconvolution the past is not strictly limited to life, in fact, the life issue is often secondary to the geology issues. Almost all of our dating techniques are not dependent on biological remains, but isotope levels of common metals (potassium) and radioactive elements and for things younger than 50,000 years, carbon-14. And its not a trivial argument because many fossil studies have gotten the geological dating wrong and had to be later revised. The geological context is very important for determining the redox state of the environment and bioavailable nutrients (iron for example becomes much less bioavailable as redox potential in fluids rise). Molecular paleontology has also had its fair share of date miscalculations.

Deconvolution the past is not just about one set of observations, but trying to collect and mesh as many observations as possible. Getting out of the 3.3 billion year old box is a license to explore possibilities, but some of the trails lead off into an abyss, some are traps and some are hidden from view. One thing we tend not to think about is one theory of early earth is that a rather large protoplanet might have slammed into earth the dust of which created our moon and is particularly enriched on earths surface.