Hacker News Reader

Comment by marginalia_nu | Hacker News Reader

It's fair to characterize life as a process that locally decreases entropy. Non-living processes tend to increase entropy. Even if you have a rock that looks exactly like a human being, say one of Phidias' statues, it will decay until it looks like every other rock given enough time and exposure to the elements. It rarely goes the other way around, the wind does not sculpt David out of marble.

It's probably not a sufficient criterion for life, but it's at least necessary to avoid granting life to a wildfire or some other process most would not consider living.

View on HN

All true, but equally important is that life is rapidly increasing entropy in a larger, but also local, frame. It’s like a fire. So, in a closed system, living things must exhaust their source of energy and reach equilibrium with its surroundings.

>Even if you have a rock that looks exactly like a human being, say one of Phidias' statues, it will decay until it looks like every other rock given enough time and exposure to the elements.

The same is true of living things. Given enough time, anything living is reduced to a pile of oxides.

Some physical processes decrease entropy locally, sometimes by incredible amounts. That's how we have the sun. What we really take from the sun is not energy, but its stored negative entropy; energy is just the medium to move it around.

Right, but it must be granted that the moment of death is a point where this decay accelerates greatly.

But qualifying any particular moment as "death" already implies that we've granted the thing in question the quality of being alive at some point, so it's not a useful way to distinguish living from non-living.

But that is how we talk about living things.

A man is alive, has a fatal heart attack, then he is not alive.

I don't think anyone would argue he was never alive, or is still alive after having gone cold.

Studying this transition from life to non-life, or vice versa should be most informative in distinguishing the two states.

Sure, but the question is not about distinguishing living from dead things, but about distinguishing currently and formerly living things from things that were never alive. More specifically, about distinguishing them structurally, not comparatively. If "death" is the point where the decay into entropy of something greatly accelerates then hypothetically we could say a rock "dies" when it's melted, or alternatively we could say that it doesn't "die" because it was never alive. The distinction does not lie on the thing itself, but on which things we assign the property of being able to "die".

The argument has never been that this is a sufficient criterion for life, but rather that it is necessary and not merely accidental the same way you can be alive and not have legs.

There does not to my knowledge exist things that are considered living that don't have this property of locally decreasing entropy.

Fair enough. It is indeed true that living things seek to maintain their entropy low.

For the majority of life existence as we know it (meaning, a proton gradient separated by a membrane), living systems were immortal.

Some external circumstances could change reducing them to a pile of oxides, but the systems themselves were perfectly self-perpetuating -- i.e. time-independent, or put another way 'given enough time' is not sufficient.

Of course, before the heat-death of the universe entropy-reducing systems will break down, but that's not an interesting observation.

> Some physical processes decrease entropy locally, sometimes by incredible amounts.

I think that's the more interesting observation. How to qualify life beyond only 'local entropy reduction systems'.

My uninformed opinion is that it is both 'entropy-reducing' and 'context-dependent' systems that should be considered 'alive' (not sure though that this is sufficient). Dependence on their context means that those systems will have a feedback-loop with their environment, allowing for example non-linear causality and thus complexity to emerge.

A star radiating away their stored negative entropy is an all give and no take, no dependency, full autonomy system. Of course they can still be impacted by external effects, but sustaining their own internal processes does not require such inputs, they are incidental to them emitting away their stored energy. Once started, their schedule (when they will deplete themselves, collapse, etc) is known and predictable, thus not alive.

> Of course, before the heat-death of the universe entropy-reducing systems will break down, but that's not an interesting observation.

Excuse the snark, but it's exactly as interesting as saying that some living systems are immortal until they're no longer immortal. If a protocell or a virus stops being alive (if we grant that it's alive) when it falls into a fire then it's not immortal just because it has no metabolism.

>Once started, their schedule (when they will deplete themselves, collapse, etc) is known and predictable, thus not alive.

Actually, no. For example, a star could hypothetically fall into another much larger than it and prolong its life, if we were to say that it lives. The only reason it's predictable is because a star can't move on its own to hunt weaker stars, and because (at least at this time in the history of the universe) stars are far enough apart that they rarely fall into each other. If that was a common occurrence it would be much more difficult to predict how long until a star burns out.

I think you simply missed my point twice.

>> "Some external circumstances could change reducing them to a pile of oxides"

(or some other transformations I guess in case of fire.)

That was precisely not what I was talking about (or put another way if you still require abundant explicitness): not the kind of immortality I was describing. Maybe I was not clear enough.

> For example, a star could hypothetically fall into another much larger than it and prolong its life

Well, once again:

>> Of course they can still be impacted by external effects, but sustaining their own internal processes does not require such inputs

> It's fair to characterize life as a process that locally decreases entropy

I would characterise life as a process that copies its information into the future.

Maybe the idea of copying and that of decreasing entropy are closely related. It's interesting that life needs digital codes to evolve, they seem a requirement for both maintaining low entropy and copying.

It's certainly not a sufficient criterion. I argue nothing about my air conditioner is living and is no closer to living than any other rock.

This isn't a fully formed point, just a thought.. but your air conditioner wouldn't exist without a living system directing matter to form an air conditioner; at least, it's so unlikely as to be practically impossible. So an air conditioner may not be a living system, but it's the product of a living system (at some point; maybe there are other non-living systems in the pipeline, e.g. factory robots).

I don't understand a heat pump to be an inherently artificial thing, but I am struggling to think of an example of a naturally occurring heat pump. If there were natural valves and compressors (which there might be from certain perspectives) then I could see a naturally occurring heat pump being possible.

It seems strange to say process devices are the thing that make a system living or not.

Definitely not a sufficient criterion. A solar panel that charges a capacitor would be alive.

Why not consider it alive? The same with the fires mentioned in other comments. In The Andromeda Strain (the novel, don't remember if it's in the film as well), there's a scene where the characters are trying to determine a single quality they can look for that signifies life. They come up with energy conversion, and they do consider luminescent watch hands and geologic formations. It's been a long time since I read it, I don't recall them ending with back-patting and "mission accomplished" cheers, but the point of the scene wasn't to make a definite answer for the readers nor the characters - it was that when you're looking for life, perhaps your definition of "life" will need expanded. So why do we need a definition that absolutely includes This, and absolutely excludes That and The Other? What if we settle on a definition and found out it does apply to The Other, why must we change the definition instead of changing how we categorize The Other? Both are adjusting our model to fit new evidence. Why be so determined that Life Must Be Like Us And No Other?

Generally with life we require some form of replication or it would quickly go extinct. Now, if you add in a 3D printer/assembler to the aforementioned solar-panel/battery the loop is getting rather complete for something to be considered life.

99% of mules are sterile. And I'm not sure what the current debate on viruses and prions is. It's a difficult question to generalize.

Oh, for sure. Like worker bees and ants, they don't breed themselves, but feed the collective that does.

I guess one of the rules of life will be "It's not an object, it's a system"

Oh that's good, I like that one! But a counterpoint for fun: weather is a system. I wonder if we can find a rule that seems sane but would classify an inert gas as alive? Imagine defining life and someone rebuts with "Argon."

Reply on news.ycombinator.com