Articles like this indicate we should lock down the definition of "computation" that meaningfully distinguishes computing machines from other physical phenomena - a computation is a process that maps symbols (or strings of symbols) to other symbols, obeying certain simple rules[1]. A computer is a machine that does computations.
In that sense life is obviously not a computation: it makes some sense to view DNA as symbolic but it is misleading to do the same for the proteins they encode. These proteins are solving physical problems, not expressing symbolic solutions to symbolic problems - a wrench is not a symbolic solution to the problem of a symbolic lug nut. From this POV the analogy of DNA to computer program is just wrong: they are both analogous to blueprints, but not particularly analogous to each other. We should insist that DNA is no more "computational" than the rules that dictate how elements are formed from subatomic particles.
> a computation is a process that maps symbols (or strings of symbols) to other symbols, obeying certain simple rules[1]
There are quite a number of people who believe this is the universe. Namely, that the universe is the manifestation of all rule sets on all inputs at all points in time. How you extract quantum mechanics out of that... not so sure
I don't think it's necessary to completely discard the idea. However, I do think it's important, at the end of it all, to ask: Okay, so what's the utility of this framework? What am I getting out of setting up my point of view this way?
I'm reminded of an old YouTube video [0] that I rewatched recently. That video is "Every Zelda is the Darkest Zelda." Topically, it's completely different. But in it Jacob Geller talks about how there are many videos with fan theories about Zelda games where they're talking about how messed up the game is. Except, that's their only point. If you frame the game in some way, it's really messed up. It doesn't extract any additional meaning, and textually it's not what's present. So you're going through all this decoding and framing, and at the end your conclusion is... nothing. The Mario characters represent the seven deadly sins? Well, that's messed up. That's maybe fun, but it's an empty analysis. It has no insight. No bite.
So, what's the result here other than: Well, that's neat. It's an interesting frame. But other than the thought to construct it, does it inform us of anything? Honestly, I'm not even sure it's really saying life is a form of programming. It seems equally likely it's saying programming is a form of biochemistry (which, honestly, makes more sense given the origins of programming). But even if that were so, what does that give us that we didn't already know? I'm going to bake a pie, so I guess I should learn Go? No, the idea feels descriptive rather than a synthesis. Like an analogy without the conclusion. The pie has no bite.
our relationship to computation got weird when we moved to digital computers. Like, I don’t think anyone was saying “life is like millions of slide-rules solving logarithms
in parallel”. but now that computers are de-materialized, they can be a metaphor for pretty much anything
Good point - maybe the analogy to computation arises simply because digital computation and the synthesis of DNA, RNA and proteins are all performed by discrete-state machines?
I don't think they are. The things analog computers work on are still symbolic - we don't care about the length of the rod or what have you, we care about the thing the length of the rod represents.
analog computers don't generally compute by operating on symbols. For example see the classic video on fire control computers https://youtu.be/s1i-dnAH9Y4?t=496
OP's specific phrasing is that they "map symbols to symbols". Analog computers don't do that. Some can, but that's not their definition.
Turing machines et al. are a model of computation in mathematics. Humans do math by operating on symbols, so that's why that model operates on symbols. It's not an inherent part of the definition.
No, analog computers truly are symbolic. The simplest analog computer - the abacus - is obviously symbolic, and thus is also true for WW2 gun fire control computers, ball-and-shaft integrators, etc. They do not use inscriptions which is maybe where you're getting confused. But the turning of a differential gear to perform an addition is a symbolic operation: we are no more interested in the mechanics of the gear than we are the calligraphy of a written computation or the construction of an abacus bead, we are interested in the physical quantity that gear is symbolically representing.
Your comment is only true if you take an excessively reductive view of "symbol."
In what sense? I agree the tech industry fucking sucks right now, but I don't see how this has anything to do with that.
A physical computer is still a computer, no matter what it's computing. The only use a computer has to us is to compute things relative to physical reality, so a physical computer seems even closer to a "real computer" or "real computation" to me than our sad little hot rocks, which can barely simulate anything real to any degree of accuracy, when compared to reality.
I love thinking about life as computation. Cells are computers, enzymes are functions, ribosomes are compilers, nucleic acids are source code...
Enzymes in particular are a lot like unix pipelines. An enzyme catalyzes its substrate's conversion into its product which is the substrate of another enzyme. When cells ingest glucose, it flows through the glycolysis metabolic pathway until it becomes pyruvate, and may be reduced even further depending on available resources. It's a huge pipeline of enzymes. They just kinda float around within the cell and randomly perform their tasks when their substrates chemically interact with them. No explicit program exists, it emerges from the system within the cell.
Cell - Computer
Enzyme - Function / Process / Filter
Substrate - Data
Product - Data
Metabolic pathway - Program / Script
I've been playing in my mind with an idea for an esoteric programming language modeled around enzymes. The program defines a set of enzymes which are functions that match on the structure of data, automatically apply themselves to them and produce a modified version of the input which may in turn match against other enzymes. The resulting program metabolizes input by looping over the set of enzymes and continuously matching and applying them until the data is reduced to its final form. If no enzymes match, the output is the unmodified input.
I think the issue with this way of thinking is that humans think in abstractions.
Abstractions don't really exist, they're a product of the human mind, but then we apply them to nature. Calling DNA code, comparing NNs and the brain, etc. But those abstractions fall apart when you look a little too deeply at what actually happens in nature.
Is DNA code? Or is it more like a machine? Is it neither, or is it something embedded in such a complex space that our simple abstractions can't capture the full nature of its being?
When you look at the nature of DNA, it does more than simply act as code. It can edit and self-modify, self-assemble, self-replicate, it can turn genes on and off, it can perform what can be argued as computations itself. If you limit yourself to thinking of it as code, you might miss crucial ways it exists/performs in real life.
> When you look at the nature of DNA, it does more than simply act as code.
> It can edit and self-modify, self-assemble, self-replicate, it can turn genes on and off
Unless my knowledge of biology is very outdated or incomplete, all of those things you cited are done to DNA. They don't happen spontaneously.
DNA doesn't self-replicate, a whole bunch of enzymes come and actively copy it. Genes don't spontaneously turn on and off, some enzyme comes and attaches or removes a methyl group. DNA doesn't self-assemble, it is actively coiled around histones to form nucleosomes. Bacteria have a huge variety of enzymes for manipulating native and foreign DNA, they have their own CRISPR mechanisms.
I'm thinking more of early RNA and DNA life, where ideas like the RNA-world might have happened and applied. RNA can assemble, replicate, and catalyze to form deoxynucleosides in a proto-DNA, without "outside" work needing to happen from enzymes/proteins/etc.
Similarly, RNA and DNA "machines" could have existed before cellular life, in which genetic material self-assembled, transferred genes horizontally/vertically, etc, blurring the lines between genes as "code" and something else.
I think RNA (in particular, ribozymes) does those things.
But DNA is effectively separation of concerns: RNA systems evolved to RNA mediated systems with DNA as more inert and reliable storage and enzymes as more effective catalysts. Or so the RNA world hypothesis goes.
You're one of those cats that provides a subtle reminder that Dr. Alan Kay (invented the tablet/Xerox ALTO interface) was first a biologist. Thank you for the enlightened smalltalk! (;3)
This feels like the kind of popsci that's written for people who already agree with the author - there's nothing resembling an argument, or even a definition of "computation." There are nods to Church-Turing, but the leap from "every effectively calculable function is computable" to "life is a computation" is larger than anything you could fit in a book.
Yes, the article appears to be a short excerpt from a book and probably loses a lot of context because of that. I am interested in the questions raised by the author but will wait for the book to come out. The good news is that it appears the book will be open access - MIT Press seems to be encouraging this lately (at least by allowing this as an option for authors).
I'm not too impressed with this article since it doesn't really give a definition for computing, just picks a few similarities between what we see as computing (in the practical sense) and what cells do.
It's a shame because there *has* been a lot of deep work done on what kind of computer life is.
People often use the Chomsky Hierarchy (https://en.wikipedia.org/wiki/Chomsky_hierarchy) to define the different types of computer vs automata. Importantly, a classical Turing machine is Type-0 on the Chomsky Hierarchy. Depending on what parts you include from a biological system, you could argue it's anywhere from Type-0 to Type-4.
Interestingly, the PhD thesis of well-known geneticist Aviv Regev was to show that certain combinations of enzymes with chemical concentration states are enough to emulate pi-calculus, and therefore are Turing machines!
https://psb.stanford.edu/psb-online/proceedings/psb01/regev....
This is the kind of evolved computer science that was going on when I was a teenager. Have an upvote eig!
My addition: it's funny for how much speculation we get in the, "hard cognitive science" (RIP) that in lieu of the big insights we get from Godel, Turing, Russell that many/most undergraduates and even post-graduates still haven't internalized Wittgenstein's work especially the Tractatus. I feel like it gets us to, "the questions you're asking about how life works and the questions about what is at the core of logic and mathematics (language) are definitely related but not in any of the fundamental ways you hope they are..."
For the uninitiated-- try reading the thing in one sitting. It takes about an hour:
I don't see the point of asking this question. Like, sure, all physical systems follow certain rules, so any such process will develop in a way that it look like a computation of an algorithm. Also, evolution itself is constantly optimizing organisms to best adapt to their environment, just like a computation.
So asking if life is a computation seems mostly like a semantic musing. Define "life" and define "computation", then see if they're the same.
Evolution is not optimizing anything. What's happening in the biosphere is a process of mutation & selection, it's not optimization towards any particular goal or objective. Furthermore & slightly more abstractly, b/c of conservation of mass & energy, what's actually happening is re-organization of existing biomass into different life forms enabled by solar radiation.
I suppose I fail to see why evolution through natural selection is not optimizing. That was Darwin's big idea, right? That given heredity, selection, and variation you end up with life forms we'd consider optimized for their environments?
Or do you mean that optimization by definition must include intent, and evolution as a mindless process has no intentionality?
The title should definitely be "Is it possible to simulate living organism?" given the last sentence is "Simulations like these show how computation can produce lifelike behavior across scales".
Nothing about life is discussed here, it's not even defined once.
> Also, evolution itself is constantly optimizing organisms to best adapt to their environment, just like a computation.
There is no optimization, if organisms can reproduce, they'll continue to exist. That does not mean they are the "best adapted" or on a trajectory toward better adaptation.
It's entirely possible for a germ line to become less fit over time, even to the point of extinction, and that's still evolution. Time has shown that is the case for most germ lines.
I like the idea instead that some biological components have deterministic optimizations because they’re closer to a molecular form, like DNA, RNA, some protein machinery, etc. because essentially these are driven by some kind of chemistry and physics. Whereas higher level, emergent biological forms are more stochastic in their function, like organelles, an organism, or populations, etc. In that sense, there’s no computation to life, more that life is constrained by the physics of the world in which it develops.
It’s likely if different life forms on another planet, it will have a different “computation” model because its defined by different physics that it experiences during evolution. Though I suppose there will some similarities depending on some fundamental rules of the universe. Will propagation molecules like RNA or DNA always look like helixes, or will the radiation or physics of another planet create another form of propagation molecule we haven’t yet observed. Might make for an interesting experiment to simulate.
Allegedly it can find the Answer, unfortunately it'll take so long to do it we'll forget what the Question was before life is done computing the Answer!
Surprised the article didn't mention the most fitting sense in which life is computation, which is at the biosphere level. Life can be characterized as a (recursive) search function executing continuously over billions of years. It seeks out new environments, reacts to changes in the environment at all scales between molecules and mountains. Life is a vast, distributed visitor pattern whose payload is itself.
The basic parameters of affective neuroscience make it difficult to conflate actions with computations. Because there isn't a content to thought, thoughts aren't about things, brains/CNS/bodies lack any units that could be computed, there's only an arbitrary sleight of hand linking life and computation.
there are criticisms of life as classical computation or in a more restricted context, cognition as computation [1] - one of which amounts to this: for any computation, there is a frame of reference in which that computation can be modeled, and if so, that frame of reference itself cannot be modeled by said computation.
No, and this is a very philosophically confused post because it weirdly does not really give any definition of computation.
Computation really is a fancy word for calculation. What matters about computation is that its teleological. Computers are physical systems designed towards a particular end. A computer is, physically, no different than any other system. What differentiates it is that it's designed and we're interpreting its behaviour in a particular way.
Unless you're trying to make a grand theological argument in which "life" is taken to be some Hitchhikers Guide-like machination towards some end, it's not a computation. Life doesn't compute anything, the same way a falling pen doesn't compute gravity unless in a metaphorical sense.
The article is a pretty good example honestly of the problems of taking metaphors literally, common in the AI space where the author hails from. A similar case "artificial neurons" which are really metaphorical neurons. You have to be particularly careful when making comparisons between intentionally designed technological artifacts and biological and physical processes.
No, obviously not. This is just clickbait and self-congratulation for the tech industry. Computation is not the end-all of every process or entropy flow. Please get better philosophy.
"However, we should be careful with the metaphors and paradigms commonly introduced when dealing with the nervous system. It seems to be a
constant in the history of science that the brain has always been compared
to the most complicated contemporary artifact produced by human industry
[297]. In ancient times the brain was compared to a pneumatic machine, in
the Renaissance to a clockwork, and at the end of the last century to the telephone network. There are some today who consider computers the paradigm
par excellence of a nervous system. It is rather paradoxical that when John
von Neumann wrote his classical description of future universal computers, he
tried to choose terms that would describe computers in terms of brains, not
brains in terms of computers."
I have no idea what the submitted MIT article is trying to say. Does the MIT article try to make the point that neural networks can be used for computation given ridiculous amounts of memory? They can, but that still does not explain real intelligence. Otherwise, the article makes the same mistakes as pointed out in the above quote.
To me, the article just ask "Is it possible to simulate living organism features?" and say a small yes by saying "Simulations like these show how computation can produce lifelike behavior across scales".
I'm not expert to judge the result of "drawing a missing hand by using neural network on each pixels"(if it's what it's done? Again not an expert).
this question reminded me of the poetry of terrence mckenna. "Technology is the real skin of our species. Humanity, correctly seen in the context of the last five hundred years, is an extruder of technological material. We take in matter that has a low degree of organization; we put it through mental filters, and we extrude jewelry, gospels, space shuttles. This is what we do. We are like coral animals embedded in a technological reef of extruded psychic objects. All our tool making implies our belief in an ultimate tool. That tool is the flying saucer, or the soul, exteriorized in three-dimensional space."
In that sense life is obviously not a computation: it makes some sense to view DNA as symbolic but it is misleading to do the same for the proteins they encode. These proteins are solving physical problems, not expressing symbolic solutions to symbolic problems - a wrench is not a symbolic solution to the problem of a symbolic lug nut. From this POV the analogy of DNA to computer program is just wrong: they are both analogous to blueprints, but not particularly analogous to each other. We should insist that DNA is no more "computational" than the rules that dictate how elements are formed from subatomic particles.
[1] Turing computability, lambda definability, primitive recursion, whatever.
There are quite a number of people who believe this is the universe. Namely, that the universe is the manifestation of all rule sets on all inputs at all points in time. How you extract quantum mechanics out of that... not so sure
I'm reminded of an old YouTube video [0] that I rewatched recently. That video is "Every Zelda is the Darkest Zelda." Topically, it's completely different. But in it Jacob Geller talks about how there are many videos with fan theories about Zelda games where they're talking about how messed up the game is. Except, that's their only point. If you frame the game in some way, it's really messed up. It doesn't extract any additional meaning, and textually it's not what's present. So you're going through all this decoding and framing, and at the end your conclusion is... nothing. The Mario characters represent the seven deadly sins? Well, that's messed up. That's maybe fun, but it's an empty analysis. It has no insight. No bite.
So, what's the result here other than: Well, that's neat. It's an interesting frame. But other than the thought to construct it, does it inform us of anything? Honestly, I'm not even sure it's really saying life is a form of programming. It seems equally likely it's saying programming is a form of biochemistry (which, honestly, makes more sense given the origins of programming). But even if that were so, what does that give us that we didn't already know? I'm going to bake a pie, so I guess I should learn Go? No, the idea feels descriptive rather than a synthesis. Like an analogy without the conclusion. The pie has no bite.
[0]: https://youtu.be/O2tXLsEUpaQ
OP's specific phrasing is that they "map symbols to symbols". Analog computers don't do that. Some can, but that's not their definition.
Turing machines et al. are a model of computation in mathematics. Humans do math by operating on symbols, so that's why that model operates on symbols. It's not an inherent part of the definition.
Your comment is only true if you take an excessively reductive view of "symbol."
A physical computer is still a computer, no matter what it's computing. The only use a computer has to us is to compute things relative to physical reality, so a physical computer seems even closer to a "real computer" or "real computation" to me than our sad little hot rocks, which can barely simulate anything real to any degree of accuracy, when compared to reality.
Enzymes in particular are a lot like unix pipelines. An enzyme catalyzes its substrate's conversion into its product which is the substrate of another enzyme. When cells ingest glucose, it flows through the glycolysis metabolic pathway until it becomes pyruvate, and may be reduced even further depending on available resources. It's a huge pipeline of enzymes. They just kinda float around within the cell and randomly perform their tasks when their substrates chemically interact with them. No explicit program exists, it emerges from the system within the cell.
I've been playing in my mind with an idea for an esoteric programming language modeled around enzymes. The program defines a set of enzymes which are functions that match on the structure of data, automatically apply themselves to them and produce a modified version of the input which may in turn match against other enzymes. The resulting program metabolizes input by looping over the set of enzymes and continuously matching and applying them until the data is reduced to its final form. If no enzymes match, the output is the unmodified input.Abstractions don't really exist, they're a product of the human mind, but then we apply them to nature. Calling DNA code, comparing NNs and the brain, etc. But those abstractions fall apart when you look a little too deeply at what actually happens in nature.
Is DNA code? Or is it more like a machine? Is it neither, or is it something embedded in such a complex space that our simple abstractions can't capture the full nature of its being?
When you look at the nature of DNA, it does more than simply act as code. It can edit and self-modify, self-assemble, self-replicate, it can turn genes on and off, it can perform what can be argued as computations itself. If you limit yourself to thinking of it as code, you might miss crucial ways it exists/performs in real life.
> It can edit and self-modify, self-assemble, self-replicate, it can turn genes on and off
Unless my knowledge of biology is very outdated or incomplete, all of those things you cited are done to DNA. They don't happen spontaneously.
DNA doesn't self-replicate, a whole bunch of enzymes come and actively copy it. Genes don't spontaneously turn on and off, some enzyme comes and attaches or removes a methyl group. DNA doesn't self-assemble, it is actively coiled around histones to form nucleosomes. Bacteria have a huge variety of enzymes for manipulating native and foreign DNA, they have their own CRISPR mechanisms.
Similarly, RNA and DNA "machines" could have existed before cellular life, in which genetic material self-assembled, transferred genes horizontally/vertically, etc, blurring the lines between genes as "code" and something else.
But DNA is effectively separation of concerns: RNA systems evolved to RNA mediated systems with DNA as more inert and reliable storage and enzymes as more effective catalysts. Or so the RNA world hypothesis goes.
I learned something new today! Thank you.
It's impressive that RNA of all things can be folded in such a way that it also acts like an enzyme.
"It's not even wrong" - Pauli
It's a shame because there *has* been a lot of deep work done on what kind of computer life is. People often use the Chomsky Hierarchy (https://en.wikipedia.org/wiki/Chomsky_hierarchy) to define the different types of computer vs automata. Importantly, a classical Turing machine is Type-0 on the Chomsky Hierarchy. Depending on what parts you include from a biological system, you could argue it's anywhere from Type-0 to Type-4.
Interestingly, the PhD thesis of well-known geneticist Aviv Regev was to show that certain combinations of enzymes with chemical concentration states are enough to emulate pi-calculus, and therefore are Turing machines! https://psb.stanford.edu/psb-online/proceedings/psb01/regev....
My addition: it's funny for how much speculation we get in the, "hard cognitive science" (RIP) that in lieu of the big insights we get from Godel, Turing, Russell that many/most undergraduates and even post-graduates still haven't internalized Wittgenstein's work especially the Tractatus. I feel like it gets us to, "the questions you're asking about how life works and the questions about what is at the core of logic and mathematics (language) are definitely related but not in any of the fundamental ways you hope they are..."
For the uninitiated-- try reading the thing in one sitting. It takes about an hour:
https://wittgensteinproject.org/w/index.php/Tractatus_Logico...
So asking if life is a computation seems mostly like a semantic musing. Define "life" and define "computation", then see if they're the same.
Or do you mean that optimization by definition must include intent, and evolution as a mindless process has no intentionality?
I'm just not sure what you're driving at.
And the flux of geothermal and chemical energy
Nothing about life is discussed here, it's not even defined once.
There is no optimization, if organisms can reproduce, they'll continue to exist. That does not mean they are the "best adapted" or on a trajectory toward better adaptation.
It's entirely possible for a germ line to become less fit over time, even to the point of extinction, and that's still evolution. Time has shown that is the case for most germ lines.
It’s likely if different life forms on another planet, it will have a different “computation” model because its defined by different physics that it experiences during evolution. Though I suppose there will some similarities depending on some fundamental rules of the universe. Will propagation molecules like RNA or DNA always look like helixes, or will the radiation or physics of another planet create another form of propagation molecule we haven’t yet observed. Might make for an interesting experiment to simulate.
[1] https://www.youtube.com/watch?v=0FUFewGHLLg
[1] https://plato.stanford.edu/entries/computational-mind/#GodIn...
Could you pull out the specific list of ATCG and make a brain
Life is truly weird sometimes
Computation really is a fancy word for calculation. What matters about computation is that its teleological. Computers are physical systems designed towards a particular end. A computer is, physically, no different than any other system. What differentiates it is that it's designed and we're interpreting its behaviour in a particular way.
Unless you're trying to make a grand theological argument in which "life" is taken to be some Hitchhikers Guide-like machination towards some end, it's not a computation. Life doesn't compute anything, the same way a falling pen doesn't compute gravity unless in a metaphorical sense.
The article is a pretty good example honestly of the problems of taking metaphors literally, common in the AI space where the author hails from. A similar case "artificial neurons" which are really metaphorical neurons. You have to be particularly careful when making comparisons between intentionally designed technological artifacts and biological and physical processes.
https://www.inf.fu-berlin.de/inst/ag-ki/rojas_home/documents...
"However, we should be careful with the metaphors and paradigms commonly introduced when dealing with the nervous system. It seems to be a constant in the history of science that the brain has always been compared to the most complicated contemporary artifact produced by human industry [297]. In ancient times the brain was compared to a pneumatic machine, in the Renaissance to a clockwork, and at the end of the last century to the telephone network. There are some today who consider computers the paradigm par excellence of a nervous system. It is rather paradoxical that when John von Neumann wrote his classical description of future universal computers, he tried to choose terms that would describe computers in terms of brains, not brains in terms of computers."
I have no idea what the submitted MIT article is trying to say. Does the MIT article try to make the point that neural networks can be used for computation given ridiculous amounts of memory? They can, but that still does not explain real intelligence. Otherwise, the article makes the same mistakes as pointed out in the above quote.
I'm not expert to judge the result of "drawing a missing hand by using neural network on each pixels"(if it's what it's done? Again not an expert).
Some prefer GUI