Monday, 17 August 2015

Everything is a Computer


 

The words 'computer' and 'computing' are both vague and very broad. Even some of those involved in the field of artificial intelligence believe that (some?) molecules are computers. Or, more precisely, they argue that molecules are “closed physical systems” which compute. That is, molecules carry out information processing. Or, alternatively, they receive input, work on that input, and then produce/deliver output.


Indeed in one place I came across this representation of the DNA molecule as a Turing machine:


      Tape = DNA
      Head = Ribosome
      State Register = RNA
      States = Amino acid
      Instruction Table = DNA codon table
      Output Tape = Proteins


(A distinction has to be made here between seeing the DNA molecule as a computer and the possible use of biochemical materials – such as DNA - as 'hardware' for computers.)


This position on omnipresent computation reaches its zenith with what's called pancomputationalism. This is the view of digital physics which says that the entire evolution of the universe is (or has been) a computation. That may mean (to some) that God Himself is a computer scientist or programmer.


Again and again it seems to come back to the broadness or vagueness of the word 'computation'. One way this can be put is to admit that in certain ways and senses the mind-brain is indeed a computer or that it does carry out computations. However, all sorts of philosophers have argued that computation isn't definitive of mind: it's not even important to being a mind. Or, it may be important, though only in the sense that, as Hilary Putnam puts it, "every ordinary open system realizes every abstract finite automaton” (1991).


John Searle agrees with Putnam on this. He wrote the following about the vagueness and/or broadness of the term 'computation':


... the wall behind my back is right now implementing the WordStar program, because there is some pattern of molecule movements that is isomorphic with the formal structure of WordStar. But if the wall is implementing WordStar, if it is a big enough wall it is implementing any program, including any program implemented in the brain.” (1992)


The Digital Window: window open = 1/ window shut = 0
Elsewhere Searle also writes:












... the window in front of me is a very simple computer. Window open = 1, window closed = 0. That is, if we accept Turing’s definition according to which anything to which you can assign a 0 and a 1 is a computer, then the window is a simple and trivial computer.” (1997)


Of course what can be said here is that there are certain things that computers (or Turing machines) do which Searle's wall or window doesn't do. That's true; though that are also indefinitely many things things that the mind-brain does that computers can't do. However, that doesn't seem to stop people claiming that the mind-brain is a Turing machine or a computer.


When people say, “The brain is a Turing machine”, many of them really mean: “The mind-brain sometimes - and in some ways - behaves like a Turing machine.” Of course I said “some people”. That means that other people believe that the human brain is literally a Turing machine or computer.


To put some more meat on the qualified claim that “brains behave like Turning machines”. This amounts to saying that some higher-level processing done by brains parallels - to some extent - what Turing machines do. The problem with that is such higher-level processing may also have some similarities with what goes in a cell or even in a inorganic/inanimate system. Indeed the brain's own neurons process input in ways similar to a Turing machine.


The crunch question would therefore be whether or not the mind-brain and Turning machine are alike when it comes to processing abstract and highly complex cognitive tasks.


Modelling Physical Systems



Where does the idea that the brain is a computer (or Turing machine) come from? I believe it mainly comes from the following.


Firstly, there's a strong link which is made between brains, mathematical models and Turing machines. When taken together (or sequentially) we get the idea that the brain is a Turing machine.


This (very roughly) is the argument:


i) Once we have mathematically described all the workings (or processes) of the brain


ii) then a Turing machine can model the brain.


Some go so far as to say that “mathematics is synonymous with computation”. And through maths/computation, we can model all of reality (or each bit separately). Thus a Turing machine can model and compute the whole of physical reality.... including the brain.


Workers in artificial intelligence (AI) are keen to tell us that physicists have created accurate models of all aspects of physical reality (including at the quantum level). All these models are essentially mathematical in nature. Thus it's only one step on from there to say that they're also computable.


So we're talking about models of physical reality. More accurately, we're talking about mathematical models of physical reality. These models, by definition, will be computable.


Thus brains, being part of physical reality, can be mathematically modelled. Therefore brains themselves are computable. They are computers.


Indeed the brain, like any spatio-temporal region or an atom, is said to be a “subset of the physical universe”. And every subset of the physical universe is computable – at least in principle.


Other people talk about “simulating physical systems” rather than modelling them. So, to quote such a person, this is what he concludes:


... if the brain is a purely physical object, which is the only option consistent with our understanding of how the universe works, there is no reason it cannot be simulated.”


The logic is simple:


i) All physical objects/systems can be simulated/modelled.


ii) The brain is a physical object/system.


Iii) Therefore the brain can be simulated/modelled.


The problem here is the slide from X's being computable to X being a computer.


Even if the brain (or its workings) were computable, that wouldn't automatically make it a computer. Searle's wall or window is (digitally) computable; though that doesn't make it a computer. Sure, we can carry out an act of stipulation and say: If the brain is computable, then it's a computer. Then again we can do the same to Searle's wall/window. If that wall/window is computable, then it's a computer. At this rate, almost everything is a computer.


A computer, on the other hand, is both computable and a computer.


References


Putnam, Hilary. (1991) Representation and Reality
Rana, Fazale. (2012) 'Biochemical Turing Machines “Reboot” the Watchmaker Argument'
Searle, John. (1992) The Rediccovery of Mind (pages 2008/9).
-- (1997) The Mystery of Consciousness

No comments:

Post a Comment