The famous Chinese Room thought experiment - John Searle (1980)

that's all by definition we ever tell digital computers how to do and so digital computers are inside the chinese room in a certain sense today we're talking about john searle's famous super famous chinese room thought experiment since it's a thought experiment that just means that it's an imagined scenario a scenario that you have to imagine and once you imagine it cyril thinks you're going to get a certain kind of imaginative result and that's going to prove something about a certain theory of the mind and if you're going to understand the philosophical theory of mind that this thought experiment is trying to disprove then we need to do a super super quick recap of the theories of mind that we've talked about in this course and so i'm going to write some stuff up on the board now so these are the theories of mind that we've already talked about in this course you can go watch my previous videos to get a recap of all of this sort of stuff very very quickly dualism is the theory that the mind and the body are different kinds of stuff they're distinct there's two kinds of things in the universe mental stuff and physical stuff behaviorism identity theory which is short for the mind brain identity theory and functionalism these three they are all versions of physicalism which is the view that there is only physical stuff in the universe the only things that exist are physical things and so the mind is just part of the physical world behaviorism is the version of physicalism that says that the mind is just behavior identity theory is the version of physicalism that says the mind is identical to the human brain and functionalism is the version of physicalism that says that mental states or mental processes are just functional states or functional processes functions you'll remember and this is the most important thing to catch up on so i'll put in the description the link to the video about functionalism functions are defined in terms of causal inputs and outputs that's going to be important for what we're talking about today before i can say what the chinese room thought experiment is we need to just understand what searle's target of attack is the thing that he's attacking it's a theory of mind that he calls strong ai strong ai is short for strong artificial intelligence and this is a theory about the human mind as cyril understands this label which he is introducing strong ai is the theory that the human mind all of your thoughts and memories and beliefs and desires and feelings and sensations all of that stuff that make up your mind the human mind is a program that's being run on the brain just in the same way that there's some software there's an application or a program being run on the hardware of a computer so according to this theory the mind is the software and the brain is the hardware that's the theory that searle is attacking and here's what he says about that theory this is on page 669 in the reading if you're following along this view that is strong ai has the consequence that there is nothing essentially biological about the human mind the brain just happens to be one of an indefinitely large number of different kinds of hardware computers that could sustain the programs which make up human intelligence on this view any physical system whatever that had the right program with the right inputs and outputs would have a mind in exactly the same sense that you and i have minds so for example if you made a computer out of old beer cans powered by windmills if it had the right program it would have to have a mind and the point is not that for all we know it might have thoughts and feelings but rather that it must have thoughts and feelings because that is all there is to having thoughts and feelings implementing the right program if you've been taking this course up until now you should recognize strong artificial intelligence the idea that having a mind just is running a certain program defined in terms of inputs and outputs right that theory should seem familiar to you because it's one of the theories that we've already been talking about it's this one it's functionalism strong ai is a version of functionalism the idea is simply that if you have the right functional structure you take the right inputs and and give the right outputs then you are a mind or then you have thoughts and feelings and desires and emotions and sensations and so forth if you think it's possible to take a human mind take all of a person's memories and thoughts and that sort of thing and download them onto a computer if you think that's possible then you are an adherent of strong ai this version of functionalism and what cyril is going to do when presenting this thought experiment is he's going to attack that theory he's going to say that theory doesn't work that's not the way the mind is one last thing before we get to the thought experiment itself here's just another passage from the reading this is on page 670 and it's just some characteristic john searle snark this is just him saying that there really are scholars theorists researchers whatever who actually believe in strong ai here's what he says herbert simon of carnegie mellon university says that we already have machines that can literally think there is no question of waiting for some future machine because existing digital computers already have thoughts in exactly the same sense that you and i do well fancy that philosophers have been worried for centuries about whether or not a machine could think and now we discover that they already have such machines at carnegie mellon that's some classic john searle snark which we're going to put aside for a minute so that we can talk about what he means by the terms that he uses in this reading syntax and semantics let's start with syntax the word syntax just means grammar right so the english language has certain rules of grammar that say that the verb has to go here and the noun has to go there and you put a comma here or whatever cirl is using the word syntax in a broader sense to refer to all of the formal features of some language here's an example to illustrate this i'm going to write a word on the board that i assume you don't know okay so i needed a language that doesn't use the same alphabet as english and so i've got this word here right you don't know what this is i do but let's say you don't know what it is you can know lots of things about these symbols right here right like you know that like this symbol right here is taller than the other symbols right this one you know it's like taller it's like it's like up here to down here whereas this symbol is only is only only goes up to there right and this one is sort of a straight line and there's a dot over it i could have left the dot out and then and then this one is sort of like a swoopy thing that looks like the english letter e you can know all sorts of stuff about these symbols without knowing their meaning you know something about their form their shape right those formal features the features that have to do with their form that's what cyril means by syntax all the stuff like the order in which the symbols appear you you can understand that right if i were to switch the symbols around in order or write one on top of the other instead of next to the other all that sort of stuff you could understand that without understanding the meaning and so all of those facts about these symbols fall under the category of syntax for cyril then there's the actual meaning of these symbols these symbols aren't just some shapes they're a word in a language they're the word shalom in hebrew this is hebrew i wrote this on the board shalom is the hebrew word for peace it's also used as a greeting you use it to greet someone you say hello shalom sometimes you also say goodbye by saying shalom but i think less frequently it's mostly used as a greeting when you when you first encounter someone to sort of mean hello this is a hebrew word and now you know the meaning of it the meaning is the semantics semantics just means the meaning of it what this word represents it represents the idea of peace okay so do you understand the difference between syntax and semantics as cyril is using these terms the syntax is the are the formal features of the language the shape of the letters and that sort of thing the order in which they appear that sort of stuff and then the semantics are the actual meaning the content what these symbols represent the shapes of the symbols their form and what the symbols represent syntax and semantics okay we're almost there we're almost ready to talk about the chinese room thought experiment but not quite first we need to introduce the idea of a digital computer and to understand what we mean by digital computer we have to talk very briefly about a guy who lived in england in the middle of the 20th century named alan turing turing was a mathematician and computer scientist living in england uh he there's a movie about him benedict cumberbatch plays alan turing i don't love the movie but it's fine i guess whatever there's some problems with it in my opinion the point is this dude invented computer science invented computers used the computers to crack the nazi enigma code defeated the nazis and then later in his life because he was gay the government of the united kingdom forced him to get electric shock therapy to cure him of his homosexuality which drove him to commit suicide so not a happy ending there are a lot more interesting things to say about alan turing the person but we're just going to talk about a certain kind of digital computer a theoretical computer called a turing machine a turing machine is a theoretical type of machine although actual computers are in some broad sense all real touring machines but here's the way to think about what a touring machine is and this is a little bit oversimplified imagine you've got a piece of tape it's a long piece of tape and the tape is split up into cells little units or whatever and on each of these cells there are some symbols written in fact there are not just any old symbols there are digits and there's only two digits written on these right either the digit one or zero this is where binary code comes from there's two digits there's only two by right there's ones and zeros and this code can be used to represent all sorts of stuff the touring machine is this piece of tape plus a little thing we'll draw it with like little wheels it's like a little box with wheels on it right and this little box um it can drive along the tape from one cell to the next one and so on it can go backwards and forwards right and it can erase whatever is inside the cell and it can write either ones or zeros in there and different turing machines can run different programs but we're going to imagine that this turing machine is an adding machine it adds up numbers so here's what's going happen we feed into the machine this piece of tape we've got three ones and then a zero and then two ones and then three zeros right this adding machine is going to add these numbers right so we've got the number three one two three and then we've got a space and then we've got the next number one two two so if this machine does its job right then it's gonna add three and two and get the answer five here's how the machine works the machine doesn't know that these ones and zeros represent numbers it doesn't know anything about numbers all it knows is that when it sees a one it ignores the one and moves on to the next cell and it sees another one it can read the one and it says nope moving on to the next cell remember it doesn't know that this thing is called a one or that this one is called a zero or anything no it can just recognize the syntax it can just recognize the shapes of the symbols and perform functions as a result of that it doesn't know the semantics it doesn't know what any of this means right it doesn't know that this is a 1 it represents the notion of one or that these three ones put together represent the number three it doesn't know anything about the meanings of these symbols it just can respond to their form or their shape it can follow instructions okay so it sees the one it keeps going when it gets to the zero it sees a zero and it has some different instructions for what to do when it sees this round zero shaped thing when it sees a round zero shape thing like that it erases what's on the cell and it replaces it with a one those are the instructions just erase the zero replace it with a line okay and then you it switches itself into a different internal state things get a little complicated we don't need to worry about that too much but then it knows it just keeps going it sees a one it keeps going it sees another one it keeps going it hits its second zero when it hits the second zero it's got instructions for what it's supposed to do it's supposed to go back one spot so it just it's got little wheels it just back tracks again and then it has as its instructions to erase the symbol in the in the cell that it goes back to and write in a zero and then its last instruction in here is to halt halt means stop that's all that the touring machine did it just looked at some symbols moved to the next one sometimes it erased sometimes it's not it didn't it didn't understand the meaning of anything that it was doing but look what it did it produced the correct answer five one two three four five five ones in a row represents the number five in the in the system that we're using in the language that we're using so even though this damn machine didn't understand anything about addition didn't know anything about numbers didn't have any meaning or semantics at all no semantic understanding of any kind all it could do was manipulate symbols but nonetheless it successfully added three and two by just following some basic instructions that's what a touring machine is computers work this way they manipulate symbols at least the computers that are well digital computers that run on binary code they just take strings of ones and zeros very very long strings and they move the symbols around that's it the key point is this a digital computer is a very specific type of thing all it is is a digit manipulator all it is is a device that moves around symbols that's all it does that's all it ever can do if it's going to be a digital computer now we are ready we're ready to get to the famous chinese room thought experiment what this thought experiment is going to try to show is that strong ai is false and thereby that functionalism is false right and it's going to show that by showing that a digital computer can't understand the meaning of anything computers like this that just manipulate symbols they can have syntax but they can never have semantics so they can't have human understanding they can't have thoughts in the way that human beings have thoughts because our thoughts involve the understanding of the meanings of symbols this should all become clearer i'm going to read a long passage which explains the chinese room thought experiment this is on page 671 if you're following along in the reading imagine that you are locked in a room and in this room are several baskets full of chinese symbols imagine that you like me do not understand a word of chinese but that you are given a rule book in english for manipulating these chinese symbols the rules specify the manipulations of the symbols purely formally in terms of their syntax not their semantics so the rule might say take a squiggle squiggle sign out of basket number one and put it next to a squoggle squoggle sign from basket number two now suppose that some other chinese symbols are passed into the room and that you are given further rules for passing back chinese symbols out of the room suppose that unknown to you the symbols passed into the room are called questions by the people outside the room and the symbols you pass back out of the room are called answers to the questions suppose furthermore that the programmers are so good at designing the programs and that you are so good at manipulating the symbols that very soon your answers are indistinguishable from those of native chinese speakers there you are locked in your room shuffling your chinese symbols and passing out chinese symbols in response to incoming chinese symbols on the basis of the situation as i have described it there is no way you could learn any chinese simply by manipulating these formal symbols okay so uh let's draw the scenario as cyril is imagining it so we've got a room and then we've got a person in the room right and the key thing is just that this person doesn't understand chinese but does understand english and searle you'll have noted refers to the chinese symbols as squiggle squiggle and squaggle swaggle and you might think that seems a little racist i agree it does seem a little racist but we can make the point without any of the squiggle squiggle talk right what we have here are some pieces of paper and on the pieces of paper are some symbols in chinese or hebrew or arabic or one of a gazillion other languages that you inside the room don't understand you don't understand the semantics but you can still understand the shapes the forms the syntax so we've got some symbols on there that you don't understand and these pieces of paper get passed in through a slot and inside the room you have a rulebook and the rulebook has some instructions on it and the instructions just tell you how to manipulate symbols the same way that computers manipulate symbols when they see ones and zeros they erase some of them they replace some of them that sort of thing all these instructions ever do is tell you how to deal with the syntax of these symbols they never tell you what the symbols mean and you look up in your rule book it says when you get a symbol like that you should pass out a piece of paper that has a symbol like this searle's point is this you can get very good at this you could get very good at recognizing these symbols and knowing what page the rule is for that symbol in the book so you can do it very quickly and you could become very accurate always passing out the symbols or the whole series a long series of symbols right that you're supposed to pass out of the room whenever you get this sort of input you could do this for a thousand years you could think really hard you could be a super genius in there you would never be able to figure out what these symbols mean no matter how many symbols you processed following your rule book reading every word of the rule book which is a million pages long and you can fit it all into your brain doesn't matter you're never going to be able to figure out semantics from syntax i mean think about it i'm assuming that you didn't know hebrew you could never figure out the meaning of these symbols just by looking at the symbols and seeing how tall they are or seeing you know which other symbols get spat out when these symbols come in there's no amount of information about the form of symbols and how those symbols are supposed to be manipulated that could ever tell you the meanings of those symbols no matter how smart you are that's searle's point and searle's point moreover is that if you think about it that's the situation that a digital computer is in we never teach digital computers because of the definition of what they are as digital computers they're just symbol manipulating machines we never give them anything other than instructions about how to manipulate symbols which zeros to erase and replace with ones and which ones to erase and replace with zeros and how to move around the symbols that's all by definition we ever tell digital computers how to do and so digital computers are inside the chinese room in a certain sense they're just given information about how to manipulate symbols and they get lots of practice manipulating symbols but they're never given any information they can't be given the type of things that they are they're never given any information about the meanings of those symbols and so therefore digital computers can't understand languages they can move symbols around in languages but unlike human beings who understand the meanings of words computers digital computers never and never could understand the meanings of words and so there's this thing there's this mental state this state of the mind we could call it understanding and that's a state of the mind that a mere functionally described machine could never have according to searle if we interpret the chinese roon thought experiment in the way that he wants us to interpret it it shows that digital computers can never have understanding now notice here that there's a difference between a digital computer which is a rather specifically defined type of machine and just machines in general searle is not saying that we could never build a machine that had understanding and consciousness and thoughts and whatever that's not what he's saying because indeed in a certain sense maybe the human brain is a machine if we understand machine in the broader sense that it can be made up out of brain goop and neurons and whatever then sure machines could have thoughts machines could understand things but searle's not talking about machines in the broadest sense such that the brain the human brain is a machine also no no he's talking specifically about digital computers and he's saying that digital computers can't have thoughts and the reason that they can't have thoughts is because they are just syntax manipulating machines and syntax is never enough for semantics or to put it another way form is never by itself sufficient for meaning okay so that's his point computers as we have them all of the computers that we have which are digital computers they can't think here's something that searle says to sort of make this point directly this is on page 673 in the reading if you're following along is instantiating or implementing the right computer program with the right inputs and outputs sufficient for or constitutive of thinking the answer is clearly no and it is no for the reason that we have spelled out namely the computer program is defined purely syntactically but thinking is more than just a matter of manipulating meaningless symbols it involves meaningful semantic contents these semantic contents are what we mean by meaning okay so the thought is functionalism is false because having the right inputs and outputs is just a matter of syntax a digital computer that has all the right inputs and outputs that kind of computer all the computers we have they can simulate understanding of the chinese language for example but they can't actually have understanding because all they ever get like the person inside the chinese room all they ever get are facts about syntax they never get any information about semantics as just a last little thing we should talk about one response to the argument that searle is making that says that cyril is just getting it wrong and that's called the system's response the idea behind the system's response is that sure searle's right this dude inside the chinese room he can't ever understand chinese but the thing that is understanding chinese according to the functionalist according to the strong ai theorist or whatever is not a person inside of the chinese room it's rather the whole damn system this whole system is the thing that understands chinese there's no part inside the system that understands chinese of course that's true but understanding chinese just is being a system that does all of this if there was a person inside of a chinese room and that person could take in chinese symbols as inputs and give the right chinese symbols as outputs then according to this response that whole system not the person inside that whole system would understand chinese because that's all there is to understanding chinese searle considers the system's response in the reading that we read for today in the original article about the chinese room and here's what he says about the system's response you can judge for yourself whether or not you think this is an adequate response this is on page 672 some people attempt to answer the chinese room example by saying that the whole system understands chinese the idea here is that though i the person in the room manipulating the symbols do not understand chinese i am just the central processing unit of the computer system they argue that it is the whole system including the room the baskets full of symbols and the ledgers containing the programs and perhaps other items as well taken as a totality that understands chinese but this is subject to exactly the same objection i made before there is no way that the system can get from the syntax to the semantics i as the central processing unit have no way of figuring out what any of these symbols means but then neither does the whole system okay that's searle's response the point is it doesn't matter if you include the whole system the whole system is just a symbol manipulating system like a digital computer but symbol manipulation is never enough for actual understanding it's not crazy that kind of seems right and if that's right then strong ai is false and maybe as a result functionalism in general is also false

Channel: Jeffrey Kaplan

Views: 435,025

Keywords:

Id: tBE06SdgzwM

Length: 28min 29sec (1709 seconds)

Published: Tue Sep 01 2020