Martha the Symbol Transformer
I posted this in the other place, so I'll restate it a bit differently here.
As a twist on Searle's Chinese Room argument, let's say that Martha, who speaks !Kung, has been trained to meticulously follow whatever rules she is handed for transforming symbols from set A to set B. Sometimes the symbols are arbitrary and made up (with arbitrary rules for how to combine them), and sometimes they are human languages, but ones Martha is not familiar with, such as English and Chinese. And sometimes it's a mix, such that English is translated into B, or A is transformed into Chinese*.
Now I think this demonstrates clearly that symbol manipulation does not result in understanding, because the system (Martha, rules and room), does not require meaningful symbols in order to perform a translation from one set to another. As far as the system is concerned, there is no difference between English or Chinese and some arbitrary symbols, except probably in the complexity of the rules needed to compute the translation (for human languages).
As such, symbol manipulation cannot be what underpins understanding or meaning. Searle was correct about that. Whatever it is that we do, it's something more than symbol manipulation.
* Of course the generated Chinese doesn't stem from anything meaningful.
As a twist on Searle's Chinese Room argument, let's say that Martha, who speaks !Kung, has been trained to meticulously follow whatever rules she is handed for transforming symbols from set A to set B. Sometimes the symbols are arbitrary and made up (with arbitrary rules for how to combine them), and sometimes they are human languages, but ones Martha is not familiar with, such as English and Chinese. And sometimes it's a mix, such that English is translated into B, or A is transformed into Chinese*.
Now I think this demonstrates clearly that symbol manipulation does not result in understanding, because the system (Martha, rules and room), does not require meaningful symbols in order to perform a translation from one set to another. As far as the system is concerned, there is no difference between English or Chinese and some arbitrary symbols, except probably in the complexity of the rules needed to compute the translation (for human languages).
As such, symbol manipulation cannot be what underpins understanding or meaning. Searle was correct about that. Whatever it is that we do, it's something more than symbol manipulation.
* Of course the generated Chinese doesn't stem from anything meaningful.
Comments (488)
What about maths? Isn't that just symbol manipulation? Given the input 2[sup]2[/sup] I've been told to output 4. This is even more evident when we start to use imaginary numbers and derivatives and whatnot. Surely this constitutes understanding?
Seems to amount to symbol manipulation.
Yes, but does this competence with symbol manipulation constitute understanding? That you are able to manipulate the symbols in accordance with rules immediately shows, at most, that you understand the rules; that is, that you understand what it is permissible or mandatory for you to do in order to correctly apply them in particular cases. This, however, isn't the understanding that is at issue in the debate regarding functionalism, or computationalism, in the philosophy of mind.
The understanding at issue rather is the understanding of the meaning of linguistic symbols (and hence also the understanding that grounds intentionality -- e.g. reference to extra linguistic items in the general case). Functionalists claim that competence in following the syntactic and/or logical (i.e inferential) rules that govern the use of the symbols is sufficient for constituting an understanding of their meanings. Searle disputes this. Hence, for him, an ability to "understanding" the rules (i.e. display an ability to comply with them) falls short from understanding the language.
That's the question I asked. When it comes to maths, doesn't understanding consist in knowing how to manipulate the symbols, or at least knowing what to do with the input (e.g. plot a graph)?
So she doesn't understand Chinese or English but she does understand Martha-Chinese and Martha-English.
No, understanding math is like understanding programming. You can use both in situations you haven't encountered before.
Quoting Michael
And what does that get you? How do you use it to solve problems or accomplish tasks?
But by this, did Wittgenstein mean knowing how to transform one sentence into another, or did he mean knowing how to use it in the world?
It's the difference between looking up a word in a dictionary, and being able to use that word in various contexts, such as might come up in conversation.
Forgive my ignorance, but my initial reaction was that the "systems reply" was still suitable. The symbol manipulation has to be meaningful to someone on the output end otherwise the rules are arbitrary. To the output recipient, the system understands the language even if the language is some hybrid of existing language(s).
This seems question-begging. I just don't see how the Chinese Room demonstrates one way or the other that humans understand symbols in a different way than the aggregate of the system. Or if humans do understand symbols differently, why we should exclude the notion that a sophisticated system can also understand symbols, albeit differently.
If one performs mathematical processes without any understanding of the meaning of the symbols or of the reasons behind the processes, then that person would be like the English speaking person in Searle's Chinese Room. You've just restated the thought experiment using math as an example as opposed to the Chinese characters.
The question is whether such examples are akin to what we actually do when we speak and convey our thoughts to other people. I'd say it's pretty dissimilar, considering in the Chinese Room we have no idea what thought we're conveying, but when we say "I'm hungry" (for example) we do.
Let's put it a different way. Symbols stand in for whatever it is that we understand. They're an abstraction. The claim Searle is making is that no amount of symbol manipulation gets you to understanding, because understanding isn't in the symbols. The symbols represent or encode for some agreed upon meaning.
What I was trying to do with Martha the !Kung speaker, and the arbitrary sets of random symbols is to show that understanding is something other than symbolic computation (manipulation). To Martha, Chinese and English are the same as some random symbols that don't mean anything. And that is exactly what symbols are to a computer.
But that's not understanding. 1 + 1 means something. It means you can take one of any individual item and put it together with one of any other individual item and have two items. And that's how kids start off learning how to do basic math. They use beads or marbles or whatever. They don't just blindly memorize the rules for symbol manipulation.
That comes in later math classes, which has been a bone of contention and motivation for educational reforms in math (although it goes back to having kids use the multiplication table to get the result of say 3 times 7 without understanding what that means).
Wouldn't a deeper understanding of maths involve knowing why symbols are manipulated the way they are on the basis of the kind of understanding of the way math is built up from the primitive logic of counting that pointed to?
å ß ?
Then they write down:
ç
Now do these symbols mean anything? Does the set of rules for computing the correct symbol (or symbols) result in some sort of understanding?
And I'll grant that there is an understanding (given a certain meaning of the word) in how to go from one set of symbols to another (based on the rules). But is that what we're doing when we speak? Searle's contention is that it is not, such that the Chinese room can't be said to understand Chinese.
That depends on how the sentence is supposed to be used. If it's supposed to be used in the world then to understand it is to understand how to use it in the world. If it's supposed to be used in a transformation then to understand it is to understand how to use it in a transformation.
And as the sentences Martha is given are supposed to be used in a transformation then to understand them is to understand how to use them in a transformation.
Saying that Martha doesn't understand the sentences because she doesn't understand how to use them in the world is like saying that I don't understand the word "bite" because I don't understand how the French use it.
It got me an A in Maths.
Which is the same as saying that Martha doesn't understand Chinese, right? The point being that languages are used in context of a world, not a lookup table or by consulting a dictionary, or applying some bayesian algorithm.
Sure. But she understands the sentences in the Martha-specific language. Just as I understand the word "bite" in English.
We can agree on that. Searle's contention is stronger. He was arguing against the notion that a computer could understand Chinese like a human being does. Applying a different meaning to "understand" and then claiming that Searle has it wrong is to miss what he was arguing against.
If we want to say that the Chinese room understands Chinese in a rule-following or symbol manipulation manner, then okay. That would be like saying that Siri understands when it's cold outside because she says, "Brrrr, it's 15 degrees outside". But of course she doesn't know what it means to be cold.
True. But only because the machine is being taught to use the words in a different way to how a human does. So obviously it won't understand it like a human does. But what if the input is sensory information rather than sentences? The machine detects water falling from the clouds and so outputs "it is raining". This would be a proper way to consider computer understanding.
The problem with Searle's argument is that if a human was put under the same conditions as a computer then the human wouldn't understand (in the same way as a human in a traditional situation). But a human is still conscious. So that a computer wouldn't understand (in the same way we would) under those same conditions is not that it is not conscious. He needs to put the computer under the same conditions that a human would be under to understand the sentences.
The computer understands it in a propositional sense. Let's make this more complex. Let's say the computer has been programmed to read faces and emotion at a funeral. It then tells a grieving person that it's very sorry for their loss.
Does the computer understand what it means to lose someone?
What does fully understanding maths consist of? Knowing the axioms, the rules of inference, and then being able to apply the latter to the former? So I'm provided with some input sentences, told what to do with them, and then output the result.
So let's say you stub your toe. I say that looks painful. Are you going to doubt that I understand what being in pain is? Or if you tell me about a strange dream. Do you doubt that I will understand having a strange dream?
I might or I might not. And I might or might not doubt the computer that says the same thing. I'm not really sure what this question is supposed to highlight. Perhaps that we dogmatically believe that people understand but computers don't?
Right, because he was attacking symbol manipulation as a form of understanding.
Quoting Michael
Okay, so there's consciousness-based understanding where the words, "I'm sorry for your loss" don't imply understanding unless symbol producer has experienced loss or can empathize with losing someone.
And then there's the question of intentionality. How do symbols refer? How is it that 1 stands for any single individual item?
How would you not? Are you supposing that I have some condition where I can't experience pain or fatigue (I'm not aware that there are any humans immune to fatigue).
We understand that people are doing something more than manipulating symbols. When I say that I understand your loss, you take it to mean I can relate to having lost someone, not that I can produce those symbols in the right situation, in which case I'm just formally being polite. If a machine says it, it's understood that someone programmed a machine to say it in those circumstances, which might come off as incredibly cold and insensitive, or downright creepy (if it hits the uncanny valley). What we don't do is think that the machine feels our pain or empathizes.
It's the same with Siri telling me it's cold outside. It's cute and all, but nobody takes it seriously.
When the input is "•" the output is "1". When the input is "••" the output is "2", etc. We're taught what to say in response to something else.
But anyway, I'll try another approach which isn't about consciousness. Once we humans have an understanding of 1 + 1 (to use a trivial example), we can universalize it to any domain. A computer can't do that. It has to be programmed in different scenarios how to apply 1 + 1 to achieve whatever result.
Sure, the computer always knows how to compute 2, but it doesn't know how to apply addition in various real world situations without being programmed to do so.
If it's not dogma then there's evidence. What evidence shows that the computer who says "I'm sorry" doesn't understand and that the human who says "I'm sorry" does?
That's not what reference means at all.
Humans form emotional bonds and machines don't. Do you need some scientific literature to back this up? Humans also grieve when those bonds are broken and machines don't.
What evidence shows that humans can form emotional bonds and grieve but that computers can't? You can't use science because science can only ever use observable behaviour as evidence, and the premise of the thought experiment is that the computer has the same observable behaviour as a human.
The mathematical symbol "1" means any item or unit ever, in the context of counting or sets. You can use it to denote any one thing.
If I made up some word, say "bluxargy", and then defined with some other made up words, what does it reference? It references nothing, so reference can't be symbol manipulation.
Okay, let's set aside empirical matters and just accept that humans do experience emotion. What about Turing machines? Can a Turing machine, in just its abstract form, experience grief? Does that make any sense?
What I mean is, say some brilliant mathematician/programmer defined the algorithm that some theoretical computer could use to compute being in grief, and wrote it down. Would that algorithm then experience grief? Let's say they pay someone to illustrate a Turing machine manipulating the symbols needed to compute the algorithm. Whole forests are cut down to print this thing out, but there it is. Are the symbols sad?
Sure. And you asked how it's come to mean this thing. I pointed out that we're provided with some input (of which there may be many that resemble one another in some empirical way), e.g. "•" or "••", and are told what to output, e.g. "1" or "2".
Again, that's not what "1" or "+" or "2" means, at all.
If we're just going to accept that the humans experience emotions then why not just accept that the Turing machine does? And if it doesn't make sense to say that the Turing machine can grieve then why wouldn't it not make sense to say that the human can grieve? What's the difference between humans and computers that makes emotions in the former reasonable but not the latter?
Because symbols are abstractions from experience. They stand in for something else. An emoticon isn't happy or sad or mad. It just means that to us, because we can be happy, or mad or sad.
So when I say "there's 1 apple" I'm saying "there's any individual thing in context of counting or sets apple"?
You're shifting the terms of understanding. If understanding is granted to the system for the accurate manipulation of the symbols, then human understanding is likewise granted for accurate manipulation of the symbols. It's not enough to have the symbols, one has to have the rules to manipulate the symbols. Searle, and perhaps you, seems to want to isolate the understanding of the Chinese Room participant from the entire system, which includes the set(s) of rules. Martha doesn't need to know the meaning of the output, because the meaning is supplied by the entire system and not a single part of it. Your tongue doesn't need to know the meaning of the words in order to get into the right position to make a sound. The aggregate system is demonstrative of understanding: input to output and all the various computational places in between.
You might object that the computational theory of mind begs the question as well. Humans do have an understanding not present in the Chinese Room, but I don't think appealing to the intuition of the scenario is going to lead us to any insight about what is going on there. The Chinese Room and !Kung-speaking Martha are inadequate to settle the matter because one cannot see past the preconception brought to the example.
Right, so what makes a computer different than an abstraction, like a Turing Machine (of which a computer is a finite realization)? Is it that the computer is made of matter instead of symbols?
You seem to be avoiding the question. Why is it reasonable to assume that humans can experience but that computers can't? What's the evidence? Is there any? Or is it just dogma?
Right, and I'll accept that this is one notion of understanding, being that words can have multiple meanings. Siri knows how to tell me what the temperature outside is. "She" understands how to compute that result.
Quoting Soylent
But Searle's point is that it doesn't matter, because it's still just a form of symbol manipulation. He thinks we do something fundamentally different than following rules to manipulate symbols when we speak English or Chinese, although of course we are capable of computing symbols, albeit not usually as well as a computer.
So it is matter that gives meaning to symbols?
And an abstract Turing Machine can't be said to be using symbols, even if we wrote out the entire computation for being in grief, but a computer can, because it has electricity flowing through it between different parts?
No; that is what proofs in a formal deductive system are. The most interesting mathematical proofs are not formal.
Do you actually have anything meaningful to say about the difference between humans and computers such that we have reasons to believe that the one can understand grief and the other can't? Or are you just trying to avoid answering the question? This is a two-way conversation. It would be nice if you could actually answer my questions too.
And what makes that meaningful?
Quoting Michael
Humans give meaning to symbols, not the other way around. What a computer computes is only meaningful to the degree it's meaningful to us. We built them, after all, to compute things for us.
1 + 1 = 2 is only meaningful to the extent that we give it the symbols meaning. Otherwise, it means nothing.
You're just reasserting the claim that humans can understand and computers can't. I want to know what evidence supports this claim.
We build humans too, you know.
Being used. I've just said.
Computers are instantiations of Turing machines (limited by physics), correct? You agreed that an abstract Turing machine can't compute grief. What makes an instantiated Turing machine different?
You might retort that abstract machines don't compute, but that's not quite right, because we can write out the algorithm for whatever computation, if we wanted to take the time and effort (within the limitations of our resources).
So if there exists an algorithm for grief, why wouldn't the algorithm itself feel grief, or a written out version of Turing machine computing that algorithm? Is there something that an instantiated computer does with symbols that an abstraction doesn't?
Is it the electricity flow through the gates? Does electricity give meaning to symbols?
It's a real thing. An abstract person can't compute grief, either.
An actual computer actually does something, just as an actual person actually does something. Whereas a hypothetical computer or a fictional person don't actually do things.
But here's the thing. The computer is taking in symbols, manipulating those symbols, and outputting symbols, correct? So what's the difference between that and a human writing out the algorithm for computing grief?
A human could take the symbols for a funeral, write down the computations a Turing Machine would make, and output the symbols for grief, or whatever. In theory. Maybe a billion Chinese could do it. Would that system grieve?
The correct question is "what's the difference between a computer taking in, manipulating, and outputting symbols and a human taking in, manipulating, and outputting symbols?" It's the one I've asked you, and it's the one I'm still waiting an answer for.
Humans provide meanings to the symbols in the first place, which is what you're ignoring.
Which means what? And what evidence shows that humans can provide meanings to the symbols but computers can't?
Again, you're just dogmatically asserting that humans can understand and computers can't.
We use symbols to communicate meaning.
Quoting Michael
Searle's argument, as I understand it, is that computers (or any system) are unable to do this if all they're doing is manipulating symbols. Humans are doing something in addition when we produce symbols. The fundamental reason is that symbols aren't meaningful, rather they connotate meaning. They're symbols for a reason.
As I've said before, this doesn't work because the computer is put under different conditions to the person (a person under the same conditions also wouldn't understand - at least not in the same way as the person who's taught different rules for how to use the language). A proper analogy must have the computer and the person having the same input (e.g. sensory detection of water falling from the clouds) and the same output ("it is raining"). Where's the evidence that shows that human understands but the computer doesn't.
And I want to know what evidence shows that people can experience grief but that computers can't.
Or think of it this way -- tables calculate pressure by their tolerance for pressure before they break. So if they break, they've in a way outputted 'I'm broken,' or a cipher of it -- so do tables understand that they've been broken?
Then what does it mean to understand "it is raining", and what evidence shows that humans can and computers can't?
What exactly do you think grief is?
I think it's a serious hypothesis. When I consider my own understanding of "it is raining" all I can consider is the input and the subsequent output.
Perhaps the input to which "grief" is the output? And if we go with something like the James-Lange theory then the input is physiological arousal.
Grief is the input to which "grief" is the output? As in, the English word? What does that even mean?
How about this: grief is a feeling.
It means that if when presented with something I consider "I am grieving" to be the appropriate response then that thing is grief.
And feelings are? And what evidence shows that humans can have them but computers can't?
So people who don't speak English can't feel grief?
Quoting Michael
I understand what a feeling is better than what a symbol is. We think other people feel because we relate to them in certain ways, and we don't relate in those ways to computers.
Are you seriously claiming computers feel?
No. I didn't say that "I am grieving" is the only appropriate response.
So it is just dogma?
I didn't say that. I'm asking what evidence shows that computers can't. Marchesk said that there is evidence. I think it's just dogma.
???
So what is grief then???
Quoting Michael
The evidence is that they display none of the qualities that make us think people feel, such as rigorously inspiring empathy.
First of all, let's try to keep the inflammatory commentary to a minimum.
Second, simply because we do not relate to a computer as well as we do to other humans doesn't mean a computer doesn't feel. The recent movie Ex Machina explores this. To treat humans above computers simply because we don't have an emotional attachment to the latter is to have an anthropic bias.
Furthermore, consciousness could be an emergent property of a system.
So what is grief then???[/quote]
I've already told you. Consider, you might ask me what a horse is. I'd say it's the thing I'd name "horse". You then ask me if I'm saying that the French don't ride horses. I'd say no, because the thing I'd name "horse" can be named something else by other people.
So the presence of emotions is determined by public behaviour. Then if a robot behaves the same way a person does, e.g. saying "I'm sorry for your loss" when you tell them that your father has died, accompanied with the appropriate facial expressions and body language, then the robot has demonstrated his capacity for emotions.
Then I'd ask if you were an idiot. That's not the appropriate way to answer that question, obviously.
Horses aren't just the things we call horses -- for example, if we stopped calling anything 'horse,' there would still be horses. Unless you have the incredible power of making all horses disappear by never uttering 'horse.'
I'm sorry if my fuse is short, but I'm so tired of this line of thought.
I'm not saying that if we stop calling this animal "horse" then it disappears. I don't know how you've managed to derive that from what I've said.
Then how would I answer it? Perhaps by showing you a horse? So let's say I show you two animals. Which one is the horse and which one is the rabbit? The horse is the one that, if shown to you, would predictably have you respond with "yes, that is a horse" (because the output "yes, that is a horse" is (one of) the appropriate thing(s) to say given the input).
So I think my answer was quite appropriate.
If a horse is that which you call 'horse,' it follows that if you do not call anything 'horse,' there are no horses. If there used to be horses but aren't any more, that means all the horses disappeared. QED.
Quoting Michael
YES. That is what a SANE person would do, as opposed to a brain-damaged OLP philosopher.
Quoting Michael
No, the horse is the one that HAS A MANE. Christ.
I'm not saying "X is a horse iff I name it 'horse'".
Quoting Michael
iff clauses are definitions. If someone is asking you what a horse is, they're roughy asking a definitional claim: what are the conditions that make something a horse?
"The things I call 'horse'" is obviously a totally uncooperative answer to that question, and everyone except OLP philosophers understands this.
"The things I call 'horse'" is obviously a totally uncooperative answer to that question, and everyone except OLP philosophers understands this.[/quote]
I can't list the conditions that must be satisfied to make something a horse (if we use your example above then consider that not all horses have manes), or show you a horse (unless we met in real life). What I can do, however, is ask you to consider the sort of thing that you'd have to see to respond with "that's a horse". Well, that thing is a horse.
Obviously it won't help someone who wouldn't know when to say "that's a horse", but I'm confident that you're not such a someone.
No, you can't do that, because then there's just a question, okay, so what the hell do you call horse?
In response to that, you'd do what a SANE person would do, and POINT TO A HORSE. I find it unbelievable that you think that "consider the sort of thing that you'd have to see to respond with "that's a horse"" is something people ever actually do in response to these questions (they don't), or that it is in any way elucidating (it isn't).
If you ask "what's a horse?" and someone responds "the thing you call 'horse'" you know they're (1) an idiot or (2) a philosopher.
And how can I point to a horse during an online discussion? All I can do is tell you something. So what sort of thing can I tell you? I could perhaps list the necessary and sufficient conditions, but of course such a thing is impossible, and then of course someone could ask me what legs and manes and horseshoes are.
At some point one has to assume that the other understands the description. And what does understanding the description consist of? Knowing the empirical situation in which such a description is the appropriate thing to use. So I've just cut to chase and gone straight to this.
And when it comes to something like grief or red or happiness, I can't even break it down into some components parts (like I could do in a generalized way with horses). All I can do is say that they're the things that I name "grief", "red", or "happiness", and hopefully there are things that you name "grief", "red", and "happiness", and so you understand what I mean.
No. Understanding the description is understanding what sorts of things fall under it.
Quoting Michael
And yet there is no way in which 'grief' means 'thing I call 'grief'.' This is a gross misunderstanding.
I'm not saying that "grief" means "thing I call 'grief'". I'm saying that grief is the thing I call 'grief'.
"Ian" doesn't mean "my father" but Ian is my father.
And what does it mean to understand what sort of things fall under "horse"? I'd say it's knowing what sort of phenomena is named "horse".
Not necessarily. Which is why when someone asks what grief is, and so wants a kind of characterization or definition, saying that grief is what you call 'grief' is completely uninformative. And notice that that is what you did to me when I asked.
Quoting Michael
That you know, for example, that things like rabbits, which don't have manes, and have prominent hind legs and small arms, don't fall under it, whereas things with giant penises do.
Yes, it's uninformative. But how can I provide an informative account? Your response "grief is a feeling" is also uninformative because plenty of feelings aren't of grief. Which feeling is grief? What's a feeling, even?
And knowing what falls under "rabbit", "mane", "small arms", etc.? You can't avoid the fact that when push comes to shove knowing what these words means is knowing when to use them. At some point you tie them to some empirical situation in which such words are the appropriate response. Given this input (some sensory stimulation and the question "what is this?") the correct output, given the established rules, is "a rabbit". A computer can do this too.
Just because it isn't maximally informative doesn't mean it's uninformative: there are also lots of things that are not feelings that this rules out.
Also, your offer is not just uninformative, but outright misleading: when someone asks you what something is, they're asking you to characterize it, viz. to give an account of what sorts of qualities make it that sort of thing. But your calling something a 'horse' in no way makes it a horse. So it is not an appropriate answer to the question.
Quoting Michael
Yes, you tie them to empirical situations, but the relevant criteria are then what they look like, how they feel, and how they act, not circular appeals to linguistic behavior!
And there are lots of things that are not named "grief", so in saying that grief is that thing we call "grief" I've ruled them out.
But as I've said before, in the case of grief there are no component qualities, and in the case of horses there are no necessary and sufficient conditions. The only way to understand grief or horses is to know in what sort of empirical situation you would say "I'm grieving" and "this is a horse". Of course, this doesn't help the person who doesn't know when they would say such things, but I assume you're not such a person.
And even if I were to provide some informative, generalized list of conditions that must be satisfied for a thing to be a horse, why are these the conditions that must be satisfied for a thing to be a horse rather than a rabbit? Well, because this is the sort of thing we call a horse.
I didn't engage in a circular appeal to linguistic behaviour. I referred to the empirical situation. But the only way I can refer to the empirical situation to which the word "grief" is tied is to use the word "grief".
I didn't say that calling something a "horse" makes it a horse. Just as I wouldn't say that calling someone "Ian" makes them Ian. But Ian is the man I call "Ian" and a horse is the thing I call "horse".
It rules out nothing, because to know what it rules out, you'd have to already know the answer to precisely the question you just asked.
Quoting Michael
But this is horseshit. For one, you can understand them without speaking English, or without English even existing! For another, you can understand grief and horses say by feeling grief or by riding horses!
Yes, and as I said, I assume you know the answer to the question "what is grief?". If you don't then nothing I can say can help you understand, so it's a futile question.
I meant to quote "grief" and "horses" in that sentence: The only way to understand "grief" or "horses" is to know in what sort of empirical situation you would say "I'm grieving" or "this is a horse".
But I don't know the answer because you are a philosopher and have idiosyncractic, non-intuitive ideas about what it means to understand things, for example, or what it means to feel grief. And you cannot explain those conceptions to me.
Quoting Michael
And whats situations are those? Presumably, you must answer, situations involving grief and horses. Which are what...? Those in which people use the word 'grief' and 'horses?' No; that is not an appropriate answer.
Can you? Explain grief to me. Explain understanding to me.
It's the only answer I can give. My answers can only ever be in English. You either understand the English words I use (which is to know when to use them) or you don't.
The best way to explain grief to you would be to kill one of your family members.
Quoting Michael
Except that's NEVER the answer anyone gives unless they're a philosopher. Instead, they do things like POINT TO HORSES.
Then why did you ask me what grief is? Presumably you wanted an answer, but don't want me to kill your family.
So given two animals, which one do I point to? The horse? Which one is the horse? The one with a mane and a saddle? Why is the animal with a mane and a saddle the horse? Presumably because the animal with a mane and a saddle is the animal English speakers name "horse"?
I asked what you thought it was, or were claiming to in guise of philosopher.
Quoting Michael
If you show me the two animals, I can show you by pointing! I certainly don't say 'the one I call 'horse,'' which would be stupid.
And if I were to then ask you why that one is the horse and not the other?
And then I'd ask you why this size, this shape of legs, and having a mane makes a thing a horse rather than a rabbit?
One response to Searle's Chinese Room thought experiment is the system reply. Another one is the robot reply. Those two responses are quite different in character. That's because Searle's original contention was that however our brains "generate" understanding -- as they allegedly can, according to him -- can't be something that occurs in virtue of computation alone. All the system does still merely consists in manipulation of input symbol strings in accordance with syntactic rules. A robot does other things. It can actively gather data (not restricted to symbols provided to it) and behave in the world.
It seems to me that Searle ought to be able to consistently accept that the robot can manifest understanding while still denying that its "brain" understands anything. Searle's rejoinder to the system reply still validly applies to claims that the robot's brain (i.e. its central controlling system) understands. Searle ought to grant that, if the robot manifests understanding of Chinese (and of its surroundings) in its public behavior, then this understanding may be enabled by its brain functions but isn't constituted by those functions.
Of course Searle actually reject the robot reply. But that's because he is an internalist about intentional content. As I phrased it above, he believes that human brains "generate" understanding on their own quite appart from their embedding ("embeddedness"?) in animate bodies, and the embedding of our living bodies in our social and natural world. He defends a view of intrinsic intentionality according to which meaning and understanding (and reference) are produced by some irreducible and emergent property of biological brains -- irreducible, that is, to computations or syntactic manipulation of symbols. If we dispense with Searle's intentional internalism (i.e. the idea that mental states supervene narrowly on brain states, and depend only on them) then we ought to be happy to deny, with him, that computers understand anything. But we can also accept, unlike him, that sophisticated robots could conceivably understand, and yet deny, unlike him, that our own brains understand anything.
The difference between a computer and a human is flesh. Flesh and computer metal are both matter, but flesh has the biological characteristics that produce the experience of grief (or any other emotion). What is that experience? Flesh experiences pain, for instance, and weakness, excitation in the diaphragm, swelling (a lump in one's throat), fatigue, and a whole set of emotional states and feelings we call grief (or happiness, excitement, remorse, anger, etc.). We experience them because they happen in our flesh (they aren't symbolic) and we have the capacity to experience and interpret our fleshly state. If you burn your hand, you feel intense and enduring pain. You can see that the skin on your hand has been severely damaged by heat, and you can feel it. Lots of animals can match this experience because they have flesh, nerves, and the capacity to feel pain.
Beings or devices which lack the means to experience their own flesh (insects, worms, wood ticks, etc. or computers which have no flesh at all, can not have feelings or experiences. "Experience" belongs uniquely to flesh and the ability to apprehend. That which is without flesh cannot have experiences. What a human experiences in their flesh isn't symbolic, numerical, computational, or algorithmical. It's biological.
If you built a suffering computer, it would consist of many biological features, and have the capacity to experience emotions and pain. Then a computer could suffer. Suffering begins in a kind of substance which machines (metal, gears, wires, transistors, semiconductors, quantum effects et al) don't have.
Interesting that you mentioned that movie, since the machine in the movie manipulated the feelings of the protagonist in order to accomplish some other goal. The protagonist felt empathy for the machine and wanted to help it, not realizing that he was being fooled.
There is another movie along these lines where journalist with a background in robotics is invited to do a piece on a successful roboticist. Turns out this person has managed to create a very human-like android and he wants the journalist to examine it. She ends up falling in love with the roboticist, and a bit disturbed by the android, because it's awkward in conversation, and begins to exhibit signs of jealousy and sexual interest.
Turns out, the roboticist is actually the android, and the android is the roboticist. She's been fooled to see if the mimicry could be carried out convincingly, which it has, since she's fallen in love with a machine that's been programmed to mimic being a self-confident genius. The real person is a less convincing, awkward, but brilliant human.
No, it's not. A person can fake emotions, afterall. I might be convinced that you're sorry (or the robot), but maybe it's just mimicry. Maybe you don't actually feel sorry. Maybe you didn't like the person who died, or me, or just aren't close to the situation. Maybe you just aren't feeling empathetic. But you want to maintain a polite appearance.
Eventually you're going to have to concede that we use the word "horse" to talk about this type of animal rather than another, and so that's why this type of animal is a horse rather than something else.
I was suggesting that understanding maths consists in more than merely knowing how to manipulate symbols; that it also consists in knowing why the symbols are manipulated the way they are. In any case it also depends on what you mean by "manipulate symbols". Can a calculator manipulate symbols? A computer?
Two questions here:
1. Who or what determines what the proper output is?
2. Do computers have physiological arousal?
This is like asking how do I know computers/robots can't be sexually stimulated just because it can be faked.
The linguistic community.
Maybe. What evidence allows us to justify an answer either way?
I don't know. How do you know a rock can't be sexually stimulated?
And computers form a linguistic community?
Quoting Michael
Something about machines not being animals, probably.
And what evidence shows that only animals experience sexual arousal?
They could. Or we could. After, all we're telling the computer the appropriate output given the input, just as we tell the young child the appropriate thing to say given what he sees.
I don't know. I guess hurricanes might be aroused when they hit shore of a major city.
It probably has to do with animals being sexual, and needing to reproduce.
Are they, though? Have computers formed a linguistic community? Have they told us what the symbols of that community mean (or how they are used to use your definition of meaning)?
It seems clear to me that you don't have any evidence that humans can feel but that computers can't. It seems clear to me that this is just a dogmatic assertion. I'm not sure why you're so unwilling to admit this.
You can't be serious.
First off, you agree that there is something more to feeling than producing a symbolic representation of feeling in the proper context, correct?
Actually, my contention was that symbol manipulation alone doesn't result in understanding. If a computer can be arranged to do more than symbol manipulation, then I'm not claiming it can't understand, because I don't know at that point.
Searle's contention was that computers only manipulate symbols, however sophisticated.
This cuts both ways though, do humans/animals do something more than produce a programmed/hard-wired output in the proper situations?
Computers and robots have shown creativity and novelty within a specific domain.
So I'm still waiting for the evidence that shows that people are doing more than manipulating symbols and that computers aren't.
The reason is because symbol manipulation alone undermines itself. In order for there to be symbols to compute, the symbols have to be defined. Chinese symbols without Chinese speakers aren't actually symbols. They're random markings.
The word or emoticon for grief isn't a word or emoticon if there is no grief. It either means something else, or nothing at all. You have to have the grief first before there can be a symbol invented to represent it.
The argument here is that symbols can't be primary or fundamental. They are derived, invented, created to aid in communication or thinking.
Yes, since they don't always produce the same output. Animals, and particularly humans, display a great deal of flexibility and variability There is also a question of what determines the proper situation. What is proper in a given situation? Often, human culture defines that.
An example for the wild is an offspring nest where a video camera was setup and streamed online. The mother, for unknown reasons, started attacking the offspring chicks, and failed to feed them properly. That doesn't make much sense from an evolutionary point of view, but life is messy.
Quoting Soylent
True.
Before language, there were animals who experienced and felt. That's what's fundamental. Language is late in the game. Symbols are parasitic.
You ask how I know that a computer can't feel. That's missing the point. Symbols can't feel. To the extent that a computer only manipulates symbols, it isn't feeling or knowing anything, because there is no knowledge or feeling in symbols themselves, only what they stand in for.
Indeed. Michael earlier remarked that we teach children how to respond appropriately to linguistic inputs. He meant to analogize this with the action of programming a computer. But your remark here illustrates why this analogy is misleading.
Before children come to master appropriate rules of grammar and grasp semantical world-word connections, they already have desires, sensations, yearnings, bodily skills and social relationships. Proper linguistic performance isn't taught to them by means of explicit instructions; the teaching of language rather consists in a further re-shaping of an already actualized embodied form of life. This immature (i.e. pre-linguistic) form of human behavior sustains the ascription of meaningful (albeit merely proto-conceptual) mental states. The mode of teaching is broadly proleptical (anticipative) rather than explicit. That is, there is no need to assume that the child understands what it is shown to him/her that s/he ought to do and say in response to determinate circumstances or verbal instructions. The child's (proto-)linguistic behavior is shaped holistically in a way that sustains intentional ascriptions to him/her of determinate (conceptual) thought contents only very approximately at first, and then, gradually, more determinately, as the meanings of the terms that he/she uses progressively dawn of him/her (as a constitutive result of his/her ability to use them properly in wider ranges of circumstances).
Again, no, horses aren't horses because we call them 'horses.' That's dumb, because they would go on being horses even if we called them something else. In fact, most people in history have called them something totally different, yet they were still horses for all that.
As the old analogy goes, a Martian who knew how to manipulate a chess board to produce all legal moves would still not know how to play, without understanding that one is trying to win. Focusing on symbol manipulation only ignores semantics and pragmatics, without which language is incoherent, and whittled away to an idle assembly of abstractions.
This is like saying that even if I change my name to Andrew then I would go on being Michael. But this is wrong. I'm Michael because I'm called Michael, and if I was called Andrew then I'd be Andrew.
You confuse "to be a horse is to have qualities A, B, and C because we use the word "horse" to name those things which have qualities A, B, and C" with "X has qualities X, Y, and Z iff we use the word 'horse' to name it" (as if calling a thing by that name gives it those properties and not calling a thing by that name removes those properties from it). I'm saying the former, not the latter. Nobody says the latter.
However, being a horse requires you to be a certain kind of creature -- it does not require you to be called 'horse,' nor does your being called 'horse' make you a horse. This is obvious from the fact that there were horses before the word 'horse' existed, and further calling them something different would not make them not horses, nor would calling rabbits horses make them horses, and not rabbits. Consider that if your position were correct, we could literally turn rabbits into horses by changing the way we called them. But this is not so.
Quoting Michael
But your own position is committing you to the latter, which is what I am trying to show you. That you deny it doesn't matter. If your position is:
"To be a horse is to have qualities A, B and C because we use the word 'horse' to name things which have qualities A, B, and C:" it follows from this that if we use the word 'horse' instead to name things that have the qualities of rabbits, then rabbits would be horses. And since they were not horses before, surely even you will admit, you are committed to saying we can turn rabbits into horses by calling them 'horse.' But since we can't, your position is wrong.
That's not how it works. We have two sets of properties; {A, B, C} and {X,Y, Z}. At T[sub]1[/sub] we say that those things that have the properties in the first set are named "horse" and those things that have the properties in the second set are named "rabbit". So to be a horse is have to properties A, B, and C and to be a rabbit is to have properties X, Y, and Z.
At T[sub]2[/sub] we decide to name those things that have the properties in the first set "rabbit" and those things that have the properties in the second set "horse". So to be a horse is to have properties X, Y, and Z and to be a rabbit is to have properties A, B, and C.
At T[sub]1[/sub], Animal 1 has properties A, B, and C, and so is a horse. At T[sub]2[/sub], Animal 1 has properties A, B, and C, and so is a rabbit.
I'm certainly not saying that because at T[sub]2[/sub] we name Animal 1 "rabbit" that it has the properties X, Y, and Z. What you're doing is conflating.
That you cite an example of programming gone wrong might handicap rather than help your cause. When everything goes as it should there is mystery, but when it goes awry we make judgements about software or hardware malfunctions. The mother attacking the chicks is outside the normal output behaviour and curiosity might lead to an examination of the mother's physiology, wherein a discovery of some physical abnormality is given explanatory power for the behaviour. Computers can malfunction in much the same way. When the output is wrong, we can examine the software code or the hardware parts for a flaw.
The problem I see with the Chinese Room and your above example is that if you buy into the computational theory of mind you can see how each respectively fits into the theory. Alternatively, if you think there's something missing, you see how each respectively demonstrates that position as well. The analogies seem only to illustrate confirmation biases in intuition rather than insight into what is really going on.
This feels like a faulty analogy as well. The manipulation of a chess board to produce legal moves would include rules about situational moves implicit with an understanding of the object of the game (e.g., what to do when in check). In this case, we are making an intuitive judgement that knowing all the legal moves is insufficient to produce an understanding of the game, but we are doing so from a state of ignorance. The scope of knowing ALL the legal moves might in fact entail an understanding of the object of the game.
Fair enough. I think I've made the mistake of accepting Searle's setup. If I don't buy into the computational theory of mind, why would I expect the Chinese Room to work? Why would I expect a symbol manipulating system to pass the Turing Test (in a strong way)?
Google had their DeepMind machine learning software learn various Atari 2600 games. For some of them, it excelled. But it struggled with others. It scored a zero on Montezuma's Revenge, because the score doesn't change unless you're able to navigate across a room with obstacles and get the key. DeepMind has no understanding of objects in any of the games. It only knows pixels and the score, which it's trying to maximize. To do well in this game, you need to know that the key is something to aim for. Any human would quickly figure that out.
But admittedly, that is different than a chess playing program. Humans labored to program chess software to play the game. It didn't learn how.
What do you mean, 'so?' There is no 'so' about it: being a horse consisting in having certain properties is not causally dependent on people choosing to call creatures with those properties by the name 'horse.' Again, they were already horses.
Quoting Michael
No, what the fuck? You literally just said that changing what words we call animals would change whether they are rabbits and horses around. Assuming that properties of the animals themselves don't change, you just said that the ones that bore the same properties at t1 that used to be horses will now be rabbits, and vice-versa, by virtue of our swapping the names around.
Surely you see that this is insane?
Quoting Michael
This makes no sense whatsoever. If it has the same properties, then it hasn't changed, and so it can't have changed from a horse to a rabbit. The only way to change from a horse to a rabbit is to change properties, but you've stipulated the properties haven't changed.
What it means to be gay, or to be a horse, depends on how we use the words "gay" and "horse". If we change the way we use the words "gay" and "horse" then what it means to be gay or a horse changes. So there's nothing insane or nonsensical about claiming that if we decide to use the word "horse" to refer to a particular type of small mammal in the family Leporidae then Thumper is a horse -- because to be a horse is to be a particular type of small mammal in the family Leporidae, and Thumper is a particular type of small mammal in the family Leporidae.
They became appropriate referents of the word 'gay' when the new word was formed, but as I said, 'gay' does not mean 'appropriate referent of the word 'gay'', which is the underlying prejudice you are holding onto. Rather, it means attracted to the same sex. They were already attracted to the same sex, and so were already gay, long before they were called 'gay,' if indeed that is what the word means.
As I said before, what it means to be gay (or a horse) depends on how we use the word. If we change the way we use the word then we change what it means to be gay (or a horse).
Those things that weren't horses according to the old use of the word are horses according to the new use of the word, just as those people who weren't gay according to the old use of the word are gay according to the new use of the word. Saying that Thumper isn't a horse even after the change in how we use the word "horse" because he doesn't satisfy the old use is like saying that homosexuals aren't gay even after the change in how we use the word "gay" because they don't satisfy the old use.
And clearly since to be be member of the set of things referred to by "gay" is just to be homosexual, and they already were, then they already were members of this set.
It does not depend on how we use the word. To be a horse is to be a certain kind of animal, which is possible even if there are no words. Horses were in existence before language. This is impossible to square on your view, and it is insane. Nor does it change when we change the words -- as I've said, your position literally commits you to claiming we can turn people gay by calling them 'gay.'
Quoting Michael
There is no such thing as being a horse 'according to the use of a word.' They were horses -- period, simpliciter. And they were so long before anyone called them anything. They were not the referent of the word 'horse' -- but then, the point is that 'horse' does not mean, 'referent of 'horse'' as you seem to think -- it means, a certain kind of animal.
Quoting Michael
If Thumper were a horse, he would have a long face, and a mane, and a horse cock. But he doesn't so he's not a horse. That's not going to change if we start using 'horse' to refer to rabbits. He will still be a rabbit, just a rabbit that is the appropriate referent of 'horse.' This does not change him into a horse.
On the contrary, homosexuals were already gay, even before the new use of the word. This is obvious from the fact that to be gay just is to be homosexual, and they were ex hypothesi already homosexual.
Your criticism conflates "horse" (and "A") as used at T[sub]2[/sub] with "horse" (and "A") as used at T[sub]1[/sub]. I'm no more saying that rabbits are equines than I am saying that ¬Xs are Xs.
Only according to the current meaning of "horse". But I've changed it. You might as well say that if so-and-so was gay then he'd be carefree and happy.
But this is wrong. Horses don't stop being horses when we stop calling them 'horse.' They are still horses. This is because to be a horse is to be a certain kind of animal, and they are still that kind of animal.
Rabbits don't become horses because we change their names, which is what you are proposing. Consider how ridiculous that is.
Yes, and at T[sub]2[/sub] to be a horse is to be a member of the rabbit family.
You're using a straw man interpretation. I'm not saying that if I call a rabbit a "horse" then it becomes the sort of animal that competes in the Grand National. I'm saying that if we, as a linguistic community, call rabbits "horses" then to be a horse is to be a member of the rabbit family.
You haven't changed the meaning of "horse". You've exchanged the word for another. Now you call a "horse" a "rabbit", but you still mean horse.
A horse by any other name.
No, at T2 to be the referent of 'horse' is to be a rabbit. To be a horse is still -- to be a horse, not a rabbit!
This is like saying that at T[sub]2[/sub] to be the referent of "gay" is to be homosexual but to be gay is still to be happy and carefree, not homosexual.
Yes, and "horse" means "rabbit". Just as now I call a homosexual "gay", and I still mean gay -- because "gay" means "homosexual".
And if we coin a new word "horse" that means "rabbit" it follows that rabbits are, and were already, horses.
This is like saying that if you change the meaning of 'gay' as a linguistic community you have changed which people are the appropriate referents of this word, not what gays are, or what it is to be gay.
You don't still mean happy, you mean homosexual now. So you don't still mean "gay".
And if we say that rabbits were gay, we mean that horses are homosexual, right?
So let's invent a new word called horsexual, and let's say that gay now means "horsexual". Now what?
No, it does not. Rabbits were never horses. That's absurd.
Quoting Michael
But thats true! You haven't changed what it is to be gay, ie. homosexual, by inventing a word! Are you crazy?
I was aiming for humor there, because the conversation was starting to make me to laugh.
Gay already means something, so I picked a meaningless word to transition to. Then you can see that meaning doesn't change when the word changes.
Here's a question: since to be any kind of animal is just to be a referent of the word referring to the particular species, were there any animals in existence before people called them anything? If so, how is that possible on your account?
Exactly the way it worked. When we used the word "gay" to talk about the happy and carefree, then to be gay was to be happy and carefree. Now we use the word "gay" to talk about homosexuals, so to be gay is to be homosexual.
I'm not saying that to be an animal is just to be a referent of the word referring to the species. I'm saying that if we use the word "equine" to refer to horses then to be an equine is to be a horse and if we use the word "equine" to refer to rabbits then to be an equine is to be a rabbit and if we use the word "gay" to refer to homosexuals then to be gay is to be homosexual and if we use the word "gay" to refer to the happy and carefree then to be gay is to be happy and carefree.
The word used changes, but a horse is a horse, of course. The meaning of horse remains the same. We don't mean that rabbits are now horses. We mean that horses are horses.
Okay, consider the following.
"Gay" means "homosexual." Homosexuals were already homosexual before the word "gay" was invented. It follows therefore that they were already gay. Please epxlain to me what is wrong with this argument.
Quoting Michael
But this is wrong. Consider: "gay" means "homosexual." If it is true that if we call something "gay," then it is gay, it follows that if we call something "gay," then it is homosexual. But this is refuted by your own examples, since we called things "gay" that were not homosexual
"Gay" means "homosexual." Homosexuals were already homosexual before the word "gay" was invented. It follows therefore that they were already gay.[/quote]
Okay, consider the following.
"Horse" means "rabbit." Rabbits were already rabbits before the word "horse" was invented. It follows therefore that they were already horses.
Your continued rejection of this is nonsensical hypocrisy.
And I'm saying that we could change the way we use the word "horse" such that it then means "rabbit".
Now tell me what is wrong with the argument I presented.
Now please address my argument above as I presented it.
[quote=The Great Whatever]Yes, but that does not mean horses are rabbits.[/quote]
And if I were to say "if we use the word 'gay' to refer to homosexuals then 'gay' becomes a synonym of 'homosexual'" would you respond with "yes, but that does not mean gays are homosexual."?
Right, but what exactly are you trying to claim? That the meaning of rabbits or horses is contained in the word we use, such that if we use another word instead, the meaning changes?
Quoting Michael
No it isn't. A counterfactual premise has the form, "If it were the case that p, then it would be the case that q."
But I'm not making the claim in the language as it is now. As I said before, "horse" means "equine" at T[sub]1[/sub], and means "rabbit" at T[sub]2[/sub], where the language at T[sub]1[/sub] is language as it is now and the language at T[sub]2[/sub] is a hypothetical future language. The conclusion that at T[sub]2[/sub] horses are rabbits applies the language at T[sub]2[/sub], not the language at T[sub]1[/sub]. That's why I accused you of conflation; you interpreted the conclusion using the language at T[sub]1[/sub].
I would have thought the implicitness of the "if" was obvious.
But we use the word "horse" to refer to animals with certain properties, and that's where the meaning comes from. The meaning of "horses" is horses. And if we decide to use "rabbit" in the future, it will also mean horses.
You are claiming now, using the English language that horses would be rabbits at T2. There is no question of which language you intend it to be 'about.' It is now, and you are making the claim now in this language. And according to the rules of the language as it is used now, the claim you just made, that rabbits would be horses at T2, is false. Of course rabbits wouldnt be horses -- that's ridiculous. If you want to make a claim specifically about the language as it exists at that time, you can do that truly, such as : "In the language at T2, 'rabbits are horses' would be true" (which is true), or "At T2, rabbits would be the reference of 'horse,'" which is also true but that is not what you are saying. You are instead saying that rabbits would be horses at T2, which is false. And as soon as this confusion is cleared away, the bite you want to have to your position deflates entirely.
Quoting Michael
No, because your argument was meant to parallel mine as a reductio, and mine was obviously not counterfactual. So leave out the implicitness, and you will either have an unsound argument, or one that does not make the point you intended to make. Either way, it doesn't work.
No, I'm using the T[sub]2[/sub] language to claim that horses are rabbits at T[sub]2[/sub].
No, you are not, because it is not T2, that language doesn't exist, and you can't use a nonexistent language. You cannot just change the rules of the language as it exists now to make a claim in a future language. You can make a claim about a future language, using the present langaue, as in one of the formulations I list above: but this is not what you are doing. If you were using the language as it existed at T2, it would be T2, but it is not, it is now.
It wasn't meant as a reductio. It was meant to bring to light your hypocrisy. You didn't have a problem with the claim "homosexuals are gay" given the change in how the word "gay" was used, but then found a problem with the claim "horses are rabbits" given the change in how the word "horse" was used. And your problem wasn't with the premise, as that was obviously intended as a counterfactual, but with the conclusion - despite the fact that the logic mirrors the gay/homosexual case.
The meaning is horse, not the words "horse" or "rabbit".
The validity isn't in question, the soundness is. The conclusion follows from the premises, but the first premise is obviously false, so it doesn't matter.
So when I output "horse" instead of "rabbit" at T2, not knowing a word of English, what do I mean? Do I somehow manage to mean rabbit? How?
My experience with Michael tells me you are wasting your time; he will never admit he is wrong (which he is).
It's not a non-existent language. It exists in my use of it in stating the conclusion.
If it's valid, which it is, and if the meaning of "horse" is determined by what sort of things we use it to talk about, which it is, then if we decided to the use the word "horse" to talk about rabbits then at that time "horse" would mean "rabbit" and so in that language, at that time, horses are rabbits.
If I change my name to "Andrew" then I would be Andrew. If we change the name of rabbits to "horses" then rabbits would be horses.
It's so simple.
There is no language in which "rabbit" and "horse" are synonymous. You are just using English as it exists now, and making a false claim in it. You are not inventing another language and then using that one in the same sentence. That doesn't even make sense.
Quoting Michael
Fucking...no. Yes, "horse" would mean "rabbit." No, horses would not be rabbits. They would be horses.
Of course it makes sense. We do it all the time when using symbolic logic.
Let "P" mean "philosopher". The Great Whatever is a P. Am I calling you a philosopher or a letter? I'm calling you a philosopher.
Honestly, if this is so hard to understand then I'm wasting my time.
That you can coherently and consistently make a claim in the present language that is true now, because it would allegedly be true in some hypothetical future language.
If "horse" meant "rabbit" then horses would be rabbits. If Hillary Clinton won the election then she would be president.
What is so difficult about this?
This is not symbolic logic. There is no language in which "rabbit" and "horse" are synonymous. A fortiori you are using no such language.
This is getting ridiculous.
You can't use present language and pretend you are using a hypothetical future language; if you could do that then everything would be up for grabs.
If I were to say "If 'squilooples' meant 'rabbits' then rabbits would be squilooples", would you find something wrong with this?
So your claim is that, if we called rabbits "horses," they would become horses. Prima facie, this claim is absurd. My claim is that they still would be rabbits, but we would call them "horses."
You say your claim rests on the ability to simultaneously invent a nonexistent language and then somehow use it in the same sentence.
Why should I believe your absurd suggestion that rabbits would become horses if we changed what we called them, which in turn is bolstered by this absurd stipulation?
There is no actual point in our present language that you are making then.
To be a horse is to be an equine animal. This is true regardless of what the word "horse" means. Horses were horses before the word "horse" existed.
It is not. Being a horse is not determined by how we use "horse;" it is determined by having certain physical characteristics.
To be a horse is to be an equine animal only because we use the word "horse" to refer to equine animals.
To be gay is to be homosexual only because we use the word "gay" to refer to homosexuals.
I am Michael only because I use the word "Michael" to refer to myself.
No. Horses were horses, because they had certain characteristics, long before any such word existed.
Quoting Michael
This is true (to an extent -- it has to do with the speech community, not just you), but names are different from common nouns.
No, you're mixing up the meaning with the word being used.
This is wrong; the meaning of 'X' is determined by how we use 'X', but what it means to be X is not. What it means to be X is what allows you to identify X. Consider this scenario; you are out in the country by yourself and you see a horse. How do you know it is a horse? Because you know what it means to be a horse, but that is not merely a matter of knowing how we use 'X' in the sense of following a rule, it is also a matter of being able to recognize the situations in which it is appropriate to use 'X', and that recognition is partly at least pre-linguistic and thus cannot be exhaustively given by knowledge of rules of language use alone.
What it means to be a horse consists in a set of perceptible physical characteristics, not in language games. So your argument really is making no point at all, other than that common nouns are just common nouns, and could in principle be interchanged with one another. This is a trivial point insofar as the interchanging of common nouns can have no significant effect on the state of the world.
It doesn't matter if the name starts with a capital or a small letter or if it refers to an individual or to a group. The logic is the same.
You seem to think that I'm saying that this animal belongs to the group of equine animals iff we call it "horse". That's not what I'm saying. I'm saying that equine animals belong to the group "horse" because we use the word "horse" to name the group of equine animals.
Tell me where this is wrong: we used the word "gay" to refer to the light-hearted and carefree, and so at that time the sentence "to be gay is to be light-hearted and carefree" was true; we now use the word "gay" to refer to the homosexual, and so at this time the sentence "to be gay is to be homosexual" is true.
And why are those the characteristics that are what it means to be a horse? Because we use the word "horse" to name things which have those characteristics.
So as I said, to be a horse is to be an equine animal (to have those set of perceptible physical characteristics) because we use the word "horse" to refer to equine animals (those that have that set of perceptible physical characteristics).
It was never meant to be anything but trivial. It was a simple remark that The Great Whatever then latched onto and claimed to be absurd/ridiculous/incoherent/nonsense. I've just being trying very hard to point out the simplicity of it.
If "P" means "man" then The Great Whatever is a P.
Yes, it's trivial. But I'm not the one who's then claiming "but the Great Whatever isn't a letter of the alphabet, so what you're saying is idiotic!".
Of course it matters. Proper names and common nouns are different sorts of words, and the latter are property-denoting while the former aren't. Although I should say even with that said, the way you formulated it is not quite right: you are a Michael only because you are called 'Michael;' but even if you were called 'Andrew,' you would still be Michael, viz. since you are Michael, you would still be yourself even if you were called something different.
Quoting Michael
You said that to be a horse consists of having certain qualities only because we use the word 'horse' in a certain way:
Yet there were horses before there were any words at all, and they were still horses in virtue of having those exact same qualities. So you're wrong. Our use of words has no effect on which things are horses. It is not as if what words we use to label animals affects what sorts of animals they are; they were already horses before we called them anything.
Or again, I will put this argument to you, which you still have not addressed:
(1) "Horse" means "equine animal."
(2) Equine animals were equine animals before we called them "horse."
(3) Therefore, equine animals were horses before we called them "horse."
Please explain to me what is wrong with this argument, and do not ignore it again. (2) is clearly true, but to say (2) just is to say (3), since "horse" and "equine animal" are synonyms. Yet you are inconsistent in affirming (2) yet denying (3).
Another way of putting this: to belong to the group of horses just is to belong to the group of equine animals: these are the very same thing. So you cannot say, on the one hand, that an animal is not an equine animal simply because we call it "horse," but on the other, say that it is a horse simply because we call it "horse." This is a contradiction.
(1) "Horse" means "equine animal."
(2) Equine animals were equine animals before we called them "horse."
(3) Therefore, equine animals were horses before we called them "horse."
Please explain to me what is wrong iwth this argument, and do not ignore it again.[/quote]
There's nothing wrong with this argument. It just doesn't address what I'm saying. I'll paraphrase what I said to John:
If "P" means "man" and if you are a man then you are a P.
So if "horse" means "rabbit" and if Thumper is a rabbit then Thumper is a horse.
The problem you're having is that you're interpreting the conclusion of the second sentence as "Thumper is the sort of animal that we race in the Grand National" (which is of course false), but that's like interpreting the conclusion of the first sentence as "you are a letter" (which is of course false).
Your interpretation is wrong, and your criticism depends on this interpretation. Therefore your criticism is against a straw man.
It addresses what you are saying because you have repeatedly denied the conclusion of the argument. So now are you saying you accept it? If there were horses before we called them "horse," it cannot be, as you have claimed, that certain kinds of creatures are horses only because we use the word "horse" in a certain way, since those sorts of creatures were [i]already[i] horses prior to this use. You cannot maintain this conclusion and accept the argument.
Quoting Michael
No, this is not true; If "horse" means "rabbit," then Thumper is a rabbit that is the appropriate referent of the word "horse." That does not make him a horse. Thumper is a rabbit, and not a horse -- hopefully we can agree on this much. Now, changing what he was called would not make him a horse -- rather, he would still be a rabbit. The fact that "horse" now means "rabbit" means that he can be called "horse" as well as "rabbit" -- but [i]that does not make him a horse.[i] He is still only a rabbit.
Quoting Michael
If he is a horse, then he must be such an animal, because to be that sort of animal and to be a horse are the very same thing. You cannot have one and not the other.
What's wrong with this?
A pig is a certain sort of animal, and I would not become that sort of animal just because I could now be called "pig." To become a pig and to become that sort of animal are the very same thing; and since obviously the latter would not happen, neither would the former. I would simply stay a man, and not be a pig, although a new word, which used to refer to pigs alone, would now refer to men as well, and so I could be called by that word, too. You seem to think this would make me a pig; but it would not, since I would be a man, and not a pig.
Again, being called a "pig" does not make you a pig; being a certain sort of animal makes you a pig. To be that sort of animal, and to be a pig, are the very same thing.
For anyone interested in this topic still, this argument is a textbook case of a simple use-mention error:
https://en.wikipedia.org/wiki/Use–mention_distinction
It seems to me everybody in this thread, including you, agree on 90% of the basic underlying assumptions regarding the conventional element in meaning attributions to words of language. Much of the difficulty comes from your using "is" as a word that signals an implicit stipulation about linguistic meaning (i.e. a copula, as in A is the referent of "B") rather than identity.
For instance, you take the two following sentence to mean the same thing:
(1) If "horse" was used to refer to rabbits, then rabbits would be properly called "horses"
(2) If "horse" was used to refer to rabbits, then rabbits would be horses.
The trouble is that you alone are interpreting the meaning of the consequent in the second counterfactual conditional (i.e. "rabbits would be horses") to be interpreted according to the use that is stipulated in the antecedent (i.e. '"horse" was used to refer to rabbits'). The trouble is that everybody else in this thread understand (correctly, in my view) the words "rabbits" and "horses" to have their ordinary (actual) meanings as used in the consequent of (2), and the phrase "would be" to signify identity rather than meaning stipulation. There indeed can't be any meaning stipulation there since there is no mention of the words "rabbits" or "horses".
Rather, in the consequent: "then rabbits would be horses", all the words are being used. And your interpretation of them as being used in accordance with the stipulation that figures in the antecedent of the counterfactual conditional makes nonsense of most ordinary counterfactual conditional statements.
So, almost all of the disagreement in this thread stems from your using the verb "to be", and your interpreting the consequent clauses in counterfactual statements, in non-standard ways. So, you are right that many of your claims are being misinterpreted, but you have some responsibility for that, and you equally often misinterpret their claims for the very same reasons.
Yes, I'm aware that this is going on, and have tried to point this out. But contrary to what you say here, it is in fact wrong to understand the word "horse" according to its current meaning rather than according to its stipulated meaning in the antecedent.
Consider the example I have already offered: If "P" means "man" and if The Great Whatever is a man then The Great Whatever is a P. This is a valid argument. To read "is a P" according to the current meaning of "P" – "is the letter 'P'" – and so therefore reject the conclusion because The Great Whatever isn't the letter P is to equivocate. For the purpose of the argument "P" must be understood according to the stipulated definition in the antecedent, and so to be a P must be understood as to be a man.
For reference, here is the invalid argument that hinges on a use-mention error.
No I don't. This is a straw man.
Not it's not. The use-mention error is when you say "'X' is the same as X" and so claim something like "rabbits are made up of seven letters".
What I'm doing is asserting the synthetic proposition that if "bachelor" means "unmarried man" then to be a bachelor is to be an unmarried man.
This is simply wrong. I can use any kind of language I like; be it current English, current French, archaic English, or a stipulated pseudo-English where the word "horse" means "rabbit". You interpreting the conclusion according to its meaning in current English and so rejecting it is equivocation.
I haven't switched languages. The entire argument is presented in the constructed language where "horse" means "rabbit". Your reading of the conclusion in English proper is the equivocation. When I conclude that rabbits are horses I am not saying that rabbits are one of two extant subspecies of Equus ferus. This reading of it, and so the subsequent rejection, is misplaced.
When I engage in symbolic logic and say "All As are Bs" you can't reject this by saying "we're speaking English, and in English the letters A and B are different".
If the whole argument is in a constructed language (which you've never said before, I think because you didn't intend this to be the case), then what relevance does it have to us here? If we're all supposed to speak a different language temporarily, all you've said is that rabbits are rabbits, which means none of your linguistically-motivated arguments go through, since for them to be interesting, they would have to be expressed to us in English as we use it now. It is precisely this attempt that would make the conclusions non-trivial, and this is why you have not said before now that the whole thing is supposed to be in a made-up language.
And it is not equivocation not to use a made-up language rather than English, that's ridiculous, since you've never stipulated until right now that the whole argument isn't in English but in a made-up language. Equivocation happens only within a language with respect to different interpretations of the same symbols.
If your whole case depends on switching what language we're arguing in temporarily, then what is the point of any of this? You're not saying anything substantial about the way words work, and your original point is not made.
Quoting Michael
"A" and "B" in this case are variables, which is a different thing entirely.
And it is not equivocation not to use a made-up language rather than English, that's ridiculous, since you've never stipulated until right now that the whole argument isn't in English but in a made-up language. Equivocation happens only within a language with respect to different interpretations of the same symbols.
If your whole case depends on switching what language we're arguing in temporarily, then what is the point of any of this? You're not saying anything substantial about the way words work, and your original point is not made.[/quote]
The premise explicitly tells you that the word "horse" is to be understood in a novel way. How much more apparent do I need to make it?
It's no different. Whether "A" or "horse" we're using a string of symbols to denote something else.
Is that premise made in English or made-up-English? If the former, then you are wrong that the whole argument is in a pseudo-language, and as I said before only the consequent is. You are therefore switching languages mid-argument, which is equivocation, and so the argument fails.
If the latter, then the 'if' clause is superfluous, since if the language as you present it already has 'horse' and 'rabbit' being equal in meaning, then there is no need to stipulate conditionally that they are. You could simply say, ''horse' means 'rabbit''. But then, if you translate the argument back into regular English, it will not work, because your conclusion will be translated as 'rabbits are rabbits,' which is not what you intended to say.
What I intend to say is that in those languages where "X" means "Y" the sentence "X is Y" is true. In those language where "bachelor" means "unmarried man" the sentence "bachelors are unmarried men" is true. In those languages where "gay" means "light-hearted and carefree" the sentence "gay people are light-hearted and carefree" is true. In those languages where "horse" means "rabbit" the sentence "horses are rabbits" is true.
This is true. But it does not mean that X is Y, which is where the claim becomes substantive. And it undermines your original point, which was that what it means e.g. to be a horse depends on how people use the word, which is false.
If the truth of "X is Y" is dependent on "X" meaning "Y", and if "X" meaning "Y" is dependent on how we use the word "X", then the truth of "X is Y" is dependent on how we use the word "X". And if that "X is Y" is true is that X is Y then that X is Y is dependent on how we use the word "X".
If the truth of "horses are equine animals" is dependent on "horse" meaning "equine animal", and if "horse" meaning "equine animal" is dependent on how we use the word "horse", then the truth of "horses are equine animals" is dependent on how we use the word "horse". And if that "horses are equine animals" is true is that horses are equine animals then that horses are equine animals is dependent on how we use the word "horse".
This is the consequence of Wittgenstein and Tarski.
This is wrong. "X is Y" is not true just in case X is Y. The words can be used to mean anything you like, and in particular, if "X" and "Y" meant something other than they do now, the truth of "X is Y" would clearly in no way guarantee that X is Y.
For example, if "horse" meant "rabbit," "Horses are rabbits" would be a true sentence, and yet horses would not be rabbits (which is absurd). This is the prejudice that is mistaken.
Quoting Michael
Again, this is wrong. Horses are equine animals; this is not because of, or equal to, any words meaning the same thing. Horses are equine animals even if there is no language at all.
Horses were equine animals before there was any language, because to be an equine animal just is to be a horse, ergo, it is false that 'horses are equine animals iff "horses are equine animals" is true,' since they were equine animals (that is, they were horses), long before there were any sentences to be true.
"Horses are equine animals" is true iff horses are equine animals.
Here is the proof that this is false. By the right-to-left of the biconditional, this follows:
If horses are equine animals, then "horses are equine animals" is true.
Now consider a case before the advent of language. Since there is no sentence "horses are equine animals," a fortiori such a sentence cannot be true. But then, by modus tollens, it follows that in this case, horses are not equine animals.
But to be a horse just is to be an equine animal. Therefore, it follows that in this case, horses are not horses.
But this is a contradiction; so you are wrong.
Is what you say here true? If so, what does its truth have to do with horses being equine animals? Nothing? So I can, in principle, accept the truth of your claim "to be a horse just is to be an equine animal" but not accept that to be a horse just is to be an equine animal, or accept that to be a horse just is to be an equine animal but not accept the truth of your claim "to be a horse just us to be an equine animal"?
Doesn't this strike you as nonsensical?
Surely if I accept the truth of your claim "to be a horse just is to be an equine animal" then ipso facto I accept that to be a horse just is to be an equine animal, and vice versa.
In the context of such a theory, it may be correct to say, for instance, that:
(1) "Horses are equine animals" is true iff horses are equine animals
According to the very same semantic model for the modern English language, one could translate this shema in French thus:
(2) "Horses are equine animals" est vrai ssi les chevaux sont des animaux équestres.
The same object language is the topic -- English -- but the meta-language has been switched to French. Yet the shema says exactly the same thing. This ought to be clear to TGW (and to Tarski and Wittgenstein), but it ought to be baffling to Michael, it seems to me.
That's because in the second shema -- (2) -- the sentence mentioned on the left has no incidence whatsoever on the meaning of "cheveaux" and "animaux equestres" as they are being used in the meta-language. But this is true also for the case of the first shema -- (1). It is, on the contrary, the independently understood meanings of the words used on the right hand side that are being relied on order to specify or define the meanings of the words of the object-language that are being mentioned (and not used) on the left hand side of the shema.
The main point is that the truth conditions expressed by the sentences being used on the right-hand side of both shemas only depend on what is the case in the world (i.e. the extra-linguistic world) regarding horses and equine animals, and don't depend on any kind of linguistic stipulation embodied in the truth theory which the T-shema is a theorem of.
Doesn't this then entail that the below is correct?
"Horses are rabbits" is true iff horses are equine animals
Yes, that's why I said it.
Quoting Michael
In the language as it is now used, it reports that horses are equine animals, which is true. This in no way means, as you think, that horses being equine animals is dependent upon the language I speak existing at all. They are, and always were, equine animals regardless.
In principle, yes, if the language were different! This is precisely the point. If 'horse' meant 'rabbit,' You could very well accept the claim that 'horses are rabbits' is true, yet for all that you would not accept, as you seem to think, that horses are rabbits. Rather, since 'horse' meant 'rabbit,' you would be accepting that rabbits are rabbits.
Quoting Michael
No; I think, again, you are deeply confused about use and mention. The above argument, to which you are not responding, is meant to show this. But to bring it home, let me generalize to the worst case.
According to your claim, with the biconditional, for any sentence "P," if P, then it must be that "P" is true.
Now, it follows from this that before language existed, there was nothing, as follows:
Consider a case where there is no language, and so there are no sentences. You have agreed that whether a sentence is true or not depends on the way it is used; and since no language exists, a fortiori no language is used, and therefore no sentence is true. So I can take any P, and it will not hold, since nothing can hold unless the corresponding sentence "P" is true.
So since in such a case "something exists" is not true, since there are no sentences and so no true sentences, it follows that it is not the case that something exists.
And you can do the same for any sentence you like, to prove any absurdity you like.
Your problem is in thinking that everything depends on language as it is used now in order to be so; but it does not. And this is why the iff schema you present is clearly false.
Quoting Michael
You do, if the language is as it currently is, but counterfactually, if the words mean different things, you obviously do no such thing. Yet these counterfactual situations are precisely what is of interest in the iff claim.
This might be true in relation to some language where "rabbits" is used to refer to what we are referring to, in English, with the phrase "equine animals". But I don't see your point. Horses still are equine animals whatever linguistic stipulations might be in use in this or whatever alternative linguistic community.
You're still missing the point. You say that there were bachelors before we started using the word "bachelor". I ask you by what virtue. You say by virtue of there being unmarried men before we started using the word "bachelor". I ask you what unmarried men have to do with bachelors. You say that to be a bachelor just is to be an unmarried man. I ask you what makes this the case.
So what makes it the case that to be a bachelor just is to be an unmarried man? The fact that we use the terms "bachelor" and "unmarried man" to refer to the same sort of thing.
No, bachelors are unmarried men regardless of what words we used. The terms are synonymous now, which means that no matter what, bachelors are always unmarried men. This persists even if the language changes. It is not as if the way we use the words makes bachelors unmarried men; to be a bachelor simply is to be an unmarried man, period, regardless of what language is used. All the language does is decide that two terms hook up to the same interchangeable group.
I'm saying that to be a bachelor is to be an unmarried man iff we use the words "bachelor" and "unmarried man" to talk about the same thing.
I'm not saying that those people who are bachelors are unmarried men iff we use the words "bachelor" and "unmarried man" to talk about the same thing.
You seem to think I'm saying the latter, but I'm not. I'm saying the former, which is different.
Nope. To be a bachelor is to be an unmarried man, period. In Rome, unmarried men were already bachelors, even though the terms were not used.
Quoting Michael
Explain to me how this differs from what you just said above. It seems to me you are just asserting then denying the same thing.
The sentence "to be X is to be Y" is equivalent to the sentence "'X' means 'Y'".
The sentence "Xs are Ys" is equivalent to the sentence "those things referred to by 'X' are those things referred to by 'Y'".
No, it isn't. To show this, it suffices to show that one can be true, while the other false.
Suppose that 'horse' meant 'rabbit.' Then to be a horse would not be to be a rabbit (which is absurd). To be a horse would still be to be a certain ind of animal, the same kind as before.
The sentence mentioned is in the same language as the sentence used.
If we change the meaning of "horse" then what it means to be a horse (in the updated language) is different to what it meant to be a horse (in the archaic language).
And the thing being referred to is non-linguistic. Tying this back to the Chinese Room argument, Searle's contention was that correctly outputting the right word in a given situation is not meaningful, because meaning is in the reference (to horses, rabbits, unmarried men). Meaning is about something, not when to use a symbol.
Consider that it's perfectly possible to say the right word in a conversation without knowing what it means. A person can fake knowing what a word means. So can a machine.
Although, there would still be horses, just not as we understand. Those animals would be incomprehensible to us.
No. To be a horse is to be a certain kind of animal, regardless of the language. You cannot change what it is to be a horse by changing what you use 'horse' to refer to. If you could, then rabbits would become horses because you called them 'horse,' but they do not become horses, they remain rabbits. And you cannot say 'I don't mean a biological transformation...' because that is precisely what would be required. A rabbit could not change into a horse without changing biologically. That is, it could not change into a horse without changing into a horse.
Yes.
No I haven't. I have repeatedly said that I haven't. I have gone to great pains to avoid any such equivocation. You're trying to force such an equivocation upon me, against my wishes, and despite my calling you out on it.
No it wouldn't. If we translated it back then it would be "horse".
If you're not going to listen to what I say then I really am going to just stop responding to you.
Now, when you translate back, one of two things will happen. Either you will have found that your argument merely states a tautology -- that horses are horses or rabbits are rabbits -- which is not what you wanted to show (rather, you wanted to show something about the relation between language use and what it is to be some animal), or your argument will come out false.
Do you see the problem? Notice that the use of Spanish is irrelevant to the argument, but you want the use of your made up language to be essential to your argument. This is I suspect because your argument hinges on an equivocation, and you want to use the made up sense of the word while slipping through the back door an implication that you have made some point, in English, about how language affects which creatures are which animals, or what conditions it takes to be a sort of animal.
None of this validates the disquotational schema, which remains false, and it remains false that to what it is to be a horse in any way depends on language. To be a horse is to be a certain kind of animal; what the language has to say about it doesn't matter.
Proper nouns and common nouns are different linguistic items. They have different morphology, syntax, and semantics. Common nouns are proprety-denoting; they are true of individuals that bear a certain property. Proper names are directly referential; they refer to individuals independently of what properties they bear. Whether or not one bears a name is a linguistic matter, but whether one bears a certain property is not.
This is confused because the sentence mentioned on the left-hand side of the T-shema doesn't stipulate anything. Rather the whole T-shema expresses one specific consequence of a Tarskian truth theory for the object-language. The meanings of the words used on the right-hand side of the shemas are the meanings that they have in the meta-language used by the theorist in order to state the consequences of the theory. Those meanings are presupposed in the act of stating the theory. Hence they can't be affected by the stipulations expressed by the T-shemas.
I am unsure what you are trying to say. If the words ""horse" and "rabbit" mentioned on the left-hand side mean what they do now, then your T-shema would express incorrect truth conditions for sentences written (or spoken) in the English language. That's because the antecedent would be false in circumstances while the consequent is true (i.e. the actual circumstances where horses indeed are equine animals, but horses aren't rabbits).
Wouldn't you say that to be a horse is to have the properties denoted by the word "horse".
Absolutely not, and this is the core of the confusion. To be a horse is to have the property of being a certain animal. It so happens that to be that kind of animal, and to be the bearer of the property denoted by 'horse,' accidentally coincide in current English. But they need not.
If the word "horse" denotes properties A, B, and C (or the things that have them) then to be a horse is to have properties A, B, and C. I can't make sense of it any other way.
They state truth conditions -- i.e. in what conditions the sentence mentioned on the left hand side is true. The sentence used on the right hand side may state something that is always false (i.e. in all circumstances). In that case the sentence mentioned on the left hand side would be false in all circumstances also. The stipulation of the truth conditions, on the right hand side, just are the stipulations of the conditions under which the sentence mentioned on the left-hand side would be true as interpreted in the object-language.
To be a horse is to have properties A, B, and C, regardless of the words used or regardless of whether there is any language at all. There were horses, and they were horses precisely because they were a certain kind of animal, long before there was any language. The creation of language did not create horses, nor what it takes to be a horse.
What part of this do you disagree with?
Do you agree that:
1) There were horses before the word 'horse'
2) They were horses because they were a certain kind of animal
3) Now, horses are horses because they are a certain kind of animal
So clearly, what it takes to be a horse didn't change, and so is in no way dependent on, the existence or meaning of the word 'horse.'
That's because the way you are reading it, as applied to the description of counterfactual linguistic stipulations (e.g. a hypothetical language as used in counterfactual circumstances) has an incidence on the meaning of the terms used on the right hand side. Hence you have a habit of saying such things as 'If "rabbits" meant "horses" then rabbits would be horses'. And you invoke Tarski's T-shema as a support for the intelligibility of this use. But the disquotational shema doesn't warrant such a use. What it warrants may be the shema:
(1) "Rabbits are horses" is true iff rabbits are horses
This homophonic shema works for the English language since both the antecedent and the consequent are false in all circumstances. But it doesn't warrant your counterfactual conditional claim.
But horses had properties A, B, and C before we called them horses. And that's why we know them as horses and not rabbits or any other animal.
What part of this do you disagree with?
Do you agree that:
1) There were horses before the word 'horse'
2) They were horses because they were a certain kind of animal
3) Now, horses are horses because they are a certain kind of animal
So clearly, what it takes to be a horse didn't change, and so is in no way dependent on, the existence or meaning of the word 'horse.'[/quote]
If I say that "horse" denotes having properties A, B, and C and if I say that this animal is a horse then I am saying that this animal has properties A, B, and C.
Your claim that I'm saying that the animal has properties X, Y, and Z, where having properties X, Y, and Z is what "horse" denotes in current English, is a misinterpretation. Your criticism is directed against a straw man.
I'm not saying they didn't.
(1) "Rabbits are horses" is true iff rabbits are horses
This homophonic shema works for the English language since both the antecedent and the consequent are false in all circumstances.[/quote]
And the schema works for the New English language, where "horse" means "rabbit", since both the antecedent and the consequent are true in all circumstances. Your interpretation of the sentence in English proper is a misinterpretation.
Yes, it would work in New English as used by New English speakers. But then you have to specify in advance that, when you are stating such a shema, you are meant to be understood as speaking New English. Else you are inviting equivocation. When you say something like 'If "horse" meant the same thing as "rabbit" then rabbits would be horses' this is still nonsense in English and equally nonsense in New English since, while the consequent might be true as expressed in that language, it doesn't depend on the truth of the antecedent. From the point of view of speakers of New English, the thought expressed by them when they use the sentence "rabbits are horses" is true quite independently of any linguistic convention.
As I had suggested, properly interpreted (as Tarski meant it to be interpreted as a theorem in a recursive truth theory for a formal language), the biconditional:
(1) "Horses are rabbits" is true iff horses are rabbits
would be true, but the counterfactual (subjunctive) conditional:
(2) If "horse" meant the same as "rabbit" then horses would be rabbits
would still be nonsense. In both cases the consequent is expressed in English. The antecedent of the subjunctive conditional claim doesn't tell you in what language the consequent must be read. It rather tell you relative to which counterfactual circumstances the claim expressed (in English) in the consequent ought to be evaluated.
I have repeatedly said that the conclusion is to be understood as speaking New English, where "horse" means "rabbit", and have repeatedly said that The Great Whatever's criticism rests on the very same equivocation which you mention - as he interprets the conclusion in English proper. See my post to him above.
You need to read it like this:
Given that "horse" means "rabbit" in this language, horses are rabbits.
Even with the provision of this explicit disclaimer, as I explained, the counterfactual conditional statement still is nonsense since the truth of the consequent (even understood in New English) is unconditional. But what you mean to say is that it is conditional on the truth of the antecedent.
If the truth of "horses are rabbits" depends on "horses" meaning "rabbits" (in this language), and if horses are rabbits if "horses are rabbits" is true (in this language), then horses being rabbits depends on "horses" meaning "rabbits" (in this language).
The Great Whatever accepts the first premise, the T-schema shows the second premise, and so the conclusion follows.
And if we apply the above to current English then we can say that because "horses" doesn't mean "rabbits" (in this language) horses are not rabbits. Using the rules of language we can conclude a fact about the world.
The above statement may be true as written in 'this language', but it is false as written in English.
Compare:
(1) Given that Germans put verbs at the end of their sentences, they sausages eat.
This is nonsense because stating a convention that applies to another language in the antecedent of a conditional doesn't entitle you to switch language mid-sentence.
Compare:
(1) Given that Germans put verbs at the end of their sentences, they sausages eat.
This is nonsense because stating a convention that applies to another language in the antecedent of a conditional doesn't entitle you to switch language mid-sentence.[/quote]
I haven't switched languages. The "this sentence" is a recursive reference. The sentence "given that 'horses' means 'rabbits' in this language, horses are rabbits" is written in a language in which "horses" means "rabbits" – and the sentence explicitly tells you this. Thus any equivocation with English proper isn't justified.
OK. You don't mean "recursive". You mean "self-referential". In that case, sure, if the sentence is allowed to claim of itself that it is to be understood in accordance with the linguistic stipulations stated in the antecedent, then, it is true. But it is then equivalent to the following:
(1) If "horses" and "rabbits" are synonymous in some language (that has the same syntax and verbs as English), then, in that language, "horses are rabbits" expresses a true claim.
If you would always say it that way that wouldn't invite any equivocation. But I am usure what philosophical lesson could be drawn from this trivial claim.
Apply the logic to English, where English is both mentioned and used. If "horses" and "equine animals" are synonymous then "horses are equine animals" is true. If "horses are equine animals" is true then horses are equine animals. Therefore if "horses" and "equine animals" are synonymous then horses are equine animals.
Now inject some Wittgenstein. If we use the words "horses" and "equine animals" in the same way then "horses" and "equine animals" are synonymous. Therefore if we use the words "horses" and "equine animals" in the same way then horses are equine animals.
This is what I've been trying to say. The Great Whatever doesn't like it.
If you say that an animal is a horse, you are saying that it has properties A, B, and C, regardless of what words you use to say that it is a horse.
Your confusion is that you think that to say something is a horse is to actually use the word 'horse,' but this is not so. Some uses of 'horse' cannot be used to call something a horse, as when the language is different; and many instances of calling something a horse do not use the word 'horse' either, as when 'caballo' is used instead.
To call something a horse is not to use the word 'horse' to refer to it, except accidentally in cases where these two things coincide, as they do in English. So your conditional is pointless: to call something a horse is to say that it has such properties regardless of the language, and so your point is not made: it in no way depends on what you say 'horse' means, what properties you say something has if you call it a horse.
The T-schema is false, at least as you interpret it, as I have repeatedly shown you.
Get it?
In that situation if "horses are rabbits" is true then horses are rabbits (where the language of the sentence mentioned is the language of the sentence used).
You're still equivocating by considering the sentence used to be in a language where "horses are rabbits" isn't true.
No, they are not; horses can't be rabbits, that's nonsense. What that sentence's truth means is that in that situation, the sky is blue. It means nothing about horses or rabbits at all, since in that situation "horse" does not refer to horses, and "rabbit" does not refer to rabbits, since ex hypothesi the words mean something else.
Quoting Michael
Then translate the argument back into English; it should still be the same argument, and therefore just as good as it was before. Why do you insist on having to use a made-up language in order for your argument to make sense? I am trying ot show you that the reason you do this is because the argument does not make sense, and rests on an equivocation. If you translate the argument back again, you will see this; translation does not affect the soundness of arguments.
Thinking that a true sentence that mentions "horses are rabbits," i.e. "in that situation, 'horses are rabbits' is true" (which is true) is the same as a use of that bit of language, i.e. "horses are rabbits in that situation," which is obviously false (horses can't be rabbits; any six year old can tell you this).
And you say things like:
Confusing the mention of a piece of language in order to predicate something about it, with a use of that very same piece of language.
These mistakes are impossible to make if the use-mention distinction is understood.
Your conditional seems to run the wrong way. It is because we know that (and only as long as we know that) horses are equine animals (assuming "are" here signifies necessary identity of extension) that the expressions designating them are synonymous. It's not the other way around. If an expression is introduced in the language as synonymous to another expression that already has a referent, then one will be able to use both expressions to make (trivial) identity claims. That's because this specific way of introducing the new word into the language (as synonymous to another one) insures that it has the same Fregean sense as the old one. But, generally, if two words that are in fact co-referential have different Fregean senses, then the fact that they are indeed co-referential is something that might need to be verified empirically; and this will not ensure synonymy unless the knowledge of the identity becomes widespread in the linguistic community and this knowledge would also be taken to be a criterion for understanding both expressions.
If you mean "using in the same way" to imply that referents are identical then in order to know that "horse" and "equine animals" are indeed used in the same way by us, in the case where we already know how to use them, would require that we check that any horse necessarily is an equine animal and vice versa.
Which part? You agreed with 'If "horses" and "equine animals" are synonymous then "horses are equine animals" is true' in your previous post and 'If "horses are equine animals" is true then horses are equine animals' is the T-schema, which you accept. The conclusion 'therefore if "horses" and "equine animals" are synonymous then horses are equine animals' simply applies the transitive relation.
We need to know that the things we call "horses" are the things we call "equine animals". Which is to say that we need to know that we use the words "horses" and "equine animals" to talk about the same thing. And what does talking about the same thing consist of? What's the metaphysics behind talking about the same thing? I'm loathe to any interpretation that claims there's more to talking about things than behaviour, intention, and the empirical contexts that influence and measure them. How can anything else become a part of language, meaning, and understanding? This was Dummett's point.
You are reading the T-shema in the wrong direction (from left to right rather than right to left) because the T-shema arises in the context of a truth theory that derives truth conditions for sentences of the object-language (mentioned in the left hand-side), and states those truth conditions in the meta-language used by the theorist -- in our case, English. Hence the meanings of the terms used on the right hand side of the biconditional are assumed to be their ordinary meanings in English. What is allowed to vary, in a range of counterfactual circumstances, isn't the meanings of the object-language sentences mentioned on the left-hand side, but rather the worldly circumstances in which their truth values are evaluated.
For instance the (homophonic) T-shema:
(1) "Snow is white" is true iff snow is white
just like the T-shema (stated in French):
(2) "Snow is white" est vrai ssi la neige est blanche
both state exactly the same thing, e.g., that the object-language (i.e. English) sentence "snow is white" is evaluated true in circumstances where snow is white and is evaluated false in (counterfactual) circumstances where snow isn't white. The T-shema is never concerned with counterfactual circumstances where the meanings of the words (of either the object- or meta-language) would be allowed to vary. On the contrary, the meanings of the terms of the meta-language are assumed to be understood and the meanings of the terms of the object-language are assigned with the use of the meta-language. (Those atomic meanings assignments to individual words actually are stated in the axioms of the Tarskian truth theory, while the T-shemas are theorems that are recursively deduced on the basis of those axioms.)
I agree with everything in this paragraph of yours.
Even then, that I can state the T-schema in a language other than English, e.g. French, is that I can state the T-schema in a language other than English, e.g. New English.
Also, the T-schema is biconditional so it can be read either way. We can say that "snow is white" is true iff snow is white or we can say that snow is white iff "snow is white" is true. It's an iff, not just an if.
In that case your example doesn't have anything to do with the T-shemas that occur in a Tarskian truth theory, and so it's unclear why you attempted to rely on this notion. In the context of such a theory, a T-shema states general truth conditions for a sentence expressed in the object-language and hence has the force of a subjunctive conditional where the truth value of the antecedent is defined as true or false in all possible circumstances, accordingly, whether the condition stated in the consequent is satisfied or not in those circumstances.
You can express the T-shema (and the whole truth theory this shema is derived from) in whatever language you like, including "New English". But such a T-shema tells you nothing about the meanings of the words used on the right-hand side of the shema. Those meanings are assumed to be understood by the theorist who uses the meta-language to state the truth conditions of the sentences mentioned on the left-hand side of the shema.
The reason why it's a biconditional simply is because if the T-shema were rather a simple conditional such as:
(1) "The cat is on the mat" is true if the cat is on the mat,
then this would leave the truth value of "the cat is on the mat" undetermined in all cases where the cat isn't on the mat. But we want to stipulate that "the cat is on the mat" is false when the cat isn't on the mat; hence the biconditional. Tarski's intention never was to imply that truth values of object-language sentences determine what can be truly be said in the meta-language.
(1) "The cat is on the mat" is true if the cat is on the mat,
then this would leave the truth value of "the cat is on the mat" undetermined in all cases where the cat isn't on the mat. But we want to stipulate that "the cat is on the mat" is false when the cat isn't on the mat; hence the biconditional. Tarski's intention never was to imply that truth values of object-language sentences determine what can be truly be said in the meta-language.[/quote]
It might not have been his intention but the logic of a biconditional is such that it can be read in either direction.
I'm not sure how this makes a difference. You accept that if "X" and "Y" are synonymous then "X is Y" is true and you accept that if "X is Y" is true then X is Y. So it's a straightforward transitive relation to conclude that if "X" and "Y" are synonymous then X is Y. If the premises are true and the conclusion is a valid derivation then the argument is sound.
As I explained, just because the connective "if and only if" is used doesn't entail that the conditionals used signify material implications rather than subjunctive conditionals. In Tarski's case, it's the latter that's signified since the circumstances where the antecedent is evaluated range over all possible circumstances (and not just actual circumstances) where this mentioned string of words might be used. Both the "if" and the "only if" signify subjunctive conditionals, and both of those must be read from right to left, since we want the meaning and truth value of "the cat is on the mat" to be determined in all circumstances including circumstances where the cat isn't on the mat.
For instance, the T-shema instanciation:
(1a) "The cat is on the mat" is true iff (i.e. in all cases and only those cases where) the cat is on the mat
is equivalent to the conjunction:
(1b) "The cat is on the mat" is true if the cat is on the mat and "The cat is on the mat" is false if the cat isn't on the mat.
Those all are subjunctive conditionals and they are all meant to be read from right to left; that is, the meanings of the words used on the right hand side are held fixed for purpose of stating unvarying truth conditions meant to apply in the whole range of possible circumstances where the truth of the mentioned sentence (on the left-hand side) is to be stipulated.
This is why use of a biconditional is needed. But it doesn't really matter. You don't need to change the meaning of Tarski's T-shema in order justify your own use of it in a different context; and I think I now have a better grasp of what you are driving at.
Yes, if, in fact (i.e. in the actual world) "X" and "Y" are synonymous, and hence have the same referent (Bedeutung) and the same Fregean sense (i.e. they have the same use in the language) then it is also true that X is Y. Hence you can say that:
(1) If "X" and "Y" are synonymous then X is Y
But this must be understood as a material implication, and not a subjunctive conditional. It says that if the antecedent is true, in the actual world, then so is the consequent. It doesn't say anything about counterfactual circumstances. (Though it might be construed as a subjunctive conditional where the antecedent ranges over epistemically possible circumstances rather than alethically possible circumstances; this would make sense if we don't actually know whether, in the actual world, the antecedent it true; that is, s/he who makes the statement doesn't know what either "X" or "Y" mean).
So, this sensible reading would seem to be sufficient to support your point that meaning is use but need no land you in a pickle where you seem committed to infer:
(2) If "X" and "Y" were (counterfactually) synonymous then X and Y would be numerically identical (even though they actually aren't numerically identical).
which is either a misuse of language or expresses a metaphysical impossibility due to the necessity of identity (argued for by Ruth Barcan Marcus and Saul Kripke). But you really don't need that in order to convey your main point, it seems to me.
(1a) "The cat is on the mat" is true iff (i.e. in all cases and only those cases where) the cat is on the mat
is equivalent to the conjunction:
(1b) "The cat is on the mat" is true if the cat is on the mat and "The cat is on the mat" is false if the cat isn't on the mat.[/quote]
1b) is:
(C ? P) ? (¬C ? ¬P)
Using transposition this gives us:
(C ? P) ? (P ? C)
Which is material equivalence.
Nothing. I take it that there are certain things implicit in that syllogism, so that it can be reformulated as follows:
If "P" means "man" in language L, and I am a man, then I am a P in L.
If "P" is replaced with "rabbit", then the argument is still sound.
Of course, I am not a rabbit (in accordance with the English language). But if the English language were to change such that the current meaning of "rabbit" became obsolete, and if it also gained a new meaning equivalent to that of the current meaning of "man", and if I am still a man at that time, then it would be correct at that time for me to state in the English language of that time that I am a rabbit.
It should go without saying that it does not follow from the argument in the paragraph directly above that by making that statement, I would be implying that I am a small mammal in the family Leporidae.
I would not have become a rabbit. I was a man before, and would remain a man. It'd also be correct at that time to state that I was a rabbit before, and am a rabbit now. That's because the meaning of "rabbit" at that time wouldn't cease to apply to past times when it wasn't used in that way. Otherwise, if you apply that rule in general, you end up with all sorts of absurd logical consequences, like that two homosexuals weren't gay before the word "gay" was used to mean homosexual. On the contrary, they were gay on account of their homosexuality.
Only if you insist on reading "?" to signify material implication. And this is a rather bad misconstrual of the significance of the T-shema instantiation, as I have explained. But you had suggested that the biconditional form shows that the correct reading is material implication rather than subjunctive conditional. This is a non sequitur since the fact that the statement can be written "(C ? P) ? (¬C ? ¬P)" or "(C ? P) ? (P ? C)" tells you nothing whatsoever about the significance of "?". Instead, you have to reflect a little about the pragmatic significance of the shema in the context of the truth theory it is pulled from. It is this pragmatic significance (i.e. how Tarski's truth theory is meant to be used) that recommends the subjunctive conditional interpretation, as I have explained.
Yes, I agree that that's a valid [i]reductio ad absurdum[/I], and it seems to me that @Michael must either revise his position or bite a bullet that makes his position implausible.
However, one could avoid your above criticism if one commits to an altered version of the claim:
For any sentence "P", if P, then "P" is true for all cases in which "P" can be formed; and for all cases in which "P" can't be formed, then "P" would be true [i]if it was formed[/I].
Quoting The Great Whatever
And, again, to avoid the above, one could revise one's commitment to account for such circumstances. In this case, one could commit to the claim that whether a sentence is true or not depends on the way that it is - or would be - used in the the relevant circumstances. Then one can accept that the universe existed before language existed. At that time, "The universe exists" wouldn't be true, but it would be true if the aforementioned sentence was formed, and if the words of which it consists were used in the relevant way.
However, abandoning the standard formulation might have unintended logical consequences that I haven't considered.
You don't actually have to tie up the truth of an assertion, or of the linguistic expressions of a thought (that may have a force different than that of assertion or belief) to the circumstances that hold at the time of the utterance. One can equally say yesterday, or today, or tomorrow, in different manners, that Smokey the cat was on the mat yesterday at 11 o'clock. This very same thought would have been expressed yesterday with the situational sentence "Smokey the cat was (or is, or will be) on the mat today at 11 o'clock" or expressed the day before with the situational sentence "Smokey the cat is going to be on the mat tomorrow at 11 o'clock". This whole system of situational sentences enables one to express the same thought, with the same truth conditions, at different times, while making use of the time of elocution, in addition to the form of the speech act used, to determine the temporal thought being expressed.
Hence, one could say that, e.g.:
(1) "There were/are/will be triceratops roaming the Earth" is true iff there were/are/will be triceratops roaming the Earth.
In this case, "were/are/will be" signals the availability of a system of situatonal sentences. This means that the sentence "There are triceratops roaming the Earth" could (conceivably) have been used to express a truth 68 million years ago. But, more importantly, it also means that whatever though would have been expressed back then in that way is the very same thought that we can express now with the sentence "There were triceratops roaming the Earth 68 million years ago". Hence, the statement of the truth conditions of (the thought expressible by) a sentence doesn't require that there actually be anyone able to utter the statement at the time when its truth value is being evaluated, since we still are able to evaluate the truth of the very same thought (concerning past events) as expressed now with the use of a situational sentence that is part of the very same unitary system that allows the expression of this thought at any time.
(This is further discussed in Gareth Evans' The Varieties of Reference, under the heading of "dynamic thoughts" and in Sebastian Rödl's Categories of the Temporal, from whom I borrow the phrase "situational sentence")
Don't you? But if you don't, then that'll have logical consequences which might be unacceptable.
Quoting Pierre-Normand
But it's not the very same thought, is it? Nor is it saying the same thing in different manners; it's saying different things in different manners. None of the thoughts, nor the sentences to which they respectively correspond, are equivalent to each other - despite them all referring to the same event (or potential event, in regard to the future).
Quoting Pierre-Normand
This part of your post made more sense to me. So, to avoid TGW's criticism, you're saying that at a time before sentences such as "There are Triceratops roaming the Earth" could have been expressed (given that there was no one there to express such a sentence), but in which there would in fact have been Triceratops roaming the Earth, we can use "were/are/will be" in the relevant sentence to mean that there could [i]conceivably[/I] have been someone there to express such a sentence (despite there not [I]actually[/I] having been anyone to do so), and, by implication, such a sentence would have been true at the time?
If that's a better way of doing it then my suggestion, then great.
Thoughts (or judgments) that are thus kept track of through displacement in space, or through keeping track of the passage of time, are called dynamic thoughts by Gareth Evans. To be able to entertain such dynamic thoughts, and master the system of situational sentences that relate their different forms of expressions at different times (and different locations) is a condition for being able to entertain them at all. For else, one would never be able re-express the very same empirical thought at two different occasions, and one wouldn't even be able to re-affirm, or contradict, or empirically verify, or infirm, an empirical thought that one had previously entertained.
I think this may clash a little bit with intuition not because it contradicts common sense but rather because it doesn't mesh well with the way propositions with empirical content are commonly treated in philosophy, and, in particular, with the use of predicate logic and tense logic. There is a tendency, in analytic philosophy, to make time figure as part of the content of empirical propositions. This is implicitly assumed when time is represented by the tense of a verb.
When time is rather understood, in a more Kantian way, not as part of the content of an empirical thought (i.e. a thought that relates directly to a possible experience) but rather as part of the form of this thought, then counterintuitiveness of the claim that the two sentences "I ate eggs this morning" and "I ate eggs yesterday morning" can express the same thought is alleviated. We have to remember that we don't perceive the time at which we perceive sensible things in addition to perceiving those things (even when there is a clock nearby). One's ability to rationally relate those two forms of expression (about eggs) is constitutive of one's ability to keep track of time and hence, also, to think temporal thoughts and grasp their logical forms. This may need to be argued more fully, but I am veering off topic. I was hoping to come back to Martha's ability (or rather, lack thereof) to genuinely express thoughts.
Quoting Pierre-Normand
The discussion veered (or at least drifted) off topic many, many pages back.
But, going back to the topic, I doubt whether I'd have much to say that hasn't already been said, and probably said in a better way than I could. I found myself in agreement with early posts by yourself and Marchesk, which reflect the view of Searle, so I am therefore more in agreement with Searle with regards to the Chinese Room than against him.
This post earlier on made a good point, I think:
Quoting Marchesk
Yes, I agree with Marchesk, and with Searle, that an ability merely to respond to external stimulations in accordance with algorithmic rules can't, in itself, constitute understanding.
Searle, however, believes that human verbal behavior is meaningful because the intentional content of speech acts are derivative from the intentional contents of the mental acts standing behind them, and he also believes that the latter contents ("intrinsic intentionality") are an emergent biological property instantiated in some mysterious way in human brains.
I agree with Michael that this is a mistake and that the proper place to look for understanding and intentionality is the public behavior of an agent in the world; and meaning is thus best reflected in the use of linguistic expressions. I think such an agent, though, must be a living rational animal and can't be a computer. Michael may be disagreeing with this. Our long digression may have just begun to touch on some disagreement about what forms of public behavior are constitutive or expressive of genuine ("intrinsic") understanding and intentionality.
"the cat is on the mat" would be true if the cat were on the mat and "the cat is on the mat" wouldn't be true if the cat were not on the mat.
Which is (from here, where ">" differs "?"):
(C > P) ? (¬C > ¬P)
And transposition applies to subjunctive conditionals as well, which gives us:
(C > P) ? (P > C)
And this subjunctive biconditional, like any biconditional, can be read in either direction, and so as:
The cat would be on the mat iff "the cat is on the mat" would be true.
I agree. The Great Whatever's criticism is due to (mis-)interpreting the statement as the above.
X iff "X" is true follows from "X" is true iff X.
"X" is true iff X follows from If X then "X" is true and If not X then "X" is false.
So if you want to reject X iff "X" is true then you must reject either If X then "X" is true or If not X then "X" is false. But such a rejection would allow for the situation where the cat is on the mat but "the cat is on the mat" is false or where the cat is not on the mat but "the cat is on the mat" is true. Isn't that a reductio?
I don't think that that [i]reductio[/I] is more compelling than TGW's. Being forced to conclude that nothing existed before language is worse than allowing the logical possibility that, e.g. the cat is on the matt, but "the cat is on the matt" isn't true. But the latter isn't necessary, anyway. It's only necessary for times in which language doesn't exist. One can maintain that for all times in which language exists, it is impossible. Or one could maintain that it is possible at all times, but rarely - if ever - has been the case since the existence of language.
Why is that the conclusion? All I've done is reversed the order of the T-schema. X iff "X" is true means the same as "X" is true iff X.
So let's see if it does:
Premise 1: nothing existed before language iff "nothing existed before language" is true
Conclusion: Therefore nothing existed before language
Certainly doesn't follow. There's a missing premise in The Great Whatever's alleged reductio.
And yet that would entail dialetheism, wouldn't it? So supposing that the conclusion is that nothing existed before language, isn't this metaphysical reconsideration more reasonable than the logical possibility of a contradiction?
I'd say it is. The choice between a counterintuitive conclusion and a contradiction is an easy one.
How can you explain the state of affairs before language? Would that not be an example of X without "X" being true, on account of there being no language, and therefore no "X"?
By using language. How do we explain anything?
Are you trolling me? I'm not amused by your reply. Please explain to me in sufficient detail how you would explain the state of affairs before language without having to concede that at that time, it would be the case that X, but "X" would not be true, and therefore, that "X iff 'X' is true" was not true at that time. Or perhaps you accept that "X iff 'X' is true" was not true at that time. Do you?
The T-schema doesn't say X happened iff "X is happening" was truthfully said at the time; it says X happened iff "X happened" is true.
If dinosaurs walked the Earth then "dinosaurs walked the Earth" is true. If dinosaurs didn't walk the Earth then "dinosaurs didn't walk the Earth" is false. Therefore "dinosaurs walked the Earth" is true iff dinosaurs walked the Earth.
The argument is valid. So if you want to reject the conclusion then you must reject one of the premises. Which one do you reject?
And I didn't say that the fact that the sentence wasn't said at the time is problematic. I said that the nonexistence of language is problematic. If language didn't exist, then sentences didn't exist. If sentences didn't exist, then they could not have been true.
Do you therefore accept that "X iff 'X' is true" was not true at that time?
This is like asking "do you accept that the King of France is not bald?" It's a nonsensical question.
Again, if you want to reject "X" is true iff X then you must reject either "X" is true if X or "X" is false if not X.
So which one do you reject?
How so? There is no King of France, yet there was a time before language. I am simply asking whether or not you think that, at that time, it would have been the case that X, but "X" would not have been true.
Quoting Michael
What you've said doesn't make it irrelevant for reasons I've already explained. Are you trying to shift the burden? Being forced to commit to the claim that [i]nothing existed before language[/I] is a damning logical consequence, and the only way that I see to avoid it is to accept that "X iff 'X' is true" was not true at that time. If you would word that last part differently, then speak up.
You're asking if a non-existent sentence was or wasn't true. But that's like asking if the non-existent King of France is or isn't bald.
I'm not shifting any burden. I'm telling you that if you reject "X" is true iff X then you must reject either "X" is true if X or "X" is false if not X. I want to know which. You're free not to answer, but I'd consider that quite telling.
I'm not agreeing with that conclusion. I don't think it follows. But even if it did, being forced to reject either "X" is true if X or "X" is false if not X is a much more damning consequence as it leads to contradictions. Rather undermine your ontological commitments than undermine logic.
Which is like accepting that the King of France is not bald in the present.
Apologies (oops, I did it again).
Just as there is no King of France to be or not to be bald, there were no sentences to be or not to be true. Therefore, it could not have been the case that there was a true sentence, despite there being a state of affairs which would correspond to such a sentence if such a sentence existed.
Quoting Michael
I am rejecting the claim that [I]"X entails 'X' is true"[/I] has been the case since before language existed. Feel free to address that rejection and it's logical consequences, and I'll cooperate.
Quoting Michael
It is one horn of a dilemma which follows from the premise that whenever X has been the case, "X" has been true, [i]and[/I] the premise that there has been a time in which X was the case, but in which there was no "X" to be true. In order to avoid contradiction, you must reject at least one of those premises. I suggest rejecting just the former premise, because I think that it's more plausible that the former premise is false, and that the latter premise is true. If you reject the latter, but not the former, then you face absurdity, such as that language has always existed and/or that nothing existed before language.
Quoting Michael
I can rephrase it in a way that might avoid that objection, and I believe that I have basically done so in my first paragraph in this post. Rather than implying that there was a sentence which was not true, I am implying that there was not a sentence which was true.
I know. As I said before, the T-schema doesn't say "X is happening" was truthfully said at the time iff X happened. It says "X happened" is true iff X happened.
But nobody has said that. What I've said is that "X" is true iff X.
No it doesn't. You can't derive "X is happening" was truthfully said at the time iff X happened from "X happened" is true iff X happened.
I know that it doesn't say "X is happening" was truthfully said at the time iff X happened. But, unless it is limited to a certain period of time, it does imply that if X happened, then "X" was true. You yourself said that it works both ways.
Quoting Michael
I know that nobody has claimed that. I have said that it is a logical implication. You've also said that the biconditional works both ways, such that X iff "X" is true, which, unless it is only applicable to a certain period of time, implies that when X was the case, there must have been "X", and "X" must have been true.
Are you, or are you not, claiming that your principle applies only to a certain period of time?
Quoting Michael
To be more precise, it is one horn of a dilemma which follows from the premise that whenever X has been the case, "X" has been true, and the premise that there has been a time in which X was the case, but in which there was no "X" to be true.
Please see the edited version of my comment.
Quoting Michael
That's not what I said. Unless you're now backtracking, you said that the biconditional works both ways. I'm addressing it the other way around, and precisely as I worded it.
The T-schema doesn't say "X" was true iff X happened. It says "X happened" is true iff X happened.
What I said is that the T-schema can be read in either direction, either as 1) "X happened" is true iff X happened or as 2) X happened iff "X happened" is true. This is an elementary fact about material equivalence. X ? Y can be read as Y ? X.
And notice that 2) isn't X happened iff "X" was true.
But that's not the premise. The premises are:
"X" is true if X
"X" is false if not X
The conclusion "X" is true iff X necessarily follows and is equivalent to X iff "X" is true. So as I said, if you want to reject this conclusion then you must reject one of the above premises, but to reject one of the above premises is to invite contradictions.
But your T-schema wasn't that specific. It was general enough to imply the second version of 2).
Quoting Michael
See? Let X be: the pre-linguistic universe exists. It follows that if the pre-linguistic universe exists, then "the pre-linguistic universe exists" is true. You must maintain that this was the case at the time for sake of consistency, or accept that the principle in bold applies only to a certain period of time, or abandon the principle in bold as formulated.
"My" T-schema is just the T-schema, which is: "X" is true iff X. It doesn't imply the second version of 2).
So I must maintain that it was the case that if the pre-linguistic universe exists then "the pre-linguistic universe exists" is true? The grammar of this is all wrong. You're mixing your tenses.
Rather I must maintain that if the pre-linguistic universe existed then "the pre-linguistic universe existed" is true. Which is the T-schema.
If I abandon X iff "X" is true then I must abandon "X" is true iff X, and if I abandon "X" is true iff X then I must abandon "X" is true if X or "X" is not true if not X. This leads to contradictions.
So I ask you, ought I do this? Do you do this?
Yes, that is what you must maintain, although that is not the only way that you could word it, and I don't recall having worded it that way, so if there is a problem with the grammar in the paragraph quoted above, then it seems trivial and of your own creation.
Unless you qualify that the principle is not timeless and that it depends on certain circumstances, then you must commit to a version of what I'm trying to say which preserves the meaning. You must word it in such a way as to show that the principle would have applied at that time, in the present tense.
You must maintain that when it was the case that the pre-linguistic universe existed, at that time, it must also have been the case that there existed a true sentence: "the pre-linguistic universe exists".
Quoting Michael
You must maintain that as well, but that is beside the point.
Quoting Michael
No, in a suitably revised or qualified form, it need not lead down that path.
Quoting Michael
No, you should do what I've suggested.
There is no way to avoid it. I've provided you the argument:
"X" is true if X
"X" is false if not X
The conclusion "X" is true iff X necessarily follows and is equivalent to X iff "X" is true.
Do you agree with the premises?
No I don't. The T-schema doesn't imply this. The T-schema only says that if the pre-linguistic universe existed then "the pre-linguistic universe existed" is true.
Assuming that the conclusion is true, does it follow that it has always been true? No. Which is my point. I am arguing in favour of the position that the conclusion is dependent on time and circumstance.
Quoting Michael
I have given you the opportunity to qualify whether or not the principle is timeless and dependant on certain circumstances, but you have not done so. I don't know what else I can do. I have demonstrated that it does imply what I've said that it implies if it is timeless and independent of circumstance. However, if you qualify that it's dependent on time and circumstance, then TGW's criticism - which I have reiterated - can be avoided.
That doesn't seem right.
*edit
I apologize for my brevity; it's more of a place holder while I collect my thoughts. I'm a slow thinker, but wanted to put something down to commit to add more.
Edit:
Quoting Soylent
Ok.
It's true and never has been or will be false. Furthermore, its truth is not dependent on special circumstances. It's necessarily true.
Your words are sounding like the King of France example again.
It was false at the time of the pre-linguistic universe, but you don't realise that, and want to have your cake and eat it.
It can't be false. It follows from premises which can't be false.
The conclusion was false at the time, because it was not the case that there was a corresponding true sentence. The contrary is far too implausible to accept. So, if your argument is valid, then, at the time, at least one of the premises must also have been false.
They can't be false else we'd have a contradiction.
"P" is true if P
"P" is false if not P
We can't allow that "P" isn't true even if P or that "P" isn't false even if not P.
If logic is inconsistent with intuition then so much the worse for intuition.
No, you're overlooking the circumstances of the scenario under discussion. In a pre-linguistic universe, there is no "P" to be true or false. The concept of truth and falsity requires language.
Then there is no T-schema to be false. So, as I said, the T-schema must always be true.
As does the concept of cats on mats. ;)
I didn't say that there would be a T-schema in a pre-linguistic universe. You have claimed or implied that the T-schema would still apply in that circumstance, and I have rejected that. It would've been the case, at the time, that the pre-linguistic universe exists, but there wouldn't have been a corresponding true sentence.
You said "[t]he conclusion [the T-schema] was false at the time" (and "at the time, at least one of the premises must have been false"). How could it (or they) be false if they weren't "there"?
No I didn't. I have only ever said "X" is true iff X (and X iff "X" is true). I've never said X happened iff "'X' is true iff X" was truthfully said at the time.
Which is why I said that the pre-linguistic universe existed iff "the pre-linguistic universe existed" is true and not that the pre-linguistic universe existed iff there was a corresponding true sentence at the time.
Also, I think your tense is wrong again. Would have been the case that the pre-linguistic universe exists? Compare with; would have been the case that the Nazis win the war. Does that make sense? I don't think so. Should be; would have been the case that the Nazis won the war, and so; would have been the case that the pre-linguistic universe existed.
Fine, then I retract that claim. It isn't necessary. I was suspicious when you brought up falsity, and now I realise that I was right to be suspicious.
There would not have been a true conclusion, nor any true premises, at the time. Nor would the latter part of the T-schema apply at such a time, which is to say that there would not have been a corresponding true sentence. This is what I have consistently claimed, with the exception of your successful - albeit shallow - ploy.
Quoting Michael
What? You seem to have reverted back to a misinterpretation that I thought you'd set aside. I have addressed what's in the brackets. It wasn't true in the pre-linguistic universe. There were facts, but no sentences, and therefore no true sentences.
What's the problem?!
The former is a separate, nonequivalent claim than the one that I've been addressing, and the latter is more or less what I think you have - perhaps inadvertently - implied.
Quoting Michael
That's misleading, if not outright incorrect. The pre-linguistic universe existed, and, at the time, it would be the case that the pre-linguistic universe exists. Similarly, the Nazis shot Jews, and, at the time, it would be the case that the Nazis are shooting Jews.
The qualification "at the time" means that you must consider that point in time with regard to what would have been the present state of affairs, and the present state of affairs would have included: the pre-linguistic universe exists.
What's the problem?![/quote]
The problem is that I've never said that there were true sentences, and nor have I implied it. I've simply taken two necessarily true propositions and performed the relevant logical manipulations.
The important thing to note is that the following two sentences are equivalent:
It is the case that P
"P" is true
So what you're saying can be re-written, with no change in meaning, into "the universe existed and, at the time, 'the universe exists' would be true".
No, the problem is that you don't realise what you've implied, but how about we just agree to disagree, and leave it at that?
Rather than focus exclusively on what it looks like, you ought to focus a little more on how it is used. Remember that it is you yourself who introduced explicitly a contentious counterfactual conditional statement, insisted that it be read as such, and invoked Tarski's T-shema in order to justify it. My rejoinder was that you were reading the T-shema in the wrong direction.
The sort of claim that you wish to defend is:
(1) If "Smokey the cat is on the mat" is true, then Smokey the cat is on the map.
And you also insisted that your claims should be understood as counterfactual conditionals (one such claim was concerning horses, rabbits and synonymy).
It is true that (1), interpreted as a material inference, can be validly inferred from
(2) If "Smokey the cat is on the mat" is true if and only if Smokey the cat is on the map.
However, when read as a subjunctive conditional in which '"the cat is on the mat" is true' is the antecedent, (1) becomes a misrepresentation of (2) as it is intended to be read in the context of a Tarskian truth theory.
That's because instanciations of the T-shema (as derived from the meaning assignment axioms of the theory) are meant to be interpreted as a recipe, or instruction manual, that tells you how, given specific worldly circumstances (e.g. circumstances either actual or counterfactual where horses are or aren't rabbits, or where Smokey the cat is or isn't on the mat) sentences in the object language ought to be evaluated as true or false.
This is why (2) must be understood as stating the conditions under which "the cat is on the mat" is correctly evaluated, in whatever object-language is being formalized by the truth theory, rather than as stating what the conditions in the world would be (counterfactually) if the mentioned object-language sentence were true. It can't mean both, for in that case the account would be viciously circular.
Not at all.
Then "of course it is" is equivalent to "of course it isn't".
It is the case that my name is Michael
"My name is Michael" is true
"My name is Michael" is true and it is not the case that my name is Michael.
Yes that's correct so long as you hold fixed the range of actual+counterfactual (i.e. 'possible') circumstances in which the implication sign can be interpreted as material implication. Saying that P > C is equivalent to saying that in all possible worlds at which P, it also is the case that C. Hence, relative to this very same range of possible worlds, it also follows that whenever ¬C, it also is the case that ¬P. But this needs not have the same significance as ¬C > ¬P, since the range of possible worlds that ¬C singles out may be a different range. And that is indeed the case where your contentious interpretation of the T-shema is concerned. For the case where the T-shema is correctly interpreted, the relevant range of counterfactual circumstances includes, precisely, worldly circumstances (i.e. ways for things to be) relative to which the truth of object-language sentences are to be evaluated in accordance with their (indirectly) stipulated meanings. There is no ranging over other possible (i.e. non-actual) reference assignments to the object-language terms.
Hence, you are licensed to say, on the basis of the T-shema instanciation previously discussed, that in all cases where Smokey the cat is (or would be) on the mat, "Smokey the cat is on the mat" (in English) is (or would be) true, and in all cases where Smokey the cat isn't (or wouldn't be) on the mat, "Smokey the cat is on the mat" (in English) is (or would be) false. Hence, material equivalence holds relative to a specific range of circumstances. This range of possible circumstances, envisioned by the T-shema instanciation, is a range of circumstances in which Smokey is located at various places, not all of them "on" the mat. But, relative to all the possible worlds in that range, the meanings of the English words used to state the theory, and the meanings of the object-language words (which can be the same language as the meta-language) are held fixed. It's precisely because they are held fixed that the T-shema doesn't licence the ¬C > ¬P subjunctive conditional claim that you want to derive from it, where they would be allowed to vary over circumstance of linguistic use in which meaning assignments to the words of the object-language would be different than those that are intended by the specific truth theory that this T-shema instanciation is a theorem of.
It follows from this that in all cases where "Smokey the cat is on the mat" (in English) is (or would be) true, Smokey the cat is (or would be) on the mat, and in all cases where "Smokey the cat is on the mat" (in English) is (or would be) false, Smokey the cat isn't (or wouldn't) be on the mat.
If we go back to the example you gave earlier...
... we can consider the case of how given the worldly circumstances some dessert ought to be evaluated as a red velvet cake (or not):
This is a red velvet cake iff this recipe was successfully followed.
If the above is true then the below is true.
This recipe was successfully followed iff this is a red velvet cake.
So it doesn't matter whether you explain it in terms of material or subjunctive equivalence or in terms of instructions for evaluation; it can be read in either direction.
I knew that is what you would say, that you would pick the second option. I was interested to hear what Pierre would say.
1) It is the case that my name is Michael and My name is Michael are equivalent.
2) My name is Michael and "My name is Michael" is true are equivalent.
3) It is the case that my name is Michael and "My name is Michael" is true are equivalent.
The first premise should be self-evident. The second premise follows from a) "My name is Michael" is true if my name is Michael and b) "My name is Michael" is not true if my name is not Michael. The conclusion applies a straightforward transitive relation.
The only way you can reject the conclusion is if you reject either a) or b), but we've already established that that leads to contradictions.
I don't know how much simpler to put it.
I know that your question wasn't addressed to me, but that isn't going to stop me from answering. Only the latter follows, or at least it does so if you're talking about the history of Earth, and not some hypothetical Earth which could have an alternative history.
The disquotational schema as you are using it is simply false, and your whole notion of how these things works seems to be predicated on it. So it would be best to return to why it's false.
The proposal:
"X" is true iff X.
Let's show that neither direction holds. First, left to right:
If "X" is true, then X.
Suppose that in an alternate state of the language in the future, "X" means the same as "not X" means now; that is, "X" means that not X. Suppose that further, in this situation, not X. In such a situation, "X" (that sentence) is true, yet it is not the case that X. In fact, the truth of that sentence guarantees just the contrary, that not X. So this conditional is false.
Now the right to left direction:
If X, then "X" is true.
Take the case of a time before there is languages, and let X be that there are dinosaurs. In this case, there are dinosaurs, yet it is not the case that "there are dinosaurs" is true, since no such sentence exists ex hypothesi, and a fortiori no such sentence is true. So this conditional is not true either.
So the disquotational schema is false, since neither of its conditional directions holds. Without it, this entire line of thought is fruitless.
---
I think, in the end, you confuse claiming that X with something like saying "X," which is the only way I can make sense of your line of thought -- but this is fallacious, as shown above. Sentences do not in any way talk about themselves and claim themselves to be true -- that would require a mention of the sentence in order to predicate truth of it. Yet when sentences are used, they talk about the things they talk about, like horses and rabbits. The truth of a sentence, and the obtaining of the state of affairs that a sentence describes, are simply not the same thing; your philosophy of language is basically and fundamentally wrong and hinges on a use-mention error.
They are not equivalent; one is about a sentence, the other about a name. The properties of the sentence might change, while the name stays the same, or vice-versa.
Look, none of those are equivalent in meaning, no matter how determined you are in your attempt to make them so. If you can't see the admittedly subtle differences, then that is not my problem. They are all similar and related, but that does not make them equivalent.
The closest in meaning are the two sentences in 1), but even those two sentences do not have exactly the same meaning. Although, if the latter is asserted, then the "It is the case that..." would be implicit. In that sense, and in that sense only, are the two equivalent.
Quoting Michael
No, that doesn't follow. That one logically implies the other does not mean that they're equivalent in meaning. That is your arbitrary interpretation. They aren't equivalent in meaning for the reason that I stated.
I thought that you'd notice the difference, too. I'm glad that I'm not the only one to object to Michael's attempt to conflate the two.
I'd add that one is about that which is true (a sentence which satisfies certain truth conditions; language), whereas the other is about that which is the case (a fact or state of affairs; the world).
That's exactly right. But notice that it doesn't follow that in all cases where "Smokey the cat is on the mat" is (or would be) true in some other language than English, in 'New English', say, Smokey the cat is (or would be) on the mat. And yet this is what you were saying. Even though you actually meant your claim to be interpreted as if you were expressing the consequent in 'New English', this use isn't warranted by the normal interpretation of the T-shema.
(On edit: I had missed the second part of your comment, so I responded to that below)
Yes, I agree with Michael. Judgments that are true at the time when they are judged or expressed are true at all times. If one correctly judges at one time that Smokey the cat is (at that time) on the mat, then this judgment remains true at a later time when Smokey has wandered off the mat. The very same judgment (the content, not the speech act) then can be re-expressed with the use of a different 'situational sentence' that includes a verb in the past tense.
It seems to me that you and TGW may be making too much of that. Even though one sentence has an English expression as its grammatical subject, it mentions it (and refers to its meaning) as an indirect means of making a claim about what is the case in the world, whereas the other sentence makes that very same claim directly. Michael is right to point out that they are logically equivalent in that respect -- in that they logically imply one another -- assuming only that the meaning of the mentioned sentence is held fixed and taken to be its ordinary meaning in English. It is through forgetting this necessary assumption (in counterfactual contexts) that Michael sometimes run into trouble, it seems to me.
There is a crucial disanalogy that you are overlooking. Correctly following the recipe for a velvet cake ensures the production of a velvet cake, let us assume. However, correctly following the semantic rules of a language doesn't ensure that "Smokey the cat is on the mat", when correctly evaluated to be true according to those rules, implies that Smokey the cat is on the mat. That's only guaranteed to be the case when the semantic rules are those of the English language. If they are the semantic rules for another language, then it may be the case that "Smokey the cat is on the mat" is correctly evaluated to be true according to those rules while Smokey the cat isn't on the mat.
OK, so, that dinosaurs were walking the earth, although true now, was not true at the time. But now that we have judged it to be true that they were walking the earth it will be true for all time, even at some time in the future, when there are no humans?
No, that's not what I meant to imply. That dinosaurs were walking the Earth is both a fact and the content of a true judgment. The content of the judgment is what is shared between different people who judge this content to be true. It also can be the content of other propositional attitudes such as hopes, fears, or it can figure in more complex thought such as, e.g. being the antecedent of a conditional.
All I meant to convey rather is that, if there had been people present at the time when dinosaurs were roaming the Earth, and who would have been in a position to judge this to be the case, and who may or may not have expressed this judgment using whatever language that had sufficient conceptual resources for expressing it, those people would have been entertaining the very same thought content that we now are able to express with a past tense statement. The very same judgment expressed by them, then, can be expressed by us now. So, its being true doesn't depend on them, or us, existing at all. The content of any judgment actually entertained by someone at a specific time (i.e. its truth conditions) necessarily must be ascertained as a function of the concepts employed (reflected in the use of the words that express them) but the truth value of those judgments only depend on what is the case in the world at the relevant time (and not necessarily at the time when the thought is entertained).
I had explained this earlier in rather more details here, here and here.
I stand by my claims. The two sentences (both in their more generalised and logical form, and in certain specifically worded forms) are not equivalent in important ways: both in terms of meaning and in terms of logical consequence. It is because of this that it is the case that P does not entail that "P" is true (but only if certain conditions are satisfied), and that, therefore, no contradiction is implied by committing to a case in which the former holds, but not the latter. And, because of this, the pre-linguistic universe counterexample stands.
Yes, I think we can agree that in the distant past, when there weren't any language users around, and hence there were no rules governing the use of the words employed in "P", it was still the case that P. That it was the case that P can be expressed by us with the sentence "P", which is true if and only if P, right? Hence it is correct to say that the two sentences (1) "P" and (2) '"P" is true' are logically equivalent, which can be expressed thus:
"P" is true if and only if P
For instance:
"There were triceratops around 68 million years ago" (as expressed by us now) is true if and only if there were triceratops around 68 million years ago.
So, it's not really the disquotational shema in itself that is the source of Michael's trouble. (And indeed, most logicians and philosophers of language don't have any trouble with this schema, though they may disagree on their detailed accounts of truth and meaning.)
Again, no. This is the error. Whether a certain sentence or string of words is true or not in a hypothetical situation (not now) does not guarantee that the situation that is described by that string of words in the language as it currently is now, holds in the hypothetical situation. In the hypothetical situation, "P" might very well mean not P, and so the truth of "P" could very well imply not P, rather than P.
Obviously, you overlooked my explicit qualifier "by us". This is always assumed by the logicians and philosophers of language who make use of the disquotational shema. We are not talking about counterfactual situations where the mentioned sentence has a different meaning. We are rather talking about counterfactual circumstances where its truth value varies as a function of the way the world is in those circumstances.
It's implicit in the schema that the sentence mentioned on the one side is the same sentence used on the other side.
So:
"X" is true iff X
The bits in bold are the same sentence.
If X then "X" is true follows from if "X" is not true then not X, as per transposition. And given that it's implicit that the sentence mentioned is the sentence used, if "X" is not true then not X must be true to avoid contradiction. Therefore If X then "X" is true must be true to avoid contradiction. Unless you can show that the logic of this fails (and not just that any ontological implications are counterintuitive) then any argument you have against it must be wrong in some respect.
I'm stating the T-schema where the sentence mentioned on the one side is the sentence used on the other side. So whatever language it's in, with this condition the bidirectional equivalence holds.
"X" is true iff X and X iff "X" is true.
They reference the same truth condition. So in that sense they mean the same thing, even if the cognitive content has a different focus. Consider the sentences "you are a parent" and "you have a child". The cognitive content of the first focuses on what you are and the cognitive content of the latter focuses on what you have, and yet they both reference the same truth condition and so amount to the same claim.
Yes, if you settle on a specific language, whatever this language might be, then the material equivalence holds. But then it can't be interpreted as a subjunctive conditional. And you also need to indicate that, when you choose some language L different than English, then you mean your statement of specific shema instanciations to be interpreted in L rather than in English. The default is to interpret sentences that are used as being written in English (on this forum, anyway).
That depends. If it's the disquotational shema (or the homophonic case for the T-shema) that is at issue, then, yes, the sentence mentioned is the same as the sentence used (and both are interpreted in the same language). But in the T-shema derived from a Tarskian truth theory, there is no requirement that the object-language and the meta-language be the same. The meta-language has its own semantic rules fixed and is used to specify truth conditions for the sentences of the object-language. And in both cases (either the simple disquotational schema or the T-schema instanciations of some Tarskian truth theory) the interpretation of the schema instanciations as counterfactual conditionals, where the antecedent specifies some counterfactual semantic rule for the mentioned language, is incorrect.
Yes, that is quite correct, though I would be tempted to nitpick on behalf of Frege and say that they have the same truth conditions and hence reference the same truth values in all circumstances.
Maybe "truth-maker" or even just "referent" is the better term? Although I guess this is largely semantic and makes no significant difference to the issue at hand.
That's why I clarified the previous example by saying:
Given that in this language "horse" means "rabbit"...
So given that in this language "horse" means "rabbit", 1) horses are rabbits, 2) "horses are rabbits" is true, 3) "horses are rabbits" is true iff horses are rabbits, and 4) horses are rabbits iff "horses are rabbits" is true.
That's not sufficient. You also have to say: "... and given that I've decided to write this post in this language..."
It's unclear because it sounds like "this" is used to single out the language you are making claims about, not to specify the language in which you've decided to write your post, or part of your post, let alone the second part of one single sentence in your post. Also, there are less confusing ways to make your point about meaning being determined by use, it seems to me.
Yes, it's true that nothing much hangs on the use of "refers to" here. But "refers to" usually is understood as a relation between singular terms and the objects that they refer to, or between concept words and the Fregean concepts (or determinations) that they refer to. In the case of whole sentences, they are said by Frege to refer to truth values, and the thoughts that they express are their senses. Knowledge of the truth conditions of sentences would be equated with knowledge of the senses that they express. One can know the sense of a sentence and not know whether it is true or not. One needs in addition to know the references (Bedeutugnen) of the terms this sentence is composed of. When this is known, then the truth value can be assessed through checking what's up with those Bedeutungen in the world (e.g. does the object referred to by the singular term have the determination referred to by the concept word?). This is why Frege locates the truth value of the sentence at the level of reference, and the meaning of the sentence (its truth conditions) at the level of sense (Fregean Sinn).
I do not agree with this part of your post. I don't think that (1) and (2) are logically equivalent, and I find the biconditional in that formulation problematic, because then I couldn't deny the converse, or deny one part of it, without contradiction, right? But I think that it is possible in certain circumstances that [i]P[/I] and not [I]"P" is true[/I].
It's also problematic to lose the biconditional, because then I couldn't exclude other possible truth conditions which I'd want to exclude on the basis that they wouldn't make sense.
This is false. The fact that the sentence mentioned is the same as the one used in no way shows that this must be true to avoid contradiction. Perhaps you can show how this is a contradiction?
I can show that it's not, by showing that in a given scenario, one can be true while the other is false:
Suppose X = "there are no more dinosaurs," where we slot in a use of this sentence for the variable "X."
Now, suppose that in Future English, "there are no more dinosaurs" means the same thing that "there are still dinosaurs" means now; that is, it means that there are still dinosaurs.
Now, take a situation in which people speak future English, in which there are no more dinosaurs, since it is the future. In that situation, ex hypothesi," "there are no more dinosaurs" is not true, since it means that there are still dinosaurs, and there aren't. If that sentence were true in the future situation, then there would still be dinosaurs, since that is what the sentence "there are no more dinosaurs" means in this alternate situation. Since there are no more dinosaurs, the sentence is false.
But, ex hypothesi, there are no more dinosaurs in this situation, so X obtains in it, that is, in this situation, X (there are no more dinosaurs).
So this is a situation in which "there are no more dinosaurs" is false, yet there are no more dinosaurs (precisely because "there are no more dinosaurs" means that there are still dinosaurs, and there are not by hypothesis).
There is nothing contradictory about such a situation, and therefore there is no contradiction as you suppose. Somewhere you have gone wrong in your reasoning, and this counterexample demonstrates that: it is another question where you went wrong, and I suspect, as I have said, that you are deeply confusing use and mention.
---
I have shown you why this is not a contradiction as you claim. Why do you think it is a contradiction? I've seen no defense of this other than asserting it, but given that (a) this assertion looks absurd on its face, and (b) we can construct a situation to demonstrate that it is wrong, it seems to me that you cannot go on claiming this unless you respond to this situation.
So given that both the sentence I'm using and the sentences I will mention are to be understood as English proper, if my name is Michael then "My name is Michael" is true and if my name is not Michael then "My name is Michael" is not true. To claim otherwise is to invite contradiction.
So given that both the sentence I'm using and the sentences I will mention are to be understood as English proper, if X then "X" is true and if not X then "X" is false. To claim otherwise is to invite contradiction.
"My name is Michael" is true and my name is not Michael.
How much more evident can a contradiction get?
As I see it, a sentence is a certain grammatical object -- a string of morphemes or words, or a syntactic structure, whatever you like -- and that same object can receive different interpretations. It seems for your objection to make sense, you would have to claim this is effectively not possible, which would commit you to a substantial view on the identity conditions of sentences (I'm not sure what they would be, but it looks wrong).
Thus I am not talking about some sentence in English and then some other sentence in the new English, and so I 'switch' nothing. There is just one sentence, namely this one --> "there are no more dinosaurs," and that sentence means one thing now, but could mean something else later.
For me, it's evidently just not a contradiction, and it's puzzling to me why you think it is. You have a claim about a sentence on the one hand, and a claim about a name on the other; presumably, to say these contradict is to say that one side of the conjunction cannot be true while the other is false: but clearly this is possible, as I showed above, so I'm not sure why you think there is any 'evident' contradiction at all.
There is likely a missing premise that you cannot articulate, and I suspect that when that premise is spelled out, the use-mention error will become more clear.
What circumstances would that be? Remember that philosophers who employ the disquotational schema (e.g. Tarskian truth theorists, disquotationalists, deflationists, minimalists, identity truth theorists, or prosentential truth theorists) all are using it in contexts where it is assumed that the truth conditions of the mentioned sentence are determined by means of the used sentence (on the right hand side of the biconditional). Hence, circumstances where its truth conditions would be different from the truth conditions of the used sentence are ruled out. So, it's rather like I were saying that if a natural number N is smaller than 3 then,
N is prime if and only if N = 2,
And you were to object that this biconditional is false because some natural numbers are prime other than 2.
My name is Michael. The previous sentence is false.
They can't both be true. That would be a contradiction. If one is true then the other must be false.
The example was a false analogy as the sentence mentioned was in a different language to the sentence used. In the schema I'm using the sentence mentioned on the one side is the sentence used on the other side.
[quote=The Great Whatever]@Michael, if I understand you correctly, your claim hinges on saying that the same sentence cannot exist in more than one language. Is that correct?[/quote]
No. I'm saying that given the sentence mentioned on the one side is the sentence used on the other side, "X" is true iff X. That's not to say that some other schema won't work given that different sentences are used.
I think there indeed is an additional premise that Michael is successfully articulating but that you keep ignoring, for some reason. The additional premise is that the languages of the used sentence and of the mentioned sentence are the same. When this additional premise, which amounts to a range restriction on the identity of the languages relied on to understand the sentences, is provided, then Michael's biconditional is true. The only trouble that was apparent to me was his earlier reliance on this biconditional to support some contentious counterfactual conditionals about, e.g., rabbits and horses.
I don't understand. I just showed above that this isn't true, by showing a counterexample. You responded that I switched sentences.
But I did not -- I used the same sentence, viz. "there are no dinosaurs," in both cases. In what sense are those two not the same sentence?
You say that one is in English, and the other is in the new English. I say this makes no sense -- the same sentence is being used in both languages.
But I did not -- I used the same sentence, viz. "there are no dinosaurs," in both cases. In what sense are those two not the same sentence?
You say that one is in English, and the other is in the new English. I say this makes no sense -- the same sentence is being used in both languages.[/quote]
That the same string of symbols are being used is not that the same sentence is being used. As you said, '"there are no more dinosaurs" means the same thing that "there are still dinosaurs" means now; that is, it means that there are still dinosaurs'. The sentence you're mentioning is in New English and the sentence you're using is in English.
But I'm telling you that the schema I'm using has the sentence mentioned and the sentence used in the same language.
If by "the previous sentence," you mean the sentence "My name is Michael," (and what else could you mean?) then this would obviously not be a contradiction. Viz., if "My name is Michael" meant instead "my name is not Michael," then this sentence could be true in a situation in which your name was nonetheless not Michael (in fact, given what the sentence meant, it would have to be that your name wasn't Michael).
In the current situation, it is of course true that whenever this sentence is true, your name is Michael, but that is because the sentence meaning that your name is Michael, and having the form "My name is Michael," accidentally coincide in the language as it is spoken now. However, this is not necessary: in situations in which these things come apart, one can be true while the other is false.
The latter is the meat of your proposal: that in any situation in which the sentence is true, your name is Michael. But this is false.
(Also, don't use sentences with indexicals like "my," which makes your claim doubly false for obvious reasons, but I have ignored it here for clarity).
Then we agree. What's left to discuss? Your criticisms are directed against straw men.
"X" is true
and
X
mean the same thing. As I have been at pains to show you, they do not. They are not equivalent, and one being true while the other false is in no way a contradiction. Getting to the point of this discussion, what this means is that how we use words like "horse" has nothing to do with what a horse is, or what it takes to be a horse. The latter is only intelligible if you mistakenly think, like you do, that the above two are somehow equal in meaning, rather than both accidentally being true at the same time in the present, in virtue of the current meaning of the sentence.
You're consistently ignoring this and addressing a straw man analogy where the sentence mentioned and the sentence used mean different things. I'm explicitly not claiming that the T-schema formulation applies in these circumstances.
Second of all, you are here confirming something that you a moment ago denied -- that your position depends on the claim that the same sentence cannot occur in more than one language. At least that seems to be what you say when you say this:
Quoting Michael
If I understand you correctly, your position hinges on a substantial thesis about the identity condition of sentences -- that the same sentence cannot exist in more than one language. Otherwise, the above counterexample works. Do you agree?
How do you get from:
1) "X" is true iff X
2) The "X" mentioned on the left of the above and the "X" used on the right of the above are the same English sentence and so mean the same thing
To:
3) No other language can have a sentence which uses the string "X" or have a sentence which means the same thing as the English sentence "X"
I'm asserting 1) and 2).
3) doesn't follow.
Quoting Michael
Makes no sense unless you assume the same sentence cannot mean two different things in two different situations. If one sentence can change meaning over time (which it seems to me it obviously can), then what you follow with, 'and so mean the same thing,' cannot be asserted.
If I say that you and I have the same job, I'm not saying that you and I must always have the same job. So, when I say that the sentence mentioned means the same thing as the sentence used, I'm not saying that they must always mean the same thing.
But, given that they do mean the same thing, the T-schema holds; just as given that we do have the same job, we're both professional Xs.
So, as I have repeatedly said, given that the "X" mentioned on the one side means the same thing as the "X" used on the other side, "X" is true iff X.
So in this situation, "X" is true, but it is not the case that X. -->
"There are still dinosaurs" is true, but it is not the case that there are still dinosaurs.
Therefore, the equivalence schema is false.
It fails as soon as you say "instead means that there are no more dinosaurs".
How many times do I have to repeat myself? The sentence mentioned means the same thing as the sentence used. So "there are still dinosaurs" means that there are still dinosaurs.
But you just said it doesn't always have to mean the same thing in the future. Ex hypothesi we are dealing with a situation in the future in which the sentence has changed meaning. You cannot simply stipulate that such a situation cannot happen; in fact you are committed to it being able to happen.
With respect to the language as we speak it now, I am not saying the sentence means two different things; nor am I switching sentences or even languages. All I am saying is that there is a possible situation in which that sentence is true, and yet it is not the case that there are still dinosaurs. This situation is imaginable, no matter what you say about the sentence being both mentioned and used has to mean the same thing now. And if you mean that the sentence mentioned and used have to mean the same thing always, you contradict what you just said above.
I'm not saying that it can't happen. I'm saying that the T-schema formulation that I'm using applies if the sentence mentioned means the same thing as the sentence used.
So, one last time:
If the sentence used on the left means the same thing as the sentence used on the right then "X" is true iff X.
As soon as you consider a case of the mentioned statement meaning something different to the used statement you're ignoring the antecedent of the material conditional.
If however you mean not a material conditional, but the 'in any situation...' claim, then adding this if-clause as a material condition does not help you. For this situation we're in now is a situation in which the antecedent is met; yet this situation would still be one in which the biconditional in the consequent is false, as I have showed you. So adding this condition does not seem to help you in the way you think it does.
This sort of ambiguity about the individuation condition for sentences of a language can be circumvented with the use of the word "statement" to refer to speech act forms -- i.e. expressions of determinate thoughts in language. In that way, "Snow is white", as used by English speakers and "La neige est blanche", as used by French speakers, make the same statement using two different sentences. Conversely, the same sentence can be used to make two different statements in two different languages. When Michael thus refers to an "English sentence", he is talking about the statement that is made when this sentence is used by English speakers.
I know it's trivially true. I've been trying very hard to show how trivially true it is. And yet there's been so much disagreement. And I was never making any grandiose claims. I was making the same trivial claims that I'm making now. And yet there's been so much disagreement.
It has me exasperated.
For example, the following material equivalence is true:
Russia is the largest country in the world iff my name is Patrick.
So is this material equivalence:
London is in France iff Paris is in England.
But in the way you want to use the biconditional, that is, to claim an equivalence in meaning between the thing on the left and the right, such that there is no situation you can find in which Russia is the largest country in the world, but my name is not Patrick, or vice-versa, this equivalence is clearly false.
The way it is used in the literature on the philosophy of language and theories of truth, the disquotational shema always is meant to express a biconditional that holds over a range of possible worldly circumstances. It says of the mentioned sentence that it is properly evaluated as true in the object-language (i.e. it expresses a true statement with the use of the object-language) whenever, and only when, circumstances in the world are as described by the used sentence. So, you are free to interpret the biconditional form as the conjunction of two subjunctive conditionals, or as the statement of a material equivalence. You have to remember that the possible circumstances of evaluation range over ways the world might be (e.g. where Smokey the cat may or may not be on the mat) but hold fixed, and indeed uniquely determine, the semantic properties of object-language. (And that the meta-language also is held fixed ought to go without saying),
It follows from the premises:
1) if X then "X" is true and 2) if not X then "X" is not true
If the "X" mentioned and the "X" used mean the same thing then these two premises must be true else we have a contradiction. As per transposition 2) is equivalent to 3) "X" is true if X. 1) and 3) make for a material equivalence.
It's the fact that X and "X" is true mean the same thing (where the "X" mentioned and the "X" used mean the same thing) that justifies 1) and 2).
So in this sense I'm stepping beyond the T-schema and arguing for a logical equivalence rather than just a material equivalence (and so arguing for a deflationary approach).
Sorry for quoting myself, but I want to add this precision:
The intended interpretation as a conjunction of subjunctive conditionals is equivalent to saying that, in whatever worldly circumstances you might find yourself, then, in those specific circumstances, the T-shema interpreted as a statement of material equivalence must be true. And this means that the truth value of the mentioned sentence must be ascribed to it accordingly.
For instance, if you were to find yourself in circumstances where Smokey the cat is on the mat, then for the following statement of material equivalence to hold in those circumstances,
(1) "Smokey the cat is on the mat" is true iff Smokey the cat is on the mat
the truth value "true" must be ascribed to the sentence "Smokey the cat is on the mat" in order that it be properly interpreted and used as a statement in the object-language.
But I'm getting tired of this too. I still think you haven't answered the criticism and are fundamentally mistaken, but truth be told I don't really care that much.
If there is no language.
Quoting Pierre-Normand
Yes, I know:
"P" is true if and only if P
But through biconditional elimination:
If P, then "P" is true
Which I have a problem with. Hence, the biconditional is problematic for me.
Quoting Pierre-Normand
They are ruled out because of the biconditional. Which is why it's also problematic to remove the biconditional and replace it with a material conditional. I don't want to allow the logical possibility of inappropriate truth conditions:
"P" is true if Q.
"The cat is on the mat" is true if the dog is on the bed.
Quoting Pierre-Normand
No, I don't think that that's a true analogy of my objection.
There being language users in the vicinity is not a feature of the circumstances that has any relevance to evaluating whether the English sentence "Smokey the cat is on the mat" is true when Smokey the cat indeed is on the mat in those circumstances. We can imagine some circumstance in the distant past, in the distant future, or in a distant galaxy far away, when, or where, there are no language users around. If, in those actual or counterfactual circumstances, Smokey the cat is (was, or will be) on the mat, then the English sentence "Smokey the cat is (was, or will be) on the mat" as used by us now to describe what is (was, will be, or would have been) the case in to those actual or counterfactual circumstances is true.
I don't understand this. This last statement can't be derived from the disquotational shema where it is assumed that the mentioned sentence belongs to the same language as the language in which the truth conditions are stated. I think part of the confusion comes from your considering the mentioned sentence as a free standing material object, or uninterpreted syntactical object, such as an inscription on a billboard, that is envisioned to have different conventional meanings relative to the circumstances where it is being employed (e.g. in different cities where different languages are spoken). But the sentence being evaluated (and mentioned) rather always is the sentence used by us, in the present, and in English, in order to describe what is or would be the case in a variety of possible circumstances.
No, not language users; language. Without language, there can be no sentence. If there can be no sentence, then there can be no true sentence. Yet, at that time, it would be the case that the universe exists (but not that "the universe exists" is true).
Q.E.D.
On edit: If you will bear with me, I will make my main point more explicitly in the last paragraph below, also added on edit.
The statement '"Smokey the cat is on the mat" is true' (if you allow me to stick with this example) isn't qualified by a time. It is a statement that we are making right now. It doesn't really make sense to ponder over what change it would make to the truth value of this statement if we were to evaluate it as it would have been made at a different time. Likewise, if the English statement "Smokey the cat is one the mat" is true, then it doesn't really make sense either to inquire about its truth value if the statement itself had been made by me three feet further on the left. Likewise with 'moving' the expression of this statement three hours or three billion years in the past.
One difficulty that seemingly arises is due the the tense of the verb that occurs in the sentence used to make the statement. This tense seems to make the statement dependent on the time when it is made in order to determine what strate of affairs it is describing. But this is an illusion that stems from confusing (1) the situational sentence used with (2) the statement made with the use of this situational sentence, as I had earlier suggested.
If I say right now that I am at home, and I am thereby making a statement that is true, my statement remains true in the future when I am not at home anymore. I could then re-express the same statement, if I wanted to, with the use of a different situational sentence that included a past tense verb. Those difficulties usually are glossed over in discussions of the T-schema through assuming that the truths being evaluated are eternal rather than temporal (e.g. we assume that Smokey always has been on the mat or never was). But if we are to make the account more general, we have to interpret all the indexical words and verb tenses of the mentioned sentence relative to its intended circumstance of expression in order to determine the statement being made, and hold this statement fixed while considering the range of circumstances relative to which its truth value is evaluated.
On edit: The range of circumstances just mentioned, relative to which the truth value of the statement are evaluated -- and, indirectly, the truth value of the sentence used by us to make this statement -- includes past, future, and counterfactual circumstances where we weren't yet around, aren't around anymore, or never were around, respectively.
I asked you this:
Quoting John
Michael responded with this: Quoting Michael which I take to show that he does not believe that "dinosaurs roam the earth" was true at the time that they roamed the earth. You responded by saying that you agreed with Michael. Hence I asked this:
Quoting John
And you replied that you did not mean to imply that "dinosaurs roam the earth" was not true at the time. This seems to contradict your earlier statement that you agreed with Michael, so now I am confused as to what you do beleive.
That's because I understand "...is true" as predicated of a statement to be tenseless. The statement at issue is the statement expressed by us, in the present, with the use of the mentioned sentence. Statements that are true at one time are true at all times. This is more apparent in the case of so called eternal truths, such as mathematical truths, but is also true of temporal truths (i.e. truths that pertain to something being in a definite state at some definite time). The latter is somewhat hidden by the fact that we represent statements with the sentences used to express them. For instance, if Smokey the cat is on the mat right now, the sentence used by me in the future to express this very same fact (which concerns what's up with Smokey right now) will be different. This sentence will use a past tense.
So, when someone claims that "dinosaurs roam the earth" was not true at the time, one may mean that the present tense sentence "dinosaurs roam the earth" would not have expressed a truth if it had been used by an English speaker in the distant past. It could also mean that the statement made by us through using that sentence right now doesn't truly ascribe a property to the Earth that it would have exemplified in the distant past. (It doesn't really matter if you hold either "... roam the earth" or "dinosaurs roam ..." to be the predicate). But on that second interpretation, the sentence is understood as an open sentence that includes a time variable. Saying then, now, that it "was true at the time", is meant to fill up the time variable in the mentioned sentence in order to express with it a definite statement (expressing a timeless truth) that is then properly evaluated true, and tenselessly so.
It is the second interpretation that I favor, since it is how the T-schema, and disquotational schema, are meant to be understood, with the mentioned sentence held to express a definite statement (i.e. a definite Fregean though) rather than as an open statement (i.e. a Fregean concept) that is predicated of a time.
So, I am agreeing with Michael that the disquotational schema, or the T-schema, are valid even as they refers to past circumstances before any human language was in use. But I disagree about the interpretation of the schema often incorrectly foisted on it, possibly by Micheal, but also by other participants in this thread too. And this is not just a matter of convention, but reflects are deeper philosophical point about the tenselesness of "...is true" as revealed by logical grammar, though somewhat hidden by English grammar.
OK, thanks Pierre, that clears it up somewhat. I have had extensive arguments with Michael about this very point in past threads. The one I can remember (and if I remember it incorrectly he can correct me) is that he claimed that although Everest may have been the tallest mountain at some prehuman time, that it could not have been true at that time that it was the tallest mountain, because no statement of that truth could have been uttered at that time.
This seemed just plain wrong to me, although I acknowledged that no truth-apt statement could have been uttered at that time. And to assert this wrongness does seem to commit one to the idea that truth is not a property of statements but of the propositions that are expressed by them and also to the idea that there must be, in some unimaginable way, unexpressed propositions. Then this begins to look like a form of Platonism.
Yes, this raises issues regarding the ontological status of propositions (Fregean thoughts). But maybe this is a topic for another thread; or possibly suited for the now dormant Pattern and Being thread ;-) (Hint: a proposition can be regarded as, in a sense, a possible pattern -- i.e. a possible way for our world to be -- while a true proposition would be grasped when one objectively refers to an actual pattern -- i.e. a way our world can intelligibly be thought to be, and, indeed, is. Also, in the previous sentence I have used "to be" and "is" tenselessly. Our world being different in the past makes a difference to how it "is" in that sense)
I am interested in these kinds of questions, and I want to get back to the Haugeland paper, and hopefully be able to make some comments about it, but I've been so busy with work lately, and now I'm a bit burned out, so...I'll have to return to it when I can.
No worry, and no hurry. The end of this month isn't a deadline either. The discussion about Brassier's paper spilled over the next couple months and there are outstanding issues there too (such as the "argument from ancestrality".)