E
epan
Guest
Don’t malign omphaloskepsis. I’ve seen some great navels.
- Status
I
inocente
Guest
Some people have very interesting hobbies.
P
Peter_Plato
Guest
Apparently, you have no “I” since you refuse to answer a simple “I” question. Therefore, statements such as "I’m a winner,’ or “I’ll ask again,” have no referent and are meaningless (which is the reason I did not directly respond to “your” post, and to whom I no longer have an obligation to fulfill a promise not to ever reply to your posts again since clearly there is no “you” to whom I am obligated.
Or are you claiming there is a “you,” after all?" I.e., By making statements such as “I’m a winner,” and “I’ll ask again.” So, apparently, the experiment has flushed out the evidence: your claim to “be” a winner is a claim to exist as a subject.
In which case, you can ignore this post since it isn’t addressed to you anyway.
M
MPat
Guest
You have offered one yourself:Sounds like I’m on a winner hereso I’ll ask again - You claimed “qualia are perfectly falsifiable”. Come on then, where’s the experiment?
That experience of “seeing a true color flag” (just like “seeing a false color flag” or “seeing the font color on instructions”) is just what “experiencing qualia” is. Therefore, if you see it, you can be certain that you do “experience qualia”.
So, that’s one unnecessarily complicated method you can use if you to want to find out if you experience qualia. It might not be very useful if someone else wanted to find out if you experience qualia, but that is not required (that “someone else” can find out if he experiences qualia himself). Also, we can guess that you performed the experiment yourself and the results were just as they were supposed to be.
I
inocente
Guest
When I said true color, I meant true to the colors you would normally see (the sensations you call red, white and blue).
But you have no way of proving that your sensation of redness is the same as my sensation of redness. The sensation is completely subjective. This is the standard problem in any discussion of qualia, it is why qualia are not even wrong, and the reason why I’m on a winner.
The point of the flag exercise is that the after-image is created by compensation mechanisms within the physical perception system. Either (a) the system is itself sensing and recognizing the after image or (b) an holmunculus is doing it. But (b) just invents a little man, it doesn’t explain how the little man is doing it. It just chases its own tail in an infinite regress.
M
MPat
Guest
Sure. And the image we were looking at originally had “false colors”.
That is not being claimed. The claim “You experience qualia.” means just that you do indeed have some sort of “sensation of redness”. Nothing more. And you admit having a “sensation of redness” yourself, therefore, you do admit that you do experience qualia - or you expect “qualia” to mean something else, in which case we are “talking past each other”.But you have no way of proving that your sensation of redness is the same as my sensation of redness. The sensation is completely subjective. This is the standard problem in any discussion of qualia, it is why qualia are not even wrong, and the reason why I’m on a winner.
Maybe we should agree on what we are discussing before trying to look for an explanation…?
Anyway, after we’ll get there, there will be more questions to consider. For example, what should be included in the “system”…? And what should count as “sensing and recognizing”…? And is it something that a purely material system (of the kind we can build) can, in principle, perform, or is something else necessary…?
But all that doesn’t matter if we are going to use the term “qualia” differently.
P
polytropos
Guest
I don’t have time to respond to anything at the moment.
However: a thought experiment.
Suppose someone says to you, “I was thinking about you last night.” Obviously this statement’s truth value cannot be verified by an external observer. However, it seems incoherent to suppose that it is not a truth functional proposition, ie. every one of us could walk up to any person and say, “I was thinking about you last night,” and it would be either true or false (depending on whether we were, in fact, thinking about the person last night).
The question, then, is, what are the necessary conditions for such an utterance to be truth functional? It seems like passing the Turing test is not sufficient for the statement to be truth-functional. Further, it doesn’t seem like the presence of data corresponding to “you” (whatever that might mean in the context of a Turing-like conversation) is sufficient either, for surely (on a materialistic view, and I would say on a coherent dualist view) such data is present in our brains whether it is consciously recalled or not. If it’s present in the machine’s “brain,” then, we do not have a sufficient basis for saying that the machine was thinking about you. For the statement to be true, there seems to be a requirement of representation to the person as a subject.
This latter requirement produces a conundrum: the statement, despite epistemological issues, is clearly truth functional when uttered by a human. But when uttered by a machine that managed to pass the Turing test, it is possibly vacuous. (All that is needed is that it is possibly vacuous; this ensures that the Turing test does not deliver credible evidence of thinking, ie. even if a machine could think, the Turing test would not be a sound epistemological method for us to discover it.)
However: a thought experiment.
Suppose someone says to you, “I was thinking about you last night.” Obviously this statement’s truth value cannot be verified by an external observer. However, it seems incoherent to suppose that it is not a truth functional proposition, ie. every one of us could walk up to any person and say, “I was thinking about you last night,” and it would be either true or false (depending on whether we were, in fact, thinking about the person last night).
The question, then, is, what are the necessary conditions for such an utterance to be truth functional? It seems like passing the Turing test is not sufficient for the statement to be truth-functional. Further, it doesn’t seem like the presence of data corresponding to “you” (whatever that might mean in the context of a Turing-like conversation) is sufficient either, for surely (on a materialistic view, and I would say on a coherent dualist view) such data is present in our brains whether it is consciously recalled or not. If it’s present in the machine’s “brain,” then, we do not have a sufficient basis for saying that the machine was thinking about you. For the statement to be true, there seems to be a requirement of representation to the person as a subject.
This latter requirement produces a conundrum: the statement, despite epistemological issues, is clearly truth functional when uttered by a human. But when uttered by a machine that managed to pass the Turing test, it is possibly vacuous. (All that is needed is that it is possibly vacuous; this ensures that the Turing test does not deliver credible evidence of thinking, ie. even if a machine could think, the Turing test would not be a sound epistemological method for us to discover it.)
M
MPat
Guest
Actually, I was thinking that, perhaps, the ones who find the Turing test very persuasive disagree with us about the definition of “truth” as well…? If we accept correspondence theory of truth, the whole idea that we need a method to find out the truth for it to “count” looks suspicious - if not absurd. So absurd that it is hard to take it seriously. But maybe they subscribe to some “consensus theory of truth”, and then it might lead to something like this:
Which, of course, is absurd under correspondence theory of truth…
P
polytropos
Guest
Perhaps. That said, to deny that the statement “I was thinking about you last night” is truth functional seems to be exactly what the most extreme eliminativists do, and from what I can tell, the materialists on this forum have been at pains to distance themselves from that level of reductionism. inocente also seemed to endorse a correspondence theory of truth when he said:
But the issue seems to be that there seems to be no reason why the reports of internal states of a machine that passes the Turing test should be truth functional. Unless it thinks in the way humans do (for that’s what the statement means in the English language which it was designed to speak in), then there is no reality to which the statement corresponds (nor even a negative state of affairs); it is semantically meaningless. Its replying “I was thinking about you last night” need only fit into whatever conversation it is having at the moment. The representational threshold for such a statement to be true need not even be possible for a computer to utter it in conversation.
One might also intentionality a similar intentionality-related point. It is conceivable that a machine could pass the Turing test in one language, but its conversation could have an entirely different meaning in another language (one need only posit a community who interpret the English language in a consistent but incompossible way - which is not even metaphysically doubtful). But this implies likewise that semantic content and truth functionality are irrelevant to the “thought” of the machine as delivered by the Turing test, while they certainly are relevant to us.
I
inocente
Guest
You started this conversation by claiming you were the first person in history to devise an experiment which shows the existence of qualia. Now you’re saying you didn’t.
I think we’re all agreed that we experience subjective sensations. Calling them qualia makes the implicit assumption that they can be treated as having an independent existence. That assumption can’t be proven and some of us find it highly counter-intuitive.
Things get worse, as the assumption leads to the notion of p-zombies, people without qualia, which is a refined variation on changeling folklore (where parents believed that their baby had been replaced by a exact replica troll). It’s like saying there’s such a thing as zombie water, where the wetness has somehow been removed.
We can avoid all of that by instead assuming that sensations are integral to the system, they are part and parcel of how the (nervous) system works. If that assumption is wrong, we’ll find out after further research. The principle is not to invent castles in the sky such as qualia.
J
Jewel34
Guest
I hope you will have time sometime in the future.
This is a good example of the problem of asserting something that is not part of reality. The “past” is not part of the reality, just like the “future” is not part of it either. Is there a truth value of the proposition: “I will think about you tomorrow evening”? Right now, there is no truth value associated with it.
There are two kinds of propositions about the past. One would be talking about an event which left a physical mark on the physical reality, which can be examined now. Example could be: “Last night a meteor slammed into my car”. We can go and check the car, observe the meteor and decide if the proposition is true or not. Such events are rare, however. Is there a general epistemological method, which can find out if a proposition about the past is true or not? Not that I am aware of. There is the reliance on “testimonials”, but that is unreliable (sorry for the pun
), even if there are first hand witnesses. Otherwise we only have hearsay…Not all propositions have (or can have) a true-false value associated with them. (There are some philosophers, who do not accept these kinds of sentences as “propositions” - but that is their problem.) If there is no epistemological method pertaining to such propositions, they firmly belong to the “who, the heck cares?” category.
Of course a computer with a long term read only memory and a query method to display that memory would offer a foolproof method of deciding if the “I thought about you last night” proposition is true or false. But then again, this read only memory is a physical “trace” of what happened in the past. A human with a perfect recall could also offer such evidence, as long as we can be sure of his trustworthiness.
I suggest we return to the previous conversation, if you have time. Especially to this one. Because the fundamental problem (so far not addressed) is “at what point does the emulation become so precise that it is impossible to tell it apart from the real one?”. The sometimes repeated “but ontologically they are different, even if it is impossible to distinguish them” is irrelevant.
I
inocente
Guest
“I was thinking about you last night” could mean “I have objective evidence of thinking about you” but more usually would only mean “I remember thinking about you”, in which case how do you know the memory is real?
More importantly I’d say we can’t assume our mind architecture finds greater intrinsic meanings than any other. Memories don’t physically appear to be stored in anything like a neat manner, and thoughts light up whole areas of the brain, which means complicated things are going on under the covers.
The truth is whatever is left after we have tried our hardest to prove ourselves wrong.
If it’s conceivable for a machine to do it, why not a human?
The following talk by Hod Lipson is 55 minutes but well worthwhile as he starts with about evolutionary robotics and ends by talking about truth, reality and such stuff.
youtube.com/watch?v=Xja6sLl6dVg
P
polytropos
Guest
I am not convinced that there is enough symmetry between the past and the future for your point to hold. To set aside the question of private statements, say I predict: “The Eagles will win the SuperBowl.” You are saying that that is not truth functional because it refers to the future?This is a good example of the problem of asserting something that is not part of reality. The “past” is not part of the reality, just like the “future” is not part of it either. Is there a truth value of the proposition: “I will think about you tomorrow evening”? Right now, there is no truth value associated with it.
There are two kinds of propositions about the past. One would be talking about an event which left a physical mark on the physical reality, which can be examined now. Example could be: “Last night a meteor slammed into my car”. We can go and check the car, observe the meteor and decide if the proposition is true or not. Such events are rare, however. Is there a general epistemological method, which can find out if a proposition about the past is true or not? Not that I am aware of. There is the reliance on “testimonials”, but that is unreliable (sorry for the pun
Not all propositions have (or can have) a true-false value associated with them. (There are some philosophers, who do not accept these kinds of sentences as “propositions” - but that is their problem.) If there is no epistemological method pertaining to such propositions, they firmly belong to the “who, the heck cares?” category.
That said, the need for a physical mark seems too strong for coherence. Suppose I snapped my fingers yesterday. I also clapped my hands at a separate time. The snapping of my fingers was caught on tape. The clapping, alas, was not; you’ll have to take my word for it. Is my reporting “I snapped my fingers yesterday” truth functional, since it can be verified, while “I clapped my hands” is not? I don’t feel like that is very plausible. It seems that both are truth functional, and we just lack a method to figure them out. Their truth values are undetermined, but clearly they have one.
(Consider, further, that I just thought that the clapping was not caught on tape. Are we to believe that the statement was meaningless until someone reveals that they actually have evidence of it, and so it then becomes truth functional? It seems that only weirdness ensues on such an interpretation.)
It seems that considerations of the past and present will not allow us to brush off the statement, so we would have to hold that internal reports about thinking are meaningless, if we are to say that there is not a truth value associated with “I was thinking about you last night.” ie. we would have to take the eliminativist position.
My goal is to respond this weekend. I apologize for the lengthy delay.
P
polytropos
Guest
In any given case, we don’t need to know that the memory was real, we just need to know that sometimes when we tell someone, “I was thinking about you last night,” we were in fact thinking of them last night, or perhaps we were not (It seems like it would be truth functional even if we did think of them and then forgot about it, for “The earth orbits the sun” was truth functional even when no one knew that it was true). The issue is that the fact of its being uttered does not mean that it has any semantic content. Even if a particular machine does give it with semantic content, the Turing test couldn’t distinguish such a machine from a machine that doesn’t, ie. even if we did make a thinking machine (whatever that would mean), the Turing test is not able to separate the wheat from the chaff. (Perhaps no test could. But that doesn’t vindicate the Turing test.)
When a human is having a conversation, he is aware of the semantic content of his utterances (generally! I recently heard of an actor who did a movie in French, despite not knowing French). But when semantic content is externally defined, there is no need for the machine to evaluate it one way, or the other. If it fits into a set of linguistic rules, then it passes the Turing test. The issue is that even if the machine were aware of the semantic content (however that might happen!), the Turing test would not help us to distinguish it from a machine that doesn’t. We simply can’t draw a conclusion from the consistency of responses.
(Another consideration is that any series of symbols can have some meaning attributed to it. So you could create a machine that spouts pure gibberish and construct a linguistic system to attribute semantic content to its gibberish, even though the machine doesn’t care what the gibberish means, and no one in the universe can figure it out. You could, perhaps more professionally, create a complex set of linguistic rules, but rules that don’t correspond to any extant language. The question then is whether the machine understands the semantic content of some new language that no one understands, or whether we’ve simply created a machine that responds to certain (name removed by moderator)uts according to a complex set of rules.)
M
MPat
Guest
I guess that is not a strawman but rather a joke that is not really directly relevant to the discussion, but meant to lighten it up… Still, I’d say that a something like “And I thought you were you were the first person in history to devise an experiment which shows the existence of qualia!” would have been better joke…You started this conversation by claiming you were the first person in history to devise an experiment which shows the existence of qualia. Now you’re saying you didn’t.
What makes you think of such an assumption…? The shorter name is used simply because we do not to repeat the description again and again and again. It does not mean that qualia can exist apart from the human who experiences it (or do you mean something else by “independent existence”…?).
OK, let’s take a different word: “dream”. Would you say that the use of that word (instead of, let’s say, “things one sees while sleeping”) implies that dreams can exist independently, apart of the human who dreams…? Also, the fact that someone else dreams is also hard to check directly…
I don’t think that anyone claims that philosophical zombies actually exist… They are meant for thought experiments only.
How…? What kind of “further research” would you expect?
OK, I think that gives me an excuse to ask you a question I wanted to ask some time ago… You write you are a Baptist. In that case, do you have some answer to the atheists who would consider God or soul as similar “castles in the sky”…?
J
Jewel34
Guest
There is no real “symmetry”. Propositions about the future are undecidable, while propositions about the past are “dubious” – unless there is a physical evidence that can be independently ascertained.
It may or may not be a “true” proposition. We simply do not know and cannot know until the game will be finished.
As said before, the ontology without epistemology is a vacuous statement. Whether a proposition is true or false is of no significance if there is no way to find out which one it might be.
It may well be that you also clapped (and you say you did), for which there is no physical evidence, but that is a much lower level of reliability. Generally speaking, if the only evidence is a “testimonial” evidence then there is no good reason to accept it uncritically. This demand does not “accuse” the witness of dishonesty of course. We are all aware (or should be) that even the eye-witness testimony of scrupulously honest witnesses is unreliable. People are sometimes simply mistaken, or incorrectly interpret the events.
polytropos said:
Instead of “meaningless” I would prefer “undecidable”. It is a coherent, meaningful proposition, but it cannot be assigned a truth value. (Even the proposition of “This statement is false” is a syntactically well-formed proposition, but it has no “truth-value” associated with it.)
Again, I do not doubt that there is a “truth” value to the proposition of “I was thinking about you last night”, I only say that there is no objective epistemological method to decide if it is true or not – at least in the case of humans. In the case of a well-designed electronic being with a “read-only” log, we do have the necessary evidence.
Now, this is a very fun conversation. But I fail to see what is its significance in relation to the Turing test. Shouldn’t we leave it to a separate thread of its own?
Please take your time. There is no hurry.
P
Peter_Plato
Guest
I think there is an even larger issue with the eliminativist position, an issue that makes it untenable, not merely from a metaphysical perspective but from a human one.
When we, or at least I, use the words, “I love you,” I do not mean - as an eliminativist would have to by definition - "I love this physical body and all the neuro-chemical synapses that occur inside of it. My meaning is “I love the person that is you, including all of the mystery of ‘you,’ the human person/being that I have been honoured and blessed to catch glimpses of every now and then in the real world that I intimately share with you.”
To reduce the reality of ‘persons’ or even ‘human beings’ to some evidential test is, at base, an odious idea. It is to assume that persons are ‘vacuous’ entities or that the idea of person is vacuous to begin with. The truth is that the Turing Test shows that eliminative materialism is an inherently feeble metaphysic.
I find the thought of reducing a person to mere neuro-chemical processes (or observable behaviours) to be, even at a shallow level, defective and inherently an insult to the person I know is much more than specifications of mere physical processes or observed behaviours that can never come close to properly characterizing what a ‘person’ actually is, as a reality.
Persons are not ‘objective’ in anything like the way that things in the physical world are. To insist that persons ought to be or will simply be disregarded owing to their irreducibility, is, to not mince words, insane.
Eliminative materialism is not only lacking as a metaphysic since it can’t explain a great deal about reality, it is severely deficient as a ground for morality (which is a whole other thread), and it simply misses the mark completely concerning reality since it entails an essential denial of what we do when we relate to each other as “persons.”
It sets the standard for human relationships to ‘proof’ rather than personal intimacy and trust. That idea is ridiculous.
The Turing Test implicitly adheres to and requires an eliminative materialist view in order to be sustainable as a test for human intelligence. It fails because it makes completely irrelevant the inclusion of “persons” as conceptual possibilities.
J
Jewel34
Guest
Peter Plato:
Is this an example of a civilized, respectful, dispassionate, “charitable” discussion?
I
inocente
Guest
A group of alien siliconites land outside your house, question you for a while, and conclude that your statements have no semantic content. They pat you on the head, as they are fond of biological robots for our entertainment value, while bemoaning the fact that we don’t even have a diagnostic port.
In other words I’m not sure why your logic could not be used by machines against humans in their version of the Turing test.
The alien siliconites determine that our brain makes associations, it’s actually really good at making associations. It’s not quite as good at breaking associations which turn out to be wrong however, and this loyalty is sometimes beneficial, sometimes not. Their analysis concludes that when there are sufficient associations at a deep enough level, we say we have found meaning, but all we are doing is using the associations to make analogies which give an illusion of meaning, unlike the real meaning which they experience.
P
Peter_Plato
Guest
Why should a discussion have to be dispassionate? Are we machines engaging in it?
As for civilized, respectful and charitable, if all those depend upon being dispassionate, in your view, then we differ on the application of those, as well.
There is a difference between a point of view and the person holding it. Being respectful and charitable apply to other people, these do not apply to points of view. Why should anyone have to be charitable towards a “point of view” or respectful towards a logical argument? These are not persons.
Of course, if persons in your view are indistinguishable from neurochemical processes then I can understand why you would also confuse persons with their arguments or their points of view. But I don’t feel any such compulsion to respect arguments as if an argument is indistinguishable from the person making it. Arguments, in my view, are fair game.
I do call ideas ridiculous and odious, a certain metaphysic inherently feeble and a position untenable, but nowhere do I make any remark which is disrespectful or uncharitable towards any person. Do you not see a difference? Perhaps you have lost the capacity to differentiate between persons and their arguments, in which case that may be a symptom that reductivism has become systemic. In that case, a dose of personalist thinking might help.
By the way, you didn’t answer any of my arguments, but merely aimed squarely at MY PERSON by calling ME uncharitable, disrespectful, uncivil and passionate (although that one I take as a complement) with regard to how I carry out a discussion. In other words, it is only YOU who are committing the foul you are trying to call on me.
Will you at least do me the respect of answering my argument? You can be as disrespectful of IT as your heart desires, but, at least, attack it with sound reasoning and not substitute logic with weak ad hominems as you did in the above post.
- Status