6 ms·

I believe it's fundamentally impossible to test for consciousness. You can't know if it feels like something to be any other thing but yourself. The domain of s

I believe it's fundamentally impossible to test for consciousness. You can't know if it feels like something to be any other thing but yourself. The domain of scientific knowledge is completely cut off from subjective experience. Even if the octopus could speak and told us in English "I feel therefore I am" we could not know for certain that it produced those words having a conscious understanding of what something feels like. Feeling is fundamentally in a different class of thing from those things which the scientific world can explore. And that to me, is absolutely bizarre. What does that mean? How can that possibly be true? Is it definitely true? What on earth is "feeling"?! What is qualia. It's the deepest question and I think probably holds the key to really understanding why the univers exists.

the8472 7y ago
I would argue it's moot to discuss consciousness unless someone prefaces the discussion with a specific definition, preferably a scientific one and not some woo. Philosophical consciousness suffers from too many issues for day to day discussion, some previous arguments I made in this direction before: https://news.ycombinator.com/item?id=20501688 https://news.ycombinator.com/item?id=20501688
- plutonorm 7y ago
  Yup I agree, the hard problem of consciousness should be deliberately excluded from discussions on various topics, particularly AI. It's not useful, its fascinating, but it's irrelevant and distracting for most purposes.
blotter_paper 7y ago
It's not even clear to me that consciousness is "real" at the level of base reality, but I do have a proposed test: link yourself up to an octopus using the same BMI setups that are used to allow two rats to share experiences and memories. If you start experiencing/remembering the life of an octopus, the brain on the other side of those wires is almost certainly conscious. The inverse is not true, of course; we could simply fail to grasp the differences between the brains of cephalopods and mammals, resulting in a bad connection to a perfectly fine brain. (I'm more than 99% sure that octopuses experience the world, but I think this test can be generalised to simpler life forms in principle.)
- MereInterest 7y ago
  I don't think that would be a good test, either. All that would prove is that there exists a storage device, and a method by which to read it. It doesn't imply that the recording device is conscious.
  - blotter_paper 7y ago
    Disagree, it would additionally show that the storage device was in a natively consciousness-readable format, and that's just for memory sharing. Experience sharing, if that's what the "light in one cage, button in another, reward for both" study really shows (as opposed to, say, short-term memory sharing that appears like experience sharing to an outside observer) would seem to directly suggest the circuitry of consciousness. Of course it can't be completely proven, but neither can anything else.
nessus42 7y ago
> I believe it's fundamentally impossible to test for consciousness. You are talking about "phenomenal consciousness", which is sometimes also referred to as "the hard problem of consciousness". (Though not all philosophers would equate these two terms.) There are quite a few other useful definitions of consciousness, and consequently, all of the different notions of consciousness should not all be lumped together. I actually have a Philosophy degree from MIT, and I spent almost all my philosophical time worrying about and studying this issue. It still keeps me up at night sometimes to this day. I agree with you that the hard problem of consciousness is not amenable to scientific study, and consequently, I don't think that we can ever understand it, unfortunately. I also agree that it is the deepest and most important question about the nature of reality that we could ever have, and it is quite frustrating that we will never know the answer. Back to other notions of consciousness, though: Many of them are interesting, and I think amenable to scientific inquiry. So we shouldn't let the hard problem of consciousness deter us from studying easier problems of consciousness.
- the8472 7y ago
  > I agree with you that the hard problem of consciousness is not amenable to scientific study What if we assume a brain simulator with good debugging tools?
  - NickM 7y ago
    A perfect brain simulator could answer many questions, but it could not answer the hard problem of consciousness. Science is all about objectivity, but there's an element of consciousness that is inherently subjective, and that part - the part the hard problem deals with - is not objectively measurable.
    - nessus42 7y ago
      > Science is all about objectivity, but there's an element of consciousness that is inherently subjective Yes, that is exactly right. Science is all about that which can be studied objectively and the "hard problem of consciousness" is specifically about the component of consciousness that is purely subjective. How can you study something that is purely subjective using objective methods? A. You can't! Here's another question that science can't answer: The Many Worlds interpretation of quantum mechanics and the Bohm interpretation of quantum mechanics posit vastly different things about the nature of reality. Unfortunately, they are experimentally indistinguishable from each other. So, if somehow we could narrow QM down to either MW or Bohm, at this point we are stuck. We can never know which interpretation is the right one, even though the answer to this question is up there in importance with the hard problem of consciousness!
carapace 7y ago
> I believe it's fundamentally impossible to test for consciousness. You can't know if it feels like something to be any other thing but yourself. I used to believe that too, but it turns out that consciousness can interact with consciousness. E.g. find someone you love and trust, sit facing each other, hold hands, stare into each other's eyes, and breath in synchrony. For some cases nothing really happens in other cases... shared subjective experience. > The domain of scientific knowledge is completely cut off from subjective experience. Shared subjective experience is still subjective.
rland 7y ago
I think you're right, but taking this approach leads to some strange conclusions. For example, that there is no way for you to know that everyone you have ever known is conscious.

the8472 7y ago

I would argue it's moot to discuss consciousness unless someone prefaces the discussion with a specific definition, preferably a scientific one and not some woo. Philosophical consciousness suffers from too many issues for day to day discussion, some previous arguments I made in this direction before: https://news.ycombinator.com/item?id=20501688 https://news.ycombinator.com/item?id=20501688

plutonorm 7y ago
Yup I agree, the hard problem of consciousness should be deliberately excluded from discussions on various topics, particularly AI. It's not useful, its fascinating, but it's irrelevant and distracting for most purposes.

blotter_paper 7y ago

It's not even clear to me that consciousness is "real" at the level of base reality, but I do have a proposed test: link yourself up to an octopus using the same BMI setups that are used to allow two rats to share experiences and memories. If you start experiencing/remembering the life of an octopus, the brain on the other side of those wires is almost certainly conscious. The inverse is not true, of course; we could simply fail to grasp the differences between the brains of cephalopods and mammals, resulting in a bad connection to a perfectly fine brain. (I'm more than 99% sure that octopuses experience the world, but I think this test can be generalised to simpler life forms in principle.)

MereInterest 7y ago
I don't think that would be a good test, either. All that would prove is that there exists a storage device, and a method by which to read it. It doesn't imply that the recording device is conscious.
- blotter_paper 7y ago
  Disagree, it would additionally show that the storage device was in a natively consciousness-readable format, and that's just for memory sharing. Experience sharing, if that's what the "light in one cage, button in another, reward for both" study really shows (as opposed to, say, short-term memory sharing that appears like experience sharing to an outside observer) would seem to directly suggest the circuitry of consciousness. Of course it can't be completely proven, but neither can anything else.

nessus42 7y ago

> I believe it's fundamentally impossible to test for consciousness. You are talking about "phenomenal consciousness", which is sometimes also referred to as "the hard problem of consciousness". (Though not all philosophers would equate these two terms.) There are quite a few other useful definitions of consciousness, and consequently, all of the different notions of consciousness should not all be lumped together. I actually have a Philosophy degree from MIT, and I spent almost all my philosophical time worrying about and studying this issue. It still keeps me up at night sometimes to this day. I agree with you that the hard problem of consciousness is not amenable to scientific study, and consequently, I don't think that we can ever understand it, unfortunately. I also agree that it is the deepest and most important question about the nature of reality that we could ever have, and it is quite frustrating that we will never know the answer. Back to other notions of consciousness, though: Many of them are interesting, and I think amenable to scientific inquiry. So we shouldn't let the hard problem of consciousness deter us from studying easier problems of consciousness.

the8472 7y ago
> I agree with you that the hard problem of consciousness is not amenable to scientific study What if we assume a brain simulator with good debugging tools?
- NickM 7y ago
  A perfect brain simulator could answer many questions, but it could not answer the hard problem of consciousness. Science is all about objectivity, but there's an element of consciousness that is inherently subjective, and that part - the part the hard problem deals with - is not objectively measurable.
  - nessus42 7y ago
    > Science is all about objectivity, but there's an element of consciousness that is inherently subjective Yes, that is exactly right. Science is all about that which can be studied objectively and the "hard problem of consciousness" is specifically about the component of consciousness that is purely subjective. How can you study something that is purely subjective using objective methods? A. You can't! Here's another question that science can't answer: The Many Worlds interpretation of quantum mechanics and the Bohm interpretation of quantum mechanics posit vastly different things about the nature of reality. Unfortunately, they are experimentally indistinguishable from each other. So, if somehow we could narrow QM down to either MW or Bohm, at this point we are stuck. We can never know which interpretation is the right one, even though the answer to this question is up there in importance with the hard problem of consciousness!

carapace 7y ago

> I believe it's fundamentally impossible to test for consciousness. You can't know if it feels like something to be any other thing but yourself. I used to believe that too, but it turns out that consciousness can interact with consciousness. E.g. find someone you love and trust, sit facing each other, hold hands, stare into each other's eyes, and breath in synchrony. For some cases nothing really happens in other cases... shared subjective experience. > The domain of scientific knowledge is completely cut off from subjective experience. Shared subjective experience is still subjective.

rland 7y ago

I think you're right, but taking this approach leads to some strange conclusions. For example, that there is no way for you to know that everyone you have ever known is conscious.