On Philosophy

September 19, 2006

Animal Consciousness

Filed under: Mind — Peter @ 12:46 am

The question of whether animals are conscious has ramifications for our understanding of human consciousness, because to argue that animals are or are not conscious is to make claims about what is required for consciousness and how we can know that other beings are conscious. One argument against animal consciousness is, of course, the dualist position, that only people have the necessary immaterial mind. But how then do we know that other people are conscious? Perhaps there is some “supernatural” connection that allows us to conclude that the man at the next table in a restaurant is conscious. But, since we would also think that a perfectly constructed holographic recording of that man was conscious, I find this unlikely. Perhaps then we simply conclude that consciousness is somehow responsible for what we are seeing. If that is true we should look for what is responsible for people’s behavior, which turns out to be their brains, and conclude that it is conscious. We therefore would conclude that human brains are somehow responsible for consciousness. However, animal brains are very similar to human brains, so why not conclude that they are conscious too? The dualist can have no satisfactory answer to this, he simply knows. And thus many who want deeper explanations seek materialist explanations of consciousness.

One feature that is thought to be essential to consciousness in general is self-consciousness. Self-consciousness is simply our conscious awareness of information about our current state, as well as the intuitive knowledge that our experiences and feelings are our own. Given this we can ask: can animals possess self-consciousness? If they don’t it might be a good reason to believe they aren’t conscious at all. There is, however, good reason to believe that animals are self-conscious, since they behave as if they were. (In theory this behavior could be produced by unconscious systems. However, the behavior does indicate that if they are conscious they are self-conscious, and thus we can’t argue that animals aren’t conscious because they aren’t self-conscious.) One example of such self-conscious behavior is the ability of animals to feed themselves when hungry, allowing us to conclude that they “know” the sensation of hunger they feel to be theirs. This is stronger evidence for self-consciousness than it may seem, since many of the mammals possess a “sympathy” system. This system produces in the animal sensations similar to those that it guesses the other animals around it are feeling. For example, when a chimpanzee sees a researcher eating some ice cream the chimpanzee’s brain displays activity that is very similar to the activity that occurs when the chimpanzee is eating a peanut. This strengthens the evidence for self-consciousness in animals because the animals do not act on these “sympathetic” sensations as if they were their own, thus making more likely that the animal experiences normal sensations in a distinctively self-conscious way.

Some would grant that animals display evidence of being self-conscious and still deny that they could be conscious. They argue that not only is the simple self-consciousness discussed above necessary for consciousness but also a more sophisticated “conceptual” self-consciousness. Conceptual self-consciousness is being conscious of one’s self not in just a direct fashion, for example experiencing some experiences as one’s own, but also as an individual with certain properties. Specifically, conceptual self-consciousness requires the ability to think about ourselves as we appear to other people, and to be able to use such self-indexicals such as “I” and “me”. I grant that conceptual self-consciousness requires a level of intelligence that is not found in most animals (although experiments with chimpanzees and mirrors have shown that they have some form of it). What I won’t grant is that conceptual self-consciousness is necessary for consciousness in general. There are two reasons to believe that consciousness doesn’t require conceptual self-consciousness. One is that conceptual self-consciousness is not found in infants and sufficiently young children, but some people have memories of this time, and remember themselves being conscious. Another reason is that while the basic self-consciousness is present whenever consciousness is, in our experience, we are often conscious without using, or being aware of, any conceptual self-consciousness. Conceptual self-consciousness seems to be employed only in planning interactions with other individuals, and perhaps when reflecting philosophically, and so if it was really essential for conscious experience we would expect most of our experience to be unconscious, which it is not. (And it doesn’t make sense to say that it is simply the capacity for conceptual self-consciousness that makes us conscious, since for all we know animals may have that capacity and be simply not using it; thus it can’t distinguish us from animals unless we are going to simply assume that animals don’t have such a never-used capacity without evidence.)

Let me now turn to an argument for animal consciousness. A quick rule of thumb for the existence of consciousness, developed by Nagel, is that when there is consciousness there is “something it is to be like”. It seems obvious to most that there is something it is to be like at least some animals, and thus we conclude that they are probably conscious. Now it is possible that we are simply anthropomorphizing other mammals, but I think this is an unlikely explanation, since most of the behavior that convinces us other people are conscious is present in animals. The only thing people have, that animals don’t, which makes it seem indisputable that there is “something it is to be like” them, is language, and it seems unlikely that language is somehow essential for consciousness. (Again, consider infants, people who never learned to speak, the fact that people learn different languages.) If we can be convinced that someone speaking a language we don’t understand is conscious based on their behavior then I think we can conclude that some animals are conscious as well. Another objection to arguing that animals are conscious based on this principle is that we can imagine that there “is something it is to be like” various non-conscious things. For example some people can imagine what it is like to be a computer or a mountain. This is a valid objection; to overcome it we must modify our rule of thumb about consciousness. We should say instead that we should think that a system is conscious only when it is conceivable that there is something it is to be like that system and the existence of this consciousness is indicated by behavior. So even if we can imagine that there might be something it is to be like a mountain the mountain doesn’t behave in ways that are indicative of consciousness (and neither do the current generation of computers). Animals do, however, and thus I conclude that it is likely that they are conscious (although I may be proven wrong if we discover the biological basis of consciousness and find it to be missing in animals).

So if we have about as much reason to believe that animals are conscious as we do to believe that other people are conscious why do so many people object to the idea? I think it is because many people assume that we have ethical obligations to people because they are conscious. If you believe this, and believe that animals are conscious, then you would be forced to conclude that we have ethical obligations to animals as well. And this would mean that many people have been acting unethically, for example by eating animals, and people hate being told that they are doing the wrong thing and need to change. Fortunately for meat eaters, our ethical obligations to other people don’t necessarily arise because they are conscious. Previously I proposed that we have ethical obligations to other people because we are both part of the same community or could be part of the same community. We could then deny that animals are intelligent enough to be part of these communities, and thus that we don’t have ethical obligations towards them. However, our ethical obligations to animals, if any, are outside the scope of today’s post, so I will leave detailing them further for a future date.

Advertisements

8 Comments

  1. I would say we can distinguish between animals and people because, even well trained gorillas fail the Turing Test. (Seriously, that Koko has what would be a terrible case of ADD for a human. Yes, she’s expressing herself, but herself is about a deep as a rain puddle on a sunny day.)

    That said, I’m not especially convinced that animal consciousness is qualitatively different, but I do think it is quite clearly quantitatively different.

    Meh.

    Comment by Carl — September 19, 2006 @ 2:20 am

  2. I thought that since we had already established consciousness and intelligence are independant that the turing test is no good.

    Comment by Peter — September 19, 2006 @ 3:23 am

  3. While I would hazard to agree that consciousness and intelligence are at least somewhat independent, I never agreed with your argument that Turing tests are no good. You seemed to be saying that a Turing Test could be bluffed by a non-conscious machine with enough pre-programmed answers, but I say, our ordinary means of telling the human from the not-human are fooled in this way under ordinary circumstances as well. To wit, some of my friends have set up joke answering machine messages, where they say, “Hi! … Oh. … Uh-huh. … Yeah,” just to fake you into believing you’re talking to a person. Of course, it’s easy to fake me out for a little while, but as time goes on, the chances that I’m being faked out by pre-programmed responses decreases exponentially, since the conversation has too many directions which it could potentially take for my friend to put the “yeahs” in the right place. Similarly, you could make a computer that could pass a Turing test for a short while with just some “yeah”s and “uh-huh”s, but to make one that passes a prolonged (multi-day, possibly multi-year) test would require the functional equivalent of a real human-like consciousness that, for my purposes, we are justified in calling a “real human-like consciousness.”

    In any event, while I think that while passing a Turing test is only a probabilistic indicator of the existence of an intelligent consciousness like my own, for a motivated participant to fail a Turing test the way Koko did is a definite indicator that either a) Koko is intentionally deceiving us into thinking she’s not a conscious intelligence in the way that we are or more likely, b) she really isn’t a conscious intelligence in the way that we are.

    Comment by Carl — September 19, 2006 @ 3:47 am

  4. I’ll agree that it proved that Koko, who knew only a few words, isn’t as intelligent as us, but I didn’t need a test to tell me that. And until you can refute the arguement I consider it to stand.

    Comment by Peter — September 19, 2006 @ 4:06 am

  5. What am I supposed to refute? The argument that the Turing test doesn’t work? Could you lay out the premises for that better than how I summed it up in comment 3? If you just think it doesn’t work because we can be “fooled” then I think, of course. We are easily temporarily fooled into thinking that there’s a real human on the other side of the phone network and whatnot, but fooling us for the long term is impossible, since there are too many possible for a conversation to take to pre-program them all.

    Comment by Carl — September 19, 2006 @ 4:31 am

  6. Argument’s conclusion: the Turing test is for intelligence not consciousness. Nothing you have said refutes that conclusion. And if there can be consciousness without intelligence then failing the Turing test in no way shows that the animal is not conscious. That animals can be conscious is the point of this post, not that they can be intelligent as we are. And intelligence is not a good reason to consider ones self to be special or better than animals, unless you are perpared and willing to accept that smarter people should consider themselves better than dumber people (or that dumber people are somehow closer to animals).

    Comment by Peter — September 19, 2006 @ 4:37 am

  7. Ah. Yeah, I guess I can agree with it, if you put it like that. However, if you decouple “consciousness” from “intelligent self-consciousness” (which I think may be fair to do), you run the risk that possessing “consciousness” is no longer a very interesting fact about humans after all, since it’s a property that even non-moral beings can potentially possess.

    Comment by Carl — September 19, 2006 @ 5:03 am

  8. Oh consciousness is interesting, it’s just not uniquely ours. And some people, sociopaths for example, are biologically immoral, but no doubt conscious, so consciousness is already de-coupled from morality in at least one direction.

    Comment by Peter — September 19, 2006 @ 5:08 am


RSS feed for comments on this post.

Blog at WordPress.com.

%d bloggers like this: