IUBio Biosequences .. Software .. Molbio soft .. Network News .. FTP

Nonhuman empathy

Michael Edelman mje at mich.com
Fri Feb 12 15:27:59 EST 1999

Matt Jones wrote:

> In article <36C4373F.A466D92D at mich.com> Michael Edelman, mje at mich.com
> writes:
> >The difference between the human and the animal case is that in the human
> >case, you may have been wrong in asssigning specific states to the actors,
> >but you can safely assume that being human they share with you certain
> >states. Humans can be sad, happy, etc. and will have similar external
> >correlates to these states.
> You say I can "safely assume that being human they share...certain
> states", and I agree. But I would go further and say that being animals
> with a lot of shared biochemistry and evolutionary history, we all share
> certain states.

How far down the tree can you assume that? And on what do you base your opinion?
Surely we have shared biochemistry and evolutionary history with fish, but do fish
feel jealosy?

> >The fallacy committed by the turtle owner was not just in assuming the
> >existence of certain states to his turtle, but more importantly, assuming
> >that the turtle was experiencing a certain state based on a behavior that
> >would correlate with that state in humans.
> This is not a fallacy, any more than assuming that a human is in a
> certain state based on external correlates. We don't have ANY clue as to
> another human's internal state other than what we see externally.

But- we assume that being human, they have certain states in common with us, and
by observation and more importantly, communication, we can not only verify the
existence of these states, we can also assign common labels to common states. So
if you say you're sad, assuming you're not using some non-standard label, I know
what you're feeling. If I see you and you look sad, I can verify that through

But if your turtle "looks" sad, you *may* be (1) misinterpreting the signs or (2)
assigning a non-existant state. If you claim that turtles can be sad, you need a
better argument than to say they share a common biochemical and evolutionary

> Different humans have different external manifestations of internal
> states that they would verbally identify with the same word. For example,
> when "angry", some people jump up and down and scream while others merely
> wrinkle their brow, while others display little if any external cues.

Doesn't matter for the reasons outlined above- the common language. But that
reinforces my point about non-human species- what is the justification in
assigning a specific internal state to some behavior?

> These external displays can be modulated intentionally (i.e., poker
> face). So it is in fact a fallacy to lump all these cues together as if
> there were a simple external "anger" display in humans.

That is not an issue up for debate.

> There isn't. Once
> we admit that we need to have some leeway in correlating human internal
> states with external ones, then we must admit to a similar leeway when
> translating across species.

Except for the common language and common experience. That common experience is
what makes language translation possible.

> ...
> >I don't think anyone would deny that cats and dogs and turtles feel pain-
> >they certainly show aversions to stimili we find painful, and we can probably
> >be safe in calling that pain.
> >
> >But there are organisms with extrememly simple CNS- a few hundred neurons-
> >that also show aversion to certain stimuli, and yet we'd be hard pressed to
> >say they have the phenomenological experience of pain.
> Yes, there is a spectrum of nervous system sophistication. A spectrum,
> not discrete categories. This is a consequence of us having evolved from
> simpler ancestors.

So where do you draw the line, and more importantly, how do you make that

> >As for something like jealosy- now you're in an area that's much harder to
> >support. Do cats get jealous? I'm not sure how you could divine that.
> Do people get jealous? I'm not sure how you could divine that. Oh, I
> know. You look at their behavior and infer an internal state.

No- you ask them. That's the difference.

> When my cat
> suddenly starts pissing on the furniture the day after I bring home my
> newborn daughter, I attribute it to jealousy.  I could be wrong, but I
> can also be wrong in making such attributions with other humans. So
> there's nothing inherently fallacious about TRYING to make such an
> attribution.

In this sense there is: You're using a behavioristic definition of jealosy, i.e.,
a set of specific behaviors that correlates to a set of inputs- bringing home the
new daughter. But here you're assigning a state bbased on human behavior.

Cats are not social animals in general. They are solitary hunters. There are
always exceptions, of course; perhaps you have a jealous cat. But it is as likely
that the cat was reacting not to something that would take attention from it, but
to a new threat to its territory. I'm not saying that this is what happened, but
rather pointing out the pitfalls of assigning interpretations of internal states
based on human states that would lead to that behavior. And *that* is what the
anthropomorphic fallacy is.

> >> Are we somehow chemically
> >> priviledged that we have the machinery for those emotions, but cats don't
> >> (the answer is no).
> >
> >You must differentiate between the machinery and the emotional state- as with
> >the simple organisms mentioned earlier. Ajellyfish shows aversion to stimuli
> >associated with harm to the organism, but it's hard to imagine a jellyfish
> >having the phenomenological experience of pain in the same sense of a human.
> Well, yeah, I don't think that jellyfish have the "phenomenological
> experience " (whatever that means) of pain in the same sense as a human.
> But I also don't think that all humans share the same experience of pain.
> People are said to have high or low pain thresholds, they respond
> differently to similar stimuli. I'd say it's hard to imagine defining any
> single "phenomenological experience" that applies to humans equally.

You're confusing intensity of sensation with the actual experience of pain. Say we
both go to the dentist, and he hits a nerve root. I scream and hide in the corner,
whimpering. You say to the dentist "uh, that was rather painful". Different
intensity, but we both experienced pain. Recoiling from stimulus is not the same
as experiencing pain.

> >> Or is our brain organized in such a way that we can
> >> somehow generate emotions through a feat of supreme mental computation,
> >> whereas creatures with less cognitive and logical horsepower can't
> >> (answer: no).
> >
> >I think you need to provide a better argument than  flat out assertion that
> >the capacity of an organism to experience complex emotional states is
> >independant of the complexity of the brain.  Does a sea snail feel jealosy?
> >Does an earthworm experience existential angst?
> First, let me reiterate that I'm not saying animals and humans experience
> EXACTLY the same mental states (that would be a stupid thing to say). I
> also don't think that any two humans experience EXACTLY the same internal
> states (equally stupid, in my opinion).

I think you're arguing something not at issue. I don't argue for two humans having
the identical physical state, but rather similar states that correlate with
similar *emotional states*, and with what might be termed experiential states. You
know what I mean by "pain", and it goes beyond behavioral correlates. You do not
conceptialize pain internally  as something that makes you withdraw from a
stimulus; it goes beyond that, right? There is what I've been calling a
phenomenological experience of pain. When the dentist hits that nerve root in your
tooth, that's what you experience.

> So, yes, of course there are
> differences between us and them. But at the bottom of it all is plenty of
> shared genetics, biochemistry, physiology and behavior. We can study
> "aversion responses" in snails and in neuroscientists alike. There is a
> core of shared experiences (at least at a very low level) because we are
> all products of the same basic environment (i.e., carbon life forms,
> convert glucose to energy, undergo sexual reproduction, the really basic
> stuff is the same).

But that still does not establish the existence of the same experiences of pain,
lonliness, whatever. You're positing it and saying we're all alive, we're all
animals, ergo we must all share these states. At the same time you do state that
not all living things share these states, so there must be some other factor that
determines why a dog can be angry, and a squid cannot. What then are these
factors, and how do you determine that?

> {...}
> No, exactly the opposite. I'm a lumper, not a splitter. I'm arguing that
> you are defining emotion too narrowly and in too anthropocentric (pardon
> my gender bias) terms. I didn't say that complexity is not an issue, but
> I do not think it is a hard and fast requirement either (there probably
> is some minimum level of complexity required for emotional states, but
> that level is far far below the complexity of the human or turtle nervous
> system). It seems to me that you are pushing a dualist viewpoint if you
> wish to separate human behavior from that of the rest of the animal
> kingdom. Do you think that only human beings have consciousness? I don't.

No, and that was never as issue. I assume many animals have conciousness, and I
assume many can feel pain, anger and so forth. But- I am very careful about how
and when I assign those emotions to organisms I cannot communicate with.

I would accept that chimps can feel jealosy, dogs can feel anger and cats can feel
pain. I doubt that any reptile, turtles include, can experience much of what we we
think of as conciousness. Too little cerebral cortex.

Complexity of behavior is no indicator of higher levels of conciousness. Sharks
have an amazingly efficent and complex array of behaviors, yet they're all
cerebellum. Are sharks self-aware? I have my doubts. >

Michael Edelman     http://www.mich.com/~mje

More information about the Neur-sci mailing list

Send comments to us at biosci-help [At] net.bio.net