At the level of personal experience, there is nothing that seems easier to understand than pain. When I jam my finger in a doorway, I have no difficulty at all recognizing the sensation that results. But this superficial simplicity covers up a world of complexity at the level of brain mechanisms, and the complexities are even greater when we try to identify pain in other people or other species of animals. Some of the complexities are purely scientific, but others are caused by moral or philosophical issues getting mixed up with scientific issues.

My provocation for writing this post was a blog post called Do Octopuses Feel Pain?, by Katherine Harmon, who writes the blog Octopus Chronicles, It's basically a nice article—there's nothing objectionable about it—but it pressed one of my buttons. She made a number of important points, and altogether what she wrote is well worth reading, but nevertheless the result left me with a feeling of dissatisfaction, as do most scientific discussions about pain in animals. I'd like to try to explain where that discomfort comes from.

In her blog post, Harmon listed three elements that are involved in feeling pain: (1) nociception, that is, having mechanisms in the body that are capable of detecting damage and transmuting it into neural signals; (2) the experience of pain; (3) the ability to communicate pain information from sensation to perception. I'm not sure I understand the third aspect, but I take it to mean the ability to transform nociception into experience.

In any case, the essence of pain as most people understand it is aspect 2. Most people think of pain as a particular type of experience—as something that happens inside our minds and can only be observed by ourselves.

But as philosophers are well aware, there's a big problem with that approach. The problem with thinking of pain as a private experience is that it leaves us helpless to identify pain in other people, much less in other species of animals. If it is unobservable from the outside, how could we possibly know whether it exists? This is what philosophers call the Problem of Other Minds.

The standard solution, implicit in most scientific discussions of pain, is that the more closely an animal resembles us, the safer we are in attributing experiences like ours to it. Other people have bodies and brains very similar to ours—the argument goes—so we can safely assume that they have pains like ours, especially when they tell us about them. Some types of animals, particularly mammals, have brain structures and nociceptive systems so similar to ours that it seems only reasonable for them to have pain experiences like ours. And so on.

But here's the thing. That solution just doesn't work. It is wrong even for mammals and other people, and if you try to apply it to octopuses or lobsters, it doesn't produce anything except bafflement.

To help you see why the similarity-to-us approach is off the mark, I'd like to present you with four simple thought experiments. I claim that the pain-as-experience and similarity-to-us theory gives answers to them that most people will feel are wrong. If we don't have clear answers to these sorts of scenarios, we will not be in a good position to make use of scientific data about octopi or other animals—maybe not even data about ourselves. I have my own answers to them, which I'll explain in a followup post.

Here are the four thought experiments:

1. Suppose we are confronted with aliens from another planet, who are broadly humanoid in appearance, with two arms, two legs, an erect bipedal stance, and a head with two eyes and a mouth. They speak a language, but it does not translate easily into English, and it is unclear whether they have any word that corresponds to "pain". Suppose also that their internal organs are completely different from ours, and in particular their nervous systems bear no resemblance to ours on any level. Imagine that we poke one of these aliens in the arm with a needle, and see that the alien jerks back its arm, scrunches up its face, and lets out a screech. Should we conclude that the alien is experiencing pain?

2. Same story, but suppose now that we are dealing with a humanoid-appearing robot rather than an alien. Should we conclude that the robot is experiencing pain? If you give a different answer for the robot than for the alien, can you justify it?

3. Suppose we are dealing with an ordinary human, a man named Wally Wallace. He claims to be in constant excruciating pain, but we cannot see any evidence of it in his behavior. He seems completely relaxed; he laughs and jokes; he moves without any appearance of constraint; he seems happy. Nevertheless he says that he is in agony. Of course he might be lying, but is there any possibility that he is telling the truth? What would it take to convince you?

4. The reverse scenario. Suppose Sally Sanders has a broken leg. She claims that she doesn't feel any pain at all, but she is pale and moves stiffly; she gasps and winces whenever anything touches the injured area; she seems to move in a way that keeps the injured area from touching things. Nevertheless she says that she is pain-free. Of course she might be lying, either to herself or to us, but is there any possibility that she is telling the truth?

If you find this challenging, I hope you won't be too distressed about it. The fact is that pain is a complicated thing. In humans it involves a mixture of sensory perception, motivation, and evaluation—a combination of brain activity, "experience" (whatever that is), and behavior. Applying such an intricate concept to a species that we don't understand very well is never going to be easy.

Image: Anneli Salo