When I was very young, the top of my feet started itching, so I started scratching. The itching continued for weeks and months, with me constantly scratching. My poor mother, seeing my red and scratched feet, implored me to stop. But the itching was relentless and I was maybe five or six, so I kept scratching. Until my feet were a bloody torn up mess. Eventually I was diagnosed with a milk allergy.
The thing is, I never remembered feeling pain when I scratched the itch, only relief, even when my fingers were coming away bloody. To be sure, there was pain, but not when I was scratching, only in the periods when the itching subsided. And of course, there was plenty of pain as I healed once the itching was gone.
I recalled this episode while reading Jennifer Corns’ The Complex Reality of Pain. I became interested in Corns’ book after Richard Brown interviewed her on his Consciousness Live podcast. Corns is a scientific eliminativist toward pain, that is, she thinks pain exists, but that while it remains a productive concept in our everyday “folk” language, it isn’t a productive concept in scientific and medical investigations. This makes her position less stark than Daniel Dennett’s, who is a full eliminativist when it comes to pain, in that he doesn’t even see it as productive to retain it as a concept in everyday language.
Now, your first reaction may be to wonder whether these people have ever had a headache, tooth ache, stubbed a toe, broken a bone, etc. But there is a logic to their position.
Consider our folk version of pain. When we suffer some damage to the body, we feel it as a brute and unanalyzable experience. Under this understanding, nociceptors in the body detect damage, or potential damage, and send their signal to the brain, where it is felt as pain. The End.
Except, it turns out pain is poorly correlated with actual tissue damage, or even threat of damage. There is my personal story above, where I didn’t feel something as pain that under different circumstances clearly would have been painful. (My experience is somewhat explainable under gate-control theory.)
There are also plenty of cases where people feel pain in the absence of any damage. I have a cousin, let’s call him “Jake”, who had serious back problems, leaving him in terrible pain. Jake was prescribed opioids. Eventually he had back surgery to fix the issue. But the pain continued, with Jake having to take ever higher dosages of the opioids, until he overdosed and nearly died. He was then pulled off the opioids, and suffered agonizing pain. But as the withdrawal symptoms eased, so did the pain. Jake’s pain, after the actual body issues had been corrected, was caused by the addiction.
But aside from such addictions, post-healing pain is not unusual.
Corns points out that even viewing pain purely from a phenomenological perspective, it isn’t a simple thing. The pain of a stomach ache is very different from the pain of stepping on a nail. Pains can be throbbing, shooting, stabbing, pinching, cramping, burning, or stinging, among many other disparate sensations.
And the science is that there is no single mechanism that accounts for pain, no single pathway that can be traced from the sensory receptors all the way through the brain. There are in fact innumerable mechanisms. Corns points out that pain is actually an extremely idiosyncratic process, with each individual instance having a unique combination of mechanisms. These mechanisms eventually converge on our categorizing it as pain, causing us to communicate about it. But prior to that convergence, there is no single unifying mechanism that can be objectively pointed to.
Apparently in recognition of this situation, the International Association for the Study of Pain (IASP) defines “pain” as:
An unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage.
In the recently updated version on their site, they expand this definition with key notes, some of which include:
Pain is always a personal experience that is influenced to varying degrees by biological, psychological, and social factors.
Pain and nociception are different phenomena. Pain cannot be inferred solely from activity in sensory neurons.
Through their life experiences, individuals learn the concept of pain.
A person’s report of an experience as pain should be respected.
In other words, pain only seems to exist subjectively.
But Corns is not even satisfied with this definition. She attacks the notion that pain is an evaluated negative threat affect, an emotional feeling related to an interoceptive (internal body) sensation. She points out that the affect and the sensation can be disassociated, with people insisting that they are in pain, but not bothered by it.
So, pain doesn’t exist? While the steps that bring us to this conclusion seem to make sense, it feels like something must have been missed.
My take is that while Corns is right to reject Dennett’s full throated eliminativism, her position of scientific eliminativism has its own issues. If the term “pain” remains useful in everyday language, we should examine why it remains useful, and strive to account for that usefulness scientifically.
Now, it seems plausible that pain doesn’t exist at certain levels of organization, similar to how chairs and tables don’t exist in particle physics. Pain may be something we talk about at the psychological level that maps to only disparate mechanisms at the neurobiological level. But again, I think we want to see those mappings accounted for. Dismissing pain from science entirely feels like overkill.
I also think Corns is hasty in rejecting the affect / sensation pairing. It is true that we’re not talking about just one affect here, but more of a category of disparate threat prediction affects that we associate with our pain concept.
And yes, the affect and sensation can be dissociated, but the dissociation seems to involve lobotomies, cingulotomies, various pathologies, or use of drugs such as opioids. In other words, the dissociation happens in brains that are damaged, or whose workings are being interfered with. When a healthy brain feels pain, there seems to be both an interoceptive sensation and an affect about that sensation. If one or the other is missing, it doesn’t seem like we consider that to be pain.
In cases of dissociation, the person may feel a sensation they remember previously being painful, but without the negative affect, and so say they’re in pain but not bothered by it. But if they had never had a negative affect associated with that sensation, it’s doubtful they ever would have categorized it as pain.
That doesn’t mean the interoceptive sensation necessarily originates from sensory neurons in a body location. Jake’s opioid addiction resulted in him feeling an interoceptive sensation about his back that, after he was healed, no longer originated from his back.
Where then did it come from? Warning: this is my own speculation. It’s grounded in the neuroscience I’ve read, but I haven’t seen it explicitly within the context of pain research, and there may be complications I’m ignorant of.
First, we have to acknowledge that a pain affect is an extremely complex and idiosyncratic psychological state, with many causal factors including memories, perceptions, levels of fatigue, and mood, in addition to the interoceptive sensation itself. That means that once “learned”, that affect could be triggered by those other causal factors, not just the sensation coming in from the peripheral nervous system.
But wouldn’t that just leave us with the affect itself without the sensation? Here I think it’s important to remember that most of the connections between regions in the cortex are reciprocal. So if region A fires a pattern that excites region B, region B probably has connections back to A, enabling it to recurrently excite A. This allows activity in the two regions to reinforce and bind with each other, which probably happens when a sensation leads to a negative affect. But it also means that a negative affect caused by non-sensory factors, can effectively back-propagate to the sensory cortices, inducing them to fire as though they had received a signal from the body region.
In other words, a psychological state can lead to real pain in the absence of actual body damage, or it can lead to signals from actual body damage being inhibited. This would explain placebo effects, as well as their reverse: nocebo effects.
However it happens, pain can happen in the absence of body damage, and there can be bodily damage without pain. Corns points out that a lot of chronic pain sufferers aren’t suffering from a body condition, but a brain one. That doesn’t make their pain any less real, or any less tragic in the effects in can have on their lives.
It’s important to emphasize that these factors are generally not conscious ones. In other words, this isn’t an argument that people consciously choose to be in pain. But the mind, primarily the unconscious mind, has a powerful effect on what we do feel.
It’s a very counter-intuitive conclusion, but we get those when we study the mind.
Unless of course I’m missing something?