It sounds weaselish but it depends on what you count as pain. If you count 'response to injury stimulus' then it just needs to be a productive action in respons

It sounds weaselish but it depends on what you count as pain. If you count 'response to injury stimulus' then it just needs to be a productive action in response to the injury. Plants feel pain by that definition. Bizarrely plants can be anesthetized which freezes autonomic responses like rotating to face sunlight or opening of flowers. For an example of a useful injury response without learning: something like 'invest energy in upping the poison production if something bites my leaves' would be a useful response with no learning involved. Of course under that definition you could call alarms that go off when a window is busted a form of pain for the building. Which I suppose highlights just how loose our definitions really are. I'm no pain semantics expert but I expect rightfully or wrongly our definitions of pain are animal neurology centered.