The brain itself cannot feel the surgeon

The cerebral cortex has no pain receptors. Surgery on it can be done with the patient awake, speaking, while the scalp and the lining around the brain are the parts that hurt.

Pain from the head comes from the scalp, the skull, the blood vessels, and the meninges, especially the dura. The cortex, where the reporting is done, does not have the receptors that would make a cut painful. Neurosurgeons have used that fact since Wilder Penfield mapped the motor and sensory homunculus in awake patients in the 1930s: open the skull under local anaesthetic, stimulate a point on the cortex, and ask the patient what happened. A hand moves. A sensation flashes on the tongue. A word will not come. The patient does not feel the electrode as pain.

Awake mapping is still how speech and movement are protected during surgery near them. The person names pictures while the surgeon stimulates, and a site that interrupts the naming is a site to leave alone. None of that would be possible if the tissue under the electrode hurt. Headaches are not a counterexample. They are the other tissues, the ones that do have receptors, complaining. The thinking part cannot tell that it has been touched.