Let's grant that my ability to use my hands is normal. Suppose a world-class violin maker loses the little finger on her non-dominant hand. This slightly impedes her ability to use her hands. But the incredible amount of gross and fine motor skills that a top violin maker needs to have exceed my own merely normal skills to such a degree that she is going to do better in any non-gerrymandered manual activity (wiggling ten fingers is gerrymandered!) than I.
But my own abilities are normal. So if her abilities exceed mine, how can hers fail to be normal? Yet it seems clear that to the extent that reattachment of the finger would be treatment rather than enhancement, even though it takes someone whose abilities are above normal, and raises her even higher above what is normal.
So we should not define the kind of abnormalcy or disability that calls for medical treatment in terms of a below-normal degree of overall function. For overall function is affected by compensation—the violin-maker's manual skills compensate for her genuine loss. Rather, we must look at something like local function, the function of a particular bodily subsystem. And here it is clear that when she lost her finger, she lost the full use of a subsystem. Disability is the loss of the full functioning of a subsystem, not necessarily of the whole.
But now here we have an interesting thing. An operation that destroys the functioning of a bodily subsystem that otherwise would have functioned properly, even if it does not adversely affect—or maybe even enhances—overall functioning of the person, nonetheless is producing a disability. Now a physician should be a healer. Sometimes to heal one must destroy a subsystem—amputating a gangrenous limb is an example. Even in those cases, the destruction is a moral reason for the physician not to do the operation, though a reason that may well be outweighed (as it is in the gangrenous limb case) by the need for healing.
But this is far more problematic when the destruction of a subsystem is not done in order to heal the system as a whole, even if in some way the person as a whole benefits. Suppose Sam has a job that consumes all his waking hours and involves no contact with people, and his normal interest in social relationships makes him less good at his job. Moreover, suppose the sad economic realities are such that he has no hope of another job. He is going to live a life of loneliness and unfulfilled sociality. Should we give him drugs that would destroy his sociality? Such drugs would improve his life, after all. Yes: but they would do so by destroying a subsystem. And their positive effect would not be a form of healing—at most a form of enhancement at adaptation to unfortunate circumstances. So there is strong—I think conclusive—moral reason why a physician should not give Sam the drugs.
And the same line of thought applies in a much more controversial, because more realistic, case: sterilization.