Right-to-life conservatives and right-to-die liberals have about exhausted their rhetorical arsenals, with the former calling the latter secularist killers and the latter calling the former hypocrites, theocrats, and (gasp) tramplers of states' rights. Meanwhile, many in the media who gleefully trumpeted how Terri Schiavo's case had turned rule-of-law conservatives against right-to-lifers have noticed with disappointment that it also had turned disability-rights activists, and liberal lions Jesse Jackson and Sen. Tom Harkin, D-Iowa, against the right-to-die crowd.

With the left-right morality play marred by these untidy alliances, this may be as good a time as any to reflect on how best to handle all of the other Terri Schiavos whose fates our society must decide one way or another in the coming decades. There will be many millions—not only the few who languish in persistent vegetative states, but the many more who will be so demented by Alzheimer's disease (like my late mother), otherwise incapacitated beyond hope of recovery, or wracked by painful terminal illnesses that their lives may be deemed not worth living.

The choices that we face, both individually and as a society, are all the harder because they are not only about religion, secularism, clashing moral values, disputed medical prognoses, and debatable evidence of patients' intent. They are also about money.

President Bush's assertion that "it is wisest to always err on the side of life" is morally compelling. But if taken to absolutist extremes, it is on a collision course with the need to control the soaring costs of medical care. The number of terribly diminished lives that we prolong will depend, in large part, on how many trillions of dollars we divert from other priorities.

Pretending otherwise, as right-to-lifers are wont to do, impedes sensible policy-making. So does seizing upon this case mainly as an opportunity to bash Bush and Congress, as liberals are wont to do, for jumping into the save-Terri movement while seeking to cut funds from the Medicaid program that helped to sustain her.

The expense of sustaining Terri was relatively modest, but far from trivial. The average cost per patient is $80,000 per year at the Florida hospice where she died. Far greater is the cost of keeping countless terminally ill old people on respirators, kidney dialysis, and other life-prolonging machinery and treatments. Some 28 percent of this year's Medicare budget of $290 billion (projected to grow to $649 billion by 2015) will be spent on people in their last year of life. In many cases, the main effect will be to prolong the pain of impending death.

If the right-to-lifers put our money where their mouths are, we may spend more tax dollars to postpone the expiration of post-sentient Alzheimer's patients than we spend to educate poor children. Christopher Hitchens, in Slate, caustically imagines "rows and rows of deceased Americans, all connected to life support until the crack of doom."

On the other hand, if the right-to-die crowd has its way, euthanasia is unlikely to stop at such reasonable moral choices as honoring the clearly expressed prior wishes of now-incompetent patients not to be kept indefinitely on life support, or giving those in unbearable pain an alternative to blowing their brains out with a bullet. Before too long, harried families and medical cost-cutters might be euthanizing nonterminal patients with treatable pain, mothers suicidal with grief at the deaths of children, teenagers severely depressed after being paralyzed in accidents, newborns with Down syndrome, and so on.

Do I overstate? Well, Princeton University bioethicist Peter Singer has famously argued that it would "be right to kill" severely disabled newborns at their parents' request, including "a hemophiliac [who] can be expected to have a life worth living [but] not quite as good as a normal baby."

And consider how legalized euthanasia has progressed in the Netherlands, as chronicled by Washington lawyer Neil Gorsuch in a Wisconsin Law Review article last year. In 1984, the Dutch supreme court exonerated a doctor for granting a request by a 93-year-old woman who wanted to die because she was bedridden by a hip fracture and unable to eat or drink. A decade later, the court extended the euthanasia defense to a psychiatrist who had supplied lethal medication to a physically healthy 50-year-old woman who wanted to end her "unbearable" psychological suffering from misfortunes that included the suicide of one son and the cancer death of another.

In 2001, the Dutch parliament authorized euthanasia at the request of any patient—including a child as young as 12, if the parents consent—whose physical or psychological suffering is "lasting and unbearable," no matter how robust his or her physical health. And late last year, the Groningen University Hospital proposed guidelines for killing "suffering" malformed infants if their parents don't want them.

Could that happen here? Should it? Euthanasia puts us on the slipperiest of slopes. The only stopping points can be seen as arbitrary: the line between deciding not to plug a patient into respirator, and pulling the plug; between such an "artificial life-support" machine, and a simple feeding tube; between withdrawing a tube to induce slow death by dehydration, and using a lethal dose of morphine to make it fast; between obtaining a competent patient's considered consent, and implementing an unconscious patient's living will; between that, and inferring intent from disputed testimony by feuding family members who claim to know what the living will would say had the patient thought of creating one; between the mental incompetence of a 90-year-old Alzheimer's patient, and that of a newborn; between brain-dead, and persistently vegetative; and so on. Thus can the right to die morph into a right to kill.

The Schiavo case was scrutinized by more judges, doctors, ethicists, legislators, and chief executives named Bush than any other right-to-die dispute in history. It unfolded under a well-thought-out state law allowing removal of Schiavo's feeding tube only if there was "clear and convincing evidence" that she would have wanted it removed. Yet still, the outcome is disquieting, and not only because we watched the state-sanctioned death by dehydration of a 41-year-old woman over the passionate objections of her parents.

While I want my feeding tube to be disconnected if ever I am in Terri Schiavo's state, I question the courts' finding that this is what Terri (a Catholic) would have wanted. That finding hinges on Pinellas County Circuit Judge George Greer's less-than-convincing February 11, 2000, opinion crediting the unavoidably self-interested, initially undisclosed, and disputed recollections of Michael Schiavo (and his brother and sister-in-law). The same Michael Schiavo wants to marry the woman whose two children he has fathered while seeking to speed his wife's demise. This is "clear and convincing evidence"?

Perhaps the law should be changed to bar such an outcome in any future case in which the patient's parents or spouse object. Still, any error by Judge Greer or the many other judges who have upheld his decision comes nowhere near justifying the ill-conceived effort by Congress and the president to have the federal courts second-guess the state courts' final judgment, in a fully and competently litigated case involving a tragic family dispute.

We will see many more cases that present such tragic choices and ugly disputes. Especially if we fail to prepare or refuse to honor living wills. University of Rochester economist Steven Landsburg has argued (in Slate) that these wills should be disregarded as "the preferences of someone who is [too incapacitated] to care what we do." Princeton jurisprudence professor Robert George has claimed (in a National Review Online interview) that people cannot make sensible "decisions in advance about whether to have themselves starved to death if they eventually find themselves disabled."

To the contrary, living wills can do a lot of good. They can ease the awful burden of decision that faces an incapacitated patient's loved ones. They can avoid or resolve many (albeit, not all) family disputes such as the one over Terri Schiavo. Perhaps most important, they may chart the only safe course for our society to steer between the Scylla of wasting scarce dollars prolonging the lives of people who would rather be dead and the Charybdis of rampant involuntary euthanasia.

"We've got a duty to die and get out of the way, with all of our machines and artificial hearts, [and let] our kids build a reasonable life," former Colorado Gov. Richard Lamm said in a much-criticized 1984 speech. He had a point. But this is not a duty for which we should be conscripted. It is one for which we should volunteer.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.