Atul Gawande has a nicely turned health reform piece in this week's New Yorker that makes quite a few useful points but ultimately says surprisingly little. His thesis statement is simple enough: Countries do best to build atop what they have. He presents this as a rejection of the single payer utopians on the left and the free market ideologues on the right. He backs it up with a very quick tour of other systems around the world: Britain built atop the medical infrastructure developed during World War II, France built atop some voluntary-payroll tax arrangements that had organically emerged, and Switzerland built atop its private insurance system. It's an interesting history but not a thorough one: Any of those systems could have made very different choices and still have been building atop past successes rather than creating something radically new. What Gawande presents as institutional determinism only works in hindsight, and even there it doesn't work all that well. Gawande's conclusion, however, is well aware of his argument's weaknesses. "Building atop what works" is a common trope in American health care circles. It tends to mean subsidizing the current system, putting in place some alternative insurance options for those who sit outside the employer-based market, and generally patching some holes and tightening some screws. But that's not Gawande's conclusion. Rather, he smartly notes that our current system is "more flotilla than ship," and we could build off virtually anything. We could build atop the VA, he says, accepting its offer of better care at the price of less choice. Or we could expand Medicare, recognizing tat it offers full choice but has been ineffective at cutting costs. Or we could choose some hybrid. "We could have Medicare for retirees, the V.A. for veterans, employer-organized insurance for some workers, federally organized insurance for others." That's true, we could. Gawande doesn't say which approach would be most ideal. Rather, he concludes that most any reform will prove preferable. "The system will undoubtedly be messier than anything an idealist would devise," he says. "But the results [will] almost certainly be better." And that's probably correct. Health reform can be very far from perfect and still very much improved over the status quo. But one implication of Gawande's piece is quietly radical: It would be no further from the American experience to institute Medicare-for-All than to strengthen the private market. Indeed, it's actually a lesser leap: Most all Americans will eventually be on Medicare. We are comfortable with that notion. But not all Americans will have private insurance, much less a particular private insurer, much less be required to buy into a private insurer. That, however, is what most people mean when they say we should build on the current system. The center, sometimes, can be far more radical than the fringe.