The problem with liberals, I think, is that they're just not paranoid enough. And so, they frequently agree with ideas, policies and legislation that look good on paper, but turn into something completely different once Republican or Blue Dog politicians get their hands on them.

This post by economist Jared Bernstein is a good example. He's defending the liberal Center on Budget and Policy Priorities (where he works) from an attack by the Washington Post's Robert Samuelson that accuses them of being the "other extreme" from Grover Norquist:

CBPP does not consider Social Security and Medicare “too popular to assail.” The Center has long said that while vulnerable beneficiaries need to be protected, this should not preclude changes to the programs’ benefit structures. It has, for example, supported the use of the chain-weighted CPI to adjust for price changes in Social Security benefits for current and future retirees, if the move to the chained CPI is applied government-wide (including to the tax code) and accompanied by a measure to moderate the effect on the oldest and most vulnerable beneficiaries.

Okay, first problem: "Accompanied by a measure to moderate the effect on the oldest and most vulnerable beneficiaries." Bernstein worked in the White House, for heaven's sake. He must know how unlikely it is that this Republican house (or any other Republicans, for that matter) will ever support adequate supports. (Hell, look what they did to premium supports in the optimistically-named "Affordable Care Act.") And even if a Democratic majority passes an adequate support, a Republican majority will cut it again.

It's what they do.

The Center has consistently supported charging higher Medicare premiums to upper-income beneficiaries; it recommended the changes included in the Affordable Care Act in this area well before that law was enacted. And it has emphasized that Medicare should lead the way in slowing the growth of health care costs. None of these were or are popular moves with many progressives — they would slow the growth of benefits relative to their current path or pare benefits back — but the Center’s support for them is based on policy, not political, analysis.

Yes, I get that. I understand that's how think tanks have to operate -- but it's also like leaving a loaded gun on the table. Once Medicare is means-tested, it becomes just another welfare program. And we just saw what happens to any funding that doesn't have a powerful political constituency.

In short, CBPP has publically disagreed with those who say Social Security benefits can’t be touched at all and that all changes affecting Medicare benefits should be put off limits.

CBPP has sounded the alarm for many years about long-term deficits and has certainly not shied away from addressing the need to raise some taxes on people below $250,000, as well as those above it. One of our most pressing concerns in recent years has been the need for more revenue to adequately meet present and especially future needs of our population, and in this regard, CBPP and Bob Greenstein have on numerous occasions advocated full expiration of the Bush tax cuts., not just the tax cuts for people above $250,000. Yet Samuelson implies our lack of support for even the expiration of the high-end cuts.

In sharp contrast to Norquist, CBBP does not ask members of Congress to sign pledges and discourages them from doing so.We have not come to these positions lightly. Anyone who knows the Center’s work knows how hard we have fought over the past three decades for programs that protect vulnerable Americans. But when our analysis leads us to positions that are potentially unpopular with one side or the other, we do not shy from those conclusions.

They're so earnest, aren't they? And don't get me wrong, CBBP does really good things -- but their work doesn't happen in a vacuum. Here's the problem: They're thinking like scientists. Just like J. Robert Oppenheimer, they don't foresee the harm that may result.

It's a think tank, they'd be crippled if they only considered what was politically feasible -- but they should at least acknowledge the practical danger.