Yeah, I definitely don't agree with you there.  Economics has weak scientific credentials, but that doesn't mean that good faith efforts to apply a scientific approach with transparent methodology are automatically inferior to common sense.  "Common sense" is another way of saying a priori.  It was sensible, at one point, to say that the world was flat.  Eratosthenes didn't figure out that the Earth was curved by just using common sense.  For that, he needed data.

I also don't understand eschewing data entirely.  Numbers don't lie; people do.  Take, for example, one of the arguments that Amity Shlaes and others are making about the New Deal.  They say that the New Deal didn't decrease unemployment.  However, the lie isn't really the numbers.  The lie is that they don't consider the people who were working for the WPA and other government entities to have actually been employed.  That's not a "job" to them, it's "work" (which is a direct consequence of their particular brand of economic dogma).  It's not statistics that are to blame here.  It's plain old, run of the mill semantic sophistry.  You start there and then the numbers follow.

Incidentally, you can make the same criticism of every single social science out there.  If you understand math and methodology, it's really not difficult to tell good statistics from bad.  If you either don't understand or aren't willing to, then perhaps all statistics seem equal, but that doesn't mean that they are.  Ultimately, the major intellectual leap that science made (thanks, Galileo!) was to understand that we need to check our assumptions against what happens in the real world.  Economics, like other social sciences, does lack a laboratory, but that doesn't mean it can't observe.  Even some of the hard sciences have this problem, geology being one example.

When you say that you don't trust data generally, I can't understand that.  I'm not even sure that you really mean it.  You keep talking about the ratio of debt to GDP.  That's data.  The old rag about how you can say anything you want with statistics has the ring of truth, but it's shallow.  What I hear when people say this is, "People lie."  Or perhaps more accurately, "People are inclined towards beginning with an ideological construct and trying to fit information into this construct rather than letting the information inform the construct."  That's precisely why common sense just isn't good enough.  Relying on common sense is how you get a Fed chairman who thinks that financial markets are self-policing.  Common sense didn't tell us that was wrong, reality did.  That's the importance of data.

In that sense, I think the mission is two-fold.  One, you have to actually assess the data that's being presented.  That means asking whether the math and methodology are sound.  Hard sciences aren't immune to this either.  Remember the cold fusion guys?  This is why a healthy peer review culture is important.

The other part is more difficult and it's what I was trying to do in my most recent post.  That's to try and get down to the ideological frames that are driving the discussion.  Common sense is a fine starting point, but to me the most important factor is whether we are doggedly adhering to conclusions that we came to prior to observation, despite the fact that observational data might contradict these beliefs.  That's really the most important lesson of science: Skepticism.  It requires the simultaneous recognition that our current understanding is always subject to being incorrect or incomplete, and therefore subject to revision, and that the only way to actually improve this understanding is to allow this process to occur.

If economics can make progress by refining understanding through applying scientific methodology, then it's scientific enough.  You'll notice that no one really debates the foundations of microeconomics too much right now.  That's probably because it's easier to understand what's going on with a single firm or individual than it is to understand what's going on with all firms and individuals simultaneously.  It's also probably because our understanding of micro concepts isn't being deftly kicked in the shins by reality.  The neoclassical economic model was really the first attempt at giving micro foundations to macroeconomics.  Neokeynesian economics do the same thing for the Keynesian model.  In the meantime, we've come to understand better how Adam Smith's original common sense about markets wasn't the complete picture.  We understand things like market failures and externalities and we can measure these things in many cases.  We also have newer concepts like stickiness and incomplete information that sharpen our understanding.

In short, I think that you're throwing out the baby with the bathwater, which always seems easier, but doesn't really produce the desired result (unless you're really trying to throw out the baby).  Also, as a friend of mine is fond of saying, "Common sense ain't so common."  It's worth noting, too, that there are still people who argue that the Earth is flat.  Their existence alone isn't grounds to simply throw up our hands and dismiss all of the findings of physics because we don't all agree on the implications.  If universal agreement is a requirement of bona fide understanding, then we don't understand anything about anything and we never will.