Models are no substitute for observations, but field measurements taken contemporaneously do little good either. The measurements that have to be gathered, such as ice cores, tree rings, and other historical data points are capable of being interpreted in many ways, and are often incomplete or inconclusive. Unfortunately, that's all we have. It goes without saying that the models are untrustworthy, we can't even predict hurricanes with any degree of accuracy. But they are an essential tool in sorting this out, not a guaranteed solution to the problem.

Models of stellar interiors (calculated by hand!) were used extensively back in the '50s to help work out the details of stellar structure and evolution, and they were always constrained by the observations (i.e. they were fudged to give the right answers). Sure, there was controversy and contradiction, but things eventually settled down to a consensus that held up fairly well to new observations. Of course, there was no ideological component to the debate back then. As non-climatologists, the question we should really be asking is, "why is there one now?"

According to the debate on this issue, we are forced to choose between two options, or perhaps some combination of the two:

1) The bulk of the scientists are, consciously or unconsciouly, allowing political and career considerations to color their interpretation of the facts.

or

2) Business interests fear that they will be forced to lose profits because of regulations instituted to control their activities.

If forced to make that choice, my money's on number 2.