Indeed, there's a long tradition of skepticism that goes back to ancient Greece. This tradition is about holding up a high standard in epistemology, and trying to dismiss everything that falls below that standard. With the rise of science, that standard was identified with science and with the gathering of empirical evidence. So in practice this skepticism often became scientistic.
I don't have a prejudice against this empiricist skepticism. I just watch as it collapses on its own accord. The austere threshold can't be sustained, as the positivists eventually saw. Huxley's formulations are incoherent, as I showed in the article. The positivists turned Popper's falsification philosophy into the falsification criterion of meaning, which turned out to be unfalsifiable and therefore meaningless.
My point about the screwdriver is that Huxley was implicitly lowering the bar from a need for certainty to a more relaxed, pragmatic standard. Also, workability is about know-how and the application of values and interests, not just logic or data. If we can reject a screwdriver for its practical deficiencies, we should be able to reject religion on the same pragmatic basis. So why be agnostic rather than atheistic about religion, given Huxley's principle that we should hold fast only to that which "works," after trying everything else?
If a skeptic rejects religion for being based on fairytales, how would that skeptic's agnosticism be different from atheism? "Agnosticism" means the lack of knowledge, but here we'd have the knowledge that religion is based on fairytales. So you're painting the agnostic into a corner, whereas Huxley thought agnosticism was an alternative to "immoral" atheistic certainty.
The atheist who rejects religion for being clearly fictional could be skeptical without being agnostic. That is, the skeptic might be initially agnostic in that she suspends judgment until the facts come in, but once they come in the skeptic can render a negative verdict. If religion's clearly based on fairytales, we ought to be atheistic, not agnostic about religion. So when the evidence pertaining to religion is clearly negative, agnosticism makes way for atheism.
Everyone's agnostic in the trivial sense that if we're uncertain, we think we should be more cautious in suspending our judgment. Obviously, when we commit ourselves, we think we know what we're doing and talking about. We can be mistaken about that, as when the religious person takes faith to be a type of knowledge. Huxley and the skeptic come along and say that faith isn't a reliable standard. Instead we need logic, data, empirical certainty, or workability. So the real debate here is about the choice of epistemic standard.
