A reasonably mature thinker holds beliefs in terms of Bayesian estimates, and as Dennet says, one should always be willing to entertain evidence that contradicts your current Bayesian estimates. That doesn't mean you don't have beliefs.