If goodness from an epistemic point of view is what we are interested in, why shouldn't we identify justification with truth, at least extensionally? If the name of the game is the maximization of truth and the minimization of falsity in our beliefs, then plain unvarnished truth is hard to beat. This consideration, however, has not moved epistemologists to identify justification with truth ... But why should this be? It is obvious that a belief might be [deontically justified] without being true ... but what reason is there for taking [evaluative justification] to be independent of truth? I think the answer to this has to be in terms of the "internalist" character of justification. When we ask whether S is justified in believing ... we are ... asking a question from the standpoint of an aim at truth; but we are not asking whether things are in fact as S believes. We are getting at something more "internal" to S's "perspective on the world" ... With respect to [evaluative justification] the analogous point is that although this is goodness vis-a-vis the aim of truth, it consists not in the beliefs fitting the way the facts actually are, but something more like the belief's being true "so far as the subject can tell from what is available to the subject". In asking whether S is [evaluative] justified in believing that p, we are asking whether the truth of p is strongly indicated by what S has to go on; whether, given what S had to go on, it is at least quite likely that p is true. We want to know whether S had adequate grounds for believing that p, where adequate grounds are those sufficiently indicative of the truth of p. (1989:99)
There's a lot packed into this.
I've seen no explanation as to why a false belief might be deontically justified. Let that pass. I think it's funny for Alston to press this 'not internal enough' complaint in arguing that the justified belief need not be true. Leafing through the pages of Epistemic Justification, we see Alston frequently formulates perspectival internalism as follows:
PI: Justification is internal in that it depends on what support is available from "within the subject's perspective", in the sense of what the subject knows or justifiably believes about the world (1989: 107)
If you have a justified belief concerning any proposition, is that not a justified belief that the proposition in question is true? I don't see any inconsistency between the claim that justification is internal in the sense outlined by PI and the claim that there are no false, justified beliefs. If your belief concerning p is justified and that belief just is the belief that p is true, the truth of the relevant proposition is internal or "within the subject's perspective". If, however, the subject's belief concerning p is not justified, the truth of p is not internal or "within the subject's perspective". On its face, it seems that insisting that all justified beliefs are, among other things, true beliefs is consistent with perspectival internalism. So, how's truth not internal enough?
It seems that while truth might count as internal according to PI, adequacy of grounds as understood in the way that Alston does would not count as internal. While the grounds might count as internal, their adequacy would not since their adequacy would require that those grounds generally lead to the truth. I think the 'internal enough' stick is an odd one for Alston to use on those who would say that truth is necessary for justified belief.