First we should start with clarifying what it means to validate a theory as a candidate for true belief.
One of the most popular theories of justification in epistemology at the moment is the idea that a theory is justified when it is shown to be 'likely true'. This position is an attempt to get around the traps fallibilism present in the vindication approach, without having to give up on its merits. Basically, if you can show that something is likely true, this makes it an appealing candidate for true belief, in the same sort of way that showing a theory is definitely true would.
It doesn't really work, though. To say something is likely true is still to make an appeal to certainty. Now, instead of saying that you are certain that p is true, you are saying that you are certain that p is likely true. The difference is not fundamental.
A second, less popular approach is to say that justification is to show a theory is a 'possible truth'. This gets around any of the problems of making an appeal to certainty, but again, it doesn't really work. It's such a weak requirement of a justification to say that 'it could be true' that you could say it of anything, and it therefore becomes meaningless.
Popper made one of the only serious attempts to make the 'possible truth' route work. He argues that a theory is justified as a candidate for true belief, if it is a possible truth in the sense that it has not yet been shown to be false. His 'reasons to believe something' are negative, then, not positive. It's not that the theory has any particular strength we're pointing to to indicate that there's good reason to believe it, it's that it's the only surviving theory left.
But the problem with the Popperian approach is it assumes too much of falsification. Now Popper did admit that a falsification could be wrong, but he failed to recognise the impact this has on his theory. It's only that we're able to eliminate some 'possible truths' in the Popper approach, that makes preferring one possible truth over another meaningful. If we actually can't eliminate any theories, the Popper project falls down.
But validation theories needn't be as strong as to make claims to likely truth or as weak as making claims to possible truth, there is a middle ground, one that is often overlooked.
For part 1 click here