Laurence BonJour belongs to a group of philosophers who are re-evaluating aprioristic/rationalist philosophy. More concretely, these philosophers are both reassessing the a priori as well as mounting a defence of it. Some commentators, however, may say that it's an exaggeration to claim that such philosophers are also re-evaluating rationalism; rather than the more selective apriorism.
Perhaps BonJour goes the furthest towards the kind of rationalism we would recognise from the Rationalists of the late 17th and early 18th centuries. Nevertheless, there are substantial differences between BonJour’s work and these philosophers. The principle difference, to my mind, is BonJour's acceptance of the possible experiential/empirical defeasibility of a priori claims, ‘reasons’ and beliefs.
Is Thought Itself A Priori?
BonJour argues that in reasoning one can't ever rely solely on experience. Even in the case of reasoning about premises which are empirical or experiential, something a priori is required to be added to the reasoning process.
BonJour particularly talks in terms of the relations between the antecedent/s and consequents of arguments (or claims). His position comes in three parts:
i) “If all the things for which there are direct experiential reasons are already contained in the antecedent.
ii) “and if the consequent genuinely goes beyond the content of the antecedent.
iii) “then experience can offer no direct reason (and no indirect reason without assuming some other conditional of the same sort) for thinking that such a conditional proposition is true.” 
What's being stressed here is that the a priori move (or inference) – i.e., from the antecedent to the consequent - can't itself rely on experience. Experience itself doesn't provide the grounds for such a move.
What's stated in i) to ii) BonJour argues can be said of all reasoning and indeed of thinking itself. If the a priori is rejected, it will lead to
“a deep and pervasive version of scepticism, one in which we have no reason for thinking that any of the various seemingly empirical claims that are not directly justified by experience are true”. [2008: p. 102]
The point which is repeatedly made by BonJour is that even if our claims or premises are empirical (indeed even if they're directly or fully empirical), the empirical or experiential itself can't account for the transitions between these empirical/experiential premises and other empirical claims or conclusions.
It can of course be said that such logical moves have been made - and experienced - before and that's an empirical fact. Despite that, experiencing logical moves (in the past) and the logical moves themselves aren't the same thing. We can make an inductive inference about the kinds of moves "average reasoners" make; though that won't tell us anything about the move itself or whether or not it's a priori in nature.
The basic point is that conditionals, for example, can't be shown - or known - to be true by recourse to experience itself; even if the antecedent and the consequent of a conditional are themselves empirical/experiential in nature.
According to BonJour, naturalists and Quineans do believe that
“we never have a priori reasons for thinking that if one claim or set of claims is true, some further claim must be true as well”. [ibid., p. 103]
If such people do believe that, then what does BonJour think will follow from this? He believes that
“there is simply nothing that genuinely cogent reasoning could consist in... the rejection of a priori reasons is tantamount to intellectual suicide”. [ibid., p. 103]
What is BonJour’s problem? Why these severe words? If a priori logical moves are really necessary and fundamental, then even if Quineans reject the a priori in word - they won't do so in deed. After all, BonJour has already argued that they can't do so. Thus there couldn't be any such thing as “intellectual suicide”- according to BonJour’s own lights!
BonJour, Laurence. (2008) 'In Defence of the a Priori', in Contemporary Debates in Epistemology (edited by Matthias Steup and Ernest Sosa).