In response to certain questions posed by John Bishop and others regarding states of belief and how they are possible, on the one hand, and whether they are justifiable, on the other, I presume to advance here a definition for doxastic states.Footnote ¹ I further presume to rest upon this definition an assertion that the only possible justification for any belief will be empirical in nature; that to attempt by other means to vindicate belief is to toil under the weight of circular assumptions and to flirt unnecessarily with nonsense. But my programme is not that of the logical positivist. I do not wish to embarrass the faithful on the ground that they lack a foundation, or that the logic of faith is more puzzle than proof. The project here is less ambitious than that, yet in some sense more radical, for ultimately I take aim not at the content of any particular state of theistic belief, but at the very state itself, and I mean to suggest the possibility that it might not exist, or at least that we require further understanding of the elements of belief in order to offer a doxastic venture or a leap of faith as a reasonable stance of the faithful.

It has been claimed, by Ayer for example, that language and logic must always render certain claims of belief nonsensical, that such claims exhibit the grammar but not the content of real statements or propositions.Footnote ² Such claims as would reference the perfection or the infinite character of God, or similar objects lying outside earthly experience, will always suffer from a deficit of verifiable particulars, and hold no chance of synonymy between terms. This I offer as a rendition of the strong claim against metaphysics, and though I do not dispute it I would pursue a weaker claim, as follows.

The limits of language and logic may render speech wholly insufficient to expressions of belief, where we understand especial instances of belief (such as theistic belief) to refer to internal states describing or defining (or divining) such matters as the ultimate ends of humankind, or universal morality, or theology – in other words, supposed transcendent realities which are beyond experience but which nevertheless are said to be in some sense believed in. I trust that I am safe in placing the Christian belief Bishop wishes to justify under this rubric, and while I am not competent to say that transcendent beliefs cannot be held, I do mean to say that claims of belief, as commonly expressed, must often be necessarily hollow; that their formulation, in logical terms, can result, as Ayer and others have suggested, in nonsense.

In his 2002 paper, and his 2007 book, John Bishop does not ask whether transcendent claims make sense, per Ayer's analysis. Instead, he asks: is it ‘justifiable to ‘have faith’, in the sense in which Christians ‘have faith’?’Footnote ³ Therefore the first claim I wish to make against Bishop is simply this: that in order to discover whether such states are justifiable, we are obliged first to determine that they do make sense. Such a determination is mandatory, and in moving quickly around the issue Bishop oversteps himself, and proceeds in advance of reason. But let the reader be advised that in singling out Bishop and his recent essays I merely take aim at the living. John Bishop, a serious philosopher and an engaging writer, is here merely identified as the latest and perhaps the best standard-bearer of the time-honoured claims I wish to examine.

So what really do we talk about when we talk about believing? The logic of belief – the systematic examination of doxastic states – takes for its field of reference the conditions and facts that justify belief in particular instances, and examines the dynamics of that belief vis-à-vis change, on one hand and volition, or doxastic voluntarism, on the other. Belief change is the more straightforward of these problems, as it involves merely a kind of fattening or winnowing or sorting of doxastic states via the transmission of new information. Doxastic volition is an altogether different beast. Yet it is relevant here, because it touches on the work of three philosophers whose claims this article aims to address: Bishop himself, who posits the ‘doxastic venture’ as a model of Christian faith, and two of his logical predecessors: William James, who a century earlier had framed such ventures as ‘leaps of faith’ in his lectures (notably ‘The will to believe’ and ‘Is life worth living?’);Footnote ⁴ and Blaise Pascal, whose famous wager, placed before us in the middle of the seventeenth century, was in its own way the proposal of a scheme somewhat akin to Bishop's.Footnote ⁵

In order to discuss Bishop's programme, and whether such an approach is reasonable and useful in justifying faith, I want to ask three questions here, to wit: (1) what are the particulars of a state of belief when such a state is supposed to incorporate no transcendent or religious matter? (2) Are we warranted in admitting of doxastic states in which transcendent matter does play a role? Or (3) is it the case that once we march beyond the boundaries of what is verifiable, our needs for justifications evaporate along with the substance of our beliefs? Answers to these questions should suggest to us whether it is permissible to hold to a claim of belief whose components fail of articulation in logical terms – in other words, a claim of transcendent belief. This article will first undertake to examine a state of belief that is purposefully devoid of transcendent matter, and to regard that doxastic state as, to appropriate Bishop's gambit, a doxastic venture; and then proceed to explore whether and how transcendent beliefs differ fundamentally from the exemplar of mundane or terrestrial belief outlined. We will then ask whether transcendent belief states can ever be warranted. In essence, we will be asking whether, if transcendent belief is unwarranted, it is perhaps merely an epistemic or linguistic vice – again, per the logical positivists – that allows them into common discourse in the first place. If such is the case, they require no justification, but neither can we stand upon them as the foundations of our believing.

All this being stipulated, let us make an example of a doxastic venture devoid of any transcendent content, as follows.

At a certain place of employment, a, b, c, and d are members of a group of workers G upon which we are focused; there is yet a fifth person, f, who stands outside the group, but who is being considered for admission. As a group G distinguishes itself by the following behaviour: each Wednesday there is an event at the Apex Hotel Lounge, between five and seven o'clock p.m., at which for a modest charge of twelve dollars admission per person, each individual of any group of five persons is entitled to two beers and all the fried chicken livers he can eat; between those same hours of five and seven, additional beers and fried treats can be purchased at prices much lower than those offered elsewhere, so long as the group of five remains intact. Once a member leaves, or if he arrives with less than the initial fee, no part of the offer holds, and if the other members wish to eat and drink they must pay the high prices on offer at all other times for all other comers. To this Happy Hour, group G has habitually retreated, week after week, because a previously unmentioned member, e, rounded out their fivesome; but e has left to accept employment in another state, and can no longer make them five, hence consideration of f as a new member.

Now these five individuals have in common many mundane details – for example their work, their standards of living, and their linguistic and cultural backgrounds (as well as a fondness for beer and chicken livers). They know certain obvious things about one another. Other things, such as what an individual might say about the make-up of his family, the plumbing in his house, or the mileage he gets in his car, they are prepared to believe. It being of no particular importance to b whether a's commode is easily overflowed, he takes it on a's word that such is the case. But it is stressed here that a feature of the relationships between them all is an absence of need for especial trust of any kind. It does not matter to b whether the toilet at a's house will flush without incident, or how many children live in that house, or whether or not those children are top-notch students: if a makes a claim about any of these things then b, c, and d are prepared to believe it. The ease of this system of knowledge and belief between them all has never encountered an embarrassment – no one, so far, who complains of a stubborn commode has been found out in reality to enjoy a high performing one. So we can fairly say that states of belief among these group members are warranted, or not, in virtue of these states of affairs: (1) there are simple things they clearly know about one another; (2) credulity in small matters constitutes no risk; and (3) the atmosphere they enjoy is one in which so far credulity has paid off, or at least done no violence to trust. States of belief among them have warrant only up to a point; they are not called upon, nor do they voluntarily attempt, to draw grand, shocking, or unnecessary inferences about one another which would require more robust states of belief and hence more knowledge of one another.

Now we return to Happy Hour. Group G requires a fifth member to guarantee the continued benefit of its weekly ritual, and f presents as a candidate. The others must believe only a few things about f to ratify his membership: he must be willing and able to dedicate the hours between five and seven p.m. every Wednesday evening to the enterprise; and he must have, every week, twelve dollars at his disposal to contribute his share. Up to this point, every other member has demonstrated his allegiance, never missing a Wednesday, always staked to at least twelve bucks, hence never been responsible for a forfeiture of those good times. Now the question falls to f, as does the responsibility to offer warrants for the others’ faith in his candidacy. Yes, he has told them, ‘I can spare all my Wednesdays, and I always have at least twelve dollars on me’. It happens, this fateful afternoon of decision, to be Wednesday, and it is nearly five o'clock. The group has offered a place to f and he has accepted. The other members, loitering about the mezzanine waiting for the end of the business day, discuss the likelihood of success with f. ‘Does he have twelve bucks?’, asks a, to which c replies that earlier he saw f at the lunch wagon purchasing a two-dollar burrito, which he paid for with a twenty-dollar bill; so at that time, at any rate, he had at least eighteen dollars. But b chimes in that an hour later f had sought him out to repay a recent loan of five dollars, so that among them they can now only know for certain that f had – at that time – thirteen dollars in his pocket (and that he has demonstrated the personal honour to make good at least one small debt). At last it is five o'clock, and f has just sprinted round the corner to join them on their way to Happy Hour. As they walk up the street towards the Apex Hotel, a down-on-his-luck gentleman steps into their midst, looks f directly in the eye and asks if by chance he might spare a dollar or two, to which f responds, perhaps with a touch of reluctance in his manner, by going into his pocket – this pocket which our group was at one point certain contained at least a twenty-dollar bill, then eighteen dollars, and (finally) thirteen dollars – and retrieves two one-dollar bills, handing them over. In spite of the limitless possibilities regarding how much cash is actually on f's person, our group now shares only this certain information: f has at least eleven dollars in his pocket, which would stand him for a single gin-and-tonic at the airport lounge, sans gratuity, but not for Happy Hour at the Apex, and if that is all he has his status as a member of group G is reversed and simultaneously the group is robbed of its ritual. The Apex now stands a half block away, a moment of truth is near at hand, and nervous glances ricochet between members of our group. We are at a point at which it matters to all the members of G whether f is good for his share, and their credulity is brought to bear in a most serious instance. In a few moments’ time G is mustered at the door of the Apex Hotel Lounge. All have proceeded onward, with f bringing up the rear, on the belief that f will make good his pledge in the bargain.

Here is a shared state of belief, the shared belief of the first four members, based upon certain knowledge of the eleven dollars f has on hand, bolstered by a presumption – in principle verifiable, and soon to be known for sure – that f must have upon his person at least twelve. This presumption is based, one must admit, upon reasonable inferences: that f probably started out with more than the original twenty-dollar bill with which he was seen to purchase his lunch; that he would not have proceeded this far if such were not the case; that he would not have unbelted more than a dollar for the panhandler a few moments ago if it had meant a forfeiture of his status in the group, etc. And in those moments, as they all exchange money and pleasantries with the Apex doorman, and turn to watch as f proceeds through the door, the members stand vindicated in their belief, their hearts warmed towards their newest member as he goes yet again into his pocket and produces a second twenty-dollar bill, thus gaining his own entrance and assuring G another blessed recurrence of its Happy Hour ritual.

Now there are no heroes in this tale, no miracles, and no acts of great generosity or feats of especial daring. Nevertheless it encompasses a scenario in which belief plays a role, and in which belief, in some small but materially genuine way, is vindicated through what I feel justified in calling a doxastic venture. The members of G start with some certain information about their Happy Hour rookie. The inferences that serve to bolster that certain knowledge as it bears on a particular, wished-for outcome are scientific in that they are expressible in logical terms and knowable in principle. The above transaction among believing agents fulfils, I wager, Bishop's doxastic venture, which he alternately allows himself to describe, following William James, as a ‘leap of faith’. Some necessary information – necessary, that is, if we were to transpose this situation from one in which some measure of belief plays a part to one in which agents act on certain knowledge – is missing. No member of G wished to reveal himself as an insufferable bore by insisting upon proof, from f, as to his cash-on-hand; no member of G was willing to go to extraordinary lengths to get a first-hand look at the money contents of f's pockets on this Wednesday afternoon. No, but hidden in the depths of f's trousers, all along, were the answers to all the other members’ questions. And it was those hidden answers – again, knowable in principle and therefore not freighted with any transcendent qualities – whose unknown status necessitated some degree of belief, which made necessary their joint doxastic venture, their ‘leap of faith’.

Now Bishop distinguishes doxastic venturing from what he understands as the fiducial model of faith thus: the doxastic venture endorses ‘believing beyond what the evidence warrants’, whereas a fiducial model does not, but merely challenges the believer, in a circular and safe gambit, to ‘trust the trustworthy’.Footnote ⁶ In our illustration the limits of what each participant knows are plain to see; for certain of their beliefs there is no immediate evidence, if only for the reason that they refrain from demanding to see that evidence; therefore their faith in f and their prospects of success with him rest on believing beyond what the evidence at hand warrants. Thus their belief remains identical to Bishop's model of Christian, or more generally, theistic faith. Now the reader may note with grief or astonishment here that I ask equal value be given to God-in-Heaven on the one hand and a twenty-dollar bill on the other. I can only admit without immediate comment that the reader is not mistaken on this point.

Let P be the proposition that f would qualify as a member of G at the crucial moment; then with reference to a member of Group G we can now state with confidence that BaP, or that member a (for example) believed, before his knowledge was certain, that P. The same sentence would hold for all the other members of the group. Once the entire group is safely inside the Apex Lounge, and once f has vindicated their beliefs, we can fairly state that KaP (a now knows that P) to express the new state of affairs wherein a's knowledge is certain, and a has passed through his doxastic venture into a state of certainty. We can say the same for every member of G.

Of special note here is the transformative aspect of our illustration. When we talk about believing, we are talking about doxastic states, and doxastic states are those states that can be expressed in terms equivalent to something like, for example: all the members of G believe that P. When they comport generally with our illustration, they are mental states – shared or not – poised on the verge of transformation, by observations, experiences, or events, into states understood as knowing states, such as expressed in K(∀xGx)P; thereby they yield valid inferences, such as it is demonstrably the case that P. Under this scenario, a leap of faith is altogether an appropriate metaphor for that interim or transformative status between believing, on the strength of probability and sound inference, and knowing in virtue of observed events. But note that the leap of faith we describe here is not followed by an endless free-fall into some transcendental ether. Here, a state of belief only precedes a state of knowledge; it presupposes, without necessarily predicting, some knowing state as its consequent. Note too that regardless of f's performance our end-state would still have been one of certainty; if disappointed in their expectations for f, group G would not have gone from being believers to unbelievers; instead they would have taken their leap of faith believing in the truth of one proposition, and hit the ground knowing the truth of another. If f had failed them, had reached for his money and come up a dollar short, we would simply rewrite our earlier inference to reflect the fact that, upon acquiring further knowledge, group G found itself no longer warranted in the belief that P. They would have made a leap of faith, and fallen upon the hard ground of discovery. Their bruises would then become emblems of new states of belief.

The implication here is that doxastic states are prior states; that the believing states of the members of G were not worthy of being entertained on their own merit or for any great length of time, but merely as the precursors to knowing states. This point is important enough to warrant repetition: a state of belief, in this scenario, serves the believer only in so far as it sustains him in a shift from not knowing to knowing. It is perhaps merely an intuitive notion, this precursor quality of doxastic states, but experience tells us it is an epistemic commonplace, and that leaps of faith should result in changes of state and in positive gain: namely, certain knowledge. Under these criteria we may say that any state of belief is warranted – or not – not in virtue of its content, but in virtue of the believer's epistemic expectations along with his plan of fulfilling those expectations. The believer, in holding to a belief, declares a stake in his own doxastic venture, and is therefore interested in carrying his venture to its conclusion – taking that leap of faith. After such a leap, belief is not strengthened or upheld but instead abandoned or neutralized in favour of knowing. Thus is belief justified, whether it prove true or false.

It remains now to be seen how doxastic states differ when they include transcendent or metaphysical elements, and whether a state of belief whose contents include transcendent matter can adhere to this doxastic rule. The question is: how do beliefs whose contents are beyond experience, hence held aloof from the possibility of transformation to knowing states, differ from the mere terrestrial beliefs of the kind that comport with our Happy Hour illustration? The transformative state of belief experienced by G can be understood as a phenomenon guiding them in predicting a small part of their future in a useful way. If the illustration comports with what states of belief really are, it demonstrates then that the state can also be described as a function, and that it is a function that bears out the efficacy (the truth or fitness) of the state. If one accepts this definition, one is forced into the understanding that doxastic states, when they are associated with transcendent matter, necessarily fail in function: they present themselves as endless states, which cannot change in any way that would vindicate or falsify any particular belief. If God is not known, and cannot be known, to the believer, his belief frustrates its own function.

Just as there are logical rules of inference, there are applicable rules for states of belief – doxastic rules that must be in force for any authentic state of belief to take hold. These rules constitute the criteria under which we believe in belief. To entertain claims of belief we must discover and apply the rules. One rule is, as discussed above, that belief must be understood as a precursor to knowledge. If such is not the case, a state of belief has everything in common with a state of ignorance, without promise of enlightenment. An agent who claims belief in an object without any plan for discovering the truth of his object is pledged to a state of confusion, and whether or not the goals be transcendent, or noble, or beautiful, we must ask: where is the value in such a condition as that? And where is the logic? And we must further ask: does this agent in fact believe anything? If belief presupposes the aim of knowing, reason forbids the holder of any belief to fail to answer questions about what sorts of evidence would constitute vindication of his belief. At the very least, a state of belief unaccompanied by an experimental regime of one sort or another has a perfect logical equivalence to a state of primal ignorance. In other words, if x is said to believe that P, and if x is unable to submit criteria by which P might be known, then we can only say that a statement which claims x believes that P is in all ways logically equivalent to a statement which merely claims x does not know that P. At some important juncture – precisely that juncture at which we justify our beliefs – states in which x believes as opposed to states in which x does not know become indiscernible. In such cases believing is merely equivalent to a poverty of the means of knowing; the status x occupies as a believer in P is not discernible from a condition of agnosticism regarding P. Under this scenario belief is agnosticism and nothing more. A proposition constituted of elements that lie beyond empirical means is, necessarily, a claim of agnosticism in regard to those elements and therefore in regard to that claim. I do not pretend here to have positively falsified any particular transcendent belief allegedly held by x. But I do question such a belief's status as a belief. And I suggest to the reader that by assigning x this new status as an agnostic, I excuse him from any obligation in justifying himself, for agnosticism is not a state one can be forced to abandon: if one does not know, one just does not know. Furthermore, if x is not seeking for evidence of his claim, then neither can all the evidence against it have any meaning for him. For him it is the claim that is worth holding to – not a process of discovery, not truly a doxastic venture as undertaken above by group G and its thirsty members. Thus the claim of belief held by x seems purely aesthetic or emotional or formal in character: there is some immediate satisfaction in the claim itself, and perhaps if it were tested and found to be true it would lend balance to his world and legitimacy to his acts, but its truth is not at issue so long as he is unwilling to submit it for proof. A person unwilling to see his belief falsified is not undertaking a doxastic venture, not making a leap of faith; such a person only repeats a static mantra: God is infinite; Life is Beautiful; the World is underlain with Cosmic Purpose … Such claims function only as retreats to ignorance as the most tenable option; or, in other words, agnosticism.

As believing and discovering agents, the members of G distinguished themselves by a material interest in transforming their beliefs into states of certainty. Such was the substance of their doxastic responsibility; such was each individual's ethical justification as a member of a group for believing – as a responsible member of a group sharing a material goal. Believers in transcendent claims, like agnostics, distinguish themselves by aiming at no such goal. The members of G, prior to arriving at proof, were not agnostics because the conditions under which their belief could be favourably acquitted were spelled out in the terms of their venture. They undertook their leaps of faith not to enjoy the religious sensation of falling but to land on their feet in a place of new and useful knowledge born of their initial hopes and beliefs. They were not agnostic with regard to the probability that f would lead them triumphant to a reward of cold beer and fried chicken livers, even if initially they were in a state of incomplete knowledge about the likelihood of such a triumph, and even if they might have been proved wrong. The members of group G, in this scenario, stand out as true believers, justified. I am curious to know, but not competent to say, how believers in transcendent claims can match the robust nature of the belief demonstrated by a, b, c, and d. More to the point, I am curious to know under what terms belief in unknowable and untested objects can be termed belief at all. Here is the pivot upon which my unease with transcendent doxastic ventures tilts. If doxastic venturing provides a scenario by which we more firmly may conceptualize belief – by which we believe in belief – and if we are to trust it, then what precisely is the element that joins a logical concept of mundane belief to a salvageable logical concept of transcendent belief? These two types of venture must share not merely a common but a fundamental aspect; otherwise we are left with two varieties of belief sharing no meaning between them – making a new logical framework for doxastic ventures necessary. But I am on the whole satisfied with the old logic of belief, and do not wish to discard it. I am left feeling that the doxastic venture as proposed by Pascal, by James, and by Bishop, remains an unwarranted concept.

For, in a word, how does the transcendent doxastic venture pay out? If it pays at all, it does so by empirical means. In our model, beliefs held this moment – if they are to serve their functions – are held in order that we may see them vindicated in the future. We act upon them in accordance with the ways in which they comport with what is immediately known; it is from the combination of our knowledge and our beliefs that we infer likely outcomes, and good reasons for acting. To hold to a belief whose core is forever beyond our knowledge, and to say we are justified in the belief on account of its place in the calculus that spurs our actions, is to say we must always act partly out of ignorance. The gravity of even a beautiful mystery is no basis for decision-making; even a beautiful mystery represents, in the puzzle of ethics, a formless, invisible piece, which in virtue of its formless nature can never complete the whole. Bishop wisely has stipulated that ‘To believe is, indeed, to believe true[.]’Footnote ⁷ I only wish to point out that this very claim (which I do not dispute) implies the possibility of discovery. What is true, as we all know, is demonstrably true. What is not demonstrably true is either false or unknown. The overriding claim I wish to defend here is that a state of logical acquittal is implicit in the language surrounding any so-called leap of faith or venture: one chooses to believe, on the understanding that soon enough one will know. James, for example, illustrated his ‘leap’ by asking his listeners to envision an actual leap – over a yawning chasm – in which the leaper chooses somehow to become a container for the belief that success shall be his. In holding the belief, the adventurer assures his chances, and thereby preserves his very life. James did not offer an illustration in which a material outcome played no role, and neither, in my reading, did Pascal. Illustrations for a commitment to transcendent ‘realities’ are always laid upon material foundations, and in making the ‘meta-leap’ from rocky crags to faith in God, philosophers err, or they implore their readers to err: they fail to bring along fully one half – the material half – of their argument, and so consign their faithful doxastic adventurers to mere agnosticism, in the best case. This is a failure not strictly of philosophers, but of their evidence, to be all of one kind, a failure of all their terms to exist comfortably in one world. It is not a consequence of bad faith on the part of theologians and philosophers of a certain stripe, and I beg the reader to accept that it is only these incongruities that constitute the kernel of dispute on which I now stand, unwilling and unable to accept any so-called leap of faith or believing venture as helpful or justifiable.

In the above scenario the members of group G shared a belief or a series of beliefs about f, and more broadly about some state of affairs that would come to pass if and when f was vindicated. We have pursued a simplistic illustration of those states of belief; in their simplicity they stand correctly as models for actual states of belief. We noticed certain things about them: they formed around certain known things; they formed in reference to and in anticipation of certain unknown things; all things in those two categories were fitting objects of scientific, earthly enquiry, because all were quantifiable and knowable in principle; nothing in either category made reference to transcendent matter. A looked-for state of affairs was at issue; certain facts about the realization of that state of affairs were understood to already be in place, certain others were unknown but probable, or unknown but knowable. Now the object is settled for us, in that we have seen the venture through to its conclusion, and imagined a number of possible outcomes, so that we can agree upon a definition of belief and we can begin to understand the dynamism of such belief.

Over a century ago, William James suggested his audience go venturing as Bishop has done, on the assumption that ‘we are supposed to gain … by our belief, and to lose by our non-belief, a certain vital good.’Footnote ⁸ This expression echoes Pascal's claim that choosing to believe was altogether the only sane option, owing to its super-dominance in terms of possible outcomes. James, like Bishop, was concerned about the justifications for transcendent belief, stipulating: ‘[O]ur faculties of belief were not primarily given us to make orthodoxies and heresies withal; they were given us to live by.’Footnote ⁹ And he further advised students and philosophers alike to understand that the attempt to pry action apart from belief was a logical barbarism. James's logic, along with Pascal's and including Bishop's, is impeccable. My contention here is merely this: that it is a logic that survives only when matters of the unseen and the unknown are removed from its midst. If faith is in some sense a gamble, let me advise the gambler that the added weight of the unknown, dark and amorphous as it is, will not improve the odds of a lucky strike. And what is to be hoped for in expressions and acts of faith is that the manifest goodness of such a lucky strike will be known to the venturer by its earthly parts, its material outcome, the acquisition of knowledge, and by no other means. The doxastic venture is possible, and it is real. But it does not account for a state of mind in which theistic or otherwise transcendent beliefs are held true.

Bishop claims the case is otherwise, and key to his assertion is a strategy that marks most clearly his departure from his predecessors, namely his use of so-called doxastic frameworks, which I understand to delineate states of transcendent belief as something apart from the mundane doxastic states I have tried to describe.Footnote ¹⁰

So before we finish, let us pause here and discuss the programmes of Bishop, of James and even of Pascal – in a word, those whom we find recommending to us that we believe in unseen things – within the context of a philosophy of language, in which we concentrate on how things are said, as well as how they are meant. Now it happens that I am privileged to know, by the author's own response to my claims, that John Bishop would dismiss my argument partly on the ground that his term Doxastic Venture is a technical one and as such is likely to mislead the novice and the at-large congregation, as it were. Under these circumstances it would seem that Bishop must believe his expression's status as an instance of specialized nomenclature confers immunity upon the thesis it covers, immunity from attack from such quarters as these, where our Happy Hour Group resides, and where its members expect to derive not only practical but philosophical pleasure from the expressions they encounter when those expressions are meant to convey information about what there is in the world. To such persons I say here that there is such a thing as belief in this world. But also I say that if there is such a thing in the world, analytical philosophers have a duty in telling us why and how such a phenomenon is possible, and where we can find its limits. So I take this moment to assure any sceptics that I may of course be wrong, and that I am always willing to be proved so, but that I assuredly do not misunderstand Bishop's programme. To that end I undertake this opportunity to discuss Bishop's ‘framework’ approach to venturing into transcendent belief, for this framework seems to be the space in which his venturing seems to consider itself safe from comparisons to my more mechanical accounts of believing. In this way might I clarify my argument to his satisfaction, and bring my reader one step closer to a position from which he may endorse it or not, without any confusion arising from questions as to whether I have taken the professor too literally, too generally, or under the handicap of a layperson's all-too credulous ear.

First let me repeat that it troubles me to imagine that there are two fundamental varieties of belief. For if there are, we live in constant danger of entertaining beliefs that fatally contradict one another, because the standards by which it becomes permissible to believe that P might differ from those under which it is permissible to believe that Q, yet we may find ourselves making faith-claims about both, and living under the assumption that Q and P are fundamentally compatible or even interdependent. Under the assumption that P and Q are both vital, this situation is untenable. Rational things that we are, we are constrained to recognize the limits logic will impose upon us as we undertake to describe reality. So belief in the existence of twenty dollars in our hero's pocket, and belief in the presence of God in Heaven must, in my view, be two examples of the exact same kind of psychological event, or else we are sunk. It is not, as some may wish to claim, merely a case for decision theorists when we speak about the terrestrial doxastic venture undertaken by group G. Belief is the business of the day, not what one can or may do or decide in virtue of the particulars of a state of belief. I take it for granted that when it comes to belief, decision theorists are in fact barred from examining the case at all: no one in our example has legitimately made a decision to believe; it is not a venture in that sense. It is a venture only in the sense that one undergoes some small travail with the expectation of realizing a reward – or at least arriving at some state that is predicted or hoped-for. One's venture, under this argument, is one's state of belief. One's claim is one's travail.

Examine the content of a particular belief, for example the belief you hold that your friend's checking account contains at least $250. Notwithstanding the obvious fact that you could demand proof, or that soon circumstances may lead you to discover whether your belief is warranted (say, when you try to cash her $250 check), we can agree that it is a belief. It is a belief that holds certain facts to be true in the world, and while available evidence may support the belief, it is not proved. So it is a claim against reality. It is not a matter of choice on your part, any more than other less personal facts are matters of choice for you: that the oceans are deep, that Mars is a long way off, that your house features copper plumbing and termites. These are all claims against reality, just as a personal cheque is a claim against funds supposed to be accessible from the bank, and though you may consider that at this point in life you know them all to be true, at some point in the past you held them as beliefs. All the foregoing examples rest upon frameworks of one kind or another – whether intuitive frameworks about the physical world, or socially agreed-upon ways of making guesses about the world you inhabit; they all stand as susceptible of being found warranted or not, and all the frameworks involved serve really only one purpose: to make conversations about them intelligible. A claim, a belief, or a framing principle all share this trait as they are discussed among philosophers of language: their truth is not at issue until their truth comes to bear upon claims about the real world. That is to say, any framing principle is valid from a merely logical standpoint. The truth of Christ's divinity is valid in the context of any discussion of the meaning of His crucifixion; however it is no longer valid in a discussion that requires for its legitimacy proof of said divinity. In terms of presuppositions or framing principles that truth then suddenly disappears. We need something further, something more fundamental in which to believe in order to continue that discussion.

In normal discourse we take much for granted merely so that we may have conversations that make sense. Everything that falls under the category of what we take for granted we call useful presuppositions – we might even openly call them useful fictions. We could just as easily and just as properly call them framing principles, and what is interesting about them in the context of this essay is this simple fact: we need not believe in them beyond their immediate logical utility. In fact we seldom bother to enquire whether or how deeply we believe in presuppositions, because their usefulness is in furthering discussion, not in describing reality. I am suggesting here then that Bishop is not misunderstood in the context of this article with regard to frameworks; instead, the concept of a framework of presuppositions is misunderstood or misused by Bishop in his own programme generally. In his haste to establish so-called framing principles as a bulwark against confusion with a fiducial model of faith, he has rushed to claim an unearned distinction. Bishop has in fact failed to distinguish doxastic venturing from the fiducial model of belief from which he seems sincerely to want to distance his thesis. So we are left, under Bishop's guidance, with the old-fashioned option of trusting the trustworthy. I say respectfully that this remains insufficient.

Whether or not it represents an instance of especial, technical nomenclature, I fail to understand how Bishop's concept of framing principles, and hence his model of a doxastic venture, differs in any substantial way from the garden-variety presuppositions to which we have everyday recourse in understanding how speech works in clarifying beliefs and predictions under whose frameworks we may have conversations about any number of things. I would hasten to reiterate here that such discussions need not of necessity take for granted the truth of the beliefs being discussed; they need only clarify the terms under which discussion is possible. A framework of presuppositions can be and often is arbitrary – any framework of presuppositions will do in order to give sense to a conversation or discussion or claim about belief (or anything else), and no party in the conversation need hold such a framework to be fundamentally true (although any party may do so) and even if understood as an outright fiction a framework may serve its function as foundational. I offer these examples: my stated belief that John is next to Mary presupposes that John and Mary exist, at least hypothetically, and that my listener knows where Mary is, and needs to know about John's spatial relationship to Mary and the world; but by the same token, my stated belief that Sherlock Holmes is next to Dr Watson depends upon and offers warrants for precisely the same kinds of presuppositions, and a conversation about that belief will be conducted along exactly similar lines as one about John and Mary. The fact that Holmes is a fiction and John is my co-worker has no bearing on the ways in which these conversations logically play out, or whether and how they make sense. It is only when I tell my listener where I believe Mr Holmes to be standing, and tell her to go to him now, and ask him if he might loan us twelve dollars for Happy Hour, that our presuppositions, as framework principles, begin to shake under the weight of possibly misguided or doomed ventures. And so it is when speech bears explicitly on the nature of reality that we are obliged to discard arbitrary frameworks in favour of what we take to be discreet, real objects in the world. Bishop's apparent admission that his ventures are not possible without a doxastic framework, in which framework principles (as, say, the divinity of the Christ) are held true without further argument, seems to suggest this arbitrary quality. If it is true that the Christ is or was the word of God incarnate, I may embark upon my venture into believing (but not knowing) what the Christian cosmos holds in store for me, or what it wishes me to do. But only if.

Likewise might I venture into some believing wager regarding a world in which leprechauns are on guard over pots of gold, if I take for granted the framing principle that the idea of leprechauns makes sense. In such a case I can have a conversation about what leprechauns may require of me as a believer, and what I may seek in them as objects of faith, and it will be a conversation conducted entirely upon reasonable grounds. It may even help suggest to my listener or myself what we ought to do or what we reasonably can be expected to do as moral agents in a leprechaun-based moral world. But when I turn to an earnest conversation about what there is in that world, I have to dismiss the leprechauns, regardless of how helpful they may be as the emblems of certain presupposed principles. So it stands with Christian presuppositions, or any other sort of transcendent ideas said to be believed in yet founded in untested principles. The guideposts of an earthly doxastic venture must necessarily be of a terrestrial nature. And because humans are of an earthly nature, transcendent ventures are, I say, unavailable to them.

Though it may be an unpopular revelation I feel bound by the rules of logic and language to hold to this primary tenet: that we are confined in our beliefs – this is a sad fact but a legitimate one. That we are confined in our beliefs I take to be an ancient dictum, one expressed in the quote from Maimonides that begins this article.Footnote ¹¹ It is the purpose of this essay to hold William James and Blaise Pascal and John Bishop all accountable for insinuating into philosophical discourse the mistaken idea that such confinement is irrelevant or weak or non-existent. It is real and I hold the strong view of its truth, though I would with patience and even admiration attend to the lecturer who devises a real argument that shows otherwise. For now, that lecturer being as yet not present at the podium, I say let a divine intelligence be guided by divine principles, but let a terrestrial intellect proceed on terrestrial assumptions.