1. Introduction

This paper seeks to evaluate the scope of the link between the evidential problem of evil¹ and the role of intuitions as a case of sui generis a priori epistemic access. In order to do this, I will first address the theoretical assumptions present in the understanding of what an intuition is and, consequently, what an intuitive belief as a case of non-inferential and a priori belief is. Following ^{Church et al (2020)}, one main point of this research would be to highlight the link between intuitions and evidentialism, as it is considered in the evidential problem of evil. This will allow a moderate and empirically informed rationalist analysis of what is understood as evidence regarding the possible (or probable, in some readings) non-existence of God, if it is the case that there is at least one instance of gratuitous evil (pointless evil) in the world². To strengthen this reading in favor of an a priori - a posteriori link as a way of responding to the argument of evil and its role of undercutting defeater (^{Oliveira, 2020}), two assumptions present in the discussion and their plausibility will be considered: i) that a priori reasons are not strictly understood as evidence for religious belief of a proposition p; ii) that the irrationality of religious belief arises from the absence of justification for the occurrence of an event of evil. I maintain that there is a tension between i) and ii), specifically regarding the absence of evidence that allows us to sustain the incompatibility between gratuitous evil and the non-existence of God. That is, it is about addressing the inferential limits of the available evidence, and about how the role of the absence of evidence is understood as counter evidence by default. An important aspect of this analysis is that the suspension of judgement, as an alternative in the absence of evidence, results in agnosticism and is an option that skeptical theism should discard.

On the other hand, a subsidiary objective here is to discuss the metaphysical and epistemic commitments of skeptical theism, in particular the metaphysical realism that discards the evidentialist alternative defended from certain positions framed in the so-called new atheism (^{Rowe 1979}, inter alia): absence of evidence and epistemic limitation of access to certain evidence, which are not coextensive phenomena.

The above stresses the skeptical theist position, since rules out suspension of judgment (or withholds belief) and requires that subject S (who sustains it) affirms that the evidence available does not correspond to the total evidence, despite not having the epistemic means to sustain this. More importantly, when religious beliefs are understood as justified from the evidentialist perspective, the skeptical theistic strategy assumes a commitment to the evidentialist position, understanding as justified and rational only those beliefs that fit the available evidence.

Following ^{Forrest (2017}), a way to discard this restriction would require to maintain that a religious belief is self-evident, which is problematic, among other reasons, because they are understood as a case of causal inference. Here the role of intuitive beliefs might help to solve this problem, without discarding the practical relevance of religious beliefs (reasons for action for example). In this sense, a probable link between intuition and evidence would help in solving this potential evidentialist demand of skeptical theism, although mainly with specific problems regarding the so-called golden triangle of necessity - apriority -meaning and its scope regarding conceivability as a metaphysical position (mere possibilia).

2. Natural evil and doxastic rationality

"According to contemporary atheists such as me...the evidence available today against God's existence vastly overwhelms the alleged evidence pro, if there can be evidence at all for such a thing. Therefore, well-informed and intellectually honest human beings should become atheists." (^{Philipse, 2019}, 191³)

This quotation is pronounced within the framework of a reflection on certain evidentialist objections to a theistic position⁴. Among these objections, the one referring to the problem of evil and its role in the Philipse makes explicit the example of the Spinozist conception of God in order to specify that, even when it could be understood as monotheistic, it does not coincide with the one enunciated above and constitutes a panentheistic proposal and, from there, heretical for the traditions made explicit. Philipse then cites the theistic understanding epistemic position plays a preponderant role. According to ^{Calder (2020)}, the concept of evil includes those states of affairs which are evil for various reasons (morally wrong actions, weakness of character). It can be noted that, this description of the notion of evil is included as part of the description of the evil events that are cases of evil. This raises the question of whether there is such a thing as evil events per se, or whether they are always so in view of how they affect the human being. This would constitute (as it will be stated below) an anthropocentric topic. Whatever the answer, in order to make a precise distinction between the atrocities caused by human beings and those that are not, a distinction is made between natural evil and moral evil. Usually, the greatest argumentative weight against the existence of God is given to cases of natural evil, in order to avoid pro-free will theodicies as a greater good. In this line of thought, ^{Calder (2020)} argues:

"Natural evils are bad states of affairs which do not result from the intentions or negligence of moral agents. Hurricanes and toothaches are examples of natural evils. By contrast, moral evils do result from the intentions or negligence of moral agents. Murder and lying are examples of moral evils."

^{Tooley (2019)} highlights the epistemic character of the problem of evil, which consists fundamentally in a questioning of the reasonableness of belief in God, understood in this case as the superior being who has what ^{Michon (2015)} calls omni attributes: omnipotence, omniscience and omniperfection. Generally speaking, it is the epistemic position of belief towards God that is the object of criticism, and it is usually identified as irrational. It is part, therefore, of a broader strategy from both atheistic and agnostic positions, in view of relieving the rationality of already suspending judgment or not believing that God exists. The problem of evil, for a theistic position, has at least two widely discussed strands, which attempt to answer different questions, such as these, following ^{Meister & Moser (2017}, 12): "Why has God permitted evil in a world that God created?" and "Is it reasonable to believe that God exists when there is so much evil in the world and, if it is reasonable, on what ground?"

In the framework of the discussion on the rationality of believing in the existence of God despite the existence of evil events in the world (especially those that seem unjustified or pointless), an edge that has not been sufficiently developed is the role of intuitions in the justification of both the pro and the con positions of the so-called evidential argument of evil against the existence of God. The role of intuitions in this and other debates is usually associated with the theoretical presuppositions on which the further discussion is based, for example, the definitions of central concepts such as God, evil, event, among others. It is correct to affirm that finding a certain discrepancy in the concepts used by the different positions in the debate leads to an initial disagreement as to what these terms mean and refer to. Regardless of this, the central focus of this reflection will be on what role intuitions play in the beliefs that justify a position for or against the evidential argument from evil. In particular, we wonder about the assumption that only empirical evidence is admissible, leaving aside, for example, certain a priori inputs, such as the role of intuitions in the beliefs of a certain subject S. What is complex, considering i) and ii) above, is to sustain the implication that absence of evidence is strictu sensu evidence against. If the restrictive status of evidence from an inductive perspective is empirical evidence, it is relevant to ask whether that restriction precludes holding, among other points, that the absence of evidence is empirical evidence. On this point, ^{Philipse (2019}, 192) argues:

"The evidence for or against a specific thesis or theory consists of empirically known facts that render it more or less probable that this thesis is true. If we want to make up our mind conscientiously about the (epistemic) probability that a specific factual claim is true, we should take into account all available evidence pro and contra. This requirement is often called the principle of total evidence. It may be that with regard to a specific claim the total evidence available is insufficient for concluding anything significant about the probability that the thesis is true. If so, further research is needed in order to discover new empirical evidence. In the meantime we should remain agnostic."

This quote emphasizes the factual character of an assertion, thus restricting the reasonableness of belief to what is probable on the basis of certain empirical evidence. Of course, this excludes the rationality of beliefs not founded on evidence and renders them irrational. Hence, it establishes a methodological and normative requirement as to what is to be believed and on what basis, without empirically justifying this normative-methodological presupposition. It is not a matter, as it might seem at first sight, of questioning the relevance of the evidence, but of evaluating its scope⁵.

It should be borne in mind that this debate is intertwined with one of a methodological-metaphysical nature, specifically referred to the naturalism-rationalism debate. Since it is understood that a naturalistic posture relieves the link with certain contributions from the sciences and that the evidential problem of evil would precisely rely on the available evidence that does not allow ruling out the existence of gratuitous or pointless events of evil, it is argued that this absence of evidence results in the irrationality ofa theistic posture.

^{Oliveira (2020}, 323) proposes on this point that this is observed in the inductive accumulation of instances without known justification: "(Instance) S's seeing no justifying reason for an instance of suffering is evidence for S that there is no justifying reason for at least some instances of suffering." The central focus is, therefore, on the epistemic accesses of certain subjects S. In turn, ^{Wright (2002)} affirmed the existence of a dilemma referred to what he called anthropocentric subject matters, among which the author included the moral, semantic and psychological realm. His conception of naturalism was expressed as "a stance which inflates the concepts and categories deployed by (finished) physical science into a metaphysics of the kind of thing the real world essentially and exhaustively is." (p. 401) This led to a commitment either to a reductionism (naturalistic in this case) or to a defense of a supernaturalism. Thus, the methodological edge consists in determining how best to approach the various domains of philosophical inquiry, the problem of (natural) evil in this case.

Naturalizing philosophical research implies bringing the theoretical and practical advances of philosophical analysis closer to successful methodologies in the empirical domain, and/or the other way around. In this case, it is what David ^{Papineau (2020)} frames as Methodological Naturalism. Some dissenting voices to this effort are collected in ^{Corradini et al (2006)}, whose approach, although it can be interpreted as anti-naturalistic, has as its main task to evaluate the plausibility of the contributions made without these contributions, rather than rejecting their relevance. From this standpoint, the aim is to avoid naturalistic reductionism, understood as the restriction of reasons and arguments to a naturalized perspective.

A question that guides this analysis is whether naturalism is a desirable theoretical approach in the framework of philosophical research, and whether by not opting for it, relevant information is lost for the understanding and eventual resolution of the evidential problem of evil, as in this case. It is desirable to reach and strengthen the link with the inputs that the various sciences provide to philosophical debates, without this forcing to restrict the discussions to positions based only on such evidence. This strains and questions the relevance of conceptual argumentative strategies that eventually do not consider in their premises and/ or conclusions empirical information (ruling out in this decision a case of deliberate ignorance).It is then a matter of highlighting the plausibility of a moderate rationalist proposal.

On this point, the criticism towards naturalism assumes a close nexus between this philosophical stance and the so-called scientism. Ted ^{Peters (2016)} has recently discussed the scope of a conceptual distinction between science and scientism, as well as its links with a skeptical stance regarding certain religious beliefs and their supposed groundlessness. The criterion of distinction between both notions is, however, logically inconsistent, since the definition of scientism defended by Peters includes in his definiendum the following concept of science: "a form of pseudo religion - actually an anti-religious religion- based upon a faith commitment: the only reality is material reality and the only source of knowledge is science" (223).

Understood in this way, the notion of scientism constitutes a metaphysical and epistemic commitment, insofar as it is affirmed that only the material is real, and that only this is knowable. It reflects, at the same time, an asymmetrical dependence of the epistemic on the metaphysical, assuming a realist stance: what a subject S knows or can know is determined by what is real, and not the reverse. While in other contexts this assumption might seem plausible, its use in the debate between theism and atheism places this metaphysical commitment as a way of exceeding the epistemic limits of scientific inquiry. This is further complicated by the fact that it seems initially correct to assign a naturalistic commitment (with metaphysical scope) to both the methods and the results of science(s). That said, from the very beginning of the discussion, intuitions play a central role. This is observed, following ^{Church et al (2020}, 7) in ^{Draper's (1989)} proposal and his hypothesis of indifference as an alternative naturalistic explanation, distinct from theism. They state the following:

"So what is the alternative to theism that Draper wants to propose? He proposes what he calls the Hypothesis of Indifference (hereafter HI): HI: neither the nature nor the condition of sentient beings on earth is the result of a benevolent or malevolent actions performed by non-human persons (^{Draper 1989}, 332). A hypothesis like HI is perfectly compatible with naturalism."

The hypothesis of indifference is not neutral. The stated indifference inclines the discussion towards an atheistic stance, without going deeper into the fact that the dismissal of such divine intervention is only one side of the discussion. Even if there is no such intervention, the central point is whether there can be one and whether there should be one, in order to prevent evil events in the present world: it is a modal, doxastic and deontic evaluation, which is not always reducible to empirical evidence, nor to implications such as HI, without being in itself an empirical hypothesis (at least not of a causal or directly testable nature).

Church et alia put forward this hypothesis as a better explanation of certain evil events, obviating that the central point under discussion is not about a conflict or tension between explanatory hypotheses, but rather about justified reasons that a subject could access to understand why A occurs, and not to explain why A occurs. It is evident that the conceptual dividing line here is very subtle, but trying to look for alternative justifications of an event that, strictly speaking, is sought to be sustained from an atheistic or agnostic position, has no ground, it is gratuitous. A causal explanation, for example, explains how an event occurs, but not why it occurs, and that is the focus of the debate between theism and atheism. The probability of the antecedent, understood in this case as the inferences that derive from the observation of evil in the world, is the point that Draper relieves in view of the probability of HI, of epistemic probability⁶.

3. Logical and inductive incompatibility: the relevance of the consistency assumption

At the beginning of his writing, ^{Oliveira (2020}, 319) identifies the various interpretations from which this problem can be approached. In each of them a central presupposition is to accept the premise that there is evil in the world⁷, and that point is important because it does not require evidence for its substantiation or, rather, it is assumed that such evidence is available in case one intends to question it. From there, at least two atheistic approaches emerge⁸, the logical and the evidential:

"According to the logical problem of evil, the mere existence of any evil whatsoever is logically incompatible with the existence of a perfect God. According to the evidential problem of evil, the existence of evil gives us sufficient evidence against the existence of such a God, despite being logically compatible with it."

The incompatibility or logical compatibility is a differentiating characteristic between both positions, both fulfilling the criterion that the existence of evil is a sufficient condition to reject the existence of God or, from its probability, the rationality of not believing in it. This is the central point of the initial reflection developed by ^{Mackie (1955)}, to show the contradiction and inconsistency between the propositions:

This is also the point of the analysis developed by ^{Hasker (1992)}, in what initially might seem to be an oxymoron rather than a counter-argument to the evidence of evil in the world: The necessity of gratuitous evil. It is not that every instance of gratuitous evil is necessary, but that it is conceptually feasible to sustain some compatibility between the two sides of the argument. He argues:

"That evil as such is inconsistent with God's existence is not much held anymore. It is evident that some evil could be a necessary condition for the existence of a greater good, and if this is so allowing the evil might be consistent with God's goodness. But evils which serve no such good purpose, gratuitous evils, are still widely thought to be inconsistent with theism and to provide the basis for a strong anti-theistic argument." (^{Hasker 1992}, 23)

The evidential argument is modified by restricting this incompatibility to specific cases of natural evil, and not necessarily by ruling it out. This point is not the standard interpretation of the evidential argument from evil¹⁰. Indeed, it stresses the potential consistency between evil events and the existence of God as an important distinction. For example, ^{Meister& Moser (2017}, 13):

"The evidential problem of evil suggests that, given the reality of evil, theism is probably not true, even if it is logically consistent. While there are various types of evidential arguments, the kind of reasoning employed is usually inductive. Such arguments also generally rely on actual cases of evil and suffering, sometimes described in graphic detail. A typical claim is that the existence of evil in its vast amounts and horrible forms provides reasonable evidence that the God of traditional theism (probably) does not exist."

It is not only the qualitative aspect of certain evil events, but their quantity, which in the standard perspective is what lends some weight to the argument. It is worth noting the reference to actual or real cases, which strains the potential generalization intended to be drawn from enumerated inductive cases. Oliveira emphasizes that the evidence must be sufficient and logically compatible in a case. It is important to note that logical compatibility or incompatibility is not directly established from any specific evidence; it can be argued that it is of an a priori nature.

An interesting, analogous case is observed in the reflection developed in ^{Morgan (1987)}. At the beginning of that paper, Morgan emphasizes that if, in order to win in a game such as golf, compliance with certain rules is required, then playing against the rules (cheating) is incompatible with winning in the game. In this example, winning and cheating are the incompatible concepts. In the case of the problem of evil, it is that the properties attributed to God would exclude the mere possibility of the existence of evil in the world. It should be noted, moreover, that this relation of incompatibility does not presuppose the existence of evil, nor that of God, but focuses on establishing potential logical relations with epistemic and metaphysical implications (e.g., if there is evil, God does not exist). Holding incompatible beliefs, in this framework, is tantamount to holding contradictory beliefs. Therefore, it is intuitively correct to hold that it is irrational to hold contradictory beliefs. The weight of the evidential argument rests on this a priori grounding link and is strengthened by holding that there is a posteriori evidence that tips the balance towards one of the two contradictory beliefs (that there is evil in the world, or in the case of the evidential argument, that there is an instance of gratuitous evil). This is referred to in psychology as cognitive dissonance¹¹, although the emphasis in this case is on the practical consequences that such tension generates, that a subject S believes A and acts against that belief (e.g., that S believes that smoking is harmful and yet smokes); to dismiss such a practical edge in the evidential argument from evil might be hasty. In any case, the analysis of the argument attempts to rule out both the rationalization of certain consequences that are intended to be avoided and the confirmation biast hat might condition the selection of certain evidence¹². It can be assumed that if a subject S holds contradictory beliefs, he does not meet certain basic standards of rationality. ^{Van Fraassen (2004},184) offers an example in which believing certain contradictions seems to be at least admissible, the paradigmatic case being Priest and his paraconsistent stance:

"What if I detect a straightforward contradiction in someone's beliefs, conclude that he has sabotaged himself in the management of his opinions, and he turns out to be Graham Priest? Priest happily admits to believing that certain contradictions are or may be true."

The evidential argument of evil, on the other hand, assumes as a necessary cognitive value that of coherence between the beliefs of a subject S, otherwise it would be self-sabotage. However, these are beliefs that imply some contradiction between them (in specific cases of evil events), and not directly contradictory beliefs (S believes A, S believes not A). That two beliefs oppose each other in this way implies incompatibility, albeit of a fallible nature. Other cognitive values relevant in decision making regarding what to rationally believe are: simplicity, fruitfulness, unity, inter alia (see ^{Kusch, 2020}, 33). The discussion as to which cognitive values should be prioritized is the focus of intense debate. Other criteria that are not necessarily opposed, may but eventually cause tension with those recently mentioned (especially coherence) are, following ^{Longino (2008)}, novelty, empirical adequacy, heterogeneity, decentralization of power, reciprocity of interaction, and inter alia¹³. According to Priest, ^{Kusch (2020}, 33) stresses that coherence can be gradual. This point seems to initially grant a certain advantage to the atheistic position, if one accepts that incompatibility is admissible. In that framework, the critique developed by ^{Weaver (2019)} regarding the logical objections that can be made to an atheistic stance allows one to make visible the initial tension that exists if one modally accepts even the possibility of logical Evil-God incompatibility. Weaver emphasizes that the logical evaluation of atheism¹⁴ requires determining which logical system allows to resolve conflicts or disagreements in the best way, both in the propositional realm (classical propositional logic or CPL) and in the modal realm¹⁵.

Thus, the central point to rescue is that affirming logical incompatibility supposes the commitment, in this case, with classical first-order logic (CFOL), since the incompatibility supposes that if an existential proposition about the existence of evil is true, the existential proposition about God must be false. From the analysis carried out by Weaver, this rules out the possibility of logical pluralism as a way of solution to the atheistic or real disagreement, fundamentally because in that case it is a matter of committing to a principle of tolerance a la^{Carnap (1935)}, which allows rather the dissolution of certain incompatibilities instead of their resolution. For their part, ^{Church et alia (2020)} point out that ^{Plantinga (1984)}, among other authors, opts for discarding the initial incompatibility¹⁶.

It is in this sense that the evidential argument from evil is understood as weaker, or more modest, than the incompatibilist position. It is curious that it is not sufficiently analyzed that, regardless of the scope, the incompatibility between certain events of evil and the existence of God persists in the evidential argument. Oliveira, following ^{Rowe's (1979)} reflection, emphasizes that the central focus of discussion is a particular type of evil, the so-called pointless suffering. The strength of this argument, as has been widely discussed in the last twenty years, lies in the intuitive lack of justification for the occurrence of any event of evil. We call it intuitive in the sense that it seems plausible even without prior substantiation and evaluation of reasons that there is such an absence of justification. This is the starting point of the strategies framed in the so-called skeptical theism. As Oliveira points out, the epistemic focus of the argument is on the rationality or, more precisely, the reasonableness of belief in God¹⁷, this being the conclusion of the argument developed by Rowe. It is relevant to highlight that, in the way Rowe enunciates the argument (and which is reconstructed on the basis of its subsequent improvements by Oliveira), it includes the non-existence of God in one of its premises, specifically in premise 4:

The Evidential Argument from Evil (EAE)

P1. After careful reflection on our experience and knowledge, we see no justifying reason for many instances of suffering in the world.

P2. If, after careful reflection on our experience and knowledge, we see no justifying reason for many instances of suffering in the world, then it is reasonable to believe that there is no justifying reason for at least some of all that suffering.

P3. So, it is reasonable to believe that there is no justifying reason for at least some of the suffering in the world.

P4. If there is even one instance of suffering for which there is no justifying reason, then God does not exist.

C. So it is reasonable to believe that God does not exist (^{Oliveira 2020}, 319 -320)

Premise 4 thus establishes the link between an argument that is fundamentally epistemic (what is reasonable to believe considering the available evidence) and its metaphysical consequences (the non-existence of God)¹⁸. This is developed by the argument via a priori, that the role of certain evidence is found in the antecedent of the conditional (if there is even one instance of suffering for which there is no justifying reason), and even that antecedent does not rely solely on empirical evidence (an instance of suffering), but rather relays the implications regarding the absence of justification. The tension between what is and what is not evidence has implications also with respect to the role of testimony as a type of admissible evidence. If, in order to determine the soundness of the evidential argument from evil, it is established that the prior beliefs of a subject S are not relevant, it is complex to explain how certain views that are at the basis of the argument should not be dismissed as gratuitous and unfounded. On this point, Benton (2018, 186) argues:

Benton follows the same line as Priest in arguing that:

"If Uniqueness is false, there is room for thinking that there could be reasonable disagreements, even among peers who have the same evidence. Denying Uniqueness amounts to endorsing a permissive view of what rationality requires, such that one's evidence may rationally support more than one doxastic position."

In the case of the evidential argument from evil, what is under discussion is which doxastic position is reasonable, while assuming that both cannot be. They could not, because of the metaphysical consequence stated in premise 4¹⁹. Without this premise, the discussion regarding the reasonableness of the atheistic position does not lead us to conclude (even if one admits the possibility of rational disagreement) that God does not exist as the conclusion of that doxastic tension. By relying on prior beliefs, on diverse intuitions that are not subjected to direct scrutiny but constitute the presuppositions of the argument, it follows that the evidential argument from evil cannot reach the metaphysical conclusion that it seeks as its ultimate goal and, thus, that it is rational to raise some remarks from the available evidence. This requires, first of all, to analyze what evidence is admissible and what evidence is excluded.

The cases considered by ^{Oliveira (2020}, 320) as examples of justifying reasons understood as "some consideration in favor of allowing the relevant instance of suffering that would mitigate its existence" are enlightening in two central aspects: i) there are certain instances of suffering that are allowed, which implies that they might not be²⁰; ii) the justifications are external to the evaluated event: they are not extracted from the event of evil in isolation but in correlation or causal relationship with other events or other potential events that are either prevented or promoted. It happens this way, for example, with the vaccinated child that Oliveira makes explicit, his central point of criticism being that "EAE simply exploits the common impression that we are unable to produce some such justifying reason for so much of the suffering that we see around us." Theistic skepticism rejects this presupposition and considers it unlikely that human beings can access such justifications, even if they exist. The point is, therefore, about how certain cognitive capacities allow a subject S, in the absence of access to certain evidence, to hold that there is no evidence simpliciter: "...if we cannot expect to see God's justifying reasons in the first place, how could not seeing them provide us with evidence against God?" (^{Oliveira, 2020}, 320)

It is interesting to stress how the evidential argument from evil assumes that there is only evidence against the existence of God, and not in favor. This aspect is recently analyzed by ^{De Cruz (2019)}, so that the idea that atheism is more plausible than theism is not accepted as intuitively correct²¹. Regardless of this, the central focus of Oliveira's analysis is the inductive weight that the accumulation of instances has as a way to weaken a position and ultimately make it irrational to maintain that belief: this is the modus operandi of the so-called undercutting defeater. Now, in the discussion on the evidential argument from evil, what is observed are usually types of cases, hypothetical cases rather than real and/or actual cases stricto sensu. That the use of the evidential argument is made via hypothetical cases indicates, at least methodologically, a close link between what is assumed as evidence (paradigmatic cases of gratuitous evil), without the need to resort directly to such evidence: they are intuitive cases in the sense that they are assumed to appear to be true cases of gratuitous evil. Usually, this aspect of the evidential argument is not emphasized, among other reasons, because its consequences regarding the rationality of a certain belief would not vary if it is made explicit that, strictly speaking, evidence is not being gathered through real cases regarding the inductive weight of the evidential argument: the hypothetical emblematic cases are not, in this sense, empirical but conceptual evidence, a priori, and their intuitive character resides in the role of assumptions that they play in the argumentation.

The above, following the analysis proposed by ^{De Cruz (2019)}, brings into discussion whether the burden of proof of the evidential argument resides only in the theistic stance or whether empirical evidence and not only counterfactual evaluation is required in the evidential argument from evil²². It is important to consider the probabilistic observation made by Plantinga about the evidential argument, which boils down to how likely is it that a finite entity such as a human being can access and evaluate the reasons or absence of reasons for allowing an event of evil. Even the methodological basis of the argument may be subject to such criticism²³.

4. Skeptical theism and intuitively pointless evil: the relevance of what seems true

^{Oliveira (2020}, 324) understands ST as follows:

However, this is not the standard option of a skeptical theistic stance, which aims to rule out rational justification pro atheism. This is picked up in ^{Church et alia (2020}, 5) by stating that skeptical theism requires the pointless concept to establish the epistemic demand to justify how a subject S knows that a certain event is not of evil simpliciter, but of gratuitous evil. What can be asserted is only that it seems to be the case, that it is intuitively pointless: "Let's grant that evils like E1 seem pointless, and that that seems to suggest that they are indeed likely pointless (which, of course, is a point many skeptical theists would want to deny)."

Oliveira observes that this is the negation of what is stated in (Instance), the center of the debate being whether the absence of evidence is evidence against, which is what the undercutting defeater suggests. It is important to note two aspects that bring theistic skepticism closer to the same problem that occurs to the evidential argument as will be seen: it focuses on what appears true to a subject S. It explicitly enunciates "seeing no justifying reason", thus restricting the scope of the claim to the epistemic limits of the one who does not see those reasons, and how for those subjects it is or is not evidence for something. Indeed, this is what Oliveira calls Limitation: "We cannot expect to see a justifying reason for instances of apparently pointless suffering, were such reasons to be there."

Following ^{Wykstra's hypothetical case (1996}, 129-30), Oliveira stresses that if epistemic access limitations prevent affirming A (in this case the presence of a virus in a vaccine), it is epistemically unreliable to affirm that there is no virus, if one does not have the epistemic accesses or the tools that allow one to substantiate it. This corresponds to the now classic argumentative path called the Nosseum argument. It is, therefore, intellectual opinions and different rational intuitions that guide the positions under discussion. The problem is how to establish which intuitions are correct and which are not. A way of response that could be unsatisfactory from the outset but adjusted to the contexts of the cognizing subjects is that, in their respective perspectives, both are justified in sustaining both the absence of justification and the suspension of judgment. In other words, Skeptical Theism does not affirm that there is justification, because another important assumption is that in that case, according to evidentialism, it should already offer reasons or evidences that support that position: it would not be enough that this is what seems to a subject S. This reflects a common point for both: ignorance and limits of certain knowledge as initial conditions of the problem statement and its search for resolution, assuming that only sustaining the equitable validity of any position in view of these limitations is not the best option: that would correspond to an abuse of the value of epistemic humility.

Conclusion

The evidential problem of evil can be understood as a specific case of epistemic disagreement, which ^{Benton (2020)} calls religious disagreement. Phrased in this way, it is in principle troublesome that two of the three disputing parties (atheist and agnostic) could discuss whether the denomination of religious in this disagreement is correct, if this presupposes an essential characteristic of this case, if it requires, in other words, a common ground characterized as religious, in a loose sense. Benton establishes a relationship between disagreement and diversity, and this diversity, in turn, has its correlate in the variation of intuitions that guide certain beliefs. He argues that "Contributing to this diversity are the many non-religious or irreligious, particularly atheists, who think that nearly all positive claims (at least about the existence of the supernatural) are false." (^{Benton 2020}, 188)

Understanding disagreement commits atheism to admitting that propositions about the supernatural have meaning and truth-value, a requirement for holding that those that make positive claims about, for example, the existence of God, are false. Leaving aside that this admissibility of an a priori truth- value, not evidentially, supposes an absence of probabilistic evaluation such as that present in the problem of evil, it should be said that what is understood by evidence is still a matter of discussion, as well as whether or not the disagreement is finally based on evidence. This is why, although there are religious diversities (generally between religions), it is not entirely appropriate to call the epistemic disagreement between non-religious positions and their affirmative counterpart in this way. Benton emphasizes that a philosophical argument attempts to stand out for its neutrality, and thus an evidential argument such as the argument from evil suggests such neutrality, by deriving some (partial, not total) evidence for the more rational epistemic position. The testimony, the subjective religious experiences, remain in an unfavorable position in this scenario, lacking this apparent neutrality. It is apparent, as has been discussed, since it is based on what seems true to a subject S. Aligned with what Church et alia maintain about the intuitions of experts, Benton states:"... the epistemic force of that testimony may depend on whether the testifier functions as an authority or as a kind of expert advisor." (^{Benton 2020}, 189) Although at this point he does not refer to intuitions but to different grounds, in both cases some validity is assumed prior to a reflective evaluation of those initial statements:

"...the existence of many who disagree on the basis of different grounds should, if acknowledged, induce one to reconsider. For acknowledging such disagreements involves learning that others have different grounds which one might find compelling, if only one would give them an honest hearing." (ibid.)

The role of intuitions here is not reduced to what Benton describes as pre-reflexive, for that would preclude any progress in the further deliberation they make possible, if such progress in essential disagreements is possible at all. At least, it is so in the weak sense that the disputing parties recognize their epistemic limitations, which preclude prima facie holding epistemic certainty about what S believes in each case: It is expected from this analysis that, at least on the epistemic edge, the parties recognize the possibility of error, including the evaluation of the available evidence and the metaphysical implications³¹.

In assessing the epistemic validity of intuitions and their variability across subjects and cultures, ^{Church et alia (2020}, 8) highlight how the outcome of testing these hypothetical cases of gratuitous evil depends on how they are stated, the role played by emotions in the reaction to them, among other factors that (strictly speaking) are not truth sensitive, do not determine whether what S believes is true or false, but rather explain why S believes p:

"...many accounts of the evidential problem of evil proceed by first placing the reader in an affective state by describing an example of suffering and evil with sufficient detail. (Rowe's example of a fawn's-not even an adult deer but a baby deer!-tragic death is a good example of this.)."

In considering the diversity of intuitions and elucidating what is assumed to constitute the common ground, it should be emphasized that lack of variability does not mean that the intuitions in question are some true and others false. The central point there is what aspect of certain arguments, in this case the evidential argument, considers among its evidences the intuitions that support such argument. It is in this sense that the incompatibility or rationality of certain beliefs cannot be measured only from the available evidence, since it is interpreted from certain presuppositions that, being different leads to different conclusions and implications. This shows that, independently of the epistemic merit that the evidential argument of evil may have by revealing, for example, eventual central disagreements, it is not possible to draw metaphysical conclusions that exceed its scope, unless this is another presupposition that is intended to be discarded and that, therefore, metaphysical affirmations supervene to the doxastic states of certain subjects. That would have the strange consequence that God exists for those who believe and does not exist for those who do not. But, if that were the scope of the discussion regarding the role of intuitions in the evidential argument from evil, it is important to note that the argument is not relevant, and that what is obtained as a result is an epistemic pluralism, supported by the incommensurability of opposing reasons and constituting the basis (again intuitive, but of second order) of a meta paradigm that frames epistemic disagreements as a case of diversity and, therefore, valid per se. This is probably the main methodological problem of pluralist proposals, insofar as it does not satisfy the parties involved in the debate, but this is perhaps so precisely because of what has been analyzed: there is no common ground that allows for a fruitful dialogue. This is clearly expressed by ^{Kerr (2019}, 199):

"Since around the start of the 21st century, philosophers and psychologists have been exploring the possibility that many of the assumptions that form the agreed basis of current philosophical and psychological debates might in fact not be generally accepted amongst different populations. Intuitions, previously thought to be beyond empirical investigation, may in fact be a source of disagreement."

Oliveira anticipates the negative consequences of sustaining skeptical theism, since it leads to questioning the scope of knowledge via testimony, or to having to admit, following ^{Hudson (2014)}, that God can even deceive in view of a greater good, if that is the case. That detracts from the discussion of gratuitous evil events, implying that every evil event is thus justified by default, which entails a conclusion that overgeneralizes the scope of the evidential argument and that, as absurd as it may seem, ends up legitimizing an understanding of God that skeptical theism tries to discard. In this sense, it is worth asking whether the conditions required for a skeptical theist are compatible with a Christian theist, which ^{Oliveira (2020}, 332) assumes is so in view of avoiding the hopeless fallibilism that seems to follow from skeptical theism, thus preventing a reliable epistemic access to God as a cost of the rejection of the evidential argument: that cost seems too high if the consequences of the argument can be avoided from another perspective and the questioning of the intuitive basis of the argument are a way to at least demand suspension of judgment regarding the evidential argument from evil. The burden of proof does not fall exclusively on the theistic stance, for it cannot be assumed by default that the atheist is more rational.

In closing, it is worth recalling what Haack (2009, 31) argues, in that there is a central difference between what most people think (that it is morally better in their example), and that what they think is the case (that it is actually morally better). Something similar happens with intuitive judgments, therefore, care must be taken that the search for reasons to support a point does not result in an omission of other equally valid reasons, only in order to maintain a position. For Haack, this results in scientism and is a risk run by the evidential argument from evil.