Rowe’s Argument from Evil – A Case of Illegitimate Noseeum Assumptions: Part I

By Fahim Faruk (Guest Contributor) 

Original Submission: PP209 (course code: Philosophy of Religion), Wilfrid Laurier University 

Throughout the history of the philosophy of religion, a plethora of arguments in natural theology and atheology have been developed. However, relatively few arguments enjoy the sort of universal recognition that the forceful problem of evil has attained especially in Western philosophy. In Western philosophy, various forms of the argument from evil, directed chiefly against Abrahamic conceptions of God, have been seriously entertained by both theistic and non-theistic philosophers. These theistic philosophers include those within the Abrahamic religious traditions of Judaism, Christianity, and Islam. Although, depending on creedal and doctrinal specifications of each religious tradition, the argument’s form, effectiveness, or relevance may vary, it nonetheless bears significance and has laid a challenge to the foundations of theism. One particular elucidation of the argument from evil is presented by philosopher William L. Rowe in his essay Evil Is Evidence against Theistic Belief. Note that Rowe emphasizes that his argument is directed against restricted theism, which is simply the affirmation of the existence of an all-powerful, all-knowing, and perfectly good God, without entailing specific doctrinal beliefs of the Abrahamic traditions. The argument is deductive in nature, and bears two premises and a conclusion:

1 There exist horrendous evils that an all-powerful, all-knowing, perfectly good being would have no justifying reason to permit.
2 An all-powerful, all-knowing, perfectly good being would not permit an evil unless he had a justifying reason to permit it.
3 God does not exist (Rowe, “Evil Is Evidence against Theistic belief,” p. 184).

The argument has received sound criticism from two prominent theistic critics, Daniel Howard-Snyder and Michael Bergmann, by means of identifying Rowe’s reliance upon an illegitimate noseeum assumption. However, it is first essential that Rowe’s argument is put into clearer perspective so as to set the stage for such criticism; a significant distinction must be made between the logical and inductive argument from evil.

The logical argument from evil posits a contradiction between the very existence of evil and attributes of God such as omnipotence, omniscience, and perfect goodness, arguing that therefore God does not exist. In contrast, the inductive argument from evil instead places its focus on the “claim that evil constitutes (sufficient) empirical evidence against the existence of God” (Alston, The Inductive Argument from Evil and the Human Cognitive Condition, 97). Rowe presumably recognizes issues with the logical argument from evil, and therefore posits a version of the inductive argument from evil. He relies heavily upon premise one of his argument, which refers to the presence of particular horrendous evils. Moreover, he emphasizes that it is the “enormous amount of apparently pointless, horrendous suffering occurring daily in our world that grounds the claim in the first premise” (Rowe, “Evil Is Evidence against Theistic belief,” p. 185). Thus, the focus is exerted on the level and magnitude of evils in the world as an empirical fact. Rowe asserts that at least some of these instances of intense suffering have no justified reason behind them. Yet, premise two of his argument requires that God have a justified reason for every evil, and even the presence of just one instance of gratuitous evil presents a problem. Clearly, Rowe recognizes that the conclusion of his argument is probabilistic in nature, and rests upon the veracity of the first premise.

The framework of his case begins by neglecting any other arguments for or against the existence of God, so as to focus primarily on the impact of his argument from evil on the probability of God’s existence. For this reason, Rowe alleges that in the absence of a positive case, God’s existence cannot “reasonably be assigned any probability beyond 0.5 – where 1 represents God’s existence as certain and 0 represents certainty that God does not exist” (Rowe, “Evil Is Evidence against Theistic belief,” p. 183). His thesis, then, asserts that upon this framework the evils that occur in the world render atheistic belief more plausible than theistic belief. That is to say, they bring the probability of God’s existence downward from 0.5 towards 0. As such, Rowe explicitly clarifies that his argument is not intended to be a proof for the non-existence of God. Proofs, he explains, require that the premises of a deductive argument be known with complete certainty and therefore lead to a sound conclusion. His argument, however, is a deductive argument whose premises are not known with complete certainty, but are alleged to be more plausible than their antitheses.

Rowe’s second premise presumes that God would not permit an evil without sufficient justifying reasons—a premise many theists accept. Thus, he focuses his substantiation on premise one. Rowe emphasizes that only two options seemingly remain explaining why God could permit any evil phenomena: God’s justifying reason for permitting [evil phenomena] would have to include…some out-weighing good that, all things considered, he wishes to realize and cannot realize without permitting that evil, or some equal or worse evil that, all things considered, he wishes to prevent and cannot prevent without permitting that evil (Rowe, “Evil Is Evidence against Theistic belief,” p. 184).

Further, Rowe goes on to provide two illustrations which seek to establish a personal impact so as to emphasize his point; that of an innocent little girl raped and beaten to death, and a fawn left to gradually die on its own after suffering from devastating wounds. Evil, argues Rowe, is gratuitous in the world and numerous evil acts appear to be utterly purposeless, devoid of greater goods unattainable without such evils, and avoidable through a different arrangement of affairs in the case of an omnipotent, omniscient and perfectly good God. Additionally, he emphasizes that although this cannot be known with certainty, it is rationally justified and so too is the conclusion of his argument. After all, argues Rowe, if there exists justified reasons in the knowledge of God for permitting such horrific evils, surely humans would be able to conceive of at least some of them, even if they are yet to or ever will manifest. Hence, criticism by Daniel Howard-Snyder and Michael Bergmann, has been dealt chiefly to the first premise of Rowe’s argument from evil.

Moreover, such criticism is built upon the framework of rejecting Rowe’s argument from evil as a good argument, as it fails to meet a minimal standard: “every premise, inference, and assumption on which the argument depends must be more reasonable for us to affirm than to refrain from affirming” (Howard-Synder and Bergmann, “Evil Does Not make Atheism More Reasonable than Theism,” p. 193). They level the charge of illegitimate noseeum assumptions against Rowe’s noseeum argument, which is contrary to his assertion that he is not in any form arguing from ignorance.

Noseeum arguments are those rooted in what Howard-Synder and Bergmann have dubbed the noseeum assumption. Such a premise renders its conclusion more likely than not on the condition that more likely than not the item in question would be detectable if existent. This principle explicates why Rowe’s argument fails to meet the minimal standard set by his detractors. Thus, with respect to the established minimal standard, a good noseeum argument must be comprised of noseeum assumptions that are more reasonable to affirm than to refrain from affirming. Here, it is helpful to consider examples of legitimate and illegitimate noseeum assumptions. Suppose, for instance, a Muslim scholar having memorized and repeated the entire Qur’an numerously since childhood and under a plethora of teachers, is met with the false claim that within the Qur’an there exists a particular collection of verses in chapter two.

The implication then being that the scholar has allegedly missed these particular verses during each and every single one of his examinations and recitations throughout his life, in every single copy of the Qur’an accessed and under various teachers all having also memorized the Qur’an. Clearly, here the scholar and his teachers are completely justified in asserting that if any such collection of verses existed in the second chapter of the Qur’an, they would have come across them, and due to the fact that they have not—such a collection does not exist. This is an example of a legitimate noseeum assumption, because the scholars are sufficiently equipped to discern the matter at hand, especially with respect to their consensus. However, consider the example of a child just beginning his Islamic studies with the memorization of the Qur’an alongside classmates of similar capacity and knowledge. If this child was met with the same false claim, dismissing it as erroneous based on his initial studies and the consensus of his classmates would constitute an illegitimate noseeum assumption. This is due to the fact that they are unequipped to discern the matter at hand. In their case, it is not more likely than unlikely that such a collection of verses does not exist in the second chapter of the Qur’an, whereas in the former case it is.

Rowe, however, attempts to establish that he is not arguing from ignorance, and is justified in presuming that more likely than not, if there actually existed justifiable reasons; at least some of them could be conceived of. Further, he includes God’s comforting reassurance as a form of justification akin to a parent reassuring their child unaware of possible goods in hardships. In sharp contrast, Howard-Synder and Bergmann soundly argue that Rowe relies on illegitimate noseeum assumptions. Firstly, there is no reasonable basis upon which Rowe can posit that there are no goods beyond human comprehension in the knowledge of God, that justify evils in the world, as he has not sufficiently demonstrated that humans are equipped to discover such goods to begin with.

Clearly, to assume so commits the same fallacy elucidated in the example of an illegitimate noseeum assumption, provided above. Just as the child is unequipped to affirm the falsity of the false claim that there exist particular verses in the second chapter of the Qur’an, so too is Rowe unequipped to assess the likelihood that humans could conceive of evil-justifying goods in the knowledge of God, if they existed. Howard-Synder and Bergmann provide two cogent points of consideration against Rowe’s noseeum assumption, demonstrating its fallacious nature. The first point of consideration explicates the notion that finite and fallible humans attain insights representative of the sort of reasons an omniscient and omnipotent God may possess. Disputes are ripe within virtually all branches of science, and in many within philosophy. For instance, quantum mechanics yields multiple conflicting interpretations. Moreover, concepts relevant to the world, such as causality, are continuing to develop as humans acquire further information.

These observations indicate that the human mind is far from even approaching mastery of the world, let alone access into the transcendental omniscience of God. In fact, Howard-Snyder and Bergmann allude to this realization in their second consideration where it is argued that “it would not be surprising at all that humans discovered various fundamental goods over tens of thousands of years separated by several millennia-long gaps in which nothing was discovered” (Howard-Synder and Bergmann, “Evil Does Not make Atheism More Reasonable than Theism,” p. 198).

Secondly, Rowe’s illegitimate noseeum assumption extends to claims regarding God’s nature and relationship with evil, and value judgments. Consider that Rowe’s argument presumes that his understanding of the logical connections between actualizing particular goods and permitting certain evils is representative of reality. For instance, he claims that “the idea that none of those instances of suffering…quite beyond belief” (Howard-Synder and Bergmann, “Reply to Rowe,” p. 206). In addition, Rowe argues that unlike the situation of a good parent reassuring their child, wholly unaware of possible goods within a difficult trial, God remains silent and offers no such assurance. Here, he argues that this casts doubt on the existence of God. However, as Howard-Synder and Bergmann argue, there is no reasonable basis upon which he can posit that there exist no goods beyond our ken justifying God’s apparent silence. As has been illustrated, suggesting otherwise amounts to an illegitimate noseeum argument, which is clearly pointed out by Howard-Synder and Bergmann.

Rowe’s methodology is also fallacious in that he neglects doctrinal specifications which may shed light on the nature, and permission of evil in the world by God. For example, there exist reasons within particular theistic traditions reconciling God’s apparent silence in the face of evils. Such reasons have been popularized as theodicies, only a few of which are attempted to be addressed by Rowe. However, his assessment of the weight of possible evil-justifying goods in relation to those evils is entirely subjective and therefore limited, which is rightly pointed out by Howard-Synder and Bergmann. As such, Rowe’s evaluation of various theodicies loses credibility; he also fails to assess them in light of the larger framework of the doctrines and various conceptions of God they fit into. Thus, Rowe’s consideration of what is entailed by God’s omniscience, omnipotence, and perfect goodness is inherently limited.

Overall, the criticisms raised by Howard-Synder and Bergmann successfully undermine Rowe’s first premise. Rowe argues that by means of evaluating goods we know of as failing to justify certain evils, being unable to even conceive of evil-justifying reasons that may exist in God’s knowledge, drawing connections based on what he believes is entailed by God’s omnipotence, omniscience and perfect goodness, and finally the apparent lack of success of certain theodicies—he builds a successful case demonstrating that aside for positive reasons for God’s existence, his argument from evil lowers the probability of God’s existence. In response, however, Howard-Synder and Bergmann accuse him of resting his case upon illegitimate noseeum assumptions. A close examination of the entire exchange reveals that Rowe fails to address his reliance on illegitimate noseeum assumptions and therefore presumably modifies his argument by incorporating a new Bayesian argument.

As such, the charge of relying on illegitimate noseeum assumptions against his former argument goes unchallenged. In their brief exchange, Howard-Synder and Bergmann overlook one core point raised by Rowe, however. Rowe argues that as per their case, it would be unjustified to affirm premise one even if life was but a consistent series of animal and human suffering. That is to say, given any possible arrangement of evil in the world, gratuitous or otherwise, it is possible to object that affirming premise one amounts to an illegitimate noseeum assumption. However, this simply seems absurd in the mind of Rowe. He feels as if the human intellect ought to be justified in inferring things about the likelihood of God’s existence given at least some experiences or realities. Whether or not this is true is unclear. Although it may be emotionally unappealing, it may nonetheless remain true that even in such an arrangement, affirming premise one relies on an illegitimate noseeum assumption. Such a realization does not undermine the oppositions’ case. Though, since Howard-Synder and Bergmann offer no explicit reply, their position on this matter remains unclear.

In conclusion, Howard-Synder and Bergmann provide a weighty refutation of Rowe’s inductive argument from evil. They argue that he relies upon illegitimate noseeum assumptions, and therefore remains unjustified in affirming premise one—that God has no justifiable reason in permitting the horrendous evils of the world. Such illegitimate noseeum assumptions extend to ignorance of the weight of goods we can conceive of in relation to the weight of evils, as it presumes that we are in any such position to measure such weights. Further, they also extend to our understanding of the omnipotence-constraining connections that may exist and whether these are actually representative of those that exist. Rowe, however, maintains that his limited human conception of the world, what is entailed by the nature of God, subjective assessment of the weight of goods and evils, and the logical connections between the obtaining of goods and the permission of evils in relation to God’s nature are sufficient to conclude the unlikelihood of God’s existence. It has been thoroughly demonstrated, however, that Howard-Synder and Bergmann’s contentions undermine such fallacious reasoning.

Only one of Rowe’s core points goes unaddressed explicitly; the charge of illegitimate noseeum assumptions may be raised against any argument from evil pointing to any arrangement of evil in the world. This remains the only issue on which no light is directly shed by Howard-Synder and Bergmann, at least in their brief exchange. Ultimately, just as atheists contend that positive arguments for the existence of God fail and therefore do not raise the probability of God’s existence, it is clear that Rowe’s argument against the existence of God fails to meet the minimal standard of a good argument set forth, and therefore does not lower the probability of God’s existence from a neutral o.5.

In part II, I shall examine how Orthodox Sunni thought frames and addresses this particular adaptation of the Argument from Evil, and I will contrast this with the methodology taken above by Daniel Howard-Snyder and Michael Bergmann.

May Allah سبحانه و تعالى guide us to sound understanding and submission to His Divine Will, and increase us in our love for His Beloved Muhammad صلى الله عليه وعلى آله وسلم. Ameen!


  1. Howard-Synder, Daniel, and Michael Bergmann. “Evil Does Not Make Atheism More Reasonable than Theism.” Trans. Array PP/RE 209 Online Learning Readings. Wilfrid Laurie University, 2013. 193, 198. Print.
  2. Howard-Synder, Daniel. The Evidential Argument from Evil. 1st ed. Bloomington: Indiana University Press, 1996. 97. Print.
  3. Rowe, William. “Evil Is Evidence against Theistic Belief.” Trans. PP/RE 209 Online Learning Readings. Wilfrid Laurier University, 2013. 183-185. Print.

Leave a Reply

%d bloggers like this: