Next Article in Journal
Political Economy as a Test of Modern Judaism
Next Article in Special Issue
Jewish Diaspora and the Stakes of Nationalism: Margarete Susman’s Theodicy
Previous Article in Journal
Gender and Space during Guatemala’s Holy Week: An Ethnographic Account
Previous Article in Special Issue
Job and the Bible’s Theo-Political Divide

The Book of Job as a Thought Experiment: On Science, Religion, and Literature

Institute for the History and Philosophy of Science and Technology, University of Toronto, Toronto, ON M5S 1K7, Canada
Religions 2019, 10(2), 77;
Received: 17 December 2018 / Revised: 16 January 2019 / Accepted: 18 January 2019 / Published: 24 January 2019


This paper presents a philosophical critique of the proposal that the Book of Job is a theological thought experiment about divine providence. Eight possible objections are entertained. They guide the discussion of the proposal. It is concluded that the proposal has more merits than perils.
Keywords: metaphilosophy; metatheology; Biblical hermeneutics; evil; literary cognitivism; fiction; revelation metaphilosophy; metatheology; Biblical hermeneutics; evil; literary cognitivism; fiction; revelation
This paper is not an exercise in Biblical Studies. And yet, its point of departure is a particular view on the literary genre of the Book of Job. Nevertheless, it does not assume a historical–critical and/or a biblical–theological position in relation to that view. Instead, what follows is a philosophical critique of the proposal that the Book of Job is a thought experiment: “The scene in heaven, like the speeches of Job’s friends, is part of the literary design of a thought experiment to generate discussion about how God runs the cosmos.” ( The story is a thought experiment. Its aim is to facilitate a discussion. Its topic is divine providence. The Book of Job is, therefore, a theological thought experiment. That is the proposal, and in what follows, I am submitting it to a careful critique.

1. The Implausibility of the Proposal

It seems utterly implausible to think of the Book of Job as a theological thought experiment. The following objections come to mind readily:
First, the genre of thought experiments did not come into existence before the period of modern science (see Kühne 2005, pp. 89–90). That was a long time after the Book of Job was canonized. The proposal seems, therefore, to suffer from an unfortunate anachronism.
Second, the Book of Job may convey a thought experiment. But it is not one. Thought experiments and their conveying medium should not be conflated (see Gooding 1992; Nersessian 1992, p. 292; 2007, pp. 143–44). The proposal seems, therefore, imprecise in an important respect.
Third, the Book of Job presents a much longer narrative than what is typical of thought experiments. Thought experiments are brief and to the point. This is an essential characteristic of thought experiments (see Sorensen 1992, p. 224). Thus, the proposal seems highly inaccurate.
Fourth, science and religion are absolutely independent realms. Thought experiments belong to the scientific realm because only here can we make meaningful cognitive use of them (see Bealer 1998, pp. 207–8; Thagard 2014). We are blurring the demarcation line between science and religion when we read a religious text as a thought experiment. This is dangerous for religion in particular, because it raises expectations as to the cognitive nature of Scripture that must ultimately lead to disappointment. Hence, the proposal has unfortunate implications for the relationship between science and religion.
Fifth, the cognitive efficacy of thought experiments stems from the propositional lines of reasoning that are hidden in them (see Norton 1996). And those are only as good as the empirical validity of the premises and the logical validity of the inferences that they display. The literary features of the narrative of a thought experiment are cognitively irrelevant. But the literary features of the Book of Job seem to be an essential part of its role as divine revelation. The proposal seems not to express a satisfactory appreciation of the literary character of the Book of Job, one can conclude.
Sixth, thought experiments and real-world experiments constitute a dialectical relationship (see Buzzoni 2008). Without the former, the latter are blind; without the latter, the former are empty. It is impossible, however, to run the human experiment of the Book of Job in the real world. This is true ethically and metaphysically. The proposal renders the Book of Job cognitively empty. But the faithful do not receive the Book of Job that way.
Seventh, to categorize the Book of Job as a thought experiment does not contribute in any meaningful way to a better understanding of the complex text. Obviously, the Book of Job does provoke discussion about the meaning of suffering and divine providence. We do not need to call it a thought experiment to appreciate that much. The proposal seems, therefore, poorly motivated.
Eighth, thought experiments are pieces of fiction insofar as they satisfy two conditions of fictionality: First, the “make-belief” condition; second, the “non-fidelity-constraint”. The former requires that fictional narratives are the result of someone’s intention to make believe the content of a narrative instead of actually believing it. The second condition is about the particulars of the content of the narrative; events may be constitutive for the featured scenario that are not believed to be true (see Davies 2007, pp. 31–32; Davies 2018, pp. 517–18). The Book of Job, however, is about divine truths. The proposal downplays the significant difference between literary fiction and divine revelation.

2. The Experimental Character of the Book of Job

These are serious objections that warrant a closer look at the proposal that the Book of Job is a thought experiment, and the discussion must begin with the glaring experimental character of the book’s scenario. God presents Job to Satan (“Accusing Angel”) as an exemplary human being. It is God who directs Satan’s attention to the perfect integrity of Job. Satan suspects, however, that Job’s conduct of life is determined by the good life he enjoys:
“‘Doesn’t Job have a good reason for being so good? Haven’t you put a hedge around him—himself and his whole family and everything he has? You bless whatever he does, and the land is teeming with his cattle. But just reach out and strike everything he has, and I’ll bet he curse you to your face.’”
God allows Satan to reach out and strike everything Job has: “‘All right: everything he has is in your power. Just don’t lay a hand on him.’” (Mitchell 1987, p. 6). Great disasters befall Job. And so it happens that in one stroke, everything is taken away from Job, including his children:
“In these disasters […], and in their combination and suddenness, there are two different kinds of suffering for Job. […] On the one hand, there is suffering such as physical pain or the psychological pain of sudden reversals of fortune. […] By contrast, the pain stemming from the death of a child is an agony of an inward sort. […] both kinds of suffering come to Job in the most intense way, not only because the disaster affects all his substance and every one of his children, but also because both kinds of losses happen at once. […] In addition, […] when disasters have such suddenness and such genesis, they rattle the psyche, so that a person has trouble trusting things in the way he did before.”
But Job continues to “bless the name of the Lord and God” confronts Satan with Job’s display of perfect integrity, “‘even after you made me torment him for no reason.’” (Mitchell 1987, p. 8) Satan is not impressed: “‘So what? A man will give up everything he has, to save his own skin. But just reach out and strike his flesh and bones, and I bet he’ll curse you to your face.’” (Mitchell 1987, p. 8) And God responds: “‘All right: he is in your power. Just don’t kill him.’” (Mitchell 1987, p. 8) Job ends up with boils.
“Job’s suffering in this second wave, however, is hardly limited to boils, even when we take into account the added psychic dimension of this physical disorder. In the first place, there is also the ordeal of the social shunning that Job’s suffering brings him by the time of his affliction with boils.” (Stump 2010, p. 182)
Added to all this is an “evident alienation Job endures from those who should be most supportive of him, most caring for him. […] His suffering has plunged him into a state of mental disorder that Job himself describes as God’s afflicting him with terror…” (Stump 2010, pp. 182–83).
Everything that is happening is greatly unjust because Job “was a man of perfect integrity, who feared God and avoided evil.” (Mitchell 1987, p. 5) Yet, Job does not leave his path of righteousness, although he questions God; he demands an explanation, as he sees no guilt on his part. God appears after some lengthy dialogue between Job and his friends about the tragic events that have befallen Job. It is a tasking dialogue for Job; he suffers and finds his perfect integrity questioned by his friends, while himself remaining in conversation with God. When God finally makes an appearance, the reader is greatly surprised that God does not offer Job any explanation: “The one thing Job wants, as he says over and over, is for God to explain to him why he suffers; and, on the common reading of Job, that is the one thing Job never gets from God.” (Stump 2010, p. 184) Instead, God actually dares put Job into his place: “Where were you when I planned the earth?” (Mitchell 1987, p. 79) A litany follows that lists all the great powers and achievements of God. Not a single word about God’s bet with Satan. Job resigns nevertheless:
“I am speechless: what can I answer? I put my hand on my mouth. I have said too much already; now I will speak no more.” (Mitchell 1987, p. 84) And, following that, “the Lord returned all Job’s possessions and gave him twice as much as he had before. […] After this, Job lived for a hundred and forty years.”
Overall, this is a disturbing story, “because at the end of the story the ten children of Job’s who died in the catastrophes of Job’s affliction are still dead. Although he has ten more children, these children do not restore to him the children he had lost.” (Stump 2010, p. 180) This certainly does not make God look good. From a historical–critical perspective on the text, it might seem attractive, therefore, to dismiss the part about the bet between God and Satan (the “legend”), and to focus on the poem that features the dialogue between Job and his friends: “we mustn’t take the legend too seriously.” (Mitchell 1987, p. xi) However, this seems problematic—a selective reading to avoid uncomfortable questions about the text in its canonized final form. Reading it as a thought experiment presents itself as a reasonable option at this point.
The book’s story is very similar to scenarios that we see featured in many of the thought experiments that facilitate discussions in ethics. A case in point is the very popular “trolley experiment”, of which many variations exist. It continues to attract scholarly discussion, functions as a text book case in many disciplines (such a philosophy and psychology), and serves various pedagogical purposes (to exemplify the merits and perils of intuitions, the demonstrate the importance of thought experiments in ethics, etc.). Here is one version of the “trolley experiment” (see Myers 2004, pp. 41–43): Imagine you see a train approaching a group of five construction workers doing work on the train tracks. You know they will be killed unless the train is stopped somehow. You see the train while standing on a bridge. There is a very fat person standing next to you. You know that his weight will be enough to stop the train. Therefore, when the train reaches the bridge, you throw that person down in front of the train. You seem like a hero, because you have saved five human lives, although one human life had to be sacrificed. But: No, no, no, an intuitive response goes, every human life is of intrinsic value. The numeric outcome of the scenario is irrelevant for the moral quality of the action under consideration. Every human life is unique. The death of the fat man is an insurmountable loss. You have acted morally wrong by killing the fat man. Only numerically, but not ethically, is there a difference between the alternative actions of either watching the train run over the five construction workers or throwing the fat man down the bridge in front of the train. And, indeed, most people responding to this thought experiment would not throw down the fat man. They would opt for the first course of action.
Now imagine you are observing the same situation, but this time from a position that gives you access to a switch that will allow you to divert the train to another track. On that track is only one construction worker. You activate the switch to divert the train to that track because you would save five lives, namely the lives of the construction workers on the track from which the train is diverted. Interestingly enough, most people responding to the thought experiment would activate the switch to divert the train. This creates a bit of an explanatory problem, because this action violates the ethical principle that was suspect to drive people’s response to the first thought experiment. The outcome is identical: Five lives saved, and one life taken. We need a different explanation, therefore, to account for the difference in the way people respond to the two thought experiments.
One explanation is that the first thought experiment requires you to actively kill a person, while the second only requires that you make use of a switch to divert the train. The train, not you, kills the construction worker who is working on the track to which you have diverted the train. Whatever the merits of this account, the thought experiment that warrants such considerations seems like a very powerful epistemic device. It seems to allow us to have a closer look into important moral issues: “we can learn from considering unreal circumstances truths about our actual circumstances. To deny this is to accept an absurdly limited account of inquiry.” (Cameron 2015, p. 30).
It is important to consider at this point the view that “the God of the Hebrew Bible is not in the business of demanding belief in some fixed body of propositions. The biblical God is portrayed as revealing his truths and unleashing his deeds in response to man’s search for truth.” God wants humanity to question God. God wants humanity to search for God. God has a clear preference for individuals that question God’s decrees and actions: Abraham, Moses, Gideon, David, Isaiah, Jeremiah, Ezekiel, Havakuk, Jonah, and Jacob. Jacob struggled a whole night with God and earned, eventually, the name Israel. A careful reading of the Hebrew Bible leaves us “without the possibility of seeing Hebrew Scripture as calling for man to adopt a life of submissive belief in a catechism we regard as repugnant to reason.” The Book of Job proves this point. “As Job tells his companions: ‘Let me alone that I may speak, and let come on me what will…Though [God] may slay me, yet will I trust him. But I will maintain my own ways before him. This also will be my salvation. For a flatterer will not come before him.’” (Hazony 2012, p. 235).
Thought experiments have a very good reputation for facilitating such scrutiny. If the Hebrew Bible is, generally speaking, normative textual testimony to humanity’s scrutiny of God’s action, then it does not seem utterly implausible to read the Book of Job as a thought experiment that facilitates such scrutiny.

3. Thought Experiments in Theology and Science

There is further reason to take the proposal seriously that the Book of Job is a thought experiment. It concerns the role of thought experiments in the encounter between modern science and theology. John Polkinghorne was probably the first to claim that thought experiments constitute a natural link between science and religion (see Polkinghorne 2007, pp. 93–94). This is a controversial idea indeed, although not as far-fetched as it may seem from a position of a principled opposition to the enterprise of a dialogue between theology and science (see Fehige 2012). Consider only the gospel of Luke’s Jesus, who bases his encouragement to pray “on the following TE (thought experiment, Y.F.): Is there anyone among you, who if your child asks for a fish, will give it a snake instead of a fish? If you then, who are evil, know how to give good gifts to your children, how much more will the heavenly father give the Holy Spirit to those who ask him? (Luke 11:12–13).” (Gregerson 2014, p. 6). It is indeed “easy to see that many kinds of stories are like thought experiments, including […] the parables of Jesus.” (Davenport 1983, p. 283) They may be understood as “vehicles for exploration and discovery, providing contexts in which features may be demarcated, their interplay examined, their implications drawn out.” (Elgin [1991] 1993, p. 25) There is no significant gap opening up here to scientific thought experiments insofar as the fictional character of the stories is concerned. Even scientific theories
“maybe composed of both factual and fictional sentences, and the fictional sentences may play any of several different roles. This means that to understand what a theory conveys, and to understand the phenomena in terms of the theory requires sensitivity to the different roles the different sentences play. And to assess a theory requires determining whether the component sentences are true enough for the parts they are assigned to play. […] it is not plausible to think of an acceptable theory as a mirror of nature. Even if the goal of a theory is to afford an understanding of a range of facts, it need not approach or achieve that goal by providing a direct reflection of those facts.”
In positive terms, we may say that theories “are ways of looking at the facts—pairs of spectacles through which to see the world differently. What makes theories persuasive in the first place is […] something in them which answers to a wider need. There is always an imaginative appeal involved as well as intellectual thirst for understanding.” (Midgley 2001, p. 26; italics in original).
Polkinghorne seems in a good position when he proposes to think of some of the New Testament’s passages as thought experiments. Some of those, he argues, serve cognitive purposes that seem identical to some that we find in quantum physics. We find them employed especially in explorations of those abstract concepts whose referential context is significantly removed from the reality as we experience it. Such concepts leave us no choice but to perform thought experiments in order to make sense of their cognitive core and, thus, of the reality to which they refer. This is the case in matters eschatological, for instance; and similarly, physicists rely on thought experiments to struggle with such notions as Heisenberg’s uncertainty principle. Christian theology would be severely impoverished without thought experiments, just like philosophy and physics would be. This is the upshot of Polkinghorne’s reflections. And, indeed, particularly in discussions among theologians in the analytic school, we find plenty of thought experiments. For example, Richard Swinburne’s proof of the coherence of the classical theist’s claim that there is an incorporeal personal God, who is “lacking a body yet being a person” (Swinburne 1993, pp. 106–7), and this in a very different way than “a ghost or poltergeist”, involves a thought experiment. We are to imagine ourselves “gradually ceasing to be affected by alcohol or drugs”. Manipulation of our brains will not affect the coherence of our thinking. “Imagine too that you cease to feel any pains, aches, and thrills, although you remain aware of what is going on in what has been called your body.” Then you reach the state of becoming aware of the events in other bodies other than our own. We are further to imagine that we gain awareness of “other material objects at any place in space” without the assistance of means comparable to those that our body provides us with. You are to imagine assuming a position that allows you “to see things from any point of view, which you choose”, and “to move directly anything which you choose”, and “to utter words which can be heard anywhere”. While you are “gaining all these strange powers, you remain otherwise the same—capable of thinking, reasoning, and wanting, hoping and fearing.” Swinburne finds that surely anyone “can thus conceive of himself becoming an omnipresent spirit. So it seems logically possible that there be such a being.”

4. Literary Fiction and Thought Experiments

The extensive use of thought experiments in analytic discourse on matters theological provoked a noteworthy intervention by Eleonor Stump (see Stump 2009, 2010). Stump argues that the popular method of thought experiments in analytic theology is a symptom of the inability to relate in a meaningful way to “the part of reality that includes the complex, nuanced thought, behaviour, and relations of persons.” (Stump 2010, p. 25) As a remedy, she proposes that analytic theologians abandon the pieces of bad literature that we have come to call thought experiments and include, instead, literary narratives in philosophical analysis. This will enable philosophers to access the unique kind of knowledge that pertains to the reality of persons, “especially those in which the qualia of the experience are among the salient parts of the knowledge. Another important species of such knowledge is acquired in direct interaction with other people.” (Stump 2009, p. 258) Such knowledge cannot be expressed in propositional knowledge; all we can do is “to re-present the experience itself in such a way that we can share it with others who were not part of it, so that the knowledge of persons garnered from the experience is also available to them.” (Stump 2009, p. 258).
The problem of suffering is intrinsically linked to that reality of persons. From Stump’s theistic perspective, it is primarily a problem about the relationship between God as a person and humans as persons. In other words, through the inclusion of literary narratives, philosophy will be much more resourceful to address the problem of suffering in a meaningful way: “So one way to compensate for the limitations of analytic philosophy as regards philosophical problems such as the problem of suffering is to reflect on these problems by drawing on the insights of narratives as well as the results of contemporary analytic discussions.” (Stump 2010, p. 25).
I have great sympathies for Stump’s project, especially in light of Bas van Fraassen’s critical take on the sterile nature of much of analytic theology in its richness of conceptual precision and technical inquiry: “Riches come with a temptation, a tempting fallacy, namely, to have us view them as all there is to be had, when they are so much. This is true of all riches, and it is true of the riches of objective knowledge. Poor are the rich who succumb to this fallacy.” (van Fraassen 2002, p. 195) The encounter with God is a personal encounter. At the core of that encounter we find a mystery, and it is not God as such: “The mystery is that of the ‘real presence.’ How can this thing that is not a thing but a perspective, appear in the world of objects where it occupies no place? How is it that we can not only address the other, but actually encounter him in the empirical world?” (Scruton 2014, p. 97).
All this is to say that I can appreciate Stump’s central concern. Good reasons are available to believe that the limits of our propositional knowledge are not the limits of our world. Stump’s pejorative stance on thought experiments, however, is unnecessary, I find. What puzzles me in particular is that the possibility escapes her that the cognitive power she attributes to literature is a function of thought experiments. Peter Swirski, for example, has accounted for the cognitive power of literature in terms of thought experiments (see Swirski 2007). He claims that “[e]motions and qualia are, and forever will be, central to the experience of literature in a manner they are not to philosophy and science. But that in no way precludes studying the non-subjective elements of literary fictions with a view to non-subjective knowledge.” (Swirski 2007, p. 20) This cognitive role, he argues, is possible due to thought experiments, which in turn is to be explained in its cognitive power by the evolution of storytelling that is essential to humanity. He does not downplay the fact that “in literature the process [of experimenting in thought, Y.F.] is more diffuse, instinctive, and incomplete than in philosophy of science.” Still, “the structure and global strategy of thought experiments all three employ are not always that different.” (Swirski 2007, p. 123).
What Swirski proposes is certainly not implausible, although it raises many questions, of course—philosophers do find fault. Cognitivism in literary theory is contentious, and evolutionary accounts of cognition raise eyebrows. But the link to literary fiction seems compelling insofar as thought experimental narratives indeed satisfy the conditions of fictionality, as we have seen above. It is, therefore, a serious option to think “that some works of fiction are properly viewed as much more fully elaborated [thought experiments].” (Davies 2007, p. 33).
Catherine Elgin is in agreement and presents an account of the cognitive power of fiction in arts and science in terms of what she terms exemplification. An exemplar both refers to and instantiates a feature. This way, it can enhance understanding of that feature. This is possible because it affords epistemic access to the feature it exemplifies:
“Exemplification is […] ubiquitous in arts and science. An experiment is designed to exemplify aspects of the phenomena it investigates. It does not purport to replicate what occurs in the wild. Instead, it selects highlights, controls, and manipulates things so that certain features are brought to the fore and their relevant characteristics and interactions made manifest. […] Thought experiments function similarly.”
Elgin finds it perfectly reasonable to think that there are short and austere thought experiments and long and more elaborate thought experiments. The latter may be a short story or a novel:
“I suggest that literary fictions are extended, elaborate thought experiments. So if thought experiments afford epistemic access to cognitively important factors, works of fiction do, too. […] Like an experiment, a work of fiction selects and isolates, manipulating circumstances so that particular properties, patterns, and connections, as well as disparities and irregularities are brought to the fore.”
Important to note is that the literary aspects of the narratives we are presented with contribute to their cognitive function, and not merely their entertainment value. This must seem implausible from the perspective of an epistemology that “valorizes truth” (Elgin 2004, p. 113), and which operates with a notion of cognition according to which there is only one epistemic goal, namely “to augment our stock of justified true beliefs.” (Elgin [1991] 1993, p. 13).
Such an epistemology drives James R. Brown’s Platonist account of thought experiments, for instance. It conceives of them as “windows” into a realm of abstract entities. Thought experiments allow us to see what is happening in that realm. What we see can help us to justify new true beliefs about the concrete empirical world. Observable regularities can be turned into necessities of nature this way. Galileo Galilei discovered a new law of motion just by thought experiments (see Brown 2011, pp. 98–124). We have discovered genuinely new truths about the world.
The empiricist’s travel with a truth valorizing epistemology results in a much more explicit disregard for the literary features of thought experimental narratives. It becomes part of the definition: “Thought experiments are arguments which (i) posit hypothetical or counterfactual states of affairs, and invoke particulars irrelevant to the generality of the conclusion” (Norton 1993, p. 129). The cognitive power thought experiments can exercise is a function of propositional reasoning only. In principle, therefore, thought experiments are dispensable. We could spell out the thought experiment in argument form and unleash the same cognitive power.
Platonists and empiricists agree that the literary features of thought experimental narratives are cognitively irrelevant. And this is partly the result of a commitment to an epistemology according to which we should accept only what we consider true, take pains to ensure that the claims we accept are in fact true, and promptly repudiate any previously accepted claims upon learning that they are false. According to a prominent view in analytic theology, there is nothing wrong with this epistemology. The “best kind of faith [is] a faith in which the agent follows a religious way which is good to follow and the best one to follow given the different probabilities on the agent’s evidence, obtained after adequate investigation, that the creeds of different ways are true. A faith will be rational to the extent to which it begins to approximate to that ideal of rational faith.” (Swinburne 2005, p. 226) If we think of science as a first class exemplar of rationality, however, this ideal of rational faith seems less compelling than one might think. There are several problems.
For starters, it suffers from a “‘closure obsession’, which has been prevalent in the twentieth century not only among scientists, but among philosophers and historians of science as well.” (Chang 2012, p. 87) This, in turn, seems partly the result of a questionable commitment to the metaphysical view that “the nature of the world is such that it can, at least in principle, be completely described or explained by” a “single, complete, and comprehensive account of the natural world (or the part of the world investigated by the science) based on a single set of fundamental principles.” (Kellert et al. 2006, p. x) There is less empirical evidence than one might think to suggest that this metaphysical view is correct. Why would we think of the nature of the world in such a way when it comes to the scientific investigation of the world? There are good reasons to think that empirical evidence actually rather suggests what has been called a “dappled world” (see Cartwright 1999), although I myself would not want to endorse the idea to counter a problematic metaphysics with more metaphysics. The best we can offer to scientific monism are temporally limited a posteriori reasons concerning the monism/pluralism divide, I think.
Pluralism in science is a viable option. And this includes a pluralism in matters of methodology, for example, “the laws, models, idealizations, and approximations which are acknowledged not to be true, but which are nonetheless critical to, indeed constitutive of, the understanding that science delivers.” (Elgin 2004, pp. 113–14) Similarly, the “non-truths that constitute fiction are cognitively valuable because they equip us to discern truths that we would otherwise not see or see so clearly.” (Elgin 2007, p. 53) And it extends to pluralism in matters of explanation. “Explanation is not a standard item. It is whatever information or reasoning will solve the particular problem that is causing trouble at the time.” (Midgley 2001, p. 124).
The Book of Job has been used to advance an explanation of human suffering. Difficulties arise when a narrow definition of explanation is assumed, according to which “asking for an explanation […] is simply asking for […] a physical condition that produces it.” (Midgley 2001, p. 124) This is reasonable because many phenomena do require exactly that. But human suffering is not among them. Stump makes a very good point in this respect: “nothing in any conceivable and reasonable explanation of suffering could undermine the fact or the force of that suffering.” (Stump 2010, p. 180) Stump seems right here if I am correct in assuming that what she means by explanation is the search for a physical efficient cause of suffering. But when it comes to Stump’s skepticism about thought experiments in comparison to literature, I have difficulties to follow her. This is especially true in light of her own reading of the Book of Job as partially explanatory in the sense that it contributes to the solution of the theological problem that the fact of suffering of innocent human beings creates.
I cannot discuss Stump’s own reading of the Book of Job at length (Stump 2010, pp. 177–226) in the present context, but I would be remiss if I did not state at least why I think that her reading confirms much of what the previous discussion entails with respect to the link of literary fiction and thought experiments in philosophical analysis proper. For starters, she rightly conceives of Job as an exemplar and thus uses the Book of Job as an important piece of textual evidence to develop her own “Thomistic” theodicy, which is basically the idea that there is some good in human suffering that warrants belief in God as the cause of such suffering insofar as God is believed to be the morally perfect and all powerful creator of everything. Moreover, she can actually motivate her own reading of the book only due to the thought experimental character of the book. Stump’s reading is directed against what she claims to be the “common reading” of the Book of Job. According to this reading, the book does not offer any solution to the problem of human suffering. No explanation is provided to make sense of the suffering that Job had to endure unjustly. Stump rightly notes that Job submits to God nevertheless, even though God does not respond to Job’s questions. She finds this very puzzling and basically asks the reader to run in her mind the scenario of the book: Job was appalled by what was happening to him and ready to challenge God. But when God appears and simply demonstrates his divine power, Job submits to God, and he submits in a most submissive way. This reading cannot be right, she claims. Stump obviously has a particular image of Job in mind, an image that has some textual basis but mostly relies on intuitions about human behavior in similar situations.
The image guides me to very a very different conclusion than Stump. I cannot bring myself to agree with Stump in her creation of a puzzle inherent in the common reading of Job’s response to God presence: “God Himself is present: Heart, be stilled before Him: Prostrate inwardly adore Him.” This is what we read in Rudolf Otto’s seminal study on holiness (Otto 1936, p. 18). And: “Job has been reasoning with his friends against Elohim, and as far as concerns them he has been obviously in the right. They are compelled to be dumb before him. And then Elohim Himself appears to conduct His own defence in person. And He conducts it to such effect that Job avows himself to be overpowered, truly and rightly overpowered, not merely silenced by superior strength. Then he confesses: Therefore I abhor myself and repent in dust and ashes. That is an admission of inward convincement and conviction, not of impotent collapse and submission to merely superior power.” (Otto 1936, p. 80) There is no need, however, to invoke much theory to see my point. I do not find it remarkable at all that Job submits the way he does. Just imagine a father—something Stump wants you to do, too—who has done wrong to his child. The child rightly protests. When fathers invoke their authorities as loving parents in such events and in a manner comparable to the way God does in response to Job’s challenge, children do submit in ways comparable to what we find in the Book of Job. The very presence of God exercises overwhelming loving authority; humans can submit in most humble ways if that happens. There is nothing remarkable here, I find. What is remarkable, however, is the bet between God and Satan, and God’s silence about it when facing Job. From a historical–critical perspective, there is a very good explanation available for God’s silence: The “legend” of the bet between God and Satan is imposed on a different text that features Job’s struggle with God in light of disasters that had befallen him and at whose core we find various dialogues between Job and his friends. Stump does not want to read the book in accordance with such historical–critical findings. She perceives a complex text with multiple layers featuring different kinds of interpersonal relationships which are in superposition, as it were: The relationship between God and Job, the relationship between God and Satan, Job’s relationship to himself, and Job relationships to his wife and friends. This complexity, I suggest, are most successfully navigated if we take the proposal seriously that the Book of Job is a thought experiment. And my sense is that the proposal actually strengthens what Stump herself has to say about the central message of the book with respect to a theodicy:
“In the divine speeches, then, there is not a claim but a suggestion, a picture, that leads us to think God operates on the principle I attributed to good parents. This principle applied to God is not equivalent to the justly ridiculed Leibnizian position that this is the best of all possible worlds in which everything that happens happens for the best. Rather, this is a principle just about one necessary condition for good parenting—namely, that, other things being equal, the outweighing benefit that justifies a parent in allowing some suffering to an innocent child of hers has to benefit the child primarily.” (Stump 2010, p. 191)
The problem is, of course, that no relevant benefit is discernible in the Book of Job. In fact, the “legend” suggests the absence of any benefit for humanity whatsoever—not to mention the many widely noted shortcomings of Stump’s Thomistic theodicy, including this one:
“At one point Stump writes, ‘if it turned out that most suffering really is pointless, as many people suppose, that would be evidence against the Thomistic defense’ (p. 457). Elsewhere she states that ‘if our own experience or the experience of others were always only of psychic disintegration in consequence of suffering, that would count against the defense’ (p. 460). However, these are pretty low bars to set for the omnipotent, omniscient, omnibenevolent God presented in Stump’s defense. Why would we not expect this God to be primarily successful in the employment of the medicinal means of suffering? Based on the narrative examples Stump focuses on, with their uniformly positive outcomes, it seems like we should be able to do so.”
I cannot see that her reading of the Book of Job should make us optimistic in this respect. I must leave it at that in the present context.

5. Conclusions

The principal aim of the previous discussion was not to present a cognitivist account of literary fiction as part of a theology of revelation. The point of the epistemological qualifications that were introduced is threefold:
First, they were meant to indicate clearly that literary cognitivism is a viable option. Second, with their help, an awareness arose as to the limits of a theological method that excludes literary fiction as part of philosophical analysis proper. The qualifications are not meant to defend the claim that the story of Job is indeed divine revelation and, thus, a normative text for theological discussions about human suffering from a Jewish, Christian, and Islamic perspective. It is reasonable that this much can be assumed without further justification in the context of the Special Issue to which this paper makes a contribution. Third, the epistemological qualifications were instrumental in countering the objections raised at the beginning of this paper.
They helped to establish that the fictional character of the narratives of thought experiments must not necessarily clash with the Book of Job’s position in the Tanach, Old Testament, and Quran—sources of divine revelation and truth. This was the eighth objection that was introduced at the beginning of the paper. As for the seventh, the previous discussion has exposed important parallels between the debate over literary cognitivism and the controversy over thought experiments. What is interesting is the paradoxical situation that the fact “that TEs (thought experiments, YF) are fictions has been taken (by some) to call into question the very thing that is supposed to be established (for others) by the fact that fictions are TEs.” (Davies 2018, p. 518) The proposal that the Book of Job is a thought experiment directs, therefore, our attention to the right questions concerning the book’s cognitive role in theological inquiry. And this seems the right direction, especially in light of the glaring experimental character of the book’s story that the previous discussion looked into. The sixth and fifth objection that were listed at the beginning of this paper seem to be a function of an epistemology that “valourizes truth”. The discussion above raised some important questions about the absoluteness of such an epistemology. In particular, we encountered good reasons to think of the experimental character of the Book of Job in other terms than the submission of propositional claims to confirmation in a realm of entities that are independent of mind and language. For instance, we touched upon the idea that experiments are essentially about exemplification, in contrast to the view that experiments are primarily a technological–operational realization of an empirical test of a well-ordered set of beliefs. The work of John Polkinghorne on thought experiments in physics and Christian theology helped us to gain some critical distance to the fourth objection. Of course, the discussion of Polkinghorne’s views on thought experiments cannot be taken to translate into a knock-down argument against the claim that science and theology are absolutely independent of each other. This was also not the task at hand. Rather, the concern to be addressed was that the use of thought experiments would bring theology and science close enough to cause damage to religion. This does not seem to be true. As for the third objection, there are much longer thought experiments than the Book of Job if it is true that we have good reasons to believe that we should conceive of great literary masterpieces as thought experiments. Hence, the length of the Book of Job is not to be taken as powerful evidence to reject the proposal under discussion. The second objection has a good point, however. Narratives and thought experiments are indeed not identical. The discussion above established that much. That being said, it seems that nothing follows from that concerning the plausibility of the proposal under consideration. The same seems true of the first objection. We can appreciate that the conceptual history of the notion of thought experiments begins only with the rise of modern science, because our “conception of ‘thought experiment’ to make the specific sense it has (however vague it may be), […] requires a context in which the concept of ‘experiment’ has some currency among a relevant community of thinkers.” (Corcilius 2018, p. 57) This important historical qualification does not translate into a serious problem for the proposal that the Book of Job is a thought experiment. For starters, the proposal is well outside the boundaries of a historical–critical analysis of the book. And this fact in itself is not an insurmountable problem, because: “Biblical hermeneutics […] is a part of the general hermeneutics applying to every literary and historical text […].” (Pontifical Biblical Commission 1993, A.2) Philosophers use the category of thought experiments when studying the work of premodern thinkers. They find thought experiments, for example, in the pre-Socratics (Rescher 1991), and in Plato (Becker 2018), as well as in Aristotle (Corcilius 2018), and also in the writings of thinkers of the medieval period (see King 1991)—in and outside the Christian “West” (McGinnis 2018). Biblical hermeneutics should not be deprived of the conceptual tool of thought experiments. Such a concession does not amount to a denial of the relative uniqueness of biblical hermeneutics. On the contrary, there can be no doubt that biblical thought experiments add in significant ways to the degree of complexity of the topic of literary cognitivism: “Inspired by God, the Bible is the trustworthy revelation of God. It is Holy Scripture because it just is the Word of God.” (McCall 2009, p. 172) But this fact in and of itself does not render the proposal under consideration implausible, especially in light of claims that the Book of Job can be read as a philosophical text (see Hazony 2012).

Conflicts of Interest

The author declares no conflict of interest.


  1. Anderson, Alva K. 2012. Review of Eleonore Stump: Wandering in Darkness: Narrative and the Problem of Evil. International Journal of Philosophy of Religion 71: 163–66. [Google Scholar] [CrossRef]
  2. Bealer, George. 1998. Intuitions and the Autonomy of Philosophy. In Rethinking Intuitions: The Psychology of Intuition and Its Role in Philosophical Inquiry. Edited by Michael DePaul and William Ramsey. Lanham: Rowman & Littlefield, pp. 201–39. [Google Scholar]
  3. Becker, Alexander. 2018. Thought Experiments in Plato. In The Routledge Companion to Thought Experiments. Edited by Michael T. Stuart, Yiftach Fehige and James R. Brown. London and New York: Routledge, pp. 44–56. [Google Scholar]
  4. Brown, James R. 2011. The Laboratory of the Mind: Thought Experiments in the Natural Sciences, 2nd ed. New York and London: Routledge. [Google Scholar]
  5. Buzzoni, Marco. 2008. Thought Experiment in the Natural Sciences. Würzburg: Königshausen & Neumann. [Google Scholar]
  6. Cameron, Ross P. 2015. Improve your Thought Experiments Overnight with Speculative Fiction. Midwest Study in Philosophy XXXIX: 29–45. [Google Scholar]
  7. Cartwright, Nancy. 1999. The Dappled World: A Study of the Boundaries of Science. Cambridge: Cambridge University Press. [Google Scholar]
  8. Chang, Hasok. 2012. Is Water H2O? Evidence, Realism and Pluralism. Dordrecht: Springer. [Google Scholar]
  9. Corcilius, Klaus. 2018. Aristotle and Thought Experiments. In The Routledge Companion to Thought Experiments. Edited by Michael T. Stuart, Yiftach Fehige and James R. Brown. London and New York: Routledge, pp. 57–76. [Google Scholar]
  10. Davenport, Edward. 1983. Literature as Thought Experiment (On Aiding and Abetting the Muse). Philosophy of Social Sciences 13: 297–306. [Google Scholar] [CrossRef]
  11. Davies, Dave. 2007. Thought Experiments and Fictional Narratives. Croatian Journal of Philosophy VII: 29–45. [Google Scholar]
  12. Davies, Dave. 2018. Art and Thought Experiments. In The Routledge Companion to Thought Experiments. Edited by Michael T. Stuart, Yiftach Fehige and James R. Brown. London and New York: Routledge, pp. 512–25. [Google Scholar]
  13. Elgin, Catherine Z. 1993. Understanding; Art and Science. Synthese 95: 13–28. First published in 1991. Midwest Studies in Philosophy 16: 196–208. [Google Scholar] [CrossRef]
  14. Elgin, Catherine Z. 2004. True Enough. In Epistemology: Philosophical Issue. Edited by Ernest Sosa and Enrique Villanueva. Hoboken: Blackwell Publishing, pp. 113–31. [Google Scholar]
  15. Elgin, Catherine Z. 2007. The Laboratory of the Mind. In A Sense of the World: Essays on Fiction, Narrative, and Knowledge. Edited by John Gibson, Wolfgang Huemer and Luca Pocci. London and New York: Routledge, pp. 43–54. [Google Scholar]
  16. Fehige, Yiftach. 2012. Quantum Physics and Theology: John Polkinghorne on Thought Experiments. Zygon® 47: 256–88. [Google Scholar] [CrossRef]
  17. Gooding, David. 1992. The Procedural Turn; or, why do Thought Experiments work? In Cognitive Models of Science. Edited by Ronald Giere. Minneapolis: University of Minnesota Press, pp. 45–76. [Google Scholar]
  18. Gregerson, Niels H. 2014. The Role of Thought Experiments in Science and Religion. In The Science and Religion Dialogue: Past and Future. Edited by Michael Welker. Bern: Peter Lang, pp. 130–40. [Google Scholar]
  19. Hazony, Yoram. 2012. The Philosophy of Hebrew Scripture. Cambridge: Cambridge University Press. [Google Scholar]
  20. Kellert, Stephen J., Helen Longino, and Kenneth Waters, eds. 2006. Scientific Pluralism. Minneapolis: University of Minnesota Press. [Google Scholar]
  21. King, Peter. 1991. Medieval Thought Experiments: The Metamethodology of Medieval Science. In Thought Experiments in Science and Philosophy. Edited by Tamara Horowitz and Gerald J. Massey. Savage: Rowman & Littlefield Publishers, pp. 43–64. [Google Scholar]
  22. Kühne, Ulrich. 2005. Die Methode des Gedankenexperiments. Berlin: Suhrkamp Verlag. [Google Scholar]
  23. McCall, Thomas. 2009. On Understanding Scripture as the Word of God. In Analytic Theology: New Essays in the Philosophy of Theology. Edited by Oliver D. Crisp and Michael C. Rea. Oxford: Oxford University Press, pp. 171–86. [Google Scholar]
  24. McGinnis, Jon. 2018. Experimental Thoughts on Thought Experiments in Medieval Islam. In The Routledge Companion to Thought Experiments. Edited by Michael T. Stuart, Yiftach Fehige and James R. Brown. London and New York: Routledge, pp. 44–56. [Google Scholar]
  25. Midgley, Mary. 2001. Science and Poetry. London and New York: Routledge. [Google Scholar]
  26. Mitchell, Stephen. 1987. The Book of Job. New York: HarperPerennial. [Google Scholar]
  27. Myers, David. 2004. Intuitions: Its Powers and Perils. New Heaven: Yale University Press. [Google Scholar]
  28. Nersessian, Nancy. 1992. In the Theoretician’s Laboratory: Thought Experimenting as Mental Modeling. In PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association, Vol. 1992, Volume Two: Symposia and Invited Papers. Chicago: University of Chicago Press, pp. 291–301. [Google Scholar]
  29. Nersessian, Nancy. 2007. Thought Experiments as Mental Modelling: Empiricism without Logic. Croatian Journal of Philosophy VII: 125–61. [Google Scholar]
  30. Norton, John D. 1993. Thought Experiments in Einstein’s Work. In Thought Experiments in Science and Philosophy. Edited by Tamara Horowitz and Gerald J. Massey. Savage: Rowman & Littlefield Publishers, pp. 129–44. [Google Scholar]
  31. Norton, John D. 1996. Are Thought Experiments Just What You Thought? Canadian Journal of Philosophy 26: 333–66. [Google Scholar] [CrossRef]
  32. Otto, Rudolf. 1936. The Idea of the Holy: An Inquiry into the Non-Rational Factor and the Idea of the Divine and Its Relation to the Rational, 6th rev. ed. Translated by John W. Harvey. Oxford: Oxford University Press. [Google Scholar]
  33. Polkinghorne, John. 2007. Quantum Physics and Theology: An Unexpected Kinship. New Haven: Yale University Press. [Google Scholar]
  34. Pontifical Biblical Commission. 1993. The Interpretation of the Bible in the Church. Available online: (accessed on 24 January 2019).
  35. Rescher, Nicholas. 1991. Thought Experimentation in Presocratic Philosophy. In Thought Experiments in Science and Philosophy. Edited by Tamara Horowitz and Gerald J. Massey. Savage: Rowman & Littlefield Publishers, pp. 31–41. [Google Scholar]
  36. Scruton, Roger. 2014. The Soul of the World. Princeton: Princeton University Press. [Google Scholar]
  37. Sorensen, Roy A. 1992. Thought Experiments. Oxford: Oxford University Press. [Google Scholar]
  38. Stump, Eleonore. 2009. The Problem of Evil: Analytic Philosophy and Narrative. In Analytic Theology: New Essays in the Philosophy of Theology. Edited by Oliver D. Crisp and Michael C. Rea. Oxford: Oxford University Press, pp. 251–64. [Google Scholar]
  39. Stump, Eleonore. 2010. Wandering in Darkness: Narrative and the Problem of Suffering. New York and Oxford: Oxford University Press. [Google Scholar]
  40. Swinburne, Richard. 1993. The Coherence of Theism, rev. ed. Oxford: Clarendon Press. [Google Scholar]
  41. Swinburne, Richard. 2005. Faith and Reason, 2nd ed. Oxford: Oxford University Press. [Google Scholar]
  42. Swirski, Peter. 2007. Of Literature and Knowledge: Explorations in Narrative thought Experiments, Evolution, and Game Theory. London and New York: Routledge. [Google Scholar]
  43. Thagard, Paul. 2014. Thought Experiments Considered Harmful. Perspectives on Science 22: 288–305. [Google Scholar] [CrossRef][Green Version]
  44. van Fraassen, Bas C. 2002. The Empirical Stance. New Heaven and London: Yale University Press. [Google Scholar]
Back to TopTop