Cognitive Unnaturalness of Science

As McCauley explains at length, practicing science is far from doing what ordinary human minds are inclined to do on their own. Some ways of thinking are easier than others because they suit the kinds of minds humans have. We make assumptions about how the world is, the properties of physical objects, the nature of living things, and how each other think and feel in large part because certain ways of thinking are how our minds usually do what they do. These conceptual tendencies may have arisen because they were adaptive for our ancestors.

Further, we appear to have a host of domain-general reasoning biases and heuristics that help us solve problems pretty well most of the time. Nonetheless, these reasoning abilities also have some weaknesses. Some of these weaknesses are so pervasive that we take them for granted. For instance, none of us match the mathematical computational fluency of a 1980 Texas Instruments pocket calculator, even if that machine pales in comparison to the processing power of our brains in so many other domains. As statistics and research methods instructors are painfully aware, we also are pretty shoddy at paying sufficient attention to base-rate probabilities or attending to disconfirming cases. Similarly, introductory logic teachers may resonate with McCauley’s (2011, 55) observation:

In dozens of previous studies of a familiar reasoning problem in the psychological literature (namely, the Wason selection task) experimenters had found that people generally do not do well in carrying out some basic logical inferences when they involve statements about conditions between propositions.

Doing science requires being able to surmount these natural limitations as well as knowing when and how to override or improve upon our reasoning biases and heuristics.

The discoveries of the sciences also commonly run counter to our more domain-specific ways of understanding physical objects, living things, and human minds. For instance, the ideas that solid physical objects are mostly space or that there are no clear and fixed boundaries between species of animals appear to run counter to naive physics and naive biology, respectively (e.g., Sperber, Premack, and Premack 1996, pts. II and IV). This counterintuitiveness of many scientific claims, and the need to effortfully override reasoning biases and heuristics, means that doing science is rarely or never characterized by the fluency, automaticity, and ease of cognitively natural activities such as using language or recognizing faces.

What McCauley has called the relative cognitive “unnaturalness” of science entails a need for extra cultural support for scientists to do what they do. In contrast to, for instance, gossip or engaging in play, modern science does not simply emerge in essentially any cultural environment by virtue of people with ordinary biological and psychological endowments doing what they ordinarily do together. To do science, we need special instructional efforts, institutions, and artifacts.

How to Seek Knowledge of the Natural World

Recognition of basic cognitive limitations implies that we need effortful correctives to improve our conceptual fidelity. If I cannot simply rely on my intuitions and straightforward perceptions of the world around me, what should I do to improve? Scientific approaches to the study of the natural world appear to be impressive improvements over reliance only on folk wisdom, testimony, personal experience, intuition, and careful observation. Something extra has been added to our toolkit for learning and discovering. More accurately, a number of somethings have been added to our toolkit.

Commonly, teachers of science research methods and philosophers of science point to numerous features of modern science as useful for drawing stronger inferences: experimental methods; careful or controlled observation, counting, quantification, and mathematical modeling; generating and testing falsifiable predictions; specification of data collection and analysis methods so that others may inspect them; replicability; representative or random sampling; and a host of other techniques and considerations. Scientists also share with other scholarly pursuits general ways of seeking knowledge such as intuition, application of deductive and inductive logic, personal experience, and the testimony of others. But the “add-ons” that characterize modern science were very rare in the study of the natural world that preceded its emergence, and remain rare outside the sciences.

Does science need some kind of theology to ground the use of these intellectual techniques? In a logical sense, perhaps not. But functionally, perhaps so. Historical and philosophical analyses gesture toward an affirmative answer. Approaching the topic historically, Peter Harrison (2021, 1) writes regarding “the scientific revolution” of the sixteenth and seventeenth centuries in Europe that though “it must be acknowledged that there was science (or something like it) well before this and in places other than Europe,” what we want to explain is how different commitments, motivations, and techniques for learning about the natural world congealed as they did to make something importantly different. “So, it is important to establish at the outset what, if anything, was distinctive about the scientific culture that arose in early modern Europe” (Harrison 2021, 1).

Instead of merely providing and explaining a list of characteristic features of modern science, Harrison (2021, 2) identifies a smaller number of features of this new natural philosophy that arose, at least in part, due to Christian theological considerations in Europe in the early modern era: “[T]hree features of modern science—laws of nature, mathematical and mechanical models, and experimental method—are indebted to religious considerations [i.e., theology].” Harrison (2021, 2) argues that a critical intellectual shift that gave rise to modern science was away from the Aristotelian notion “that nature was ordered according to the intrinsic properties of natural things” and toward “the idea that nature was governed by laws imposed upon it by God.” Pioneered by the likes of René Descartes and Robert Boyle, this move arose from theological considerations. “The specific theological impulse for this shift was a renewed emphasis, most conspicuous in the theology of the Protestant reformers, on the sovereignty of God and the primacy of the divine will” (Harrison 2021, 2–3).

Regarding the use of mathematics, there was a shift from thinking of mathematical models as useful fictions to tracking reality. “Advocates of the new mathematical astronomy . . . argued that God had stamped mathematical relations onto the natural order and that these mathematical relations were not simply human constructions” (Harrison 2021, 3). Consistent with its law-likeness, “nature was increasingly thought of as analogous to a machine, rather than as self-organised and goal-directed living thing” (Harrison 2021, 4). After all, “a direct implication of the Christian doctrine of creation was that the world, in an important sense, was an artificial creation”; that is, “God made the world, and therefore it was analogous to an artifact” (Harrison 2021, 4).

Even more important to the basic epistemological assumptions of modern science was a shift away from the Aristotelian notion that humans are “rational animals” that can more or less reliably perceive and comprehend the world. The trend was toward a greater pessimism concerning human will and intellect, a pessimism arising from the Reformation emphasis on the “fallen condition of human minds and senses, and because the world itself had deteriorated since its original creation” (Harrison 2021, 4). The sinful condition of humans and a fallen creation more resistant to detecting God’s law-like order and purposes required something new:

What was now called for were repeated sets of painstaking observations, conducted under experimental conditions. As for fallen human senses, these now had to be augmented through the use of artificial instruments such as the telescope and microscope. Whereas in the past scientific knowledge could be attained by the wise few, now it was understood the call for the labours of many individuals carried out over long periods of time. Science would necessarily be corporate and cumulative. (Harrison 2021, 5)

As McCauley (2011, 139) observes, this socially distributed dimension of science remains one of its critical features for overcoming the individual conceptual and motivational errors that often are a product of cognitively natural dispositions:

Nonetheless, however poor scientists may prove to be as critics of their own views and of views that they favor, the history of science supplies compelling testimony about their abilities as critics of other scientist’ views and views that they oppose. Scientists may suffer from an all-too-human penchant to confirm the proposals they prefer, but they suffer from no such failing when it comes to assessing competing positions.

Harrison’s historical analysis is complemented by Alvin Plantinga’s (2011) philosophical one. Plantinga too offers a list of ways in which Christian theology provides intellectual support for doing modern science. In many cases, he suggests that theology, especially Christian theology, provides better intellectual scaffolding for doing science than a naturalist worldview. The topics he includes in his analysis are an intellect that fits the world, induction and learning from experience, the reliability and regularity of the world, the law-likeness and constancy of the world, the astounding efficacy of mathematics to track what appear to be deep truths about how the world really is, and the contingency of the natural world. A review of all of these topics would exceed the scope of this article. I raise them primarily to demonstrate that, together with Harrison’s historical analysis, theological considerations may provide strong intellectual support for doing modern science. That is, it would be assuring for scientists and their supporters if the scientific enterprise seemed to be based upon good enough intellectual grounds. Coming from theology, such assurances go part of the way towards overcoming the expense and unnaturalness of science.

Objection. It appears that a particular cluster of theological commitments helped give rise to and may continue to provide substantial intellectual support for modern scientific inquiry. Do such theological commitments remain valuable, let alone needed, for contemporary scientists, scientific institutions, and their efforts? Even if these various positions are consistent with theological claims and arose out of theological reflections, it does not follow that theological foundations are needed to hold these assumptions about how the world is, how human minds work, and how knowledge is formed about the natural world as a consequence. Even more strongly, it may be that, similar to how early modern thought supplanted more Aristotelian medieval thought, leading to an improvement in how to learn about the natural world, scientific efforts were further improved by divorcing the metaphysical grounds of modern science from their theological foundations.

Response. If it remains true that thought about one or more gods is broadly natural for humans living in communities, as many cognitive and evolutionary scientists of religion have argued, (e.g., see Boyer 2001; McCauley 2011), and if the existence of gods and similar “ultimate concerns” are deeply motivating for individuals and their communities, as psychologists of religion have argued (e.g., Emmons 1999), then theological reflections are likely to remain important resources to fortify the intellectual grounds for investing in and doing science. Though many contemporary scientists may not see a need to locate their intellectual foundations about human minds and the world within a theological framework, it does not follow that theological sources are not continuing to form the cultural backdrop that make confidence in the sciences seem more intuitive than it may actually be.

My central thesis is not that science needs one particular theology, such as that held by the pioneers in the sixteenth and seventeenth centuries. It may be that other theological perspectives and commitments serve to intellectually support scientific inquiry as well or even better. What Harrison’s analysis suggests is that if the theological moorings change substantially enough (e.g., from an Aristotelian-inflected theology to a more Reformed theology), the resulting systematic study of nature changes from one sort of thing to another. Premodern science changed, over time, into what we now recognize as science. If the theological or philosophical moorings change substantially enough again, we may have something different still. I have no position regarding whether postmodern “science” or post-scientific exploration of the natural world needs theology. More importantly, it is not clear whether such repositioning—such as adopting methodological naturalism—actually constitutes an improvement in learning about the natural world (Harris 2026). We cannot know with certainty, because the sciences have continued their work in societies importantly shaped by theologically derived values and populated by people holding theological commitments.²

Nonetheless, Plantinga plausibly argues that many of the foundational assumptions of contemporary science do not merely find their historical roots in theology but are more coherent when situated within certain theologies. For instance, concerning the fit between the world and our cognitive faculties, Plantinga (2011, 269–70) observes (with the help of the pioneering cognitive scientist and linguist Noam Chomsky):

[T]here is a match between our cognitive or intellectual faculties and reality, thought of as including whatever exists, a match that enables us to know something, indeed a great deal, about the world—and also about ourselves and God himself. According to Noam Chomsky, “This partial congruence between the truth about the world and what the human science-forming capacity produced at a given moment yields science. Notice that it is just blind luck if the human science-forming capacity, a particular component of the human biological endowment, happens to yield a result that conforms more or less to the truth about the world.” From the point of view of theistic religion, this is not blind luck. It is only to be expected.

Plantinga (and Chomsky, it seems) take for granted a sort of intelligibility regarding the natural world when it comes to science. That is, we may take for granted that the world exists independently of our thoughts about it, and yet, out thoughts can grasp, however imperfectly, truths about that reality. Plantinga can locate this intelligibility in a creation theology.³ Chomsky can only regard it as a happy accident. It seems plausible that a creation theology is a more secure support for science than counting on blind luck.

Though coherence is not strictly necessary from a psychological perspective, blatant incoherence in an intellectual domain is unlikely to be helpful. Plantinga argues that many of the foundational assumptions of modern science sit as or more comfortably in some kind of theistic framework than others. Perhaps most famously, he argues that one particular assumption—that human reason of the sort that generates scientific theories is generally truth aimed—is not just coherent with theism but inconsistent with naturalism when combined with contemporary evolutionary theory. He argues that naturalism coupled with human evolution by natural selection does not yield confidence that our reasoning faculties can be trusted to be largely truth aimed. To simplify the argument, without some notion of a creator God who brought about humans with truth-aimed reasoning faculties, natural selection provides no solid grounds for thinking that our reasoning faculties are truth aimed, only fitness enhancing. After all, natural selection is blind to conceptual states; it only cares about adaptive behaviors. Even if we grant the contentious claim that conscious, reflective conceptual states are causally potent, plenty of evidence exists for a decoupling of truth-aimed mental states—particularly, higher-order reasoning—and adaptive behaviors. For instance, reasoning could be tuned to avoid catastrophic behavioral mistakes instead, known as “error management,” which may include tuning toward false, but not maladaptive, beliefs (McKay and Dennett 2009). Or it may have evolved primarily to persuade others and generate consensus (Mercier and Sperber 2017). Even if the naturalist can construct a compelling argument that in order for reflective reasoning about mundane aspects of life to be evolutionarily adaptive, it must be at least somewhat truth aimed, it does not follow that the abstracted reasoning of philosophers and scientists need be adaptive. Indeed, McCauley’s point about the cognitive unnaturalness of science is that it supplants or improves upon cognitive systems that, apparently, lead us to less trustworthy belief states because those cognitive mechanisms were designed by natural selection for other purposes. Consequently, under natural selection without God, we have little reason to trust our abstract, scientific reasoning, including our reasoning that led to a belief in natural selection (Barrett 2018, 162–64).

Objection. Even setting aside Plantinga’s particular argument regarding evolution and our reasoning faculty, suppose that Plantinga is correct, following Alfred North Whitehead, that “[f]or science to be successful, the world must display a high degree of regularity and predictability” (Plantinga 2011, 271) and, “[f]urthermore, science requires more than regularity: it also requires our implicitly believing or assuming that the world is regular in this way” (Plantinga 271–72). Could it be that this assumption, and many others Plantinga points to as critical for doing science, are ordinary tacit assumptions that people make as part of just being humans? If so, it may seem that we do not need theology to provide us with these assumptions. It could be that as long as scientists do not too critically question their science-supporting assumptions or find inconsistencies or tensions with their broader belief and value systems, all is well.

Indeed, the sciences could ride the momentum of their intellectual ancestry for as long as certain assumptions and values remain part of the cultural fabric that produces scientists and supports scientific enterprises. If these viewpoints are part of ordinary human cognition—independent of the reflections of philosophers and theologians—all the better. Some of the foundational assumptions Plantinga points to seem to be of this sort. We need reasons to think the world is not real, cannot really be grasped (at least to some extent) by human reason, and is unreliable or irregular rather than reasons to think these things. Even if, strictly speaking, such positions would benefit from a broader metaphysical framework in which they are nurtured, practically speaking, scientists can just keep doing what they do.

Response. Or can they? The truth of the matter is likely to be complex. Some of these assumptions will be safe when not explicitly challenged (e.g., that there is a real world we can at least begin to understand), but others may not be part of “natural” cognition at all. They may seem natural but are instead the product of a certain modern enculturation that itself is historically peculiar and open to change. For instance, the idea that the mathematical patterns derived from experiments in one time and place safely generalize because of the stability and predictability of the world is almost certainly the sort of thing that needs to be taught or acquired through enculturation. These have not been universally held perspectives, nor easily derived from common interactions with the world.

It may be natural and intuitive to think that yesterday, today, and tomorrow are similar to each other and that what is true about how things go in my neighborhood is also true on the other side of the village, but these uniformity, stability, and predictability assumptions may fail to apply beyond very local extensions. Why not assume that the world is basically chaotic and changing with islands of stability instead of the other way around? Stories about how things used to be (but are not now), unpredictable calamities, chaos in weather, wild variation in the success of hunts or harvests, and the oddities of distant peoples are the ordinary stuff of human existence and evidence that we do not simply take broad uniformity for granted. It is easy to be impressed with the variability, unpredictability, and seemingly chaotic nature of the behavior of humans, animals, weather systems, and ecological dynamics. Outside temporally, physically, and culturally local environments, uniformity is not at all obvious. We seek to find uniformity in natural patterns hidden behind apparent complexity, change, and chaos when motivated and tutored to do so. The idea that a divine being providentially holds back the chaos or has created the cosmos with deep regularities that may be hard to see may be fruitful support.

In this vein, McCauley’s point about modern science is that even though it does make use of some natural cognitive inclinations, as a whole, it is radically unnatural and characterized by counterintuitive ways of thinking. Its operating assumptions require at least periodic refreshment from thinkers who are trying to preserve and fortify its strengths and are committed to its wellbeing. Learning to be a scientist requires special training, including reflections upon why scientists do what they do and think how they think. If this training has shaky intellectual grounds, it will be less persuasive and more subject to competition from non- or anti-scientific pursuits. It is not hard to find thoughtful, educated people in universities around the world who are far from impressed with modern science, in part because they reject many of the assumptions that seem necessary to its character.

Methodological Naturalism and Methodological Providentialism

Though not an essential feature of modern science, one McCauley (2011) identifies as pushing modern science in the direction of being cognitively “unnatural” is a growing resistance against explanations that appeal to design, purpose, and especially intentions of minded agents. From early in childhood, people readily make sense of the world around them through appeals to the aims, desires, perceptions, beliefs, and actions of intentional beings, including humans, many animals, and gods (Barrett 2012). Resisting such intuitive explanatory strategies requires motivation and diligence. In contemporary sciences, this effort is fortified by the doctrine of methodological naturalism (MN).

Plantinga, following Bas van Fraassen, identifies three features of MN. First, the scope or model of the data to be explained relevant to scientific inquiry will not include supernatural agents (including God) “or employ what one knows or thinks one knows by way of revelation” (Plantinga 2011, 172). Second, an explanatory scientific theory should “not to include reference to God or any other supernatural agents . . . and the theory, like the data set, also can’t employ what one knows or things one knows by way of revelation” (Plantinga 2011, 172). Third, the evidence base that can be drawn upon to decide whether a theory or explanation is plausible “will not contain propositions obviously entailing the existence of God (or other supernatural beings); nor will it include propositions one knows or thinks one knows by way of revelation” (Plantinga 2011, 173). Plantinga goes on to note that there are stronger and weaker forms of MN. The weaker form is simply silent about supernatural entities; the stronger form includes the denial of their existence for the sake of the scientific enterprise.

In the broad sense that I have been using the term “theology” here, MN amounts to a theological position. To guide scientific inquiry, MN must specify what constitutes the natural and distinguish that from the non- or supernatural. If I do not have a theory about the supernatural, I cannot identify when and where it is causally potent and relevant for a particular question. This theorizing is theological in that it concerns the supernatural, divinity, and the like in relation to the rest of the world. In practice, MN amounts to the proposition that our understanding of the world is best done by presuming that, for the sake of generating sufficient explanations, the activity of God or other divinities or supernatural beings and forces are causally irrelevant. They do not generate relevant data, nor help account (enough) for patterns observed. If theology includes how all things—including their nature, purpose, and flourishing—relate to divinity, then the claim that divinity is explanatorily irrelevant to the nature, purpose, and flourishing of the material stuff of this world is a theological claim. Similarly, judging that God or gods exists or not or that they are causally inert or irrelevant are (a)theological judgments.⁴

As much as many contemporary scientists might recoil at founders of modern science finding intellectual support in theology, what we now call the “scientists” of the sixteenth through early nineteenth centuries would probably find MN shocking and superfluous (or worse). But this tension can be resolved if we recognize the considerable overlap between the constituent features of the earlier theological viewpoints and the contemporary atheological ones. For lack of a better term, I call the deliverance of the earlier theological commitments methodological providentialism (MP), and it includes the following elements:

• The world exists independent of human minds and is contingent and so must be explored empirically.

• It is law-like and so relatively constant, reliable, regular, and predictable in its deep, nonobvious structures and causal dynamics.

• Importantly, exceptional supernatural action (by God or other agents) will not fundamentally occlude these regularities and so does not play a prominent role in describing how the natural world typically works, or the proximate causes of events.

• The typical operations of the world are describable mathematically with accuracy (not merely useful models).

• Humans have a truth-aimed intellect that fits the world, which enables induction and learning from experience.

• But human intellectual powers are limited and distorted by ignoble aims, which requires special socially distributed practices and mechanisms for reducing or managing these limitations and distortions, through which we can remain optimistic in improving our understanding the world.

Importantly, this cluster of assumptions is largely shared by practicing scientists today, at least tacitly, not just by those pioneering natural philosophers who laid the foundations for modern science. And yet, these assumptions are far from universally affirmed by thinkers and scholars historically or even across universities around the world today.

These assumptions do, however, find a cozy home in Christian and neighboring theologies that include some key components: a creating and sustaining deity who freely created and ordered the world with law-like regularity, predictability, and deep, stable structures, has given humans truth-aimed conceptual systems but the freedom to become corrupted and mis-aimed in our thought, and exercises enough sovereignty over the created world that neither humans nor other intentional beings and forces (natural or otherwise) can fundamentally change these regularities. To sum, what is required is an assumption of divine providence: that God freely gives and constantly preserves a stable, orderly creation and gives humans the gift to accurately—even if in limited ways—grasp truths about that creation.

From a Christian perspective, the very reason we can explore the natural world as if it is regular, law-like, and predictable is because God causes those regularities to persist through God’s providence. As the Heidelberg Catechism answers question 27:

Providence is the almighty and ever present power of God by which he upholds, as with his hand, heaven and earth and all creatures, and so rules them that leaf and blade, rain and drought, fruitful and lean years, food and drink, health and sickness, prosperity and poverty—all things, in fact, come to us not by chance but from his fatherly hand.

God is not in the business of changing creation willy nilly or arbitrarily, or tinkering here and there in irregular and unusual ways without good reason. This is God’s world, not a world governed by lesser deities monkeying with nature and making datasets largely unpredictable. These theological commitments mean that we need not constantly appeal to new divine action or the work of spirits and lesser gods to account for features and processes of the natural world. Nature seems to have its own independent dignity precisely because God constantly gifts nature with that dignity. So, what looks like MN to an atheist or agnostic looks like MP to a theist who believes in such a creator God.

It just so happens that the theistic notion of divine providence in these ways maps onto the “blind luck” (to quote Chomsky) of MN. In many topical spaces in which one was already unlikely to regard supernatural beings (including God) as part of the data to be explained, as relevant proximal causal agents, or as part of the immediately relevant database, whether one adopts MN or MP will yield very similar results. If, for instance, I want to examine how de-icing salt on roads impacts macroinvertebrate diversity downstream from bridges, I am not investigating the presence of spirits or angels and am unlikely to invoke their activity in explaining findings; whether or not they exist is not part of the relevant evidence base that might make my explanations more or less plausible. I am likely to stick to an explanation that involves how the chemical properties of road salt change rivers’ chemical properties, which may lead to changes in the habitat suitability, fertility, and fecundity of the macroinvertebrates that live in those rivers and streams. The situation is not unlike a scientist who fully believes in dark matter but has no need to bring it into their current study of macroinvertebrate diversity. Such a scientist may be practicing methodological a-dark-matterism.

Though grounded in a high view of God’s providence, MP could be adopted by a non-theist, much like how supernaturalists adopt MN even if they reject metaphysical naturalism. It is methodological too for theists who think that God and God’s agents (or demonic powers) do occasionally act in special ways in the created world. For the sake of typical scientific inquiry, unusual intervention by spiritual beings is demoted in explanatory priority.

MP is not just comparable to MN, it is better. First, it is more flexible. Like MN, MP expects nature to be causally regular, without arbitrary and capricious monkeying by supernatural beings, but MP leaves open the possibility of special divine action in theologically understandable cases. Methodologically, both MN and MP would treat unusual causal events as initially just noise in the data, but MP would have motivation to track this “noise” and not ignore it. For instance, miraculous healings are reported all over the world. MN requires that they be ignored or even discarded as part of scientific practice, leading to an impoverished dataset or (unfair?) claims about the reliability or honesty of the sources of these reports. MP can accept that they are possibly genuine miracles without having to discard ordinary natural causal pathways.

Second, MP is more coherent than MN in relation to background assumptions concerning our ability to form true beliefs about the natural world. As Plantinga argues, naturalism gives very little reason to expect that our science-relevant reasoning is truth aimed. So, if the sciences are an epistemology, and MN is an assumption of that epistemology, this epistemology has an assumption that actually weakens our confidence in its success, particularly the farther we explore questions not plausibly related to human fitness (Barrett 2018). If we have no strong reason to think human reflective thought—the kind of thing scientists do—is part of our make up in order to give us true beliefs, then why put so much confidence in it? Under MP, we have no such worries. MP claims that our minds are created and sustained by a creator God who has equipped us to seek and grasp truths, even if imperfectly.

A third virtue of MP over MN is that it creates space for sciences of minds and other features of creation that are difficult to classify in terms of being natural versus non- or supernatural. In his critique of restricted MN, Joshua Lee Harris (2026) describes how MN has struggled with an intelligibility problem in that no consistent, principled, and satisfactory way of demarcating the natural from supernatural is has emerged. What is “natural” may turn into nothing more than a crude heuristic concerning those things that are nonphysical, hard to pin down ontologically, or just a bit spooky by contemporary standards of a given era. So, for instance, the behavioristic rejection of mental states as a legitimate area of psychological science in the early twentieth century arose in part because it seemed to fit better with MN:

To venture a historical generalization, we might say that behaviorism was behaviorism just to the extent that it was naturalistic in spirit—with increasingly austere positivity and eventually Quinean banishments of the “non-physical” (often a euphemism for the super-natural) from the explanatory domain entirely. (Harris 2026, 33)

That is, in this case at least, a certain variation of MN proved to be too restrictive, rejecting mental states as legitimate areas of scientific inquiry. In primarily drawing an ontological line between God and Creation instead of ambiguously between “nature” versus all else, MP is more open to investigating created-but-not-physical entities such as percepts, thoughts, emotions, and consciousness.

MP also has resources for de-centering the sciences from selfish human aims that MN does not possess. If one approaches the sciences with the provisional operating assumption that the world is productively investigable because of some kind of divine providence, then one may be more likely to receive the potential of the sciences with gratitude instead of proprietary entitlement. Assuming some kind of divine benefactor, I may wonder for what purposes that benefactor has gifted humans, among the creatures, the ability to discern deep patterns and law-like regularities with some fidelity. Perhaps it is for divine aims and not merely for technological achievements or to satisfy human curiosity. I turn to these considerations more directly in the sections that follow.

To sum, because doing modern science is cognitively unnatural, it requires special conditions for its pursuit. In addition to using careful quantification and research instruments as cognitive and perceptual prosthetics for individuals, science is also socially distributed and regulated: the cognitive shortcomings of individuals are partially remediated through these social mechanisms (i.e., public reporting, data sharing, insistence on replicability, peer review, etc.). Special research methods such as experimentation or use of statistical control improve the ability for scientists to draw reasonable inferences from their observations. This whole complex of activities, however, requires a certain amount of intellectual support to maintain investments of individuals and institutions. Originally, this intellectual support came from theology. The key bundle of theologically derived assumptions may be called MP, but they are more flexible and expansive than today’s MN, though they closely parallel.

Even if one is unpersuaded regarding MP’s ability to intellectually support the sciences and prefers MN, MN itself bears many of the marks of a theological position. It generates normative claims about how one should or should not go about participating in a highly influential and authoritative form of intellectual inquiry that demands massive amounts of public support in terms of educational institutions and financial investment, and these claims are derived from attempts to specify what is or is not supernatural and whether their properties or activities are of the sort that they should be included as relevant to the public exercise of this form of intellectual inquiry we call the sciences. Such considerations are claimed by philosophers, but they are simultaneously theological. Theology of one sort or another then has played a key role in providing intellectual support for scientists and the scientific enterprise more broadly.

Much more briefly, I turn to two more forms of support that theology gives in keeping science healthy and productive: motivational support and directional support. I regard these as perhaps more important than the intellectual support already discussed.

Whose Directions?

Motivating science by way of the goods it may (or may not) bring about introduces the fact that such motivations concern particular directions or aims for specific scientific enterprises. If you are motivated to cure disease because your worldview regards disease curing as an important good, such motivation will likely encourage enthusiasm for various biological sciences more than for astronomy, theoretical physics, or geology. Depending upon one’s values, only some scientific endeavors may be regarded as worthy of investment.

Utilitarian approaches to the sciences will run into this trouble of particularities: not all areas of scientific research are plausibly worth investment because of the goods they may bring about. Many practical problems facing humanity or the world more generally could be more efficiently and effectively addressed using nonscientific means (e.g., trial and error, analogically reasoning from neighboring problem spaces, local expertise, etc.). Furthermore, some scientific areas will have as much or more potential to bring about ills than goods, or the “goods” they bring about are narrowly aimed back to the scientists and technologists themselves. Pharmaceutical companies, military-industrial complexes, and tech-entertainment conglomerates may generate lots of scientific work for which the primary aims are very provincial “goods.” Less dramatically, individual scientists, lacking a sound and salient theology, may be driven to conduct research aimed at self-promotion or professional security instead of some higher, sacred good.

Science needs motivational and directional support to keep it healthy. Specifically, it needs compelling reasons for individuals to go into the sciences, for societies to equip their children with the extensive prerequisite educations for going into the sciences, and for the public or wealthy citizens to financially support scientific institutions and activities. The needed motivation will also interact with the questions and problems the sciences are directed towards. None of these motivations or directions are intrinsic to what modern science is but must be derived from other values and commitments. Whence these values and commitments? From reflections on what we hold to be ultimate concerns, sacred principles, the nature of people, and the aims of lives worth living. In short, the values and commitments that will motivate and direct the sciences will come from theologies and theology-adjacent philosophical ruminations.

The trouble with pretending that science is autonomous of theological considerations is that scientists and scientific institutions will merely enshrine their own theologies—perhaps deifying “discovery,” “progress,” “knowledge for its own sake,” or even “Science” itself—as ultimate goods that demand the allegiance of all who do science. Such sacralization may be tacit or effervesce passively (at first) from more localized efforts to curry popular favor and support. Nonetheless, it is not hard to see that the authority of science can be appealed to regarding not just its epistemological strengths but its particular aims. Resistance is bullied as “anti-science” and ignorant. The power of government is used to take resources from citizens to support “Science,” whatever It decides to do, because its aims are unquestionably good.

Unchecked by broader considerations—theological and philosophical ones—scientists collectively risk producing their own science- and self-serving theology: whatever science does is good and worth public support, even borrowing against our children’s futures. From such a viewpoint, all knowledge is good in and of itself, even if those who possess it lack the wisdom to steward it well. Surely this is not a rightly aimed condition for science.

Science Stewardship

To summarize the preceding reflections, I suggest that what modern science requires for its enduring thriving is a theological grounding I term science stewardship. Requisite components of science stewardship are to see the sciences as good tools that must be cared for and employed to virtuous ends. They cannot be taken for granted or left to generate their own morals and purposes—a condition identifiable as scientism.⁵ A Christian version of science stewardship would be to view modern science as a good gift from God that, like other cultural developments, needs to be cared for and deployed judiciously toward right ends. As God, in the Genesis account, commanded Adam and Eve to care for the garden (an artifact or cultural gift), science too is a cultural gift, a set of tools, for better understanding ourselves and the world around us. Used well, it enables us to better serve as God’s representatives or stewards of God’s creation, in part through better discerning God’s wisdom in the world—its patterns and purposes—and in part by generating interventions and technologies that should be aimed at promoting God’s purposes for the world as we best understand them.

This emphasis on God’s purposes is consonant with MP. MP posits that our ability to successfully discern deep causal structures is due to some kind of divine providence. The world is not ours but God’s, and our ability to successfully explore it is a gift. From that perspective, it is easy to wonder why God has gifted humans, among the rest of the creatures, in this way. Does God have purposes for our scientific enterprises that are not merely to satisfy human curiosity or desires? Theological motivations for doing the sciences may be particularly apt for refocusing scientific endeavors away from selfish and human-centered aims.

The health and goodness of scientific work also benefit from certain personal virtues of its practitioners. McCauley (2011, 275) observes that modern science “requires dedication, honesty, and lots of hard work.” Additional virtues include “humility, open-mindedness, provisional trust of experts, an inclination to seek better evidence, recognition of personal blind spots, and a willingness to remedy these blind spots” (Barrett, 2025, 45). Within a theology that commends such virtues, as well as righteous ends, the sciences become powerful tools for good.

Nonetheless, in actual practice, scientists and the scientific community, like any other culture, can fall victim to vices. Curiosity put to virtuous aims can slip into curiosity for self-gratification, prestige, or avoidance of responsible action . . . Much scientific work is driven by the hope of gaining power and control . . . As many sci-fi books and movies such as Ex Machina and The Terminator teach us, because the sciences have the potential to increase power, they should not be taken lightly. (Barrett, 2025, 45–46)

Part of not taking the sciences lightly is to insist that they are surrounded, supported, and guided by sound theology.

Not All Theologies Will Do

My argument is that doing the work of participating in and caring for the cultural complex we call “science” will not happen without sufficient intellectual and motivational support, and this support is best located in theology. But clearly not just any theology will do. The preceding analysis included historical observations of how the shift away from some aspects of medieval theologies to early modern and Reformation theologies sparked the scientific revolution. Similarly, my point in rehearsing what I see as the components of MP and science stewardship is to identify just which sorts of theological claims are supportive of scientific inquiry and institutions. Implicit in such an account is that not all theological positions are equally fertile soil for science. This was true in the sixteenth century and remains true today. For instance, a theology that suggests that realities of the world are largely inscrutable, or one that asserts that we each have our own individual, equally good “science” and idiosyncratic constructions of the world, would hardly drive its adherents to invest in the modern sciences as we know them.

Open Questions

As mentioned at the outset, science is a distributed enterprise undertaken by individual people sometimes working alone and often working collaboratively. As rhetorically efficient as it may be to discuss whether science needs theology in the abstract, ultimately, better questions would be more specific. They must concern the degree to which people doing or supporting the sciences in some sense need particular theological considerations to motivate, sustain, or inflect their efforts. Though I have argued that science needs theology, perhaps the more interesting questions remain open. These questions concern the degree to which theology is needed for individual scientists and the various communities of scientists.

It may be that we need some, but not a majority of, scientists to be motivated and instructed by theological commitments. The situation may be a bit like claiming that rock ‘n’ roll requires a drumbeat and therefore requires drummers. But not all or even a majority of rock ‘n’ roll musicians have to be drummers or even like the drummers in their bands. Science requires theology, but not all scientists have to do or even appreciate theological reflection. The distributed nature of the scientific enterprise means that, once instituted, it can keep running without reflection on its foundational commitments, at least for a while. (But how long?)

The analogy with music suggests a related question: Could it be that particular sciences are more in need of theological support than others? Drummers are necessary for rock ‘n’ roll but not bluegrass or chamber music. Marching band music requires a prominent role for percussion. Perhaps some areas of the sciences are more or less dependent upon theological contributions, for instance, to keep their activities properly aimed, to fortify the waning support of the public, or to help scientists re-attend to necessary intellectual virtues.

So, even if science needs theology in the general sense I have presented here, it is reasonable to wonder: What proportion of scientists, and/or the general public from which those scientists draw support, needs to have the right sort of theological convictions, to what degree of salience, and in which particular spheres of the sciences? This complex of questions is surely one that will benefit from historical, philosophical, and even social–scientific analyses. These are, at least in part, empirical questions.

My argument also presumes that some degree of coherence between one’s theological positions and the assumptions behind one’s scientific efforts at least occasionally matter. That is, a theological or atheological foundation that does not cohere with how the sciences are done will lead to tension, demotivation, or distortion. It may be, however, that in the context of either doing or supporting the sciences, a fair amount of incoherence has little impact on the wellbeing of scientific enterprises. Perhaps I have overestimated the psychological harm of incoherence, or the degree of incoherence is what matters. For instance, even if MP is a stronger, more coherent foundation for doing science than MN, the relative weaknesses may not be especially harmful to the collective enterprise.⁶ But there may still be a degree of incoherence that, if reached, will prove fatal to modern science. What is that degree?

A Challenge

A reasonable person may object that what I have argued for is that science needs philosophy. Especially when it comes to intellectual support, the commitments that underlie scientific efforts are the sorts of things philosophers of science talk about. I do not object to the claim that science needs philosophy. Adjudicating the boundaries of philosophy and theology is not my interest. Much like how the boundaries around “science” are the result of historical and cultural particularities, the purported boundaries around philosophy and theology are not fixed, nor tracking some immutable essences that govern these areas of inquiry.

For the Christian or Jewish theist, the love of wisdom (philosophy) springs from right understanding and relationship with God, a lived theology. “Fear of the Lord is the beginning of wisdom,” after all (Proverbs 1, 2, 9; Psalms 111). There are certainly contemporary descriptive distinctions between what philosophers do and what theologians typically do. My point is that philosophers cannot claim undisputed sovereignty over the love and pursuit of wisdom, understanding, and knowledge. Even today, the culturally constructed boundaries between theology and philosophy are porous. Moreover, given that the commitments I suggest are needed for the wellbeing of science concerns claims about divine/supernatural entities (e.g., whether or not they uphold or interfere with the causal dynamics of the world, where to draw the line between nature and super-nature) and the aims and values one holds preeminent or sacred, these strike me as very much theological concerns, even if they are also philosophical. For instance, Plantinga’s reflections I have cited here are simultaneously Christian philosophy and theology.

Whether theologians have risen to the task of developing convincing and compelling treatments of the scientific enterprises is another matter. If I am right and science needs the right theology, then theological scholars who value the sciences need to develop and disseminate an adequate theology of science. As scientist-turned-theologian Alister McGrath (2022, 66–67) writes:

To simplify a complex argument . . . science helps us understand how things work; theology helps us grasp what things mean. We need to hold these together, seeing them as integral elements of a “Big Picture” of reality. And that, in my view, is what theology helps us to do – to recapture the coherence and unity of our world, which is so easily lost, compromised or obscured by the rise of disciplinary specificity on the one hand, and reductionist approaches to nature on the other.

McGrath’s picture of the relationship between theology and science is not one in which the scientific enterprise goes about its business, generating new findings, problems, theories, and technologies, and then theology integrates those findings into a “‘Big Picture’ of reality.” To the degree that the scientific enterprise is part of the “Big Picture” itself—not merely its products but its people, relationships, institutions, sources of material support, and aims—it too is subject to theology. Just as theological reflections on the family, work, government, and the arts are legitimate and potentially valuable guides to the health of these aspects of human experience, so too is the theology of science for properly encouraging the thriving of science.

As much as theological scholars are needed to generate and communicate adequate theologies, these theological contributions will not penetrate the sciences as practiced passively. Those who work in the sciences must mindfully and judiciously bring their theology into their scientific work. But are we willing to consciously and conscientiously practice MP and science stewardship, or retreat to passive acceptance of MN and allow underexamined motives to direct the sciences?