I was honored and delighted to be invited to contribute to this special issue marking the retirement of Alister McGrath. As readers of this journal well know, for many years Alister has been a leading figure in the science‐and‐religion field, and while his remarkable expertise spans a much wider range of concerns, this field owes him a special debt. On a more personal note, Alister was a colleague and close neighbor during my time at Oxford—we were the sole occupants of rooms in the tower above the dining hall at Harris Manchester college. He was a source of sage advice during that period, and I remain grateful for his guidance in navigating the numerous idiosyncrasies of the Oxford system (if “system” is the right word). It was also Alister who invited me to participate in a 2008 Conference on “Beyond Paley: Renewing the vision for Natural Theology,” an occasion memorable not only for the quality of the discussion but also on account of the atmospheric venue: Oxford's Museum of Natural History. I distinctly recall from that event Alister's nuanced paper on William Paley which led me to reconsider some of the standard prejudices that I had long cherished against what I imagined to be Paley‐style natural theology. That occasion also afforded me the opportunity to reflect on the close historical connexions between science‐and‐religion and natural theology which has been an interest ever since and which brings me to the focus of this article.

The brief I was offered was the theme of natural theology. Natural theology is often identified as the key site of Christian theology's interaction with the natural sciences. It is a topic to which McGrath (shifting register here from encomium and memoir to academic commentary) has given long consideration, the fruits of which may be found in four monographs: The Open Secret: A New Vision for Natural Theology (2008); A Fine‐Tuned Universe: The Quest for God in Science and Theology (2009); Darwinism and the Divine: Evolutionary Thought and Natural Theology (2011) and, most recently, Re‐Imagining Nature: the Promise of Christian Natural Theology (2017). I should also add that this series of works does not represent minor variations on the same theme but illustrates both the development of McGrath's thinking on this challenging topic and, in the case of the 2009 and 2011 books, the specific application of his general approach to the physical and biological sciences, respectively. I thought to use this occasion to offer some reflections on the general category of natural theology, taking as a point of departure the central arguments of Re‐Imagining Nature (unattributed page numbers in parenthesis throughout will be to this work).

Let me begin by saying that I am very much in sympathy with the substantive aims of McGrath's conception of a Christian natural theology. The recurring theme of Re‐Imagining Nature is that we must challenge “the established conception” of natural theology that understands it exclusively as “the enterprise of providing support for religious beliefs by starting from premises that neither are nor presuppose any religious beliefs” (16, 181). Against this, McGrath offers a much broader conception: “Natural theology, as I understand the notion, is not about discovering persuasive grounds of faith outside the bounds and scope of revelation, but is rather a demonstration that, when the natural world is “seen” through the lens of the Christian revelation, the outcome is imaginatively compelling and rationally persuasive” (131). Elsewhere, the style of natural theology being advocated involves “ways of thinking, both rational and imaginative, that arise naturally within human minds on account of innate cognitive processes” (24); and again, “a certain way of thinking about the natural world and God, or a way of seeing the natural order, rather than any explanatory or normative theory about the world or God” (24, 128). On the negative side, McGrath thus eschews the idea of natural theology as a neutral philosophical enterprise (23) and proposes a broad conception of what is “natural” in natural theology. This takes in not just the deliverances of the natural sciences as if they were only authoritative sources for nature, but also includes a range of other considerations that range from the rational to the imaginative. All of this is commensurate with what we see in the relevant historical practices.

In the final pages of Re‐Imagining Nature, McGrath articulates the worry that he may have overstated his criticisms of the established conception of natural theology (p. 181). My worry is a slightly different one, and I wonder rather whether his criticisms have gone far enough. Following through on those criticisms would, in my view, lead to a conclusion slightly different to that drawn by McGrath. My suggestion would be that the putatively “established” conception of natural theology is actually incompatible with the version of natural theology that McGrath seeks to advance and is, in any case, largely absent from the tradition within which it is said to be established. This also leads me to wonder whether the expression “natural theology” can be meaningfully extended to the wide variety of activities that have been so labeled, and whether it might be time to consider abandoning it.

It would be hubristic to think that in the span of brief paper these issues could be settled beyond dispute. There is a vast literature on natural theology and its history (e.g., Muller 2003; Sudduth 2009; Brooke, Manning, and Watts 2013, pt 1), and this is not a field in which I can claim any special theological expertise. The plan, then, is to offer a select number of historical observations that support the general thrust of McGrath's position on natural theology, but which at the same time push it toward what I regard as its logical conclusion. In keeping with the readership of this journal, a particular emphasis will be the history of the relation of science to theology. Part of the argument will be that the ultimate incompatibility of different versions of natural theology is related to fundamental differences in how key concepts are understood: “nature,” “natural,” “reason,” and “rational.” The apparent stability of the relevant terminology disguises the shifting meanings that lie beneath the common vocabulary (although as we will see, the vocabulary also changes in ways that support diversity rather than unity). Once we have arrived at an understanding of the historical instability of these notions, the “established conception” of natural theology can be seen for what it really is: namely, a modern construct that bears little resemblance to anything that we encounter within the tradition itself. More generally, it appears that this conception has been constructed to suit the agendas of various constituencies, ranging from “insider” theological critics (such as Karl Barth) to “outsiders” (such as Deists, philosophers of religion, and scientifically motivated religious skeptics). This construct has been projected back onto the tradition, leading to the illusion that it was there all along.

Varieties of Natural Theology?

It is characteristic of much of McGrath's work that even in the context of setting out a new approach he is at the same time able to offer an authoritative overview of the topic at hand. This typically involves a helpful systematization and classification which is educative and illuminating. In the case of natural theology, the opening chapter of Re‐Imagining Nature sets out a number of different activities that might be, or have been, designated “natural theology” (18–21). The six themes are set out below:

1. “What human reason unaided by revelation can tell us concerning God.” This is theology that comes naturally to the human mind (and which I take to be restricted to a priori considerations).

2. Physicotheology. The demonstration of the existence of God from the regularity and complexity of the natural world. I take this to be distinguished from (1) by its a posteriori focus.

3. Intellectual outcome of a natural desire for God. (Think here of Aquinas's interior instinctu (Interior instinct), or Calvin's sensus divinitatis (sense of divinity)).

4. Explores an isomorphism between our reason and the structure of reality. This approach, characteristic of Joseph Butler and John Polkinghorne, has the more modest objective of establishing the possibility of congruences between specific claims of Christianity and what we know of the natural world from other sources.

5. Approaches that seek to show that purely naturalistic approaches to the world are deficient or in some way self‐defeating. (Alvin Plantinga would be exhibit A here.)

6. Natural theology as a “theology of nature,” reflecting a specifically Christian understanding of the natural world. Framed in terms of a doctrine of creation.

The first two senses combined come close to “the established conception” which, recall, involves “starting from premises that neither are nor presuppose any religious beliefs” (16). McGrath concedes that “there is no predetermined essential form of nature or natural theology’ and recognizes that the various forms that natural theology takes are critically dependent on context. Yet this is accompanied by a conviction that it is nonetheless possible to understand the variety of activities that might be labeled "natural theology" as aspects of “a single coherent project” (2). On the specific question of the “established” notion, which already looks like the odd one out, McGrath suggests that the Christian imaginarium can “accommodate and position this classical notion of natural theology” (36). I would suggest instead that while it can position it, it cannot easily accommodate it. In support of this contention, I propose a consideration of some of the historical manifestations of these themes, paying particular attention to presence (or absence) of the “established” conception.

Medieval Natural Theology?

It is significant that the term “natural theology” (physicam theologiae or theologia naturalis) was little used in the Middle Ages, even by those with whom it is most commonly associated. Augustine, for example, essentially followed the usage of Roman writer Marcus Terentius Varro (116–27 BC) who had referred to three kinds of theology: “natural theology” (physicum theologiae), “mythical theology” (essentially euhemerism, the worship of dead heroes), and “civil theology” (state‐sponsored worship of images) (City of God, VI.5; cf. Tertullian, Ad nationes II.ii, cf. McGrath 2017, 12–13). Augustine thus equated natural theology with pagan thinking about the gods. Centuries later, Thomas Aquinas would rehearse the same threefold division and refer to these forms of theology as instances of “superstitious idolatry” (Summa theologiae 2a2ae, 94, 1). This usage of “natural theology” persisted until well into the eighteenth century (e.g., Gale 1672, 111; Leng 1730).

The negative connotations of “natural theology” were not unrelated to the fact that “theology” itself tended to be regarded as a suspect term until the twelfth century, again on account of its pagan associations. (Plato had coined the term, and he and Aristotle had established its standard meaning.) In the early Middle Ages, theological reflection was typically designated sacra doctrina (or sacra pagina, or lectio divina). Peter Abelard (1079–142) pioneered the use of the term “theology,” initially against powerful objections. This resistance was partly motivated by the fact that what was on offer was not simply an innocent change of terminology. Influenced by the circulation of Aristotle's newly translated logical writings, Abelard's “theology” involved the introduction of dialectical reasoning into theological reflection and a move away from the more contemplative approaches that had characterized sacra doctrina and lectio divina. (The latter involved the dual contemplation of both scripture and nature, which were connected through the process of allegorical interpretation.) Advocates of the traditional contemplative approach, such as Bernard of Clairvaux, were bitterly opposed to what for them was the analytic, logic chopping “theology” of Abelard (Evans 1980; Brown 1990, 82–97). This, then, looks a little like an earlier version of Blaise Pascal's objections to “ the God of the philosophers”, Kierkegaard's fulminations against Hegelian‐inflected religious thought, and indeed of recently expressed reservations about “analytic theology.”

The controversy generated by Abelard's new approach bears directly on what theology of nature/natural theology might entail. Consider what is involved in the traditional practice of lectio divina. The premise is that God has invested the objects of the natural world with theological meanings. Through a process of contemplation, reading, and prayer, these meanings reveal themselves (Evans 1984; Harrison 2015, 57–63). The significations of natural objects, moreover, did not amount to what we would now regard as the conclusions of “established” natural theology—the existence of God and evidence of his power and wisdom. Rather, the contemplation of nature was thought to reveal the same truths as those encountered in scripture: the “book of scripture” guided the reading of “the book of nature,” and the two books communicated a common message. That “message” encompassed the mysteries of the Incarnation and the Trinity. Bonaventure (1221–74), for example, tells us that “the creature of the world is like a book in which the creative Trinity is reflected, represented, and written” (Breviloquium II.12). Recall, also, Augustine's earlier suggestions about ways in which Trinitarian patterns emerge from introspective reflections on the human soul (De Trinitate, passim).

On this view, when the Psalmist announces that “the heavens declare the glory of God’ (Ps 19:1), the proposal is not that the study of astronomy yields the premises of a rational argument from which can be inferred the existence of a deistic God; it is rather that God's presence is directly evident to those able to see the world in a particular way. (For Augustine's extended riffing on this passage, see Confessions XIII.15). That way of seeing the world, moreover, does not arise out of a dispassionate, naturalistic appraisal of the mechanics of nature, but is part and parcel of a formative practice or regimen of spiritual exercises involving prayer, meditation, and reading (cf. McGrath's pertinent discussion of habitus, 69–72). The parallel structures of nature and scripture were still being urged in Raymond of Sebonde's fifteenth‐century Liber Naturae sive Creaturarum to which the title theologia naturalis was appended in 1485 (cf. McGrath, 13–15).¹ For Raymond, we encounter salvific truths in both nature and scripture. The early medieval approach seems clearly discontinuous with the idea of a natural theology understood as “what human reason unaided by revelation can tell us concerning God.” It does, however, nicely comport with McGrath's own emphasis on “a certain way of thinking about the natural world and God, or a way of seeing the natural order, rather than any explanatory or normative theory about the world or God” (24). But the movement is from what is revealed to the natural world. This is entirely in keeping with the familiar maxims of Augustine and Anselm concerning the need to believe in order to understand (crede ut intellegas; credo ut intelligam), and the order of reasoning that proceeds from faith to understanding (fides quaerens intellectum).

The theological and moral orientation of premodern philosophy and natural philosophy are also significant in this context. To consider some brief examples, in the Timaeus, Plato proposed that study of the heavens resonates with the internal divine principle, renews human nature, and contributes to the life that the gods intend for mankind (90d). The second‐century astronomer Ptolemy also insisted in his Almagest—the most influential scientific text of the Middle Ages—that familiarity with the revolutions of the heavenly bodies reformed one's character and reduced it to a spiritual condition (Prologue). The Neoplatonic philosopher Simplicius similarly held that natural philosophy was to be pursued for moral and spiritual ends and that the study of “physics” was a ladder that led toward the knowledge of God (Harrison 2015, 26–34). The interaction of these traditions with Christianity would then not take place in some neutral philosophical space, since they were understood more as competing spiritual practices that shared some of Christianity's basic theological presuppositions. With some important qualifications, we can say that this model of a natural philosophy that was always and already theological persisted into the modern period, with Isaac Newton, for example, declaring that discourse of God was a part of natural philosophy (Newton 1934 [1713] 546). The idea of neutral or nonreligious premises did not enter into it.

It has been argued that from about the thirteenth century onward the idea of neutral God‐talk, or notions of a purely rational preamble to revealed theology got under way in earnest. The move from a more contemplative lectio divina to a more dialectical theologia gives some credence to this view. Thomas Aquinas is thus often identified as the progenitor of natural theology in this sense (e.g., Barbour 1998, 6–8, 98–99; Kretzmann 1998). However, the activity that Aquinas undertakes in the “five ways” or the Summa contra gentiles is not that of attempting to reach God from nonreligious premises. Consistent with what has already been said about Plato and Ptolemy, Aquinas tells us that “almost all philosophy is directed towards the knowledge of God“(Summa contra gentiles, I.4.3, vol. 1, p. 67). To begin with the shared premises of Jewish, Greek, and Islamic philosophers, then, are already to have a religious starting point. Accordingly, the aim of the exercise is not to establish the existence of God and some of his attributes on the basis of reason alone in order to provide a foundation for revealed theology. Rather, it is to show how Christianity succeeds in fulfilling the unrealized goals of Judaism and Islam, along with the inchoate religious aspirations of the philosophical schools. As Stanley Hauerwas and others have suggested, modern natural theology (i.e. in the ‘established’ sense) is not an enterprise that Aquinas would have recognized (2001, 23; cf. Kaspar 1984, 78; Wolterstorff 1986, 39; Hibbs 1995; Moore 2003, 31–3). In any case, the “without recourse to divine revelation” condition is somewhat confounded by Aquinas's assumption that it is by faith that we know that God's existence can be demonstrated (i.e., on the basis of Rom. 1:18‐20, Summa theologiae 1a. 2, 2).

The Age of Physicotheology

It may seem that a much stronger case can be made for the emergence of natural theology in the “established” sense in the early modern period. The seventeenth and eighteenth centuries were, after all, the golden age of the design argument (or “physicotheology,” although as we will see this is not quite the same thing) and we encounter frequent references to the metaphor of God as the divine watchmaker during this period. With the emergence of modern experimental natural philosophy and the gradual growth in its prestige, a neutral, objective “science” might now be thought of as providing raw materials for natural theological speculation and providing it with a secure foundation. Again, however, it is not quite as straightforward as this. Amos Funkenstein aptly referred to early modern natural philosophers as “secular theologians” (1986, 4–10).

This is because the new science or natural philosophy was built upon theistic foundations. Equally importantly, it also needed religious legitimation. The context of the new sciences thus brings a different complexion to natural theology. It is tempting to regard this “secular theology” as an exercise in Christian apologetics. But arguably it was needed more by the sciences to establish their metaphysical foundations and secure their religious and social respectability. The new science was vulnerable both on account of its novelty and because it drew upon on an Epicurean theory of matter which had long been associated with materialistic atheism. Its exponents thus found it necessary to explain why the new physics was not incipiently atheistic. Their efforts in “secular theology” were directed partly toward demonstrating the theological respectability of their scientific endeavors.

The disciplinary configuration known as “physicotheology” appears superficially to resemble “established” natural theology, and the abstract notion of “the physicotheological argument” (an expression belatedly coined, as far as I can tell, by Immanuel Kant) leads to a natural assumption that physicotheology involved the constant repetition of an argument from design, furnished with an evidence base drawn from the dispassionate, scientific reading of nature. Closer examination of the relevant practices reveals a different picture. For a start, physicotheology initially included topics that fell within the scope of what we regard as revealed theology. Robert Boyle authored a short work on the physicotheology of resurrection, while the biblical Deluge, transubstantiation, and the virgin birth were all physicotheological topics (Blair and von Greyerz 2020). This blurring of the boundaries between natural and revealed may also be found in Johann Alsted's earlier Theologia Naturalis (1615), which suggests that there is a “certain image” of revealed mysteries in nature which is declared rather than demonstrated. For Alsted, moreover, we are natural theologians “by the grace of God” (12).

The stated goals of physicotheology also included the promotion of personal piety and morality. For some, this was because natural theology could be understood in terms of the older Aristotelian categories which held “science” to be an intellectual virtue and related it to human ends to be perfected through habits (Harrison 2015, 11–16). The physician Gideon Harvey defined natural theology as “a natural habit of possessing the greatest good, and living in the greatest happinesse, that a natural man may attain unto in this world, and in the world to come”; while “Supernatural Theology is a supernatural habit of possessing the greatest good, and living in the greatest happinesse, that a man may supernaturally attain unto in this material, and in the next spiritual world” (1663, pt. 1. bk. 4, p. 4). It was certainly common to stress the moral, as opposed to the intellectual ends of “natural theology” (e.g., Harvey 1663, pt. 1, bk. 4, p. 5) and, at the very least, knowledge of God was always annexed to “our duties to him” (Barker 1674, Epistle Dedicatory).

Even for those less inclined to own up to Aristotelian teleological commitments, this moral and spiritual orientation was a central element of the union of natural philosophy and theology. In his impassioned defense of the early Royal Society, Thomas Sprat thus contended that experimental science inculcated the habits of “spiritual repentance” and “spiritual humility” (Sprat 1667, 341, 349f; see also Trepp 2005). Supposed pioneers of natural theology such as Robert Boyle and John Ray tended to link the study of nature to religious devotion, rather than rational speculation. Natural philosophers could understand themselves to be “priests of nature,” with the natural world a “‘temple,” harking back to older sacramental views of the cosmos. Boyle proposed that the exercises of natural philosophy, when rightly conducted, would inspire “true Sentiments both of Devotion and of particular Vertues” (1688, 88, Harrison 2014). The study of nature was philosophical worship of God (1664, 54–55, 101, 115–16). In his classic work of physicotheology, The Wisdom of God Manifested in the Works of Creation (1691), the naturalist and Fellow of the Royal Society John Ray similarly maintained that the formal study of nature would “Stir up and Increase in us the Affections and habits of Admiration, Humility, and Gratitude” (1691, Preface). This is consistent with the origins of the work which lay in morning divinity exercises delivered in the chapel of Trinity College, Cambridge. For Ray, the whole work was “rather Theological than Philosophical” (1691, Preface). Clearly these are not exercises devoid of explicitly religious premises or goals, and in spite of a superficial similarity to dispassionate rational arguments about God's existence, something quite different is going on here.

Paley's classic Natural Theology (1802), perhaps the most celebrated work in the genre, also turns out to be an awkward fit for the genre of “established” natural theology. Paley's text is typically understood as “a work of religious apologetics and philosophy of religion,” to borrow the description from the relevant Wikipedia entry. But as even a quick perusal of the Table of Contents indicates, the vast bulk of this work is neither apologetics nor philosophy but is given over to natural history. Only three of 27 chapters deal explicitly with theological themes, the rest being devoted to anatomical and physiological features of living things. (It is no surprise, then, that this work provided the explanandum for Darwin's subsequent theology of evolution by natural selection.) This is a further instance of natural theology serving the natural sciences, in this case by providing a medium for the organization and communication of scientific facts. This pattern was to continue well into the nineteenth century. The Bridgewater Treatises (1833–36), often taken to be the high‐water mark of English natural theology, were thus produced not by theologians, but by the leading scientific figures of the day, and conveyed to popular audiences the most up‐to‐date findings from various scientific fields.

Natural history delivered in this mode was still conceived of as an exercise in moral and religious formation, however. Paley's natural theology/natural history was aimed at inculcating "a habitual sentiment of our minds” which, he contended, constitutes “the foundation of everything that is religious.” Like Boyle, Paley also alludes to the commonplace that the natural world is a temple, concluding that its contemplation is “one continued act of adoration” (1825 [1802] 374f.) It is thus the inculcation of a religious sentiment at which the work aims, not a set of dispassionate inductive conclusions. Something similar might be said for the eight volumes of the Bridgewater Treatises, which despite somewhat varied goals, shared an orientation toward “the management of religious experience” (Topham 2022). This is not quite the “established” version of natural theology. The natural sciences themselves, then, were not understood as morally or religiously neutral enterprises—far less was natural theology—but were conducted within a framework of Christian theological assumptions.

It is certainly true that from the seventeenth century onward there was an increasing emphasis on the notion of evidences and the need to convince “atheists,” of whom there were imagined to be increasing numbers. The Boyle Lectures are a celebrated instance of this phenomenon, dedicated as they were to “proving the Christian religion against notorious Infidels, viz. Atheists, Theists, Pagans, Jews and Mahometans” (Topham 2010, 64). But “proving” here almost certainly retains some of its original connotation of “testing” or “putting to trial,” as opposed to offering demonstrative proof. “Atheist” was also understood in a much broader sense, taking in those who denied providence, for example. It is also significant that the Boyle Lectures took the form of sermons delivered in London churches—events unlikely to have been oversubscribed by infidel and atheist auditors. This suggests that they were primarily aimed at fortifying the faithful.

All of this said, something closer to “established” natural theology did appear in the early modern period. Deist writers were concerned to highlight the superiority and universality of supposedly rational arguments for God's existence, as compared to what for them was a less reliable (and dispute‐engendering) revealed theology. These deistic works were often motivated by confessional conflicts and a perception that competing confessions all implausibly claimed to be the legitimate interpreters of divine revelation. The situation of religious pluralism led to a quest for rational and dispassionate criteria for determining religious truths, along with an attempt to identify a minimal number of religious “fundamentals” upon which all could agree. “Natural religion” was thus often proposed as the solution to apparently irresolvable conflict of revealed religions. It was crucial, given this context, that it declare its complete independence from revelation. But this Deist stance represents the position of a small minority.

Orthodox opponents of Deism might have allowed the conclusions of natural theology, but they remained critical of the status accorded it by deistic writers. As Henry Fish put it: “the Infidel maintains the all‐sufficiency of natural theology” (1840, 1). He implies, interestingly, that persuading “infidels” by means of natural theology was not the problem: getting them to move beyond it was the goal. For Christian writers, then, the legitimacy of natural theology depended on its standing in a particular relation to revealed theology. On its own, it remained as dubious as the traditional “pagan thinking about the gods.” Natural theology “creates an appetite which it cannot quell,” as one writer put it (Chalmers 1840, vol. 2, 399, cf. Chadbourne 1875, 296f). Natural theology outside of a Christian context was deemed pointless. So we do encounter something that resembles “established” natural theology in the early modern period, and we will find dictionary definitions such as this: “Knowledge we have of God from his works, by the light of nature, and Reason” (Chambers 1728, vol. 2, s.v “theology”). Later in the century, the Enlightenment encyclopaedists Diderot and D'Alembert's will specify “by reason alone,” (1772, 55). But these still definitions exclude the condition that the enterprise be conducted “without religious premises,” not least because reason and “the light of nature” tend to be understood theologically, although this, too, was changing. And they leave open the question of the relation of natural theology to revealed theology.

A final development of the eighteenth century that is directly relevant to the belated appearance of the “established” notion of natural theology may be found in more strictly philosophical accounts of arguments for the existence of God. David Hume and Immanuel Kant offer instructive examples. Hume's subtle treatment of natural religion sought to undermine Deist claims for the “naturalness” of religious conviction while also casting doubt on the validity of specific arguments for God's existence. On the first point, he argued that the rational grounds of religion were different from their origins in human nature (or history). Against the Deists, who typically held that natural religion (a religion grounded in natural theology) was the first and universal religion of the human race, he contended that the first religious sentiments did not arise out of rational reflections on the natural order, but from more basic “hopes and fears“(Hume 1957 [1757]). A simple, rational monotheism was not, then, the natural, default condition of the human race. As for the validity of rational arguments, he advanced penetrating criticisms of popular versions of the argument from design—although at times also seems to grant it some cogency (Hume 2007 [1779]).

Subsequently, Kant provided a celebrated treatment of arguments for the existence of God in the Critique of Pure Reason (1781/1787). Kant announces that there are three modes of proving the existence of a Deity through reason, and three alone. These are the physicotheological argument, the cosmological argument, and the ontological argument (Kant A 591/ B 619; 2007, 500). In Kant's hands, the activity of physicotheology was effectively reduced to an argument or proof derived from speculative reason (although Kant had earlier acknowledged the moral and affective dimensions of “the physicotheological method” (2002 [1763], 159). While Kant mentions some of his philosophical predecessors—Descartes's version of the ontological argument, Leibniz's version of the cosmological argument—his enumeration of arguments (and his limiting of them to three types) does not arise out of a survey of natural theology (in the broad sense) and its various historical forms. It is rather a creature of the logic of his own philosophical position and his focus on establishing the limits of rational speculation. Leaving aside the force of his impressive treatment of the arguments, Kant's significance for the history of natural theology is that he neatly packages three arguments for posterity. Overstating the matter somewhat, subsequent philosophers, not given to reading much of anything but philosophy, tended to take Kant's systematic presentation of the arguments as definitive of natural theology and its history, and imagined that it had been characteristic of every thinker from Plato, through Anselm and Aquinas, to the early modern physicotheologians.

In this way, natural theology became “established” as an analytic category in spite of the fact that it mapped poorly onto the historical realities that it was meant to characterize. Severed from their original religious context, what were once pious contemplations on nature were destined to become three contestable philosophical arguments on whose validity the truth of Christianity was thought to depend. When criticisms of such arguments invited refutation, these were now conducted on the home ground of philosophical discourse. Irrespective of the plausibility of the arguments on either side, this situation represents a significant shift in the grounds of the discussion. Accordingly, the impulse toward the rational defense of Christianity in this new dialectical form had unintended secularizing consequences. This is the truth behind Anthony Collins’ mischievous suggestion that no‐one thought to question the existence of God until the Boyle Lecturers set out to prove it (Brooke and Cantor 1998, 198, Brooke 2020).

“Established” natural theology thus emerged belatedly as a construction based in philosophical treatments of the eighteenth century. Crucially, no‐one, aside from a relatively short‐lived coterie of Deists, placed much credence in natural theology conducted in this mode. From the perspective of the orthodox, it was insufficient, based upon an unacceptably narrow conception of reason, and it improperly excluded religious sentiments of reverence and piety. For philosophical critics, such as Hume and Kant, the arguments themselves were not decisive. On the religious side, Blaise Pascal, as already noted, spoke dismissively of “the God of the philosophers.” The German jurist and philosopher Christian Thomasius (1655–728) stressed the continuity between the natural theology of his era and the older notion of “natural theology” understood as pagan thinking about the gods, concluding that natural theology leads to atheism (Hunter 2007, 65–73; cf. McGrath, 129). Kant, in fact, combined philosophical critique with theology sympathies: “It was not the will of Providence,” he maintained, “that the insights so necessary to our happiness [i.e. God's existence] should depend upon the sophistry of subtle inferences” (Kant (2002 [1763], 111). In the nineteenth century, John Henry Newman (1801–90) was similarly unimpressed by the “boasted demonstrations” of “the Age of Evidences” (i.e., the eighteenth century). It was the world's “fatal error,” he maintained, “to think itself a judge of Religious Truth without preparation of heart” (1843, 197f.) (McGrath develops a very similar notion in his wide‐ranging discussion of the Christian imaginarium, and the idea of natural theology as a habitus, 69–73). The very idea of a religious system based upon “unaided reason” was doubtful, Newman insisted: ‘We know of no such system, because we know of no time or country in which human Reason was unaided’ (1843, 17). For Newman's Danish contemporary, Søren Kierkegaard (1813–55), the rational defense of Christianity amounted to “collusion with the enemy.” Kierkegaard pointed out that the proofs had a whiff of the irrational about them: “if the God does not exist it would of course be impossible to prove it; and if he does exist it would be folly to attempt it” (1941, 218; 1964, 49). The significance, in the twentieth century, of the famous Barthian “Nein” to natural theology need not be labored. Admittedly, the specific social context played an important role in this negative verdict (McGrath, 102). But it is equally true that Barth's rather severe assessment conforms to a long‐standing theological animus against the basic tendency represented by the “established” version of natural theology.

In sum, the enduring legacy of the seventeenth and eighteenth centuries in relation to natural theology was twofold. First, the practice of physicotheology established a lasting precedent that bound natural theology to the discoveries and theories of the natural sciences. Second, the philosophical prepackaging of natural theology by philosophers such as Kant offered an abstract version of the enterprise, now understood as repertoire of arguments—typically three—which, owing its user‐friendliness and susceptibility to potentially endless philosophical analysis, became the “established” version: this despite the fact that from a historical perspective it was anything but.

“Reason” and “Nature”

Newman reminds us that the validity of any system grounded in human reason requires us to have some prior commitment to the reliability of our rational faculties. When he asks rhetorically, whether human reason was ever “unaided” he invites the question of whether the basic components of natural theology, such as “reason” and “nature,” have ever been innocent of theological content. If not, then they cannot satisfy the definition of natural theology as proceeding from nonreligious premises. This is not the occasion to offer a history of these conceptions, but a few examples show the force of Newman's question.

“Reason” was long understood within the Western tradition as a divine element within the human soul (e.g., Plato, Timaeus 90a–d, Aristotle, De anima III 5, 430a17–23, Eudemian Ethics 1248a21–29; Nicomachean Ethics 1177b27–34, 1178b20–24). In Christianity, this impulse was linked to notions of the image of God, or reason as a divine gift or “light.” Directly relevant to the present discussion, reason was also understood as a “natural” (understood here in the sense of divinely implanted) propensity to receive divine revelation. These ideas were also connected to Aristotelian/Thomist understanding of the “natural” ends of human beings. Aquinas proposed that human beings have a natural desire for the beatific vision: that is, a “natural” desire (desiderium naturale) for the “supernatural,” which suggests that the “natural” and “supernatural” do not represent distinct orders. ² (This is the background to Gideon Harvey's earlier mentioned reference to natural theology as a “habit” related to the fulfillment of a natural end.) In this general connection, McGrath, for example, also notes that numerous Christian thinkers from Athanasius to Emil Brunner understood the imago dei as priming the human heart for knowledge of God (39). Theological anthropology is thus, in this sense, prior to “natural” theology and to any conceptualizing of the origins and competence of reason.

A typical early modern example of this understanding of reason is offered in Matthew Barker's Natural Theology (1674), one of the first English books to have the expression in its title. Barker informs us that it is “the very Light and Law of Nature” understood as a God‐given faculty, that preaches the existence of God and attendant moral duties (Epistle Dedicatory). We know this about the light of nature through revelation (1–2). Better known, perhaps, are the ubiquitous references of the “Cambridge Platonists” to reason as “the candle of the Lord.” Benjamin Whichcote writes that reason is “a candle in man lighted by God and doth discover God” (1751, vol. 3, 144). Thus understood, reason is a God‐given propensity to receive divine revelation. Hence, our confidence in the workings of reason, upon which the prospects of the success of natural theology rest, derives ultimately from revealed truths about human nature. On this common understanding, like the prevailing natural philosophy reason is already “religious.” Moreover, it is the “natural” function of reason to accept revelation.

However, an alternative understanding of reason was emerging at this time. Whichcote's contemporary Thomas Hobbes maintained that reason “is nothing but Reckoning (that is, Adding and Subtracting) of the Consequences of generall names agreed upon, for the marking and signifying of our thoughts” (2012 [1651], vol. 2, 64). John Locke would set out the related view that reason is concerned with the connection between ideas (that were in turn based on sense impressions). As is well known, Locke firmly rejected the idea that reason might have any substantive content (Locke 1975 [1690], 4.17.2, p. 669). While “reason” during this period is understood in a number of different ways, then, we can identify two tendencies—the richer theological conception and the more analytic instrumental one. Charles Taylor has spoken in this context of “substantive” and “procedural” rationality (Taylor 1989, 121–24, 242–47; cf. Wolterstorff 1986, 238–42). The notion of a theological project grounded in “reason alone,” then, means quite different things depending on which version of “reason” is being appealed to. To put it another way, on the older model the very notion of “reason alone” verges on incoherence (which is Newman's point).

Conceptions of nature, and especially human nature are implicated in these discussions too. As already implied above, “nature” and “natural” are susceptible to similarly divergent considerations. Much of the difference hangs on the different senses in which the relevant terms are being used: whether there is a strong disjunction between reason and revelation or between natural and revealed; and on the latter score, how to adjudicate between competing conceptions of nature as, say God's book, or the theatre of God's glory, versus nature as the naturalized (which is to say detheologized) and material object of scientific enquiry. The relevant categories for most of the common era were “creation” and “the creatures” which were understood in the modern period as synonyms for “nature.” With the former categories it might still be possible to draw a meaningful distinction between what could be known of God from the creation and what from revealed truths in the Bible (although as we have seen, for the early middle ages there was direct overlap). But this cannot be equivalent to a distinction between “nature” (in the modern sense) and revelation, because like the traditional conception of reason, “creation” and “creatures” are already theological.

This remains the case for much of the early modern period, when “nature” and “creation” are often used interchangeably. For our purposes, the most important transition in the meaning of “nature” takes place in the nineteenth century, when the object of the natural sciences is stipulated to be the sphere of natural operations, understood as excluding any reference to the supernatural. The older “natural philosophy” (as understood by Newton) and “natural history” (as understood by Paley) admitted theological involvement at least at some level; the new “science” and the new “biology”, both of which take on their modern meanings during the nineteenth century, do not. The sciences are now characterized as “naturalistic” in a sense opposed to “supernaturalistic”. Admittedly, the terminology of a natural/supernatural distinction goes all the way back to Aquinas (de Lubac 1946). But the ideas of naturalism and its contrast case, supernaturalism, date from the nineteenth century. ³ This subsequently makes possible the position that the scientific study of nature must be characterized by methodological naturalism, an expression first coined in the early twentieth century (by a Christian philosopher, Edgar Brightman 1937, 157–58; cf. De Vries 1986).

If the secularization of reason begins in the seventeenth century, the naturalization of nature takes place in the nineteenth century when we see a new conception of nature understood as the object of a naturalistic science (Lightman and Dawson 2015). Clearly, this understanding of nature cannot be equated with “creation” and is arguably incompatible with it. But it is this assumption—nature as thoroughly detheologized—that also places the sciences in the frame as the most reliable source of facts about the natural world “which neither are nor presuppose any religious beliefs”—a core condition of “received” natural theology. And so scientific naturalism is implicated in the “established” conception of natural theology.

Two things follow. First, to return to what might now seem like a tired refrain, a natural theology that begins “from premises that neither are nor presuppose any religious belief” cannot have been established before the transitions described above. Second, proponents of natural theology in the “established” sense face a dilemma: either the relevant conceptions of nature and reason are given in (revealed) theology, in which case the “established” sense becomes a contradiction because of the “nonreligious premises” condition; or, we accept the late modern scientific understanding of nature as providing the model for an appropriately nonreligious neutrality, in which case we are again confronted with a contradiction since the presumed object of inquiry—God—has already been ruled out by the methodological approach said to guarantee the neutrality of the investigative process and the data that it relies upon.

Disestablishing “Established” Natural Theology

If the historical analysis set out above is on the right track, we are still left with a puzzle about how the “established” version of natural theology ever got to be accorded that status in the first place, and why it still seems to be a live option. In this final section, I offer some observations about this, organized around four themes: first, contemporary philosophy of religion; second, critics of natural theology; third, Christian apologetics; fourth, the field of science and religion.

Part of the answer to this puzzle comes through a consideration of who has an interest in sustaining the “established” version. One group that springs to mind is contemporary analytic philosophers of religion. It was the secular, instrumental conception of reason that we have identified as emerging in the early modern period that made possible the modern practice of the philosophy of religion—a locution that traditionally would have made little sense. (The phrase itself does not appear until the nineteenth century (Dick 1826)). Modern philosophy might now be considered a practice in which this newly neutralized reason could be applied as an analytical instrument to theological topics, while remaining independent of any commitment to them. Philosophy would then seem to be the obvious disciplinary base for exercises in the neutral practice of natural theology. Contemporary philosophers of religion thus have an interest in sustaining discussion of the arguments that constitute “established” natural theology because it is their day‐job.

This brings us back to the neutrality of this enterprise where, again, we encounter some difficulties. My suspicion is that the putatively dispassionate philosophical verdicts rendered on the validity of arguments for God's existence—positive or negative—will tend to match the preexisting convictions of those engaged in the analysis. This seems to be the most plausible explanation of why philosophers disagree about the cogency of the arguments in question. Indeed, there is some empirical evidence to support this suspicion (De Cruz and De Smedt 2015, 198–99; Tobia 2016; De Cruz 2018). The celebrated case of Antony Flew's “conversion” would then represent a rare exception (Flew 2007). On the other hand, if we grant the neutrality of the relevant philosophical enquiries, then the overt loyalty of contemporary philosophers engaged in exploring these questions, qua philosophers, would not be to any theological, agnostic, or atheistic position per se, but to philosophy and its capacity to adjudicate the arguments. There is nothing theological about this. Arguments in philosophy of religion would then become theological only by accident—in the event that they yield positive conclusions about the existence and attributes of God.

Practices in the contemporary philosophy of religion that sustain the “established” version of natural theology might seem a harmless enough diversion for contemporary philosophers. But the impression is often given that this is some kind of perennial exercise that engages with a repertoire of “classical” arguments that have been doing the rounds since antiquity and which have been core to Christian self‐understanding (see, e.g., Chignell and Pereboom 2020; Craig and Moreland 2009). This amounts to the projection of a set of modern philosophical practices and concerns onto activities that, I hope to have shown, meant something quite different to the historical actors themselves. Such projection is often motivated by the understandable desire to show that contemporary philosophy is relevant to real life.

A second group with an interest in promoting the “established” version is those who wish to set it up as a “straw man” against which to define their own preferred vision of things. On one side are theological critics such as Karl Barth. As McGrath rightly observes: “Barth's formulation of natural theology has little, if anything, in common with what we find in Augustine of Hippo, Gregory of Nazianzus, or Thomas Aquinas; it is specifically tied to the paradigms of natural theology which emerged during modernity”. He goes on to point out that Barth's primary target, Brunner, does not accept this formulation of natural theology either (145). But this could be parsed the following way: Barth was right to reject “natural theology” as he saw it, yet wrong to think that any significant Christian thinker had ever endorsed it. We might even say that Barth is constructing the “established” version of natural theology for his own purposes, much as modern philosophical opponents of Christian revelation had done. This is an old gambit that has been deployed in other contexts. It has been plausibly argued that medieval heresies, for example, were primarily the construction of inquisitors (see, e.g., Zerner 1998; Pegg 2001; Moore 2007).

The more common position lies on the other side—religious skeptics for whom it is important that Christianity be grounded in arguments that they consider vulnerable. To some degree, Hume fits this description. Although this is a little unkind to Hume, we get distant echoes of his approach in attempts by contemporary scientific critics of religion such as Richard Dawkins (2006) and Lawrence Krauss (2012) to take on arguments for God's existence which they mistakenly think are: (a) foundational to the religious life, and (b) susceptible to their amateur attempts at philosophical demolition. Overall, my suggestion would be that natural theology, understood as an arsenal of rational arguments for God's existence, turns out to be more important to critics of Christianity than to most of its adherents—although, as mentioned above, professional philosophers of religion also have a stake in maintaining it as a going concern.

The maladroit efforts of writers such as Dawkins and Krauss, it must be conceded, do present an almost irresistible target. This brings me to a third group which tends to perpetuate the idea of an “established” natural theology: Christian apologists. It is perfectly understandable that defenders of Christianity would wish to deflate the philosophical pretensions of those who propose, on dubious grounds, that religious belief is somehow irrational. But this may inadvertently lend a certain legitimacy, not so much to any particular argument, but to the activity and the grounds on which the arguments proceed. It might be specified that the exercise is one of highlighting the weakness of these critical arguments considered on their own terms. That is fine, but those terms are not native to Christian theology, and arguably the key concepts such as “nature”, “reason”, and even “God” do not, or should not, mean the same things for the respective protagonists. The question is whether, when imported back into Christian theology, the alien arrivals, like an introduced species, pose a threat to the native ecosystem. My objection, to be very clear, is not to apologetics per se, which seems perfectly in order as an extra‐curricular activity. But I am not sure that it is any kind of theological endeavor, and this is why it should not be comprehended within the category of natural theology (in a range of senses, with the exception of the “established” version).

A possible response to this exclusionary stance is to suggest that, irrespective of the absence of the “established” conception for much of Western history, in the late modern period it has become part of the tradition. On this view, philosophical conceptions once constructed by outsiders have been internalized by religious communities who now see these considerations as theologically significant. ⁴ The extent to which this is true is partly a sociological question, and as far as I am aware there are no data available that might help us determine the extent to which it is true. (We do know the curious fact that in the U.S. knowledge about religion is highest among Atheists and agnostics, Jews and Mormons, which might suggest something similar for knowledge of the arguments of “established” natural theology (Pew Forum 2010)). But there is also a normative question: whether these forms of reasoning and their accompanying presuppositions should be taken on board by religious communities and should be incorporated into Christian theology. That is not for me to judge, but I would say: first, that I cannot see why critics of Christianity or philosophers of religion should determine the theological agenda; second, that importing this kind of reasoning into Christianity should not be based on the mistaken supposition that the “established” version has been there all along; and, third, that there needs be some recognition of both the historical novelty of the “established” version and of the historical circumstances that gave rise to it.

Finally, to a topic that is of particular interest to readers of this journal: science and religion. There is much that could be said here but a few preliminary suggestions will need to suffice. If science is the preeminent authority on nature, it follows that any informed natural theology insofar as it, too, is concerned with the natural world, will need to take on board scientific theories and discoveries. If we buy into the “established” model of natural theology, moreover, we have an additional reason to be paying close attention to science because it also seems to fit the model of the neutral enterprise that begins from “premises that neither are nor presuppose any religious beliefs.” The stance of methodological naturalism is thought to ensure this. I have offered some reasons why we might not wish to buy into that model. But here are some further thoughts specifically on where science fits into this discussion.

Equating naturalism with neutrality would be valid only if naturalism turns out to be true. In the case that the natural world is God's creation and a theatre of divine activity, however, then the properly neutral starting point would be a theistic one. (Such a starting point, I have argued, was characteristic of scientific practice pretty much up to the nineteenth century, not least because the relevant conceptions, “nature” and “reason” were already theologically inflected.) A possible way of resolving the question of the correct starting point would be to propose naturalism as a provisional methodological premise and then see where it gets us. It turns out that it gets us quite far, at least in a certain direction, and the remarkable accomplishments of the modern sciences are an ongoing reminder of this. For this reason, some philosophers have argued that the success of science, conducted on the model of methodological naturalism, is a confirmation of the truth of the starting premise (for discussion of this claim, see Harrison 2020a). This, then, supports naturalism proper, a conclusion that advocates of a Christian natural theology would no doubt wish to resist. But how?

One response would be that the perspective on the world offered by naturalistic sciences is limited or partial, or, more radically, in some tension with a robustly religious understanding of creation. On the face of it, proceeding to study nature on the premise that God is not to be found there will conflict with most approaches characteristic of natural theology (excepting, again, the “established” version). Admitting to this tension need not amount to a negation of the impressive achievements of science, or a denial of the material benefits that it has bestowed upon us. It is rather a matter of placing these things in perspective and seeing why, from a theistic outlook, science in its modern form cannot be taken as offering the last word on nature. Consider, again, the interpretative frame offered by the Psalmist who declares that “the heavens declare the glory of God”, or the comparable sentiment in “Gerard Manley Hopkins” “The Grandeur of God”. Hopkins tells us that in spite of the modern instrumentalization of the natural world, “nature is never spent/ There lives the dearest freshness deep down things”. These “deep down things” are not objects of scientific inquiry, and what is on offer here is a very different nature to that viewed through the restrictive lens of methodological naturalism. It is a nature more continuous with the metaphors of nature as a “temple” or God's “other book”, which is to say, with more traditional theistic readings of nature, characteristic of natural theology and the sciences before the nineteenth century.

How to make room for such alternative readings? One possible move on the theistic side, in answer to the proposal that the success of science warrants metaphysical naturalism, would be to understand at least some of the sciences in empiricist or instrumentalist terms. On this view the modern sciences yield predictions and technologies, but not truth with a capital “T”. ⁵ There are clear medieval precedents for this kind of approach—the approach of “saving the appearances”—and it has influential advocates among philosophers of science. But this option has not attracted many takers in the science–religion space (but see Reeves 2018; Harrison 2021). McGrath, for example, is typical in plumping for a version of critical realism (161). The instrumentalizing move would enable an appreciation of the practical boons delivered by the sciences without allowing them a monopoly on truths about nature.

A further consideration here, specifically in response to the claim that it is naturalism that underlies the success of science, is that this is simply not true for most of history (Harrison and Roberts 2019). Newton's physics, for example, would still count as “science” on most descriptions and yet was not pursued on the basis of any rigorous naturalistic principle. This also raises the interesting possibility that modern science, in spite of its overt naturalism, has actually retained a number of tacit theological assumptions. If true, this would offer an intriguing new way of thinking about contemporary science–religion relations, premised on making explicit the hidden theology of the modern sciences.

This leads me to one last point on the science–religion question, again informed by the history of science. We have seen that many of those who engaged in physicotheology thought of themselves as engaged in a form of spiritual practice. What was foremost in their minds was not the logical cogency of some inductive argument, but how the scientific study of nature inspired religious reverence and awe. We might then inquire whether, and in what sense, modern scientific practices might also be imagined as retaining some aspects of the traditional physicotheological approach. To the extent that this is the case, we might then rethink science‐Christianity relations in terms of competing or complementary “spiritual practices” (cf. Ip 2021). Disciplinary formation in naturalism, even as a methodological stance, would suggest an inevitable conflict with the theistic commitments of Christianity. But there is so much more to the modern sciences than methodological naturalism and an accompanying reductive materialism. Methodological naturalism may be theoretically required as an overt commitment for the conduct of modern science, but arguably that mandate is most visible in the policing of the boundaries of science, and among those concerned with theoretical criteria of demarcation. Whether it constantly occupies the waking thoughts of practicing scientists is another question. Arguably, there are other elements of scientific formation that are continuous with the older physicotheological stance. McLeish (2019), for example, has powerfully argued for the creative and affective dimensions of contemporary scientific practice, suggesting that it has much closer affinities to poetry and music that we typically imagine (cf. McLeish 2021, McGrath, 41–50 and passim). This suggests that it is possible to misconstrue scientific practices, and with the same kind of bias that has given rise to the construction of “established” natural theology, as a form of bloodless analysis and rationalization. If there is anything to this proposal, it suggests an alternative space for thinking about relations between science and Christianity to that offered by “established” natural theology, shifting the focus away from doctrines to virtues and practices (cf. McGrath, 110–16).

Abandoning “Natural Theology”?

By way of conclusion, I want to address the question of dispensing with natural theology, foreshadowed in the subtitle of this piece. As I hope is clear by now, my main concerns about the validity of natural theology lie not with most of the activities that McGrath includes in his characterization of the enterprise, nor with the kind of project that he seeks to promote. It is rather the label “natural theology” that strikes me as ambiguous and unhelpful. This is because I believe that “established” natural theology—which I readily concede is what most people think about when they think “natural theology”—cannot be brought under the broad umbrella of a Christian natural theology, and that attempting to do so is likely to be confusing and counterproductive. Of course, we can use the term “natural theology” to mean anything we like, but the issues that I have identified are not semantic ones that can be resolved simply by stipulating our way around the difficulties. Providing a clear definition is well and good—“natural theology is ….”—but the question still needs to be asked whether the definition matches any relevant practice. “Established” natural theology fails that test by my reckoning, although it is instantiated in groups of “outsiders” who, for the most part, have different goals to those engaged in Christian theology. In this sense, our present, established “natural theology” has a good deal in common with the physicam theologiae that medieval thinkers viewed with some suspicion.

Happily, workable alternatives to “natural theology” are readily at hand. “Theology of nature” is an option, although it does not quite cover the range of options that McGrath, for example, would wish to include in the portfolio. Perhaps McGrath's own “Christian natural theology” offers the best solution. The qualifier “Christian” clearly signals that this is not an activity that rests upon nonreligious premises. It is not, in other words, “established” natural theology. Rather it is a long‐standing tradition of thinking from an explicitly Christian perspective about aspects of our natural and social worlds. ⁶