Geology, naturalism, and the problem of miracles

My blog posts have become rather rare events over the last year, while I work on a book project on the ‘rare event’ in science (especially geology) and its implications for the prevailing view of naturalism in philosophy and theology, and especially for the problem of how to define miracle. In order to work through my argument, I gave a seminar paper on this to our Theology and Ethics seminar at New College in Edinburgh. Here’s the text of my paper (warning: it’s over 6,000 words).

Title: ‘The Stone the Builders Rejected’: Geology, Naturalism, and the Problem of Miracles

I should warn you at the outset that this paper contains very little theology, and even less ethics. I want to explore how the natural sciences can be used to explore the theology of miracle, so starting from the science end of the science-theology conversation. I should emphasise that I’m condensing an ongoing book project into one paper here, hoping to get a sense of whether the argument holds together. It means that I’ll need to apply a very broad brush, and skip over the interesting details and debates. So I’d be interested in your comments afterwards.

In summary, I want to explore the relationship between naturalism and miracle, at the roots of nature. The current way of doing this – where physical law is taken as the basis to nature – is profoundly unhelpful, I believe. I want to suggest a different way forward, using two historical methodologies (philosophies?) from geology: uniformitarianism and catastrophism.


My main title says it all, in a coded way. You’ll be familiar with the line from Ps.118 – ‘The stone the builders rejected has become the chief cornerstone’, quoted by Jesus to his critics in the Temple.[1] I’m not making a Christological statement myself, more an historical judgement about the history of geology and its impact on wider science and theology. Of particular relevance here is what’s called the uniformitarian-catastrophist debate, which I’ll explain later. For now, I’ll just say that when the C19 builders of the new science of geology rejected the stone of catastrophism as their founding methodology and embraced uniformitarianism in its place, they lost something crucial, namely their ability to describe a truly remarkable historical event in scientific terms. It wasn’t clear at the time that anything had been lost, because uniformitarianism was a key move towards the scientific naturalism we take for granted today, and it was also a key move in the parting-of-the-ways between science and religion. However, this naturalism can’t easily recognise the significance of unique events, contingency and novelty in nature. And it’s this naturalism which, I’ll suggest, has created endless problems for theologians and philosophers engaging with the idea of miracle. By the time of the late C20 though, geologists were realising that what had been rejected, namely catastrophism, should in fact be something of a cornerstone in geological method, and they began the task of reconciliation between catastrophism and uniformitarianism. C21 philosophers and theologians are behind the times here, still working with a naturalism based on law which automatically marginalises contingency as a component of natural reality. Towards the end of this paper, I’ll go on to suggest that, by returning to the uniformitarian-catastrophist debate, as today’s geology has in fact been doing, we can start to see a new way through the problem of miracle.


As you’ll know, treatments of miracle in theology and philosophy tend to revolve around David Hume’s famous definition of miracle as a transgression of the laws of nature by the deity. Hume is almost inescapable here, despite the many problems identified with his definition. But he’s inescapable because, if you hold physics as the most fundamental science (and most scholars seem to), then you’ll inevitably end up with a naturalism where physical law is the basis (cause and explanation) for everything, where the laws of nature become the most real things there are, where it’s the laws that define the boundary between nature and supernature. This is why it’s Hume’s definition of miracle dominates, problematic as it might be. Another name for this trap is reductionism: the metaphysical assumption that all of the natural sciences reduce down to physics at the end of the day, and probably particle physics at that. Notice that you can’t demonstrate this result empirically: you can’t test, either by theory or experiment, that particle physics is the most fundamental branch of physics even within physics. To repeat: reductionism is a metaphysical belief, and one that’s challenged by many branches of science these days that deal with complexity and emergence, including my own branch of physics (condensed matter physics). Reductionism is also challenged by the simple empirical observation that there are many natural sciences that don’t relate naturally to physics, except in trivial ways like, gravity acts on all things, or all things are made of physical stuff.


So what I want to do in this paper is to explore what happens when you question the assumption that physics is the primary science, and that all others are derivative in some sense. The landscape changes, I’ll suggest, particularly regarding the problem of miracle.



Naturalism, and the unity of science
Central to everything here is what we mean by ‘nature’, or the ‘natural world’, and conversely, what it stands over and against, i.e. ‘supernature’. These terms tend to be highly elusive to define, and the problems revolve around knowing what holds nature together, what unifies it? The main way into this is to look at the natural sciences: what holds them together, what unifies the sciences? The easy answer is provided by the reductionist, who sees physics as the glue that holds all of the sciences together through its fundamental laws, which, the reductionist would say, undergird all of the sciences alike. But if you’re not a reductionist, the unity of the sciences is much more elusive.


Theologians try to escape the problems here by talking about ‘creation’ as the unifier for the sciences, but this solves nothing. It simply places the natural sciences in a box which is larger than they are, but they rattle around inside it. Trying to unify the sciences by talking about creation does nothing to tell us how the individual sciences should relate to each other in ways they can apprehend, through their own individual aims and methods; for instance, how geology should relate to physics. You’re back at reductionism again as the only viable solution.


As a way forward, I want to explore naturalism, another elusive term. The most widespread version, what’s often called ‘scientific naturalism’, but which is really ‘physics naturalism’, is based on the laws of nature, and is therefore heavily reductionist. Take cosmologist Sean Carroll’s definition, which is entirely typical:[2]


‘Naturalism comes down to three things:


1.      There is only one world, the natural world.

2.      The world evolves according to unbroken patterns, the laws of nature.

3.      The only reliable way of learning about the world is by observing it.


Essentially, naturalism is the idea that the world revealed to us by scientific investigation is the one true world.’


Carroll appears to regard this definition as both fundamental and obvious: he makes no attempt to establish or justify it by further argumentation. I, on the other hand, find that it begs four questions.


First, there’s the question of why the natural sciences should be assumed competent to define the ‘one true world’: after all, no scientific experiment or theory is able to do this, still less provide the reason why the natural world should be amenable to our scientific enquiries; it’s a deep mystery in physical terms that begs a metaphysical answer.[3] Even the discovery of a so-called ‘theory of everything’ would presuppose the laws of mathematical physics, but wouldn’t explain those laws, nor why there are laws in the first place; we’d still be left with metaphysical questions about why this theory and not another. Second, there’s the question of whether scientific naturalism is able to recognise the reasoning by which it’s derived, including other abstract realities that aren’t described directly by the natural sciences, realities such as mathematics, money, and the lived human experience. To be fair to Carroll, he attempts to account for these other realities by formulating what he calls ‘poetic naturalism’, but he still regards the law-based scientific form of naturalism as fundamental; the upshot is that non-scientific realities like money and human experience are epiphenomenal, not as real as the laws of nature. So, third, there’s the question of whether scientific naturalism can account realistically for human subjective experience of the world. After all, my own first person subjective experience as a living, thinking human person is the most empirical fact in the entire world to me, but scientific naturalism, which can only speak in the objective third person can’t account for it. Fourth, and most significant I think, there’s the question of whether scientific naturalism is actually scientific. After all, it doesn’t reflect the empirical reality of the natural sciences themselves as they’re actually practiced on the ground, in all their diversity and complexity. Instead, scientific naturalisms like Carroll’s rely on a metaphysical agenda of their own, the ‘Myth of the Unity of Science’ as philosopher John Dupré puts it.[4] An empiricist such as myself (born out of decades of experience in a laboratory) would emphasise the basic empirical observation that the natural sciences are highly distinct from each other, even if there is overlap. Indeed, this compartmentalisation of our knowledge of the natural world into different disciplines is such an inescapable feature of the working scientific landscape that one might almost wonder whether the natural world itself is similarly ‘compartmentalised’ rather than unified. I don’t want to drive wedges between the natural sciences here, but I do want to take their individual aims, objectives and methodologies more seriously than does the prevailing scientific naturalism, like Carroll’s version.


My complaint with a scientific naturalism like Carroll’s isn’t just the inherent reductionism—which is too metaphysical for an empiricist like me—it’s the fact that an over-reliance on the laws of nature (which are, by definition, timeless and a-historical) dominate over historical contingency. The natural world becomes defined by necessity over chance, by impersonal and eternal causes over contingent events, or, in short, by natural philosophy over natural history. Not only does this rely on an unstable view of the natural sciences, it can’t make sense of human experience of the natural world, because it downplays contingency and history (among other things). Hence, I want to suggest, against scientific naturalism, a more ‘natural’ approach to naturalism, and ironically, one which is also more helpful for the theology of miracle.


For, if there’s one thing that indisputably unites the natural sciences, it’s the conviction that ‘there will be no miracles here’. Every natural science without exception follows this rule, whatever compartment of the natural world it’s concerned with, that it should continue to ask questions and to investigate, and never be satisfied with explanations that call upon mysterious personal forces and inexplicable happenings (miracles, in other words). This means that, if we want to understand how the natural sciences might be united, we shouldn’t call upon reductionism, or the laws of physics, we should focus on miracle.



Hume’s definition of miracle
‘A miracle may be accurately defined, a transgression of a law of nature by a particular volition of the Deity, or by the interposition of some invisible agent’ (Enquiry X.12, Endnote [K]).


David Hume’s definition of miracle from the chapter ‘Of Miracles’ in his Enquiry Concerning Human Understanding is so well known that I needn’t dwell on the many problems that are routinely identified. Instead, I want to place it in its wider context.


That wider context is Hume’s philosophy of induction, which, coupled with his rigorous empiricism, is absolutely central to the natural sciences. In his Enquiry, Hume insists that we can’t make reasonable claims about the world that go beyond our pragmatic familiarity with it: all knowledge of the natural world must be based on observation and experience (i.e. experiment) rather than theorising from first principles on the nature of things. This point has been so influential in the development of contemporary scientific method that it’s universally taken for granted as scientific ‘common sense’.[5] Hence, it’s no surprise that Hume’s definition of miracles as transgressions of the laws of nature is effectively contained within scientific naturalism, where the supernatural and the natural are so discrete and so mutually-exclusive that a miracle can only be a forced disturbance of the natural order, as represented by the laws of nature. Critics of Hume’s definition of miracle often point out that it requires a very high view of our ability to know the laws of nature, such that they become virtual a priori principles stacked up against the miracle-working Deity. But if that’s so, I ask, why does Hume have a relatively low view of the laws of nature elsewhere in his Enquiry? Is Hume inconsistent, or have we simply misunderstood him? Note that scientific naturalism certainly works with a high view of the laws of nature, so if this is indeed an inconsistency, it’s shared by theologians and philosophers who assume that scientific naturalism defines the natural-supernatural divide. Which is probably why, although those scholars invariably critique Hume’s view of miracle, they keep returning to it; it’s inescapable if you believe that physics in any way underpins the other sciences.


There’s a way out of this vicious circle though, and that’s to stop placing the laws of physics at the basis of natural reality. In fact, Hume himself provides the way out earlier in his Enquiry, and that’s through the principle of uniformity, our assumption that the future will resemble the past. His famous definition of miracle might lead you to think the laws of nature underpin Hume’s whole philosophy of how we reason about the world; in fact it’s uniformity which is central for him. It’s uniformity which allows us to interpret our experience of the world and to predict the relationship between cause and effect. Now uniformity is closely related to the laws of nature, and indeed, the laws are really a special case of the uniformity of nature, but uniformity is a wider—and to my mind much more fertile—concept than laws, especially for thinking through the problem of miracle. For one thing, if we tend to believe that the laws of nature are out there, embedded in nature herself, Hume’s uniformity is at least as much in here, a cognitive tool which we develop and refine by means of our lived experience of the world. In fact, Hume’s treatment of uniformity emphasises this latter lived experience, where uniformity is an assumption that we must make in order to go about our business: I can never prove a priori that the future will resemble the past, that when I go to bed tonight, for instance, I’ll wake up tomorrow. My own experience has so far been 100% consistent with that expectation, but I know from wider human experience that it’s entirely possible I might not wake up tomorrow. Nevertheless, my own experience so far is consistent with the expectation that I will. And so I can make a working hypothesis of uniformity between my past, present, and future, allowing me to extrapolate from the past in order to infer the future. Here’s the famous passage:


For all inferences from experience suppose, as their foundation, that the future will resemble the past…If there be any suspicion that the course of nature may change, and that the past may be no rule for the future, all experience becomes useless, and can give rise to no inference or conclusion. It is impossible, therefore, that any arguments from experience can prove this resemblance of the past to the future; since all these arguments are founded on the supposition of that resemblance (Enquiry IV.21).


So uniformity is the inference from the known to the unknown, the assumption that our experience is reliable and uniform, that ‘the future will resemble the past’. Uniformity is related to the laws of nature, but it’s internal to my reasoning as much as it is external, in the regularity of the natural world itself. Uniformity is also an important principle in geology, it’s what uniformitarianism is all about, and I’ll have more to say about that shortly. For now, focussing on miracle, I want to underscore Hume’s emphasis on the provisional nature of uniformity, that uniformity is a ‘supposition’, a working hypothesis. It can never be confirmed by inference from what we’ve already experienced, because that inference would be made by assuming uniformity to hold in the first place.


The conclusion I want to make here is that Hume’s ubiquitous definition of miracle as a transgression of the laws of nature is misleading when taken out of the context of his wider (and more cautious) account of inductive reasoning by means of uniformity. Theology has struggled to maintain a reasoned approach to miracles ever since this misreading of Hume became widespread, but theology hasn’t helped itself by assuming that the only version of naturalism which really matters is the reductionist one based on physics. As is hopefully becoming clear, I maintain that such a fixation with physical law as the most ‘real’ basis to the natural world misunderstands the sciences, and misunderstands Hume, as the founding father of naturalism. Taking Hume’s wider context into account, I suggest that his view of miracle is best articulated as a disturbance in the uniformity of nature by a supernatural agent, rather than a transgression of the laws of nature. Both are compatible with each other, but by highlighting miracle as a disturbance in the uniformity of nature we acknowledge that miracle is as much a disturbance in our internal uniformity as in the external uniformity of the world. In other words, this definition is simultaneously realist and anti-realist about miracle; it includes both objective and subjective elements. To see this more clearly, we need to look at what it means to speak of the ‘uniformity of nature’.



The Uniformity of Nature
When physicists speak of the uniformity of nature, they tend to think of it as an objectively real property of the natural world, manifest in physical laws embedded in nature, but also in symmetry, and mathematical properties in nature such as the ubiquity of the golden ratio (1.618…).


Geologists, are more cautious than physicists about the uniformity of nature, realising, thanks to the uniformitarian-catastrophist debate, that there’s an important distinction between saying that uniformity is truly real in nature, and saying that uniformity is our working hypothesis (an ‘instrument’) in our study of nature. This, of course, is the realist/anti-realist dichotomy.


Hume got there before the geologists. My own analysis of his thought on the uniformity of nature divides up into three categories, which cut across the realist/anti-realist dichotomy:


Epistemological uniformity, i.e. our internal (anti-realist) assumption that the future will resemble the past, which enables us to hope that we can gain traction on category 2.
Empirical uniformity, obtained by the scientific process of experiment, simplification, and modelling. Newton’s laws of motion are examples that Hume discusses, and he appears to hold an instrumentalist take on them. But he nevertheless holds that such ‘general laws’ advance our understanding, in the sense of providing us with improved tools to predict nature’s behaviour. But Hume is careful not to equate ‘general laws’ like Newton’s with the ‘ultimate springs and principles’ at the bedrock of reality.
Metaphysical uniformity, the realist belief in a uniformity that’s embedded in nature, what Hume sometimes refers to as ‘ultimate springs and principles’, or ‘secret powers’. Hume is extraordinarily cautious about saying anything concrete about 3., although he acknowledges his belief in this kind of uniformity.
The weight of Hume’s presentation falls on category 1., the most anti-realist type of uniformity. Note the contrast here with scientific naturalism, which claims to draw from category 2. (empirical uniformity) to determine physical laws, but in elevating those laws to the ultimate expressions of reality it actually puts it weight on category 3. (metaphysical uniformity). To emphasise: the weight of Hume’s uniformity falls on 1.; the weight of modern scientific naturalism falls on 3. It’s no wonder then, that if we read Hume’s definition of miracle in terms of modern scientific naturalism (where the emphasis is on 3.), then of course we’ll misunderstand Hume (whose emphasis is on 1.), and we’ll find problems with his definition, without being able to escape from it.


Scientific naturalism might belong largely to category 3., but the day-to-day work of the natural sciences is spread across all three of Hume’s categories. Working scientists are well aware that their theories are highly approximate, analogical, and provisional—today’s stunning insight into reality will certainly become tomorrow’s instrumental working approximation—and yet there’s still a basic conviction that nature is fundamentally uniform in itself, metaphysically. Look at any scientific paper reporting experimental results or observations, and you’ll see all three types of uniformity coming in and out of focus. Scientific naturalism completely misses this point: it assumes that, when it talks about the laws of nature as the ground of reality, these are the same laws of nature that scientists determine empirically. They’re not. Real science is considerably more subtle, more provisional, and more messy, than scientific naturalism assumes. This point is abundantly clear when you look at how uniformity is used in geology, where miracle comes back into the picture.



The Uniformitarian-Catastrophist Debate
One of the big problems about using physical law as the basis to your view of reality is that you marginalise history. A laboratory scientist like a physicist or a chemist might hope to determine the law-like, regular behaviour of a phenomenon, so she would carefully isolate the contingencies in the laboratory, and try to determine what holds in all times and all places. A law of nature. Many branches of geology simply can’t do that. And neither are they much interested either. The thing about determining laws is that you lose the uniqueness of time and place. Geology is unusual among the sciences in that it welcomes the scandal of particularity: it’s not that it’s disinterested in laws of nature, so much as it wants to understand the specialness of earth history: the unique times and places. So it’s not that the laws of nature are irrelevant, more that the focus is on contingency, or if you must think in terms of laws of nature, how the laws have played out in specific times and places. For instance, take one of the most impressive geological features on the earth: the Grand Canyon. Now you could derive a law of canyons which would tell you everything you needed to know about how all canyons form everywhere in the universe. And if you were a reductionist, you’d want to reduce that to more fundamental laws of physics. But none of these various laws would tell you anything about why the Grand Canyon itself is so uniquely interesting to geologists,[6] nor would this law of canyons tell you what geologists have learned about the earth’s history from the Grand Canyon, above all other canyons.[7] In other words, reducing the Grand Canyon to one instance of a general law would tell you nothing about why this particular canyon is of unique scientific importance in itself.


So geology has an intense interest in the uniqueness and contingency of history, a peculiar feature of the science which is very relevant, I think, for trying to develop a naturalism that appears more natural for historical beings like us, who are also naturally interested in the uniqueness of time and place. Such a naturalism, I think, would also be more productive for a theology of miracle. I suggest that uniformity is the key here, and coincidentally, of all the natural sciences, it’s geology that’s thought most deeply about uniformity.


For much of twentieth century geology, uniformity was seen as so fundamental to geological method as to form a virtual law of nature itself, and a specifically geological law at that, representing the conviction that geological processes have been uniform throughout the earth’s history. That is to say, those geological processes we see now—uplift and erosion, sedimentation, volcanism, seismicity, and of course many more—have operated throughout the past in comparable ways to what we see now. This position, widely known as ‘uniformitarianism’ is clearly relatable to Hume’s principle of uniformity, where we assume that the future will resemble the past. In uniformitarianism though, the temporal perspective works in the opposite direction to Hume, and you get the famous maxim of uniformitarianism which expresses the difference with Hume well: ‘the present is the key to the past’. So while Hume expressed uniformity in terms of our assumptions about the future, based on the experience we’ve built up and the evidence available to us now, uniformitarianism attempts to reconstruct the unexperienced past on the same evidential basis, what’s available to us now.


A quick historical sketch of how geology came to this position, which takes us back to the so-called uniformitarian-catastrophist debate of the 1830s. I should note that this is an important precursor to the Darwin debates of the second half of the C19, and, like the Darwin debates, it’s similarly been caricatured as a science vs. religion debate. I’m going to give you the caricatured version first, such as it was taught to generations of geology undergraduates until approximately the 1980s. (The caricatured version is actually the easiest way in). So in the early C19, prominent geologists like Georges Cuvier, William Buckland, and Adam Sedgwick believed that the earth’s history had been marked by gigantic cataclysms, with relatively stable, quiet periods in between. The Alps were often cited as a prime example of how the earth must have been wrenched by tremendous forces acting very rapidly. After all, looking at the jagged and extreme immensities of the Alps, especially in places like the Dolomites, it seems to go against all common sense to say they were formed by the infinitesimally slow geological processes we mostly see happening around us. Surely the Alps must have been formed by a rapid and violent convulsion of the earth, like a gigantic earthquake.


There was one notable exception to this widespread conviction that the earth had been shaped by geological catastrophes though: Scottish geologist Charles Lyell who, in the 1830s published his monumental Principles of Geology. Lyell’s famous critique of the prevailing consensus was that it was ‘parsimonious of time, and prodigal of violence’.[8] Lyell pointed out that if we take the idea of deep time seriously, that the earth has been in existence for millions of years, then much of the earth’s geological record can be explained by those imperceptible processes we see now, but operating over unimaginable expanses of time. Lyell’s approach came to be known as uniformitarianism, working with the principle that the geological processes of the ancient past are uniform with those of the present. In contrast, the prevailing view, which emphasised tremendous cataclysms, came to be known as catastrophism, and so the uniformitarian-catastrophist debate was born, with Charles Lyell and his disciple Charles Darwin on the uniformitarian side, and everyone else on the other. Thanks to the later success of Darwin’s theory of evolution by natural selection, which also emphasised gradual naturalistic change acting over vast periods of time, uniformitarianism slowly came to dominate over catastrophism, replacing it completely by the C20. And so generations of geology undergraduates in the C20 were taught that Lyell was the great hero of scientific rationality against the unscientific and superstitious catastrophists. The uniformitarians made an important leap forward in the development of scientific method, while the catastrophists were forced to call upon miracles to explain the remarkable features of the earth. This is why catastrophism should be relegated to our unenlightened past.[9]


That’s the usual presentation of the C19 uniformitarian-catastrophist debate, until recently. But, like the Darwin debates, it’s been re-assessed by historians over the last few decades and shown to be yet another version of the myth of conflict between science and religion, where heroic men of science such as Lyell and Darwin labour to free geology from the shackles of religion. I won’t go into the details of the historical revisions here because they’re not especially relevant to this paper. What is relevant, is the climate in modern geology in the 1970s and 80s around the time this debate started being re-assessed. For, a very serious interest in catastrophes started taking hold of geology again at this time. And the turning point came in 1980, when an article was published in the high-impact journal Science, which suggested that the reason for the mass extinction at the end of the Cretaceous period, 65 million years ago (which included the dinosaurs) was because of a massive asteroid collision which caused widespread ecological and climatic disruption. A huge catastrophe in other words. The popularity of this ‘extraterrestrial impact’ model has gathered momentum, so that, even though there are detractors,[10] it’s reasonably close to being the consensus explanation for what’s generally referred to as the K-T (or K-Pg – Cretaceous-Palaeogene) extinction, and this model has, in any case, become firmly entrenched in the popular imagination. Also, this model has inspired something of a renaissance in catastrophism, now widely dubbed ‘neocatastrophism’. Here, the primary claim of Lyell’s uniformitarianism – ‘the present is the key to the past’: that the rates, intensities and types of geological processes have remained uniform over the history of the earth – is seriously undermined as the methodological basis to geology. So, in coming to terms with this revival of interest in mass extinctions and their causes, contemporary geology has been through another wave of debate about the uniformitarian-catastrophist controversy, and it’s reached a kind of via media which accepts elements of catastrophism quite readily, especially the point that much of the earth’s geology is episodic and evolutionary, rather than smooth and cyclical (as Lyell’s uniformitarianism would have it). This means that there’s no longer any real anxiety in mainstream geology about watering down its commitment to uniformity between the past and present, as if this was an abandonment of scientific method to bring back miracles, superstition and apocalypticism (which is how the rather shrill rhetoric of earlier C20 geology saw catastrophism). In effect, what modern geology has discovered is that there was never a meaningful gulf between uniformitarianism and catastrophism in the first place: they should belong together to form a complete scientific methodology for the earth’s history. And in fact, modern geology’s via media has come to look rather like Hume’s understanding of uniformity, with his three components: metaphysical, empirical, and epistemological.


But what does this mean for the theology of miracle? Well, once you begin to incorporate catastrophic (that is, highly unusual, extreme, and perhaps even unique) events into your scientific history of the earth, our habitual thinking about miracles (that they’re impossible events, naturally speaking, transgressions of the laws of nature) becomes largely meaningless.



A Science of the rare event
Another way of talking about the impossible is to consider the idea that there might be a science of rare events. Remember that the natural sciences are usually said to be concerned with patterns and regularities in nature—hoping to unearth metaphysical uniformity, in other words—so the idea that there might be a scientific description of rareness, perhaps even uniqueness, is somewhat counter-intuitive. But, from the human perspective, that’s exactly what geology does. We humans have an extreme cognitive disadvantage when it comes to grasping the primary variable in geology, time. When you have 4.5 billion years at your disposal (the age of the earth), almost anything becomes possible, but we find it difficult to imagine time beyond a few thousand years. Remember that uniformitarianism emphasises the immensity of geological time to argue that even tiny causes can have immense effects. The catastrophist side of that same coin says that over the immensity of geological time even highly improbable but immense effects can happen. Modern geology has combined these viewpoints.


Hence, to re-iterate, when you have deep time at your disposal almost anything becomes possible in the normal run of nature. The improbable becomes inevitable, and the incredible becomes probable.[11] Is it still meaningful to distinguish between what’s possible and what’s impossible in the natural scheme of things, as Hume’s definition of miracle is normally understood? Probably not. When an event is so improbable that the likelihood it might happen in any given year dips below 1 in 10-11, then it’s effectively impossible on the timescale defined by the age of the earth’s crust, which is a reasonable geological yardstick.[12] When we speak of such tiny probabilities, there becomes no practically-discernible difference between an impossible event and an extremely unlikely event. Note that this is very much the same point as that made by the multiverse hypothesis in modern cosmology, to explain away fine tuning without needing to call upon a Creator God. The idea here is that when you’ve got many, many universes, one of them is bound to turn out right for life, even with the tiny probabilities involved. You don’t need a God to explain fine tuning. Geology is much more secure than cosmology though: the multiverse is an entirely untestable hypothesis, whereas the immensity of geological time is as close as you can get to solid empirical fact in science. So what I’m suggesting here about a science of the rare event is not some untestable speculation on the edge of science: this is mainstream.


Again, let’s assume that the collective memory of the human race is about 5,000 years (which would take us back to beyond the oldest written texts). We might think that an amazing natural event that happened only once during that time was as good as being unique, or at best extraordinarily improbable, such that the highest probability we could assign to it would be once every 5,000 years, at tops. And yet a quick calculation[13] shows that this would be such an inevitable, regular and probable event on many geological timescales as to appear practically continuous. So even the concept of an event, a happening at a moment in time, is a relative concept in light of the immensity of geological time. The almost impossible event (on our timescale) becomes probable (on nature’s timescale), a regularity even. ‘Time itself performs miracles’.[14]


This, I think, makes a very useful point, that when we look at natural timescales—that is, geological timescales as well as the timescales over which laboratory scientists live and work—the whole concept of miracle has to be re-thought. A miracle is at least as much in our perception of what constitutes an event, and an extraordinary one at that, as it is in the probability that such an event might happen. Subjectivity can’t be overlooked, as I’ll suggest in my final section.



Naturalism and miracle
I spoke about scientific naturalism earlier on, with its emphasis on laws of nature, but with no real interest in historical contingency. Let’s think about how to formulate a geological naturalism now, one that respects the immensity and the particularity of time. This problem has been hampered by the fact that few philosophers of science have taken an interest in geology (preferring the bright lights of physics and evolutionary biology, it seems). But in my opinion, now that the prevailing physics-based scientific naturalism is looking increasingly shaky, geology has the potential to lead the way (as it did in the first half of the C19).


What we can say at this stage is that the assimilation of neocatastrophism into the mainstream has meant that geology now works with a broader, looser, kind of naturalism than the law-based version which most philosophers and theologians assume is the basis to physical reality. It’s not that geologists don’t believe in laws of nature any more. It’s more that, in many branches of the subject, laws aren’t particularly effective at capturing what’s most interesting about reality, and that’s especially the case when you’re concerned with rare and episodic happenings in nature, like mass extinctions. So geological naturalism (insofar as it’s a thing yet) isn’t just concerned with laws of nature as the bedrock to reality, but it shows an equal interest in that which is contingent, unusual, and perhaps even unique too. When you’re interested in contingencies, teleology inevitably starts to appear in your scientific description of reality: why this happening and not another. Therefore, narrative description becomes more useful than law in your scientific account of the natural world, just as narrative is our primary means of describing human history,[15] and also of describing miracles in our theological traditions too. As biblical scholarship demonstrates, the miracles of Jesus, for instance, are best approached (and critiqued) on a case-by-case basis, as narratives relating rare or unique events in a specific historical context, not within a generalised or universalised category of ‘miracle’ divorced from all particularities. The consequence here is that if your naturalism is becoming more narrative-based, and your account of miracle also emphasises narrative, then you’ll inevitably find it difficult to tell them apart. An objective definition of miracle becomes rather meaningless, since the focus is less on the mechanics and the probability of the event in question, and more on its significance. The objective side of miracle, which has tended to be our main interest in defining and determining miracle for the last few centuries, becomes less significant than the subjective side of miracle, its deeper meaning.


I want to conclude by returning to David Hume. If he’s the source of the problems surrounding the modern understanding of miracle, then I suggest that he’s also the solution. I suggested that his famous definition of a miracle as a transgression of the laws of nature is usually misinterpreted because it’s not taken in its wider context, where it would be better seen as a disturbance in uniformity. I explained that Hume’s uniformity of nature has three components, which cut across the anti-realist/realist divide. In other words, even Hume’s treatment includes an inescapable subjective element to miracle. And finally, this is why I think that Hume’s subtle view of uniformity is actually a much more promising way of setting up a useable naturalism across the natural sciences, over the prevailing scientific naturalism based on a high view of the laws of physics.


Return to summary, and draw out key points:

Introduction: ‘The stone the builders rejected’ – theology of miracle has suffered from being seen through the lens of a naturalism that can’t recognise historical uniqueness in naturalistic accounts of the world.
Naturalism, and the unity of science – physics doesn’t provide a strong basis for naturalism once you question reductionism
Hume’s definition of miracle – usually assumed to set a stratospherically-high bar for divine action, using the unbreakable laws of nature. But this ignores Hume’s wider context
The uniformity of nature – uniformity provides a more accurate understanding of Hume (uniformity has three components, cutting across the realist and anti-realist split)
The uniformitarian-catastrophist debate – the importance of the rare event
A science of the rare event – on geological timescales the impossible becomes possible. The need for a narrative account of contingent events within nature.
Naturalism and miracle – objective understandings of miracle become largely meaningless; need to incorporate subjective. Cf. Hume’s uniformity.

[1] Mk.12:10 and parallels.
[2] Carroll (2016: 20).
[3] Trigg (2015: 143-148).
[4] Dupré (2004: 39-51).
[5] Peter Millican
[6] Basically because it exposes most of the last two billion years of North American geology almost perfectly; it provides one of the world’s most complete geological columns.
[7] See the dispute between George Gaylord Simpson and Watson on this.
[9] E.g. Stephen Jay Gould’s 1965 paper, ‘Is Uniformitarianism Necessary?’, p.224
[10] The main contender at present incorporates very large-scale volcanic eruptions into the scenario, eruptions that are known to have occurred at the time in India’s Deccan Traps (G. Keller et al., “Main Deccan volcanism phase ends near the K-T boundary: Evidence from the Krishna-Godavari Basin, SE India,” Earth and Planetary Science Letters 258 (2008) 293-311).
[11] Hsu.
[12] Gretener, p.2206.
[13] Gretener again.
[14] Wald.
[15] Kosso; Cleland 2011, 53-5.

7 thoughts on “Geology, naturalism, and the problem of miracles

  1. I see that your request for comments seems to have remained unfulfilled so perhaps you will have given up checking for any. However, I would like to leave a quick question (or three) here just in case.

    Your view that science is (or should be) based on the principle of uniformity rather than natural law sounds, to me, similar to Alvin Plantinga’s view of science. Do you recognise any similarity with his ideas?

    I’m somewhat puzzled by your “Time itself performs miracles” argument. You say, correctly in my view, that given a sufficiently long period any event becomes inevitable or at least highly probable. You then seem to suggest that this allows you to regard miracles as regular, if unusual, events in the natural world. But if the period we are talking about is, as you say, the 5000 years of recorded human history (and I think you have correctly focussed on that period because, surely, the whole purpose of miracles is for God to make some moral point clear to humans) this is nowhere near the length of time necessary to make these events probable and, presumably, most miracles have happened during the period that the Bible was being written so it would be shorter than that actually.

    You say that the “prevailing physics-based scientific naturalism is looking increasingly shaky.” Could you give some examples of this?

    • Thanks for the comment – it’s very welcome! To take your questions in order:
      (1) I’m not sure what you mean by suggesting that my view of science based on uniformity is similar to Plantinga’s view. Can you explain further? (I should add that natural law is a component of uniformity, but the latter gives a wider perspective).
      (2) “Time itself performs miracles” – this is a quote from Wald’s paper on rare events in science, where he effectively points out that, with very low probability events it becomes difficult (if not meaningless) to distinguish between a ‘natural’ event, and a miracle (understood as an ‘impossible’ event). So I wasn’t trying to suggest that miracles are ‘regular’, but that from an empirical perspective they become impossible to distinguish from rare natural events. And yes, if you take the viewopint that you’ve suggested about the probability of such rare events happening in the biblical period then it follows that in probabilistic terms they become extraordinarily unlikely. However, my piece doesn’t try to make that claim, since I’m not making any theological points about the meaning of any miracles, nor the cause, but simply trying to look at what it means to rty to establish a miracle on ‘objective’ terms.
      (3) Physics-based naturalism – yes, I think that I gave four good reasons early on in the paper why it’s pretty shaky, e.g. the problem of human consciousness.

  2. Thank you for such a prompt reply. Plantinga seems to me to be putting forward a view of knowledge as containing science but having that “wider perspective” that also includes our lived experience, similar to the way you are using the idea of uniformity. I suspect that he would also agree with your view that bridge-building between scientific naturalism and subjective experience is rather self-defeating in that it implies that they really are two separate realms.

    I’m still puzzled as to what you are trying to say about miracles. You say “I’m not making any theological points” on the meaning or cause of any miracle but looking at what it means to establish a miracle on objective terms, but surely the whole point of a miracle is its meaning as a planned, supernatural event. If you take that away all you are left with is a highly unusual random event that may or may not happen again. Presumably, this could be used to argue that the laws of physics are not as secure as scientific naturalism claims but I imagine it would be quite a weak argument.

    Yes, I see the four problem areas that you’ve listed, but my question was about why you regard them as “increasingly” shaky. I would agree that scientific naturalism has an enormous problem with human consciousness, but cognitive science, neurophysics or whatever names they’ve invented for it, has actually discovered things about how the brain works in the last 30 or 40 years so, while I can see that this hasn’t added much to our understanding of mind, it surely can’t have diminished our understanding as your “increasingly” implies.

    My apologies if I seem to be nit-picking here but it’s not often I get the chance to ask these sort of questions of someone of your standing and experience.

    • Thanks. Perhaps my “increasingly shaky” phrase was more rhetoric than precision, but I had in mind the four complaints, which have all emerged from the philosophy of science over the last decade or two, in response to physics-based naturalisms. In other words, although the general perception in the public (and theological) worlds seems to be that physics says more fundamental things about nature than any other science, philosophy has moved away from the heavy emphasis on reductionism which such a claim entails.

      Your other question about miracles – yes, if everyone was to agree that a remarkable event was deliberately caused by a deity with a specific purpose which was apparent to all, and everyone agreed on that, then said remarkable event could presumably be accorded the objective status of miracle. The trouble is that I don’t know of any supposed miracle stories like that. Even scriptural accounts of miracles are full of questions when approached from a naturalistic perspective, and that even if everyone was to agree that the deity had caused the miracles for a clear purpose. Hence the discussion in this blog post, over the difficult of trying to establish what it means to talk about miracle from a naturalistic perspective.

  3. Thank you for that clarification. Of course, clarification reveals more questions! However, I’m not sure how to phrase the question I want to ask in a way that makes sense.

    You say, “My complaint with scientific naturalism…isn’t just the inherent reductionism – which is too metaphysical for an empiricist like me – it’s the fact that an over-reliance on the laws of nature (which are, by definition, timeless and a-historical) dominate over historical contingency.”

    And in your introduction you define reductionism when you say that you want to “explore what happens when you question the assumption that physics is the primary science, and that all others are derivative in some sense.”

    Now, I presume, as a physicist you’re not doubting the empirical evidence of the success of physics but that other areas of inquiry, such as social and psychological phenomena, have mistakenly assumed that following the methodology of physics would make their enterprise properly scientific and that, in your view, these areas need to take into account historical and social contexts in order to make progress.

    This is a complete turn around from the post-war view of Popper, Gellner and others who were sure that the social sciences needed to free themselves from the “dead hand of philosophy” as a way of producing real results (I remember, at university, reading a review by Gellner of Peter Winch’s book “The Idea of a Social Science” in which Gellner described Winch as “like a Philosophic Superman flying in to put things right for the earthbound social scientists”) and I agree that this attitude has apparently failed to achieve anything that looks remotely like useful science. However, one of the things that Popper was extremely concerned about was that historicism would open the door to all forms of relativism. I’m thinking, for example, of Alisdair MacIntyre writing that, “Hegel and subsequent historicists were right in claiming that morality which is no particular society’s morality is to be found nowhere.”

    This is a view that leads many Christian groups to be antagonistic towards postmodernism but MacIntyre and other British postmodernists seem to have found a solution to the problem of relativism by themselves becoming Christians, which makes me wonder if postmodernism can only work in a religious context. Is this a problem that you are aware of in any way?

    • Gosh, that’s a very interesting response, and you have made me realise (as you did with your last comment) that I should write another section in the book for clarity (this time about the unity of science). I’m afraid that I don’t know whether postmodernism only works within Christianity, but it seems like a surprising result if true!

      You are indeed right that I do not for an instant doubt the efficacy of physics, nor any other natural science, to probe the natural world, and to provide new knowledge of it. What I have grown weary of, though, is the intense reductionism that philosophers until recently used to apply to the problem of how to unify the sciences (and which many scientists seem to assume). Perhaps it’s my identity as a condensed-matter physicist which makes me doubt reductionism – after all, there is no way that the many-body problems I work on can be reduced to particle physics, and it’s not clear they ever will in any useful way – but I recently discovered that many philosophers of science have also given up on reductionism as the unifier of the scinces. Some embrace pluralism, and I think there is also a concern to avoid the postmodern relativism. For myself, I don’t quite know what pluralism means in this context, but I definitely want to avoid relativism. In any case, I cannot get away from the empirical observation that there are many sciences that do not clearly relate to each other, except through shared subject matter. This is perhaps why I am (and always have been) so impressed with the earth sciences, because they span all of the classic natural sciences (with all of the many methodologies therein, along with interests in law, uniformity, history and contingency), but are united by their subject matter. That is essentially what the book I’m working on is trying to make sense of, while also reviewing thinking on miracles simultaneously.

      I hope this is beginning to make more sense…

  4. In a way I’m gratified to find that other people are as puzzled by this as I am. I’ve spent a couple of days trying to catch up with the current views of pluralism in science but I must confess that my science education went no further than 3rd year of secondary modern (as we used to call it in the good old days) school so my understanding of this is very much in the context of political philosophy. However, this may provide an interesting perspective.

    I think your feelings about reductionism parallel my own dislike of analytic philosophy and that those feelings grow from similar roots. What interests me as much as anything is the language that the debate is expressed through. Earlier this year I was having an online conversation with someone about artificial intelligence in which I put forward the idea that general artificial intelligence is unlikely because human intelligence is embodied in an individual who exists in a particular time and place. My fellow conversationalist replied with, “Do you mean this: ‘The embodied paradigm, originally developed in the fields of theoretical biology and linguistics, holds that human cognition, even in its highest level inferential processes, is rooted in sensorimotor processes that link the agent to the world in which she is embedded’?” I had to reply that, no, I didn’t mean that because those words actually had the effect of disembodying the idea. I think philosophy often works as the result of the creative impreciseness of ordinary language. Perhaps Wittgenstein, although initially a member of the Vienna Circle, left it as soon as he realised that Carnap and the others had read something in to the Tractatus that he hadn’t intended; the academic view set in stone now is that he came back later with a completely different philosophical view but perhaps it’s the same view expressed differently.

    Where does this intersect with the debate about reductionism and pluralism? My daughter (who has the advantage over me of a Physics degree) looked at this conversation and said, dismissively, “Oh, he just means maths!” By which, I presume she meant that some science can only be expressed through maths, some by various combinations of maths and ordinary language and some by ordinary language only. Not being a scientist, I have no idea whether this is meaningful in any way. However, remembering the difficulties that postmodernist philosophy had with trying to distinguish itself from “New Age” pseudo-philosophy (the only solution being a pragmatic one – it either makes sense or it doesn’t) and thinking about the way analytic philosophy slices ideas so finely that they go beyond any possible line marking the point at which they lose their connection with normal human experience, would that type of pragmatic (and debatable) classification avoid the problem of reductionism in a way that doesn’t lose itself in the philosophical tangle of the search for absolute epistemological foundations?

    I think it’s very difficult not to see this as two distinct opposing sides, partly because that’s the ideological basis of the modern world derived from Descartes and the Enlightenment, partly because the 19th century idea of strictly-defined disciplines still has a strong influence in academic institutions, but, looked at in the wider context of the gradually spreading post-modern opposition to narrow epistemological views, I think that an oppositional analysis is unlikely to make sense of the debate. Actually, I imagine a Divinity Department is the ideal place for the sort of multi-disciplinary approach that this needs.

Comments are closed.