Joshua and Cornelius get to know each other

Yes, that was left behind long ago. Haha. Sorry. Couldn’t resist.

No, this is absolutely not true. This false claim is repeated over and over, but it is not true. Both convergence and divergence are rampant. And several other patterns, such as extreme conservation, that make no sense. Of course you can always say, “well evolution did it,” because evolution and CD are so flexible. But don’t then say the data fit the theory. They don’t. If you don’t believe me, take a look at Klassen, et. al., 1991, Figure 6, which shows the consistency index (CI) for a whole bunch of data sets that had been compiled in previous studies (data set = character data for a set of species).

The figure shows the CI value plotted as a function of the number of taxa in the data set. The CI values were consistently terrible. This was especially evident in the more meaningful studies, with greater #taxa. Above about 30, almost all the CI values were less then a half! They ranged from about 0.2-0.4, with random being around 0.15. In other words, the results were far closer to a randomized data set. This meta study incorporated a lot of data, and showed that you are usually closer to a randomized data set than anywhere close to the CD model.

Again, we all know that evolution and CD can do anything, but the claim that “common ancestry makes excellent sense” of the observations is simply false. You can always make Theory A fit Observations B; that doesn’t mean there an excellent fit. There isn’t in this case. This notion that the biological evidence just excellently fits evolution and common descent is simply a myth.


I have not finished reading your link, but I find this extract from your link interesting:

" This means that evolutionists cannot model the observed structures and pattern of distribution merely as a consequence of common descent. Instead, a complicated evolutionary history is required (Brown) where the pentadactyl structure re-evolves in different lineages, and appendages evolve, are lost, and then evolve again. And as one recent study concluded, “Our phylogenetic results support independent instances of complete limb loss as well as multiple instances of digit and external ear opening loss and re-acquisition. Even more striking, we find strong statistical support for the re-acquisition of a pentadactyl body form from a digit-reduced ancestor. … The results of our study join a nascent body of literature showing strong statistical support for character loss, followed by evolutionary re-acquisition of complex structures associated with a generalized pentadactyl body form.” (Siler and Brown)"

Can you elaborate, especially by way of considering the counter arguments proponents of CD would bring?

This is tedious. I am a natural theology/design/IDer, and I’ve explained that several times already.

Good question. Theory evaluation is tricky. It can be quite conservative; that is, proponents often can sustain a lot of contradictory evidence. There are all kinds of possible explanations, such as: (i) there’s something wrong with the observations, (ii) we’ll discover the rest of the story sometime in the future, (iii) the theory actually can explain the observations just fine with some modifications, (iv) arguments against the theory are tantamount to arguing against science itself, etc, etc.

So for this pentadactyl falsification, and many others like it, (iii) is often used. As you can see from the quote above, it if fairly effortless for evolutionists simply to say: “Oh, interesting, well I guess evolution is more complicated than we thought. Who would have thought, it evolved, and re-evolved, and re-evolved, etc, etc.”

Also, contradictory evidence can be considered as “anomalous,” whereas confirming evidence is “normative.” And so a failed theory can live on for quite some time.

Is there any attempt to use genome data with respect to a phenotype that can than be shown to change over time through discreet changes in the genome (or more generally specific mutations)?

But you refuse to engage with the questions I asked of you? If common ancestry doesn’t make sense of the data, then surely you have a better model in mind?

I can go online right now - as can you - and verify that the human-chimp DNA identity value is above 95% using publicly-available databases. I can verify that human and chimpanzees have nearly identical proteins. I can verify that humans, and other placental mammals, have vitellogenin pseudogenes. And you say these are merely rhetorical questions, and then refer to papers published in the pre-genomics era as if they present a problem for CD?

Like all scientists, I stand ready to be persuaded by your evidence for a better way of looking at the data. But you’re not making even close to a convincing case.


@Cornelius_Hunter was quoted as quoting “one recent study”: “Our phylogenetic results support independent instances of complete limb loss as well as multiple instances of digit and external ear opening loss and re-acquisition.”

The quote comes from this article…

Cameron D. Siler,
Rafe M. Brown
First published: 9 May 2011Full publication history
DOI: 10.1111/j.1558-5646.2011.01315.xView/save citation
Cited by: 15 articles

The sentence comes from the abstract, which is fully quoted below:


“Evolutionary simplification, or loss of complex characters, is a major theme in studies of body-form evolution. The apparently infrequent evolutionary reacquisition of complex characters has led to the assertion (Dollo’s Law) that once lost, complex characters may be impossible to re-evolve, at least via the exact same evolutionary process. Here, we provide one of the most comprehensive, fine-scale analyses of squamate body-form evolution to date, introducing a new model system of closely related, morphologically variable, lizards.”

" Our phylogenetic results support independent instances of complete limb loss as well as multiple instances of digit and external ear opening loss and re-acquisition. Even more striking, we find strong statistical support for the re-acquisition of a pentadactyl body form from a digit-reduced ancestor."

" Our study reveals that species of the genus Brachymeles exemplify regions of morphospace (body plans) previously undocumented in squamates. Our findings have broad, general implications for body-form evolution in burrowing vertebrates: whatever constraints have shaped trends in morphological evolution among other squamate groups (excluding Bipes) have been lost in this one exemplary clade."

" The results of our study join a nascent body of literature showing strong statistical support for character loss, followed by evolutionary re-acquisition of complex structures associated with a generalized pentadactyl body form."

1 Like

Two things:

(1) You completely ignored the DNA analysis I presented. In the ordinary world, when multiple lines of evidence confirm the same hypothesis, its validity is strengthened. In the world of Dr. Cornelius Hunter, though, it seems the multiple lines of evidence that align with the same hypothesis are really just multiple instances of circular thinking. 10,000 lines of evidence validate the same hypothesis? In the world of Dr. Cornelius Hunter, that’s just 10,000 instances of circular thinking.

(2) The inevitable result of your rhetoric is a descent into the deepest pits of post-modernism, where one’s biases and preconceptions drive every conclusion and there can be no meaningful way to sort out competing claims. Some people seem to prefer this kind of uber-post-modernist world because it allows them to effortlessly refute those who disagree with them. “You are only interpreting the evidence that way because you are engaging in circular thinking.” How can that possibly be refuted?

As for me, I don’t want to live in that world. And I am not interested in discussing disagreements with someone who inhabits the uber-post-modernist world; we have no possibility of gaining insights from one another.

We do agree on some important things, Dr. Hunter. We’ll just have to limit our discussions to those topics, I reckon.

For about 400 years, the ability of the scientific community to make progress has rested on its shared commitment to a Baconian metaphysic.

If we cannot agree to a common metaphysic of methodological naturalism, there is no point in continuing this discussion; we could not possibly come to any agreement.

Moreover, there is no point in your participating in the modern scientific endeavor. The work of the scientific community is built on methodological naturalism. If you really consider methodological naturalism to be a dressed-up version of Epicureanism, why would you even be interested in being a biologist? What is to be gained by dancing with the Epicurean devil?

Clearly you have abundant gifts that would enable you to be of service to the Lord and to the world outside the realm of biology. Heck, plenty of jobs in big data analytics are available to a talented computational biophysicist interested in making a transition into data science. Private message me if you are interested.


Chris Falter


Sure, there is plenty of evidence that supports evolution/CD. Let me make several points that I think are salient, as I have studied and thought about this quite a bit. Warning: This will not lead to a hard and fast “here’s the answer” conclusion, but will, I hope, advance our thinking.

  1. There is a lot of data that support evolution/CD. But supporting evidence does not prove a theory. Otherwise geocentrism would be proved.

  2. There is a lot of data that do not support evolution/CD. Nothing has changed, conceptually, with the genomics era. Also, I have referred to recent papers as well (such as in the AS discussion above). You can go back two thousand years, the arguments have not changed. The microscope did not change anything. Modern chemistry, molecular biology / Perutz / Watson & Crick, Edman degradation and protein sequences, DNA sequences, electron microscopes, NMR, micro arrays, genomics, etc., etc. None of these conceptually changed the origins discourse, they just amped up the data. The complexity got that much more complex, the dysteleology got that much more pronounced.

  3. In light of 1 and 2, we need to be careful of confirmation testing.

  4. Quantity, by itself, doesn’t count for much. Geocentrism is supported by a lot of data, but no one would say it is true.

  5. It is important to understand what evolution/CD (and any theory) can reasonably accommodate (Steve also made this point and we agreed). If a false prediction and unexpected observation can, in fact, be reasonably accommodated with reasonable tweaking, then don’t count it as a killer false prediction. OTH, if the observation isn’t so easily accommodated, and the theory is getting unreasonably stretched, then we’re having problems. It is important for proponents not to ignore this.

  6. There are a whole lot of instances of #5 (observation that are not easily accommodated) for evolution/CD. Like Mars in retrograde motion, we can’t just sweep these under the rug. There are serious problems for evolution/CD. Just as one example, and Joshua made this point as well above, we are finding that it is not so much the protein coding genes that create the different species, but higher-order regulation and control. Gene expression (TFs, microRNA, etc.), alternate splicing, alternate transcription initiation and termination, epigenetics, etc., are all examples of this higher-order functionality. Joshua called it remarkable. It is incredibly remarkable–a profound, general, finding, and it does not fit evolution/CD, for all kinds of reasons. For instance, they require elaborate mechanisms to be in place, a priori. They greatly enlarge the design space. And often times, you’re not going to get a fitness improvement until much later. What these findings do suggest is design. Otherwise the serendipity is enormous (e.g., you just happened to create all these genes, and all their exon boundaries, and they just so happened to be the needed parts to help make a human, given new splicing programs). Likewise, it isn’t good enough merely to say that contradictory patterns (e.g., the rampant convergence and divergence) don’t really matter because of #1, all the supporting data. Again, volume doesn’t save the day. And realize that these very terms (convergence and divergence) are theory-laden. They carry a message of common descent, even though they contradict that model. We could just as easily use some other terminology.

  7. There are all kinds of examples of dysteleology. They may be inefficiencies, or strikingly similar patterns of useless structures in different allied species, etc. These provide no positivistic support for evolution/CD. IOW, from a positivistic perspective, these are similarities between species, that’s it. They provide no more evidential support than the most complex, sophisticated structures that are also similar between species. In fact, as Sober has pointed out, they actually are slightly negative evidence for evolution/CD, precisely because of their uselessness/ dysteleology. But they provide enormous evidence/argument against design/creation. This is contrastive thinking, and it is very powerful.

  8. #7 entails an assumption of very high knowledge. It is very difficult to establish these things definitively (e.g., that “backward photocell in our retina makes no sense,” claim) Claims of inefficiencies, disutility, etc., have gone south at an alarming rate. If there ever was a lesson from the history of science (and in particular history of evolutionary thought), it would be that such claims have a poor track record.

  9. #7 reflects evolutionary thought, rather than design, in that in the former we would normally expect a lot of useless structures, and a biology that isn’t all that complex, because after all it just evolved, so we could understand it fairly easily. So #7, to a certain degree, actually entails evolutionary thought, and so is theory-laden.

  10. #7 is metaphysical because it specifies the teleology–how the world should/would be designed/created. That doesn’t mean it is wrong. But the metaphysics provides the strength. It is precisely the metaphysical assumptions about how the world should/would be designed/created, and how the observations contradict that teleology, that make the argument strong. The metaphysics is inherent in the argument. You can’t have both ways. If you try to withdraw the metaphysics, then the argument makes no sense. It has no power. All it would be is slightly negative evidence for the theory you are trying to promote.

  11. It would be very difficult to argue that #7 is biblical. John Walton notwithstanding, there is precious little, if any, biblical support for #7, and plenty of scriptural negation of #7.

  12. So you’ve got #7 and #6 going against each other. #7 has been highly influential. As usual, religion tells science what to do.

  13. One common objection, and defense of evolution/CD, is that “you don’t have anything better.” Well I hope that by spelling out the above dozen items, we can see this is a tricky question. First, some claim that evolution/CD works really well. Well so does geocentrism and the flat earth. There is nothing wrong with any of these. But working well, for whatever problem you are attacking, is a very different thing than being true. So I would caution against making that leap without careful thought. Second, I don’t see what’s so bad about a design approach. Well, you say, it doesn’t provide much detail or mechanism. True enough, but that may not be needed for many applications. Furthermore, given #6, perhaps it is better to have a less detailed theory than a more detailed theory that is problematic. In fact, this will be a shock to evolutionists, but science doesn’t actually need theories. Such sentiment is anathema to evolutionists. But this debate goes way back. In the 17th c the Royal Society empiricists argued against the rationalists, saying we’d rather pursue our investigations without undue theoretical constraints. I’m not taking sides here so much as trying to point out that there are a multiplicity of legitimate philosophies of science. The rejoinder “But you don’t have a theory,” is really a very loaded spring. There is an entire philosophy behind it, and without getting the cards out on the table, and exploring all the premises, there isn’t much to be gained by entering the trap. For instance, as Joshua and evolutionists in general often say, science must be limited to naturalistic explanations. Well that puts an entirely different spin on the question “what’s your theory?” So this whole question of “what’s your theory?” deserves a much deeper and clearer examination than the usual flip demands. There are a lot of moving parts here.

  14. I’m not expecting concurrence here, but I hope what I’ve said helps move the thought process along. With that, I have to move along. I’ll give you the last word, which I will read.

1 Like

Just one point on this thread, Chris.

I believe it was you who first raised Bacon in the rather familiar “science has given us mobile phones” style of argument.

Bacon happened to be a name that quickly came up in my brief Internet search to understand better where Cornelius Hunter’s approach is coming from, on seeing this thread. Since he was fighting on several fronts, I’m not surprised he didn’t answer your point directly. (If it had been me I’d have been sure to disrupt the thread by saying that the Cistercians gave us blast furnaces; Aristotle, via Bede, gravity; Taoism gunpowder and paper and Adolph Hitler the VW and space travel). But here’s a quote from a careful Amazon review of one of his books:

Hunter begins by contrasting the methodologies of two great thinkers: Francis Bacon and Rene Descartes. The former was an empiricist whereas the latter was a rationalist. Bacon was not very interesting in constructing a grand metaphysical or philosophical system within which science could operate. Instead, he simply advocated performing tests and following the data. By contrast Descartes was obsessed with the question of certainty, and constructed a mechanical vision of nature wherein all things were ordered by God so that they might accomplish their purpose without outside intervention. As Hunter shows, this view of nature became a powerful justification for theological naturalism. Theological naturalism, says Hunter, does not at all equate to atheism or metaphysical naturalism. On the contrary, it was often justified by arguments about the nature of God. Particularly influential was the notion that God would be greater if He were able to design a self-sufficient system which could produce all the order in the cosmos without outside intervention.

In other words, Hunter is a supporter of Bacon’s science: his objection is to the contrasting approach of Descartes (and later, the Enlightenment Deist rationalists like Leibniz - who of course was a direct metaphysical opponent of Newton in saying that God would obviously make the cosmos to run entirely on its own). Bacon, in fact, was not of this persuasion. Here’s a source on his views:

Not only did God arrange nature according to laws which made it predictable, but in his providential actions he also proceeded according to consistent principles and patterns. Although God’s actions were not predictable, except in so far as He revealed His intentions through prohecy, His consistency made the actions of the hand of providence recognizable to those who knew his ways.

Bacon, in other words, held that there are both lawlike and divinely contingent events in the world, which require to be distinguished. Descartes saw nature as a closed system in which no events were divinely directed: leading indirectly to an undeniable category of event which is not lawlike, cannot (on principle) be providential, and so must be due to a third causal category, Epicurean “chance”.

So as soon as one invokes “chance” in a scientific argument, you have opted for the metaphysics of Leibniz over Newton, and for that of Descartes over Bacon. There does seem to be a valid question of why this should be the correct option for contemporary science.


Your 14 points, in the absence of data derived from Geology and Physics, is an exercise in Sophistry.

You say nothing has changed in 2,000 years … but everything has changed in 200 years of geology and 100 years of cosmology.

George Brooks

1 Like

One of the greatest strengths of the scientific method is that NO theory is ever considered to be proven.

So what’s your point in throwing around “proof,” other than misrepresenting the most fundamental aspect of science? What’s your hypothesis, Cornelius? What empirical predictions does it make?


Hi Jon,

Hope you are doing well today.

This is a truly odd argument, as neither Bacon nor Descartes envisaged quantum mechanics, or mathematical modeling of chaotic systems like weather and DNA mutations. We have come to the place where “contingent” processes can to a very great extent, it seems, behave in a lawlike fashion. Many scientists who use stochastic models to explain nature do not feel bound by Cartesian categories of natural determinism vs. Epicurean chance. The rule of thumb you present (100% lawlike = Cartesian, X% lawlike + (100 - X)% divinely contingent = Baconian) seems far too oversimplified.

Moreover, I vehemently reject the suggestion that incorporating stochastic elements into a scientific model renders it, ipso facto, into an Epicurean exercise in which God is absent and the end results are all the product of blind chance.

First of all, Epicureanism–the view that God is aloof from His creation–can pitch a tent even in the absence of stochastic processes. Any deterministic process will do. For example, physicists model baryonic matter as elementary particles like quarks that self-assemble under the right conditions into protons, neutrons, and electrons, which then self-assemble into atoms and molecules, etc.[1] In the physics models, these particles seem to be utterly independent of any divine action or providence. To an Epicurean, the success of relativity and the atomic model can seem like a strong confirmation of their worldview.

Nevertheless, Christians affirm by faith that God upholds baryonic matter and the equations of relativity, and that He continually sustains them through His providence.

Likewise, a computational biologist can model the occurrence of various types of mutations with a probability mass function. (I haven’t done it myself, but I assume it’s what they use.) You can label that “chance,” just as you attach labels to elementary particles. Just like the elementary particles in physics models, the probability mass function in a genomic evolution model seems to be utterly independent of any divine action or providence. To an Epicurean, the success of evolution and genomic models can seem like a strong confirmation of their worldview.

Nevertheless, Christians affirm by faith that God upholds the probability mass function(s) and continually sustains them through His providence.

John Polkinghorne, with whom you are no doubt familiar, goes even further. He contends that a stochastic model is quite amenable to God’s intimate, providential interaction with His creation in a manner undetectable by science. Because of stochastic processes, God can subtly intervene through the injection of information, and no scientist can detect His hand. We don’t categorize these interactions as miracles because they seem lawlike; the probability mass function still seems to hold. So stochastic processes and God’s intimate, providential involvement with creation are not antitheses; they are in fact complementary. I’ll stand with Polkinghorne on this issue.

By imposing the false dichotomy between lawlike, providential processes and stochastic, Epicurean processes, you are unnecessarily driving a wedge between Christians and scientists, in my opinion. Theologians like Polkinghorne and scientists like @Swamidass point to a better way to deal with the intersection of stochastic models and faith.[2]

Thanks for listening, Jon. It’s always good to chat with you!

Chris Falter[3]

[1] Note that sub-atomic and atomic processes are also quite stochastic, as described by quantum physics.

[2] I want to clarify for the record that I am not addressing the question of openness vs. determinism. Both of them can be reconciled with stochastic processes. Some folks seem to think that stochastic processes can’t be squared with determinism, and might want to open a new discussion on that topic. If anyone feels so inclined, sure, go ahead and start a new thread. But don’t count on my participation; I have too much else going on at the moment.

[3] I also want to take this opportunity to mention that I am planning to formulate a reply to Dr. Hunter’s gracious and most recent post, but it will probably take a week or longer. I have other duties that press.

1 Like

In my time away from this conversation, tt seems that @Cornelius_Hunter has given his final thoughts.

As the @BradKramer -bestowed-name of this thread is for @swamidass and @Cornelius_Hunter to “get to know each other,” I will be giving my concluding remarks in a couple posts in the next couple days. I am swamped with my work, but have a high priority in making time for meaningful exchanges like this.

Before I make those comments, I want to start by thanking @Cornelius_Hunter for dialoguing with us here. It is never easy to engage an audience where a large number of people disagree with you. He has done just that in this dialogue, and that effort deserves the respect of all of us.

Moreover, I very much admire how he handled some “less than ideal” comments to him here.

I agree with @Casper_Hesp and appreciate his apology to you. I winced a bit when I read Casper’s initial comment, but then laughed out loud at your humorous reply. That was very kind of you to “overlook an offense” and really embodied the best of civil dialogue to whimsically respond with humor. I personally hope to emulate you there, and think that interaction is an example to all of us. Thank you for that.

Also, I agree with deeply with your final comment, about “not expecting concurrence here, but I hope what I’ve said helps move the thought process along.” My goal is that we might “understand and be understood.” In this, I feel like real progress has been made. I really do understand your position better, and can identify several places where I had you pegged wrong. I’m very glad to be corrected to more accurately represent you in the future.

Though I do think you are missing my position, I do hope there has been some movement to a more correct understanding of my position. Maybe my final comments, to come, maybe do that.

Regardless, I want to begin the ending of this conversation with a genuine and “thank you” to @Cornelius_Hunter. We need more conversations like this.


This is pretty tough reading …

Hi Chris,

Discussions on this forum tend to go in many directions, and I want to make one clear point regarding your comments on models and stochastic treatments. Models as I understand them, are “theory bound” and must be assessed in some manner. We may develop a “model” of say, chemical bonds, based on a theory, but the validity and relevance of any such model must be assessed by comparing its results with accurate measurements. If the comparison is good, we attribute a high degree of confidence in both the model and the theory underpinning it. If the comparison is poor, then we have a low degree of confidence in both model and theory.

Stochastic treatments are used in a very large number of scientific projects - this is a tool that can show high correlations with data, of may be a qualitative treatment - so it is part of a scientist’s toolbox. Often when a great deal of data points are considered, the correlation coefficient may be a sufficient guide to the goodness (or otherwise) of fit. For example, papers have been published in which a coefficient was provided for natural selection vs observed/measured data. The correlation was poor, but in this case I do not think they questioned a model, but rather the treatment of the data set. The results may also be used to question the way NS is understood, but the initial response is to question the stochastic treatment.

These brief remarks do not impact on epicurean vs other philosophical matters, but we should see a distinction between stochastic methods and specific models.

1 Like

Thanks, George, for these helpful thoughts on modeling and stochastic treatments.


I want to second @Swamidass ’ comments. Dr. Hunter’s posts made me think deeply about modeling, theories, data, and inference, so I can affirm that the discussion has borne some good fruit.

Thanks Joshua, I appreciate your thoughtful words.


Fine, thanks, Chris: I love my work, love the Lord and live in a beautiful part of the world. But I caught a cold from some musician.:unamused:

Let me try to peel off the science to the metaphysics beneath. Science is the study of repeatable causes. It cannot explain why a potato from your garden bears an uncanny resemblance to Donald Trump, though it might if that started to become a common occurrence.

When Fermat and Pascal started to discuss probability, and far more since Maxwell, contingent events began to be drawn into science because they followed repeatable probability distributions - and it’s that predictability that makes them “stochastic”. I discuss that history here.

What that shows is that the underlying events, though unknowable, are not indeterminate - there is actually method in their madness. I cite the statistician William Briggs in another post to affirm this.

Now you speak of chaotic and quantum events. Neither entail indeterminacy. Nothing in chaos theory requires indeterminacy: in fact it is usually assumed under naturalism that such systems are entirely lawlike, but unstable and so beyond human measurement - if we exclude those systems in which quantum events might, conceivably, be amplified and overcome physical determinacy. Maxwell’s Demon is still in a job under chaos theory, but has to work a lot harder and more accurately to predict the future.

This can be demonstrated by the case of a coin toss, which is a system carefully designed to give a binary probability of 50:50. In that second post I describe how a coin toss machine can be made to give entirely predictable outcomes. The outcomes of any coin are determined entirely by the initial conditions and the design of the system.

Quantum events, however, almost certainly have no causes within known physics (no hidden variables). But again, they are statistically so predictable that we date the world by their statistical outcomes. They are, therefore, “stochastic” - ie they too show method in their madness, and the profound mystery - to scientists at least - ought to be the source of that method.

So having stripped off the repeatable, we come to the causes of the contingent events themselves: the individual molecular trajectories in gases, the exact interactions of multiple gravitational objects in space, the individual decay of radioactive molecules, and so on. Because they are contingent, they are individually no more amenable to scientific study than your potato, and any explanation is “beyond physics” - it’s metaphysics.

For any such event (particularly those giving rise to probability curves in aggregate) there are, that I can think of, only two possible mutually exclusive explanations.

The first is that they occur without a determining cause, yet somehow become arranged into stochastic order. This was the view of Epicurus, and so is well-termed the Epicurean explanation.

The second is that they have an ordering cause beyond the scientific, which explains how they come to form Bell-curves, at one extreme, and individual human beings, at the other. This was Aristotle’s view against Epicurus, but was most developed in the Judaeo-Christian tradition as the doctrine of divine providence.

I (and an erstwile BioLogos poster “Penman”) have done a 4 part blog on the history of this. To summarise (and of course oversimplify), universal providence was held by all pre-Christian Jews apart from the Sadducees, and more or less universally by Christians until the later Deists. They, wishing to distance God from direct interaction with his creation on the grounds of rationalism and anti-supernaturalism, but finding that physical determinism wouldn’t run, reintroduced an Epicurean understanding of chance to the Christian world as a denial of providence. Part 3 of the “history” describes how that developed up to and including Darwinian theory.

And so that is why I say that for any event that one calls “chance” because, being contingent, it is beyond the repeatable patterns amenable to science, one necessarily makes the metaphysical choice of seeing it as undetermined (the Epicurean explanation) or governed by providence (the Theistic explanation).

If, as a Christian, one opts for the former in any particular case, one needs to justify why one has that explanation in ones toolbox at all - presumably using fundamental theological arguments about the nature of God and Creation other than those employed by the Enlightenment Deists, whose day is long-gone. One also has to explain how to justify applying the “chance” explanation rather than the “providence” explanation in any one case - and by definition that will have to be a metaphysical or theological justification, not a scientific one.