Yesterday I surveyed a whole category of arguments for theism, the “Science Needs God” complex. And I concluded by mentioning a (sort of) new one, by a well-credentialed professor of philosophy, Tomas Bogardus (another fashionable Protestant convert to Catholicism): “If Naturalism Is True, Then Scientific Explanation Is Impossible,” in Religious Studies 59.1 (2022). I also watched a video on Capturing Christianity (hosted by Cameron Bertuzzi) where Bogardus discusses this paper for over an hour, so I could be sure I’m reading him right. And, yeah. I am.
I have noted before that even peer-reviewed philosophy as a field is full of bad philosophy. In my randomized survey of three examples I found only one was solid, and even it under-cited the pertinent science; another half failed and half succeeded; and the third should never have passed any reliable peer review. I’ve dealt with even wholly harebrained examples (from The Blondé-Jansen Argument from Consciousness to my whole series on Thought Experiments). Peer review in philosophy is better than nothing, but the field lacks a reliable standard for what should pass. And the Bogardus thesis shouldn’t have, and for much the same reason: it contains demonstrable fallacies of reasoning; and it ignores all the contradicting science of today. No philosophy should ever pass that does either.
The approach of Tomas Bogardus is quasi-Thomist and thus already starts from a bankrupt methodological framework (see Thomism: The Bogus Science). But we’ll treat his argument on his own terms anyway. His abstract correctly describes it thus:
- Any scientific explanation can be successful only if it crucially involves a natural regularity.
- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
- Therefore, a scientific explanation can be successful only if it crucially involves a natural regularity, and this regularity does not call out for explanation while lacking one.
- If Naturalism is true, then every natural regularity calls out for explanation but lacks one.
- Therefore, if Naturalism is true, then no scientific explanation can be successful.
He then concludes, “If you believe that scientific explanation can be (indeed, often has been) successful, as I do, then this is a reason to reject Naturalism.” In Bayesian terms: since e is “Scientific explanation can be (indeed, often has been) successful,” then, given his argument’s conclusion that (as a hypothesis) “Naturalism” predicts no scientific explanation can be successful, P(e|naturalism) is low, whereas it would be high on any suitably formulated nonnaturalism—like theism, the only alternative Bogardus proposes—and therefore this e is evidence against naturalism and for (at least some form of) nonnaturalism; and Bogardus gives reasons why that would be most likely some form of theism.
One could quibble over that, e.g. Taoism would fully succeed at providing a suitable nonnatural cause, yet involves no personal consciousness, and would out-perform theism because it lacks all of theism’s explanatory defects. It’s even simpler as a supernatural alternative because in its most ancient formula, the Tao’s full gamut of powers and properties evolve inevitably from simpler originations, rather than (as with most gods still believed in today) just conveniently existing de novo without cause. And this is all vastly more so for any variety of theism I’ll bet Bogardus has any interest in believing. Theists today typically forget that their version of supernaturalism is historically and innately extremely bizarre and not at all exhaustive of available worldviews.
Since Bogardus never presents any formal argument that the nonnaturalism that must replace naturalism has to even be theism, much less whatever specifically weird form of it he wants you to buy into, his paper isn’t really an argument for God. It intends to be, but only by sloppily (and ethnocentrically) assuming his peculiar cultural shibboleth is the only solution left. So I will only treat his paper as what it formally contains, which is only an argument against naturalism. But I’ll assume he is at least proposing God as an alternative hypothesis, and thus critique that as well.
I have also written the conclusion here as a complete brief of this article, so for those who want the TL;DR and just get a summary of what’s wrong with Bogardus’s paper before going back and reading through the details, you can jump to the end. But for those who want to go on the journey, read on.
Science Illiteracy
A leading problem among Christian apologists today is an inexplicable, and inexcusable, science illiteracy. We find this plaguing nearly their every argument. For example, see Plantinga’s Tiger and Other Stupid Shit. Failing to grasp even rudimentary science is common in Christian arguments that ignore how evolution works, how biochemistry works, how thermodynamics works, even how physics works, or cosmology; and ignores what we know from other sciences, too, like anthropology, sociology, psychology, and cognitive science. Common examples: theists almost never correctly describe the Second Law of Thermodynamics (see Justin Brierley on the Science of Existence), or Evolution or Protobiology (see Three Common Confusions of Creationists), or the latest Cosmological theories (see My Fourth Reply on the KCA).
An example shows up in the Capturing Christianity interview, around minute 43, where Bogardus reveals he doesn’t know that the Gravitational constant doesn’t actually exist: it’s just a unit converter, from arbitrary British “feet” and “seconds” to the physically smallest natural units. And much else: he doesn’t know that inverse square laws are logically entailed by local geometry; that masses multiply their force interactions because that’s just iterated addition from every particle in a mass interacting with every other; that the reason the electron charge equals one is that humans decided to count it that way (change the electromagnetic force any way you want, the charge always stays 1 because that’s what we chose to count—a decision that created a problem when we discovered charges actually do have fractions; what Bogardus I think wants here is something else called the fine-structure constant, which determines the strength of a charge); and so on. One could give scientifically literate examples of what Bogardus wants to say. There are things about physics we have not yet explained, and there might be something that possibly has no further explanation at all, like the fundaments of quantum mechanics. But that Bogardus can’t come up with one is a red flag: he doesn’t know what he is talking about.
It is the first responsibility of any philosopher who intends to pontificate on a point of fact to become acquainted with the state of science pertaining. But Christians just don’t. Even after being shown their mistakes, often they will come up with reasons why they get to ignore all the findings of science, rather than realize and admit that their beliefs and arguments need to be in alignment with the findings of science. Bogardus didn’t walk across the hall and ask a scientist if his description of science actually is how science works. Is it the case that “Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one”? No. This is not a component of any scientific standard in any field. We have never needed any explanation to be complete in order to be successful. That’s how we built a technological civilization that would be described as magic by any ancient observer: by not needing our explanations to be complete to be accurate and correct, and thus successful. So Bogardus’s whole premise is bogus. It’s pseudoscience.
But it’s even worse than that. When in his set-up Bogardus opines about a passage in Aristotle—already a reliance on which is a Thomist affectation illustrative of Bogardus’s pseudoscientific (indeed outright medieval) methodology throughout this paper—we see he has practically abandoned everything modernity has taught us (and Philosophy Must No Longer Do That). Aristotle was impressively right about many things; he is the father of modern science, after all. But he was also wrong about a lot of things. Indeed, Aristotle was proved wrong about a lot of things by his own students, within just a century or two of his time. By now, modernity has moved on, to a better and more accurate understanding of everything Aristotle wrote about (such that, philosophically, we can retain only a skeleton of some of his core ideas: see Why A Neo-Aristotelian Naturalism Is Probably True).
Missed Memo: Aristotle Is Obsolete
This matters. For example, Bogardus seems impressed by Aristotle’s argument that the fact that it rains in winter when needed and abates in summer when needed cannot be explained by chance accident but only by some sort of final purpose—some kind of “design.” Likewise the fact that teeth grow reliably in the same format in most people (and thus deformities are rare). Aristotle was actually face-palmingly wrong about every component of these arguments. Bogardus should be correcting Aristotle here, not trying to build on him. All philosophy is worthless that ignores all advances in science since Aristotle.
Aristotle did not know about equatorial climates (where this attribute of seasonality is not true of rain) or how the tilt of the Earth necessarily causes the rain observations he was familiar with; or how that tilt is a chance accident, which isn’t being “randomly reset” every year because the Earth is darned big and its angular momentum gargantuan, and nothing is around that could alter it. Likewise, Aristotle didn’t know about DNA and how it stores naturally selected instructions for building teeth a certain way, such that the Empedoclean theory Aristotle thought he was refuting (that natural selection causes teeth to grow in the format they do in human mouths) is actually correct—Empedocles just didn’t know information was being preserved across generations, through a microscopic molecular process, which is not controlled by any intelligence but fully operates by the very processes of chance and necessity Aristotle was trying to deny (indeed fully explaining why deformities even exist, and why the “design” in humans, plants, and animals is so inconsistent and ad hoc—and has changed, and absurdly slowly, over deep time, going from simple to complex over a period of billions of years).
Taking DNA for a moment: the idea that this stored-information transfer could be the case would not arise in any of Aristotle’s successors, so far as we know, until Galen five hundred years later. And even Galen only had a sketched-out surmise (about microscopically-stored computer programs in bodies being mechanically transferred to fetuses, sometimes causing transcriptional errors). And he still did not yet have an understanding of genetic physics, or how evolution by natural selection would explain how those computer programs got there, which is again by chance and necessity.
Galen was instead familiar with robotic theaters, a popular novelty built by scientific engineers of the time (we have descriptions in Hero of Alexandria’s Pneumatics and Automata). They functioned by having a hidden system of ropes and wheels and pulleys, arranged in a sequence, such that once you triggered it, it followed a programmed pattern of actions, like having robotic actors appear on stage and make the requisite movements, lighting fires and drawing and closing curtains at the correct time, and so on, producing a replication of an entire multi-act play. Galen thus inferred that maybe there was something like that: not literally pulleys and ropes, but some mechanical equivalent that, once triggered, causally replicated a human body, and was also hidden (in this case inside human fetuses, too small for the human eye to see). This, he concluded, would explain why deformities occur (something goes wrong with the mechanism), but rarely (usually the machinery works).
Galen assumed God must have built these micro-programs, but only because he didn’t know about evolutionary biochemistry: DNA is the rope, wheels, and pulleys. And it replicates, with occasional transcription errors, automatically. No intelligence involved. And over time, as Empedocles predicted, better and better programs are selected, by the bad ones getting more rapidly killed off, more quickly and more often (the science behind this now is vast). So Aristotle was wrong about teeth too. It was all just chance and necessity after all.
Aristotle’s ignorance cannot become an argument. And yet this is what Bogardus proceeds to do. We get a foreshadow in a footnote here where he thinks Alan Guth’s demurring on how he (personally) might go about explaining why all the laws of physics are as they are (or how theists would) means “he has no idea how to approach that question, because he was assuming that he would have to approach it as a Naturalist, and it’s hard to see how a Naturalist could give a scientific explanation…for a truly fundamental law of nature.” But we could do that (see, for example, Superstring Theory as Metaphysical Atheism and The Argument to the Ontological Whatsit and The Problem with Nothing; or, for another example, Victor Stenger’s The Comprehensible Cosmos, where he deduces most basic physical laws from a single supposition: that there is no privileged point of view, i.e. if no one existed to decide the laws of physics, then these are the laws we’d always have in result, and it is thus their lack of further explanation that causes them to be what they are). And had Guth been asked that question, he could have easily explained to Bogardus the entire field of Theory of Everything studies. Bogardus reveals he doesn’t know any of this. He’s not up on the pertinent science. And this tanks his entire argument.
Bogardus also doesn’t know that the reason Guth finds God theories unworkable is that they make no differential predictions. They are pseudoscientific, because they start with physics as-is and simply build God’s decisions and motives around that (producing “just so” stories, which you then have to keep changing as new scientific facts are discovered), rather than (as would be required for a real scientific theory) starting with the God hypothesis and successfully deducing from it specific unique features of physics (like why there are twelve different quarks, or their exact masses and properties, or why biological order requires billions of years of a meandering random walk to generate), particularly new ones not yet discovered (the gold standard of scientific proof for any theory). There is a reason every God hypothesis failed to predict the vast age and size of the universe, the entire history of evolution, and even the existence of a quark. It makes no predictions. It’s just a giant Texas Sharpshooter Fallacy.
But Bogardus doesn’t appear to understand what a scientific theory is or how the scientific method works—nor does he know any of the science relevant to his thesis. He didn’t even know “Theory of Everything” is an actual field of study in physics. But worse than all that, because he doesn’t know any of that, he doesn’t realize Aristotle’s entire methodology is thereby proved invalid. You can’t argue “I see no reason for these regularities, therefore they must be by design.” That’s exactly the backwards medieval error-mode the Scientific Revolution was all about abandoning, and explaining why it must be. Aristotle was wrong to jump to that inference, because background knowledge entails it will too routinely not turn out the way he is inferring, owing to the alternative explanation of ignorance. Since it almost always turns out that there is an explanation for such things, the premise “I don’t see a reason for a regularity” renders the conclusion “therefore it must be by design” almost always false.
In Bayesian terms, e (“I don’t see a reason for a regularity”) is actually not at all likely on h (“It was designed that way”) but extremely likely on ~h (“there will turn out to be some combination of chance and necessity that explains it”). This is why Naturalism Is Not an Axiom of the Sciences but a Conclusion of Them. Any scientist could have explained this to Bogardus or his peer reviewers.
And Then It’s Error All the Way Down
Rather than master any needed science literacy, and thus learn that the Argument from Ignorance is a fallacy, and that the scientific method requires positive evidence for a specified explanatory model and rejecting drawing conclusions about the unknowable (if you have no explanatory model, or no specific evidence for your model as against others, you cannot claim to know what the explanation is), Bogardus literally says, “So, for Empedocles,” per Aristotle, “something that seems to call for explanation in fact has none. That’s a cost. And that’s Aristotle’s insight.” Which is literally saying this: because Aristotle was completely wrong and his inference model therefore completely unreliable, therefore we should copy Aristotle’s inference model and reach Aristotle’s conclusions anytime we are in the same epistemic position he was. Thus Bogardus is basically saying “Aristotle didn’t know how planetary tilt and DNA explains seasons and teeth, and thus incorrectly inferred chance and necessity didn’t cause them; therefore anytime we don’t know the causes of a thing, we can correctly infer chance and necessity didn’t cause them.”
Head spin? Yes. This is a face-palm moment. Bogardus documents a failed method, and concludes it’s a good method. And then proceeds to build his entire paper on that method, the method he actually just showed doesn’t work—because Aristotle was wrong; his method didn’t work. And if modern science has done nothing else, it has so thoroughly proved that that method doesn’t work that we can be reliably certain its results will never be correct. When we discover design, we do it by a sound method: which means, we start with an explanatory model, we look for differential evidence supporting that model over others, and when we find it, then—and only then—do we conclude “design.” Hence we are “right” when we find chiseled spearheads and conclude intelligent design caused them, not because we “lack” any chance-necessity explanation for them, but because we “have” an explanation that makes the features of that evidence far more probable than any known chance-necessity explanation. How chiseled spearheads come to exist and why—by human manufacture, for the function of stabbing, or by random or inevitable geological processes—entails differential likelihoods favoring design. It’s simply more probable they were made by people than by natural processes. And it is made more probable by evidence—not the absence of it.
Even when the absence of evidence is evidence (see my formal analysis in Proving History, pp. 117–19), it’s still a differential prediction: we know the Exodus never happened as described because that could not fail to generate quite a lot of physical evidence of mass human transit and habitance. As a theoretical model, it simply fails to predict the evidence, whereas the null hypothesis (that humans weren’t there en masse at any time relevant) predicts it perfectly (among much else: see A Test of Bayesian History: Efraim Wallach on Old Testament Studies). Assuming “design” always predicts the evidence better than alternatives is one of Three Common Confusions of Creationists. Because it’s based on their Texas Sharpshooter technique of always deciding on whatever elaborate theory would produce the evidence after the fact, rather than honestly asking if that theory is even probable—and being able to correctly explain why it is.
Thus Bogardus’s approach we have empirically shown almost always fails as a procedure. As with Aristotle: he ended up wrong about seasons and teeth; and so has gone pretty much every other deployment of his method in human history. The only time design gets reliably predicted is when, instead of arguing from ignorance, we argue from evidence. For example, as a Supreme Court justice once said, “Even a dog can tell the difference between being tripped over and deliberately kicked.” A dog is not aware that it has information, about people, emotions, patterns of behavior; but it nevertheless learns from that information what patterns correlate; and those correlations then become evidence for which causal model is correct, making one more probable than the other. And this is how humans actually detect design, such as in history or world affairs (and would detect it in other domains, such as SETI). When we have evidence for design being an operating cause, then—and only then—do we start to reliably discover it. That is the scientific method.
And any scientist could have explained this to Tomas Bogardus. And that should be the peer review standard. Philosophers who pontificate about science without even asking a scientist first (or at least consulting what they’ve written on the point) are bad philosophers. And they shouldn’t be able to pass peer review until they’ve proved they really met the conditions of this required step. And that means properly. Just talking to a biased creationist scientist, for example, or any other who conveniently is bad at this or doesn’t specialize in it, rather than surveying the entire field to find the best conclusions of scientific opinion (on such things as the rudiments of scientific method, for example, or the prospects of a Theory of Everything), is an invalid procedure. This is The Scary Truth about Critical Thinking: you have to try your darnedest to disprove your claims before you can justify believing them. Siloed ideology is just editorializing, not philosophy. Competent falsification tests have to be real tests. And that means not cherry picking your scientific advisers. It means honestly looking for any opposition’s steel man before reaching a conclusion. This should be the required method in all philosophy publications.
It is thus fatal to Bogardus’s paper that it contains no literature survey of leading scientific “ultimate explanations” (from M Theory to specific models like Stenger or Wolfram or Tegmark or Everett; or the likes of Lincoln and Wasser, on which see my conclusion below; and beyond). And he never examines the best of them as examples to apply his argument to, asking in what ways do these fail whereby theism does not. No paper making claims about such things without including such things should pass peer review. That’s like publishing a paper making unusual statements about the biology of crows, in which you never cite or discuss any actual science of corvid biology.
Where I Agree
Of course, Bogardus doesn’t have a theory, any more than Aristotle did. He quotes Aristotle’s hand-waving (essentially, what Aristotle said was, “How does the prime mover move anything? He just does.”) with approval (rather than correctly, as actually dodging the question). Just saying God “wants” to move things, and that this just magically “causes” them to move, is a non-explanation. Indeed, it’s a gigantic monstrosity of endless epicycles of non-explanations: see The Argument from Specified Complexity against Supernaturalism and Evan Fales’ thorough exploration of the problem of divine causation in Divine Intervention: Metaphysical and Epistemological Puzzles (Routledge 2010). There is a reason Aristotle’s second successor, Strato of Lampsacus, abandoned the idea of a Prime Mover—because it doesn’t work, and isn’t needed (see Why A Neo-Aristotelian Naturalism Is Probably True).
So one might then wonder if Bogardus is contradicting himself. If he doesn’t have a “complete” explanation either, then how can his conclusion that there is a Prime Mover be justified? Well, he could simply state that this is not a scientific explanation, but a philosophical one, and, once granted, it solves the scientific explanation problem; and that that is the reason we should adopt this philosophical conclusion. That would then make his argument a Fallacy of Special Pleading: he is claiming science is committed to a bizarre methodological requirement that other approaches to knowledge are not. And then stating no reason why. Which is what we’ll see in my next section leads to the core flaw in his argument. But apart from that, we can acknowledge that Bogardus is not saying he has a working Theory of Everything; he is saying he has a stand-in proposal that fills a gap between observation and scientific methodology that, if left unfilled, leaves scientific methodology inexplicable. That’s why this is another “Science needs God” argument. So we will have to tackle that argument as-is.
So, yes, his approach is convoluted, and ultimately bogus, but not because it is self-refuting. If it really were the case that we “could not” explain why a method worked “but for” God, that would be evidence for God. Imagine, for example, if petitions to the divine actually worked. For example, the prayer of a truly believing Christian, and only a truly believing Christian, actually could relocate a mountain on demand (but always only ever in alignment with the stated values of God, proving the power came from that God and not some wizard merely claiming it does); or if whenever we asked the Oracle at Delphi advanced physics questions, properly addressing Apollo for an answer, and were able to confirm empirically the Oracle was always correct. These would be methods hard to explain without some sort of at least quasi-divine agency. But no methods like this exist. Nor does modern scientific method require any (again, see The Myth That Science Needs Christianity). But though it lacks evidence, Bogardus’s theory isn’t self-contradictory, at least; so we can at least give it a shot.
I also won’t quibble with Bogardus’s “admittedly vague” definition of naturalism. He recognizes no one in the field can agree on one, and settles on a circular definition, whereby naturalism just means only natural causes exist, without explaining what makes a cause “natural.” One could do better than that (see my article Defining the Supernatural), and the effort to would expose flaws in Bogardus’s assumptions (the supernatural looks even less plausible the more substantively you try to define it). But for the purposes of this paper’s argument, his uninformative definition is adequate, as it includes any definition of naturalism more substantively defined—so if his conclusion carries, it carries for that, too.
I also won’t take too much issue with Bogardus’s archaic appeal to “laws of nature.” Because he is aware most science does not traffic in laws of nature per se. Psychology, sociology, anthropology, even botany, zoology, and geology involve a great deal of explanations not really formulatable as “laws.” Most of what science does is simply catalog what exists. And next after that, most of what it does is provide causal explanations, which do not have to be lawlike (they do not always require that certain conditions always produce certain outcomes). And a lot of what science has found everything reduces to is random (from scientific weather predicting to quantum mechanics); but even that, presumably, has a causal explanation—and that, Bogardus says, is close enough for what he means: there are things that beg explanation, whatever you call them.
For example, that the Mosuo people of China employ a system of open marriage whereby a mother raises her children with her brother, not their biological fathers, is a documented scientific fact; but it involves no law of nature. One can reductively explain it with laws of nature (anthropology reduces to sociology, which reduces to psychology, which reduces to biology, which reduces to neurochemistry, which reduces to biochemistry, which reduces to physics); but one does not have to. This fact of the Mosuo is a scientific fact regardless of whether or which laws of nature have anything to do with it. Likewise, anthropology and sociology do propose (and evince with evidence) certain economic, psychological, and sociological reasons why such marital systems arise instead of others, which are causal accounts, but they do not operate by anything like a “natural law.” Rather, they only attest what usually (not always) causes such outcomes, and require even for that just a basic notion of causation, not “laws” per se. That there just happen to be laws of physics underlying it all Is a Historical Conclusion of the Sciences as a Whole, not a necessary feature of any causal theory of specific human cultural developments.
But Bogardus asks us to be charitable and assume he means, simply, uniformities of nature. The Mosuo people don’t randomly wake up every day and start following a different system of marriage. Something maintains the persistence of the observation. Scientists appeal to the fixity of neurology, for example, and the documented ways limiting how (and how fast) it can change. These all assume regularities, whether involving fixed laws or not. Neurons don’t just “randomly reconfigure themselves” (at least not usually); social change requires certain processes of mass human thought; and so on. The magical notion of free will, as in a will wholly unimpacted by causation, has been documented not to exist, after all (elsewise we’d see all social customs, like ours or the Mosuo’s, dissolve overnight).
So if we set aside Bogardus’s archaic use of phrases like “laws of nature” and just replace that with a philosophical equivalent that correctly describes what scientists are doing, “natural uniformities” will suffice to qualify. This then includes “laws of nature” by any relevant definition, as well as mere “facts of nature,” and all explanatory causal models. “Why do things stay the same unless caused to change; and why do causes exist; and why do they always or usually operate a certain way,” can all be appealed to as a request for some sort of underlying “law(s) of nature” in this sense. As we’ll see, this just starts to look like another appeal to An Ontological Whatsit.
The Failure of the Bogardus Thesis
Things really go south at his Premise 2. Here Bogardus tries to insist that it “is a deficiency of a theory if the theory leaves [anything] unexplained,” e.g. if Archimedes’ Law of Buoyancy (see my pertinent discussion of that in All Godless Universes Are Mathematical) doesn’t explain the Big Bang and the Standard Model of Particle Physics and Quantum Mechanics and Einsteinian Relativity and the Periodic Table, then it is a deficient theory. This is hogswallop, of course. No scientist would recognize this as a requirement of any theory. The success of the scientific method is not determined this way, and never has been. So Bogardus wants to tell scientists how they do their jobs, without having a clue how they do their jobs. And he does this by building on what we just saw was Aristotle’s catastrophically failed method, instead of any actual scientific method.
So Bogardus’s second premise is simply false. His definition of “successful” has no recognizable place in modern science, and thus modern science (and the modern naturalism now based on it) is completely unaffected by his false declaration of its necessity. A scientific theory is successful simply insofar as it works. Does Archimedes’ causal model, based on observed facts, predict the sinking-or-floating behavior of different objects in a liquid? If yes, then that’s all it need do. The question of why the components of its causal model are that way (why is water not compressible; why does water weigh what it does; why do certain objects repel it and hold their shape; why is there even gravity and mass at all; etc.) is for other scientists to figure out. Naturalism is not the assumption that it can all be figured out without appeal to nonnatural entities or powers; it is a conclusion from having observed that it usually always turns out that way. Hence Naturalism Is Not an Axiom of the Sciences but a Conclusion of Them.
Naturalism has no necessary role to play in science; it is not a premise required by the scientific method, for example. Nor is it any criterion by which science’s success is measured. Had the supernatural existed, we’d still have all the marvelous success of science—and it would refute naturalism. We’d have faith-healing wings in hospitals and schools of wizardry. But that just isn’t the way it went. And that’s the reason we grant the prior probability is high that it’s going to turn out to be naturalism all the way down. Some of us might have additional reasons to suspect this (see The Argument from Specified Complexity against Supernaturalism), but even we start with the empirical finding. Only evidence can decide whether anything even resembling the supernatural exists. And so far, the evidence has consistently indicated it doesn’t.
This is why it is crucial that in a footnote Bogardus backtracks and explains that, really, what “I mean” is “that there is literally no explanation to be had,” not merely that we don’t have it yet. In other words, he isn’t really talking about science, or how it actually works, or what its own criteria of success are. What he is actually trying to do is pose an argument similar to mine against supernaturalism, that naturalist explanations are somehow impossible, and though he can’t produce a proof of this, he wants you to believe he can present evidence for this—as I did for the supernatural (probably) being impossible. But instead of just state that as his actual argument, he buries this mission in convoluted layers of other propositions…including crucially false ones.
And this is where his thesis most conclusively fails:
Bogardus’s entire argument depends on it being the case that “any incomplete explanation is unsuccessful.” But when he appeals to evidence that science is successful, he term-switches (a typical Equivocation Fallacy that should never pass peer review), changing what he means by “successful” back to the real one science actually uses: Does it make reliable predictions and does it continually, and testably, increase the body of such knowledge? That science can and does do that does not require that it be “successful” in this other, bizarre sense he just made up—that its every theory consist of a complete Theory of Everything explaining every single fact of existence. So he starts his argument with that definition of “successful,” but ends with a completely different definition of successful. This is really bad philosophy.
The Crucial Mistake
We can see this in Bogardus’s own example of explaining to his curious daughter why people say “God bless you” when someone sneezes. He answered that people just always do that; and then realized that’s a bad explanation. It leaves her with more questions—or rather, really, her same actual question: why do people do that? “They just do” is not an “incomplete” explanation; it’s a non-explanation. It just repeats her question back at her as a statement. This is the insight Bogardus failed to make; instead he got on this really weird idea about his explanation requiring a Theory of Everything (it doesn’t).
Of course, people don’t “always do this.” Saying “God bless you” to a sneezer is actually a historically recent and ethnographically peculiar behavior. Older and more common was simply to say “May you be well,” which is more obvious in its explanation, given the association of sneezing with illness, which used to be far more fatal, warranting wishing someone well when confronted with it, to signal your concern and compassion (or superstitious misbeliefs about the power of thought and word). “God bless you” is just a Christian colonization of this practice, sneaking “God” in where it never had any function before.
But the mistake Bogardus made is in thinking the reason his answer was unsatisfactory was that he couldn’t explain every single thing underlying that phrase all the way down to the Big Bang, Stellar Physics, and the Primordial Soup. But his daughter wasn’t asking that. All she wanted to know is why does a custom exist. If he had explained that, eventually (at some level of explanation) she would be satisfied and would require no further answer for the specific question of “Why do people do that?” The history of a custom, and the causes of human adherence to customs, are not “vacuous” explanations. They are not “unsuccessful.” They do not “fail to explain” why a behavior exists simply because they don’t keep going all the way to where photons come from, or spacetime, or logic.
And explanations do not “fail” simply because we couldn’t explain something many levels down the causal-historical chain. Say Bogardus and his daughter went back and forth on the causal-historical chain explaining sneeze customs and get to the level of explanation whereby “human brains” explains everything else all the way up to the sneeze custom. And suppose Bogardus and his daughter live in the 2nd century, and thus literally don’t know where brains come from, and thus, like Galen, fallaciously just assume, “Oh, well, Zeus makes those.” And when his daughter asks, “Why?” Bogardus answers, “Oh, well, he just wanted to.” And suppose he really thinks it stops there—that there is no further explanation why God wanted to do that, he “just did.” Or suppose Bogardus settles on, “It was just necessarily the case that God would want to do that,” but can’t produce any actual demonstration of why that would be necessarily the case; he just declares it must have been. Or even if Bogardus kept going—eventually he’s going to get to some point where there just is no explanation why God is that way, or even exists at all, or has any of the relevant powers for this causal explanation to work.
Does Bogardus’s historical explanation of the sneeze custom then become “unsuccessful”? Well, no. And good thing, too. Because, as we’ll see, everything Bogardus complains about regarding naturalism having to hit some inexplicable bottom like this is just as true of theism. It doesn’t perform any better. And thus theism, by Bogardus’s own reasoning, is also unsuccessful as an explanation. It either entails circularity, infinite regress, or some brute fact or inexplicably necessary being. In fact it’s worse, because he can’t deduce from “God exists” the sneeze custom at all, so God’s existing doesn’t even explain that. Whereas a naturalist explanation of sneeze customs does—because naturalists don’t require everything to be explained to explain any individual things. It’s only this weird theist by the name of Tomas Bogardus who does. And even if he makes a methodological exception for gods; that exception will apply to naturalism, too. “I’m not talking about science, but philosophy” rescues naturalistic explanation just as well as theistic. So all his talk about science specifically seems moot. All science does is explain particular things, and continue to crawl down the causal chain as far as it empirically can. The rest is always filled in by philosophy. So neither science nor naturalism requires Bogardus’s bizarre standard of success. They can establish the probability of their conclusions entirely without it.
Instead of catching this mistake in his own thinking, Bogardus proceeds with several paragraphs of digression having nothing to do with the question of whether science is successful in its actual sense, or whether it needs to be successful in this “other” sense. Why would it need to be successful in that sense? He never explains. Instead, eventually, after several pointless knots have been twisted around, he gets to simply trying to insist naturalism can never come up with a grounding explanation of reality. But if that’s his argument, he should have simply said so, and tried to make a direct case for it. Why all the verbal games? Why bury the lede like this?
Christian apologists’ propensity for this kind of behavior really annoys me. But now that we’ve stripped away thousands of words of impertinent baloney, maybe we can get to what his actual argument is?
The Failure of Bogardus’s Hidden Argument
Ultimately, Bogardus never makes any actual case for it being logically impossible for Naturalism to ground reality (we actually have plenty of working models of that), and he can’t make an empirical case, because the evidence consistently establishes the opposite: to borrow his analogy about explaining the Earth holding us up by appealing to a turtle holding the Earth up, we actually found the turtle the Earth rests on, and can actually see it’s turtles all the way down. It’s not a theory anymore; it’s an observation now. “But we can’t see the bottom, because it’s so far; so maybe there’s a dragon down there and it’s not just all turtles” is not a rational declaration when faced with this evidence. It looks like it’s all turtles. So…what evidence do you have that it stops at a dragon? [To be clear, this is now my analogy. The Dragon is God. I am not saying Dr. Bogardus argues there are dragons at the bottom of any turtles.]
Bogardus is thus flipping scientific method upside down, and rather than follow the actual scientific method and only derive beliefs from the evidence, he wants to declare all evidence irrelevant and claim that, even after crossing millions of hills, whenever we don’t know what’s on the other side of the next hill, we get to declare it’s a magical dragon. For…reasons? He never explains.
I’m sorry, but this should never pass peer review.
Bogardus wants the worry to be, somehow, that Naturalism can’t generate an infinite regress of explanations, and therefore “we need a dragon down there” (i.e. God). But that’s his second mistake: assuming the ultimate explanation has to be (figuratively) a dragon (i.e. God) and not a turtle (i.e. some natural fact or other). Bogardus examines all three possibilities for Naturalism (supplying here my own illustrative examples): that the explanations are indeed endless (infinitism: e.g. it’s turtles all the way down), that the explanations are ultimately circular (circularism: e.g. downward force bends in a circle and the Pillars of the Earth holding up the sky are in fact the legs of the last turtle), or that all explanation ends in some final brute fact, something that just exists for no reason (e.g. a final antigravity turtle that always falls up). But all of this analysis would hold for God as well (just swap G in for N and it’s all the same). Maybe God exists and his existence has an underlying infinite regress of explanations; maybe God created himself in some act of circular causation; or maybe God just exists for no reason. These are not logically impossible. And whichever one you choose is always going to leave something unexplained: why that thing (that infinite series, that circular causation, or that brute fact) and not something else?
One of the most common follies of Christian apologetics is to declare something impossible that isn’t. When someone makes a modal argument like “no explanation is possible,” we can refute that with only modal evidence—we need merely present a possibility. We don’t have to prove that explanation probable; we only need to present a possibility to refute the claim that there are none. So we commit no possibiliter fallacy if we do nothing more than that. “Possibly, therefore probably” is a fallacy; but “It’s possible, therefore it’s not impossible” isn’t.
Remember, Bogardus isn’t really arguing it is improbable that there will be a naturalist grounding to reality; and he can’t, because then he’d be talking about frequencies, and that makes it an empirical argument. And when we look at the actual frequency data, the rate of “it always turns out to be a natural explanation” is far beyond trillions to one now; in fact, so far, nonnatural explanations have scored exactly zero. You’d have to be pretty foolish to keep betting on that loser, right? So there is no rational case down that path. Bogardus must steer clear of it, lest he crash right into disaster. So he takes the other fork in the road and just insists “it’s impossible,” without presenting any evidence that it is. Does he survey all scientific and philosophical Theories of Everything and successfully rule them all out so as to argue that it might be impossible to come up with one? No. He doesn’t even consider a single one—and yet considering only a single one would be a Straw Man Fallacy anyway and thus still shouldn’t suffice to pass peer review; the more so, none at all.
I have already presented possible grounding explanations for Naturalism: from ten different ones in The Myth That Science Needs Christianity, to my favorite one in Superstring Theory as Metaphysical Atheism; and I explain why there is no barrier to coming up with more in The Argument to the Ontological Whatsit. And if your specific worry is “Where could it come from?” I’ve got a distinct possibility for you in The Problem with Nothing (which Christian apologists have struggled to find fault with). But there are countless others in the actual science of cosmology (see my debate with Wallace Marshall), all starting with brute facts far simpler than any God, and entirely naturalistic. So Bogardus can’t claim “it is impossible to come up with possible grounding explanations.” There are too many possible ones in evidence for such a declaration to stand. The evidence falsifies his theory.
But he doesn’t even look for that evidence. He literally just argues, “Aristotle said ignorance justifies belief in design; I see ignorance in the sciences; therefore, I get to believe in design.” That is the entire sum of his argument. He never examines any pertinent evidence regarding any component of the truth of that. I’m not kidding. He never asks whether there are possible Naturalist grounding models (much less considers any). He just says there can’t be. Yet he never presents any evidence that there can’t be. Even when he tries to defend the premise with his analogy about turtles holding up the Earth, he never explains how anything like that holds for Naturalistic explanations—e.g., he never presents any evidence that Naturalist explanation has to entail infinite regress, or that a Nonnaturalist explanation wouldn’t (likewise, circular or brute fact explanations).
So Bogardus never really establishes there to be any epistemic virtue lacking for Naturalism here that’s present for Nonnaturalism. What explains all the regularities of God? Brute fact? Infinite regress? Self-causation? That’s all the same stuff. And punting to “necessary being” gets you nowhere, because no such proof exists for God; so anything could qualify. Like an ultra-simple initial quantum vacuum; maybe that just necessarily exists. For example, in the Capturing Christianity interview (starting toward minute 52) Bogardus suggests his theory that “God creates sentient life because it is good” requires no further explanation. But…um, yes it does. Why is that good and not something else? Why does God care about what’s good? And why does goodness have any of the requisite causal powers here required? He’s just proposing another brute fact, another circular argument, that what he thinks is good “just is” good, for no reason; and moreover, he thinks “its being good,” merely thus, can somehow cause an entire bizarre superbeing to exist with a constellation of weird powers and bizarrely specific desires. Um, why? There is no explanation here. “Why would a dragon do any better than a turtle? It just does. Shut up.”
It’s only worse that God, as Bogardus is here formulating it, is then a bad explanation of observations. Because it fails by the actual standards of science: it predicts the opposite of everything we see. The universe is as poorly designed for life as it could be and still produce it; and even locally remains capriciously maleficent. The ground of all being seems conspicuously disinterested in any sentient good. So we should reject his theory, not embrace it. But even by his own bankrupt medieval method it’s a fail, because he still hasn’t explained why what he thinks is good “is” good in that sense (so this still “calls out for an explanation”), or (even more importantly) why its being good could cause a God to appear and have a lot of weird abilities (that, too, still “calls out for an explanation”). For example, I can produce an explanation for why consciousness is objectively good, but not in any way that would cause consciousness to exist—much less produce or explain a god (see The Objective Value Cascade). “I just feel like that needs no explanation” is not an explanation; it is, quite conspicuously, the absence of one. Yet that’s all Bogardus has (see the Capturing Christianity video, minute 52 to 53).
I cannot think of a more glaring logic fail than this. I struggle to understand how any peer reviewer would give this a pass. Why did none of them stop and say, “Wait, could you please present evidence for these assertions? Could you please at least survey some Naturalist theories you claim don’t exist or can’t work? Could you at minimum explain what your turtle analogy has to do with realistic, much less scientific, ontologies? And can you please prove how your alternative hypothesis survives any of the same objections, and not just handwavingly claim it does?” But, nope.
Oh No. Not Feser.
In a footnote Bogardus reveals his thesis is based on a claim made by Edward Feser. The guy who completely failed to rescue his position from Feser’s Five Proofs of the Existence of God: Debunked! even after two tries at bat (see Feser Can’t Read and Feser Still Can’t Read). Bogardus quotes Feser claiming that if A causally explains B and B causally explains C, if then we posit A (and thus, by extension, any underlying cause) as a brute fact, “then A has nothing to impart to B or C that could possibly explain their operation.” But if that were true, it would be true of God, as well. If God is just a brute fact, then, per Feser, God cannot even in principle explain anything. But few theists would grant such an absurd claim. Theists would be in a pickle if they did, of course, because all ontological arguments have failed, so theists currently lack any explanation of why a god would exist—much less their extremely bizarre version of one (this is, indeed, the Hidden Fallacy in the Fine Tuning Argument). But neither would any scientist grant Feser’s assertion regarding natural explanation either. Because it isn’t logical.
It could well be that a simple quantum mechanical principle underlies all existence and is a brute fact. But if that explains numerous bizarre and peculiar observations of physics, then in what way does it “have nothing to impart” to those facts by which to explain them? It explains them. Indeed, it would explain them so darned well at that point as to ensure it’s true. That’s far better a position to be in than we have for God. So how would realizing this principle “just is” a brute fact change any of that? Well, it wouldn’t. So there is no argument here. It’s logically possible all existence is explained this way. And there is more evidence that it is than that it is explained by supernatural agents—even now; far more so then, in the condition just imagined, when A would actually explain wild oddities about B and thus C, verifying A is indeed true, and not just hypothetical.
In The Problem with Nothing, for example, I have shown that even starting with absolutely nothing we should inevitably expect this universe to come into existence. Here we have no brute fact but nothing at all. You can’t get simpler than that. And there is no way to gainsay it; because to deny that “but for a cause, ‘nothing’ would exist” is to declare the existence of “something” necessary (as those are the only two options), eliminating any need for further explanation. And since that necessary being could be anything (there is no evidence it has a mind or makes intelligent choices, or is anything more than a simple quantum mechanical vacuum), there is no way to argue it will be nonnatural. You need evidence for that conclusion. And Bogardus has none. To the contrary, all the evidence there is on the point indicates the contrary of his thesis.
So Then What?
One approach would be to find the most probable brute fact possible, by identifying the simplest thing that would explain everything else. Which is why Bogardus’s example of mysteriously appearing oranges on doorsteps misses the mark—oranges, and doorsteps, are bizarrely complex circumstances, and thus the least likely things to happen for no reason. Whereas a quantum wave function for a complete multiverse can be described in just a dozen or so bits of information, vastly simpler than having to posit a whole complex orange appearing in hyper-specific ways—and even more vastly simpler than a God. In fact on current quantum mechanics, we actually can calculate the probability of oranges appearing exactly as Bogardus describes. It’s not zero. The same is true on classical thermodynamics or any other model of wholly ungoverned random production. The probability is always just too small to be likely to observe, which explains why it doesn’t happen—and I need no further explanation than that. That is a complete (and thus on Bogardus’s own terms successful) explanation. (See my recent article All the Laws of Thermodynamics Are Inevitable.)
Instead of discovering this, Bogardus spends pages and pages describing all three possible metamodels (infinitism, circularism, and brute-fact foundationalism) and what he doesn’t like about them, which is principally, that they all entail some unexplained brute fact—but he never presents any reason to doubt the logical possibility of, indeed, such a fact; or even its probability, given the direction all scientific evidence has consistently gone for thousands of years now. He also never explains how we are to know his God is not just another such brute fact. He never shows that “God” performs better against his completeness tests than “Naturalism.” Just saying “but I declare God to be logically necessary” still leaves you with something unexplained—in fact, two things: “What proves God logically necessary?” and “Why does logic have that power?” Is logical necessity itself the ultimate brute fact, or can we say it is successfully self-explanatory in precisely Bogardus’s own sense? (See The Ontology of Logic.)
But then what about God? If we get to say “God is necessary” when we have no evidence (no proof) that God is “necessary” in this sense, then we get to say this of any natural fact we want as well. And then we’re at an impasse. What makes “God necessarily existed” any more likely than “A primordial quantum vacuum necessarily existed?” You can’t appeal to outcomes, because they are all just as expected on either theory. And you can’t appeal to theoretical simplicity—a simple quantum vacuum wins that contest; of course theism also then loses to my Theory of Nothing, because no thing is theoretically simpler than “Nothing.” So what empirical virtues can you appeal to to get the necessary ground of all being to be God? When Edward Feser tried to answer that claim, for example, he left out anything actually getting him to a mind, and inadvertently ended up proving mindless spacetime was the necessary ground of all being. That these arguments never go the theist’s way should clue you in. They always lean on some fallacy or other. Yet true theories don’t have to do that. Only false ones do.
So it does not help that Bogardus imagines “maybe” we can prove God ontologically necessary. We still haven’t. And we can just as “maybe” prove that of any other simpler fact, like a primordial quantum vacuum, or nothing at all: In my Problem with Nothing model, for example, I consider two possible conditions (past finitism and past eternalism), and show that, no matter which is the case, if we start with absolutely nothing, logic entails our universe then inevitably would come to exist, whether at some past point in time (if there ever really was “nothing”) or not, because if a past-infinite series exists, the lack of anything to explain that would still atemporally explain why that series exists and not something else.
This is because logic entails the laws of probability, and since “nothing continuing to be nothing” is a single choice (against infinitely many “something” choices), and nothing predetermines what will be chosen (as is precisely the case when nothing exists), then “nothing” is infinitesimally unlikely to be selected as what exists—so if nothing is choosing, “nothing” is least likely to be chosen. And when we look at what the most common thing to arise by such random selection is (given that every possibility is equally likely at that point), it will be a multiverse in which our universe is functionally certain to arise (it is thus telling that there are already Six Arguments That a Multiverse Is More Probable Than a God).
This conclusion can only be evaded by insisting logically contradictory states of affairs can exist (like that somehow nothing is choosing what will exist and at the same time something will always choose nothing to exist); but once you do that, all arguments for God collapse, as then you cannot “argue” that nothing cannot just “become” our universe, because you just admitted logical contradictions can come true, so any demonstration that this would be impossible would automatically be false.
The bottom line is that “God” is no different than any of the Naturalism models Bogardus considers. God could be past eternal, and thus suffer all the same defects he sees in infinitism. God could appear out of nowhere in time, and thus suffer all the same defects he sees in brute factism. And God could be self-causing, and thus suffer all the same defects he sees in circularism. His own model suffers all the same defects he claims of Naturalism. Which is to say, it suffers just as little, because these aren’t really defects. “It annoys me that something might just exist and be the way it is for no reason” is not a rational method of deciding the matter. Facts don’t care about your feelings.
Conclusion
Bogardus offers an extremely verbose and convoluted word-wall that purports to offer some new argument for God, but in the end it doesn’t. It’s just another Argument to the Ontological Whatsit. He’s just saying “You can’t explain the regularities of nature without positing there be at least one brute fact, but I can,” and giving no evidence whatever that he can. All he has to offer in replacement is another brute fact—in fact, one vastly more bizarre and poorly evidenced than any that modern science could claim. He never presents any evidence that any necessary being exists, either, or that it has to be a personal consciousness, or mental or “teleological” in any way. So we’re left with the simple fact that reality might indeed be grounded in a brute fact, something that just exists or happened for no reason. Bogardus has no demonstration it can’t be.
Bogardus instead relies on a failed argument of Aristotle’s that attempted to derive knowledge from ignorance, which we now know is impossible. And his only syllogism trades on an equivocation fallacy in defining the word “successful” when describing any scientific theory, requiring it mean something wholly irrelevant to science at the point of his hypothesis (that “science” is only successful if it does the one bizarre thing he wants), and then switching to something else entirely at the point of his evidence for that hypothesis (that science is successful in the required sense for the argument to carry). He thus disregards all modernity, the entire Scientific Revolution and every discovery since, and retreats to a bankrupt medieval scholasticism, to argue illogically that because Aristotle thought not knowing the cause of something warrants believing it was caused by an intelligent agent, Bogardus gets to do the same thing now—even though the entire history of science has already proved that inference false.
The only way to know if a necessary being underlying all reality is a sentient immateriality or an insentient physical fact, is to present a formal proof or sufficient evidence of it. Bogardus has none. That’s also the only way to know whether the ground of all reality even is a necessary being or not. So Bogardus cannot claim to know this either. And the only way to prove the underlying ground of all reality isn’t just a brute fact, something that simply has no explanation, is to either deductively prove it (and Bogardus doesn’t; there is no formal proof of that conclusion anywhere in his paper) or prove it empirically (and Bogardus doesn’t; there is no discussion of any pertinent evidence anywhere in his paper). All he has to offer as if it were evidence for this conclusion is “that would annoy me.” But there is no logical connection between Bogardus’s feelings and the actual nature of reality.
What we have left is (1) the trendline of the evidence—which is toward any ultimate explanation being natural, whether necessary or brute it doesn’t matter—and (2) the fact that Bogardus has no proof that any necessary being exists, much less that it’s a person, or in any way nonnatural. I already have more evidence than he does that the necessary being, should there be any, is just spacetime itself—just simple, mindless spacetime. The Argument to the Ontological Whatsit doesn’t get you to God, if you care about the evidence and the logic of what a necessary being would actually have to be like. Whereas, by contrast, gods require extraordinary unexplained specified complexity, and are therefore the least likely things to be either brute or necessary facts.
In the end, if you are still left awake at night struggling to understand how any brute fact can exist (rather than not, or some other), you really should study The Problem with Nothing. If nothing exists to decide what should exist, then what will just happen to exist for no reason will be decided at random—without intelligent choice (because when nothing exists to specify a choice, any choice becomes as likely as any other); what Aristotle called “chance.” And this will necessarily be the case—what Aristotle called “necessity.” Because the absence of a choice of any one outcome over any others entails a random outcome, and of all possible random outcomes, ones that look like ours are the most common.
And yes, this is an actual peer reviewed theory in the sciences, disproving Bogardus’s claim that no such thing can exist: see Maya Lincoln and Avi Wasser, “Spontaneous Creation of the Universe Ex Nihilo,” Physics of the Dark Universe 2 (2013): 195–99.
This isn’t about your article directly, but since I saw Tomas’ name in the title and you referenced his conversion, I thought I would drop this but of info:
“Tomas Bogardus (another fashionable Protestant convert to Catholicism)”
Tomas has appeared on Cameron’s show before. The last time I saw him he was covering a paper he wrote titled “ The Problem of Contingency for Religious Belief” in which he attempted to tackle the “argument” of “if you hand been born elsewhere, you would have different beliefs”. And I use “argument” in quotes as I view it more an observation than an argument as, as far as I’m aware, it does not exist as a formal argument which is why Bogardus had to craft his own syllogisms to argue against in his paper.
Anyway, my criticism of his paper aside but very much related to it, in his interview with Cameron, Bogardus shared that he had converted to Catholicism from Protestantism in college (or grad school). When he shared this news with his grandparents, they explained to him that they had originally been raised as Catholics before they immigrated to the US. The reason for switching religions and raising their own children as Protestants? Well of course it was because the closest church to their new home was a Protestant one and the Catholic Church was too far.
I have no idea if Tomas has been successful in converting his parents to the True™ faith or what it means for their immortal soul, but I can’t help but think of this story every time I see his name come up. His paper on “The Problem of Contingency for Religious Belief” is equally as terrible as this one sounds if you ever feel like giving that one a gander!
Weirdly, it does. And it would be curious if Bogardus didn’t even know that. It’s a subvariant of the Outsider Test, formally articulated (very well IMO) in John Loftus, The Outsider Test for Faith (Prometheus 2013).
I’ll let Loftus know about his paper. He might blog about it if he deems that worthwhile to do.
Perhaps I misunderstand the Outsider Test for Faith, but I was under the impression that was about assessing claims from an outside perspective without the bias of your own beliefs. For example, judging individual miracle claims from Christians as if the same claim was coming from a Muslim.
I can see how this would be related to the “born elsewhere” observation, but I don’t see it as a formal argument such as
P1 – If you had been born at a different time or in a different place, your beliefs would be different.
C – Therefor your beliefs are wrong.
This is the type of argument I have not seen put forward (and indeed the examples Bogardus brings up in his paper to not argue for this conclusion or anything similar).
For example, Bogardus attempts to formulate multiple logical arguments (as a way to steal man those that would raise such an objection), one of which is:
3) If you had been born and raised elsewhere, else when, and formed religious beliefs using the same method you actually used, then, by your own lights, you easily might have believed falsely.
Therefore,
4) Your religious beliefs were not formed safely.
Therefore,
5) Your religious beliefs don’t count as genuine knowledge.
That’s why I recommend reading Loftus. There is much more to the OTF than just attempting an “unbiased POV.” He has whole sections on the location argument: if one’s belief is determined largely by the accident of location, then all beliefs are cast into doubt equally by that fact. Christians have to “special plead” to keep their religion immune to that result of an objective POV. And if you have to do that, you are failing the OTF. See Loftus’s book for the details.
The steel man you suggest from Bogardus is a different way to put the same point. Another way to put it is in terms of Error Theory (as applied to religion rather than, where the term originated, moral theory: see the section on that near the end of Malcolm Murray’s Atheist’s Primer).
Christians readily explain other beliefs (Muslims, Hindus, Buddhists) in terms of an Error Theory (upbringing effect, availability bias, etc.). The OTF requires you apply the same reasoning, on the same terms, the other way around (e.g. Muslims can make all the same arguments against Chrisians on this point). And Christianity would not pass this test any more than Islam does. This has become such a problem some theists have even adapted Skeptical Theism to try and escape the consequences of realizations like this (which was originally invented to escape the cognitive dissonance of the Argument from Evil).
I think framing it as “beliefs not formed safely” is a decent way to put it. But Error Theory is already a well developed subject in the philosophy of religion; there are a lot of ways it is applied to the Argument from Locality (in time and space). But they could all be subsumed under “beliefs not formed safely,” as long as one doesn’t use the abstraction to hide the particulars of why that matters (the specific ways this indicates unsafe belief and the evident impossibility of Christians escaping them any more than Hindus, putting them on an epistemically equal footing).
Notice of Correction: I heard that Dr. Bogardus mistook me as attributing my dragon analogy to him. I don’t know how anyone could think that. But to be sure, I have amended the article to make clear this is my analogy, not his. He never argues for a dragon. He argues for a god. “Dragon” is my analogy to “God” in the culturally common “turtles” analogy that Bogardus references.
Thanks for taking the time to reply to my paper, which is available here: https://philpapers.org/archive/BOGINI.pdf
I hope you’ll give me the courtesy of a reply to your objections. I initially responded point by point, since there are some pretty wild speculations and insults in your blog post. But my response started getting quite long, and I didn’t want to burden your readers with another tome, so I’ll limit myself to what I think is your main objection.
At first, it sounded like you were objecting to my second premise. You say, “Bogardus’s second premise is simply false.” And in response to that premise, you say, “Is it the case that ‘Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one’? No. This is not a component of any scientific standard in any field. We have never needed any explanation to be complete in order to be successful.” You say this premise is “bogus” and “pseudoscience.”
Unfortunately, you misunderstood that premise. In endnote 15, I clarify: “By ‘but lacks one’, I don’t mean merely that we don’t know what the explanation for this element would be, or that it’s not available to us, or some such constraint on our epistemic access to an explanation. I mean that there is literally no explanation to be had.”
You also seem to misunderstand what I mean by a “successful” explanation. But I clarify that on page 6: “insofar as there’s a connection between explanation and understanding – as Woodward (2019) puts it, ‘One ordinarily thinks of an explanation as something that provides understanding’ – this explanation has failed, since we’re not in a position to understand…” See also endnote 18, where I say, “A successful explanation can produce in us understanding of the phenomenon, an understanding of why or how it’s happening.”
You say that we managed to build amazing technologies despite incomplete scientific explanations, and in that way our scientific explanations have been successful. Sure, we have successfully developed new technologies, using our scientific theories. No doubt about that. But I hope it’s clear now that when I say an explanation is successful, I mean that it provides understanding, not technological advancement. Maybe this would help illustrate the distinction: ChatGPT may help a student pass a computer programming class, without really learning anything. This student would be “successful” in one way, but not in another. He successfully passed the class, but he didn’t gain any understanding. Similarly with scientific theories, I’d say. I’m concerned with whether science can produce understanding, if Naturalism is true. I myself think that’s an interesting question!
You accuse me of equivocating on “successful.” But I can assure you I’m not. When I say that I’m confident that some scientific explanations have succeeded, I mean that I’m confident that scientific explanations have produced understanding. (Aren’t you?) What I doubt is that they could do this, if Naturalism is true.
You say, “Bogardus’s entire argument depends on it being the case that ‘any incomplete explanation is unsuccessful’.” This is a misleading use of quotation marks. My paper does not contain those words you put in quotation marks, but your reader would certainly be excused for thinking that it does, based on what you write here. You do this several times in your blog post, including apparently attributing to me some strange claims about dragons. I encourage you to use quotation marks in a less misleading way, and I encourage your readers to look at my actual paper, to see what I actually said (and what I didn’t say). Here’s perhaps the most egregious example: “[Bogardus] literally just argues, ‘Aristotle said ignorance justifies belief in design; I see ignorance in the sciences; therefore, I get to believe in design’. That is the entire sum of his argument.” Buddy. You must know that’s a terrible abuse of quotation marks. Those words are not mine. You’re shadowboxing, and misleading your readers.
Returning to your criticism of premise (2), again, no, I don’t think that scientific explanations must be complete in order to succeed. Premise 2, again, goes like this: “Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.”
You say that “this is where [Bogardus’] thesis most conclusively fails.” But since you’re attacking a view about completeness that plays no role in my argument, I can rest easy that you’ve not identified any failure in my argument here.
In the section of your blog post titled, “The Crucial Mistake,” you again seem to be attacking a strawman, a view that I do not endorse, a view according to which any incomplete explanation is unsuccessful. So I’ll just repeat: that’s not my view. See endnote 15 again, and search for instances of “literally” in my paper, to see my attempts to forestall your misunderstanding.
I did my readers the courtesy of actually laying out my main argument step by step, as clearly as I could, with attempts to preempt possible misunderstandings. I would appreciate the courtesy of a response to things I actually said, instead of a response to things I certainly didn’t say (while audaciously putting quotation marks around them! haha).
In your final section, you say, “Bogardus never makes any actual case for it being logically impossible for Naturalism to ground reality.” Well, I’m not entirely sure what you mean by “ground reality,” but I guess I’ll just repeat what my main argument actually was, and invite your readers to check out my paper. The main argument goes like this:
• Any scientific explanation can be successful only if it crucially involves a natural regularity.
• Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
• So, a scientific explanation can be successful only if it crucially involves a natural regularity, and this regularity does not call out for explanation while lacking one.
• If Naturalism is true, then every natural regularity calls out for explanation but lacks one.
• So, if Naturalism is true, then no scientific explanation can be successful.
The paper is spent defending each premise. In the end, I invite the reader to run a modus tollens on (5), and I defend theism from a tu quoque objection. That’s it. That’s the paper. And I can hardly be accused of providing no argument for that conclusion, proposition (5) there.
If your readers are interested, they can check it out here: https://philpapers.org/archive/BOGINI.pdf
Thanks again for taking the time to read and respond to my paper. I genuinely appreciate it.
Thank you for your reply, Tomas.
I’ll give you a considered response next week when I have some time free.
Tomas:
When you say, “By ‘but lacks one’, I don’t mean merely that we don’t know what the explanation for this element would be, or that it’s not available to us, or some such constraint on our epistemic access to an explanation. I mean that there is literally no explanation to be had”, the problem is that this is false, and trivially so. Which is what Richard went on at length to point out.
Even “It’s just a brute fact” is an explanation . Is it actually true? We don’t know, but that’s not the issue at hand here. Ditto “It’s a fundamental metaphysical principle” or “It will deductively hold in all logically possible contexts”.
For you to shore up premise 2, then, by your own admission you’d have to demonstrate that no naturalistic explanation, not even something quasi-supernatural like the Tao or some kind of elan vital or organizing principle, is available. Which is impossible. And which your paper didn’t do. And didn’t try to.
Since you didn’t do that, this is still an argument from ignorance. Because you haven’t systemically proven that any explanation cannot be had. So you are only inferring from the (supposed) lack of one on offer. And that is a straightforward argument from ignorance, and, worse, a false one, because there still are proposals on offer, just ones that we haven’t confirmed, so it is doubly an argument from ignorance.
I suspect you may not be recognizing the degree to which this comes off as a motte-and-bailey fallacy, exactly as Richard noted. Here, it’s a nested one. “Naturalism can have no ultimate explanations” is conflated with “Naturalism at present does not have any ultimate explanations” which in turn is conflated with “Science makes theories that are predictable”.
Which is why Richard’s analysis is still more than sufficient here. Because, in fact, science doesn’t need ultimate explanations. Indeed, if your claim that none could ever be on offer is true, that can very easily be an indication that the very idea is nonsense.
As I pointed out, I think the very fact that, for example, there could be past eternality and thus infinite causal regress proves that your condition of ultimate explanation is incoherent .
When you say “What I doubt is that they could do this, if Naturalism is true.”, you’re actually still not responding to Richard. It does confirm for me that this is a motte-and-bailey argument that has ensnared even you, though, so I do think this is in good faith.
“I’ve made a predictive model and I don’t even care if it is metaphysically true and in fact suspect it probably isn’t because it is very useful” works in any logically available reality. Even ones that are truly random or chaotic. Even with Descartes’ demon. Yeah, the theories suck, but for you to say that the theories on offer are suspiciously good, you’d need to generate a universe which is unambiguously obviously designed and then compare theories that would be on offer in that universe and their predictive power against ours. And you didn’t. So you failed to meet your burden.
In other words, Tomas, I don’t know what a suspiciously explanatory theoretical model for a supposedly unexplainable natural universe would look like, and neither do you . I don’t know how comprehensible theories should be under supernatural metaphysicalism. Richard elsewhere has pointed out that classical theories about the world being a clockwork design or the world being a set of crystalline spheres with air in it would be simple, understandable, designed worlds, exactly as people designed when they believed in supernatural metaphysicalism honestly rather than as a woobie and as a social control mechanism. And then we found out the real world was a hostile vacuum where even the fundamental forces seem to change and where we could be living in a multiverse. And then the special pleading began.
By failing to understand what scientific claims are (and I invite you to read Hawking’s description of a theory to see that you’re being inaccurate), you are ascribing to them a strength of claim they aren’t actually presenting. That’s argumentatively a strawman, and so invalid, and it again shows that Richard is correct to argue that your position is based on scientific illiteracy. When I read a scientific theory, I don’t think they have discovered a finally-true statement of the universe at all times and all places. They’ve discovered a contingently useful model. Which isn’t suspicious at all in any logically available universe where people would be observers, and your article doesn’t argue why it would be.
As for, ““[Bogardus] literally just argues, ‘Aristotle said ignorance justifies belief in design; I see ignorance in the sciences; therefore, I get to believe in design’. That is the entire sum of his argument.” Buddy. You must know that’s a terrible abuse of quotation marks. Those words are not mine. You’re shadowboxing, and misleading your readers”, I can only speak to myself, but it is so clearly presented as not your words but his (self-admittedly reductive but in his mind not inaccurate) paraphrase that I think this complaint comes off as very disingenuous. He’s using colloquial and polemic language, but it is a reach at best to say that this is shadowboxing.
We know you made your claims and tried to defend them, Tomas. The problem is that the first two statements, as said, are false . They are just obviously false. So the best case scenario is that you are misstating your argument so badly as to cause confusion. And, yes, I think it’s reasonable enough to say that 5 holds if 1 to 4 do. But… notice how Richard spent time attacking 2 to 4. So that’s a moot complaint.
As far as the complaints about quoting inaccurately: He uses this quote, “is a deficiency of a theory if the theory leaves [anything] unexplained” . Is this an accurate quote and not leaving out major context, Tomas? Because, if so, then your definition of a theory is terrible . Again, to the point of suggesting scientific illiteracy.
Thank you so much for the response, but I really think you need to actually take Richard’s criticism seriously as something that isn’t a malicious strawman. I really do think you are fundamentally mistaken about what people who are operating from either procedural naturalism and/or metaphysical naturalism are saying theories are, and misunderstanding just how provisional even very good theories can be and how they are inherently models capturing slices of reality.
Dear “Fred B-C,” (I’d rather not use your first name, since I don’t wish to presume a degree of familiarity that we definitely don’t have. But I kind of doubt that’s your real name, and I’m not sure how else to address you, given your username.)
You said,
You quoted me saying what I mean by the phrase “but lacks one.” You then say this is trivially false. Is that what you really meant to write? That my claim about what I meant by “but lacks one” was trivially false? Why do you think I’m wrong about what I meant by that phrase? Any why is my mistake about what I myself meant by that phrase “trivially” false?
You also say,
No, to say that a fact is brute is to say that it has no explanation. And surely pointing out that a fact has no explanation is not itself an explanation.
You then say,
Here’s my second premise:
-2- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
This is just a conditional, placing a necessary condition on explanation. Premise 2 doesn’t mention Naturalism. So why do you think I need to prove something about Naturalism in order to “shore up” premise 2?
Also, when you say I need to “shore up” premise 2, this suggests that premise 2 has come under some sort of threat, that an objection has been proposed to it. But neither you nor Dr. Carrier has presented any objection to premise 2 yet, given what my paper says about the meanings of “successful” and “but lacks one.” If you want to object to premise 2, you should propose a counterexample: an example of an explanation that can be successful, while also crucially involving an element that calls out for explanation but lacks one.
Dr. Carrier tried to do this, but unfortunately he misunderstood what I mean by “successful” and by “but lacks one,” despite the attempts I made in my paper to forestall those very misunderstandings.
Do you have an objection to premise 2? A possible counterexample? If so, what is it, exactly?
You then say,
As I said in my previous comment (above), I did my readers the courtesy of actually laying out my main argument step by step. Objections are welcome, but I would appreciate it if you did me the courtesy of saying how your objections actually engage the argument. Which premise are you objecting to exactly? What, exactly, is the objection?
Saying “this is an argument from ignorance” (or, later, “this comes off as a motte-and-bailey fallacy”) doesn’t tell me which premise you’re challenging. If you wish to challenge premise 2, then please sketch a counterexample to premise 2.
You then say,
I myself am skeptical that infinite regresses of causes are possible. But, if they are, perhaps you could lay out your proof step by step. What do you think my “condition of ultimate explanation” is, exactly? And how does the alleged possibility of infinite causal regresses show that this condition is “incoherent”? And what, exactly, do you mean by “incoherent”? Do you mean logically impossible? If so, I would be very curious to see this proof of yours. If you’re willing, please lay it out step by step, using common argument forms/inference rules. It sounds interesting.
You then say,
I don’t believe I’ve said any theories on offer are “suspiciously good.” And what theories are you talking about, exactly? On second thought, there’s probably no need to reply to this. The conversation would likely be more productive if you were to just clearly state exactly which premise(s) of mine you object to, and exactly what your objection is. Much of the rest of your comment is hard to make sense of. There’s a lot of sound and fury, but it’s hard to see how any of it engages with any premise in my argument. You’ll have to show me.
I’ll reproduce my argument here, for convenience. (In my last comment, the numbers were turned into bullet points for some reason. Hopefully it works this time…)
-1- Any scientific explanation can be successful only if it crucially involves a natural regularity.
-2- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
-3- So, a scientific explanation can be successful only if it crucially involves a natural regularity, and this regularity does not call out for explanation while lacking one.
-4- If Naturalism is true, then every natural regularity calls out for explanation but lacks one.
-5- So, if Naturalism is true, then no scientific explanation can be successful.
About my argument, you said,
It sounds like you think premise 1 and premise 2 are both obviously false. Why do you think so? Each of those premises is a conditional. To show that a conditional is false, you should provide a counterexample. As far as I can tell, you haven’t yet tried to do that. I myself don’t think these premises are obviously false. In fact, I think they’re true. So, if you would, please pony up actual counterexamples to these premises, to help us see why these premises are obviously false. Merely claiming that a premise is false is not yet an objection, even if you add “obviously.”
Finally, you ask a question:
First, I didn’t offer any definition of “theory,” so I think you’re misunderstanding what’s going on in this context. Second, I linked to the paper twice, so you could have easily checked for yourself to see what’s going on with this quotation: https://philpapers.org/archive/BOGINI.pdf. Here’s some fuller context from my paper:
“First, what is it for something to ‘call out’ for explanation? Recall that Aristotle seemed to think that any natural regularity ‘calls out’ for explanation: it’s not the sort of thing that requires no further explanation – because it’s obvious, or self-explanatory, etc. – it can be further explained, and ceteris paribus it is a deficiency of a theory if the theory leaves it unexplained.”
Yes, I think Dr. Carrier’s omission of “ceteris paribus” and substitution of “anything” for “it” there does in fact change the meaning of the sentence. I’m not saying that it’s a deficiency of a theory if the theory leaves anything unexplained. I say rather that, other things being equal, it’s a deficiency of a theory if the theory leaves unexplained anything that calls out for explanation.
Also, had you read the paper yourself, you would have seen that there was an endnote immediately following that last word in the quotation there. I believe this endnote addresses your concern. The endnote says,
“But what if the theory is not intended to explain the relevant phenomenon? Could it really be a deficiency of a theory that it doesn’t do what it was never intended to do? For example, is it really a deficiency of the theory of evolution that it fails to explain the movement of quarks? Well, yes, I would say that it is. And you can see that by considering how the theory of evolution would be stronger, better, if – somehow! – the theory were also to explain the movement of quarks.”
Do you really disagree that the theory of evolution would be better if—somehow!—it also explained the movement of quarks? Why wouldn’t that make the theory better? Sounds like a pretty cool theory to me! Ceteris paribus, more explanatory power is a theoretical virtue. No?
I will read your reply with interest, and if you raise any intriguing objections that clearly engage one or more premises in my argument, I will likely respond. But I hope you’ll understand that I can’t spend a lot of time going back and forth with an anonymous internet critic down here in the comments.
No, it’s my real name, Tomas. I self-hyphenated my last name but didn’t legally change it, so I just use that as a short hand for commenting.
“It’s a brute fact” is an ultimate explanation. Is it satisfying? No. Is it likely true? Who knows. But the universe doesn’t owe us satisfying.
Ditto “It’s a metaphysical principle” or “It’s an inevitable concomitant of geometry”.
The debate isn’t whether or not there are naturalist metaphysics on offer. The debate is whether they apply.
So unless I am badly misreading you, your claim is trivially false. Sorting amongst possible explanations isn’t the same as not having any to sort amongst, obviously.
“It is the way it is” is an explanation. It’s not satisfying to us intuitively because we’re not used to parts of the universe that could even possibly be brute facts, or consequences of infinite regress, or consequences of true randomness, or anything of that sort. But that just says that our intuitive expectations are false. Just like they’re false about Schrodinger’s cat or the double slit experiment.
If I construct a model where there’s a brute fact that explains everything, and that’s how our world looks to be, I’ve done my job as a scientist. Done and dusted. Go home. As Carroll said to Craig, “But what’s the cause?” can easily be as silly a question as “But where does the film go?” in a camera phone.
Fair enough. To be precise, these objections are more to premise 4.
No, actually, we can point out that your definition of “successful” isn’t useful. You said it yourself. This is a conditional, not a definition. You aren’t defining successful in this fashion. (And you shouldn’t be, but that’d be a different problem).
But you are right, we would need to point to scientific theories as examples, rather than just point out that this is just not how science operates generally.
Newtonian mechanics.
Successful and used all of the time. Also known to be inaccurate both in its ontology and its predictions.
Mertonian strain.
Explains some range of behavior. Can’t explain all of it.
Indeed, virtually all social science theories end up either being framing approaches to explore the world which will inherently bias the data but can be starting points or explain some range of variables with some range of variables but have exceptions. And those are considered successful in their field.
It’s possible to go on and on. Lots of theories have been proposed in scientific history without an underlying causal mechanism. Newton himself despised that he had to accept strange, seemingly acausal, action at a distance. But that’s what the data said. Other theories have been proposed as just a part of the picture. I don’t think Darwin was under any delusions that selection would explain all of life’s diversity. He had found a mechanism, just not all of it. And yet no one would say natural selection isn’t a wildly successful theory.
Which leads me to worry that you’ll complain about “success”. Which leads me to think that you formulated this (possibly unwittingly) to create an unfalsifiable test, not to make a serious argument.
You don’t define any such condition in your paper, Tomas. You object to things like a brute fact explanation with what I would call specious reasoning and you absolutely commit the motte-and-bailey fallacy we are discussing there. I actually don’t see clear, unambiguous criteria you have on offer for what even makes an ultimate explanation.
And you do realize that if we’re talking about scientific models that are on offer I can’t use something like a traditional syllogism, right? This is sort of emblematic.
But, okay.
Imagine a universe that is a single particle moving upward. It moves through four separate positions. So each previous position “causes” the next. Then time resets at the fourth position and we go back through them again.
This is an infinitely looping system. (I’m doing it this way so we don’t have to visualize an actual infinity). Is it coherent? Yep. You could visualize it on a computer right now. We can construct a mathematical model that explains each step. It’s properly Newtonian in the sense that the particle just keeps moving because no force exists to stop, slow or deflect it.
Is this our universe? Nope! But it’s a universe.
This is the fundamental problem with your approach, Tomas. This is how we assess scientific theories. Are they a good model of the world? Then they’re a good theory. The goal for us in the real world is to find which model matches the universe we live in the best. (Or, in some contexts, to find a nice, simple model that works well enough. People use Newton all the time even though it’s wrong because in most contexts it’s not wrong enough to matter).
I’ll just quote in relevant part from Hawking: “A theory is not Truth with a capital T, not a rule, not fact, not the final word. You might think of a theory as a toy boat. To find out whether it floats, you set it on the water. You test it. When it flounders, you pull it out of the water and make some changes, or you start again and build a different boat, benefiting from what you’ve learned from the failure.” Similarly, “it must accurately describe a large class of observations on the basis of a model that contains only a few arbitrary elements, and it must make definite predictions about the results of future observations”.
https://www.themarginalian.org/2018/08/20/stephen-hawking-a-brief-history-of-time-theory/
Now, could Hawking be wrong? Sure! But the point is that he testifies to how working scientists actually use the concepts. Which tells us that you’re arguing from the armchair, which indicates you may have gotten the universe wrong.
So when you say “Notice that this principle – our second premise in the Main Argument – does not entail anything closely resembling a Principle of Sufficient Reason, namely any principle that says, roughly, that any contingent fact must have a cause, reason, or explanation. For it’s consistent with our second premise that explanations commonly, even always, fail. If we lived in a world in which any proposed explanation crucially involved some element that called out for explanation but lacked one, the right thing to conclude, I would say, is that we live in a world in which, ultimately, nothing could be explained. We would live in a deeply and irredeemably mysterious world. As unsavory as that sounds, it is possible. In fact, I will argue, a committed Naturalist must accept that we actually inhabit such a deeply and irredeemably mysterious world, a world impervious to scientific investigation”, you might be right, buddy! We don’t know yet. Which makes your argument toothless. But that isn’t a problem for science, because science isn’t trying to answer every mystery, necessarily. Only the ones that can be. Contrary to what you seem to be arguing, this isn’t a problem. And no one thinks it is.
Of course, it’s quite likely that there is an underlying ontology, something like a fundamental geometry, that provides an ultimate explanation. (And, of course, “A totally mysterious entity beyond our possible comprehension wanted us to be able to play in its sandbox” isn’t an explanation either, it’s just a brute fact, so you are being hypocritical about what kind of explanations you accept).
Done above, Tomas. What I expect is quibbling over either the terms successful or theory. Which I think exposes how arbitrary your conditionals are.
But the reason why I balked as this being obviously false, Tomas, is because I’ve read Hawking, and engaged with social science, and you actually learn very quickly that this hope to have some master theory is dashed all the time, and may in principle not be available to us. Including for reasons having to do with fundamental laws of computation. (You can’t simulate a thing on anything smaller than the thing itself).
Which tells me that you haven’t been through that process, and learned to adjust your expectations for what is explainable.
Which not only supports the impression that your argument hinges on scientific illiteracy (to a seriously very basic level), but also that your argument is obviously false because you are mistaken as to how actually comprehensible the universe is.
Oh, if we’re talking about scientific explanations , as your five points originally stated, then we’re actually even worse off for you. “It’s raining because there are clouds in the sky” counts as a scientific explanation. It’s not a deep theory, but it’s a connection between two facts. So the very fact that you’re using the term “explanation”, which as far as I can tell also remains undefined in your paper, is much worse (I only see five instances of the word “define” or “defining” in the paper), which is why Richard pointed out how simple a scientific fact or piece of data can be. We are steelmanning you by using scientific theory.
But this should worry you, in fact. Because maybe what you’re doing is asking from science for something, what you think of as an “explanation”, and it’s not on offer. Because the universe doesn’t offer “explanations” of the type that would satisfy you, even after hundreds of years of detailed work with the best available instruments. Because there’s, apparently, no God.
If I leave out “ceteris paribus” from “Increasing supply causes prices to go down”, I’m not strawmanning the position. The only complaint would be if someone were to nitpick about the cases where that doesn’t happen. But that’s not what we’re doing here.
I will grant, though, that you are correct that, all else held equal, a theory that leaves unexplained anything that calls out for explanation is probably worse. (By the way, everything calls out for explanation. Nothing is unimportant in a model, if we’re being careful. So this is already worrisome). But that’s for a very good reason. We’ve constructed the way we assess theories because we’re used to the fact that theories that explain their elements tend to be better. (They’re not always, of course, but that’s the ceteris paribus part). But we also have very good reason to expect, for reasons you and I share, that the underlying explanation for the universe won’t be like the explanations within the universe. (I’ll return to this in a bit).
What you seem to totally underestimate is how much work that ceteris paribus is doing for you. In reality, a theory that leaves something unexplained but is incredibly accurate and predictive is much better than one that is even only slightly less accurate and predictive. In fact, such a theory can easily be better because it highlights that an element needs further study. It’s probably a mark against a theory that parts are unexplained, but it’s just not a big one.
Richard’s argument against you still applies . Because, indeed, we don’t care that the law of buoyancy doesn’t explain the Big Bang, even though you could claim it needs to because on a long enough timeline back the Big Bang explains why the physics are such that buoyancy works the way it does. It’s not a footnote that liquids work the way they do because of underlying forces and particles. There is still something to be explained. Yet it’s successful. Obviously so. So Richard gave you all the examples you were asking for here and you didn’t respond to them. Do you see why this comes off somewhat as special pleading?
To return to the point about theories of universes: We actually have no a priori reason to think those theories will be very easily accessible or satisfying, and good reason not to. And so does the theist. “An unexplained triune mind that has no causal antecedent and no parallel wanted the universe to be understandable to some arbitrary degree” is just as brute a fact as anything else. God is incredibly finely tuned. This doesn’t bother you. So why should it bother any of us? Tomas, you and I accept that explanations that are intuitive and make sense and tidy everything up are likely to disappear when we get to matters beyond human scales and timelines. It’s as true of black holes as it is the universe. So your attempt to try to argue that this is a problem for science just comes off as such special pleading. And it comes down to this complaint, which comes down to you overvaluing an explanation in a theory.
I do agree it would do that, if it did it perfectly! But not if it did both explanations more badly than either an evolutionary or quantum theory individually and had to just stipulate similarities between quarks and cells for the theory to work. Then it’d actually be garbage. Which is why lots of theories that try to basically imagine biological principles in the universe are naked woo.
But, of course, even if evolution could later explain quarks, it would still be a good explanation for how it now explained life. And didn’t need to explain how life emerged to be evolved upon.
Seriously, dude. You picked the theory that has the hole even creationists are smart enough to pick up on that the first lifeform is not going to evolve. (Or, more accurately, the very earliest chemicals that don’t have a capacity for selection – and notice how even that had to take some time to develop as a theory). And yet it’s a theory that empirically scientists view as incredibly successful.
So, yes, if no ultimate explanations were possible, we would view our theories as less satisfying than if they were. Boo hoo. The universe doesn’t owe us the best possible theories. That would be the ultimate reaction from scientists.
As Richard has pointed out vis-a-vis parsimony, this is why we’re happy to accept the atomic theory over Aristotelian elements, and a massive array of quarks over a few simple particles. Even parsimony matters little compared to predictive accuracy. Let alone explanatory scope. We’re happy to imagine a cluttered, complicated universe with lots of things beyond our understanding if it does the actual job we need. (And, again, the fact that we’ve gotten used to that is massive disproof of God).
In fact, there’s even pressure these days to try to cut variables, even ones we know probably matter, from theories. https://saturncloud.io/blog/how-to-increase-the-model-accuracy-of-multiple-linear-regression/#:~:text=Including%20irrelevant%20variables%20in%20the,to%20generalize%20to%20new%20data. . Because unless your model is near-perfect and your data is incredibly granular, you’re actually making the net picture less clear.
This really does expose the motte-and-bailey fallacy we accused you of, Tomas. We just aren’t looking for final explanations with the fervor you imagine. Because the whole history of science has found that they’re not easily available, and in principle may be impossible in all possible worlds (yes, even your proposed theistic one, wherein only God would ever be able to perfectly model the world and that’s only because of the seemingly impossible power of omniscience and omnipotence), and so we’ve had to get used to theories that are as Hawking described.
Contrary to how you’ve accused Richard of behaving here, you’ll actually find that this is an incredibly reasonable commentariat by and large with good moderation. Of course you have countless professional and personal obligations, as do I, and I really do appreciate that you came out to defend your ideas. And I’m not anonymous, as you could have found out with a bit of Internet sleuthing, but I won’t knock you for that. Frederic Christie will get you my Quora responses, where I think you’ll see I’m hardly an inveterate religious critic.
I have indeed read the paper, and I find it unconvincing. And I really do think my points about things like showing the range of intelligibility and completeness of theories we would ostensibly expect under various metaphysics that you didn’t address still hold quite profoundly. I don’t think anyone has any of these debates very seriously (which I blame the theists for).
But the reason I find it unconvincing is what I’ve laid out here and what Richard has laid out. I think you misunderstand what the state of science is and so your argument is at best armchair prescriptivism from someone who isn’t in the fields in question. I think you are happily engaging in special pleading on how God makes the world understandable and on the nature of any God theories. I think you are vulnerable to all the criticisms Carroll made of Craig (and he does go in his work there and in God Is Not A Good Theory into some elements of theory construction). And I think you are constructing a framework to ignore that the world in fact did not proceed as if the universe were designed to be fundamentally understandable to us by a God, which is a strong disproof of your predictions and explains why science has taken its provisional, model-testing course.
This is a reply to a comment “Fred B-C” from September 10, at 6:33 PM. But I didn’t see a “Reply” button under his comment, so I pushed the “Reply” button under my previous comment. :-/
You said,
“So unless I am badly misreading you, your claim is trivially false.”
Which claim? Initially, you said my claim about what I meant by “but lacks one” was trivially false. Now you seem to have in mind a different claim, but you didn’t say which claim you have in mind. Please tell me exactly what claim you think is trivially false. In the form of a complete, declarative sentence.
In this most recent comment of yours, you seem to back off your claim that my second premise is false, and now you say instead, “Fair enough. To be precise, these objections are more to premise 4.” Here’s my fourth premise:
-4- If Naturalism is true, then every natural regularity calls out for explanation but lacks one.
What, exactly, is your objection to premise 4? Premise 4 is again a conditional, so an objection would be a counterexample. In this case, a description of how it could be that Naturalism is true, but not every natural regularity calls out for explanation but lacks one. So, an example, assuming Naturalism, of a natural regularity that either doesn’t call out for explanation, or which does but that ultimately has one.
I scrolled all the way through your comment, but I found nothing that looked like a possible objection to premise 4. If I missed it, please lay it out clearly and carefully. What is your proposed counterexample to premise 4?
In your previous comment, you made this claim: the fact that there could be past eternality and thus infinite causal regress proves that my condition of ultimate explanation is incoherent. I asked to see this proof, and I asked what you take my “condition of ultimate explanation” to be.
In reply, you said, “You don’t define any such condition in your paper.” Oh. But then why did you claim that you could prove my “condition of ultimate explanation” is incoherent, if I didn’t offer any such condition? What was in your mind when you were typing that sentence of yours, and making that rather bold claim?
If you retract your claim that I offered any “condition of ultimate explanation,” then I suppose you also retract your claim to be able to prove that this condition is incoherent.
In your previous comment, you claimed that my first and second premises are both obviously false. In my reply, I asked you to please pony up actual counterexamples to these premises, to help us see why these premises are obviously false. To my utter astonishment, you claim that you did. “Done above,” you said.
I don’t wish to be patronizing, and I’m sorry if this comes across like I’m holding your hand through this, but what exactly is your objection to these premises?
-1- Any scientific explanation can be successful only if it crucially involves a natural regularity.
-2- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
Let’s start with premise 1. What is your proposed counterexample to premise 1? A counterexample to 1 would be a case of a successful scientific explanation that does not crucially involve any natural regularity. Recall what I take “successful” to mean in this regard in my paper. I didn’t see any such case in your comment.
What is your proposed counterexample to premise 2? A counterexample to 2 would be a successful explanation that crucially involves an element that calls out for explanation but lacks one. Recall what I take “but lacks one” to mean in this regard in my paper. I didn’t see any such case in your comment.
The closest we get to an objection to either premise 1 or premise 2 is when you say you’ve “read Hawking, and engaged with social science, and you actually learn very quickly that this hope to have some master theory is dashed all the time, and may in principle not be available to us.”
But this is not a counterexample to either premise. Neither premise 1 nor premise 2 entails that we must HAVE some “master theory,” or that such a theory must be “available to us.” So, again, what you say here is simply wide of the mark. It doesn’t engage the premises as written.
I recommend that you review what I said about the “but lacks one” phrase. In endnote 15, I say this: “By ‘but lacks one’, I don’t mean merely that we don’t know what the explanation for this element would be, or that it’s not available to us, or some such constraint on our epistemic access to an explanation. I mean that there is literally no explanation to be had.” Please read the rest of endnote 15, where I offer an illustration of this distinction. https://philpapers.org/archive/BOGINI.pdf
In your previous comment, you asked about this quotation from my paper: “ceteris paribus it is a deficiency of a theory if the theory leaves it unexplained.” You said that, if this was an accurate quotation, then “[my] definition of a theory is terrible. Again, to the point of suggesting scientific illiteracy.” Strong words!
But after I gave you the full quotation, including text from an endnote that preempted your concern, you now retreat from your criticism, saying, “ I will grant, though, that you are correct that, all else held equal, a theory that leaves unexplained anything that calls out for explanation is probably worse.” Thank you for admitting that, and for conceding that your previous criticism was mistaken. It takes an honest mind to back off a criticism like that, worded as strongly as it was.
You say, “Richard’s argument against you still applies.” But: What Argument? Please lay out this argument step by step, using common argument forms/inference rules. I honestly have no idea what argument you have in mind. Show me the argument.
Hopefully, the conclusion of this argument is the negation of some premise(s) in my argument. Then it will be clear how what you’re saying is relevant to my argument.
I said that I would likely respond “if you raise any intriguing objections that clearly engage one or more premises in my argument.” I don’t think you raised any objections like that, but apparently I responded anyway haha. If you reply again, and your reply doesn’t contain any objections that clearly engage one or more premises in my argument, I don’t know if I’ll be able to refrain from responding, but if I do refrain from responding, please know why.
Actually, I continue to maintain both are false, but I recognize that your conditional could be a disguised definition, at which point I can only say that your definition is not useful or applicable.
So, to two, I respond that, while an explanation for all elements is helpful to a theory, it is not necessary for that theory to be successful. Or, to put it in your language which is less useful, an explanation need not have no element that calls out for explanation but lacks one to be successful.
And for four, I argue that naturalism in fact can support many complete ontologies where a theory that had total explanation for all elements is possible. Richard’s example of a space-time geometry that allows all possible variations we see from its properties is sufficient. At that point, the only question left is metaphysical, i.e. “Why that geometry and not some other?” But that will apply to any complete theory, i.e. “Why that God and not some other”?
Yes, that may not be our reality . But that wasn’t your objection.
Because your argument actually depends on having one. Your failure to do so means that your argument is either trivial or false. The whole explanatory scope discussion we had is emblematic. It is not a substantial mark against evolution that it also does not explain quantum theory. It would be better if it could do so equally well, but it not doing so doesn’t make it not successful.
Naturalistic theories can at this point provide everything but a metaphysically complete “Why?” picture as a hypothetical. But the same is true of supernaturalist theories, because no one has a robust ultimate answer. And the lack of a robust ultimate answer is the only thing that would make either the law of buoyancy or quantum physics unsuccessful by your standards.
I don’t mean to sound patronizing in turn, but I offered Mertonian strain and Newtonian mechanics as counter-examples, and went on at length about it, and you haven’t responded and apparently didn’t even read it. Now it’s your turn. Why are those unsatisfying? Or do you concede I am correct?
And what is your actual objection to Hawking, by the way, while we’re at it? Since his argument directly contradicts your #2 because it dismisses complete explanatory power as necessary for a successful theory. So at this point I have argued against #2 and #4 both with example and theory, to no response.
By the way, I do think #1 is also debatably false, but I’m not going to bother defended that. I do think one could come up with a “theory” or at least a taxonomy or set of laws that would work even without some regularity. One could try to arrive at rules even of a truly random infinite card deck, or arrive at a taxonomy of truly unrelated things (the way that Linnaeus effectively proceeded).
The argument that it is false that a successful explanation must explain every element, even ones that seem to call for explanation. (I say “seem” because I submit that actually none “call” for explanation, because if your explanation is good enough for what you can explain it’s moot to complain about what you can’t, but I’m trying to work within your framework which I think is flawed from the outset). And that you are making a motte-and-bailey fallacy because in your own work you then use conventional theory criteria. As Richard laid out. Which you have yet to respond to. Despite me mentioning it, and now explaining it directly.
Why is it false? For the same reasons you say the converse is true. Well-accepted principles of creating theories do view explanatory scope and elegance as useful, but not as necessary. Something that has narrow explanatory scope, or stipulated elements, or finely tuned elements, or unjustified assumptions, can be acceptable as a theory, even rather a strong one. “I don’t know how life started, but once it did as a single organism, it evolved” is an immensely powerful theory.
Take Richard’s sneeze example. It’s not a knock against the explanation “Oh, people say ‘God bless you’ because sneezing worries people and in many contexts they would say something like ‘May there be well’ but in our context it became custom to use the God formulation” that it doesn’t explain the entirety of the sociology of religion and culture or the entirety of neurology. Anyone who was actually wanting a satisfactory explanation would be happy with that. More things could be explained, as anyone who’s worked with a five year old can testify to.
Seriously, I’m not being any less precise than you were in your paper. So can you engage with what I’m saying or not? I will happily answer questions but this seems like you are being deliberately obtuse.
Tomas, I haven’t caught up on your conversation here with Fred, so I don’t yet know what issues are raised there. But the bottom line regarding my article is that Premise 2 is false and therefore your argument fails to establish its conclusion. You try to disguise this with a number of other fallacies (such as the argument from ignorance and equivocation), but being fallacies they do not rescue the argument.
To use Premise 2 as evidence for your conclusion, you need it to be the case that there are a lot of examples of science succeeding in its defined sense. There are none. So your conclusion is without evidence. You instead make the mistake of appealing to a completely different set of evidence for this move: the actual success of science, which never involves any instantiation of Premise 2. This is an equivocation fallacy. And it is fatal to your argument.
You also, separately, make the mistake of appealing to various convoluted forms of the Argument from Ignorance, and attempt to justify this with an example from Aristotle of that method failing, and then proceeding as if you had established it was successful. This results in a lot of wasted words, because the example is of a failed method, not a successful one.
In the end, all that wheel spinning just gets you back round to the original error: there is no evidence Premise 2 is true.
Your argument therefore fails.
Nothing you are saying now addresses any of this. You have no evidence that Premise 2 is true. You only have evidence that science succeeds without it.
And all your attempts to bypass this (such as by trying to revive a provably failed argument of Aristotle) are fallacious.
Thanks for your reply, Dr. Carrier. You say that “Premise 2 is false.” Just to recap, this is my second premise:
-2- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
This premise is a conditional. To show that a conditional is false, we’d need to provide a counterexample, a situation in which the antecedent is true, yet the consequent is false. In this case, that would be a scenario in which an explanation is successful (i.e. produces understanding), and yet crucially involves an element that calls out for explanation but lacks one (i.e. that element literally has no explanation).
Your comment does not supply any case like this. Do you have any case like this? If not, why do you say that premise 2 is false?
But elsewhere it seems you think rather that this is the “original error: there is no evidence Premise 2 is true.” Of course, saying that premise 2 is false is different from saying there’s no reason to think it’s true. If you have no good counterexample to premise 2, and no good evidence for premise 2, the right response is to suspend judgment on premise 2, to be agnostic. You shouldn’t say that it’s false; you should say that you’re not sure whether it’s true or false. So I’m sensing a tension in your comment: is your objection to 2 that it’s false, or that there’s no good reason to think it’s true? (Or both?)
But I did supply evidence for premise 2. I devote an entire section of my paper to defending premise 2, starting on page 5 of my paper: https://philpapers.org/archive/BOGINI.pdf
I invite your readers to check that out for themselves, and I encourage them also to read endnote 15 and endnote 18.
……………………….
You say, “To use Premise 2 as evidence for your conclusion, you need it to be the case that there are a lot of examples of science succeeding in its defined sense.”
Unfortunately, this is a misunderstanding of how conditionals work. Premise 2 says only that IF an explanation is successful, THEN it crucially involves no element that calls out for explanation but lacks one. (It’s true that I used an “…only if…” construction when I wrote the premise, but this is equivalent to an “if…then…” construction.)
And a conditional can be true even if the antecedent is false. And a conditional can be eminently reasonable, even if we have no good reason to believe the antecedent, an in fact excellent reason to think the antecedent is false. Here’s an example of all these things: if Joe Biden is a dog, then Joe Biden is a mammal. That conditional is true, and eminently reasonable to believe, even though we have no reason to believe the antecedent, in fact excellent reason to think the antecedent is false.
So my premise 2 could be true, even if the antecedent is false or unsupported. So, even if you were right that I hadn’t proven that the antecedent is true, and even if were in fact false, nevertheless premise 2 itself could still be true. So your proposed objection is consistent with the truth of premise 2, and therefore not really an objection at all.
It’s true that, later, I go on to assert that some scientific explanations have succeeded in producing understanding. Is this what you deny? Do you think that scientific explanations have never produced any understanding of the natural world? I’d be surprised to hear that’s your view, though you might accept that view and also think my main argument is sound. You might agree that, if Naturalism is true, then no scientific explanation can be successful, run the modus ponens, and just be a bit of a dour Naturalist, pessimistic about the power of science. As I say in my paper, that would be a pretty surprising turn of events, since typically it’s Naturalists who trumpet science in support of their view. Sort of weird now to say that science has never produced any understanding of the natural world, no? But is that your considered view, that science has never produced any understanding of the natural world?
……………………….
You say that I “make the mistake of appealing to various convoluted forms of the Argument from Ignorance.” Please show me how, exactly, my Main Argument is an Argument from Ignorance. Once again, here is my argument:
-1- Any scientific explanation can be successful only if it crucially involves a natural regularity.
-2- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
-3- So, a scientific explanation can be successful only if it crucially involves a natural regularity, and this regularity does not call out for explanation while lacking one.
-4- If Naturalism is true, then every natural regularity calls out for explanation but lacks one.
-5- So, if Naturalism is true, then no scientific explanation can be successful.
But instead of relying on a Critical Thinking 101 list of informal fallacies popular with the “Average Redditor”-types, it would be better if we could simply say which premise is false and why, or which inference fails and why.
You say that “Nothing you are saying now addresses any of this.” I understand that it must feel, from your perspective, like I’m not addressing your points. But I hope you can understand that this is because, from my perspective, your objections are not engaging my actual argument. I appreciate that, from your perspective, it seems like you are lobbing devastating objections my way, and I’m refusing to acknowledge how devastating they are. But, from my perspective, these objections you’re lobbing are heading in the opposite direction of my argument, and so I’m trying to explain why that is, and to encourage you to aim them instead in my direction. I’m practically standing on top of my argument, waving my hands and shouting “Over here!” while giving careful instructions on how to craft counterexamples to my premises.
In this case, again, it may feel like I’m not addressing your objection, because your objection is that the antecedent of premise 2 is false or unsupported. And I’m pointing out that premise 2 can be true even if its antecedent is false or unsupported. That’s how conditionals work. So, far from sinking my battleship, this objection is landing miles away from the game board.
But, again, I do appreciate your time and effort in crafting your responses. Thank you for that. I recall in grad school being informed that one shouldn’t expect people to agree with one’s work, and that receiving objections is the way academics compliment one another’s efforts haha. So, again, I appreciate your time and effort.
“IF an explanation is successful, THEN it crucially involves no element that calls out for explanation but lacks one.”
Evolution IS a successful explanation, BUT it involves an element that calls out for an explanation (how did life initially begin) but lacks an explanation.
The Big Bang is a successful explanation for how the universe expanded from its initial state, BUT it involves an element that calls out for an explanation (what preceded the Universe’s initial state ) but lacks an explanation.
Molkien is correct. Indeed, more so: “Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one” doesn’t even limit the declaration to science. All explanations “call out for further explanation and lack at least one.” The evidence of reality refutes the premise. We don’t even have to go as far as esoteric topics like Big Bang theory.
Of course, the trend line is that the lacking explanations will someday not be lacking, as all past scientific progress proves. And that trend line predicts the bottom of the explanatory chain will be something natural. Because explanations of explanations so far always are.
Near as I can tell, this is the sort of sticking point of our interactions based on my interaction with Tomas.
Tomas makes the quite reasonable point that, all else held equal, a theory that has greater explanatory scope or fewer unexplained elements (whether of great importance or not) or fewer elements of arbitrary finetuning or assumption to get around being unable to provide an explanation is going to be better than one that isn’t.
I think we could quibble to some extent even here, I can imagine some corner cases where a theory that is quite narrow and equal in a lot of respects could still be preferable to a broad one. But the point generally is fair. It is reasonable to look at an element of a theory that is unexplained and ask further questions. What is c? How does gravity actually accomplish strange action at a distance?
And when we do find elegant answers to these questions, we do learn some underlying truths. When Einstein saw that the speed of light and an unexplained constant in Maxwell’s equations were connected, we actually learned something about the universe. So it is fair to ask for important unexplained elements in an explanation to be explained.
What it seems to me Tomas has done is equivocate from “A theory, all else held equal, would be better if it had broader explanatory scope and fewer unexplained elements” to “A theory fails if it does not have broad unexplanatory scope and no major unexplained elements”.
Being as charitable as I can, I think this is why Tomas used the phrasing of “crucially involves no element that calls out for explanation but lacks one”. I think Tomas agrees that theories don’t need to explain everything: Minor, ancillary or even somewhat major parts of the explanation that aren’t critical components could be left unexplained. But something like Newton being unable to have even an idea for the ontology of gravity would by his #2 be an unsuccessful theory. And I can see that as being a potentially reasonable position, even if ultimately not tenable in light of the history of science. It’s sort of pre-Newtonian, but at the same time underlying causal mechanisms really are important.
Tomas calls his #2 a conditional, but I don’t think that’s really what he’s doing. I think he’s essentially making it an argument for a definition or a criterion. I think Tomas is arguing that science should take the standard that any explanation that has brute elements.
And Tomas is convinced that it is likely that only explanations like a brute fact are available. That was his footnote 16 and his analysis of Brute Foundationalism.
Of course, by not surveying options like the ones you’ve described, Richard, like the inevitability of a given universe from a geometry of spacetime or the inevitability of a profligate multiverse from a true nothing condition, he’s not really addressed the array of options on offer. And it is still an argument from ignorance, because it testifies only to his assessment of the slate of theories today. And, most centrally, none of his arguments excuse God, which is always going to be a far more unexplained, fundamentally mysterious thing.
Indeed, that would be my point to Tomas. I think Tomas is quite happy to look at God, see something with elements beyond our comprehension, and still be happy with everything that comes from that prediction. Even if we find something like a brute foundationalism, we would be epistemically justified with exactly the same satisfaction. We would look at one element that isn’t satisfying, but isn’t illogical, and then see all the other explanations.
This is the point of analogies like the yardstick coming into existence at the first inch. If we live in a past-finite universe that just was, the simplest interpretation of the Big Bang, that’s not a problem because time didn’t precede it. There’s no causal chain left unexplained. Only metaphysical “But why that universe rather than another?” question. Which I don’t think Tomas’ paper or argument is in a state to differentiate between.
Ultimately, if we put Tomas’ point as “We may have to accept a theory of the universe that has, to some degree, unexplained or unexplainable elements”, then he’s almost certainly right. But that’s exactly like Newton complaining about strange action. If it’s what the evidence says, it’s what’s most probably true. And naturalist and supernaturalist theories alike have to accept that.
“Do you think that scientific explanations have never produced any understanding of the natural world? I’d be surprised to hear that’s your view, though you might accept that view and also think my main argument is sound.”
These are two completely different meanings of the word successful. According to the conditional there’s simply no such thing as a successful theory. As the paper puts it, “the promise
of an explanation of the stationary Earth has merely been deferred, but ultimately not
fulfilled.”
Assuming we think there are successful theories, meaning for example that they have predictive power or are otherwise useful to us, that would mean the conditional is false, more akin to “if Joe Biden had wings, he’d be a dog.” Maybe a dog with wings would also be a dog, but you wouldn’t say it was a dog because it had wings.
What Frans said.
I am astonished to see Tomas continue to make this mistake: confusing his Premise 2 with a description of the actual success of science, which violates Premise 2 (it never involves meeting the conditions it states).
This is the central equivocation fallacy of his entire paper.
The “if” statement is the antecedent; the consequent is “only then is an explanation successful.” The counterexample is therefore “successful explanations that DO contain elements that call out for explanation but lack one.” Science consists entirely of such successes. The history of science thus flatly refutes Premise 2. In fact I am not aware of any explanation in history that confirms Premise 2, as in, an explanation that is “successful” and “contains no element that calls out for explanation but lacks one.” All successful explanations lack this feature. So your claim that a successful explanation requires such a feature is false.
And yes, this is a correct understanding of how conditionals work. “Only if” means there can be no other examples of successful explanations than Premise 2 defines. It is therefore refuted by the entire history of science, which consists of nothing but successes that don’t conform to your condition.
You can trivially intend this to be a redefinition of success, such that you are only stipulating a new standard of success and not describing any real one that exists, but once you do that, your evidence for Premise 2 vanishes. Hence you appeal to the actual success of science in support of Premise 2; that’s the point of your paper. So you aren’t presenting a new definition of success. You are claiming that scientific success proves Premise 2. You are therefore claiming Premise 2 is a description of actual scientific success, and therefore actual scientific success is evidence for your conclusion. Since it is not a description of actual scientific success, since science in fact refutes Premise 2 (it is NOT the definition of success in any science and never has been), your conclusion fails along with the Premise.
-:-
Though you are still failing to get even these basic points here, so you aren’t realizing this, I suspect what you want to argue (but failed to) is just what I point out in my critique (which is why I included a whole section on it as well):
You want to argue something to the effect that “science could only have been successful if there is some ontological ground to reality, some ‘final thing’ that explains all explanations” (what I call the Argument to the Ontological Whatsit). But that formulation of a Premise (found nowhere in your paper) could not get you to nonnaturalism (Premise 4 would then fail by want of proof).
Because your three stated (and by understatement, four) grounding conditions (infinitism, circularism, brute-factism, and necessary-factism) exhaust all logical possibilities (regardless of whether they are natural or nonnatural, there is no fifth possibility), and therefore both naturalism and nonnaturalism face this same tetrachotomy. Therefore, that this tetrachotomy obtains cannot be evidence for either naturalism or nonnaturalism.
For example, you never present any evidence that God is a necessary being (ontological arguments have failed for centuries despite abundant tries) or that no natural fact could be; you never present any evidence that if God can be self-causing or a brute-fact that no natural fact can be; and so on. There simply is no argument to nonnaturalism in your paper. Other than the equivocation fallacy, by which you attempt to claim the actual success of science is evidence for nonnaturalism, combined with a failed Aristotelian Argument from Ignorance bolstering Premise 4.
“Molkien,” thank you for making a clear attempt at giving a counterexample to premise 2! Nice work.
You said:
“Evolution IS a successful explanation, BUT it involves an element that calls out for an explanation (how did life initially begin) but lacks an explanation. The Big Bang is a successful explanation for how the universe expanded from its initial state, BUT it involves an element that calls out for an explanation (what preceded the Universe’s initial state ) but lacks an explanation.”
In both of these cases, I myself think the elements you mention do have further explanations. But I suspect we might disagree about that. Do we though? Do you really think that there’s literally no further explanation for how life initially began? Remember, when I say “but lacks one,” in premise 2, I mean that literally. This element that calls out for explanation literally lacks any explanation. It’s not just that we don’t know what the explanation is, but rather that there literally isn’t one. I would have thought that a Naturalist would say there must be some further explanation for how life initially began, we just don’t yet know what it is. And one goal of science in this area is to uncover what that explanation is. No?
So maybe the Big Bang example is better. Maybe there you really do believe that the initial state itself calls out for explanation but lacks one. (I myself disagree that it lacks one, so this isn’t a problem for my view.) In that case, I would say that any explanation bottoming out in such a mystery would be an unsuccessful explanation. It would be just like the turtle case.
Suppose we’re wondering about the position and apparent stability of the Earth. If, by way of explanation, you say the Earth rests on the back of elephants, that might sound like you’ve explained the position and stability of the Earth. But not so fast, I’d say: Where did these elephants come from? Why are they so stable? That is, the elephants call out for explanation. If you were to go on to say that they rest on a turtle, again that might sound like you’ve explained the position and stability of the elephants. But, again, not so fast: this turtle would call out for explanation. If you were to go on to say that this turtle literally has no explanation, then I think what we discover is we don’t really understand the position and stability of the Earth. The appearance of explanation was deceiving. Explanation kept getting deferred, down the chain (from the elephants to the turtle), but an explanation was never really delivered. We haven’t successfully explained the position and stability of the Earth after all. And, if that’s right, then we don’t have here a counterexample to premise 2. Rather, this would be an instance of premise 2.
If we add enough links in the chain (Earth, on elephants, on turtles, on bulls, on…), we may fail to notice the mystery, or succeed in putting the mystery out of our minds. To be frank, I think this is what many Naturalists do when thinking scientifically. They just put the mysterious foundations out of their minds, and manage to proceed as if there’s no mysterious, unexplained turtle down there. The purpose of my paper is to direct our attention to this problem.
Please check out page 6ff in my paper. And a response to objections in endnote 18, and this bit of endnote 16: “it’s consistent with our second premise that explanations commonly, even always, fail. If we lived in a world in which any proposed explanation crucially involved some element that called out for explanation but lacked one, the right thing to conclude, I would say, is that we live in a world in which, ultimately, nothing could be explained. We would live in a deeply and irredeemably mysterious world. As unsavory as that sounds, it is possible. In fact, I will argue, a committed Naturalist must accept that we actually inhabit such a deeply and irredeemably mysterious world, a world impervious to scientific investigation.”
https://philpapers.org/archive/BOGINI.pdf
Dr. Carrier, you say: “All explanations ‘call out for further explanation and lack at least one’.”
First of all, whom are you quoting here? Unfortunately, it looks to me like you’re doing that thing where you put your own words in quotation marks, and appear to attribute these words to someone else.
Second, I think it’s false that all explanations crucially feature any element that calls out for explanation but lacks one. If you think of explanations in mathematics, logic, morality, or philosophy (for example), where the explanans are obvious and necessary, I think you’d be able to supply a counterexample to your claim here.
You also say, “Of course, the trend line is that the lacking explanations will someday not be lacking, as all past scientific progress proves.”
This makes me worry that you’re still misunderstanding me when I say “but lacks one.” When I say that an element calls out for explanation but lacks one, I don’t mean merely that we don’t currently know what the explanation is, but maybe we’ll discover it in the future. As I say a few times in the paper, what I mean is that the element calls out for explanation but literally lacks one. There’s literally no explanation there to be discovered. I think that if you came to terms with what I mean by “but lacks one,” you might be able to see why many of your proposed objections to premise 2 don’t hit the mark.
I perfectly well understand what you mean. I even make sure to point it out in my critique.
As to the rest, see other comment.
Dr. Carrier,
Sorry, I just saw that you posted another comment. In that comment, you said, “The counterexample is therefore ‘successful explanations that DO contain elements that call out for explanation but lack one’. Science consists entirely of such successes.”
Minor point but, again, whose words are those in the quotation marks? Not mine. But, if yours, why use quotation marks?
Another minor point: you say, “the consequent is ‘only then is an explanation successful’.” No, that is not the consequent of premise 2. The consequent of premise 2 is: “it [i.e. the explanation mentioned in the antecedent] crucially involves no element that calls out for explanation but lacks one.”
Major point: No, I don’t think that science consists entirely of such successes. Molkien tried to present two such cases, and I replied to those above. And the main burden of my paper was to convince you that science does NOT consist entirely of such successes. But maybe we could just work with the examples Molkien provided, if you found my response to Molkien unsatisfactory.
Please learn the use of rhetorical quotation marks.
And you are wrong. Premise 2 contains “only if.” That makes it what I formulate, not what you are. Did you not notice this? Or do you not know the logical consequence of an “only if” operator in a conditional?
And please give me an example of an actual successful scientific explanation that does not lack any explanation it calls out for. I am most curious to see such a marvel.
If all you mean is that there are some non-scientific explanations that can claim this, then please explain: What relevance are those to your paper?
Here’s a quick reply to “Frans.” Thanks for your comment. You said,
“These are two completely different meanings of the word successful.”
True. I think Dr. Carrier and I managed to disambiguate the word up in the thread. In the paper, I’m pretty clear which sense I mean: a successful explanation is on that produces understanding. And I promise that’s the only sense of “successful” that figures into my main argument. I am not equivocating. Please read the main argument with only that sense of “successful.” That’s how I intend it to be read.
You then say, “According to the conditional there’s simply no such thing as a successful theory.”
I take it that “the conditional” is premise 2. No, premise 2 does not entail that there’s no such thing as a successful explanation (or theory). It would if you thought every proposed explanation crucially features some element that calls out for explanation but lacks one. But why think that? I don’t think that’s true.
You then say, “As the paper puts it, ‘the promise of an explanation of the stationary Earth has merely been deferred, but ultimately not fulfilled’.”
It seems like you’re supplying this quote in an attempt to prove that I myself think there’s no such thing as a successful explanation. But, no, if you read the context there, you’ll see that I’m giving you an example of premise 2, an instance of it. A case of a proposed example that crucially features an element that calls out for explanation but lacks one. True, I do think THAT explanation fails.
But by no means does it follow that I think EVERY explanation fails. I think very many explanations as successful, in exactly the sense of “successful” that I give in the paper. Unfortunately, it looks like Naturalists have to disagree, and have to say that all scientific explanations are unsuccessful, that they fail to produce understanding.
If Naturalism is true, we have to admit that we live in a world impervious to scientific explanation. And that would be a real disappointment, to say the least haha. One more reason not to be a Naturalist, I’d say.
Oof. My last comment to “Frans” has several typos. My bad. I’m typing in a bit of haste (have to run soon). I’ll do a better job of checking for typos in the future!
And yet again, “It would be nice if we found out where life began” doesn’t make evolution not a successful theory. Precisely because what we view as a successful theory, a successful explanation, is predictive success within its domain. In practice, working scientists do not view it as a knock on the theory of evolution that it doesn’t explain abiogenesis. That’s a different question. Which takes us back to the core point Richard and everyone else keeps making. What I call a motte-and-bailey fallacy at the heart of your argument is this. Science’s predictive success continues unabated despite essentially never having theories that have unlimited explanatory scope or total explanations for any elements. We always know there’s more, which is one of the reasons science is provisional. And yet science continues successfully.
If your argument were interesting and not trivial or formsfalse, you would develop a claim that a Godly universe would be the only type of universe where science would encounter even that kind of success, where science would be able to make technological progress with incomplete hypotheses because the underlying laws of physics would still be reliable enough to do so. I don’t think this argument is remotely defensible, and it’s bog-standard garbage-tier Christian apologetics, but it’s at least defensible.
There may not be an explanation for everything. Certainly not explanations available to our cognitive capacity. We may end up having mysteries in the universe. Exactly like you yourself attested to. (Just like in a theistic universe where God would be a massive mystery, forever and always). In fact, contrary to your rhetorical position here, it is only naturalism and some forms of non-theistic supernaturalism where there is even the potential for everything to reduce to comprehensible, simple forms and principles.
Science does not proceed by postulating that a theory with unlimited explanatory scope and power must be available. It only proceeds, and always has, under assumptions that we have the right tools to find such a theory if it exists. It is pretty much impossible to read Hume and not come away with that conclusion.
The only thing this tells us about the cosmos is that we happen to not live in a cosmos where non-trivial truths are easily, instantly, intuitively available. This is true of essentially all remotely plausible naturalist universes but actually nowhere near all theistic universes.
Yet science continues to make us cars and cell phones even though we have neither found the bottom turtle (or dragon) nor even concluded there is a bottom turtle necessarily or that the thing at the bottom is a turtle.
Which is the motte-and-bailey fallacy your paper covers up.
You just never explain why contingent, incomplete explanations about a certain number of elements are not useful. (Or suspiciously useful).
The fact that the arc of scientific history moved away from grand theories that explained everything, like the Newtonian promise of a clockwork world, to more contingent, provisional theories seems to be actually precisely what a naturalist metaphysics would predict.
I was really hoping that your paper wasn’t just yet another way of complaining about Hume and the world of uncertainty we live in. But it seems like it is, by your discussion here.
Okay, then your paper failed because I do not think you have laid out even one single theory that is considered unambiguously successful that has no major unexplained elements. If abiogenesis calls out for explanation in the theory of evolution, one of the clearest cases imaginable where “This theory explains a huge swath of data from point X that we stipulate to because that’s where the causal mechanisms operate in” isn’t successful, then there is nothing.
But even if science is awash with such explanations, so what? The point is that it has never needed them. By your reasoning, science couldn’t even get started, because no theory would be viewed as even successful enough to continue investigating. We’d never have moved on even from Aristotle.
“Fred B.C.”: I just wanted to quickly write to let you know that I’m not ignoring your replies. I have read them. Unfortunately, the prospect of sifting through your replies to help you untangle your subtle misunderstandings just doesn’t sound super appealing to me at the present time. Especially since I’d be largely repeating what I’ve already said to you. I hope you’ll understand.
If you really think you’re onto something here, it should be no trouble to write it up and publish it in a journal. I encourage you to try. Whether you do or not, I wish you all the best in your continued study of Philosophy.
Tomas:
Well, I am glad that you think the misunderstandings are subtle. However, I suspect that the precise fact that any miscommunication may be on subtle points actually does not bode well for the argument. I similarly suspect you may be misunderstanding, and while I actually do think I can try a different angle for discussion depending on the objections, it may indeed end up being a repeat. Because I really do think that, until you use anything like the way that scientists talk (“explanation” is just too broad, vague and undefined a term), your argument is going to be endlessly vulnerable to equivocation.
I also suspect that you will find that any response from Richard is going to center on a similar constellation of concerns and perspectives. Certainly Molkien and I essentially made the same point.
I do thank you for being polite and patient. Richard attracts really top-notch people who are ready to defend their ideas to his comments section, and you have largely handled yourself with a lot of class against a tenor of critique that I can understand one viewing as a bit harsh. While I didn’t get too much out of your paper because I can’t see far past what I think is a fundamental difference in worldview and methodology, I did appreciate that your paper took naturalistic worldviews seriously and really tried to think about naturalistic metaphysics in a way that was not dismissive. That puts you into a league of your own.
I just think that much of this kind of discussion is going to be fairly fruitless until someone does a really detailed deductive analysis of all the outcomes of an array of possible metaphysics, and I don’t think I’m suited for anything besides just noting that problem exists!
Dr. Carrier,
You suggest that I “learn the use of rhetorical quotation marks.” I encourage your readers to google that, and notice the dearth of results, none of which seem particularly relevant to your (mis)use of quotation marks.
I would also recommend that your readers check out this advice on quotation marks from Purdue: https://owl.purdue.edu/owl/general_writing/punctuation/quotation_marks/index.html
Specifically this part: “The primary function of quotation marks is to set off and represent exact language (either spoken or written) that has come from somebody else. The quotation mark is also used to designate speech acts in fiction and sometimes poetry.”
I don’t think we’re engaged in fiction or poetry here, so I’d recommend that we put quotation marks around only exact language from someone else.
……………………………….
We had a little disagreement over what counts as the consequent of premise 2. Here’s that premise again:
-2- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
You said, “the consequent is ‘only then is an explanation successful’.”
There’s no easy way to say this, but, no, that’s not the consequent. You are misunderstanding conditionals. Philosophers treat these two constructions as logically equivalent:
If A, then B.
A only if B.
For example:
If Buckley is a dog, then Buckley is a mammal.
Buckley is a dog only if Buckley is a mammal.
Hopefully you can hear the equivalence there. This is why “if and only if” expresses a biconditional. “A if and only if B” means: A, if B. And A only if B. In other words: If B, then A. And A only if B. That is: if B, then A, and if A, then B.
Anyway, we tend to use the “If A, then B” construction to emphasize the sufficiency of the sufficient condition, A. We tend to use the “A only if B” construction to emphasize the necessity of the necessary condition, B.
In these conditionals, proposition A is called the antecedent, and proposition B is called the consequent. See Definition 1.2.1 here: https://www.siue.edu/~jloreau/courses/math-223/notes/sec-conditionals-and-biconditionals.html
So, as you can now see, the consequent of my premise 2 is as I said: “it crucially involves no element that calls out for explanation but lacks one.” Where “it” refers to the explanation mentioned in the antecedent.
So, not a big deal, but yes, you were mistaken about this fundmental feature of conditionals.
………………………………..
You also say, “And please give me an example of an actual successful scientific explanation that does not lack any explanation it calls out for. I am most curious to see such a marvel.”
I think you meant to ask me to give you an example of a successful scientific explanation that does not crucially feature any element that calls out for explanation but lacks one. Well, if Naturalism is true, I think there are no such explanations! That’s the point of my paper. That’s the problem with Naturalism that I mean to point out.
If, on the other hand, we’re willing to countenance theistic explanations of natural regularities, then I think there will be many such examples of successful scientific explanations. All the ones you’d normally think of, I’d say. This is what my paper is about, it’s a pretty interesting result, and your readers can find the paper here: https://philpapers.org/archive/BOGINI.pdf
For the benefit of my readers, on the various use of quotation marks and that it’s obvious how I am using them, you’ve already been called out on that. See comment and comment. I suggest you drop that. It is unproductive wingeing at this point.
Your words, my emphasis. That is an “only if” conditional. That means the condition can only be met on that consequent. This is therefore saying there is no other kind of success than here described.
As you seem to admit:
This is exactly what I am saying.
How you are confusing me as saying anything else baffles me.
You said, exactly as I said, that an explanation is successful only if it meets that condition. That statement is false. No scientific success has ever met that condition.
-:-
You’re starting to look dense here. You said there can be no success that doesn’t satisfy that consequent. Then you said science shows a lot of success. But science has never shown any success meeting that consequent. The consequent is therefore false. You have neither described any actual success, nor have you produced any evidence of any such success in science. This is an equivocation fallacy. Exactly as I explain.
Once again, you appear to be (as I stated my suspicion before, both here in comments and in my article) wanting to have said something else in your paper: That science can only have been as successful as it has if it assumes some ultimate (as yet unconfirmed) grounding of all explanation. That isn’t in your paper. But you seem now to be acting like that is what you meant to argue all along, and I point out in my article I did indeed suspect that is what you were trying to argue but failed to correctly formulate it and instead built a convoluted and fallacious apparatus in its place.
If you want to move the goal posts thus, and completely rewrite your entire paper, such that your argument is instead that science’s actual success has required assuming there is some ontological ground (even if we do not yet know what that is), and then argue that only nonnatural grounds “work,” then your paper as-is contains no defense of that.
Instead, all you do is analyze possible grounds as infinite regress, final circular fact, brute-fact, and necessary fact, and complain about them all. But nonnaturalism suffers the same defect: it can only be one of those same four things. You present no evidence it can’t be a natural fact, or that any nonnatural fact you presented escapes being one of these four things.
So yes, rewrite your paper to have a completely different argument than you actually presented.
You still don’t get to your conclusion. Other than by fallacies (such as Aristotle’s Argument from Ignorance, which you spend a bizarre amount of time futilely defending).
And this is already explained in my article. I already anticipated this goal-post move then and already addressed it.
I suggest you re-read my article and actually think about what it says this time.
P.S. I should also add, since you seem to miss things like this a lot:
In the new formulation of Premise 2 (“That science can only have been as successful as it has if it assumes some ultimate (as yet unconfirmed) grounding of all explanation” or any equivalent statement), you have no evidence for that Premise (science has never assumed this, yet is successful; and science doesn’t have to assume this, to have been successful).
Moreover, your analysis is exhaustive of all possible grounding scenarios (infinitism, circularism, brute-factism, and necessary-factism), and thus renders Premise 2 moot. Science can be successful without assuming anything about this—as in fact it has been. It is, instead, just logically necessarily the case that naturalist explanations (as all explanations) must be grounded in one of those four ways (even if we haven’t scientifically discovered that ground yet), but so must nonnaturalism, so this being the case is no argument for or against either. It’s a problem equally faced by both.
You needed to demonstrate by some nonfallacious logic or survey of evidence that none of those grounding scenarios can work for naturalism but at least one can work for supernaturalism. But your paper contains no sound syllogistic or evidence-based demonstrations of either.
Exactly as my article takes pains to explain. Hence you might want to actually read it, and not rage-skim it.
Tomas:
I did! And I found… https://en.wikipedia.org/wiki/Scare_quotes
On the first page of results.
More importantly, Richard was, you know, talking to you. I have to imagine you got the point. You’re free to dislike the practice, but you seem to be acting as if it’s not what it transparently is.
“Fred B.C.,” I’m familiar with scare quotes (as you can see), and I found that Google result as well. I class that among the results of questionable relevance, since it sure doesn’t seem like that’s what Dr. Carrier was up to. I appreciate your loyalty to your friend, but you really needn’t defend what is pretty clearly simply misquotation. Friends are dear, but truth is dearer.
Really, though, the abuse of quotation marks is pretty low on my list of concerns. I’ve said more than I care to about it, to be honest.
Tomas:
The point is that he wasn’t quoting you, any more than someone saying that Trump is a “leader” is necessarily quoting a specific Trump speech. He’s acting as if you said something. “It’s as if he had said, ‘…'” It’s really not at all confusing.
To just quote from the Wikipedia…
That example of “groupies”? Or the Atlantic’s use of “identity politics” or “rape culture”? Those aren’t quotes from specific people. The only difference is that Richard is rhetorically attributing his phrasing of your idea to you in specific, in the same piece that he is also quoting you. Not literally. Rhetorically. It’s just not that uncommon a thing to do.
Can that be used to strawman or shadowbox? Sure. But so can the exact same misattribution without rhetorical quotation. “Tomas is basically arguing that, because Aristotle said something (specifically something stupid), science can’t work” would be exactly as critical of your position and yet wouldn’t use a quote. You’d object to that too, I imagine. So just say that Richard is using a strawman and have that discussion. Complaining about the specific rhetorical use of quotes really seems beneath you, and, yeah, makes you seem like you’re in the ivory tower.
Richard isn’t my friend. I don’t personally know him. We haven’t chatted. And I disagree with him on many issues. If you actually check out my comments here, you’ll see me taking different tacks from him all the time. You’ll also see me attempt to steelman others as best as I can. You’re free to not believe me, dude, but seriously, this isn’t a weird or uncommon thing Richard is doing . He’s just talking like a human being. It’s like him putting on a falsetto voice and saying “Dur dur dur, I’m Tomas, I think Aristotle is the tits”. It’s not misquotation because no one would actually think it’s what you said. He isn’t trying to emulate your writing style, he’s making it very reductive, <i.>on purpose. Like, you’re free to not like it, you’re free to tone police it, but it’s not misquotation.
But hey, prove me wrong. Find someone who actually reads the article and thinks that these are things you said. And I would bet you dollars to donuts that, if you did, Richard would find a way of clarifying.
Dr. Carrier,
You said, “For the benefit of my readers, on the various use of quotation marks and that it’s obvious how I am using them, you’ve already been called out on that.” I have been called out for mentioning your misuse of quotation marks? I feel like I’m in bizarro land haha. Look, I’m happy to drop the misquotation topic, especially since I notice you’re being much more careful about your use of quotation marks now. I’m grateful for that.
…………………………………………
You quoted my second premise, and you added some emphasis:
-2- Any explanation can be successful only if it crucially involves no element that calls out for explanation but lacks one.
You say a few things about this:
“Your words, my emphasis.” That’s right! Thank you for your careful use of quotation marks.
“That is an ‘only if’ conditional.” Right again. Nice work.
“That means the condition can only be met on that consequent.” Erm, this is a little awkwardly phrased. But I think I understand what you’re trying to say. By “the condition” you mean the antecedent, the sufficient condition. Yes. That condition can be met only if the consequent is true. Yep. That’s precisely what the conditional says.
“This is therefore saying there is no other kind of success than here described.” No, that doesn’t sound quite right. There may be other kinds of success (Marital success? Professional success?). But given the sort of explanatory success I’m interested in—production of understanding—yes, I’m claiming that premise 2 states a necessary condition for that kind of success.
You then quote me saying that we tend to use the “A only if B” construction to emphasize the necessity of the necessary condition, B. And then you say, “This is exactly what I am saying.” Well, I’m willing to grant that this is what you were trying to say. Sure.
But you continue, “How you are confusing me as saying anything else baffles me.” Oh, it’s because you had said, of premise 2, “the consequent is ‘only then is an explanation successful’.” And that’s false. That simply is not the consequent of premise 2. And I wouldn’t say I’m confusing what you said. It was very clear what you said, and it was very clearly false. It’s not that big of a deal, and if this conversation weren’t happening in public I imagine that you’d be willing to correct that minor error.
Anyway, about premise 2, you say: “That statement is false. No scientific success has ever met that condition.” As I said above, we disagree about that. I agree that if Naturalism is true, then no scientific explanation has ever met the necessary condition laid out in my second premise. And, so, if Naturalism is true, no scientific explanation has ever been successful. Yes. That’s basically the point of the paper. But I don’t think that follows if Naturalism is false. And, as you’ve probably gathered, I think Naturalism is false.
You continued: “You said there can be no success that doesn’t satisfy that consequent. Then you said science shows a lot of success. But science has never shown any success meeting that consequent. The consequent is therefore false.”
As I’ve said a few times now, I think scientific explanations can meet that necessary condition for success, if Naturalism is false. And, no, you’re incorrect to say that the consequent would be false, even if you were right that no scientific explanation had ever met that alleged necessary condition on success. Again, you’re misunderstanding how conditionals work. What you’re probably trying to say is that, if the antecedent were true while the consequent were false, then the conditional itself would be false. Yes, that would be correct. Not the consequent, as you say, but the conditional itself. The second premise. But, as I’ve said, I don’t think you’re right that there’s any case in which antecedent is true while the consequent is false. As far as I can tell, only Molkien has come within a bull’s roar of providing a case like that. But I don’t think the case works, and I responded to it above.
You then say, “You have neither described any actual success, nor have you produced any evidence of any such success in science. This is an equivocation fallacy. Exactly as I explain.”
Look, to show that I have committed an equivocation fallacy, you’d have to show that the truth of one of my premises requires one reading of a word or phrase, but the truth of another premise requires a different reading of that word or phrase, and yet if we read one premise in one way and the other premise the other way the argument would not be valid, and therefore the argument cannot be sound. You’d want to show that my argument features either one false premise or an invalid inference. That’s how you’d pin an equivocation fallacy on me. You can’t simply insist that I’m using a word or phrase in two different ways, especially when I insist that I’m not. I wrote the argument, and I’m telling you what I had in mind: a univocal use of “success” meaning production of understanding. I say this pretty clearly in the paper, to forestall this very misunderstanding. If you think that makes one of my premises false, then you should simply focus on that criticism.
You know what might help? I’ll just rewrite the argument for you, swapping in what I mean by “be successful” for every instance of that phrase:
-1- Any scientific explanation can [produce understanding of the phenomenon it’s meant to explain] only if it crucially involves a natural regularity.
-2- Any explanation can [produce understanding of the phenomenon it’s meant to explain] only if it crucially involves no element that calls out for explanation but lacks one.
-3- So, a scientific explanation can [produce understanding of the phenomenon it’s meant to explain] only if it crucially involves a natural regularity, and this regularity does not call out for explanation while lacking one.
-4- If Naturalism is true, then every natural regularity calls out for explanation but lacks one.
-5- So, if Naturalism is true, then no scientific explanation can [produce understanding of the phenomenon it’s meant to explain].
Now, I hope, you’ll agree that there’s no fallacy of equivocation going on here with the word “successful.” I do go on to say that, insofar as one thinks any scientific explanation has [produced understanding of the phenomenon it’s meant to explain], then one has a reason to reject Naturalism. And I do think at least some scientific explanations have done that. So, I think this argument gives me reason to reject Naturalism.
Maybe you disagree. Maybe you think no scientific explanation has ever [produced understanding of the phenomenon it’s meant to explain]. OK. You can own that, if you’d like. But it’s surprising to hear a Naturalist so pessimistic about science, and I myself consider that a reductio of your view.
……………………………………….
You then say, “Once again, you appear to be (as I stated my suspicion before, both here in comments and in my article) wanting to have said something else in your paper: That science can only have been as successful as it has if it assumes some ultimate (as yet unconfirmed) grounding of all explanation.”
Hey, thanks for resisting the urge to put that proposed revision in quotation marks and attributing it to me haha. Nice. But no thank you to the proposed revision. That’s definitely not what I want to say. I’d like to keep my main argument exactly as I wrote it.
You say, “If you want to move the goal posts thus, and completely rewrite your entire paper…”
Haha, no thank you again to both suggestions. I’m happy with the argument as I wrote it, and I encourage you to think of objections to that argument, the one that I wrote.
I think we’re probably at an impasse. We’ve both shared our thoughts, multiple times. We’re repeating ourselves. I didn’t really expect to change anyone’s mind down here in the comments. I’ve had my say and I’m content to leave things there. If you still think you’ve uncovered a serious objection, I sincerely encourage you to write it up and send it to a journal. Thanks again for thinking about my paper, and for letting me reply to your comments. Take care.
I find that in general as apologists get cornered, they generate ever longer word-walls attempting to rescue their position. The cascading length-increases of your comments are showing that.
It’s clear now that a straightforward, non-convoluted discussion with you is impossible. This forces me to atomize the conversation and walk you by hand one point at a time. So, I am setting aside every rabbit hole you have dug for yourself and choosing only one to continue with for now:
You keep contradicting yourself by admitting I’m right about what an “only if” conditional says, then claiming my description of it as an only if conditional is wrong. I fail to see any coherent answer here. You simply aren’t making any sense at this point.
Instead…
No, we don’t. You have never presented any examples of this kind of success in science. So even you have tacitly admitted no such success standard exists in science nor has ever been applied to it.
What you are “disagreeing” with is yourself:
As I keep explaining (a point you keep dodging), you wrote “can be” instead of “could have been as” in Premise 2 in your paper.
Now you want to rewrite your paper to say something else, something (it sounds) more like the latter. As I suspected even in my critique.
By not admitting you messed up the wording of the premise, and thus created an equivocation fallacy when you tried introducing evidence for Premise 2 being true, you are simply spinning wheels here.
You need to admit the actual evidence of science’s success does not evince Premise 2, not as actually worded, nor as you (it seems) want to reword it now.
To say that science, for example, “could only have been as successful as it has been if X” cannot be proved by citing the actual success of science. Because that evidence (the actual evidence) is equally expected on the truth or falsity of that Premise.
That evidence therefore does not establish Premise 2, either in your paper’s presented sense, or this new sense you want to defend (which you still have not articulated clearly but are avoiding articulating clearly, requiring me to repeatedly interpret your new intended meaning, but we’re getting you closer to admitting it…).
So you do want to redefine “success” contrary to how science defines it, just as my critique said. Excellent. Now you are starting to admit my critique is correct.
This has exactly the consequences my critique points out…
Now, with this new definition of science, Tomas, will you admit you cannot cite the actual success of science as evidence for this version of Premise 2?
Because that would be a circular argument in defense of the Premise, not a valid empirical argument. You need to establish the Premise is true, not just declare it true. So you need evidence that it is true.
And if you want to reword that premise in some other way than “Science could only have been as successful as it has been if X” then try that. But I promise you, you’ll never get a premise that both (a) is confirmed true by any evidence and (b) gets to the conclusion you want.
Just try and see.
Tomas:
Yes, I don’t think you’re going to be convincing. Because you talk about science and then cite only a scant few examples… and then when you provide examples, we point out how they don’t work.
Your newest post to Richard I think makes it much worse.
Because of this.
Tomas. Whether naturalism is true or false, scientific explanations at present do not match your criteria. Because they’re all incomplete. Just like all supernaturalist explanations, because all supernaturalist explanations are at this point naturalist explanations with inherently unexplainable cruft tacked on.
You cannot cite hypothetical explanations WE don’t have. I emphasize “we” there because you have no explanations either. “God did it”, by your own criteria of actual explanatory value, is not an explanation either. Under any supernaturalism, we as humans are exceedingly unlikely to understand anything like causal mechanisms. Anything like an actual explanation of why we say “God bless you” when we sneeze. “It’s magic” is barely any better than “That’s the custom” as far as non-answers that effectively repeat the question.
All scientific success that has occurred thus far (and you rather conspicuously refuse to clarify if you actually think science has been successful on its own terms) has been without an explanation with no important elements left unexplained. You yourself provided the example of evolution. When Darwin suggested natural selection, it was already promising. When everyone began confirming the power of his model, it was already successful. Not only was this before even the most cursory abiogenesis theories, this was even before the incorporation of genetics, DNA and RNA, understanding of the mechanisms of mutation, genetic drift, epigenetics, symbiosis, the behavior of retroviruses, pleiotropy, and all the countless things we have synthesized into the present neo-Darwinian (or really neo-neo-Darwinian) model. Anyone in the late 19th and early 20th centuries had massive unanswered questions as to why life had evolved in the specific ways it had. And yet evolution was still successful as a theory. By science’s own standards. Which you have provided utterly no reason to reject, or describe as unreasonable.
Scientists have not tried to meet your criterion in your second premise. Because your criteria are nonsense. And you steadfastly refuse to respond to either me or Richard pointing out how supernaturalism by your own Feserian standards also leaves critical elements unexplained. (Unless you can explain God. Fully and totally. Since the totality would be necessary to have a complete theory that would make perfect predictions).
Of course, you’re also just repeating an unevidenced gainsaying of the fact that, again, as we’ve been discussing here, there in fact can be quite robust explanations under naturalism that can only arguably leave metaphysical questions of the type “But why is it like that?” (which may in fact just not be meaningful questions because “why” may cease to be meaningful at the level of fundamental ontology), and indeed thereby behave much better than supernaturalism. Because only naturalism can have a universe that breaks down to irreducibly simple components, whereas supernaturalism must mandate irreducibly mental components which requires tuned complex components to any supernaturalist theory. And so your argument is just calling the trial before the jury has even started the first hour of deliberation, and is woefully (and fatally) premature.
But all of that is minor compared to this staggering new manifestation of the fact that your argument is an argument from ignorance. Whether a rooted argument from God is possible, you don’t have it, and no one ever needed it for success from science.
Which is what Richard has repeatedly said is clearly what you are actually trying to argue, which this exposes quite transparently, and yet you want to keep your original argument intact for some reason.
SMH, sir. SMH.
I love how his equivocation fallacy fatally ruins the particular argument at play. As Carroll pointed out to Craig, we just don’t think about necessary causes or Aristotelian causation demands anymore. And different scientific disciplines actually vary quite a bit (as you point out discussing anthropology and sociology) as to the ways they think about causes and explanations. In any social science, one just has to come to accept that, for the foreseeable future, we’re not going to have anything like the level of robust theory we do in astronomy or physics. We’re examining systems that are too multicausal, too complicated and too interconnected. Thinking about causes in those kinds of systems becomes really complicated, and imposing Aristotelian causation onto any of it is like trying to demand that your car literally conjure horses to pull it.
So what we call a “successful” theory in science actually doesn’t match what he says. He’s not even talking to scientists. His motte is nonsense and his bailey is barely better. In fact, scientists are quite happy to work within models they know can’t be perfectly accurate if they’re going to be more than good enough for the accuracy required in a given situation. We constantly make models, and if those models work for our tasks, job done.
What I find funny from a meta perspective is that this again is Bayesian disproof of God, and in fact you don’t even need to put evidence back in because Bogardus is providing all of the evidence. The fact that we actually don’t find ultimate explanations everywhere, that the actual scientists on the ground do use standards of utility like prediction and falsifiability precisely because accessing underlying reality is so fraught, and that the universe is not intuitively explainable, is actually pretty bizarre under a theistic model, by Bogardus’s own admission . It is precisely the fact that we live in a world that isn’t like that and that science had to emerge with that degree of certainty that is illustrative of the converse of Bogardus’ argument.
(This actually relates to an argument I’m working on that I’m calling the Argument from Moral Difficulty, that the fact that the universe makes moral decisions complicated and painful and ambiguous is also not expected under most God models).
I also find it funny that his standard as stated would actually make science definitionally impossible in some contexts. If we lived in a world with infinite past regress, for example, there could be hard reasons against finding an ultimate explanation. (And, as you allude to, Christians themselves would frequently have to admit that God is a mystery and so they don’t have an ultimate explanation either ) . Which is a sign of the hypocrisy of the argument.
I think there is definitely an argument there. It would fit alongside the Argument from Meagre Moral Fruits (on which see my discussion in re Brieley). That is about the unexpected failure of Christianity to do any actual good; yours would be about the unexpected failure of the universe to do any actual good.
Yes, and not just do actual good, but do unambiguous actual good .
Theists constantly talk about how God can allow a world with evil because it lets us test ourselves as moral beings. We can grow, God can prove our character, etc. All of that is of course nonsense moral equivocationism, but even if it’s true, what that would entail would be a world where the test is unambiguous.
In the same vein, Christians also routinely describe objective morality as being not only extant but also clear . The world is Manichaean. There is dark and light, and nothing in between.
All of that actually produces predictions. Even if we accepted a God who deliberately made a universe containing evil so that the drama of evil being defeated could play out (or. more sensibly, that a finitely powerful good entity coexists with a finitely powerful evil entity, a la Zoroastrianism), that would still imply a world where everyone could see good and evil. It’d be as metaphysically obvious as the distinction between heaven and hell, or moon and sun.
And insofar as God would want to make tests, God would want those tests to be controlled variable tests. What’s the point of putting us into tests where we always have at least some excuse based on finite willpower, knowledge, empathy, moral calculus, and impulse control?
Whereas under naturalism what we’d expect is that, if morality is objective, it is difficult to ascertain . Complex meta-ethics would need to be done. The world would not necessarily be intuitively understandable morally any more than it would be intuitively understandable physically. There would be constant situations where moral actors would have finite information, unclear tradeoffs, and moral dilemmas. We would live in a world where people would not be sure if they had done the right thing even after extensive work and thought. We would live in a world where a psychologist would wonder precisely how serious a client was about threatening an immediate crime.
What I like about this argument is that the Christians have no defense against it, even more than usual. First of all, their own rhetoric proves the point. Christians at this point tend to be smart enough to know about the problem of evil, so they tend to preemptively seed their special pleading into discussions of evil. But when it comes to moral guidance and clarity? We are told that all one has to do is consult God or read the Bible. I know for a fact that I could assemble a wide array of theists making clearly explicitly and implicitly that they think divine guidance is readily available.
Second, we all know it is false to Cartesian certitude. Essentially as close to everyone alive as makes no odds has clearly had moments where they weren’t sure if they were doing the right thing (even if they cared or not). It is essentially universal to have to navigate moral dilemmas and anxieties.
Which means that the evidence against the theistic prediction is as good as it can possibly be.
And the efforts to gerrymander around that in terms of designing a God are outright harmful. Christians benefit a lot from the fact that the idea of defeating evil in a climactic struggle is at least romantic. But muddling through ethics meetings because the world is complicated in boring and mundane ways? Not so much.
If you are looking for a reliable standard for determining good philosophical practice, the least you could do is make Bogardus’ comment public. Why should your readers not see what he has to say in response?
I’m confused. His comment is public. It might even have been public before you commented.
-:-
I should also caution you against unrealistic expectations. You evidently don’t know how things work. Because whether his comment was up or not, you assumed it’s not being so would have to be some sort of nefarious conspiracy. In reality, delays in mod queues are due to workflow.
I can’t whitelist people without personally identifying info. So I could not clear his access to instant posting until after he submitted one comment (with the correct data to mark it to my whitelist). But once that happens, he can post instantly without going into the queue.
Since I don’t check the queue for sometimes days at a time (because I have a life), first comments will always be on a delay. I can only clear them when I see them. I will not telepathically know when comments have been submitted.
But per my Comments Policy, anyone I write about can claim unmoderated commenting privileges (provided they follow all the other rules thereafter, but I have never had that problem and don’t expect it here). Hence as soon as I did see Bogardus’s comment, I cleared it and whitelisted his access from there on.
This is how it happens on this website for everyone, all the time.