People who hold the view that “there is a non-scientific source of knowledge about the natural world, such as divine revelation or the historical teachings of a church, that trumps all other claims to knowledge,” are a menace to science. That’s the claim made by mathematician Jason Rosenhouse, in his latest post over at his Evolution Blog. Science, avers Rosenhouse, is not just a collection of facts; it’s “an attitude, one that says that all theories must be tested against facts and that evidence must be followed wherever it leads.” In an earlier 2009 post, Rosenhouse criticizes the claim that “science is not the only way of knowing,” and forthrightly declares: “The ways of knowing that are unique to religion, namely revelation and the words of holy texts, have today been utterly discredited.” Is he right?
Dr. Rosenhouse is an American author and associate professor of mathematics at James Madison University, Harrisonburg, Virginia. He has been writing about creationists for some years now, and is the author of the book, Among the Creationists: Dispatches from the Anti-Evolutionist Front Line (Oxford University Press, 2012).
In this post, which I shall try to keep this post as short as possible, I’d like to explain what I think is wrong with Dr. Rosenhouse’s whole approach to epistemology.
1. Let me begin by saying that I intend to play fair. For example, it would be very easy for me to make fun of Dr. Rosenhouse’s claim that science is the only way of knowing with the standard retort: “How do you know that?” But Rosenhouse could counter this cheap jibe by rephrasing his epistemological claim as an imperative: “Don’t trust claims that there are other, non-scientific ways of knowing!” There’s nothing self-refuting about telling people that.
2. The first thing I want to say in response to Dr. Rosenhouse is that science is not a self-supporting enterprise: there are certain background assumptions that it presupposes. (I’ll list them below.) The next thing I’d like to do is spell out what that entails:
(a) if science is not a self-supporting enterprise, then science can never hope to explain everything, since science is necessarily incapable of explaining what science itself presupposes;
(b) if science is not a self-supporting enterprise, then science cannot possibly be the only way of knowing, since the way in which we know the background assumptions upon which science rests is necessarily different from the way in which we know facts which we discover by applying the scientific method itself: the former mode of knowledge is better described as meta-scientific.
3. The following is a short (but not exhaustive) list of background assumptions about the world, which the scientific method presupposes. Science would be impossible as an enterprise, if the vast majority of scientists did not hold these assumptions:
(a) There exists an external world, which is independent of our human minds: it’s real, regardless of whether we believe in it or not;
(b) Objects in the external world have certain identifying characteristics called dispositions, which scientists are able to investigate;
(c) Objects in the external world behave in accordance with certain mathematical regularities, which we call the laws of Nature, and which tell us how those objects ought to behave;
(d) Scientific induction is reliable: scientists can safely assume that the laws of Nature hold true at all times and places;
(e) Solipsism is false: there exist other embodied agents, with minds of their own;
(f) Communication is possible: scientists are capable of talking to one another, and sharing their observations, as well as their thoughts (or interpretations) relating to those observations;
(g) The senses are reliable, under normal conditions, within their proper domain, which means that scientists are capable of making measurements on an everyday basis;
(h) There exist standard conditions, under which ordinary people (including scientists) are routinely capable of thinking logically, making rational discourse possible;
(i) Scientists are morally responsible for their own actions – in particular, they are responsible for their decision to tell the truth about what they have observed, or to lie about it; and
(j) Scientists should not lie under any circumstances, when doing science.
Science would also collapse as an enterprise, if these background assumptions were not objectively true.
The inclusion of an ethical norm (statement (j)) in my list of background assumptions might raise eyebrows in some quarters. Physicist Frank Tipler argues for its necessity to the scientific endeavor, as follows:
…[A] moment’s reflection will show that the value/fact distinction is difficult to maintain. Consider the hardest of the hard sciences, physics. The real reason that we consider physics to be a hard science and the profession of politics to be a soft science (if we consider it to be a science at all) is that we trust the experimental data produced by the physicists. That is, we assume that physicists have adopted the moral precept Thou shalt not fake data. If this moral precept were not adopted in the sciences, if physicists, for example, were known to fake their results whenever their politics required it, there would be no hard sciences. So clearly, all positive science necessarily is based on normative principles.
(“The Value/Fact Distinction: Coase’s Theorem Unifies Normative and Positive Economics”, January 15, 2007, p. 4.)
Note: Although I referred to agents and their thoughts and obligations, objects and their dispositions, and the laws that objects conform to in their behavior, I took great care not to include any purely metaphysical statements in my list of background assumptions above. All entities referred to in the above list are publicly observable.
4. In addition to the above, there exists a class of statements known as synthetic a priori truths, whose truth we can know without doing any science at all. Some examples:
(a) while causes which generate effects may precede those effects, or be simultaneous with those effects (e.g. a head lying on a pillow, in which it produces an indentation), it is impossible for such causes to come after their effects;
(b) space can have a positive integral number of dimensions (e.g. 1, 2, 3, …), but it cannot have a negative number of dimensions, a fractional number of dimensions, or an imaginary number of dimensions;
(c) the flow of time is objectively real, which means that scientists’ decisions, which are made in time, really do matter in the scheme of things; and
(d) the same object cannot be red all over and green all over, at the same time.
I’m not going to offer a general account of how we know these things without doing any experiments. All I will say is that if you claim to have knowledge of any of these truths, then you have committed yourself to an extra-scientific mode of knowledge.
5. In addition, the scientific enterprise is governed by certain rationality norms, which tell scientists what they should be investigating. Failure to follow these norms is tantamount to committing the sin of intellectual laziness, and is therefore poor science. Some examples of these norms are as follows:
(a) Contingency warrants a scientific explanation: whenever a scientist identifies a non-essential property of some object in the natural world (e.g. an arbitrary numerical value of a constant of Nature), he/she should look for an external explanation of why the object has that property;
(b) Complexity of function warrants a scientific explanation: whenever a scientist identifies a natural object performing a function involving two or more steps, he/she should look for an external explanation of how the object is capable of performing that function;
(c) Complexity of parts warrants a scientific explanation: whenever a scientist identifies a natural object composed of two or more parts, he/she should look for an external explanation of what holds the object together;
(d) Coming-to-be and ceasing-to-be warrant a scientific explanation: whenever a scientist identifies a natural object coming into being, he/she should look for an external cause of the object’s coming-to-be; and the same holds true for a natural object which a scientist observes when it is ceasing to be;
(e) More generally, any question about the natural world which is not obviously nonsensical should be regarded as falling under the purview of science, and the systematic attempt to answer this question, however bizarre it may sound, should be regarded as a legitimate part of the scientific endeavor.
6. In addition to the above, testimonial knowledge (or knowledge based on a reliable source) is a legitimate (non-scientific) way of knowing something. If your geology professor tells you that the age of the Earth is 4.54 billion years, give or take 1%, then you are perfectly entitled to take your professor’s word for it, and to claim that you know that the Earth is that old, because your professor told you so. If your friend, who is widely traveled, tells you that the roads in the center of Sofia, Bulgaria, are covered in yellow brick (as indeed they are), then you are entitled to claim that you know this for a fact, based on what your friend told you. And if a child’s parents, who are a lot older and wiser than she is, tell her to stay away from a particular person because he is a bad character, then the child is epistemically warranted in assuming the same.
7. Testability is a vital ingredient of scientific knowledge, but it is not sufficient to render a claim to knowledge scientific. A person might check the reliability of one of her sources by testing that source; but that does not make her knowledge scientific. St. John tells us to “test the spirits to see whether they are from God” (1 John 4:1), but this does not refer to scientific knowledge.
8. At the same time, not all knowledge needs to be testable, in order to count as genuine knowledge. For instance, sometimes you can just see, from the expression on someone’s face, that they are telling you the truth; their sincerity is impossible to doubt. If such a person swears to you that they have never harmed or betrayed you, then you have every right to say that you know they are telling the truth.
9. The above-listed sources of knowledge, coupled with the rationality norms listed for science, are all that is needed to provide a warrant for religious claims:
(a) One powerful argument for God’s existence is based on the existence of laws of Nature (written in the language of mathematics) which not only describe how objects actually behave, but also prescribe how those objects ought to behave, pointing to the existence of a Divine Prescriber, Who made those laws. This kind of transcendental argument takes as its starting point a pre-existing epistemic commitment on the part of a scientist, who is committed to the possibility of our being able to know about the external world. The argument then proceeds to show that in order to justify that commitment, one has to invoke a Creative Mind, Who is incapable by nature of deceiving us (i.e. God);
(b) Another type of argument, known as an abductive argument, takes as its starting point some observable state of affairs in the world (e.g. the existence of astronomically improbable configurations of parts performing some complex task, or executing some program), and then argues for an Intelligent Designer as the best explanation of those facts. If it can also be shown that the cosmos itself exhibits fundamental features pointing to its having been designed, then we may infer the existence of a Designer Who is Transcendent as well. In order to justify its conclusion, however, this type of argument appeals to premises based on our past and present observations of intelligent agents, and of unguided natural processes. Probabilistic calculations are then invoked, in order to show that the probability of these state of affairs occurring, given the existence of an intelligent Cause of Nature, is much, much higher than the probability of their occurrence in the absence of such a Cause;
(c) Yet another type of argument appeals to various rationality norms, relating to the kinds of questions scientists should ask. Since (as I argued above) there’s nothing obviously wrong with the question, “Why does the cosmos obtain?”, we should treat it as a legitimate question and look for an answer in a Necessary Cause Who cannot cease to obtain. More recently, Professor Paul Herrick, in his 2009 essay, Job Opening: Creator of the Universe — A Reply to Keith Parsons, has propounded what he calls his Daring Inquiry Principle: when confronted with the existence of some unexplained phenomenon X, it is reasonable to seek an explanation for X, if we can coherently conceive of a state of affairs in which it would not be the case that X exists. Herrick uses his Principle to argue for the legitimacy of inferring the existence of a Necessary Being Who created the cosmos through an act of free choice.
In a similar vein, the other rationality norms I listed above can all be used to construct powerful arguments for the existence of God. The fact that everything we see around us is composed of two or more parts prompts us to look for a Simple Cause of their existence. The fact that observable things possess arbitrary physical properties (as shown by the constants of Nature) points to the existence of a non-arbitrary Cause. The fact that the multiverse itself (according to cosmologist Alex Vilenkin) had a beginning, points to its having had a Cause – and replacing the statement, “The multiverse had a beginning” with the more innocuous statement, “Time has a finite duration” does nothing to obviate the problem either, for we can still legitimately ask why the universe has precisely that duration (since it’s an apparently arbitrary property of the cosmos-as-a-whole).
I have only sketched the arguments for God’s existence here. I explore these arguments in far greater depth in the following posts:
Does scientific knowledge presuppose God? A reply to Carroll, Coyne, Dawkins and Loftus
Is God a good theory? A response to Sean Carroll (Part One)
Is God a good theory? A response to Sean Carroll (Part Two)
Is God a good theory? A response to Sean Carroll (Part Three)
(The last post addresses the problem of evil, and why it isn’t a good argument against the existence of God.)
The conclusion that the God of classical theism exists is not a scientific one, strictly speaking, as such a God is not only physically simple, but also metaphysically simple. In addition, the God of classical theism is not merely free from arbitrary limitations, but also metaphysically infinite: such a God is often described as Being Itself, or Truth Itself, or Love Itself. Science cannot take us that far. Nevertheless, science can take us to a Being beyond this cosmos, as the cosmos (taken as a whole, which is how the science of cosmology takes it) exhibits features which are not self-explanatory, and which therefore require an explanation.
Religious arguments for the truth of this or that religion are not merely based on private revelation and holy texts, as Dr. Rosenhouse maintains. Rather, they are typically based on a very public revelation that is vouchsafed by large numbers of eyewitnesses who attest to having seen it. In that case, the credibility of the religious claim can be assessed by performing Bayesian logic on the testimony itself, as well as any supporting documents (manuscripts containing records of that testimony). In addition, a prior probability needs to be assigned for the supernatural claim in question e.g. a resurrection form the dead). The prior probability should not be assigned a zero value; nor should it be assigned an infinitesimal value (as that would violate Cromwell’s rule, which states that only statements that are logically true – e.g. No bachelor is married – or logically false – e.g. Tom is married and single – should be assigned a prior probability of 0 or 1).
A more sensible value for the prior probability of a miracle can be computed by following Laplace’s famous analysis of the Sunrise Problem, which would mean, for instance, that the prior probability of a resurrection from the dead is around 1 in 100 billion (the total number of individuals who have ever lived). In Chapter X of his Ninth Bridgewater Treatise (2nd ed., London, 1838; digitized for the Victorian Web by Dr. John van Wyhe and proof-read by George P. Landow), which is titled, On Hume’s Argument against Miracles, the nineteenth century mathematician Charles Babbage demonstrates that the testimony of even a small number of independent eyewitnesses is sufficient to overcome Hume’s daunting odds against the occurrence of a miracle. In Chapter XIII, he calculates the number of individuals who have ever lived to be 200 billion (which is about double the modern estimate), and goes on to discuss Hume’s example of a man being raised to life. Babbage concludes that we can indeed know that such events took place in the past.
In recent years, the philosophers Tim and Lydia McGrew have written an excellent article on the evidence for the Resurrection. The best critique of their article is an online essay by Jesse Parrish, who has a great deal of respect for the McGrews, but doesn’t think that their argument quite works. (I’ll be writing a post of my own on miracles in the near future.)
Arguments for the truth of a holy text are another matter. Such arguments can only rely on the strength of testimonial knowledge. If, for instance, the subject of a miraculous claim were to testify that some book was inspired, and if this testimony were followed by a miracle, then one could reasonably take that sign as constituting powerful evidence that the statements made in the book in question were actually true. And if the statements made in that text were quite clearly at odds with the best science of the day, then it would still be rationally prudent to believe the text over the scientific claims, as the Source validating the claims in the text is a Transcendent Being, Who presumably has access to far more reliable information about the cosmos than that currently possessed by our best scientists.
In other words, belief in a young Earth is not necessarily irrational. Nevertheless, it requires a lot of conditions to be satisfied, to make it epistemically warranted. The reason why I’m personally not a young Earth creationist is that I don’t think it’s at all clear that those conditions have been met. Nevertheless, I can understand why someone might be.
10. Dr. Rosenhouse is very alarmed at the damage done to science by the stranglehold of religious claims, which can choke its progress. He is especially critical of the attitude to science shown by the Church in the Middle Ages, where natural science was treated as the handmaid of theology, the queen of the sciences:
…[T]hat attitude is practically the definition of anti-science, at least as we understand that term today. They did not believe that nature should be studied solely by natural means, or that we should follow evidence wherever it leads, or that we should test our beliefs against evidence. Rather, they believed that science was valuable only insofar as it served religious ends, and if it strayed into areas on which the Church had taken a stand that it had to be stopped. Ruthlessly, if need be…
Galileo was a threat to the Church because he suggested that science, and not scripture, should be how we learn about nature. The Church saw this as a threat to its power by challenging its claims to religious authority, so they came down on him. Hard. If you don’t see that as a conflict between science and religion, then you need to rethink your definitions.
St. Thomas Aquinas did indeed speak of sacred doctrine (or theology) as a science, and as being nobler than the other sciences. But this tells only half the story. The term “science” at that time referred to any branch of knowledge, and it did not acquire its modern meaning until the early nineteenth century, under the influence of William Whewell, John Herschel, Charles Babbage and Richard Jones.
Regarding the interpretation of Scripture, what both St. Augustine and St. Thomas Aquinas maintained was that one should hold the truth of Scripture without wavering, but that since Scripture can be explained in multiple senses, one should be ready to abandon a particular interpretation of Scripture, if it be proved with certainty to be false, lest Scripture be exposed to the ridicule of unbelievers. Augustine laid down these exegetical principles in his De Genesi ad Litteram, Book I, chapter 19, paragraphs 38-39, a commentary on the opening chapters of Genesis. Aquinas cited these principles in his Simma Theologica, I, q. 68, art. 1.
Where Galileo did part company with Augustine and Aquinas was in maintaining that the authority of the Bible is effectively limited to matters with which the natural sciences cannot deal, making science and religion two independent domains of knowledge. Dr. Gregory Dawes has described Galileo’s exegetical position in an illuminating article entitled, Could there be another Galileo case? Galileo, Augustine and Vatican II. What Galileo upheld was an early version of Stephen Jay Gould’s NOMA – and it was unworkable for exactly the same reason. Whether we like it or not, religion does have things to say about the “Big Questions” (Whence came we? What are we? Whither go we?) which have empirical implications. For instance, when the Nicene Creed describes God as “maker of Heaven and earth,” then it obviously places itself in a position of potential conflict with science: if science were to show that the universe had no beginning, then God could still be its Sustainer, but not its Maker. To require religion to forego making such empirical claims is to effectively emasculate it, and confine it to the domain of purely spiritual (other-worldly) affairs. Religion confined in this way is but a shadow of its former self. It is incapable of transforming the way in which we live: it no longer provides an all-embracing worldview, as it leaves out the material universe.
Galileo’s problem was that by the standards of Augustine and Aquinas, the evidence marshaled he had for his heliocentric theory fell a long way short of “proof.” Physics was still in a very primitive state – Newton was born in 1642, the year Galileo died, and his Principia wasn’t published until 1687 – and it would take another two centuries before stellar parallax was observed by Bessel in 1838. Had Galileo been able to provide these proofs, the Inquisition would have been forced to back down; instead, they made a very bad decision based on their faulty, very rigid interpretation of some poetical passages in Scripture, such as Psalm 104:5, which (properly translated) does not speak of the Earth as never being moved, but as never faltering.
But if Dr. Rosenhouse can’t get the bad taste of the Galileo episode out of his head, then I would urge him to have a look at the progress of science in England, from the year 1660 (when the Royal Society of London was founded under King Charles II, with the motto, “Nullius in verba,” or “Take nobody’s word for it”) to the year 1865, when Maxwell’s equations were published. During that time, science flourished as never before in human history, and the spirit of intellectual inquiry was free and untrammeled. And yet the vast majority of scientists during this period were devout Christians. Does Dr. Rosenhouse seriously want to argue that these scientists’ Christianity impeded their scientific discovery-making? Did Victorian England hamper even the work of scientists such as Charles Darwin, whose Origin of Species ushered in a period of tension between science and religion on human origins? No; he was always at full liberty to pursue his research. When I look back at the intellectual freedom of the nineteenth century, I cannot help but contrast this liberty with the stifling political correctness of the modern era, when it is impossible to publicly doubt Darwin’s theory, or the latest IPCC climate projections, without being assailed as a “denier.”
And that brings me to my final point, which is that secular humanism keeps science in a straitjacket, by failing to ask the really hard questions that scientists ought to be asking. Instead of asking, “What is a law of Nature? Why do we have the laws of Nature that we do?”, secular scientists are likely to stop their train of intellectual inquiry at a nice, neat-looking “brute fact”: maybe an equation that fits on the back of a T-shirt. “This is as far as science can go,” they’ll say. And in the process, science will be horribly stunted.
Ask yourself which attitude is really more harmful to science: the view that the whole of Creation is a manifestation of the Mind of God, Who wants His intelligent creatures (human beings) to understand as much as possible about His plan for creation, or the view that we are the product of four billion years of evolution from slime, that our brains are kluges that can’t be trusted to think straight, and that scientists’ inadequate theories will always have to be revised, but should nonetheless be accepted with Gospel fervor whenever the politics of the day demands it?
Dr. Rosenhouse should be careful what he wishes for: he just might get what he wants.