On Friday and Saturday last week, I spoke at the Ontario Christian Home Educators’ Convention in Ontario, Canada. It has been 10 years or more since I last spoke there, so it was nice to be back. The convention was held on the campus of Redeemer University College in Ontario, where the admissions director is a home educator. Like many universities, Redeemer has learned that homeschool graduates make above-average university students (see here, here, here, here, and here), so they actively encourage homeschool graduates to apply. They are also happy to support homeschooling in Canada.
The convention was very well attended, and based on a show of hands at my keynote session, about 20% of the attendees had never been to a homeschool convention before. As I spoke with individual attendees, it became clear that several of the people at the convention were considering home education for the first time. When I mentioned this to one of the conference organizers, he indicated that the new premier of Ontario is introducing a radical sex education program, and it is causing many in Ontario to look for a way out of the government school system. Based on what I read about the new program, I truly hope lots of parents remove their children from such a horrible situation!
I gave a total of six talks at the convention: Homeschooling: The Solution to our Education Problem, Be Open-Minded, but Don’t Let Your Brain Fall Out, What About K-6 Science?, Why Homeschool Through High School, ‘Teaching’ High School at Home , and How are Homeschool Graduates Doing? When I give talks in a different country, I always try to make them relevant for that country, so most of the statistics I shared came from Canadian education studies, and most of the experts I quoted in my talks were Canadian. The conference organizers really appreciated that. I guess some U.S. speakers come to Canada and just assume that all of their U.S.-based talks are relevant to Canadians, and some of them just aren’t.
One thing I have to note is that this convention really knew how to make an out-of-the-country speaker feel right at home. They arranged for me to have a home-cooked dinner the night that I arrived, and it was great! The couple who hosted me had three charming children, two of whom colored pictures for me. Those pictures are now on my bulletin board in my office. Then, each morning, another couple cooked breakfast for me. The other meals were catered by Redeemer University College. Everyone made sure I had everything I needed to be as comfortable as possible. I don’t get pampered like that very often, and it was really nice!
I got some excellent questions after each talk, and one of them was from a high school student. During Be Open-Minded, but Don’t Let Your Brain Fall Out, I made the point that in many ways, the Medieval Christian Church was more open-minded than many modern-day Evangelical churches, because it understood that you can learn from those with whom you fundamentally disagree. For example, the Medieval Christian Church required its priests to be very knowledgeable about the teachings of Aristotle, who was a pagan. While the church didn’t agree with his views on religion, they understood that Christians could still learn a lot from Aristotle, especially when it came to logic and rhetoric.
During the question/answer session afterwards, a high school student asked if I thought that some of Aristotle’s religious views ended up creeping into the Medieval Church’s doctrines, even though they were studying him for other reasons. I told the student what an excellent question he had asked and then told him that we was, of course, absolutely correct. The teachings of Aristotle, for example, strongly influenced Thomas Auqinas, who was an incredibly important figure in the Medieval Christian Church. In fact, some have called Thomas Aquinas “The Christian Aristotle.” Obviously, then, if we are going to learn from those who are not Christians, we must continually check their ideas against Scripture to make sure our theology is not unduly influenced by them!
One other question I got was from a parent. He came up to me at my speaker’s booth and asked, “What do you think about quantum mechanics.” When he asked that question, I thought, “Oh, no. He’s one of those Christians who thinks quantum mechanics is bad.” I told him that quantum mechanics is a robust, incredibly useful theory that explains many aspects of the atomic world. It might not give us the entire picture, however, since it fundamentally contradicts general relativity, which is also a robust, incredibly useful theory about space, time, and gravity. Nevertheless, it has so much evidence stacked in its favor that it must be providing a reasonable view of how things work at the atomic level.
The parent then asked, “Do you think that God uses quantum mechanics to intervene in nature?” At that point, I realized he wasn’t an anti-quantum-mechanics Christian. Indeed, he had studied it enough to know that in quantum mechanics, specific outcomes cannot be predicted. Only the probabilities of a range of outcomes can be predicted. Thus, God could “rig” the probabilities so the outcome that He wants in a particular circumstance will be the one that actually happens. That way, God can intervene in His creation without technically breaking any natural law. I told him that there are several theologians and philosophers who think that way (see here, here, and here, for example).
However, I told him I don’t think that’s the way God intervenes in nature. First, I think that rigging the probabilities is, in fact, breaking the rules of quantum theory. After all, the rules specifically dictate that the outcomes are determined randomly. If there is even one time when the event isn’t determined randomly, then technically, it is a violation of quantum theory. Second, as a scientist, I don’t have a problem with the idea of God breaking the natural laws. After all, He created them. Logically, it only makes sense that He can break them. When He wants a miracle to happen, then, He simply works the miracle, since the laws of nature have no power over Him. Of course, I don’t think God does that very often, so as a scientist, I can still count on the natural laws working normally the vast majority of the time.
I really loved my time in Ontario, and I hope it is less than 10+ years before I speak there again.
Your statement about the Medieval Church being more tolerant toward alternative philosophies than the Churches of today is quite apt. One would think Christians would have enough confidence in their faith to not be fearful of teaching or studying the logistics of competing beliefs.
Also, to play the role of the skeptic, “B-But the Inquisition!”
Zorcey, the inquisition was about combating heresy from within. Thus, it wasn’t about whether or not the church thought it could learn from those with whom it disagreed. It was about the church enforcing its authority on those who already claimed to submit to its authority. This is a pretty good discussion of the Medieval Inquisition.
I was the teaching assistant for my department’s grad quantum course last semester, so let’s hope what I’m about to say makes sense:
Quantum mechanics precisely determines the probabilities of results of measurements, but it doesn’t determine the result itself. Somehow the result just happens/is chosen with probability given by the wavefunction, and there’s no way to predict what it will actually be. Further, QM itself doesn’t posit a mechanism for this choice; for that you need stuff beyond QM (which I don’t know much about, though I’m pretty sure all the solutions to the measurement problem are still speculative at this point).
Because QM has no mechanism for providing the actual result of a measurement, I think it doesn’t break the rules of QM for God to choose the result – as long as there is some probability for that result to occur. This is different from rigging the probabilities (though I suppose there’s no reason God couldn’t choose outcomes of collisions and whatnot to make the future probabilities He wanted). Usually, “the universe” just blindly samples the wavefunction for the result, right? Sampling without a blindfold doesn’t change the distribution, and QM can only tell us about the distribution.
For example, say we sent an unpolarized beam of neutrons through a Stern-Gerlach apparatus. None of the rules of QM would be broken if, in the first run of the experiment, 60 percent of the neutrons were observed to have spin up. Now, as a physicist I’d look at that and say, “That’s ridiculous – that should never happen” (after saying “Oops – my beam must be partially polarized” and checking my equipment), but what I’d really mean is it should almost never happen. Because there is some absurdly small probability that it will happen. And unless I kept getting that same result after multiple experiments, or found some new physics to explain it – no way would a scientist stop observing after getting such a result – and other people got the same result in different labs, I wouldn’t be able to say that what I saw violated QM, any more than I could say it’s impossible for a coin to land on heads 100 times in a row. (I can’t help myself here: imagine, in the many-worlds interpretation of QM, that one poor universe in which EVERY Stern-Gerlach experiment measured ALL the particles as spin-up! Physicists would be so confused.)
Anyway, I doubt miracles occur this way; it just seems silly – perhaps too human (Isaiah 55:8 and following). But I don’t think anything I said is inconsistent with QM. Am I missing something?
What you say makes perfect sense, Jake, but I would still have to disagree with you. Sure, if you saw an event that was supposed to happen 50% of the time and saw it happening 60% of the time for a while, that wouldn’t be a surprise at all. However, let’s consider something a bit more dramatic than that. After all, this is supposed to be about miracles, right? Let’s suppose an event was supposed to happen with a probability if 1 in 650,000. Now suppose you saw it happen three times in a series of 100 trials. What would you think? You would think there is something wrong with the physics you are using to analyze your experiment or the experiment itself, because there is simply no way that such a low-probability event would happen that often in a small series of trials.
I chose those numbers for a reason. In poker, a Royal Flush (the strongest hand) occurs with a probability of 1 in 650,000 (649,739 to be exact). Now, let’s suppose you sat down and played 100 hands of poker with a player and he came up with three Royal Flushes in those 100 games. What conclusion would you draw? You would rightly conclude that he must be cheating. I see the same thing with God working through quantum mechanics. When the distributions are known to be random (regardless of whether or not we can explain why they are random), a non-random distribution is a violation of the rules.
But I think even that small probability is enough to say that the person may not have been cheating. I should have been more clear in my example: I don’t actually know what the beam luminosity was in the original Stern-Gerlach experiment (and they used silver atoms), but I was imagining at least 10 neutrons per second (the neutron beam at NIST gets around 10^5 per second). If I then let the experiment run for 100 seconds, the probability that I’d get 60 percent spin-up would be (using Stirling’s approximation on the binomial coefficient) 10^(1000((3/5)log_{10}(3/5)+(2/5)log_{10}(2/5))), or approximately 10^-300. That’s absurd; 3 royal flushes out of 100 hands is nowhere close (it’s not even 10^-20). And even 10^-300 is nowhere close to the kinds of probabilities you see in thermodynamics, where the probability of having a mole of gas entirely on one side of its container is around 2^(-10^23). These are the kinds of probabilities I’m referring to – and random sampling doesn’t preclude such events. So even though it’d be crazy, observing an event with probability 10^-300 wouldn’t make me think QM was violated – as long as I only saw it once. Since miracles themselves are rare, I think the argument works.
The problem, though, is that if my magical Stern-Gerlach apparatus kept giving me 60 percent spin-up, eventually we’d get more data going against QM than supporting it. What’s amusing to me is that the many-worlds interpretation says there are universes where only the outlying events occur. And since the many-worlds interpretation is consistent with QM itself, this line of reasoning makes QM unfalsifiable. (Of course, there’s some distribution to the number of universes, so there could be very few universes with magical S-G apparatuses. But that doesn’t help the physicists in those universes.) So for that reason I can see how the miracles-through-QM argument would fail. But at the same time, I think getting that 60 percent result after 100 s of a 10 Hz luminosity beam is still consistent with QM. Since this problem is clearly scientific in nature (how many outliers have to occur before I conclude QM is wrong?), I think it’s actually independent of the miracles issue. And I don’t like the idea of miracles through QM anyway. But it’s interesting to think about.
Sorry – it figures I’d forget to divide by the total number of possibilities (2^1000, lol). The actual probability of observing 60 percent spin-up is actually 10^-9, not 10^-300.
Well, that does deflate things a bit, but if you run my magical S-G device 200 s the probability is still smaller than 3 royal flushes in 100 games. My previous answer was bothering me; now I can get to sleep…
I am glad you caught your mistake, Jake, and it seems to make my point. As I said, “if you saw an event that was supposed to happen 50% of the time and saw it happening 60% of the time for a while, that wouldn’t be a surprise at all.” Under your first set of conditions, then, seeing 60% spin-up neutrons is possible, because the probability is 1 in a billion. That’s a very low probability, but still possible to observe. However, when you start talking about 1 in 10^20, the probability is so incredibly low that, in the end, you would have to question the randomness of the situation if you actually observed it.
And I guess that’s the point. If I saw outliers like an event with a probability of 1 in 10^20, I wouldn’t conclude that QM is wrong. Instead, since QM works so well in so many instances (and since the results are truly random in those instances), I would rather conclude that something is wrong with my experiment. Either I have set my experiment up so that the probability isn’t really 1 in 10^20, or there is some underlying process I don’t understand that it causing the probability distribution to be different from what I think it should be. Of course, if God does miracles through manipulating the probability distribution of QM, that would the the “underlying process I don’t understand,” and it would be a violation of QM.
I meant to post this earlier, but I got distracted.
I discussed this with one of my fellow physics grad students, and he put it this way: choosing results from a distribution isn’t the same as sampling from it randomly. I think this is the point you were making, but for some reason it was clearer to me put that way.
So let me see if I understand you correctly: the laws of the universe dictate some normal mode through which quantum distributions are somehow randomly sampled for the outcomes of measurements. God choosing extremely rare outcomes would correspond to “turning off” that normal mode, and would hence be a violation of QM.
The problem with this is that QM itself can produce the extremely rare outcomes God chose. To the experimenter, choosing results from a distribution *is* the same as sampling from it randomly, as long as over time the measurements balance out to the quantum distribution. There’s no way to distinguish the two. And it doesn’t matter whether the probability is 10^-9 or 2^(-10^23). The experimenter would be right to suspect something fishy happening but *couldn’t say so for sure* unless the same outlying results occurred forever – and everyone else’s experiments produced the same thing.
Not to be trite or too simple, but QM is all about absurdly improbable things occurring (loosely interpreting “improbable”) – it’s why we can joke about baseballs tunneling through walls. Now that I think about it, 10^-15 to 10^-20 sounds like the right number for the probablity of decay of a single nucleus within 1 second (assuming you need about a mole of a radioactive element to see many decays). Of course, by giving myself a time scale (1 s) the number becomes less significant.
Anyway, thanks for arguing with me about this. I don’t know if you remember, but about 10 years ago (when I was still in high school, doing your science courses) I sent you a few e-mails with questions about QM. I really appreciated that you responded (I still have your e-mails); I ended up spending part of my senior year of high school reading “Primer of Quantum Mechanics” by Chester on your recommendation. I guess what I’m saying is that, by being such an amazing person and resource, not only have you inspired people to study science; there is now a nonzero number of people – myself included – who want to be you when we grow up.
Thank you so much, Jake. I can’t tell you what it means to know that I helped to inspire you to study science!
Your fellow student expressed it perfectly. Choosing results from a distribution isn’t the same as sampling from it randomly. If QM is based on random sampling, then choosing a result is a violation of QM. Now I do agree with you that the experimenter can never say for sure that random chance was violated, but if several unlikely events happen in a short time, then he or she has good evidence that it has been violated. The same could be said of a miracle that is the result of direct divine intervention, however. If a person has inoperable brain cancer and is then healed by direct divine intervention, you couldn’t say for sure that it was a miracle. It could always be chalked up to the body healing itself.
Hmm. I guess if I want to argue this, I have to be prepared to say that QM could have parted the Red Sea and stopped the sun in the sky… I’m not sure I *have* to go that far (I doubt QM allows the latter) but all the same. This does allow the reduction of every miracle to some natural process.
I still think the argument works – and low-probability events would still be “miraculous” in a sense – but now I feel better about dismissing it. Does Popper discuss testing of probabilistic statements in The Logic of Scientific Discovery? (Also, I’ve been using HTML tags to italicizes things, but it doesn’t look like it works…)
Popper discusses this at great length, Jake. Chapter 8 of the book is entitled, “Probability,” and he starts it off like this:
I can’t do justice to his discussion in a brief reply (or even a blog post). However, he ends up saying that while it is absolutely impossible to falsify a probabilistic theory, we can deem it “logically more improbable” or “logically more probable.” If an experiment ends up producing a significantly different frequency of outcomes than predicted by the theory, the theory is “logically more improbable.” If the experiment produces a frequency distribution that is close to the predictions of the theory, it is “logically more probable.”
(I see the title of Popper’s book in italics in your comment.)
This isn’t entirely relevant, but speaking of sampling from distributions, guess where the other grad student I mentioned went to high school? In fact, out of about 110 people, three of us were homeschooled. And a fourth is coming in the fall! Though the number mirrors the homeschool population in the U.S., it still surprises me (and we’re not at a shabby program either). Although this might be evidence that we’re not adequately social…
Hehe. I would say it is probably a reflection of the fact that homeschool graduates are more likely to have an associates degree, a bachelors degree, a masters degree, or a profession degree than public and private school graduates.