00:00:00.000 Welcome to Topcast, episode 7 on the MegTrip Science, where we're going to look at what
00:00:18.240 science is and what makes it special. In the last episode, I spent a long time, perhaps
00:00:39.680 too long, in explaining exactly what science is not. When I use the word science here, I mean what
00:00:46.640 the top guide refers to as the natural sciences. For now, when you hear me use the word science,
00:00:53.440 presume I'm talking about the natural sciences. And in this sense, science is not a process
00:00:59.280 where predictions are generalized from repeated observations. That is, science has nothing to do with
00:01:05.040 the fallacious method of reasoning which is commonly called induction. Unfortunately, this is the
00:01:11.040 view you will read and hear about at various times when studying talk. Now I claimed that the
00:01:18.080 process of induction not only did not explain how scientific theories are justified, but also that
00:01:23.760 induction wasn't even a method of justification at all, plain and simple. It does not even work for
00:01:28.880 everyday justification of our theories. What I have done then is dismissed the entire problem of
00:01:34.960 induction by saying that no knowledge can ever be produced through induction. It's an unfortunate
00:01:40.240 and rather serious shortcoming of texts and resources that choose to characterize science in
00:01:44.960 this way. Science is, after all, our best attempt at trying to understand the world. It is not
00:01:50.240 a mere generalization of predictions. Once again, I urge you to research and consider alternative
00:01:55.680 points of view to my own. I'm happy to discuss these. But science is a process, just not an
00:02:02.320 inductive process. It's a process that begins with a problem, offers testable solutions, and generates
00:02:08.640 explanations which are themselves predictive. We can define science therefore as a process for
00:02:14.320 creating new knowledge, as well as the body of knowledge generated by that same process.
00:02:21.120 So on this view, science generates reliable theories about how the world works.
00:02:26.960 The best theories we have in this regard at the moment form the body of scientific knowledge.
00:02:32.160 Now, just as an aside here, I've used the word world a number of times in these podcasts.
00:02:39.040 And I'm using it in the philosophical, not the geographic sense. In other words, it's bigger than
00:02:44.480 just the planet Earth. The world in philosophy is everything that exists. It's the whole universe,
00:02:50.960 and to some extent even more than just the physical universe. It's the sum total of reality.
00:02:56.560 So when I say science is a process, and a collection of our best theories about how the world
00:03:02.960 works, I used the word theory in a special way that I've alluded to before. A scientific theory,
00:03:09.920 above all else, is an explanation. Our scientific theories are our best explanations of everything
00:03:17.520 that exists in the universe. But what is an explanation? An explanation tells you why something
00:03:25.120 happens. It gives reasons. An explanation allows you to understand a process or a system.
00:03:31.840 It is also a representation of a process or a system which closely resembles, in some sense,
00:03:37.200 that process or system. For example, the theory of how stars produce light,
00:03:42.080 consists of symbols we call words, and mathematics that represent in symbolic form this process.
00:03:48.320 Between the star itself and the explanation, there is some sort of correspondence between
00:03:56.160 individual components of what is happening in the star, and the symbols in the explanation.
00:04:02.320 Now before the scientific revolution, which happened in the wider context of the Enlightenment
00:04:07.200 centered on Europe in the 18th century, people thought that most reliable knowledge came from
00:04:12.400 authority and tradition. An authority would generally rely upon either some rule of thumb
00:04:20.000 or perhaps an ancient text. The physics and astronomy of Aristotle was followed because of
00:04:26.800 tradition. Indeed, Galileo is reported to have remarked that few of his peers wanted to look
00:04:32.880 through his telescope because most everything that needed knowing came from ancient texts and
00:04:36.880 scholars. Clearly, if Aristotle taught that the geocentric theory was what natural law
00:04:42.000 mandated, what could the telescope add or subtract from this? The church, as is famously now known,
00:04:48.560 used its authority to enforce belief in Aristotle's physics. The Bible itself is not entirely
00:04:54.560 clear on the question about whether the solar system is geocentric or heliocentric.
00:04:59.600 Taking another historical example, if say your problem was a medical one,
00:05:04.160 like say some infection, then a frequent treatment that came from traditional medicine
00:05:09.600 was bloodletting. The idea here was that letting the supposedly diseased carrying blood out
00:05:18.720 would do some good. Now of course, we did not rely upon authority or tradition to validate
00:05:24.480 our knowledge claims today. Indeed, the foremost scientific organization in the world in 1660
00:05:31.280 was the Royal Society of London. Their motto was, and remains, Nullius Inverbia. Take no
00:05:37.680 one's word for it. In so far as these days, authority or tradition do have any part to play with
00:05:43.920 knowledge claims, we understand the liabilities associated with trusting these sources.
00:05:50.240 Now, I might have to take it on authority to some extent that Andrew Wiles 100-plus page-long proof
00:05:57.840 of Fermar's last theorem is valid because I cannot understand the details. All that the best
00:06:03.040 treatment for a cerebral aneurysm is endovascular coiling because I've not had training
00:06:07.760 in neurosurgery, but I also know that these things can be checked by people who do know
00:06:13.200 and that anyone with sufficient time and intelligence can check the relevant facts.
00:06:18.160 Ultimately, such knowledge does not spring for more authority alone. It springs from scientific
00:06:24.160 or mathematical investigation into the world. These investigations lead to problems. How is it
00:06:31.680 that we explain the origin of species? How do we explain how people catch diseases? How is it
00:06:37.760 that metals corrode? Why do planets wander across the night sky? And what is the cause of earthquakes?
00:06:43.680 These problems demand solutions. What is noticeable about all of the solutions to these problems
00:06:50.960 is that we explain what we do see in terms of something we cannot see. Take all the different
00:06:57.680 animal species in the world, something we can see. Explained in terms of gradual evolution
00:07:04.640 by natural selection over many eons, something we cannot see. Sick people, something we see.
00:07:12.320 Explained through germs and viruses, something unseen. Corrosion, planetary motion and stars,
00:07:20.160 all seen. Explained through electron transfer, the curvature of space and nuclear fusion,
00:07:26.640 all unseen. Science makes conjectures about these unseen entities
00:07:34.160 to explain these processes. We do not observe these entities directly. We do not see
00:07:40.400 evolution by natural selection occurring. We do not see the laws of physics or nuclear fusion.
00:07:46.000 Indeed, to be very strict, we do not even directly see anything. When we look at one type of
00:07:51.040 finch, and notice it is different to another, we do not see a finch as it is. What happens
00:07:58.240 is that light enters our eyes and an electrical signal is sent to our brains. This electrical
00:08:03.520 signal is a pattern of impulses along neurons in our brains. One pattern of impulses,
00:08:08.400 rather than another, means that we recognize a finch as a red finch, rather than a blue finch.
00:08:14.160 Our brains interpret these electrical impulses in such a way that we experience the world
00:08:21.440 in the way we do. Our brains synthesize the pattern of electrical activity sent to it by our
00:08:26.640 sense organs and form it into a coherent hole that forms our conscious experience of the world.
00:08:35.120 How is it that our senses give us knowledge of the world? Well, when it comes to seeing,
00:08:40.640 it is just a way I have described. But notice that understanding this way of knowing
00:08:46.000 relies upon a scientific theory about sense perception. This scientific theory involves explaining
00:08:52.960 how light is reflected from the wings of finches and how light is then refracted by our eyes,
00:08:58.480 and ultimately how it causes electrical impulses in our brains that allow us to say that our
00:09:04.240 senses give us knowledge about the world. This is just what I said about how sense perception
00:09:08.800 works in episode 3 of Topcast. We do not observe the world directly, to be sure. But I emphasize
00:09:16.560 again that this is no real problem for anyone who is not interested in certain knowledge,
00:09:20.560 but rather who is more concerned about explanations. Our scientific theories about how our senses
00:09:27.120 work explain that they represent the world to us in such a way that we can gain reliable
00:09:33.040 information about the world. And the reliable information they give us about the world,
00:09:37.520 however subject to error that information might be, is useful for developing theories about
00:09:42.880 the world. The consequence for science on this view is simple. We never directly observe anything.
00:09:50.160 So observation is not the most important thing in science.
00:09:54.320 So what's the most important thing in science then?
00:09:57.840 Well, I'm glad you asked. It's explanation. And explanations can be good or bad as I said in the
00:10:03.840 last episode. Those explanations produce predictions and those predictions can be tested
00:10:09.760 against the observations. If the predictions do not match the observations, then we must change
00:10:15.840 or reject our theory. But there is more to a good theory than just this. It must also be based
00:10:21.680 upon few assumptions. And this is Occam's Razor. Occam's Razor is also sometimes known as the
00:10:36.960 principle of passimani. And although the two ideas are different in very subtle ways,
00:10:43.600 I will use them interchangeably here. Occam's Razor says in common language,
00:10:49.360 all else being equal, the most simple theory is the preferred one. Now, I prefer to say that
00:10:54.560 Occam's Razor can be expressed as the theory which makes the fewest assumptions used to be the
00:11:00.240 preferred one. And what's important to notice here that Occam's Razor is also not the most
00:11:05.040 important concern in choosing the best scientific theory. The best scientific theory must also
00:11:10.720 explain as much or more than all of the travels. In addition, of course, to being falsifiable,
00:11:15.600 Occam's Razor is just one of the features that a scientific theory must satisfy. But sometimes,
00:11:21.840 it can make all the difference. The reason why Copernicus first proposed a heliocentric theory
00:11:27.680 was because Ptolemy's idea of a geocentric theory required far too many assumptions.
00:11:33.600 In order for the predictions about the behavior of planets to be accurate,
00:11:37.360 epicycle upon epicycle had to be independently assumed. It was much easier simply to assume
00:11:43.280 that the Sun was at the center. At first, the predictions that the heliocentric theory made were
00:11:48.640 actually not as accurate as the geocentric theory. But as we can see from this example,
00:11:53.520 sometimes predictions are not what matter most. It is both Occam's Razor and the
00:11:59.440 abilities to make precise and accurate predictions of that taken together constitute important
00:12:04.240 features of a scientific theory. But there is yet another feature now to add to these two
00:12:08.880 features discussed so far about scientific theory being predictive or falsifiable and able to
00:12:14.320 satisfy Occam's Razor. The next feature is that scientific theory should also be hard to vary.
00:12:22.480 They should not be easy to change. An example might help you.
00:12:33.280 Why is it that earthquake sometimes happen in Japan?
00:12:36.640 An actual explanation provided by the Japanese themselves in ancient times involved a catfish,
00:12:43.920 a huge catfish. This huge catfish was called Namazoo. Namazoo lives in the mud beneath the earth
00:12:50.800 and is guarded by the god Kashima, who restrains the fish with a stone which is put over his head.
00:12:56.240 When Kashima lets his guard fall, Namazoo thrashes about causing violent earthquakes.
00:13:01.040 Okay, so what's wrong with this explanation? First and foremost, it is easy to vary.
00:13:09.280 Why a catfish? Why not a tuna? Or a lizard? And why the god Kashima? Why is a rock needed?
00:13:15.760 Why wasn't the catfish tied up? This explanation is easy to vary and it postulates a raft of
00:13:21.920 unsupported entities which themselves require explanation. It is hardly parsimonious. It does not
00:13:28.400 obey Occam's razor. Rather than actually explaining earthquakes, it simply postulates a bunch of
00:13:34.160 needless assumptions about gods and catfish and rocks, each of which themselves would require
00:13:40.320 explanation if we really were serious about trying to solve the problem of what causes earthquakes
00:13:46.480 in Japan. And this explanation does not generate anything in the way of accurate predictions.
00:13:51.680 Is this theory even falsifiable? Well, actually in a sense it is. If the ancient Japanese
00:13:58.720 who believe this theory knew about say the United States, which is far from Japan, and they
00:14:04.640 further knew that the United States also suffered earthquakes, then how would Namazoo explain earthquakes
00:14:11.600 there? Presumably such an observation of earthquakes in the United States would strictly speaking
00:14:19.040 falsify the theory about Namazoo. If the islands of Japan alone are on the back of the catfish,
00:14:25.440 then what about the United States? It's far larger than Japan. Presumably it would require
00:14:31.520 many catfish, or perhaps it requires something larger like salmon. So it's not the fact that
00:14:39.200 the catfish theory of earthquakes isn't falsifiable that makes it un-scientific, it's probably
00:14:44.240 falsifiable. It's not scientific because the theory is too easy to change. We can substitute
00:14:50.480 catfish for anything you like. And also, those entities which supposedly do the explaining,
00:14:57.760 the god and the catfish and the rock, actually do nothing of the sort. They complicate matters by
00:15:03.600 themselves being unexplained, so they violate Occam's razor. The theory is not simple. It's complicated.
00:15:10.320 It postulates a raft of things, all of which are unexplained. You're in a worse mess than before
00:15:15.680 you started trying to explain anything. We really are none the wiser. And now we've got a raft
00:15:21.440 of strange entities including gods, catfish, islands attached to the vertebrae of catfish,
00:15:27.040 stones used to restrain imprisoned catfish. Questions about feeding and motivation and the list
00:15:32.160 is long and preposterous. It's certainly fun to make up such theories. We could make up one now.
00:15:44.240 In the center of the earth, they've giant trolls. The giant trolls push boulders around just below
00:15:52.160 the surface of the earth. These boulders act like bearings between the surface of the earth
00:16:00.000 and layers of rock deeper down. Now and again, one of the trolls. Gary, the smartest
00:16:07.600 the rest. Let's go of his rock and it crashes into some of the other trolls. At these times,
00:16:14.800 the collisions can be felt on the surface of the earth as earthquakes. This explains earthquakes.
00:16:23.200 Again, we can go through the list of things wrong with this explanation. First and foremost,
00:16:28.080 it is too easy to vary. Why trolls? Could I not simply change trolls for something else like,
00:16:33.680 say, giant rhinoceros? Could not change boulders for huge wheels? So what is the actual theory
00:16:40.960 which explains earthquakes in Japan? And how is it any better? Well, it's the theory of tectonic
00:16:48.240 plates. Japan sits at the interface between two large solid plates that are moving at the
00:16:54.560 rate fingernails grow and are grinding against one another. Pressure builds up and is released over time
00:17:01.040 and sometimes this happens quickly. This is felt as an earthquake. This also happens to explain
00:17:06.960 why there are usually aftershocks and also why, in and around this country we find volcanoes,
00:17:12.800 the gaps between plates are places where the material below the surface, which is molten,
00:17:17.680 can break through the gaps. This also explains why the plates move. They are floating on this
00:17:24.000 very molten stuff called the mantle, which we call lava when it comes to the surface.
00:17:29.200 This explanation is shockingly hard to change. Could we simply change plate for anything else?
00:17:36.400 Now it has to have all the properties of a tectonic plate and not introduce anything
00:17:40.640 further that's unexplained. Okay, what about a big catfish that looks just like a plate?
00:17:46.000 It does everything that a plate does, doesn't it? Yes, and more. It then complicates matters,
00:17:51.280 as we would then want to know, how does this catfish eat? What does it eat?
00:17:55.760 How, if it looks identical to a tectonic plate, can we reasonably assert that it is actually a catfish?
00:18:01.680 The plate, like all aspects of the tectonic plate theory, are very hard to change,
00:18:07.840 and the theory makes predictions. It says we should expect earthquakes all over the world,
00:18:14.080 but especially where plates meet. We should further expect a pattern of volcanoes.
00:18:20.160 They should be clustered around regions where plates meet also.
00:18:23.920 It also predicts that we should find evidence in the geological record of the
00:18:27.520 continents having moved over time. For example, we might find very similar rocks now on opposite
00:18:33.040 sides of the earth, or even similar fossils on countries that now have entirely different
00:18:37.440 organisms living on them. It also fits well with the rest of what we know about science.
00:18:43.520 In this case, the theory of tectonic plates fits with what we know about paleontology.
00:18:48.320 Paleontology suggests that the continents used to be connected, as we know that the same
00:18:52.960 dinosaurs that lived, for example, in South America, also roamed in Africa.
00:19:03.200 I hope what I've showed here is that a scientific theory is an explanation.
00:19:08.080 It's an explanation that could possibly be wrong, but so far has survived all attempts at proving
00:19:13.840 it wrong. It should also be difficult to vary, and it should fit in well with everything else
00:19:19.200 that we know about science. So let us summarize what we have learned so far about what it takes
00:19:25.440 for a scientific theory to be good. Firstly, it should be able to make precise and accurate
00:19:31.520 predictions. That is, it should be testable. The science of astronomy makes clear predictions
00:19:36.560 about where the planets will be in the night sky. Next, it should rest on as few assumptions
00:19:42.880 as possible. That is, it should conform to Occam's razor. All else being equal, it should be the
00:19:48.960 simplest theory. Thirdly, it should be difficult to vary. Next, it should explain as much,
00:19:57.120 preferably more than all it rivals. It might not explain everything, but that's okay.
00:20:03.520 The classical example is that Einstein's relativity explains everything that Newton's physics
00:20:08.320 explains, but also, it can account for some things that Newton cannot. For example,
00:20:14.240 Einstein's relativity correctly explains how much starlight bends during a solar eclipse,
00:20:19.280 and it also explains what is known as the procession of Mercury's orbit. That is,
00:20:24.880 how much the axis of the orbit moves as Mercury goes around the sun.
00:20:30.000 And also, as far as scientific theories are concerned, in order for them to be good,
00:20:35.120 they should mesh well with what we already know about the rest of science.
00:20:39.280 The theory of evolution by natural selection does not contradict anything we know
00:20:43.920 in any other area of science. Indeed, the theory that species change over time due to the natural
00:20:50.000 selection, natural selection is a non-random process, of mutations. These are the random part of
00:20:57.360 evolution by natural selection of random mutations. This theory fits in perfectly well with what
00:21:03.680 we know about genetics and physics at the molecular level. We know that genes exist on a molecule
00:21:09.520 called DNA, and that these genes carry information that codes for certain features. The mutations,
00:21:15.040 which occur, can be caused by a number of things including radiation. If evolution by natural
00:21:20.400 selection was contradicted by some law of physics or chemistry, then we would have reason to question
00:21:25.200 it. It's worth mentioning, just as an aside here, that one objection raised by some about
00:21:32.560 Darwinian natural selection is whether or not it contradicts what is known as the second law of
00:21:37.680 thermodynamics. It certainly does not. The second law talks about a certain property of closed
00:21:43.600 systems called entropy. Entropy can very loosely be called disorder, the higher the entropy,
00:21:50.240 the more disordered a system is. So, for example, ice is a nicely ordered crystal, but liquid
00:21:57.600 water is disordered, with all the molecules all over the place, and far more free to move. The second
00:22:03.360 law says that for a closed system, entropy must increase. Some people loosely also contrast the
00:22:11.760 disorder of high entropy with the complexity of low entropy, and assert that evolution causes
00:22:17.280 complexity and therefore order to increase in contradiction to the second law. For example,
00:22:23.200 evolution on earth shows that natural selection is making the ordered, that is life,
00:22:28.240 out of the disordered, i.e. non-living stuff. The important point here is that the second law
00:22:34.320 applies only to a closed system, and the earth is certainly not a closed system. Energy is constantly
00:22:41.760 falling onto the earth from the sun. This energy from the sun enables life to make order
00:22:46.560 out of disorder, and complete agreements with the second law. Taken together, the sun and
00:22:52.000 earth system, all as one, the total entropy of the system does increase. I mentioned this,
00:22:58.800 because if some people who claim that there was a contradiction were right, then evolution would
00:23:04.880 indeed not mesh well with what we know about physics, and we would have to question one of these
00:23:09.920 theories. As it actually is, however, both fit perfectly well together. You simply have to understand
00:23:16.480 your thermodynamics more completely to see that there is no contradiction. Now we have five features
00:23:22.400 of a good scientific theory. It should be falsifiable. That is, make predictions. It should rely
00:23:28.160 on as few assumptions as possible. It should be hard to vary, and explain more than a
00:23:33.440 driver's, and also fit with everything we already know. Now, what else is there that science does
00:23:47.760 well? Well, here is a sixth feature of science, generally speaking. It improves. That is, it makes
00:23:56.880 progress. Science is a process of open-ended discovery. We do not reach a point and then say,
00:24:03.040 well, now we are finished. No, we continue revising our theories. We continue to refine them and
00:24:09.040 improve them. Things are objectively better now than in the past. If you have bacterial men and
00:24:15.600 jitis, the medical treatments available to you now are objectively better than they were 200 years
00:24:20.880 ago, when the disease would have been a death sentence. If you wish to get from Sydney to London,
00:24:26.960 we have superior ways of doing so now compared to any time in the past. Clearly, we have made
00:24:33.760 progress over the centuries, where once our only means of transportation was our feet,
00:24:38.480 our we can fly. And we can fly because we really have learned some things about aerodynamics
00:24:43.760 and internal combustion. If you wish to explain what those lights in the sky are now,
00:24:49.200 then it can be said that we have made progress in the last few hundred years and we continue to learn
00:24:54.720 more. Science improves. In this latter case, our knowledge of astronomy and cosmology is far
00:25:02.080 from complete, but there is no question it is an improvement upon what we once knew.
00:25:07.200 Indeed, the data comes in daily and the scientific papers are relentless in coming.
00:25:12.480 The steps sometimes falter, but this is no problem, the general trend is an improvement in our
00:25:17.920 knowledge. We know there is more to know and more to explain. Our exploration of the cosmos is
00:25:23.520 truly open-ended, indeed our attempts to explain all phenomena remain so. Yet we can still say
00:25:29.520 some things with a higher degree of reliability. We might not know everything about exactly how
00:25:34.480 nuclear fusion in the core of stars works, but we know much more than we did 50 years ago,
00:25:39.680 and we knew much more than than we did 200 years before that.
00:25:44.560 This sense in which science admits of progress can be identified with Karl Popper's idea of
00:25:50.320 verisillamitude. There are what now? There are silamitude. It means closeness to truth.
00:25:58.800 Now it might very well be an axiom of epistemology, but we can never know if we have ever described
00:26:04.160 reality as it really is. That is, we may never know if we have obtained the truth about the world,
00:26:10.000 but we can make objective claims about our theories compared to one another. Clearly,
00:26:15.360 the heliocentric theory is closer to the truth than the geocentric theory.
00:26:19.440 Anyone who denies this has probably ruled him or herself out of the discussion.
00:26:24.080 It is not a problem that we do not know for certain what the truth is.
00:26:29.040 As I have said a number of times now, almost like a mantra, a scientist is not interested
00:26:34.240 in certainty, but rather in explanations. So long as we seek to maximize verisillamitude,
00:26:41.360 closeness to truth, then we have achieved our aim in science.
00:26:45.040 If our explanations map reality as closely as possible, our only gauge of this is to assess
00:26:51.040 our theories as being better or not. I want to return for a moment to this important interaction
00:26:59.280 between what the top guide refers to as a way of knowing, namely, sense perception,
00:27:04.720 and the area of knowledge called the natural sciences. I have said already that scientific
00:27:10.800 theories explain the scene in terms of what is not seen, and what is not seen cannot come
00:27:17.840 through the senses. I think this shows that sense perception as a way of knowing is only
00:27:22.960 possible when coupled with reason. Now empiricism is the idea that knowledge is gained
00:27:29.120 through the senses alone. Clearly, such a view is not true. The laws of nature are
00:27:35.120 mathematical in form, but we do not see mathematical equations. We do not see the curvature
00:27:41.200 of space time, or we do not see photons or the origin of species. So empiricism alone cannot
00:27:47.440 be correct. As an aside here, it is worth noticing that induction also seems to assert that the
00:27:53.360 unseen, for example the next sunrise, resembles the scene, for example the last sunrise.
00:27:59.360 We've already seen that this is not true. Scientific knowledge can be tested by observation,
00:28:06.160 but can't be derived from it. We never observe something as it is in itself. We do not see
00:28:13.440 books, for example. We do not even see light. Our eyes detect light, but we are not our eyes.
00:28:20.400 We are our brains. And our brains detect electrical impulses. And we do not detect those as they are.
00:28:27.280 What they are is electricity. Instead, our brain interprets those impulses, those electrical
00:28:34.240 impulses, and synthesizes them into a coherent hole that forms our consciousness of the world.
00:28:41.280 And it is here in our consciousness that we can reason about our sense perceptions,
00:28:46.960 and form theories and criticize those theories as being good or bad explanations.
00:28:51.920 But we never perceive reality as it is. At least we do not know if we perceive reality as it is.
00:28:59.040 Rather, we reason about the contents of our mind and seek to explain what it is we perceive.
00:29:05.280 Our best systematic attempt at doing this is known as science.
00:29:11.040 Next time, I want to address exactly what the scientific method is and some more of the questions
00:29:16.640 from the guide about the area of knowledge, we call the natural sciences.
00:29:29.200 For resources and links associated with talk-ass visit the talk-ass website at www.talkass.net.
