Is Science in as Good a Place as it Might Be?

Most people probably think that science progresses through all scientists diligently seeking the truth but that illusion was was shattered when Thomas Kuhn published “The Structure of Scientific Revolutions.” Two quotes:

(a) “Under normal conditions the research scientist is not an innovator but a solver of puzzles, and the puzzles upon which he concentrates are just those which he believes can be both stated and solved within the existing scientific tradition.”

(b) “Almost always the men who achieve these fundamental inventions of a new paradigm have been either very young or very new to the field whose paradigm they change. And perhaps that point need not have been made explicit, for obviously these are the men who, being little committed by prior practice to the traditional rules of normal science, are particularly likely to see that those rules no longer define a playable game and to conceive another set that can replace them.”

Is that true, and if so, why? I think it follows from the way science is learned and then funded. In general, scientists gain their expertise by learning from a mentor, and if you do a PhD, you work for several years in a very narrow field, and most of the time the student follows the instructions of the supervisor. He will, of course, discuss issues with the supervisor, but basically the young scientist will have acquired a range of techniques when finished. He will then go on a series of post-doctoral fellowships, generally in the same area because he has to persuade the new team leaders he is sufficiently skilled to be worth hiring. So he gains more skill in the same area, but invariably he also becomes more deeply submerged in the standard paradigm. At this stage of his life, it is extremely unusual for the young scientist to question whether the foundations of what he is doing is right, and since most continue in this field, they have the various mentors’ paradigm well ingrained. To continue, either they find a company or other organization to get an income, or they stay in a research organization, where they need funding. When they apply for it they keep well within the paradigm; first, it is the easiest way for success, and also boat rockers generally get sunk right then. To get funding, you have to show you have been successful; success is measured mainly by the number of scientific papers and the number of citations. Accordingly, you choose projects that you know will work and shuld not upset any apple-carts. You cite those close to you, and they will cite you; accuse them of being wrong and you will be ignored, and with no funding, tough. What all this means is that the system seems to have been designed to generate papers that confirm what you already suspect. There will be exceptions, such as “discovering dark matter” but all that has done so far is to design a parking place for what we do not understand. Because we do  not understand, all we can do is make guesses as to what it is, and the guesses are guided by our current paradigm, and so far our guesses are wrong.

One small example follows to show what I mean. By itself, it may not seem important, and perhaps it isn’t. There is an emerging area of chemistry called molecular dynamics. What this tries to do is to work out is how energy is distributed in molecules as this distribution alters chemical reaction rates, and this can be important for some biological processes. One such feature is to try to relate how molecules, especially polymers, can bend in solution. I once went to hear a conference presentation where this was discussed, and the form of the bending vibrations was assumed to be simple harmonic because for that the maths are simple, and anyhting wrong gets buried in various “constants”. All question time was taken up by patsy questions from friends, but I got hold of the speaker later, and pointed out that I had published paper a long time previously that showed the vibrations were not simple harmonic, although that was a good approximation for small vibrations. The problem is that small vibrations are irrelevant if you want to see significant chemical effects; they come from large vibrations. Now the “errors” can be fixed with a sequence of anharmonicity terms, each with their own constant, and each constant is worked around until the desired answer is obtained. In short you get the asnswer you need by adjusting the constants.

The net result is, it is claimed that good agreement with observation is found once the “constants” are found for the given situation. The “constants” appear to be only constant for a given situation, so arguably they are not constant, and worse, it can be near impossible to find out what they are from the average paper. Now, there is nothing wrong with using empirical relationships since if they work, they make it a lot easier to carry out calculations. The problem starts when, if you do not know whyit works, you may use it under circumstances when it no longer works.

Now, before you say that surely scientists want to understand, consider the problem for the scientist: maybe there is a better relationship, but to change to use it would involve re-writing a huge amount of computer code. That may take a year or so, in which time no publications are generated, and when the time for applications for further funding comes up, besides having to explain the inactivity, you have to explain why you were wrong before. Who is going to do that? Better to keep cranking the handle because nobody is going to know the difference. Does this matter? In most cases, no, because most science involves making something or measuring something, and most of the time it makes no difference, and also most of the time the underpinning theory is actually well established. The NASA rockets that go to Mars very successfully go exactly where planned using nothing but good old Newtonian dynamics, some established chemistry, some established structural and material properties, and established electromagnetism. Your pharmaceuticals work because they have been empirically tested and found to work (at least most of the time).

The point I am making is that nobody has time to go back and check whether anything is wrong at the fundamental level. Over history, science has been marked by a number of debates, and a number of treasured ideas overthrown. As far as I can make out, since 1970, far more scientific output has been made than in all previous history, yet there have been no fundamental ideas generated during this period that have been accepted, nor have any older ones been overturned. Either we have reached a stage of perfection, or we have ceased looking for flaws. Guess which!

Advertisements

Science and Climate Change

In the previous post, I questioned whether science is being carried out properly. You may well wonder, then, when this week the Intergovernmental Panel on Climate Change issued a rather depressing report, and a rather awkward challenge: according to their report, the world needed to limit the temperature rise to 1.5 degrees C between now and 2050, and to do that, it needed to cut carbon emissions by 45% by 2030, and net zero by 2050. Even then significant amounts of carbon have to be removed from the atmosphere. The first question is, then, is this real, and if so, why has the IPCC suddenly reduced the tolerable emissions? If their scientists previously predicted seriously lower requirements, why should these be considered better? There are two simple answers. The first is the lesser requirements were based on the assumption that nations would promptly reduce emissions. Most actually increased them. The second is more complicated.

The physics have been verified many times. However, predicting the effects is another matter. The qualitative effects are easily predicted, but to put numbers on them requires very complicated modelling. The planet is not an ideal object, and the calculation is best thought of as an estimate. What has probably happened is their modelling made a projection of what would happen, and they did this long enough ago that now that they can compare prediction with where we are now. That tells them how good the various constants they put into the model were. Such a comparison is somewhat difficult, but there are clear signs in our observations, and things are worse than we might hope for.

So, what are we going to do? Nothing dramatic is going to happen on 2040, or 2050. Change will be gradual, but its progress will be unstoppable unless very dramatic changes in our behaviour are made. The technical challenges here are immense. However, there are a number of important decisions to be taken because we are running short of time due to previous inaction. Do we want to defend what we have? Do we want to attempt to do it through sacrificing our life style, or do we want to attempt a more aggressive approach? Can we get sufficient agreement that anything we try will be properly implemented? Worst of all, do we know what our options are? Of these questions, I am convinced that through inaction, and in part the structural defects of academic science, the answer to the last question is no.

The original factor of required emissions reduction was set at 1990 as a reference point. What eventuated was that very few countries actually reduced any emissions, and most increased them. The few that did reduce them did that by closing coal-fired electricity generation and opted for burning natural gas. This really achieves little, and would have happened anyway. Europe did that, although France is a notable exception to this in that it has had significant nuclear power for a long time. Nuclear power has its problems, but carbon emissions are not one of them. The countries of the Soviet Union have also actually had emission reductions, although this is as much as anything due to the collapse of their economies as they made the rather stupid attempt to convert to “free market economics” which permitted a small number of oligarchs to cream the economy, sell off what they could, use what was usable, pay negligible wages and export their profits so they could purchase foreign football clubs. That reduced carbon emissions, but it is hardly a model to follow.

There is worse news. Most people by now have recognized that Donald Trump and the Republican party do not believe in global warming, while a number of other countries that are only beginning to industrialize want the right to emit their share of CO2 and are on a path to burn coal. Some equatorial countries are hell-bent on tearing down their rain forest, while warming in Siberia will release huge amounts of methane, which is about thirty times more potent than CO2. Further, if we are to totally change our way of life, we shall have to dismantle the energy-related infrastructure from the last fifty years or so (earlier material has probably already been retired) and replace it, which, at the very least will require billions of tonnes of carbon to make the required metals.

There will be some fairly predictable cries. Vegetarians will tell everyone to give up meat. Cyclists will tell everyone they should stop driving cars. In short, everyone will have ideas where someone else gives up whatever. One problem is that people tend to want to go for “the magic bullet”, the one fix to fix them all. Thus everyone should switch to driving electric vehicles. In the long term, yes, but you cannot take all those current vehicles off the road, and despite what some say, heavy trucks, major farm and construction equipment, and aircraft are going to run on hydrocarbons for the foreseeable future. People talk about hydrogen, but hydrogen currently requires massive steel bottles (unless you are NASA, or unless you can get hydrides to act reversibly). And, of course, there is a shortage of material to make enough batteries. Yes, electric vehicles, cycling, public transport and being a vegetarian are all noble contributions, but they are just that. Wind and solar power, together with some other sources, are highly desirable, but I suspect that something else, such as nuclear power must be adopted more aggressively. In this context, Germany closing down such reactors is not helpful either.

Removing CO2 from the atmosphere is not that easy either. There have been proposals to absorb it from the effluent gases of coal-fired power stations. Such scrubbing is not 100% efficient, but even if it were, it is not dealing with what is already there. My guess is, that can only be managed by plants in sufficient scale. While not extremely efficient, once going they look after themselves. Eventually you have to do something with the biomass, but restoring all the tropical rain forests would achieve something in the short term. My personal view is the best chances are to grow algae. The sea has a huge area and while we still have to learn how to do it, it is plausible, and the resultant biomass could be used to make biofuel.

No, it is not going to be easy. The real question is, can we be bothered trying to save what we have?

From Whence Star-burning Planets?

This series started out with the objective of showing how life could have started, and some may be wondering why I have spent so much time talking about the cold giant planets. The answer is simple. To find the answer to a scientific problem we seldom go directly to it. The reason is that when you go directly to what you are trying to explain you will get an explanation, however for any given observation there will be many possible explanations. The real explanation will also explain every connected phenomenon, whereas the false explanations will only explain some. The ones that are seemingly not directed at the specific question you are trying to answer will nevertheless put constraints on what the eventual answer must include. I am trying to make things easier in the understanding department by considering a number of associated things. So, one more post before getting on to rocky planets.

In the previous two posts, I have outlined how I believe planets form, and why the outer parts of our solar system look like they do. An immediate objection might be, most other systems do not look like ours. Why not? One reason is I have outlined so far how the giants form, but these giants are a considerable distance from the star. We actually have rather little information about planets in other systems at these distances. However, some systems have giants very close to the star, with orbits (years) that take days and we do not. How can that be?

It becomes immediately obvious that planets cannot accrete from solids colliding that close to the star because the accretion disk get to over 10,000 degrees C that close, and there are no solids at those temperatures. The possibilities are that either there is some mechanism that so far has not been considered, which raises the question, why did it not operate here, or that the giants started somewhere else and moved there. Neither are very attractive, but the fact these star-burning giants only occur near a few stars suggests that there is no special mechanism. Physical laws are supposedly general, and it is hard to see why these rare exceptions occur. Further, we can see how they might move.

There is one immediate observation that suggests our solar system is expected to be different from many others and that is, if we look again at LkCa 15b, that planet is three times further from the star than Jupiter is from our star, which means the gas and dust there would have more than three times less concentrated, and collisions between dust over nine times rarer, yet it is five times bigger. That star is only 2 – 3 My old, and is about the same size as our star. So the question is, why did Jupiter stop growing so much earlier when it is in a more favourable spot through having denser gas? The obvious answer is Jupiter ran out of gas to accrete much sooner, and it would do that through the loss of the accretion disk. Stars blow away their accretion disks some time between 1 and 30 My after the star essentially finishes accreting. The inevitable conclusion is that our star blew out its disk of gases in the earliest part of the range, hence all the planets in our system will be, on average, somewhat smaller than their counterparts around most other stars of comparable size. Planets around small stars may also be small simply because the system ran out of material.

Given that giants keep growing as long as gas keeps being supplied, we might expect many bigger planets throughout the Universe. There is one system, around the star HR 8799 which has four giants arrayed in a similar pattern to ours, albeit the distances are proportionately scaled up and the four planets are between five and nine times bigger than Jupiter. The main reason we know about them is because they are further from the star and so much larger, hence we an see them. It is also because we do not observe then from reflected light. They are very young planets, and are yellow-white hot from gravitational accretion energy. Thus we can see how planets can get very big: they just have to keep growing, and there are planets that are up to 18 times bigger than Jupiter. If they were bigger, we would probably call them brown dwarfs, i.e. failed stars.

There are some planets that have highly elliptical orbits, so how did that situation arise? As planets grow, they get gravitationally stronger, and if they keep growing, eventually they start tugging on other planets. If they can keep this up, the orbits get more and more elliptical until eventually they start orbiting very close to each other. They do not need to collide, but if they are big enough and come close enough they exchange energy, in which case one gets thrown outwards, possibly completely out of its solar system, and one gets thrown inwards, usually with a highly elliptical orbit. There are a number of systems where planets have elliptical orbits, and it may be that most do, and if they do, they will exchange energy gravitationally with anything else they come close to. This may lead to a sort of gravitational billiards, where the system gets progressively smaller, and of course rocky planets, being smaller are more likely to get thrown out of the system, or to the outer regions, or into the star.

Planets being thrown into the star may seem excessive, nevertheless in the last week it was announced that a relatively new star, RW Aur A, over the preceding year had a 30 fold increase in the amount of iron in its spectrum. The spectrum of a star comes from whatever is on its surface, so the assumption is that something containing a lot of iron, which would be something the size of a reasonably sized asteroid at least, fell into the star. That means something else knocked it out of its orbit, and usually that means the something else was big.

If the orbit is sufficiently elliptical to bring it very close to the star one of two things happen. The first is it has its orbit circularized close to the star by tidal interactions, and you get one of the so-called star-burners, where they can orbit their star in days, and their temperatures are hideously hot. Since their orbit is prograde, they continue to orbit, and now tidal interactions with the star will actually slowly push the planet further from the star, in the same way our moon is getting further from us. The alternative is that the orbit can flip, and become retrograde. The same thing happens as with the prograde planets, except that now tidal interactions lead to the planet slowly falling into the star.

The relevance of all this is to the question, how common is life in the Universe? If we want a rocky planet in a circular orbit in the habitable zone, then we can eliminate all systems with giants on highly elliptical orbits, or in systems with star burners. However, there is a further possibility that is not advantageous to life. Suppose there are rocky planets formed but the star has yet to elimiinate its accretion disk. The rocky planet will also keep growing and in principle could also become a giant. This could be the reason why some systems have Neptune-sized planets or “superEarths” in the habitable zone. They probably do not have life, so now we have to limit the number of possible star systems to those that eliminate their accretion disk very early. That probably elimimates about 90% of them. Life on a planet like ours might be rarer than some like to think.

Science and Sanctions

This may seem an odd title in that most people consider science far away from describing human activities. I am not suggesting the scientific method should govern all of human activities, but I think that a little more attention to its methods would help humanity (and I try to show a little of this in my novels, although I am unsure that most would notice). The first important point, of course, is to clarify what the scientific method is. Contrary to what you may see on TV programs, etc, it is not some super geek sitting down solving impossible mathematical equations. Basically, the scientific method is you form propositions, perhaps manipulate them, then check with reality whether they might be correct. The most important feature here is, check the evidence.

What initiated this post was news that the US House of Representatives has passed a bill that will impose new sanctions on Russia, including (according to reports here) the forbidding of any help with Russia’s oil and gas industry, and President Trump has signed it into law. So, what are the premises behind this?

The first one is that foreign countries will oblige and help carry them out.

The second, presumably, is that Russia will now fall into line and do whatever the sanctions are intended to make it do.

The third is, if Russia cannot export more oil or gas, their prices will rise.

The fourth is, removing Russian hydrocarbons from the international market will lead to further markets for US hydrocarbons. Note the US now has the capacity to be a major exporter, thanks to fracking.

The first two depend on each other, and obviously, seeking evidence of the future is not practical, nevertheless we can look at the history of sanctions. Are there any examples of countries “bending the knee” in response to sanctions when they probably would not have done it anyway? I cannot think of any. Obviously, sanctions are less likely to effective if foreign countries refuse to cooperate, which is why the two are linked. The two most recent examples of sanctions are Iran and North Korea. Both have been imposed for sufficient time, and the question is, how effective are they?

In the case of Iran, one objective is claimed to have been met in that Iran argues it no longer has the capacity to make nuclear weapons, however it also claimed that was never its intention. Everyone seems to delight in arguing whether either of those statements is true, but in my opinion nuclear weapons are a poor strategic objective for Iran. I also believe they are a poor option for North Korea, but seemingly someone has to show Kim that is so. For either of them, what would it gain? Iran has opted (if truthful) to avoid nuclear weapons, but then again, what has it gained from doing so? The sanctions America imposed are still largely there. As for the effectiveness of sanctions, it appears that Iran is doing reasonably well, and a number of countries are buying its oil, including China. So I conclude that sanctions are not particularly effective there.

North Korea does not seem in any immediate hurry to “bend the knee” to the US and while it has suffered the harshest sanctions, apparently over the last few years its exports have increased by at least 40%, mainly to China. President Trump has accused China of not helping, and he is correct, but being correct does not get anyone very far. The obvious question is, why is North Korea chasing after better weapons? The answer is obvious: it is at war with the US and South Korea. The Korean War never ended formally. The sides agreed to a ceasefire, but no permanent treaty was signed, so one of the actions that America could have taken in the last sixty years or so would have been to negotiate a formal peace treaty. You may well say, the US would never launch a preemptive strike against North Korea. You may well be right, but are you that sure? From North Korea’s point of view, the US has launched cruise missile attacks frequently against places it does not like, it has significant military bases in Syria, it invaded Iraq, and so on. You might argue that the US was justified because these countries were not behaving, and you may well be right, but from North Korea’s point of view, it is at war with the US already, so it has decided to do what it can to defend itself. One approach to end this ridiculous position would be to at least offer a treaty.

The third and fourth premises are probably ones the US Congress does not advertise, because they are full of self-interest. Apparently there is enough liquefied natural gas able to be produced to substitute for Russian gas in Europe. So, why don’t they sell it? Competition is a good thing, right? The simplest answer is price and cost. Europe would have to build massive lng handling facilities, and pay a lot more for their gas than for Russian gas. And it is here that these sanctions may run into trouble. The Germans will lose heavily from the loss of Russian gas, in part because their industries are involved in expanding the Russian fields and pipelines, and of course, they would have to pay more for gas, and some equipment would need changing for the different nature of the gas.

So, if we return to the evidence, I think we can conclude that these latest attempts at sanctions are more based on self-interest than anything else. There is no evidence they will achieve anything as far as pushing Russia around goes. It is true, if imposed, they would hurt Russia significantly, but they would also hurt Europe, so will Europe cooperate?

Trump and Climate Change

In his first week in office, President Trump has overturned President Obama’s stopping of two pipelines and has indicated a strong preference for further oil drilling. He has also denied that climate change is real. For me, this raises two issues. The first is, will President Trump’s denial of climate change, and his refusal to take action, make much difference to climate change? In my opinion, not in the usual sense, where everybody is calling for restraint on carbon dioxide emissions. The problem is sufficiently big that this will make only a minor difference. The action is a bit like the Captain of the Titanic finding two passengers had brought life jackets so he confiscates them and throws them overboard. The required action was to steer away from a field of icebergs, and the belief the ship was unsinkable was just plain ignorant, and in my opinion, the denial that we have to do something reasonably dramatic about climate change falls into the same category. The second issue is how does science work, and why is it so difficult to get the problem across? I am afraid the answer to this goes back to the education system, which does not explain science at all well. The problem with science for most people is that nature cares not a jot for what you feel. The net result is that opinions and feelings are ultimately irrelevant. You can deny all you like, but that will not change the consequences.

Science tries to put numbers to things, and it tries to locate critical findings, which are when the numbers show that alternative propsitions are wrong. It may be that only one observation is critical. Thus Newtonian mechanics was effectively replaced by Einstein’s relativity because it alone allowed the calculation of the orbital characteristics of Mercury. (Some might say Eddington’s observation of light bending around the sun during an eclipse, but Newton predicted that too. Einstein correctly predicted the bending would be twice that of Newton, but I think Newton’s prediction could be patched given Maxwell’s electrodynamics. For Newton’s theory, Mercury’s orbit was impossible to patch.)

So what about climate change? The key here is to find something with the fewest complicating factors, and that was done when Lyman et al. (Nature 465: 334-337, 2010) measured the power flows across ocean surfaces, and found there was a net input of approximately 0.6 W/m2. That is every square meter gets a net input of 0.6 Joules per second, averaged over the 24 hr period. Now this will obviously be approximate because they did not measure every square meter of ocean, but the significance is clear. The total input from the star is about 1300 W/m2 at noon, so when you allow for night, the fact that it falls away significantly as we get reasonably away from noon, and there are cloudy days, you will see that the heat retained is a non-trivial fraction of the input.

Let us see what that means for the net input. Over a year it becomes a little under 19 MJ for our square meter, and over the oceans, I make it about 6.8 x 1021 J. There is plenty of room for error there (hopefully not my arithmetic) but that is not the point. The planet is a big place, and that is really a lot of energy: about a million million times 1.6 tonnes of TNT.

That has been going on every year this century, and here is the problem: that net heat input will continue, even if we totally stopped burning carbon tomorrow, and the effects would gradually decay as the carbon we have burnt gradually weathers away. It would take over 300 years to return to where we were at the end of the 19th century. That indicates the size of the physical problem. The fact that so many people can deny a problem exists, with no better evidence than, “I don’t believe it,” is our current curse. The next problem is that just slowing down the production of CO2, and other greenhouse gases, is not going to solve it. This is a problem that has crept up on us because a planet is a rather large object. It has taken a long time for humanity’s efforts to make a significant increase to the world’s temperatures, but equally it will take a long time to stop the increase from continuing. Worse, one of the reasons the temperature increases have been modest is that a lot of this excess heat has gone into melting ice. Eight units of water at ten degrees centigrade will melt one unit of ice, and we end up with nine units of water at nought degrees Centigrade. The ice on the planet is a great restraint on temperature increases, but once the ice in contact with water has melted, temperatures may surge. If we want to retain our current environment and sea levels, we have some serious work to do, and denying the problem exists is a bad start.

Why do we do science?

What is the point of science? In practice, most scientists use their knowledge to try to make something, or solve some sort of problem, or at least help someone else do that. (Like most occupations, most junior ones turn up to work and work on what they are told to work on.) But, you might say, surely, deep down, they are seekers of the truth? Unfortunately, I rather fancy this is not the case. The problem was first noted by Thomas Kuhn, in his book, “The structure of scientific revolutions”. In Kuhn’s view, scientific results are almost always interpreted in terms of the current paradigm, i.e. while the data are reproduced properly, they are interpreted in terms of current thinking, even if that does not fit very well. No other theory gets a look-in. If a result does not conform to the standard theory, the researcher does not question the standard theory. The first effort is to find some way of accommodating it, and if that does not work, it may be listed as a question for further work, in other words the researcher tries to persuade someone else to find a way of fitting it to the standard paradigm rather than taking the effort to find an alternative theory.

According to Kuhn, most science is carried out as “normal science”, wherein researchers create puzzles that should be solved by the standard paradigm, in other words, experiments are set up not to try to find the truth, but rather to confirm what everyone believes to be true. This is not entirely unreasonable. If we stop and think for a moment, an awful lot of such research is carried out by PhD students, or post-doctoral fellows. The lead researcher has submitted his idea as a request for funding, and this is overseen by a panel. If you submit something that would not get anywhere within the current paradigm, you will not get funding because the panel will usually consider this to be a waste of time. On top of that, if you are going to include a PhD student in this work, that student needs a thesis at the end of his work, and that student will not thank the supervisor for coming up with something that does not produce results that can be written up. In other words, the projects are chosen such that the lead researcher has a very good idea as to what will be found, and it will be chosen so that it is unlikely to lead to too great an intellectual challenge. An example of a good project might to make a new chemical compound that might be a useful drug. The project might involve new synthetic work, there will be problems in choosing a route, but the project will not founder on some conceptual problem.

Natually, the standard paradigm clearly must have much going for it to get adopted in the first place. It cannot be just anything, and there will be a lot of truth in it, nevertheless as I mentioned in my first ebook, part 1 of “Elements of Theory”, any moderate subset of data frequently has at least two theories that would explain the data, and when the paradigm is chosen, the subset is moderate. If all that follows it to investigate very similar problems, then a mistake can last. The classic mistake was Claudius Ptolemy’s cosmological theory, which was the “truth” for over 1600 years, even though it was wrong and, as we now recognize, with no physical basis. If you wish to find the truth, you might follow Popper and try to design experiments that would falsify such a theory, but PhD theses cannot be based like that as it is too risky that the student will find nothing and fail to get his degree through no fault of his.

What brought these thoughts on was a recent article in the journal Icarus. The subject was questioning how the Moon was formed. The standard theory of planetary formation goes like this. After the star forms, the accretion disk that remains settles the dust on the central plane, and this gradually congeals into larger bodies, which further join together when they collide, and so on, until you get planetesimals (objects about the size of asteroids) then, apart from the asteroids, eventually embryos (objects about the size of Mars) which gravitationally interact and form very eccentric orbits, and then collide to form planets (except for Mars, which is a remaining embryo). All such collisions once planetesimals form are random, and the underpinning material could have come from a very large region, thus Earth was made from embryos formed from material beyond Mars and Venus. The Moon was formed from the splatter arising from a near glancing collision of a Mars-sized body called Theia with Earth.

If you carefully measure the isotope ratios of samples of meteorites, what you find is that all from the same origin have the same isotope ratios, but those from different parts of the solar system have different ratios. As an example, oxygen has three stable isotopes of atomic weights 16, 17 and 18. We have carbonaceous chondrites from the outer asteroid belt, a number of samples from Vesta, some from Mars, and of course unlimited supplies from here. The isotope ratios of these samples are all the same from one source, but different between sources. We also have a good number of samples from the Moon, thanks to the Apollo program. Now, the unusual fact is, the Moon is made of material that is essentially identical to our rocks, at least in terms of isotope ratios.

This Icarus paper carried out simulations of planetary formation employing the standard theory, and showed that since the Moon is largely Theia, the chances of the Moon and Earth having the same ratio of even oxygen isotopes is less than 5%. So, what conclusion do the authors draw? The obvious one is that the Moon did not form that way; a more subtle one is that planets did not form by the random collision of growing rocky bodies. However, they drew neither. Instead, they really refused to draw a conclusion.

I should add that I have in interest in this debate, as my mechanism outlined in Planetary Formation and Biogenesis has the planets grow from relatively narrow zones, although the disk material is always heading towards the star to provide new feed. The Moon grows at the same distance as Earth (at a Lagrange point) from the star and hence has the same composition. The concept that the Moon formed at either L4 or L5 was originally proposed by Belbruno and Gott in 2005 (Astron. J. 129: 1724–1745) and I regard it as almost dishonest not to have mentioned their work, which predicts their result provided the bodies form from local material. Unfortunately, the citing of scientific work that contradicts the standard theory is not exactly frequent, and in my view, does science no service. The real problem is, how common is this rejection of that which is currently uncomfortable?

You may say, who cares? It may very well be that how the Moon formed is totally irrelevant to modern society. My point is, society is becoming extremely dependent on science, and if science starts to become disinterested in seeking the truth, then eventually the mistakes may become very significant. Of course mistakes will be made. That happens in any human endeavor. But, do we want to restrict them to unavoidable accidents, or are we prepared to put up with avoidable errors?