Ed Yong (@edyong209) is a well-known and highly respected science writer. At regular intervals he posts lists of links on his website, Not Exactly Rocket Science, of science stories that he has found interesting, a sort of one-man blog carnival. On his links list for 20 April he included a link to Adam Gopnik’s BBC Point of View piece, which I recently criticised, with the following description.
Galileo was a great scientist because he wasn’t afraid to admit when he was wrong. If only more of us did the same.
Now in recent months we have had a series of talks and articles by such luminaries as Paul Nurse, President of the Royal Society, making similar claims for science and scientists in general. That is that scientists are characterised by their willingness to admit that they are wrong and to give up the theories they hold that have been proved to be defective. Such speeches have had historians of science all over the Intertubes banging their heads against the wall in collective displays of disbelief because even a cursory survey of the history of science would show that the exact opposite is true, scientists hang on to their cherished theories until the bitter end against all sorts of opposition and refuting evidence and Galileo is a prime example of such behaviour. For me this attitude, and it is not necessarily one that I condemn, was wonderfully summed up by Einstein when Eddington confirmed the General Theory of Relativity empirically. Asked by a reporter what he would have done if Eddington had refuted his theory Einstein is said to have replied then I would have said that Eddington’s measurement were wrong. A certain amount of tenacity is important in the early development of scientific theories, which are seldom born complete and perfect and are brought to their optimal condition through a process of criticism and modification to refute that criticism. Not giving up supposedly refuted theories is part and parcel of the scientific process but sometimes this tenacity can be and is misplaced and Galileo is one historical figure who delivers very good examples of a man who held onto wrong theories beyond the point of no return.
Galileo is well known as a supporter of Copernican heliocentricity and as one of the founders of the new mechanics but in both theories his adherence to an antiquated theory held him back an adherence that he maintained although he must have known that it was wrong. The antiquated theory that Galileo refused to abandon was the so-called Platonic axiom in astronomy. This metaphysical axiom says that the planets move with uniform motion in circles. Like Copernicus before him Galileo’s fidelity to this axiom meant the retention of the whole Ptolemaic apparatus of deferents and epicycles. Clinging to this axiom also meant that Galileo failed to formulate the principle of inertia properly as he believed, like a good Aristotelian, that only circular motion was natural motion. All well and good but why do I claim that he should have known better? The answer is Johannes Kepler.
Before Galileo had finished writing his Dialogo Kepler had already delivered his three laws of planetary motion thus completely refuting the Platonic axiom. Kepler’s laws were strict mathematical laws derived from Tycho Brahe’s empirical observations. Here was modern science in action if ever there was and Galileo ignored it clinging to the clearly refuted Greek orthodoxy, why? Faced with this seemingly inexplicable behaviour of their hero the Galileo fan club argue that Galileo could not accept Kepler’s bizarre Renaissance meta-physics and that is why he refused to accept Kepler’s work. This seems like a reasonable argument until one takes a closer look at the evidence. Kepler’s first two laws were delivered in his Astronomia Nova (1609) a book that contains none of his possibly off putting meta-physics. Also the clearest statement of his three laws and their application to a heliocentric system is in his Epitome astronomiae Copernicanae (1617, 1620 & 1621) a textbook, which also is relatively free of anything that might have offended Galileo’s sensibilities.
I think that the answer is actually to be found in Galileo’s ego. Kepler’s work on the planetary orbits was a much better and more convincing argument for heliocentricity than anything Galileo had produced. In fact it was the Epitome Copernicanae combined with Kepler’s Rudolphine Tables that led to the acceptance of heliocentricity in the seventeenth century and not Galileo’s work. If Galileo were to include Kepler’s work in his Dialogo then he would be merely the messenger and not the creator so he simply ignored it and stuck to the Platonic axiom that he knew to be wrong.
Galileo’s second case of refusing to admit that he was wrong is even more undignified. In 1618 a spectacularly bright comet was visible over Europe, which was of course carefully observed by nearly all the leading astronomers. One notable exception was Galileo who because of ill health had been unable to take part in the observations. By now Galileo was Northern Italy’s leading natural philosopher feted for his quick wit and his even quicker slicing tongue with which he took great pleasure in dicing his scientific opponents. Asked for his opinion on the nature of the new comet Galileo, who as already noted actually knew nothing about it, took the strange step of attacking the Jesuit astronomer Orazio Grassi who had carefully observed the comet and based on his observations had correctly calculated that the comet was a supra-lunar celestial body. Galileo now claimed that Grassi was wrong and presented what was in essence the out dated and discredited Aristotelian theory that comets were sub-lunar meteorological phenomena. This notorious dispute culminated in Galileo’s Il Saggiatore with its famous “the book of nature is written in the language of mathematics” quote. Here is Galileo bizarrely lecturing Grassi that investigations of nature must be empirical and mathematical in a situation where Grassi’s investigations were just that and Galileo’s own were definitely not. Of course Galileo’s brilliant polemic crushed his poor hard done by opponent without Galileo’s claque noticing that Grassi was in the right and Galileo very much in the wrong something that the maestro almost certainly knew, however his ego would not let him admit it. He had to win the argument at all costs even if he was horribly wrong.
We now turn to the most notorious case of Galileo refusing to accept that he was wrong his theory of the tides. This was the crowing glory of Galileo’s argumentation for heliocentricity his only empirical evidence. This took up the whole of the fourth day of his Dialogo delivering the climax and was the only argument that he brought forward in 1615 when he was trying to head off a condemnation of Copernicus by the Inquisition. This was Galileo’s trump. Unfortunately it suffered from one glaring defect it was wrong. It was empirically, irrefutably, undeniably, indisputably wrong.
This theory of the tides was first formulated in about 1596 by Galileo and Paolo Sarpi in one of their intellectual sparring sessions. In fact it is not clear if the theory is from Galileo or Sarpi, 1596 being the date that Sarpi first recorded it in his scientific diary. Of itself it is actually quite an ingenious idea. If the earth is actually moving as stipulated in a heliocentric hypothesis then wouldn’t the water on the earths surface slop around like water in a bowl being carried by someone and might this not be the explanation for the tides? If this were the case it would indeed be a solid empirical argument in favour of heliocentricity. In fact John Heilbronn in his Galileo biography dates Galileo’s conversion to Copernicanism to 1596 and the formulation of this theory.
As already stated above this theory has a major problem it was empirically refuted. As formulated by Galileo/Sarpi there would only be one tide a day but as every coastal inhabitant knows there are two. There is also another problem for this theory there already existed a better theory to explain the tides, a theory that we now know to be true; they are caused by the moon. The correlation between the tides and the phases of the moon had already been observed in antiquity but as every scientist knows, or should know, correlation does not equal causation and there was no known explanation as to how the moon could cause the tides. Newton was not even a blip on the horizon at this time.
Now Gopnik argued in his two essay’s that Galileo stuck to his refuted theory of the tides because the much more rational alternative smacked too much of magic for him to accept it, action at a distance would prove a difficult point even for Newton. Gopnik has solid evidence for his claim this is exactly the argument that Galileo brings for rejecting the lunar tide theory in book four of his Dialogo. So Gopnik is right? I don’t think he is. I think Galileo is being deceptive.
Galileo is convinced that his theory of the tides can deliver the empirical proof he so desperately needs for heliocentricity. If it were true then it would indeed the only such proof he has to offer. All the other arguments he marshals in his masterpiece are suggestive that heliocentricity might be a viable alternative but none of them is a proof or anything remotely like it. He needs his theory of the tides. He spent thirty years trying to cure its very obvious defect and failed but he is still not prepared to abandon it. Now as already pointed out there existed a much better empirically based explanation for the tides the lunar theory, one that for example Kepler backed. It of course lacked an explanatory mechanism but one could set up a research programme based on the concept of attractive forces, an idea that Kepler was already playing with, to find that mechanism, which is of course exactly what Newton did. If however Galileo accepted the greater plausibility of the lunar tide theory then his only “proof” of heliocentricity was down the drain so we look for a reason to reject it. In doing so he of course rejected the possibility of following Kepler down the path of considering forces controlling the solar system; a path that interestingly Galileo’s pupil Borelli took.
To summarise I don’t think that Galileo rejected the lunar tide theory because he thought it was magical. I think he rejected the lunar tide theory because it posed a serious threat to his own mechanical tide theory the only supposed proof that he had for a geocentric astronomy and having rejected it he looked around for an excuse to justify his rejection. There are lots of other examples of contemporary natural philosophers and astronomers employing the same tactic, Copernicus, Kepler, Newton and Galileo himself but to detail them here would make an already long post even longer. I shall save them for another post on another occasion.
As we have seen far from being unafraid to admit that he was wrong on three separate and highly significant occasion, significant for the evolution of science that is, not only did Galileo refuse to admit that he was wrong although he knew that he was but he actually did his best to bamboozle people into believing that he was right.
The Renaissance Mathematicus 
Glad to see you finally got around to discussing Galileo’s tide theory. A few comments and questions.
(1) Is there any evidence that Galileo read Astronomia Nova carefully or far enough to have seen Kepler’s 1st and 2nd laws? Using only the Dialog, all one can tell is that he read the Intro.
(2) According to Donahue’s translation of the Astronomia Nova, Kepler’s Intro was reproduced as an addendum to the Latin translation of Galileo’s Dialog. Is it known whose idea this was?
(3) The Dialog has very little on contemporary geometrical astronomy; I can’t remember if Galileo even mentions epicycles. My own impression is that Galileo’s knowledge of, and interest in, this subject was rather shallow; he was really more of a physicist (obligatory anachronism apology).
(4) Actually, the Astronomia Nova does contain a lot of Kepler’s physics: see Stephenson’s Kepler’s Physical Astronomy, or Donahue’s Selections from Kepler’s Astronomia Nova. The Intro is the source for Kepler’s tidal ideas, which Galileo ridiculed. Are you referring to the stuff in Mysterium Cosmographicum, (e.g., Platonic solids) and in the Harmonice Mundi?
(5) It is worth noting that Kepler’s gravitational force was quite distinct from the forces he posited to determine the planetary orbits (see Stephenson or Donahue).
(6) Speaking of tides, the lunar theory had the same problem as Galileo’s with just one tide a day instead of two, but at least it explained the obvious correlation with lunar phases.
(7) Galileo’s tide theory is inconsistent with so-called “Galilean relativity”, as many historians have noted. Did you ever run across Harold Brown’s paper, “Galileo, the Elements, and the Tides”? Brown resolves the inconsistency in a remarkable way: he suggests that Galileo was even more of an Aristotelian than previously suggested, with different dynamical laws for air, earth, and water.
(8) You use the words “deceptive” and “bamboozled”. I take you believe Galileo fully realized his theory was wrong, and was counting on no one else to notice the defects.
Now, trying to discern the unstated motivations of historical figures is always a dicey business. The harshest assessment I’ve ever read of Galileo’s theory is in W. R. J. Shea’s paper “Galileo’s Claim to Fame: The Proof That the Earth Moves from the Evidence of the Tides”, BJHS 5:2 Dec. 1970, p.111-127. He rakes him over the coals for both the empirical flaws (making an revealing comparison with Francis Bacon’s writings) and logical problems. Nonetheless, he does not accuse Galileo of insincerity. Much the same can be said of Koestler’s Sleepwalkers.
An alternate interpretation: in 1596, as you say, the theory must have looked very appealing. Possibly it even converted Galileo to heliocentrism. If we put together wishful thinking, egocentricity, and the intellectual inertia that (as you point out) is no stranger to scientists, don’t we have a sufficient explanation for why Galileo clung to the theory long past its expiration date?
The author has not responded, but I for one do not feel that being deceptive and bamboozling necessarily means that Galileo knew that his theory was wrong. He could have been convinced of its basic rightness, but he could have been bamboozling to avoid having a real discussion since he was unsure of how to argue it convincingly. I am fairly sure I have done this on occasions.
One more comment: you wrote, “Kepler’s work on the planetary orbits was a much better and more convincing argument for heliocentricity than anything Galileo had produced.”
True enough from the standpoint of geometrical astronomy, but this has nought to do with the physical question of the earth’s motion. In other words, it doesn’t distinguish between Tychonic and heliocentric versions of Keplerian astronomy.
As you’ve pointed out before, Galileo doesn’t even mention Tycho’s theory in the Dialog. Maybe this was purely for polemical reasons. Or maybe it was partly, even chiefly, because of the physical focus of the Dialog.
Galileo also doesn’t mention any of the important geometrical advantages of heliocentrism. Here he passes up a polemical opportunity.
By the way, in (1) and (4) of my previous comment, I forgot to mention the Epitome along with the Astronomia nova.
Pingback: I’ve got your missing links right here (18 May 2013) – Phenomena: Not Exactly Rocket Science
Apologies for somewhat dim memory – I read a correspondence between Kepler and Galileo where one of them admitted that he was wrong, and (wrongly) agreed that the moon does in fact have oceans. I believe it was Kepler making the admission but am not certain.
It seems to me the article is arguing a strawman based on selection bias. It is a sound hypothesis that science and scientists in general are characterized by their willingness to admit that they are wrong. It is tested by the observation that science progresses, and much faster, than non-empirical areas. The individuals that joins such a process must implicitly show a willingness to change by admitting errors, or they won’t be successful.
However there is no statistics as such. And at the same time that those who have done science or met scientists know that they have to admit errors or mistaken judgment openly, we also see that there are exceptions. And yes, there are some areas within the process where resistance to change can be more productive (new areas, new hypotheses). Hence the basis for selection bias.
If the article is suggesting that scientists are not as knowledgeable on science history as they should be, it is correct. I think scientists can display an interest, but a) the process has long been set up to _conceal_ history by presenting minimal descriptions of what works (which is a flaw and people have started to act against it, for example by publishing what not works as well and/or the actual path taken to the results) and b) history of science is not important for success within science. A historian may be pissed and peeved about lack of interest, lack or importance et cetera, but I don’t think these historical examples move the hypothesis on science.
As for the actual example, and the opposite strawman that people impute from more or less flawed history, please have at it!
“It is a sound hypothesis that science and scientists in general are characterized by their willingness to admit that they are wrong. ” I think it probably depends on how closely associated they are with the theory. In general humans hate to lose things, and this is particularly true of arguments (or theories they themselves have come up with. Changes in science are often associated with the changing of the guard. I believe that Newton was another example of someone who would focus excessively on proving himself right and even on destroying his opponents.
@Michael Weiss:
“trying to discern the unstated motivations of historical figures is always a dicey business.”
Indeed, often scientists or other empiricists from the position of hindsight ask why individuals missed some easy pickings.
Myself, I was peeved by Newton promoting “absolute” space even if he must have known that galilean relativity would have worked as well. It wasn’t until someone mentioned that his texts are unreadable for modern readers as they are rooted in the society of the time that it clicked.
Presumably the then society had never abstracted space, but explicitly described it as for example the space between the table and the chair. It was after all Newton that introduced the concept, I take it. Newton then needed his readers to visualize an ‘absolute’ existing space first and foremost, everything else was secondary priority. His “absolute” space was likely not my “absolute space”.
And now I’m just catching up on Susskind’s -09 cosmology lectures on youtube. He similarly laments the fact that basic cosmology of the universe, including the fact that it prefers to expand and hence its age, is derivable by newtonian mechanics. He even claims Newton was on the edge of suggesting a non-static universe, that he had all the pins in a row (I haven’t checked that). But that perhaps Newton’s religious beliefs, however heretic, barred him from making the conclusion in public as (presumably) even eyeball constraints makes the universe really old.
Weinberg devotes a whole chapter in The First Three Minutes to wondering why it took so long for astronomers and cosmologists to look for the CMBR. (And then serendipity, incarnated as Penzias and Wilson, made the actual discovery.)
But at least Weinberg speaks as one who lived through it.
Pingback: Famous Astronomer 1 – Nicolaus Copernicus | Famous Astronomers
Pingback: Cosmos, Neil deGrasse Tyson, and Bruno | Darin Hayton