**Transcript**

A new book just appeared: A New History of Greek Mathematics, by Stanford Professor Reviel Netz, Cambridge University Press. Let’s do a book review.

It will be a critical review. The main theme will be the sciences versus the humanities. Note the title of the book: “a New History.” Netz’s “New History” represents the new humanities-centred dominance in the field. As opposed to the “old” histories written by more mathematically oriented people. In my opinion, “new” does not mean better in this case. And I will tell you why.

Let’s start by attacking a city. The enemy are hunkering down behind their city walls. We are going to have to scale the walls with ladders. How long should we make the ladders? The ancient historian Polybius has the answer:

“The method of discovering right length for ladders is as follows. … If the height of the wall be, let us say, ten of a given measure, the length of the ladders must be a good twelve. The distance from the wall at which the ladder is planted must, in order to suit the convenience of those mounting, be half the length of the ladder, for if they are placed farther off they are apt to break when crowded and if set up nearer to the perpendicular are very insecure for the scalers. … So here again it is evident that those who aim at success in military plans and surprises of towns must have studied geometry.”

Great stuff. But Netz gets it wrong, in my opinion. Here is how he concludes:

“And then, of course, we are supposed to apply – Polybius leaves this implicit – Pythagoras’s theorem.” (223)

I don’t think so. I don’t think that’s what Polybius intended.

Sure enough, you can solve for the length of the ladder using the Pythagorean Theorem, but that is a clumsy and inefficient way to do it. If you did this the modern way you would need to do some algebra followed by some calculation involving a square root. They didn’t have calculators on their phones back then, you know. Do you expect carpenters in the military to be able to calculate square roots by hand?

In fact, Polybius has already told you everything you need to know with his numerical example. If the wall is 10, the ladder should be 12, he says. But it scales! So what Polybius is really saying is that, whatever the height of the wall is, the ladder is always 20% longer than that. That’s all you need to know. No Pythagorean Theorem needed.

Those numbers are a rule of thumb. You can also do it more exactly if you want, according to Polybius’s more theoretical characterisation of the optimal length. But you don’t need the Pythagorean Theorem for that either. There’s a much better way, that you can easily teach to an illiterate carpenter in five minutes.

Draw an equilateral triangle, just as Euclid does in Proposition 1 of the Elements. Cut it down the middle. Now you have a right-angled triangle, where the base is exactly half of the hypothenuse. This corresponds precisely to Polybius’s rule: the distance along the ground is half the length of the ladder.

So now we have a scale model of what we want. The height down the middle of the equilateral triangle represents the city wall; the side of the equilateral triangle represents the ladder, and it is precisely half its own length from the foot of the wall, exactly as Polybius says it should be for optimal stability.

So if we are given that the height of the wall is for example 10 meters, then we divide the height of the triangle into ten equal parts. We take a blank ruler and mark those ten marks on it. Then we take this ruler, with this length unit, and measure the hypothenuse of the triangle. However many marks long it is, that’s how many meters our ladder needs to be.

Piece of cake. Easy to improvise in the field without any specialised knowledge or tools. While Netz is busy trying to teach his carpenters the algebra of quadratic expressions and how to extract square roots, I have already scaled his walls using my much quicker methods. That is what you get when you put humanities people in charge of mathematics.

So I wouldn’t trust Netz when it comes to mathematics, even when he says “of course,” as he does here.

Here is another example: Did you know that parabolas are pointier than hyperbolas? At least if we are to believe Professor Netz. This claim occurs in a discussion of Archimedes. Archimedes studied solids of revolution obtained by rotating a conic section around its axis. Here are Netz’s words:

“In the case of a parabola, this will be of a more pointed shape; in the case of the hyperbola, this may be more bowl-like.” (140)

This is BS. Parabolas are not “more pointed” than hyperbolas.

This is clear for example from the following fact: you can draw a hyperbola having any two given lines as asymptotes and passing through any given point. So in other words, you can draw a V, an arbitrarily pointy letter V, and then pick an arbitrary point inside that V, for instance a point super close to the vertex of the V. Then there is always a hyperbola that fits inside the V and that passes through the designated point. You can hardly get any pointier than that, now can you? Yet parabolas are nevertheless “more pointed”, somehow, Netz apparently believes.

By the way, this fact I just mentioned, about constructing a hyperbola within a given V (that is to say, with given asymptotes), that is Proposition 4 of Book II of the Conics of Apollonius.

Or is it? Here we have another interesting point. It seems that this proposition was actually not in the original version of the Conics. Because Eutocius, in late antiquity, needs this theorem at a certain point and he says he better prove it since it’s not in the Conics of Apollonius. But then in the text we have of the Conics, what we call Apollonius’s Conics today, this proposition clearly is there, with the exact same proof.

And in fact the standard text that we call Apollonius’s Conics today comes to us only through that very same author, Eutocius, who wrote a commentary on the Conics and also preserved the text at the same time. So it seems that Eutocius inserted this proposition into Apollonius’s original text, because he had noticed in other works that it was a useful thing to prove.

Netz describes this correctly, which is all the more reason why he should know that a hyperbola can be as pointy as you’d like, since this follows immediately from this proposition that he discusses at length.

But anyway, there is another kind of error here in Netz’s discussion of this. The point that this proposition of the Conics is an insertion by Eutocius — that insight, says Netz, is due to Wilbur Knorr, Netz’s predecessor as a classics professor at Stanford.

“No one noticed that prior to Knorr” (431-432), says Netz.

But that is not true. Wilbur Knorr was not the first to discover this. In fact, Knorr clearly says so in his own article, the very article cited by Netz, which Netz has evidently not read very carefully. Already in the 16th century, Commandino, in his Latin edition of the Conics, very clearly and explicitly made the exact same point as Knorr, using the exact same evidence and arguments. And this in turn was cited in a 19th-century German edition of the Conics, as Knorr himself says. So Knorr didn’t discovery anything except what people had already known for hundreds of years.

This is not such an innocent mistake. How are we supped to trust anything Netz says if he makes blatantly false statements that are clearly and unequivocally seen to be factually incorrect by simply glancing at the very article that Netz himself cites in support of his own claims?

But it’s even more problematic than that. Because it’s clearly not just a random mistake. It is an ideologically driven error. By saying that Stanford humanities professor Wilbur Knorr was the first to make this important scholarly discovery, Netz is obviously indirectly boosting the impression that his own claims are important and novel, since he too is a Stanford humanities professor.

Netz is not only saying that Wilbur Knorr was the first to discover this particular thing. He is implicitly saying that earlier generations of scholars missed important insights, and that only people like him — Stanford humanities professors — are true experts.

That is of course the point of the title of the book: A *New* History of Greek Mathematics. In the past everybody did it wrong, and we need people like Netz to finally do it right. There is indeed a lot of explicit posturing to this effect throughout the book.

Let’s look at another example of this. Let me read a passage where Netz is attacking Thomas Kuhn’s account of the history of astronomy. Thomas Kuhn wrote in the mid-20th century and he worked on the history of science even though his PhD was in physics. So that is exactly the kind of people Netz wants to denigrate. He wants to say that only specialised humanities professors, with their “new” histories, are actual experts in the field.

Here is what Netz says about Kuhn: “Like most nonspecialists, Kuhn supposed …” See? I told you. It’s not just that Kuhn was wrong. It is that Kuhn epitomises the kind of people (people with a PhD in physics, for example) who need to be eliminated from the field because they make so many hopelessly naive assumptions without even realising it. Anyway, let’s continue with the quote:

“Like most nonspecialists, Kuhn supposed that Aristotle was broadly canonical from the beginning and that although the ancients offered various astronomical variations, these had all to agree with the Aristotelian framework. … This is wrong. In fact, Aristotle was not canonized throughout most of antiquity; Greek philosophers were in continuous, ever-shifting debate; the very practices of astronomy went through several stages in antiquity before they became stabilized through the ultimate canonization of Ptolemy – and of Aristotle – in Late Antiquity.” (487)

Indeed, I agree with Netz that mathematicians and scientists would have ignored Aristotle. Netz says it very well:

“In the second century BCE itself, Aristotle was marginal even within philosophy, let alone to a scientist such as Hipparchus. It is quite likely that Hipparchus never even read Aristotle’s Physics.” (346) Reassessing ancient science in this light, “we come close to imagining a very Galilean Hipparchus” (347).

Yes, perfect, I agree. That is exactly what I have said before about ancient science as well. Go Team Netz on that one.

But what about poor Kuhn whom Netz uses as a punching bag? Was he really so stupid? No.

I went to my copy of Kuhn’s book on the Copernican Revolution to check Netz’s accusations, and here is what I found. Here is a quote from Kuhn’s book:

“The great Greek philosopher and scientist, Aristotle, whose immensely influential opinions *later* provided the starting point for most medieval and much Renaissance cosmological thought.” (Kuhn, Copernican Revolution, 78)

So Kuhn says exactly the opposite of what Netz accuses him of “supposing”. *Later* Aristotle provided the starting point of scientific thought. Not “from the beginning.” Later. Exactly as Netz himself argues.

Here is another quote from Kuhn’s book that says the same thing:

“Aristotle said a great many things which later philosophers and scientists did not have the least difficulty in rejecting. In the ancient world there were other schools of scientific and cosmological thought, apparently little influenced by Aristotelian opinion. Even in the late centuries of the Middle Ages, when Aristotle did become the dominant authority on scientific matters, learned men did not hesitate to make drastic changes in many isolated portions of his doctrine.” (Kuhn, Copernican Revolution, 83)

There is no way you can read this and say that “Kuhn supposed that Aristotle was broadly canonical from the beginning,” that is to say, from his own lifetime onwards. Kuhn clearly says the opposite.

Netz’s accusation is just slander. So it’s the same in both the Knorr case and the Kuhn case: Netz makes false assertions and then cites sources that clearly and explicitly say the exact opposite of what Netz alleges.

At least in these cases Netz bothered to provide references at all. More often he doesn’t even do that. He allows himself the licence to make assertions at will, which readers are supposed to accept on his authority alone. Consider for example the following rant about the alleged bias of some unnamed “past scholarship”:

“In past scholarship, this Babylonian achievement [in astronomy] was sometimes dismissed as ‘merely’ practical, the Babylonians unfavorably compared with the Greeks in that they did not produce a geometrical account of the sky, hence no physical model, so, unlike the Greeks, ‘not real science’. This is obviously an absurd special pleading, where one defines as scientific whatever it is that the Greeks do and then reprimands the non-Greeks for failing to be Greek. The Babylonian theory is in fact directly analogous to the Greek mathematical theory of music – whose scientific significance no one doubts.” (326)

Well, no wonder that we need a “new” history of Greek mathematics then, amirite? That darn “past scholarship,” you know, they couldn’t think straight back then because they were so biased in favour of the Greeks. Or why sugarcoat it, why not just come out and say it: They were all racist back then, weren’t they? Thank God we have proper humanities-trained experts like Netz at last to save us from all of that. “A New History of Greek Mathematics”. Basically code for: The First non-Racist History of Greek Mathematics.

Well, yes, the argument that Netz refutes is indeed idiotic. But what is this so-called “past scholarship” that allegedly made this idiotic and basically racist assertion that Babylonian astronomy is “not real science” because it’s not geometrical? Who ever said that? No one I ever heard of.

Maybe Thomas Heath? If Netz is the “new” history of Greek mathematics, then Heath’s famous book is obviously the old one, written more than a hundred years ago.

But no. I looked it up. Even old Heath explicitly uses the phrase “Babylonian science” with approval (History I, 8). Of course it was “science”. Perhaps Thales, in his travels, learned of “Babylonian science”, for example, Heath says (Aristarchus of Samos, 18), in exactly those words.

So who, then, is Netz arguing against, except straw men that he has made up to present himself as the anti-racist saviour? I don’t know.

But enough bickering about that. Let’s turn to a big issue of major interpretative importance.

According to Netz, “Thales and Pythagoras did no mathematics whatsoever” (17). According to Netz, earlier generations of scholars naively believed in such fairy tales because they blindly trusted a single source:

“My predecessor Heath and many historians – up until the last generation – gave credence to the view according to which Thales, and then Pythagoras, made lasting contributions to mathematics. This derives almost entirely from Proclus’s commentary, which, because of its overall sobriety, was taken seriously even for such obviously unfounded assertions.” (423)

First of all, it is not true that this “derives almost entirely from Proclus’s commentary.” It is disturbing that Netz makes this false and self-serving statement. Just read Heath, whom Netz names in this very rant. Read Heath’s chapter on Thales. Heath goes through the sources explicitly. There are several sources about Thales as a mathematician that predate Proclus. And several of those testimonies, as well as passages in Proclus, are explicitly attributed to various specific earlier authors.

So it is not the case that earlier generations of scholars uncritically and blindly relied “almost entirely” on a single biased source, as Netz dishonestly and falsely claims.

Let’s look at Thales and Pythagoras in turn. Let’s start with Thales.

I have spoken before about how the idea of Thales as the originator of formal geometry makes good sense. The way I told it was based on two theorems attributed to Thales.

The first theorem is that a diameter cuts a circle in half. I described how one can show that using a very neat proof by contradiction. The appeal, obviously, would not have been the theorem as such, but the realisation that that kind of thing can be established by a very elegant and satisfying type of reasoning, namely a rigorous argument based on paying careful attention to the definitions of concepts such as circle and diameter, and the remarkable power of proofs by contradiction for proving this kind of thing. That is exactly the same aesthetic that one finds on the first pages of almost any modern mathematics textbook in abstract algebra, for example: proofs of basic results driven by carefully formulated definitions and tidy proofs by contradiction. It makes sense that people would fall in love with this aesthetic that has stood the test of time, and it makes sense that it would have begun with a basic theorem such as that the diameter bisects a circle. Just as ancient sources suggest.

A second theorem attributed to Thales is that a triangle inscribed in a circle with the diameter as one of its sides must be a right triangle. It is natural to arrive at this insight by playing around with ruler and compass. And the aha-moment would then have been that one can prove such things. Make a rectangle, draw a diagonal, draw the circumscribing circle. Now you are in business. From playing with shapes, you have arrived at a proof of a universal truth. Pretty cool. It makes sense that the idea of proving geometrical theorems might have started with something like that, as some ancient sources suggest.

I told my own version of this story, but in broad outline something like that is a pretty standard and well-known point of view. But Netz acts as if he has never heard of any of that. He pretends that people who believe that Thales initiated geometry are simply blindly taking Proclus’s word for it without having thought it through at all. Netz says so explicitly. Here are his words:

“I suggest here that Hippocrates’s works were among the earliest pieces of Greek mathematics ever to be written.” (48)

Ok, so that’s Hippocrates, considerably later than Thales, famous for a very technical and detailed argument about the areas of lunes, a kind of shape composed from circles. This looks a lot more like a specialised piece of technical geometry from a quite mature geometrical tradition. It seems like a very odd and obscure place to start with geometry altogether. In reply to this, Netz says:

“This might seem surprising. Could mathematics emerge like that – springing forth from Zeus’s head? Would we not expect mathematics to emerge in a more rudimentary form? In fact, I think this is precisely how we should expect mathematics to emerge: from Zeus’s head, fully armed. What would be the alternative? … Of course the very first mathematical works in circulation would contain remarkable, surprising results. Why else would you even bother to circulate them? I suspect that the counterfactual is sometimes not sufficiently carefully thought through here. Just what would a more rudimentary piece of mathematics look like? Would it prove some truly elementary results, such as, say, the equality of the angles at the base of an isosceles triangle? Why would anyone care about such a treatise, proving such a result?” (48)

It is baffling that Netz allows himself to make this lazy argument, as if no one had ever though those things through. He states these rhetorical questions as if no one had ever thought of any of that. But of course people have thought about that and they have compelling answers to Netz’s questions.

I just told you what the alternative to Netz’s narrative is and why it would make sense. And I am not the first person to say this. But Netz is too lazy to engage with alternative views seriously, so instead he dishonestly says that no one has ever thought through any alternative to his view.

So that’s Thales. Netz rejects a plausible interpretation of the Thales testimonies in ancient sources by dishonestly mischaracterising as hopelessly naive any scholars who adhere to such views.

Now Pythagoras. “Heath … had three full chapters on the mathematics of Thales and Pythagoras!” (22), Netz says triumphantly, suggesting that this is proof that his “new” history of Greek mathematics is sorely needed.

Anyone who believes in Pythagorean mathematics is stupid, according to Netz, and for this he relies on a famous book by Burkert. Here is how Netz describes it:

“[Burkert’s book] Lore and Science in Early Pythagoreanism … was a more careful, professionalized classical philology, keen to understand the authors we read not as mere parrots, repeating their sources, but instead as thoughtful agents who shape and retell the evidence as suits their agenda. Pythagoras, under such a reading, crumbles to the ground: almost everything … comes to be seen as the making of later authors from Aristotle on. Never mind: the historians of mathematics went on as before.” (23)

We hear the ideological overtones here. Burkert is Netz’s kind people: he is hailed as “professionalized.” By contrast, “the historians of mathematics went on as before”. That is to say, the mathematically trained people working on history of mathematics were a bunch of fools who didn’t even realise what fools they were, and we would be much better off if “professionalized” experts such as Burkert and, presumably, Netz himself, would be given a monopoly on expertise status in the field.

I do not agree with this, neither in terms of content nor ideology.

Regarding Pythagorean mathematics, since Netz doesn’t go into any more depth, I will now analyse Burkert’s book itself, which Netz accepts as gospel truth. A book review within a book review! Here we go.

According to Burkert, “the apparently ancient reports of the importance of Pythagoras and his pupils in laying the foundations of mathematics crumble on touch”. Not that phrase: “the foundations of mathematics.” I am going to criticise Burkert, and I am going to say that Burkert makes a naive and anachronistic assumption about what “the foundations of mathematics” are. (For page references for the quotes from Burkert, see my Operationalism article.)

When Burkert speaks of “the foundations of mathematics,” he takes for granted the traditional view that a core pillar of Greek geometry was its Platonist detachment from the physical world. As Burkert says, “Greek geometry assumed its final form in the context of [Plato’s] Academy … after Plato had … fixed its position as a discipline of pure thought.”

Indeed, Burkert’s arguments against Pythagoras’s mathematical significance are really arguments that he did not advocate a proto-Platonist philosophy of mathematics. Burkert’s overall thesis is that “that which was later regarded as the philosophy of Pythagoras had its roots in the school of Plato.” And indeed he proves convincingly that there was a clear tendency to distort history in this way in Platonic sources that is not consistent with more reliable sources outside this tradition.

For example, Burkert shows that when Proclus mentions Pythagoras in his “catalogue of geometers,” and attributes to him “a nonmaterialistic procedure” in mathematics, this, unlike the rest of the catalogue of geometers, is not based on the highly credible Eudemus. Instead it is copied from Iamblichus, that is to say, from the biased Platonic tradition.

From this it does not follow, as Burkert tries to argue, that Eudemus did not mention Pythagoras as a geometer at all. It follows only that Eudemus in this place likely did not associate Pythagoras with proto-Platonic views. This is enough to give Proclus the motivation to supplement his account with phrases from Iamblichus, even if Eudemus had mentioned Pythagoras in the original.

Burkert also observes that “Aristotle [says] expressly of the Pythagoreans [that] ‘they apply their propositions to bodies’---bringing out the distinction, in this regard, between them and all genuine Platonists.” Eudemus and Aristotle are clearly much more credible than the much later, more biased, and less intellectually accomplished Iamblichus and Proclus.

Thus Burkert’s arguments that Pythagoras’s alleged proto-Platonist philosophy of geometry is a fabrication of biased sources are quite convincing. However, it does not follow from this that the Pythagoreans did not take a profound theoretical and foundational interest in geometry altogether.

Burkert conflates these two conclusions, because he sees no alternative path to theoretical mathematics than through Platonic-style abstraction and detachment from physical considerations. Burkert believes that early work on geometrical constructions “is still not doing mathematics for its own sake”; rather, the “discovery of pure theory” was a later “accomplishment,” in his words.

If you have followed what I have said in the past you know that I reject this. Burkert is naive to assume a dichotomy between constructions and “pure theory.” Constructions were not the opposite of theory, and hence the opposite of “the foundations of mathematics,” as Burkert erroneously assumes. On the contrary, constructivism *was* the foundations of mathematics.

Once we admit that possibility, there is every reason to think that earlier mathematicians, such as the Pythagoreans, could very well have made profound and foundationally sophisticated contributions, while at the same time rejecting Platonising tendencies in the philosophy of geometry.

Indeed, when going beyond his convincing case against Pythagoras the Platonist, to the more general case of trying to minimise the significance of Pythagoras and his followers in the history of geometry, Burkert find himself on the back foot. He is forced to try to explain away Aristotle’s compelling statement that “the so-called Pythagoreans were the first to take up mathematics; they advanced this study, and having been brought up in it they thought its principles were the principles of all things.”

Burkert’s thesis leaves him little choice but to dismiss the centrality of mathematics implied by this statement as “a psychological conjecture of Aristotle, which the historian is not obliged to accept.” That Proclus was wrong is plausible enough, but having to postulate that Aristotle was wrong comes at a considerably higher cost. And while Burkert was able to discredit Proclus’s mention of Pythagoras in the catalogue of geometers, he cannot deny that numerous attributions of mathematical discoveries to Pythagoreans made by Proclus are indeed based on Eudemus and hence credible, by Burkert’s own admission. Thus even Burkert must admit that “Pythagoreans made significant contributions to the development of Greek geometry.” Yet he hastens to add: “but the thesis of the Pythagorean foundation of Greek geometry cannot stand.”

Once again Burkert’s argument is based on tacitly assuming a monolithic conception of what “the foundations of Greek geometry” consisted in. The constructivist reading of Greek geometry problematises this assumption. It shows that one cannot simply take for granted that “the foundations of geometry” means what modern authors think it should mean. Constructivism offers an alternative vision, according to which much early Greek geometry may very well have been eminently foundational, but in a sense different from that commonly assumed by modern observers. This at the very least raises the possibility that early traditions such as that of the Pythagoreans may have been more foundationally significant than Burkert’s argument admits.

So much for Burkert, whose judgement Netz accepts unconditionally. Far from being an unequivocal triumph of “professionalized” expertise over previous naiveté, as Netz would have it, Burkert’s account is itself naive and by no means unquestionable.

So Netz is fond of dismissing what the ancient sources say. All the stories about Thales and Pythagoras, that’s just so much fiction. To be sure, the sources are highly imperfect and definitely contain a lot of misinformation. Nevertheless, it is surely better to try to save some meaning in these stories than to almost take it as a point of pride to dismiss as much of it as possible, as if the more sources you dismiss the more sophisticated a historian you are.

In fact, Netz continues in the same vein for later Greek geometry as well. “The stories [about Archimedes] probably are fabricated,” (128) we are told.

Stories such as Archimedes’s use of the principles of hydrostatics to detect a fake gold crown, because it did not have the right density properties. That is the “Eureka!” story.

“Biographers concoct anecdotes, based on the contents of the authors’ works. This is clearly the case here. The story of the crown is a clear echo of Archimedes’s study of solids immersed in liquids, On Floating Bodies.” (129)

Now, how would this work exactly? Let us “think through the counterfactual,” as Netz admonished others to do above.

Ok, so Archimedes wrote a sophisticated technical work on floating bodies. For some reason. Certainly not because of fake gold crowns and such things, because those are just “concocted anecdotes.” I guess Archimedes just woke up one day as said to himself: I think I will prove a bunch of theorems about hydrostatics, which nobody has done before, because I’m a mathematician and I just do things arbitrarily for no reason with no connection to the real world.

So he wrote a detailed, hyper-mathematical treatise on floating bodies. Theorem-proof, theorem-proof.

And then, maybe hundreds of years later or whatever, another guy told himself: Hey hey, I’m a writer! I’m going to write about the history of mathematics, but I won’t find out actual facts about the history of mathematics. Instead I’m going to pour over these extremely technical treatises that very few people can understand, and I’m going to master their content in great depth, to the point where I will be able to invent out of thin air real-world scenarios that involve realistic, sophisticated applications of the complicated technical results found in these treatises. And my goal in doing so is to concoct a one-paragraph anecdote about for example Archimedes making a discovery in the bath that made him run naked through the streets. Haha, what a funny image to imagine him running and screaming eureka like that. Totally worth all those probably hundreds of hours that I had to spend studying very complicated mathematics and then designing and working out my own research-level applied mathematics problem just so that I could make this little joke about Archimedes running from the bath.

Well, that’s apparently what happened if we are to believe Netz.

I very much doubt that story tellers were ever that good. The story about Archimedes and the crown is really very good scientifically. The connection with the technical details of Archiemdes’s treatise is the real deal. If this is a “fabricated” anecdote “concocted” by a biographer, as Netz says, then that biographer was not only a story teller but one of the leading scientists of their age.

Look, I teach calculus regularly, and I always try to get students to think about the physical meaning of mathematical notions and interpret results in the context of a real-world scenario. And I can tell you that that is an uphill battle to say the least.

I don’t think Netz teaches calculus so I think he underestimates how hard it is to make up stories that simultaneously make perfect scientific sense.

It is quite easy, on the other hand, to make up stories that do *not* make scientific sense. And that bring us to another one of Netz’s theories.

Netz has another book called “Ludic Proof”. Ludic as in play, playfulness. According to this theory, mathematicians borrowed stylistic approaches from poets. Poets had a fondness for cleverly constructing narratives that led to surprising twist reveals. Mathematicians shared the same aesthetic, according to Netz.

Netz, in all seriousness, proposes that this could be the main reason why Archimedes did calculus-style calculations of areas at all, and why he even turned to mathematical physics at all. The root cause is supposed to be not ordinary scientific or mathematical motivations, but Archimedes’s desire to do mathematics in the style of the poets: mathematics was “written, always, against the background of wider literary currents, emphasizing subtlety and surprise” (218).

According to Netz this is why Archimedes did calculus-style calculations of areas and volumes:

“Archimedes … picked up a particular technique, first offered by Eudoxus, because its subtlety … made a certain kind of surprise especially satisfying. Hence the infinitary methods.” (218)

And this is also what made Archimedes apply mathematics to physics:

“[Archimedes] saw the possibilities of applying geometry to a seemingly unrelated field – the study of centers of the weight in solids … – because there was a particular payoff of subtlety and surprise to be obtained by the bringing together of apparently irreconcilable, maximally distinct fields of study. This was rather like Callimachus’s poetry! Hence the mathematization of physics.” (218)

So there you go, calculus and mathematical physics are just side effects of mathematicians pursuing their true goal, which was to imitate the poets. That is some tin-foil-hat level of crackpottery, in my opinion.

It is one thing that Netz previously advanced his bizarre theory in a specialised monograph. Of course it must be possible for scholars to try out unconventional ideas. But to put this crazy stuff in a survey history with a straight face, as if this was objective information that any beginner in the field needs to learn, that is quite irresponsible, in my opinion. Certain chunks of this book are not an introduction to the history of Greek mathematics, but an introduction to the pet theories of Reviel Netz that no one but him believes.

Let’s look at some specific mathematical examples that are allegedly all about surprise, according to Netz.

For example, Archimedes found the area of one revolution of the Archimedean spiral. How do you think he’s going to prove this? Well, you have probably already seen how Archimedes found the area of a circle. Naturally readers of his more advanced treatise on spirals would already have read his more basic treatise on the circle.

Archimedes found the area of a circle by cutting it into wedges, as it were. Equal-angle pizza slices all the way around.

Naturally it makes a lot of sense to try the same idea for the spiral. The Archimedean spiral is like a circle but with a linearly growing radius. In polar coordinates, the radius r is proportional to the angle theta.

So when we apply the method we used for the circle to the spiral we get a bunch of equal-angle wedges that gradually get bigger and bigger. The radius grows linearly with the angle, so the radii of the wedges form an arithmetic progression. For every equal increment of the angle, the radii increase by the same amount, let’s say alpha. And the Archimedean spiral starts with radius zero, so the radii go: alpha, 2 alpha, 3 alpha, etc.

To get the area of the spiral we have to add up all the wedges. Obviously the areas scale like the square of the radii. Linear scaling of distances means square scaling of areas. So since the radii went alpha, 2 alpha, 3 alpha, the areas will be proportional to alpha^2, (2 alpha)^2, (3 alpha)^2, and so on.

So to get the area we have to add up a series of squares, the squares of numbers in an arithmetic progression. Indeed, Archimedes has a theorem that does exactly this. That is his Proposition 10.

Did you find any of this “surprising”? Hardly. It was a predictable extension of the idea used for the circle. And the trick of getting a complicated area or volume by an infinite series sum of simpler components is also a well established trick. Archimedes used the same trick for the area of a parabolic segment, for example, and Euclid used it too, for example for the volume of a tetrahedron. The sum of a geometric series was the key ingredient in those cases, and now for the spiral we need the same kind of theorem but for the squares of numbers in an arithmetic progression. Very predictable and business as usual for a Greek geometer.

But Netz doesn’t think so. According to Netz, the reader of Archimedes’s treatise is not supposed to have been able to see those things and instead they are supposed to have been baffled by the introduction of Proposition 10, that is to say, the sum of the series. They are not supposed to have been able to realise that this series is obviously the same kind of area calculation by series that had been well-known at least since Euclid, and that the particular terms of the series obviously correspond to the most natural way of cutting up the spiral area.

Here is what Netz says:

“Archimedes aims at surprise. The key point is that as proposition 10 is introduced, Archimedes makes all efforts to disguise its potential application. … The key observation – that the sectors in a circle behave as the series of squares on an arithmetical progression – is not asserted in advance. Instead, the application of proposition 10 is postponed and revealed only at the very last minute when, introduced in the middle of proposition 24, it finally makes sense of the argument. … Everything is designed for the sake of this denouement where, finally, the narrative of the treatise would make sense in a surprising turn. Ugly, misshapen proposition 10 is really about sectors in spirals: the duckling was a swan all along!” (149)

I think this is nonsense. I don’t think Archimedes’s readers would have been surprised at all by any of this.

Today we teach our mathematics students: when you read a theorem, before you look at the proof, take a few minutes to think about how you would prove it. Then when you read the proof you will understand it much better. You will know which parts are easy and obvious, because you have already thought of those yourself. And you will appreciate the difficult parts because you have realised when trying to prove it yourself that certain steps would have to involve some real work.

I bet Archimedes’s readers did the same. They get a treatise by Archimedes, a key result of which is the area of a spiral. Indeed, the treatise comes with a prefatory letter by Archimedes himself where he highlights the key results, so obviously you know where it’s heading. You don’t just start reading cold from A to Z.

And if you follow the elementary advice that we teach all our undergraduates, without which you will never get far in mathematics, to try to prove it yourself before reading the solution, then you will very quickly realise that the obvious approach is to cut the spiral area into wedges and sum the components, which will obviously lead to a series of squares of numbers in an arithmetic progression. So when you get to Archimedes’s Proposition 10 you will be far from surprised. On the contrary, you knew all along that he would have to do this sum.

Let’s look at another example of a so-called “ludic proof.” If you point a parabolic mirror at the sun, all the rays are reflected toward a single point, the focus of the parabola. Diocles proved this, and the “ludic” part is that he first proved some properties of tangents and normals of a parabola, and only then introduced a line parallel to the axis, which represent the rays of the sun. Surprise! It was about rays of the sun all along. Who could ever have guessed that saying something about the tangent first would be relevant to this! Except of course someone who has read the title of the treatise and has basic mathematical competence.

Here is how Netz describes it:

“[Diocles’s proof of the focal property of the parabola is] palpably Archimedean. The same emphasis on subtle surprise – down to the intentional delay in the construction of the parallel line, so that, throughout the argument, we do not yet see the relevance of any of it for the optics of rays of the sun.” (215)

So the “surprise” is that basic properties of the tangent of the parabola are relevant to the optics of rays of the sun. What a shocking reveal! Since the solar ray had not been drawn yet, there is no way we could have known this, according to Netz.

Once again, any mathematically competent person who looks at this problem for five seconds will realise that of course it is going to involve the tangent. The notion that mathematically competent readers would not have been able to see the relevance of theorems about tangents for the optics of rays of the sun is ridiculous. And yet that notion is the corner stone of Netz’s ludic proof interpretation of this episode.

There is another bit of nonsense here as well. Diocles talks about the tangent of a parabola, but Archimedes also talked about the tangent of a parabola. Aha! Therefore Diocles’s proof “is really a brilliant variation on an Archimedean theme” (215), in Netz’s words.

This is a way of thinking that perhaps makes sense in literary history. Poets and playwrights like to draw inspiration from earlier masterpieces and rework their themes in a new way. Netz tries to do the same thing for mathematics, but in my opinion the results are nonsensical.

What Netz is saying is like saying that if Person A gives a mathematical argument involving the derivative of a quadratic function, and then Person B gives a completely different argument that has nothing to do with the first one except that it too involves the derivative of a quadratic function, then Person B’s argument is a variation on Person A’s theme.

That’s rubbish. Of course tangents of parabolas show up regularly in mathematics. That doesn’t mean that anyone who talks about the tangent of a parabola is subtly reworking what earlier authors have done. That may be how literature works, but it is not how mathematics works.

So, in this case as in so many others, Netz’s “new history” is what you get when you look at Greek mathematics through eyes attuned to the humanities but not to mathematics. Indeed, Netz’s description of the mathematics is factually wrong as well. Archimedes and Diocles both state the tangent theorem in terms of “the intercept between tangent and ordinate” (215), according to Netz. No, that’s not right. It’s the intercept between the tangent and the axis. Not ordinate, axis.

But it is not my goal to catalogue all the mathematical errors in Netz’s book. If you take a humanities professor as your guide to mathematics then you have only yourself to blame anyway.

]]>**Transcript**

The discovery of non-Euclidean geometry in the early 19th century was quite a wake-up call. It showed that everybody had been a bit naive, you might say.

Here’s an analogy for this. Suppose we had all been speaking one language, let’s say English. And we were all convinced that English is the only natural language. In fact, that question didn’t even arise to us; we simply assumed that English and language is the same thing. We even had philosophers “explaining why” English is a priori necessary. These philosophers had “proved,” they thought, that without English the very notion of linguistic communication or thought is impossible.

And then we discovered that there are French speakers and Chinese speakers. Oops. Very embarrassing. English is not necessary after all. It is not innate, it is not synonymous with language itself. For thousands of years we made those embarrassing mistakes because we were not aware of the existence of other languages.

That’s how it was with geometry. What I said about English corresponds to Euclidean geometry. For thousands of years, nobody thought of Euclidean geometry as one kind of geometry. Everybody thought of it as THE geometry. Geometry and Euclidean geometry was the same thing. Just as an isolated linguistic community thinks their language is THE language.

And the philosophers I spoke of, Kant is an example of that. He argued that Euclidean geometry was a necessary precondition for having spatial experience or spatial perception at all, which is like saying that English is necessary for any kind of linguistic expression.

And already long before Kant, many people had been convinced by how intuitively natural and obvious the axioms of Euclidean geometry feel. Descartes for instance and many others made a lot of this fact. Remember how important it was to Descartes that our intuitions were truths implanted by God in our minds.

Well, we all think our native language is intuitive. And we think other people’s languages are not intuitive. This feeling is so strong that we think it must be objective. When we try to learn a foreign language, it feels impossible that anyone could think that was intuitive. And yet they do.

So apparently our intuitions can deceive us. We feel that our native grammar is much more natural than everyone else’s, but that’s a delusion. It felt like an objective fact, but it turned out to be subjective.

Could it be the same with geometry? Could the alleged naturalness and intuitiveness of Euclidean geometry turn out to be just an arbitrary cultural bias, like thinking English feels more natural than French?

So, ouch, we took quite a hit there with the discovery on non-Euclidean geometry. It exposed our insularity. It showed that things we had thought we had proven to be impossible were in fact perfectly possible and every bit as viable as what we had thought was the only way to do geometry.

And yet there is hope as well. The language analogy doesn’t just expose what an embarrassing mistake we made, or how non-Euclidean geometry hit us where it hurts. The language analogy also suggests a way out; a way to rise from the ashes.

We were wrong about the specific claim that Euclidean geometry is innate, because that’s like saying that English is innate. Nevertheless we were right too, one could argue.

Language is impossible without something innate. Everybody learns their native language with incredible fluency. Every child learns it somehow, with hardly any systematic teaching; they just pick it up naturally. And they do so at a very early age, when their general intelligence is still very limited.

Meanwhile, no animal comes anywhere near achieving the same feat. Nor any adult human for that matter. I’m better than a three-year-old child at any intellectual task, except learning French. Somehow the child is super good at that.

So clearly something about language is innate. The ability of language acquisition is innate. Some kind of general principles of language are innate.

Perhaps geometry is like language in this way. We have some innate geometry. Not specific Euclidean propositions or axioms maybe, but some kind of “geometryness” nonetheless. Some sort of more structural or general principles of geometry than specific propositions, just as our innate linguistic ability doesn’t contain anything specific to any one language but instead has to do with “languageness” in general.

In linguistics, this is called “universal grammar.” Every language has it own grammar, of course, but there are some more general or structural principles of language that are the same for all human languages. This common core is the universal grammar.

Here’s an example of such a principle that belongs to universal grammar. Consider the statement “the man is tall.” It is a declaration, an assertion. You could turn it into a question: “Is the man tall?” We turned the assertion into a question by moving the “is” to the front of the sentence. That’s how you make questions from statements: start with assertions—“the man is tall”—and move the “is” to the front—“is the man tall?” So that’s a recipe for making questions.

But consider now the statement “the man who is tall is in the room.” How do you form the corresponding question? There are two “is”’s in the sentence. So the rule “to form a question, move the is to the front” is ambiguous. Which of the is’s should we move?

It could become: “is the man who is tall in the room?” That works. But if we move the other is to the front we get nonsense: “is the man who tall is in the room?” Well, that didn’t work. It didn’t make a question, it just made gibberish. Even though we followed the same rule as before: move the “is” to the front.

If language was nothing but a social construction then it would be perfectly reasonable for children trying to form questions to come up with the gibberish one. If the child simply extracts general rules from a bunch of examples, it would have been perfectly reasonable for the child to have guessed that the general rule is: move the first “is” to the front of the sentence to form a question. Which would lead to the nonsense question “is the man who tall is in the room?”

In fact, however, “children make many mistakes in language learning, but never mistakes such as [this].” Apparently, “the child is employing a ‘structure-dependent rule’” rather than the much simpler rule to put the first “is” in front. Why? “There seems to be no explanation in terms of ‘communicative efficiency’ or similar considerations. It is certainly absurd to argue that children are trained to use the structure-dependent rule, in this case. The only reasonable conclusion is that Universal Grammar contains the principle that all such rules must be structure-dependent. That is, the child’s mind contains the instruction: Construct a structure-dependent rule, ignoring all structure-independent rules.” So although each language has its own grammar, there are some general principles like that that are universal: common to all languages. Those principles are hard-wired into the mind at birth.

This stuff about there being an innate “universal grammar” is the view Chomsky, the leading 20th-century linguist. The example and explanation I just quoted are from his book Reflections on Language. There are many debates about Chomskyan linguistics, but I’m going to assume Chomsky’s point of view for the purposes of this discussion, because its parallels with geometry are very interesting.

You might say that this view of language is Kantian in a way. We saw that Kant put a lot of emphasis on the necessary preconditions for certain kinds of knowledge. Geometry, for example, is not an external, free-standing theory that we can analyze with our general intellectual capacities. Rather, the fundamental concepts of geometry are bound up with the very cognitive structure of our mind itself.

Some things are purely learned through experience and convention, such as how to play chess or how to dance the tango. But some things are not like that. For instance, they way we experience color. The mind is made to see red and blue and to not see infrared and so on. That’s a fixed, domain-specific property of how our mind works. You can neither learn nor unlearn that through general intelligence. Color experience is just one of those basic things hardwired right into the brain.

From the Chomskyan point of view, language is like that as well. Language is not merely a social construct with man-made rules like chess or tango. Nor is it explicable in terms of general intelligence only.

Chess or tango you can learn by general intelligence. That is to say, if you spend enough time looking at people playing chess or dancing, you can eventually figure out what the rules are by general rational thinking such as pattern recognition, and forming preliminary hypotheses about how you think it works and then observing some more to check if you maybe need to revise the hypothesis to take into account some other possible circumstances or cases.

Color experience is not like that. You don’t learn to experience redness by watching other people. It just is. And if you’re not born with it, then you can’t learn it by general intelligence, like you can learn chess.

Language is similar to color and not similar to chess. You don’t learn color perception by watching others and using general intelligence to figure out the patterns and rules. General intelligence is not sufficient to sustain such a thing.

Many people overestimate the potential of general-purpose intelligence. Both Kant and Chomsky agree about this. Remember the tile of Kant’s work: a critique of pure reason. “Pure reason,” or general-purpose intelligence, is not by itself capable of generating human linguistic capacity or geometric experience.

The capacities of our mind depend much more than people realize on domain-specific conceptions. It is obvious that color experience is a hardwired specific domain of our cognitive structure and isn’t merely the outcome of some pattern-recognition process of general-purpose intelligence. But it’s less obvious that geometry is like that, or that language is like that. But Kant and Chomsky maintain that they are. According to them, we underestimate the extent to which basic geometrical and linguistic conceptions are intertwined with the very nature of our mind and our cognitive capacities.

So the wrong way to think about it would be like this. The human brain is a general-purpose thinking machine. Imagine a person in a prehistoric hunter-gatherer society. This person’s general-intelligence mind might think to itself: Well, it’s great that I’m so smart. I can learn many things, like which plants are poisonous; I can figure things out like how to make fire, how to use tools and so on. But gee, wouldn’t it be handy if I could communicate my thoughts to others. Then we could organise collaborations, learn from each other’s experiences, and so on. I know, let me invent language, that will work for this.

From the Chomskyan point of view this story is wrong because it overestimates the general-purpose mind. In fact, note that I described what the pre-linguistic mind was thinking by using language. But I was talking about a hypothetical stage in history in which there was no language. Does it even make sense to imagine such a thing as thought without language? No, according to Chomsky. The very nature of thought itself cannot be separated from language like that.

The story of the hunter-gatherer inventing language is no more plausible than the story that he invented color experience by discovering that certain wavelengths of electromagnetic radiation were associated with grass, others with fruit, and so on.

Instead of thinking of the mind as starting from general-purpose intelligence and then inventing domain-specific things like color and language, we should perhaps think of it exactly the other way around. The mind is made up of the domain-specific skills. Those are the fundamental cognitive starting-points. Insofar as we have any general-purpose intelligence, that comes from piecing together the domain-specific skills. Not the other way around.

From an evolutionary point of view, the human mind perhaps evolved by adding domain-specific modules one by one: first color, then a hundred thousand years later geometry, then a hundred thousand years later language, and so on. We don’t have general-purpose intelligence. We only have the sum of our modular parts. But eventually these modules became so advanced, and combined in such fruitful and powerful ways, that we fool ourselves into thinking that we have general intelligence, “pure reason.” But at bottom our precious “pure reason” actually still depends more than we realize on domain-specific preconceptions hardwired into our cognitive capacities. That’s what Kant said about geometry and that’s what Chomsky said about language.

So in this way we can “save Kant.” The discovery of non-Euclidean geometries was a blow to Kant’s idea of the innateness of geometry. Kant associated the intuitiveness of Euclidean geometry with its innateness. But native languages are intuitive, yet they are not innate. And geometry could be the same, because just as there are many languages there are many geometries. This shows that intuitive and innate is certainly not the same thing, so it calls into question the Kantian story that the mind is constrained by pre-programmed conceptions.

We save Kant with the rebuttal that in fact language too is innate after all. Even though there are many languages that all differ in fundamental respects, nevertheless there is some universal languageness that is common to all and without which language learning would be impossible in the first place.

Same with geometry. Instead of focusing on the differences between Euclidean and non-Euclidean geometries and concluding from this that no one geometry could be a necessity of thought, we should instead focus on the more fundamental and structural preconceptions common to all geometries, without which any kind of geometry would be unthinkable at all.

Or we can put it like this. Thought presupposes language. When you think, you think in terms of words and sentences. Of course thought does not presuppose any specific language. You can think the same thing in English or German. Nevertheless thought does presuppose that you use some language. There is no “pure thought,” or hardly any, that does not involve words.

It’s funny: thought cannot exist without language, yet you can switch the entire language and still have the same thought. So there’s both dependence and independence.

Kant says basically the same thing but for geometry. You can’t have spatial perception or spatial reasoning without geometrical presuppositions. Just as you can’t think without presupposing some language, so you can’t geometrize without presupposing some geometry.

The choice of which language or which geometry you take as the basis for thought is arbitrary. As Kant says, it’s a synthetic a priori, not an analytic a priori. That is to say, it is not logically necessary that we must use Euclidean geometry as the presupposition for all our spatial experience. But it is necessary that we must make some such presupposition.

Remember, as Kant said, we don’t have direct access to objective physical reality. We only know the outside world through perception which is always necessarily interpreted. The presuppositions of that interpretation are arbitrary—in fact, it’s arbitrary in two ways one might say: one good and one bad. It’s arbitrary in a “bad” way in that it is subjective. It lacks objective justification. But it’s also arbitrary in a “good” sense, namely that it doesn’t necessarily matter all that much which interpretation we choose.

Just like language. It is arbitrary that I’m speaking English. There’s no objective or logical reason for why English is any better than any other language. But it’s also arbitrary in that it doesn’t matter. I could have said the same things in some other language. And in fact it’s only because of my choice of some arbitrary language that I am able to say anything at all.

Same with geometry. Our minds think in terms of Euclidean geometry even though that has no absolute logical justification. Yet it would be a mistake to criticize this as arbitrary subjectivity. Because it is only because I have some geometrical preconceptions at all, no matter how subjective, that I am able to reason spatially and have spatial perception and experience in the first place.

The analogy that geometry is like language is suggestive in other respects as well. Here’s one interesting question. When a child is learning their native language by picking up the speech of their parents and their environment, how does the child know which sounds are language and which sounds are other kinds of noises? It’s a pretty difficult problem, isn’t it?

Suppose you had to program a computer to detect and recognize speech. What criteria could you define by which the computer could tell if any given sound is linguistic or not? Words come in many forms: you can scream them, whisper them, sing them. Those are very different as sounds, but somehow you have to be able to tell that they are all words. And you have to be able to tell that other sounds are not linguistic, such as a doorbell, a barking dog, the sizzling of a frying pan, and so on.

You have the same problem in geometry. Among all the sensory impressions we are bombarded with every second, which ones should be regarded as geometrical, and which not? If geometry is like a language, a child must have some criteria by which to answer this. Just like the child somehow picks out linguistic sounds from the environment and lets that shape their native language, so also the child must pick out geometric features of the environment and let that shape their native geometry. This is how their intuitive geometry can become either Euclidean or non-Euclidean depending on the environment, just as their native language can become English or Russian or whatever.

So: What parts of all our sensory impressions have to do with geometry? You must know that first, before you can start thinking about whether those impressions are Euclidean or non-Euclidean.

Poincaré had a very elegant solution to this problem. Here’s his criterion for telling geometry from non-geometry. It goes like this: Among all sensory impressions, those are geometrical that you can cancel through self-motion.

Let me explain what this means by an example. I have a piece of paper. One side is white and the other side is red. I hold the paper up with the white side facing toward you. Then I rotate it so that the red side is facing you. This is a geometrical transformation: it has to do with rotation, with position. You know that it was geometrical because you could walk around and stand on the other side and then you would see the white side of the paper again. So you could cancel the transformation in impressions, you could restore the original sensory impression, through self-motion. By moving yourself. Not by manipulating the environment, but only by moving around in it.

There are many transformation of sensory impressions that are not like that. That are not cancelable or reversible through self-motion. Including other kinds of switches from white to red. Pour a white liquid, like a lemon sports drink, into a glass. And then pour in something very red, like beet juice or some strawberry syrup. The liquid in the glass went from white to red, just like the paper did when I flipped it over.

But the liquid is different, because you can’t cancel it this time by moving around and looking at it from another point of view. This is precisely why it is not geometrical. The paper example should be interpreted in terms of geometry. If someone asks: what happened? Then for the paper example you would give an explanation in geometrical terms: the object rotated 180 degrees. But for the liquid example you would give an explanation in non-geometrical terms: the red liquid “colors over” the white one by some kind of, I don’t know, chemistry somehow; not geometry anyway.

So there you have a very clear criterion for selecting from the environment which things are to be accounted for in terms of geometry and which not. Cancelability through self-motion.

Before a child can tell if their parents speak French or Russian, they must be able to distinguish which sounds are linguistic at all. And before we can tell if the space around us is Euclidean or non-Euclidean, we must first be able to distinguish which sensory impressions have to do with geometry at all. Poincaré’s criterion in terms of self-motion answers this problem.

So this suggests that it is only through motion that we can impose a geometric interpretation on our visual impressions. It may feel to us as if our sense of sight is inherently geometrical: geometry is visual, it lives in the eyes. But Poincaré’s perspective suggests that it’s more complicated than that.

Vision becomes endowed with geometry only through its interaction with self-motion. If we could not move ourselves or our eyes, our sense of sight would be as un-geometrical as our sense of taste or smell. It would be just a bunch of qualitative impressions with no particular structure.

With sense and smell, you can tell when one thing is different from another, but you can’t do much more than that. There is no “Pythagorean Theorem of taste” that allows you to calculate the taste-distance between wine and beer if you know the distances between beer and water and water and wine. Taste impressions don’t have geometrical structure or any comparable kind of structure. And if we didn’t have self-motion then sight would be like that as well.

There’s a passage in Rousseau’s Emile that fits this perspective. It goes like this:

“It is only by our own movements that we gain the idea of space. The child has not this idea, so he stretches out his hand to seize the object within his reach or that which is a hundred paces from him. You take this as a sign of tyranny, an attempt to bid the thing draw near, or to bid you bring it. Nothing of the kind, it is merely that he has no conception of space beyond his reach.”

So imperfect capacity for self-motion goes with imperfect understanding of space, it seems, in the child. Of course Rousseau was writing long before Poincaré. I used Poincaré as the point person for this perspective about the role of self-motion in geometry but indeed the basic ideas go back centuries before. Poincaré explains his view very well in his book La Valeur de la Science of 1905. But that’s the culmination of a tradition of more than two centuries.

For example, many philosophers had debated the following question: Suppose a person who has been blind all their life has an operation that makes them able to see. Can they then, from visual impressions alone, tell for example a cube from a sphere? They already knew the difference by touch, but could they then automatically make the connection between that and sight, or would they have to learn to recognize things by sight through experience?

This is the so-called “Molyneux’s question.” Molyneux raised it in 1688. Obviously it has a lot to do with the question of whether geometry is innate, or whether it is learned by experience.

This thing about a blind person becoming sighted was not just a thought experiment. It could be done through surgery in some cases. Let me read to you a report of the experiences of such a person. This is from the Philosophical Transactions of 1728. A boy who was 13 years old and had been blind all his life got his sight back through a surgical procedure. And his reactions were as follows.

“When he first saw, he was so far from making any Judgment about Distances, that he thought all Objects that he saw touch’d his Eyes, (as he express’d it) as what he felt, did his Skin.”

“He knew not the Shape of any Thing, nor any one Thing from another, however different in Shape, or Magnitude; but upon being told what Things were, whose Form he before knew from feeling, he would carefully observe them, that he might know them again; but having too many Objects to learn at once, he forgot many of them. One Particular only (tho’ it may appear trifling) I will relate; Having often forgot which was the Cat, and which the Dog, he was asham’d to ask; but catching the Cat (which he knew by feeling) he was observ’d to look at her stedfastly, and then setting her down, said, So Puss! I shall know you another Time.”

“He was very much surpriz’d, that those Things which he had lik’d best, did not appear most agreeable to his Eyes, expecting those Persons would appear most beautiful that he lov’d most, and such Things to be most agreeable to his Sight that were so to his Taste.”

“We thought he soon knew what Pictures represented, which were shew’d to him, but we found afterwards we were mistaken; for about two Months after he [became sighted], he discovered [that] they represented solid Bodies; when to that Time he consider’d them only as Party-colour’d Planes, or Surfaces diversified with Variety of Paint; but even then he was no less surpriz’d, expecting the Pictures would feel like the Things they represented, and was amaz’d when he found those Parts, which by their Light and Shadow appear’d now round and uneven, felt only flat like the rest; and ask’d which was the lying Sense: Feeling or Seeing? Being shewn his Father’s Picture in a Locket at his Mother’s Watch, and told what it was, he acknowledged a Likeness, but was vastly surpriz’d; asking, how it could be, that a large Face could be express’d in so little Room, saying, It should have seem’d as impossible to him, as to put a Bushel of any thing into a Pint.” (That is to say, a larger volume into a smaller.)

That’s quite entertaining but also quite significant evidence for the debates we have been considering. Clearly, learning the geometry of sight was a bit like learning a language for this person who became sighted. He didn’t immediately understand the geometrical structure of visual impressions, so clearly all of that is not completely innate. So it speaks against a Kantian account that takes Euclidean geometry to be a precondition of any geometrical thought or geometrical sensory perception.

But the story of the boy who became sighted fits quite well with a Poincaré-type account in which the geometry of sight can only be developed gradually through experience and coordination with self-motion.

Nevertheless, you can still say that Kant was right in a way. Poincaré is in a sense neo-Kantian. According to Poincaré, Euclidean geometry is not innate, but some geometrical notions are. The mind is predisposed to discern geometrical aspects of its surroundings. Hardwired into the mind are not all of Euclid’s axioms but still a good bit of geometry, such as the categorisation of which perceptions are related to geometry at all, and perhaps related to this some concepts such as displacement, rotation, and so on.

So, those are the ways in which geometry is like language. Both are part innate and part shaped by the environment. To adopt a particular language or a particular geometry is to fit your thoughts into an arbitrary and subjective framework. But that’s a good thing because there are no objective frameworks, and without some such conceptual framework, thinking could never even get off the ground in the first place.

]]>**Transcript**

The mathematician has only one nightmare: to claim to have proved something that later turns out to be false. There are thousands of theorems in Greek geometry, and every last one of them is correct. That’s what mathematical proofs are supposed to do: eliminate any risk of being wrong. If this armor is compromised, if proofs are fallible, then what’s left of mathematics? It would ruin everything.

But the nightmare came true in the 19th century. What had been thought to have been proofs were exposed as fallacies. Top mathematicians had made mistakes. Mistakes! Like some commoner. It’s going to be hell to pay for this, as you can imagine.

I’m referring to Euclid’s fifth postulate, the parallel postulate. Euclid’s postulate had rubbed a lot of people the wrong way even since antiquity. It sounds more like a theorem.

The earlier postulates were very straightforward: there’s a line between any two points, stuff like that. Very primitive truths. It makes sense that the bedrock axioms of geometry should be the simplest possible things, such as the existence of a line between any two points.

The parallel postulate, by contrast, is not very simple at all. It’s not a primordial intuition like the other postulates. It states that two lines will cross if a rather elaborate condition is met. That’s the kind of thing theorems say. This particular type of configuration has such-and-such a particular property. That’s how theorems go in Euclid. Very different from the other postulates, which are things like: any line segment can be spun around to make a circle.

So, people tried to prove the parallel postulate as a theorem. Many felt that it should not be necessary to assume the parallel postulate as Euclid had done. The parallel postulate should be a consequence of the core notions of geometry, not a separate assumption.

Many people tried to “improve” on Euclid in this way. From antiquity all the way to the 19th century. Around 1800, people like Lagrange and Legendre proposed so-called proofs of the parallel postulate. Those are big-name mathematicians. Their names are engraved in gold on the Eiffel Tower. Lagrange was even buried in the Panthéon in Paris. Elite establishment stuff.

But even these bigwigs were wrong. Their proofs contain hidden mistakes. It’s astonishing that this was more than 2000 years after Euclid. People tried to improve on Euclid for millennia. And not a few claimed to have succeeded. But the fact is that Euclid was right all along. The parallel postulate really does need to be a separate assumptions, just as Euclid had made it. It cannot be proved from the other axioms, as so many mathematicians during those millennia had mistakenly believed.

The Greeks, you know, they were really something else. It’s so easy to make subtle mistakes in the theory of parallels. History shows that there are a hundred ways to make tiny invisible mistakes that fool even the best mathematicians. Top mathematicians who were never wrong about anything else stumbled on this one issue.

Somehow Euclid got it exactly right. He didn’t make any of those hundred mistakes that later mathematicians did. That’s not luck, in my opinion. Arguably, the Greeks were more sophisticated foundational geometers than even the Paris elite in 1800. Unbelievable but true. Euclid’s Elements is not a classic for nothing. Euclid is not a symbol of exact reasoning because of some lazy Eurocentric birth right. Euclid’s Elements really is that good.

When Euclid made the parallel postulate an axiom, he seems to be suggesting that it cannot be proved from the other axioms. And he was right. But, as I said, many people had a hunch that he was wrong about this. They thought it would be impossible for the other axioms to be true and the parallel postulate not true.

So many mathematicians figured they could prove this by contradiction: Suppose the parallel postulate is false. If we could show that that assumption would contradict other geometrical truths, then the assumption must be false. So this way we could prove that the parallel postulate must be true, by showing that it would be incoherent or impossible for it to be false.

Indeed, it was found that negating the parallel postulate had various strange consequences. For example, if the parallel postulate is false then squares do not exist. Suppose you try to make a square. So you have a base segment, and you raise two perpendiculars of equal length from the two endpoints of the segment. Then you connect the two top points of these two perpendiculars. That ought to make a square. In Euclid’s world it does.

But proving that this really makes a square requires the parallel postulate. If the parallel postulate is false, one can instead prove that this construction does not make a square but rather a weirdly disfigured quadrilateral. Because the last side of the “square” doesn’t make right angles with the other sides. So even though you made sure you had right angles at the base of the quadrilateral, and that the perpendicular sides were equal, the fourth and final side still somehow manages to “miss the mark” so to speak. It makes non-right angles.

It’s as if the sides are sort of bent. It’s as if you had four perfectly equal sticks of wood, but then you stored them carelessly and they were exposed to humidity and so on and they were warped. So now they’re kind of mismatched in terms of length and straightness, and when you try to piece them together to make a square they don’t fit right. They make some wobbly not-quite-square shape.

Doing geometry without Euclid’s parallel postulate feels a bit like that. It’s sort of bent out of shape and nothing fits the way it should anymore.

One person who investigated this was Saccheri. He wrote a big book discussing this misshaped square and other things like that, in 1733. Saccheri felt that he had justified Euclid’s parallel postulate by examples such as theses. The square that’s not a square and other such deformities, Saccheri declared to be “repugnant to the nature of the straight line.”

But one might say that he used this emotional language to compensate or cover up a shortcoming in the mathematical argument. He had indeed showed that if the parallel postulate is false then geometry is weird. Then you have squares that don’t fit, and other things that feel like doing carpentry with crooked wood.

But weird is not the same as self-contradictory. Despite their best efforts, mathematicians could not find a clear-cut proof that negating the parallel postulate led to directly contradictory conclusions. This is why Saccheri had to say “repugnant” rather than contradictory. You only get “repugnantly” deformed squares, not direct contradictions such as 2=1 or a part being greater than the whole. Those things would be logical contradictions and you wouldn’t need emotions like repugnance.

In fact, a hundred years after Saccheri, mathematicians came to accept that this strange non-Euclidean world of the warped wood is not contradictory. It is coherent and consistent. It is merely another kind of geometry. An alternative to Euclid.

People used to shout and scream that all kinds of things were repugnant, such as homosexuality, for instance. That doesn’t really prove anything except the narrow-mindedness of those accusers. Mathematicians had been equally narrow-minded. They had tried to justify the status quo for thousands of years. They had tried to prove that their way of doing things–their geometry–was the only right way. Only in the 19th century did they finally realize that it was much more productive to embrace diversity, to accept all the geometries of the rainbow.

For so many years mathematicians could not get away from the idea that the “straight” squares of Euclid were the only “normal” ones, and that the “repugnant” alternative squares of non-Euclidean geometry were birth defects. But they were wrong. Non-Euclidean geometry is as legitimate as any other. It was a creative watershed shift in perspective in mathematics to finally accept this instead of trying to prove the opposite.

Here’s how Gauss, the greatest mathematician at the time, put it in the early 19th century. Negating Euclid’s parallel postulate “leads to a geometry quite different from Euclid’s, logically coherent, and one that I am entirely satisfied with. The theorems [of this non-Euclidean geometry] are paradoxical but not self-contradictory or illogical.” “The necessity of our [Euclidean] geometry cannot be proved. Geometry must stand, not with arithmetic which is pure a priori, but with mechanics.”

Geometry has become like mechanics in the sense that it is empirically testable. The theorems of geometry are not absolute truths but hypotheses like the hypotheses of physics that have to be checked in a lab and perhaps corrected if they don’t agree with measurements.

For example, Euclid proves that the angle sum of a triangle is 180 degrees. But this theorem depends on the parallel postulate, just as Euclid’s proof reveals it to do. In non-Euclidean geometries, angle sums of triangles will be different. So that’s something testable. Measure some triangles to see which geometry is right, just as you drop some weights or whatever in a physics lab to see which law of gravity is right.

Let me quote Lobachevsky, one of the other discoverers of non-Euclidean geometry. Here’s how he makes this point in his book of 1855: “[Non-Euclidean geometry] proves that the assumption that the value of the sum of the three angles of any rectilinear triangle is constant, an assumption which is explicitly or implicitly adopted in ordinary geometry, is not a consequence of our notions of space. Only experience can confirm the truth of this assumption, for instance, by effectively measuring the sum of three angles of a rectilinear triangle. One must give preference to triangles whose edges are very large, since according to [Non-Euclidean geometry], the difference between two right angles and the three angles of a rectilinear triangle increases as the edges increase.” So you need big triangles to tell the difference, just as the earth is round but looks flat from where we’re standing because we only see a small part of it. In the same way we need big triangles to detect the nature of space. Therefore Lobachevsky recommends that we should use astronomical measurements for this: “The distances between the celestial bodies provide us with a means for observing the angles of triangles whose edges are very large.”

Let’s think about the logical structure involved in the realization that non-Euclidean geometry is possible. It used to be thought that Euclid’s parallel postulate was a necessary consequence of the other axioms. Although Euclid seems to have been wise enough to realize that it was not, others erroneously believed that this was a mistake rather than an insight on Euclid’s part.

So the question is: Does the parallel postulate follow from the other axioms? If the answer is yes, then the way to settle matter is to provide a proof, a deduction, starting from the other axioms and ending up with the parallel postulate. So that would be like adding another theorem to Euclid’s Elements.

On the other hand, suppose the answer is no, the parallel postulate does not follow from the other axioms. How then could we prove that? It’s very different in this case. It is no longer about proving a theorem. Rather it is about proving that something cannot be proved. It’s much more “meta” than just proving a particular theorem.

But here’s how you do such a thing. Consider this analogy. Suppose someone believes that all odd numbers are prime numbers. 3 is prime, 5 is prime, 7 is prime, and so on. So someone has become convinced that all odd numbers are prime numbers, and they set out to prove it. The start with what it means to be odd, and from that information they try to prove that that implies that it must be prime as well.

But this is of course wrongheaded. Trying to prove that being prime follows from being odd is just as futile as trying to prove that the parallel postulate follows from the other axioms of Euclid.

How could we set this mathematician straight? How could we prove that what he’s trying to prove is impossible to prove? The way to do this is not by some general proof, but by a specific example.

Look at the number 9. It’s odd, but it’s still not prime. Because it’s 3 times 3, so not a prime number.

The obvious way to interpret this is to say that the guy was wrong with his hypothesis. The claim that being odd implies being prime is false.

But from a logical point of view it is interesting to look at it in slightly different terms. Let’s not think about it in terms of right and wrong. Logic doesn’t care about right and wrong. Logic cares only about what follows from what. When logic looks at a proposition, logic doesn’t ask: is it true or false? Logic asks: does it follow from a particular set of axioms?

Logic is about entailment relations. What follows from what. Logic doesn’t care what assumptions or axioms you use. It only cares about what follows from those axioms.

So in terms of our example with the odd numbers, we shouldn’t focus on the question “are all odd numbers prime numbers?” Instead, from a logical point of view, the better question is: “does being odd entail being prime?” Or “is primeness a logical consequence of oddness?”

We had a counterexample: the number 9. From the logical point of view, we interpret this a bit differently. Not as proving the falsity of the conjecture, because we’re not interested in true or false. Instead, what the example of 9 shows is that it is not possible to derive the property of being prime from the property of being odd.

When we put it this way, we have an answer to that challenging meta question: How can we prove that it’s impossible to prove something? We just did! It’s impossible to prove primeness from oddness. Because if there was a proof that showed that any odd number must be prime, then that proof would apply to 9, since it’s odd, and it would prove that 9 is prime, which it is not. Therefore no such proof could exist.

It was the same in geometry. People thought the parallel postulate was a logical consequence of the other axioms. The way to prove this wrong is to exhibit an example in which the other axioms are true but the parallel postulate is false. Just as in the number theory case we had to find an example where oddness was true but primeness was false.

This is indeed what happened. Mathematicians discovered something that corresponded to the number 9. This proved the logical independence of the parallel postulate, just as the number 9 proves that primeness is not a logical consequence of oddness.

In the geometry case, the role of the number 9 was played by models of hyperbolic geometry. These are visualizations that prove that there are perfectly coherent worlds in which the parallel postulate is false while all the other axioms of Euclid are true.

Once mathematicians started thinking in these kinds of terms, it turned out to be not so difficult to find models like that. Mathematicians really could have done that a lot earlier. Even hundreds of years earlier, or even in Greek times. It’s a bit of an embarrassment that it took so long.

Imagine how embarrassing it would be to sit around for hundreds of years trying to prove that all odd numbers are prime numbers, and ranting about how the very idea of an odd non-prime is “repugnant to the nature of an odd number” only to then discover that, whoops, actually there’s a pretty straightforward counterexample right there, the number 9.

The mistake mathematicians made in geometry was of course not quite so glaring but still in a way it was quite similar. The counterexamples were not that difficult to find. Once mathematicians opened their minds to the possibility of such counterexample, they found them fairly easily.

Mathematicians had missed these rather simple counterexamples for thousands of years because of their closed-minded perspective and preconceived notions. Mathematicians had relied too much on emotions, intuitions, such as repugnance. And they had assumed that there can only be one reasonable geometry, because geometry must correspond to physical space.

Mathematicians could not afford to make those mistakes again. These mistakes are what made the nightmare come true, namely that what mathematicians had thought they had “proved” was actually false.

It was a time for soul searching and repentance. And the lessons from this whole embarrassment were quite obvious. The sources of error were intuitions, such as feelings about how straight lines “should” behave, as well as the notion that geometry means the geometry of the physical space around us.

Those ideas were the losers of the story. The winner was logic. The breakthrough had come by detaching geometry from intuition and reality. By abstracting geometry away to its logical structure only. That was the winning perspective.

To spell out what this means for geometry and its relation to the world, let me quote Einstein’s essay Geometry and Experience. Einstein wrote this is 1921, but he is really just summarizing a standard consensus that had been firmly established decades earlier. But why not use the words of the famous Einstein, they are as good as any to make this point. Here’s what Einstein says:

“How can it be that mathematics, being after all a product of human thought which is independent of experience, is so admirably appropriate to the objects of reality? Is human reason, then, without experience, merely by taking thought, able to fathom the properties of real things? In my opinion the answer to this question is, briefly, this: as far as the propositions of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality. It seems to me that complete clarity as to this state of things became common property only through that trend in mathematics, which is known by the name of ‘axiomatics’.” That’s Einstein’s word for what I called the logic perspective. Same thing. Einstein continues:

“Let us for a moment consider from this point of view any axiom of geometry, for instance, the following: through two points in space there always passes one and only one straight line. How is this axiom to be interpreted in the older sense and in the more modern sense?”

“The older interpretation [is]: everyone knows what a straight line is, and what a point is. Whether this knowledge springs from an ability of the human mind or from experience, from some cooperation of the two or from some other source, is not for the mathematician to decide. He leaves the question to the philosopher. Being based upon this knowledge, which precedes all mathematics, the axiom stated above is, like all other axioms, self-evident, that is, it is the expression of a part of this a priori knowledge.”

“The more modern interpretation [is]: geometry treats of objects, which are denoted by the words straight line, point, etc. No knowledge or intuition of these objects is assumed but only the validity of the axioms, such as the one stated above, which are to be taken in a purely formal sense, that is, as void of all content of intuition or experience. These axioms are free creations of the human mind. All other propositions of geometry are logical inferences from the axioms (which are to be taken in the nominalistic sense only). In axiomatic geometry the words ‘point’, ‘straight line’, etc., stand only for empty conceptual schemata. That which gives them content is not relevant to mathematics.”

“‘Practical geometry’ [arises if we] add the proposition: solid bodies are related, with respect to their possible dispositions, as are bodies in Euclidean geometry of three dimensions. Then the propositions of Euclid contain affirmations as to the behavior of practically-rigid bodies. All length-measurements in physics constitute practical geometry in this sense, so, too, do geodetic and astronomical length measurements, if one utilizes the empirical law that light is propagated in a straight line, and indeed in a straight line in the sense of practical geometry. I attach special importance to the view of geometry, which I have just set forth, because without it I should have been unable to formulate the theory of relativity. From the latest results of the theory of relativity it is probable that our three-dimensional space is approximately spherical, that is, that the laws of disposition of rigid bodies in it are not given by Euclidean geometry, but approximately by spherical geometry.” That is to say, the angle sums of triangles are more than 180 degrees.

So all of that I quoted from Einstein. But Einstein speaks for basically the entire mathematical community here. He is describing what was, in his time, the standard view that almost everyone took for granted.

Indeed, these points about mathematics turning to pure axiomatics and so on, apply not only to geometry but to mathematics as a whole. Mathematicians took that lesson to heart and never looked back, basically. So the discovery of non-Euclidean geometry was the birth of modernity, you might say, in mathematics. It led mathematicians to conceive their field exclusively in terms of logic and formalism, and forget everything about intuition or the idea that mathematics is linked to physical reality. And that’s pretty much where we are today, almost two centuries later, with few exceptions.

In the 19th century, you could be forgiven for thinking that this was a case of straightforward progress. Mathematicians had simply discovered the right way to do mathematics, or the best way known so far anyway. The new logic perspective was simply better than the old intuitive or empirical stuff. We shedded the old errors like so many superstitions and became enlightened.

Around 1900, that was a pretty credible narrative. The logic perspective had gone from win to win, and done a clean sweep of mathematics. Everything it touched seemed to become instantly clearer and better. Hilbert was a leading mathematician at this time who may be taken as a symbol of this. He turned from field to field and made everything clear and clean and modern with this logical Midas’ touch.

But the winning streak did not last forever. With one knock-out win after another behind him, Hilbert turned to the foundations of the entire subject of mathematics and tried to do the same trick there. Many people were optimistic. The trick had worked every time before, and now the world’s greatest mathematician was going to use it to definitively settle all the questions of the foundations of mathematics, such as proving that mathematics is consistent.

But the trick broke this time, even though it had worked every time before. Hopes of a quick victory proved as delusional as the equally hubristic delusions of the war planners who were marching into the First World War at the same time.

The world came crashing down around the great Hilbert. He was German, and these were not good times for Germany. First the students and younger generation died in the war. Then the many prominent Jewish faculty were driven out of the country. Hilbert’s once vibrant university was quickly turned into a shadow of its former self. Hilbert himself contracted a rare decease for which the only treatment was eating lots and lots of raw liver every day.

1933 was a year of not one but two disasters. The Nazis took power, but there was an equal blow in the world of mathematics, when Gödel proved that the logician’s dream was impossible. Logical formalism could not prove its own consistency. In other words, the program of detaching mathematics from intuition and experience turned out to be inherently limited. Its utopian dream proved to be unreachable, and demonstrably so in fact.

Kant used a beautiful analogy that is relevant here. It goes like this:

“Deceived by the power of reason, we can perceive no limits to the extension of our knowledge. The light dove cleaving in free flight the thin air, whose resistance it feels, might imagine that her movements would be far more free and rapid in airless space.”

Which is of course not true. The dove may think that air causes nothing but resistance, but if all air is removed, the dove would quickly be taught a different lesson of course. Not only would the dove crash to the ground at once, it would also suffocate in seconds.

A similar fate awaited the movement to purge mathematics of intuition and physical content. People like Hilbert were so keen to remove the old dependence on intuition and the physicality of geometry as if these things were nothing but “air resistance” that prevented the flight of pure logic in a perfectly clean vacuum.

But birds cannot fly without air, and neither could mathematics. Gödel’s theorem of 1933 proved that logical formalism cannot prove it own consistency, which in terms of this analogy is like proving that the dove cannot fly in a vacuum.

This setback within mathematics was perhaps just as unnerving to Hilbert and other mathematicians as all those jarring disasters that were piling on in the outside world. It’s cruel joke of history that it had both these worlds collapse at the same time.

Maybe the parallel extends further. World War One was a horror of horrors, but that didn’t prevent us from doing it all over again soon thereafter. And we still don’t know how to get rid of war.

Mathematics has a similar attachment to formalism and logic. As with war, the romantics among us are not too happy about formalistic mathematics. Its power cannot be denied. Some, or maybe even many, of its victories were for the best. But still it does not feel right in one’s heart to drill young people into an army of formalists. Seeing mathematics as nothing but logical inferences from arbitrary axioms is as heartless as realpolitik. It reigns to this day, despite a now checkered record, because the only alternatives are hippie fantasies with no realistic prospects of ruling. Modern mathematics and modern politics are alike in this regard.

Well, that makes for a bleak ending. Perhaps non-Euclidean geometry does not deserve to be associated with all this misery. It’s not non-Euclidean geometry’s fault that mathematicians had made mistakes about the parallel postulate. Nevertheless the impact of the discovery of non-Euclidean geometry on the mathematical psyche was dramatic and long-lasting. It sent mathematicians on a soul-searching bender, the hangover of which is still felt today.

]]>**Transcript**

Rationalism says that geometrical knowledge comes from pure thought. Empiricism says that it comes from sensory experience. Neither is very satisfactory, because geometry is clearly both: it’s too good a fit for the physical world to be only thought, and it relies too much on abstract proofs to be only experience.

So it seems the “right” philosophy of mathematics must be a little bit of both. But how? Rationalism and empiricism mix like oil and water. They are so different, so opposite, that it seems impossible to find any sort of middle ground that has half of each.

Nevertheless there is a golden mean of sorts. The philosophy of Kant. Immanuel Kant, the late 18th-century philosopher. His view of geometry is in a way the best of both worlds: combining the best rationalism with the best of empiricism. Let’s see how he pulled that off.

Kant’s theory is another innateness theory. Geometry is innate; it is hardwired into our minds. That’s what the rationalists said too, of course. Innateness was why the uneducated boy in Plato’s Meno could reach substantial geometrical insights without instruction. The innate intuitions of geometry are reliable because God is not a deceiver, said Descartes. And they apply to the physical world, because the Creator put into our minds the same ideas he had used to design the universe, said Kepler and others.

But Kant’s innateness is different. He lived a century and a half after Kepler and Descartes, and the reliance on God to support the rationalist worldview had become much less fashionable during the intervening years. And indeed Kant does away with it.

By taking geometry to be innate, Kant automatically inherits the best of rationalism. That is to say, he is able to account for the prominent role of pure reason in geometry in a convincing way, just as the earlier rationalists had done before him.

But Kant solves the weakness of rationalism very differently. If geometry is innate and susceptible to purely theoretical elaboration, then why does it always agree so well with experience? Not because “God was a geometer,” as Plato and Kepler said. No, Kant’s solution is: Not only are geometrical principles innate in our thoughts, they are also innate in our perceptions.

Aha, plot twist! The so-called success of mathematics in the real world is no miracle. It’s a rigged game. Our eyes and our senses are biased. They can only see Euclidean geometry.

It’s an illusion to think that we can compare our mathematical deductions with reality. We think we can “test” whether theoretically established theorems are true or not in the physical world; for example, by measuring the sides of right-angle triangles and comparing the results to what Euclid says it should be. But we were naive to think that these two things were really independent. Just as our thoughts are shaped by our innate intuitions, so also our perceptions of physical reality are shaped by the same intuitions.

We don’t have direct access to “raw data” about the physical world. When we think we observe the world, we really observe an interpreted version of the world. The mind can only process information that is interpreted or converted to fit a particular format. In terms of geometry, that means Euclidean geometry.

We see the world through Euclid-colored glasses. Like those pink sunglasses that John Lennon used to wear. Of course if you wear pink glasses, then everything looks pink. But of course John Lennon could not say: aha, I told you, the world is rosy; I argued theoretically that the world should be rosy, and now I have confirmed by observation of the actual world is in fact rosy, because look how everything is pink. John Lennon would be fooling himself if he argued that way.

Euclidean geometry is exactly like that, according to Kant. With experience and observation, we do not discover that Euclidean geometry applies to the real world. What we really observe is our own biases. We discover not facts about the world, but facts about what kind of glasses we are wearing.

John Lennon made the world rosy by putting on his glasses, and in the same way we make the world Euclidean by a hidden lens in the mind’s eye.

Think of a chili pepper. This pepper proves that Kant was right. How so? Think about it. What are the characteristics of a chili pepper?

It’s red. What is red? Is the pepper “really” red? That is to say, is its redness an objective property of reality? Well, yes and no, right? The way science accounts for colors is in terms of wave length of light. What we in our minds regard as red correspond in reality to a particular frequency of light waves.

So there’s something there corresponding to red, but our minds put a major interpretative spin on it. Red is really just a wavelength, “just as number,” so to speak. But our minds turn it into something more qualitative.

And the mind is very selective as well. The remote control of your TV sends its signal using infrared light, which the minds chooses not to see at all. Even though infrared light is exactly the same kind of thing as the color red from the pepper. They are exactly the same kind of waves, only with slightly different frequencies. “Objectively” they are very similar. But subjectively, in our minds, they couldn’t be more different. One is the color red that is one of the fundamental categories in terms of which we navigate reality, and the other is completely invisible.

So perception is very far from direct access to “objective” raw data. The mind interferes very heavily: it selects, transforms, distorts. Maybe when we think Euclidean geometry fits reality it’s only because of this. Only because things that don’t fit a Euclidean mold are as invisible to us as infrared light.

The chili pepper is also very hot to taste. Again, is that an objective property of pepper itself, or is it just a subjective matter of how it interacts with our tongues?

In fact, birds eat chili peppers and to them they are not hot. Chili peppers evolved to be hot because it’s better to be eaten by birds than to be eaten by mammals. The seeds spread better that way. So chili peppers developed this characteristic of being repellently hot to mammals like us, but but perfectly appetizing to birds.

Could it be the same with our geometrical intuitions? “Two lines cannot enclose a space,” Euclid says. The very thought is repellent! Well, maybe that’s just our subjective experience, just as we find it repellent to bite into a hot pepper. Maybe other creatures have completely different geometrical intuitions. Just as birds think pepper are not hot, maybe they also think space has five dimensions or whatever. Who knows.

Different animals can have all kinds of different intuitions and modes of perception that fits the way they navigate the world, the way they find food, and so on. Maybe Euclidean geometry is just something that happened to be convenient to us, just as night vision is convenient to a cat, or a heightened sense of smell is to a dog.

That would make geometrical experience quite subjective. Or at least subjectively contaminated, or entangled with subjective factors.

This is how Kant is able to bridge the gap between rationalism and empiricism. We can sit in a closed room and figure out in our heads geometry that then turns out to be true in the real world. Because, says Kant, what is really happening is that the mind is analyzing itself. What we discover through geometrical meditation is not objective facts, but facts about the geometrical preconceptions hardwired into our minds and the consequences of those conceptions. And these results apply to the world not because they are really out there but because our perception actively processes and converts and interprets any sensory input in Euclidean terms.

As Kant himself says: “Space is not something objective and real; instead, it is subjective and ideal, and originates from the mind’s nature as a scheme, as it were, for coordinating everything sensed externally.” “It is from the human point of view only that we can speak of space, extended objects, etc.”

In the introduction I framed the issue in a way that is flattering to Kant. I said: Everybody knew that you needed some kind of middle ground between rationalism and empiricism, but no one could figure out how to do it. Until Kant, he finally cracked the nut.

But maybe that’s the wrong way to look at it. It often is, this kind of narrative. “Everybody tried to do such-and-such but no one could do it, until finally one guy was smart enough to figure it out.” That’s not usually how history works. If we look into the details we often find that contextual factors explain why key steps happened at certain times.

It’s true that everyone knew that neither rationalism nor empiricism were perfect for hundreds of years. And Kant’s proposal is certainly a very clever way of tackling that problem. So why did it take all the way to the late 18th century before these ideas were proposed?

We already mentioned some relevant factors. Kant makes geometrical knowledge in a sense subjective. That’s a major disappointment, one might say. Most philosophers had certainly hoped to be able to defend a much grander claim. Kant “solves” the rationalism-empiricism problem only by as it were belittling geometrical knowledge, which is a very high price to pay.

The main alternative, as we have seen, was to give God a major role in epistemology. So there’s a trade-off: either you pin geometry to God and you can have it be the most amazing thing, the most perfect knowledge, or else you detach it from God and make it stand on its own legs, but then it’s a lot weaker; it’s a mere subjective human thing and no longer this almighty pinnacle of pure intellect.

The exchange rate, as it were, between these two options fluctuated over time. As God became less popular, the cost of switching to Kantianism went down.

But there’s another reason too why Kant’s theory made more sense in the 18th century than in the 17th. Namely what we said before about how Newton’s science was a blow to rationalism.

We spoke about how that was the case. Rationalism requires knowledge to be generated from within the mind. All knowledge needs to be gradually built up from the most simple intuitions, according to the rationalist point of view. In geometry, that meant ruler and compass and other tools for generating geometrical objects. In physics, it meant contact mechanics; that is to say, seeing complex physical phenomena as an aggregate of lots and lots of little collisions of bodies.

Newton’s physics cannot be reduced to contact mechanics. Or to any other simple intuition. It is in fact counterintuitive. So it cannot be generated from within the mind, through an elaboration in thought of the most undoubtable truths. This is why Newtonian physics is a problem for rationalism.

But the story is a bit more general than that. In fact, Newton’s physics can be seen as a blow to philosophy altogether.

From the rationalist point of view, philosophy comes before science. You start with general philosophical thought. “I think therefore I am”: That’s a very general philosophical truth, and you start there because it’s the most knowable. You start by asking yourself what kinds of things are knowable. From that starting point you arrive at the idea that in physics one of the most primitive knowable things is the contact mechanics of bodies.

From this point of view, philosophy is the boss of science. Philosophy is telling science what to do. Before even starting on science, you have already determined through introspection and meditation what the primitive intuitions of physics are. Any science that follows needs to conform to these predetermined rules that philosophy has established beforehand.

From a rationalist point of view, this makes sense. If knowledge fundamentally comes from within the mind, it makes sense to work from the inside out; to start with the most general philosophical core and then build on that to get to things like physics and other stuff that are more connected to the outside world. That’s a core commitment of the rationalist worldview. This is why it requires philosophy to be prior to science, and the boss of science.

Newton does it the other way around. To him, science is the boss of philosophy. This is a natural consequence of his empiricist, “reading backwards” mindset that we have emphasized before. Thought starts not with inward reflection on our basic intuitions, but in the wild jungle of complex phenomena. Science reasons as it were backwards from there to discover the basics principles, such as axioms of geometry and fundamental laws of physics.

If you continue this process one further step you get to philosophy. Just as the laws of physics are whatever is needed to explain the phenomena, so the principles of philosophy are whatever is needed to make that physics possible. So philosophy is subordinated to science. It doesn’t tell science what to do, but the other way around.

To the rationalists, philosophy set the ground rules that science must obey. To the empiricists, to Newton, philosophy merely describes what assumptions are necessary for science after science has already been established. To the rationalists, philosophy is prescriptive: it gives orders, it says how science has to be. To the empiricists, philosophy is descriptive: it’s an observer, a backseat journalist, that merely says how science is, without having any influence over it.

So we see how the basic outlooks of rationalism and empiricism imply these opposite views of the relation between science and philosophy. And Newton’s physics was extremely successful. So its success lent credibility to the empiricist outlook overall, including the demotion of philosophy.

But in fact this is still not the end of it. There is yet another respect in which Newton’s physics dealt an additional death blow to philosophy. Namely on the issue of absolute versus relative space.

Newton clashed with Descartes and Leibniz on this issue as well. It goes like this. What can we know about the spatial properties of a body, such as its position and velocity?

Descartes and Leibniz were relativists about space. Everything we could ever know about positions and velocities of bodies is relative. That is to say, you can only specify the position or speed of a body by comparing it to another body. The chair is so-and-so far from the table. The train is moving away from the station at such-and-such a speed. You cannot speak of the position of the chair or the speed of the train without comparing it to something. You need to relate it to some reference point.

Descartes and Leibniz insisted on this. Here’s how Descartes puts it: “The names ‘place’ or ‘space’ only designate its size, shape and situation among other bodies.” “So when we say that a thing is in a certain place, we understand only that it is in a certain situation in relation to other things.” Leibniz agreed. “Motion is nothing but a change in the positions of bodies with respect to one another, and so, motion is not something absolute, but consists in a relation.”

It takes two to tango, and it takes two bodies to be able to speak of position and velocity. Because you can only describe the position or velocity of the second body by using the first as a reference point.

If there was only one body in the universe, it wouldn’t make any sense to ask whether it was moving or not. Since there’s nothing to use as a reference point, the very concept of motion becomes meaningless is such a situation. According to the relativist conception of space.

This fits very well with our previous emphasis on operations in geometry. Relative positions and relative velocities correspond very well to operations. You can specify what it means for one object to be so-and-so far from another object, or moving with such-and-such a speed with respect to the other object, in terms of concrete measurements. I take a measuring tape, I stretch it from one to the other, that’s how far apart they are.

If there is only one object in the universe, there is no operation we can perform to check whether it is moving or not. So to introduce the idea of every body having some absolute state of motion, independently of any other body, is equivalent to introducing concepts by means other than operations. We know from geometry that this is dangerous, as we saw with the superright triangle and other examples.

Yet Newton does exactly this. Newtonian physics presupposes absolute space. That is to say, it assumes that every body has some definitive position and velocity, completely independently of any other body, and completely independently of what is measurable or knowable to us.

So from the Newtonian, absolutist point of view, if there is only a single object in the universe, then that object still has some definite velocity. It’s either moving or not. Whether it’s moving or not is physically undetectable. There is no way to tell, with a physical experiment, whether it is moving or not. Nevertheless, the question of whether it is “really” moving or not still makes sense and has a definite answer, according to Newton.

This notion–that any body has an “absolute” position and velocity–is necessary for Newton’s physics. Think of the law of inertia. It says: If there is no outside force acting on a body, then the body keeps going in a straight line with the same speed. Forever. Like a metal ball rolling on a marble table, when there is no friction and no obstacles, it keeps going with the same velocity. Without external influence, the state of motion remains the same.

But note that this law talks about the state of motion of a body without reference to other bodies. The law of inertia presupposes that the body has some inherent velocity, a true velocity. That’s the thing that stays the same in absence of interference. Obviously this is not dependent on some particular reference point. The body in and of itself has a state of motion associated with it. The state of motion of the body is an absolute property, not a relative one.

This clash between the absolute and relative space points of view is another clash between science and philosophy. Relative space is clearly the “best” view in terms of philosophy. The philosophical objections to absolute space are very compelling: Absolute space is unknowable. Absolute space introduces concepts that are empirically untestable, unverifiable, unoperationalisable.

The reply from the other side, from Newton’s side, is not to dispute that philosophy is on the side of relative space. Instead it is to belittle the authority of philosophical arguments. Indeed, absolute space makes no sense philosophically. But the conclusion from this is: tough break for philosophy.

Absolute space is a necessary precondition to state the law of inertia, and the law of inertia is an integral part of Newton’s extremely powerful physics, so inertia and hence absolute space must be accepted. Philosophy is just going to have to deal with it.

So this once again reinforces Newton’s point that philosophy is basically a spectator sport. Philosophy can’t tell science what to do. If philosophy clashes with science, as it does regarding absolute space, then philosophy has to give way.

Physicist Stephen Hawking famously declared that “philosophy is dead.” He had in mind 20th-century developments. That’s how many modern scientists think. But philosophy was dead once before. Newton killed philosophy.

If you want to get somewhere in science and mathematics, you can’t get caught up in pointless speculations and debates about “what it all means.” You just have to do the math, get on with it. That was the case in the 18th century, and again in the 20th century.

Another prominent modern physicist, Lee Smolin, put it as follows: “When I learned physics in the 1970s, it was almost as if we were being taught to look down on people who thought about foundational problems. When we asked about the foundational issues in quantum theory, we were told that no one fully understood them but that concern with them was no longer part of science. The job was to take quantum mechanics as given and apply it to new problems. The spirit was pragmatic; ‘Shut up and calculate’ was the mantra. People who couldn’t let go of their misgivings over the meaning of quantum theory were regarded as losers who couldn’t do the work.”

It was exactly the same thing in the 18th century. Then too scientists and mathematicians figured they were better off just ignoring philosophy. And with good reason since Newton’s physics was an obvious winner in terms of mathematics and science, but a complete non-starter philosophically according to many.

The greatest mathematician and physicist of the 18th century, Euler, realized this perfectly well. He knew that absolute space was junk philosophy but essential to science.

He knew that the law of inertia demanded absolute space. As Euler says: “For if space and place were nothing but the relation among co-existing bodies, what would be the same direction? Identity of direction, which is an essential circumstance in the general principles of motion, is not to be explicated by the relation of co-existing bodies.”

Euler also knew that there were powerful philosophical objections to absolute space. The objections of Descartes and Leibniz that I already mentioned. Let me quote here how Ernst Mach later made the same point in the late 19th century. Mach is basically reviving the 17th-century criticism of absolute space. Here’s how Mach puts it:

“Absolute space and absolute motion are pure things of thought, pure mental constructs, that cannot be produced in experience. [They have] therefore neither a practical nor a scientific value; and no one is justified in saying that he knows aught about it. It is an idle metaphysical conception. All our principles of mechanics are experimental knowledge concerning the relative positions and motions of bodies. No one is warranted in extending these principles beyond the boundaries of experience. In fact, such an extension is meaningless, as no one [can] make [any] use of it.”

Euler and others in the 18th century were aware of this problem with the notion of absolute space that is so essential to Newtonian science. They didn’t know how to solve this philosophical problem, except to ignore philosophy altogether. Euler pretty much says so. Listen to this quote:

“I do not want to enter the discussion of the objections that are made against the reality of [absolute] space and place; since having demonstrated that this reality can no longer be drawn into doubt, it follows necessarily that all these objections must be poorly founded; even if we were not in a position to respond to them.”

So Euler admits that he cannot answer the philosophical objections. Instead his solution is: forget philosophy. Philosophy became obsolete with the Newtonian revolution in science. It was out of touch.

Kant is the savior of philosophy. Kant makes philosophy relevant to science again, after a century of being obsolete. Kant’s theory is a way to bring philosophy up to date with science. It is a philosophy that is compatible with Newtonian science, unlike earlier versions of rationalism.

Against this background we can understand why Kant was willing to make mathematical knowledge subjective. That part of his theory was a huge betrayal of a major tenet of classical rationalism. But times had become desperate enough. Philosophy was the laughing stock of scientists. It had to do something, anything.

So Kant decided to bite the bullet on subjectivity in order to at least salvage something of philosophy. Save what can be saved.

Rationalism had once been a mighty kingdom, but it was bleeding territory. Newton’s science was taking the world by storm, and it seemed a real risk that rationalism would not only lose ground but might even be wiped off the map altogether.

Kant’s plan for saving rationalism shows how far it had fallen. In its glory days, rationalism would have scoffed at the notion that geometry is subjective. But now, it was that or death. Like royalty eating peasants’ porridge, rationalism had to adapt or die. Rationalism had to sacrifice the pride of its forefathers–the objective truth of geometry.

But despite this humiliating concession, Kant’s reinvention of rationalism was an astonishing success. Rationalism was back with a vengeance.

Not only was rationalism no longer obsolete or out of touch with science, it was even ahead of the game. Kant had not only stopped the rot but even brought rationalism back on the winning side. Kant’s account not only showed that some parts of classical rationalism could be saved; it also provided the best available account of how the success of Newtonian science could be explained philosophically. Where people like Euler had merely given up on philosophy because of the magnitude of the problems it faced, Kant had shown that philosophy could answer the challenge and more. Philosophy was relevant again. Philosophy was no longer dead.

]]>**Transcript**

Here’s a fundamental problem in the philosophy of mathematics. You can sit in an isolated room, in an arm chair, and prove theorems about triangles, such as the angle sum of a triangle or the Pythagorean theorem. When you do this, you have the feeling that you have established these results with absolute certainty. You feel that they must be true because of how compelling the proof is. And you feel that you have established this by thought alone, by purely intellectual means.

Mathematics is unique in this respect. In other subjects, thinking is a powerful tool, but it is always supplemented by observation and experience. If you spent your whole life isolated in a locked room, you would not be able to say anything about the laws of astronomy or the anatomy of the digestive system, because without observation, with only pure thought, it is impossible to even get started in those field. But you could figure out everything about triangles. If one day you were released from your prison where you had been sitting for decades, you could go out and measure actual triangles and you would find that, indeed, their angle sum is always two right angles, the Pythagorean theorem always holds for right-angle triangles and so on. Just as you had predicted by pure thought.

This is a bit of a mystery. Because it shows that there are two sides of mathematics that are difficult to reconcile. On the one hand, the internal, mental conviction that mathematics establishes absolute truths purely by reasoning. On the other hand, the external, physical fact that mathematics works in the real world.

What is the bridge between these two worlds? It is as if there is a natural harmony between our minds and the outer world. What is the cause of that harmony?

These two poles can be called rationalism and empiricism. Rationalism takes mathematics to be fundamentally a matter of pure thought. This fits well with the sense we have when doing mathematics, when reading Euclid, that we are establishing absolute truths by sheer reasoning. But it doesn’t explain why mathematics works so well in the physical world.

We have encountered some rationalists already: Plato, Descartes. We saw how Descartes solved the problem. Mathematics is pure thought, and it works in the physical world because the Creator put mathematical ideas in our minds. As the Bible says, “God created man in his image.” That is to say, God created the world based on mathematical ideas, and then created humans and sort of pre-programmed their minds with the same kinds of ideas that he had used to create the world.

So no wonder there’s a harmony between the mental and the physical worlds: they both stem from the same source, the Creator, who used the same principles when designing both. Descartes said basically this quite explicitly, as we recall. Plato pretty much hints at the same idea. God is a mathematician. That is a central belief in Platonist thought as well. And it is a necessary thesis for the rationalists to explain why mathematics works so well.

We have already encountered some empiricist as well: Aristotle, Francis Bacon. They think knowledge ultimately comes from the world around us. From that point of view, it is no mystery that mathematics works on physical triangles. It stems from physical experience to begin with, so of course it conforms to physical experience.

The challenge for the empiricists is instead to explain the mental experience of doing mathematics; our feeling that it brings absolute truth by pure thought in a way that no other subject does. From the empiricist point of view, this feeling is a mistake, a delusion. We think we are doing pure thought, but actually mathematical thought is generalized experience. We think we can sit in a closed room, an arm chair, and figure things out about an outside world that we have never even seen. But it only feels that way.

We have seen and touched many lines and triangles and squares our entire life, since the year we were born. We have internalized this experience. It has become second nature to us. Basic truths of geometry, such as Euclid’s axioms, may feel like core intuitions that are much more pure and absolute and undoubtable than things we know from experience. But that feeling is a delusion, according to the empiricists. Our minds, our feelings have imperfect self-awareness. Just as we are not aware through introspection how our digestive system works, so we are not conscious of the psychological origins of our mathematical intuitions.

I think we can agree that rationalism and empiricism both face big challenges. The challenge for rationalism is to explain why mathematics applies to the physical world. Traditional rationalism had an answer that was very compelling at the time: the explanation in terms of God, the Creator. But nowadays we may want an atheistic answer. And then rationalism is back to square one, facing the original problem all over again, without any solution in sight.

Empiricism doesn’t have that problem, but it has other ones. If mathematics comes from experience, how can it seem so absolute and undoubtable? How can an exact science come from inexact sensory impressions? If mathematics is based on experience like everything else, why does it seem to be such a different kind of knowledge in so many respects? Those are challenges for the empiricist to answer.

It matters how you answer these questions. It shapes the kind of science that you do.

Consider for instance Kepler, the 17th-century astronomer. He was another rationalist. As Kepler says: “Nature loves [mathematical] relationships in everything. They are also loved by the intellect of man who is an image of the Creator.” That’s almost word for word how I described the rationalist position just moments ago.

Kepler felt that the world was designed with the intent that we should study the universe mathematically. As he says: “Whenever I consider in my thoughts the beautiful order [of the universe] then it is as though I had read a divine text, written onto the world itself saying: Man, stretch thy reason hither, so that thou mayest comprehend these things.”

In fact, scientific facts support this view, in Kepler’s opinion. For example, as he says, “Sun and moon have the same apparent sizes, so that the eclipses, one of the spectacles arranged by the Creator for instructing observing creatures in the orbital relations of the sun and the moon, can occur.”

That is indeed a striking fact: that the moon is exactly the right size to precisely block out the sun at the moment of a solar eclipse. From the point of view of modern science, this is a remarkable coincidence. It’s pure chance that the moon is exactly the right size.

You can understand why the explanation in terms of purpose was more compelling in Kepler’s time. Witnessing a solar eclipse is a spiritual experience. It all seems so perfect. Much too perfect to chalk it up to chance. It’s very disappointing that modern science offers nothing more than this non-explanation of such an emotionally compelling spectacle.

And not just modern science. Such views were around already in Kepler’s time. Atomism is a classical worldview that is indeed happy to attribute almost everything, eclipses included, to chance and randomness. According to Kepler’s teacher, Melanchthon, such views “wage war against human nature, which was clearly founded to understand divine things.”

So here we have again that double challenge to empiricism. If mathematics is just one type of knowledge among many that we pick up from experience, then, first of all, why does the universe show so many signs of being mathematically designed? Like the thing with the eclipses, but there are also countless other examples one could use to make this point. Empiricism has no answer to this. It thinks that’s all just a bunch of coincidences, and we are just fooling ourselves by looking for purpose and design that isn’t there.

And secondly, if empiricism is right, and mathematics is just experiential knowledge like everything else, then why does mathematical reasoning feel so uniquely compelling and convincing? As Melanchthon says, mathematics is as natural to a human being as “swimming to a fish or singing to a nightingale.” Just as animals are born with these instincts, so our minds are innately predisposed to do mathematics. Empiricism does not explain why that is the case, or why that seems to be the case.

So it’s understandable that Kepler was a convinced rationalist instead. And this conviction shaped his scientific work. Astronomers are “priests of the book of nature,” as Kepler said. So he was always looking for meaning and purpose and design.

For example, the telescope was a new invention in Kepler’s time, and it was a big moment when the moons of Jupiter were discovered. Kepler immediately looked for the purpose behind the existence of these moons. He concluded that Jupiter must be inhabited. Why else would it have moons? As Kepler says: “For whose sake, the question arises, if there are no people on Jupiter to behold this wonderfully varied display with their own eyes? We deduce with the highest degree of probability that Jupiter is inhabited.”

Another of Kepler’s attempts at uncovering divine design was his theory of planetary distances. According to Kepler, the Creator had chosen the number and position of the planets according to a very beautiful and pleasing mathematical design. Namely, a plan based on the five regular polyhedra.

Euclid discusses the regular polyhedra at length in the Elements. There are precisely five of them, as Euclid indeed proves in the very last theorem of the Elements.

Kepler figured God was as fascinated by these shapes as Euclid had been. So when God asked himself how many planets there should be in the solar system, and how far from the sun to put them, God figured that the most mathematically pleasing way would be to choose six planets, and to have the spaces between them chosen in such a way that the five regular polyhedra fit between them like a nesting doll.

Kepler’s theory in fact fit the data very well. You could calculate planetary distances from astronomical measurements, and you could calculate size proportions of the regular polyhedra from Euclid’s Elements. If you put these things side by side in two columns they come out remarkably close to one another.

So again Kepler explained things that modern science doesn’t explain at all. Why are there six planets? Why are they positioned at those particular distances form the sun? Why does the moon fit precisely on top of the sun during an eclipse?

Kepler explained all of these things. If you accept the basic outlook that it makes sense to think of the creator of the universe as a Geometer, then Kepler’s explanations are very good. This is Kepler, the best mathematical astronomer of his age. These are not some whimsical religious musings. It’s very serious science. Very good science, one might argue.

Meanwhile, modern science doesn’t explain any of these things. There is no explanation, there is no why, according to modern science, of course. It’s all just chance. The solar system was formed by a bunch of random rocks getting caught in a gravitational field. Whatever positions they took up is just random.

It’s easy for us to judge Kepler. But shouldn’t science explain more things as it develops? Not fewer things. You would think that science should take things that are not explained and explain them. Instead of taking things that are already explained and attributing them to coincidence instead. And yet that is precisely what happened when Kepler’s theories were abandoned.

In any case, this Kepler stuff is interesting for all kinds of reasons, but for our purposes, what I wanted to show was that it matters whether you are a rationalist or an empiricist. Rationalism, as we saw, almost requires the hypothesis that God was a Geometer, just as Plato and Descartes and Kepler all said. And that assumption has major implications for how you practice mathematical science. It suggests looking for deliberate design put into the world by a mind that is essentially like our mind, as far as mathematics is concerned.

So that’s one way in which the rationalism-empiricism divide strongly shaped scientific practice in the early 17th century. But that was not the end of it. Here’s another example: the contrasting ways in which Descartes and Newton approached cubic curves.

Cubic curves are the next step beyond conic sections. Conic sections are curves of degree 2. They were studied in great depth by the Greeks. Cubic curves are called cubic because that have degree 3. So they are the more complicated cousins of the conic sections. In the 17th century, this was natural direction to take geometry: to understand curves of degree 3 and higher in the same depth that the Greeks had understood conic sections.

For instance, conic sections come in three classes: ellipse, parabola, hyperbola. Can one find an analogous way of classifying cubic curves? There are going to be more classes because cubics are more complicated. But maybe with the right principle of taxonomy one can impose order among their variety in way that is as useful as the division into ellipse, parabola, and hyperbola is in the theory of conics.

Newton did precisely this. He gave a very detailed and advanced technical study in which he classified cubic curves in several different ways. He divided cubic curves into “species” as he says. That’s Newton’s own term, and it’s a vivid one.

Taxonomising curves into “species” makes Newton sound like a pioneering explorer-scientist forging into unknown jungles and studying all the strange creatures. When you find a new exotic insect, you put it under a microscope and study all its properties. How many legs does it have, how many eggs does it lay, and so on. It’s the same when studying curves. How many crossing points, how many inflections points, and so on. It’s the zoology of mathematics.

This metaphor fits very well with the epistemological ideals of empiricism. You learn by studying the great diversity of things out there. Into the jungle! That’s the call of empiricism. That’s how you learn things. By immersing yourself in the unknown.

“The best geologist is one who has seen the most rocks.” That’s another slogan of empiricism. Experience is the source of knowledge, in other words. If you want to understand rocks, you need to look at a whole lot of rocks. And if you want to understand cubic curves, you need to look at a whole lot of cubic curves, first of all. Once you have built up a store of experience, then maybe you will see some patterns starting to emerge and you can begin the process of systematising or taxonomising the “rocks.”

Empiricism is all about diving in at the deep end and figuring it out as you go. This corresponds to reading Euclid backwards. You start with the complicated stuff, the Pythagorean theorem and such things. Those kinds of things are the exotic beasts that you encounter “in the jungle.” Gradually, you seek to bring order into the chaos by finding general principles that account for the phenomena you observe.

That’s empiricism. And it’s completely backwards according to rationalism. That’s not how you learn things. You can’t start with observations, with the phenomena. Perception is unreliable. Aimless exploration unguided by the intellect is bound to be a waste of time leading nowhere.

The way to knowledge is thinking. To “meditate,” as people used to say. You have heard of Descartes’s Meditations. That’s even the title of one of his works. The source of knowledge is meditation. That is to say, deep thought where you basically close yourself off from the world. Sitting in an armchair in a closed room. That’s where you make progress in understanding, not running around in the jungle.

So Newton’s way to study cubic curves was the empiricist way. Get your machete out and start chopping your way through the thick of it. Eventually you become familiar with all these wild things you encounter, and you start to see what kinds of species there are and how they are related.

Descartes was the opposite of this. A rationalist. Descartes studied cubic curves too, but through meditation. His big book is La Géométrie (1637). He doesn’t study cubics specifically, but all algebraic curves. So curves of any degree, not just degree 3.

Already we see a typical rationalist characteristic: rationalism starts from the general; empiricism starts from the specific.

Rationalists withdraw into meditation because they do not trust individual observations. Thought is more reliable. If you sit back in an armchair and introspect about what is knowable, you are bound to come up with very general and abstract truths: I think therefore I am; the whole is greater than the part; two lines cannot enclose a space. Gradually, you have to work your way from there, step by step, to any specific fact you need to explain. Just as Euclid gradually works his way up to more and more complex and detailed material by starting with very general principles that ultimately entail all the rest.

So the rationalist in interested in all-encompassing abstract law or axioms. It is important to the rationalist that all truths can in principle be deduced from these axioms. But it’s less important to actually do this. The rationalist is most interested in the fundamental axioms or laws, because those are the source of the certainty of knowledge. The specifics derived from them merely inherit their certainty from the certainty of these foundational axioms.

So the very first principles of the entire field is where you need to focus your attention if you are a great rationalist philosopher. And that’s exactly what Descartes does in his book, La Géométrie. Even the title fits with this point of view: The Geometry; it’s a very total, definitive account of geometry as a whole, just as the rationalist epistemological ideal demands.

This is further confirmed in the very first sentence of the text: “All the problems of geometry …”––that’s how Descartes opens his book. He starts with extreme generality, just as rationalism suggests one should. He wants to find the principles that can be used to solve “all the problems of geometry,” in principle.

Descartes doesn’t care so much about the details. He is very keen to explain why his principles are sufficient to solve “all the problems of geometry,” but has very little patience for actually solving any of those problems. This is reflected in the very last sentence of his book.

Descartes writes: “I hope that posterity will judge me kindly, not only as to what I have explained, but also as to what I have intentionally omitted so as to leave to others the pleasure of discovery.”

This is a bit dishonest, of course. He did not omit the details merely out of kindness to the reader, obviously. His focus on the general and lack of interest in the specific is a consequence of his rationalist outlook.

Newton is the opposite. He loves the details; he loves getting stuck in with some obscure technical problem. In fact, his long treatise on cubic curves is full of technical details but he gives very little attention to explaining any general conclusions. It’s hard to see the forest for the trees.

That’s good empiricism, of course. Rationalism thinks you can trust specific results because they are derived from reliable general principles. The certainty of knowledge resides in the axioms, the general principles. That’s where you need to focus your attention to secure the rigour and reliability of reasoning. And that’s what Descartes does.

Empiricism looks at it the other way around. It is the details, the little things, that are the most knowable. Knowledge starts from the directly observed phenomena, with all their specificity. That’s the root of reliability and certainty. Abstract principles are trustworthy only insofar as they are inferred from a large body of facts.

It’s the same in physics. To Newton, the empiricist, the starting points are specific facts. The orbital time of Jupiter, the speed of Saturn. Specific observable facts. You have to start there and then infer general laws like the law of gravity by showing that it fits a long list of facts. It is the specific facts that give credibility to the general law.

Not so to Descartes. The introspective, meditative, rationalist way of doing physics is to figure out first what properties of moving bodies are the most undoubtable. What are the things that are like Euclid’s axioms, but for mechanics?

Descartes did physics exactly this way. In his view, the most undoubtable core principles of physics are the laws of collision of two bodies. If one body bumps into another, what happens? Well, if one is twice as heavy but they have the same speed, then so-and-so happens; if one is twice as heavy but the other is twice as fast, then so-and-so happens; etc. Those are the kinds of principles that Descartes thought one could establish through pure thought and meditation.

Descartes saw this as analogous to Euclid’s geometry. Euclid’s axioms are about lines and circles: the basic building blocks of all geometrical figures. More complex figures are built up from there by combinations of lines and circle, or ruler and compass. In the same way, in physics, complex phenomena can be regarded as ultimately generated by the simple root phenomenon of the collision of two bodies.

Indeed, modern science kind of agrees about that part. If you exhale on a cold day, you breath forms a cloud that moves in complex ways. It seems to flow or float, but really it’s just lots and lots of tiny molecules crashing into each other millions of times, and that gives rise to this kind of flowing pattern that you see on a larger scale.

So simple generative principles can be enough to account for all kinds of things behind their immediate reach, though elaborate repeated composition. Just as lines and circles kind of “give birth” to all geometry, including very complicated shapes that aren’t just round or straight.

Actually, lines and circles are not enough to generate all geometry. They can’t generate cubic curves for example. Descartes is very interested in this issue. And indeed, in his book La Géométrie, he supplements the ruler and compass with another basic generative principle for drawing curves. A kind of linkage principle. You can build a sort of machine that consist of multiple rulers and pegs interlinked in certain ways, and as you push one part of the machine the other parts move in specific ways because of how all the parts are interconnected. An ordinary compass is sort two rulers nailed together. In the same way you can make more elaborate devices composed of more rulers. This gives rise to “new compasses,” as Descartes calls them. And these are sufficient to encompass “all the problems of geometry,” according to Descartes.

In a way it might seem contradictory that it was the rationalists, like Descartes and Leibniz, who were so concerned with the making of geometrical figures with concrete devices. Shouldn’t a proper rationalist hate physical instruments, like Plato did?

But there is no contradiction. Descartes cared about geometrical instruments for theoretical reasons. As I just emphasised, constructions in geometry go naturally with the general rationalist idea of the mind generating all knowledge from within itself. It’s a form of self-reliance. It doesn’t need anything from the outside world.

And earlier we have spoken about how constructions are connected to the epistemological foundations of geometry. Maker’s knowledge. Constructions are the most knowable thing, and the most secure form of geometrical knowledge, protected against many threats of paradoxes and contradictions. So that’s another way in which constructions go well with rationalism, which is of course very much concerned with what are the most undoubtably knowable things.

So these instruments like the ruler and compass and the generalisations of them that Descartes conceived are theoretical, not practical. There’s a funny anecdote that sums this up in the Brief Lives by Aubrey—a late 17th-century collection of biographical stories, maybe not super reliable exactly but this story could very well be true. Here’s what this biographer Aubrey says:

“[Descartes] was so learned that all learned men made visits to him, and many of them would desire him to show them his instruments. He would drawe out a little drawer under his table, and show them a paire of Compasses with one of the legges broken: and then, for his ruler, he used a sheet of paper folded double.”

Quite amusing, and it fits with what I said about the constructions being theoretical.

So we see that the idea of drawing curves with instruments in geometry is analogous to the idea of explaining all of physics in terms of collisions of little bodies. They are both simple, intuitable principles that generate the entire world of phenomena.

From a rationalist point of view, you need such principles. You start in the simple and pure world of meditation and you need to reason your way to the complicated and messy outside world. So you need a bridge that goes from the simple to the complex. Contact mechanics is such a bridge in physics, and ruler and compass is such a bridge in geometry.

But this is only necessary if you are a rationalist. If you insist on starting with pure intuition and thought, then you need such a bridge to the phenomena and the outside world.

But if you are an empiricist you take the outside world—the jungle—for granted as given, as a starting point, so you don’t need to explain how it can be generated by repeated composition of simple principles.

Indeed, Newton rejects both contact mechanics and geometrical constructions at the same time, for precisely this reason.

The fact that these two things are intimately related is not lost on Newton. This is why he starts his big masterpiece on physics by talking about the construction of line and circle in geometry. A very weird way to start a physics treatise to modern eyes, but it makes perfect sense if we keep in mind the background of Descartes and rationalism and everything I just outlined.

I’m referring to Newton’s Principia of 1687. Descartes was long dead by then, but his ideas about the foundations of physics were as relevant as ever. Leibniz, who was a contemporary of Newton, was a rationalist like Descartes. Like Descartes, Leibniz attached great importance to contact mechanics in physics and constructions in geometry.

So when Newton’s Principia came out, Leibniz was very upset that Newton had abandoned the principle of contact mechanics, which was so essential to the entire rationalist worldview. Let me quote Leibniz on this point. Here’s what he said: “A body is never moved naturally except by another body that touches and pushes it. Any other kind of operation on bodies in either miraculous or imaginary.”

Newtonian gravity is precisely one such “other operation”; something that cannot be explained in terms of particles bumping into one another. This is why Leibniz condemns very fiercely the notion of gravity as a foundational principle of physics: “I maintain that the attraction of bodies is a miraculous thing, since it cannot be explained by the nature of bodies.”

That is to say, Newton’s law of gravity cannot be explained or arrived at from a rationalist point of view. Newton in fact agreed. If anything, he makes this point in even stronger terms than Leibniz. Here’s what he says: “It is inconceivable that inanimate brute matter should, without the mediation of something else, which is not material, operate upon and affect other matter without mutual contact, as it must be, if gravitation be essential and inherent in it. That gravity should be innate, inherent, and essential to matter, so that one body may act upon another at a distance through a vacuum without the mediation of anything else, is to me so great an absurdity, that I believe no man who has in philosophical matters a competent faculty of thinking can ever fall into it.”

Very strong words there from Newton. And we can understand why. He wants to discard the rationalist outlook entirely. He is not interested in winning broad support for his theory by trying to argue that it sort of fits with rationalism somehow. He could have given that a shot. He clashed with many influential people: Descartes, Huygens, Leibniz. He could have tried to go a diplomatic route and try to come up with reasons for why his way of doing science was compatible with their rationalist commitments. But he chose not to. This is why he comes on so strongly in these quotes about how gravity is rationally inconceivable and so on.

In this way, Newton moves the conflict into the area of rationalism versus empiricism generally, instead of arguing about the interpretation or meaning of gravity specifically. “With the cause of gravity I meddle not,” says Newton, since “I have so little fancy to things of this nature.”

So what Newton wants to justify is not gravity specifically, but a the empiricist way of doing science generally, in which you don’t care about such questions at all. Questions such as how to give a rationalistic account of gravity, or explaining how a meditating mind in an armchair could arrive at the necessity of the law of universal gravitation. Those questions should simply be ignored, says Newton. Which makes sense from an empiricist points of view, but is sheer madness from a rationalist point of view.

So Newton bites the bullet on the cause of gravity. He says: yeah, I know my physics completely clashes with the core beliefs and methodology of rationalism, but rationalism is wrong anyway.

Now, as I said, the role of contact mechanics in physics is analogous to the role of constructions in geometry. Newton knows this, and this is why, to justify his physics, he starts by talking about how to interpret the role of constructions in geometry. Here is what he says right at the beginning of the Principia:

“The description of right lines and circles, upon which Geometry is founded, belongs to Mechanics. Geometry does not teach us to draw these lines, but requires them to be drawn. For it requires that the learner should first be taught to describe these accurately, before he enters upon Geometry; then it shews how by these operations problems may be solved. To describe right lines and circles are problems, but not geometrical problems. The solution of these problems is required from Mechanics; and by geometry the use of them, when so solved, is shewn. And it is the glory of Geometry that from those few principles, fetched from without, it is able to produce so many things. Therefore Geometry is founded in mechanical practice.”

So that’s a clearly empiricist account of geometry. Not only because it obviously grounds geometry in the physical world, in physical practice and experience. But also because it takes away the idea that the axioms need to be justified by being intuitive and undoubtable. That was important to the rationalists, but Newton does away with that.

This is how Newton can justify that he “meddle not with the cause of gravity.” Geometry likewise doesn’t “meddle” with the construction of curves, but merely postulates their description—in fact, geometry postulates these things precisely “because it knows not how to teach the mode of effection,” just as physics does not know how to teach the cause of gravity.

So Newton has twisted Euclid into support for his physics. This is why the preface to the Principia is about constructions in geometry, such as the ruler and compass of Euclid. If geometry doesn’t really know how to generate these curves, but only takes them for granted and goes from there, then physics can do the same with gravity.

So Newton and Leibniz clashed along such lines. And not only them. One could argue that there’s a geographical element to this divide. Empiricism is to some extent a British movement more generally: not only Newton but also Francis Bacon, John Locke, Wallis—just to name some people we have already encountered before. Meanwhile, Leibniz’s rationalistic tendencies in his science and mathematics were shared by his leading colleagues in Continental Europe, such as Descartes and Huygens.

By way of summary, let me read a passage by Newton on his scientific method, and I will insert comments on how what he says fits exactly with what we have discussed. The passage begins:

“As in mathematics, so in natural philosophy, …”

Already very interesting. In other words, Newton is announcing that his scientific method is based on the method of mathematics; the method of Euclid basically. Ok, so what is this this methodological principle that is common to both mathematics and science? The sentence continues:

“… the investigation of difficult things by the method of analysis, ought ever to precede the method of composition.”

Analysis corresponds to reading Euclid backwards. To analyse is to break down into smaller pieces. Composition corresponds to reading Euclid forwards. To compose is to put simpler pieces together to form more complex results. Newton continues:

“Analysis consists in making experiments and observations, and in drawing general conclusions from them by induction. By this way of analysis we may proceed from compounds to ingredients, and from motions to the forces producing them; and in general, from effects to their causes, and from particular causes to more general ones, till the argument end in the most general.”

“General” indeed: a key words here. From observations, that is to say from specific facts, one infers more general underlying principles. Empiricism goes from the specific to the general; rationalism the other way around.

It is also nice that Newton mentions that analysis goes “from compounds to ingredients”: this is precisely the chemistry or cooking metaphor that we used before when discussing how to read Euclid backwards.

Newton continues:

“This is the method of analysis, and the synthesis consists in assuming the causes discovered, and established as principles, and by them explaining the phenomena proceeding from them, and proving the explanations.”

That is to say, reading Euclid forwards is of course also essential. The method of analysis that the empiricist uses does not dispense with this directions of Euclid; it merely reveals that a preliminary stage is necessary to understand its meaning. It is through the preliminary analysis, the backwards direction of reasoning, that one arrives at the principles—not by direct intuition, as the rationalists would have it. Then the forward direction, the composition or synthesis, proves that these principles really work; that is to say, that they are sufficient to prove everything. That part is the same to both rationalists and empiricists. The key difference is how they account for where the principles or axioms came from.

So those are Newton’s own words, corresponding very closely to the story I have told. Of course Newton and Leibniz and all these guys were acutely aware of all of this. In this way they were much more philosophically conscious than most scientists of later ages. And clearly it shaped their science very profoundly, as I have shown by several examples. So that’s all the more reason to keep pursuing these questions. As indeed we will.

]]>**Transcript**

Gothic architecture is known for its pointed arches. Unlike round arches like a classical Roman aqueduct for example. Those are semi-circular, but Gothic arches are steeper, pointier. Gothic buildings, cathedrals, have these arches everywhere: windows, doorways, and so on.

Gothic arches consist of two circular arcs. You can make it like this. First make a rectangular shape. Like a plain window or door. A boring old rectangle. Now let’s spice it up. Take out your compass, and put it along the top side of the rectangle. Draw two circular arcs going up above the rectangle. Use the top side of the rectangle as the radius, and its two endpoints as the two midpoints of the two arcs you are drawing. The two arcs make a pointed extension of the rectangle. Now you have your Gothic window.

If you have your Euclid in fresh memory you will recognize at once that this is precisely the type of construction involved in Proposition 1 of the Elements. Coincidence? No, I don’t think so. The Gothic style of architecture arose in Europe in the early 12th century, within a decade or two of the first Latin translation of Euclid’s Elements. If that’s not cause and effect, it‘s an incredible coincidence.

There is little direct documentation about this, but, I am quoting now from Otto von Simson’s book The Gothic Cathedral, “at least one literary document survives that explains the use of geometry in Gothic architecture: the minutes of architectural conferences held in 1391 in Milan. The question debated at Milan is not whether the cathedral is to be built according to a geometrical formula, but merely whether the figure to be used is to be the square or the equilateral triangle. The minutes of one particularly stormy session relate an angry dispute between the French expert, Jean Mignot, and the Italians. Overruled by them on a technical issue, Mignot remarks bitterly that his opponents have set aside the rules of geometry by alleging science to be one thing and art another. Art, however, he concludes, is nothing without science, ars sine scientia nihil est. This argument was considered unassailable even by Mignot’s opponents. They hasten to affirm that they are in complete agreement as regards this theoretical point and have nothing but contempt for an architect who presumes to ignore the dictates of geometry.”

So the geometrical ethos was very strong indeed. This hardline view probably softened a bit over time. Renaissance art is more expressive, emotive, more alive, one might say, than this rigid late medieval stuff. That’s if we fast-forward two hundred years from these Gothic conferences about how art is nothing without geometry.

Then you have people like Michelangelo who said: “the painter should have compasses in his eyes, not in his hands.” I suppose it means that art should go a little more by feeling and intuition, and not be completely dictated by mathematics. But you still have “compasses in your eyes,” so there’s still a very significant role for geometry, it seems.

It is also revealing, perhaps, that Michelangelo thought it was important to point this out at all. I guess there were a lot of artists with compasses in their hands running around back then. Why else would Michelangelo feel the need to criticise that practice?

In fact, geometry proved useful to art again, in new ways, in the Renaissance. At this time artists discovered (or perhaps rediscovered) the geometrical principles of perspective.

Accurate representation of depth in a painting follows simple geometrical principles. The key construction is that of a tiled floor. Like a chessboard type of pattern of floor tiles, but seen in perspective, so tiles that are further away appear smaller in the picture. There are a lot of tiled floors in Renaissance art, because they are great for conveying a sense of depth.

You draw it like this. Draw two horizontal lines: one is where the floor starts, and one is the horizon. Divide the floor line into many equal pieces, representing the size of the tiles. Connect all these points to a fixed point on the horizon. This is because all the parallel rows of the floor will appear to converge in one point, just as whenever you are looking at parallel lines that go off into the distance, such as railroad tracks for instance, they appear to meet at the horizon.

Next, draw a second horizontal line representing the other edge of the front row of floor tiles. Now here comes the magic step. Draw the diagonal of the first tile, and extend it. Where this line cuts the other lines you have already drawn, those are all corners of other tiles.

This is because of the geometrical principle that a straight line, no matter how you look at it, from whatever angle, will always still be straight. A lot of stuff looks funny in perspective: big things can look small, parallel lines appear to meet, perfectly round things appear to be oval, and so on. Perspective distorts shapes in all kind of ways. But straight lines remain straight lines. That is an invariant in all of this. This is why the diagonal of the first floor tile is also the diagonal of successive tiles. Since that makes a straight line in the real world, it must make a straight line in the picture as well.

Once you have this diagonal it is easy to complete the rest of the floor. It looks great. It creates a photorealistic sense of depth. No wonder so many artists chose to set their scenes in locales that just happened to have lots of tiled floors.

But the insight is much deeper than this. It’s not really about tiled floors; it’s about the correct perspective representation of depth and size generally. Even if you don’t want a tiled floor in your finished picture, it’s still very useful to draw one in pencil on your canvas as a reference grid. You can use the tiles as a guide to transfer and compare sizes at different depths and distances. Then later you can paint it over with some trees or whatever, so nobody can see the grid anymore; you just used it behind the scenes to get the proportions right.

The Greeks probably knew about this stuff, but basically no paintings from antiquity have been preserved. But we know they were skilled artists. One guy is said to have painted grapes so realistically that birds came and pecked at it.

It seems the Greeks knew the geometrical principles of perspective and used it to creates scenes for the theatre for example. As Vitruvius says, “by this deception a faithful representation of the appearance of buildings might be given in painted scenery, so that, though all is drawn on a vertical flat facade, some parts may seem to be withdrawing into the background, and others to be standing out in front.”

This wasn’t just a party trick to the Greeks. It also had philosophical implications. To Plato they raised profound epistemological conundrums. He was concerned that optical illusion painting has “powers that are little short of magical,” “because they exploit this weakness in our nature,” bypassing “the rational part of the soul.” The solution to this problem, as Plato saw it, was a solid mathematical education. Since “sense perception seems to produce no sound result” with these illusory paintings, “it makes all the difference whether someone is a geometer or not.”

“The power of appearance often makes us wander all over the place in confusion, often changing our minds about the same thing and regretting our actions and choices with respect to things large and small.” “The art of measurement,” by contrast, “would make the appearances lose their power” and “give us peace of mind firmly rooted in the truth.”

Those are all Plato’s words. A rousing case for mathematics! But Plato perhaps drew his conclusions a step too far, rejecting categorically the role of observational data in science: “there’s no knowledge of sensible things, whether by gaping upward or squinting downward.” Science must be based on “the naturally intelligent part of the soul,” not observation. For example, “let’s study astronomy by means of problems, as we do geometry, and leave the things in the sky alone.” With such attitudes, perhaps it is no wonder that the Greeks excelled more in mathematics than in the sciences. But indeed the threat of optical illusions is a legitimate argument in Plato’s defense.

When the principles of perspective were rediscovered in Renaissance Italy, they were again at the heart of scientific developments, but this time on the side of empirical science. Galileo looked at the moon through a telescope and concluded that it had mountains and craters. Of course the image one sees through a telescope is flat. But mountains and craters are revealed by the shadows they cast.

This is not necessarily very easy to see, or not necessarily a very evident conclusion. Some scholars have argued that the artistic tradition, and its extensive study of perspective and shadows, was a necessary training for the eye to be able to correctly interpret the telescope data.

Is it a coincidence that Galileo the telescopic astronomer came from the same land as the great Italian Renaissance painters? Galileo was born and raised in Tuscany, right where so many of these masters had worked. Maybe only someone immersed in this artistic culture had the right eyes to interpret the heavens. A far-fetched theory, in my opinion, but it’s a nice story.

Let’s put aside the art stuff now and look at another theme in how mathematics was received in the early modern world. Namely, the status of mathematics in relation to other fields. Geometry carried a certain authority. This led to many tensions.

Let’s jump right into the action, with an eyewitness report from 1703. “There has been much canvassing and intrigue made use of, as if the fate of the Kingdome depended on it.” “On the eve of Newton’s election as president [of the Royal Society], matters had deteriorated to such an extent that various fellows could be restrained only with difficulty from a public exchange of blows (or, in one case, the drawing of swords).”

Yikes. So what was this conflict on which “the fate of the Kingdome” depended? It was a battle between the mathematical and the non-mathematical sciences within the Royal Society in London.

The “philomats” who identified with Newton thought the non-mathematical sciences were hardly science at all. Botany, geology, stuff like that. They just collect data and write down obvious things. There’s no real thinking involved, no advanced theoretical progress, no genius.

Here’s how they put it, when they made the case that Isaac Newton, the great mathematician, ought to be the new president of the society to ensure its intellectual quality: “That Great Man [Newton] was sensible, that something more than knowing the Name, the Shape and obvious Qualities of an Insect, a Pebble, a Plant, or a Shell, was requisite to form a Philosopher, even of the lowest rank, much more to qualifie one to sit at the Head of so great and learned a Body.”

So science is divided into two camps: mathematical geniuses like Newton, and then people who just know the names of a bunch of insects.

As you can imagine, the other side saw it rather differently. They identified with Francis Bacon, who had complained about “the daintiness and pride of mathematicians, who will needs have this science almost domineer over Physic. For it has come to pass, I know not how, that Mathematics and Logic, which ought to be but the handmaids of Physic, nevertheless presume on the strength of the certainty which they possess to exercise dominion over it.”

So mathematicians have an inflated ego. They are so full of themselves that they think they have the right to tell others how to think.

Here’s how this point was put in 1700: “The World is become most immoderately fond of Mathematical Arguments, looking upon every thing as trivial, that bears no relation to the Compasse, and establishing the most distant parts of Humane Knowledge, all Speculations, whether Physical, Logical, Ethical, Political, or any other upon the particular results of number and Magnitude. In any other commonwealth but that of Learning such attempts towards an absolute monarchy would quickly meet with opposition. It may be a kind of treason, perhaps, to intimate thus much; but who can any longer forbear, when he sees the most noble, and most usefull portions of Philosophy lie fallow and deserted for opportunities of learning how to prove the Whole bigger than the Part.”

So mathematics corrupts mind and soul by fostering delusions of grandeur, and by focusing on obscure technical questions instead of on what is really important.

Roger Ascham made a similar point in 1570: “Some wits, moderate enough by nature, be many times marred by over much study and use of some sciences, namely arithmetic and geometry. These sciences sharpen men’s wits over much. Mark all mathematical heads, which be wholly and only bent to those sciences, how solitary they be themselves, how unapt to serve in the world.”

Meanwhile, the mathematicians, for their part, thought that an exclusive focus on the merely practical is anti-intellectual and beneath a true thinker. Others scientists may use basic mathematics, but the real accomplishment is to understand it.

Mathematician William Oughtred put it like this: “The true way of Art is not by Instruments, but by demonstration. It is a preposterous course of vulgar Teachers, to beginne with Instruments, and not with the Sciences, and so to make their Schollers onely doers of tricks, and as it were jugglers.”

Very relatable for a modern mathematics teacher. Students are so dependent on calculators that they are “onely doers of tricks.” That’s what you get when mathematics is not respected as an end in itself, but only as a tool for what is practically useful.

There’s an interesting twist to this story though. Part of what these opponents of mathematics were attacking was the pedantic focus on theoretical subtleties. Instead of tackling real problems, mathematicians sit around and muse about nuances of definition and postulates that only matter for very subtle foundational debates, not for actual problem solving. A valid critique, you might say, after reading Euclid with all his foundational pedantry.

But here’s the twist: Many mathematicians didn’t like that stuff either. Many mathematicians in the 17th century felt that the Greek geometrical style was much too formal. They recognized the value of the Euclidean style for foundational investigations, but they felt that creative mathematics must be much more free and loose.

Here’s how Clairaut put it in the 18th century:

“[Euclid’s] geometry had to convince stubborn sophists who prided themselves on refusing [to believe] the most evident truths. It was necessary then that geometry have the help of forms of reasoning to shut the idiots up. But times have changed. All reasoning which applied to that which good sense knows in advance is a pure loss and serves only to obscure truth and disgust the reader.”

This fits pretty well with what we have said about the Greek context. Euclid’s special style of geometry arose in a critical philosophical climate. Mathematicians had to anticipate attacks from philosophers who wanted to undermine the entire notion that geometrical reasoning was a rigorous way of finding truth.

Without this external pressure from philosophy, mathematicians may have been happy with a much more informal style, as they were in other cultures and societies. And as indeed they became again in the 17th century.

Almost all mathematicians in the 17th century were very happy to take a freewheeling approach for example when exploring a lot of stuff related to what we call calculus today. For example, John Wallis, a leading mathematician, did work on infinite series that was based on daring, unrigorous extrapolations and generalisations, which he considered “a very good Method of Investigation which doth very often lead us to the early discovery of a General Rule.” In fact, “it need not any further Demonstration,” according to Wallis.

This is very unlike Euclid or anything you find in Greek sources. It’s explorative trial and error, and a readiness to trust the patterns and rules you discover without the minutiae of carefully writing out meticulous proofs of every little thing.

When mathematicians chose this approach, they did not think of themselves as going against the ancient tradition. Instead they imagined—and they were probably right, of course—that Greek mathematics too would have been developed this way, in an informal way.

Euclid’s style of mathematics is very powerful for certain foundational purposes, but of course Euclid’s proofs do not reflect how people initially discovered these things. There must have been an exploratory side to Greek mathematics that is not revealed in surviving sources.

Euclid’s Elements is the end product of a long process of discovery and exploration. That process would not have been conducted in the pedantic and overly polished style of the finished Elements. It is necessary to start with a much freer creative phase. Then its fruits can be systematized and analyzed in the manner of Euclid.

Torricelli, for example, expressed a view typical among 17th-century mathematicians: “For my part I would not dare to assert that this Geometry of Indivisibles is a thoroughly new invention. Rather, I would have believed that the old geometers used this one method in the discovery of the most difficult theorems, although they would have produced another way more acceptable in their demonstrations, either for concealing the secret of the art or lest any opportunity for contradiction be proffered to envious detractors.” Many mathematicians agreed with Torricelli on this point.

The Greek historian Herodotus says about Persian political leaders that they “deliberate while drunk, and decide while sober.” That’s how you have to do mathematics too. First you need to generate ideas. For this you have to be “drunk,” that is to say, try out wild ideas, be uninhibited. Then you have to go over the same material again while “sober”: that is to say, you scrutinize everything critically, discarding and correcting all the mistakes you made while “drunk.”

The documentation we have for Greek mathematics is only the “sober” part. But there must have been a “drunk” part too. The sober part is what gives mathematics its distinctive precision and exactness and reliability. But the sober part alone is sterile. It needs the fertile input of daring ideas from the drunk part. Creative mathematics requires both.

Note that if you want to create new mathematics, then this is essential to realize. So working mathematicians, research mathematicians, will absolutely agree with his.

But many people in the 17th century wanted to use the example of mathematics to support various agendas, without having any interest in discovering new mathematics. From that point of view, it is possible to ignore the drunk phase. If you are merely preserving and admiring past mathematics, and you don’t need creativity, you don’t need new ideas, then you can stick entirely to the sober mode, the Euclidean mode, and maintain that that alone is the essence of mathematics.

This matters if you want to use the authority and status of mathematics to legitimate other, non-mathematical agendas. Indeed, it suited some people very well in the 17th century to emphasise the “soberness” of Euclid. They wanted mathematics to be like that, because they had political or philosophical ideals that fit that image.

Amir Alexander’s book Infinitesimal has some nice examples of this. Let’s look at those. I mentioned Wallis as an example of a creative mathematician who very much embraced the “drunk” style of mathematics. His arch enemy was Hobbes, who, by contrast, appealed to the authority and rigour of Euclidean geometry as a model for reasoning as well as political organisation.

As Amir Alexander says: “Wallis and Hobbes both believed that mathematical order was the foundation of the social and political order, but beyond this common assumption, they could agree on practically nothing else. Hobbes advocated a strict and rigorous deductive mathematical method, which was his model for an absolutist, rigid, and hierarchical state. Wallis advocated a modest, tolerant, and consensus-driven mathematics, which was designed to encourage the same qualities in the body politic as a whole.”

Wallis’s vision of mathematics was very agreeable to the experimental scientists of the Royal Society. “Experimentalism is a humbling pursuit, very different from the brilliance and dash of systematic philosophers such as Descartes and Hobbes. It ‘teaches men humility and acquaints them with their own errors’. And that is precisely what the founders of the Royal Society liked about it. Experimentalism ‘removes all haughtiness of mind and swelling imaginations’, teaching men to work hard, to acknowledge their own failures, and to recognize the contributions of others.”

“Mathematics, [the Royal Society founders] believed, was the ally and the tool of the dogmatic philosopher. It was the model for the elaborate systems of the rationalists, and the pride of the mathematicians was the foundation of the pride of Descartes and Hobbes. And just as the dogmatism of those rationalists would lead to intolerance, confrontation, and even civil war, so it was with mathematics. Mathematical results, after all, left no room for competing opinions, discussions, or compromise of the kind cherished by the Royal Society. Mathematical results were produced in private, not in a public demonstration, by a tiny priesthood of professionals who spoke their own language, used their own methods, and accepted no input from laymen. Once introduced, mathematical results imposed themselves with tyrannical power, demanding perfect assent and no opposition. This, of course, was precisely what Hobbes so admired about mathematics, but it was also what Boyle and his fellows feared: mathematics, by its very nature, they believed, leads to claims of absolute truth, dogmatism, threats of tyranny.”

But note that this image of mathematics as totalitarian and absolutist is linked to its sober phase. By playing up the liberal, drunk way of doing mathematics, one changes its political implications.

So that’s how things played out in England. Conservatives appealed to Euclid’s rigour to justify hardline reactionary politics, while creative mathematicians saw the freedom of creation and discovery in mathematics as suggesting that society as a whole should have a high tolerance for unconventional ideas and novel approaches.

The situation in Italy was quite analogous. The Jesuits were the intellectual leaders of the Catholic world in the 17th century. They ran hundreds of colleges across Europe, notable as much for their “sheer educational quality” as for their doctrinal role “in the fight to defeat Protestantism.”

The Jesuit colleges placed great emphasis on Euclidean mathematics, which to them “represented a deeper ideological commitment. Geometry, being rigorous and hierarchical, was, to the Jesuit, the ideal science. The mathematical sciences that followed—astronomy, geography, perspective, music—were all derived from the truths of geometry. Consequently, [the Jesuit] mathematical curriculum demonstrated how absolute eternal truths shape the world and govern it,” thereby serving as a model for their religious doctrine and worldview. “Euclidean geometry thus came to be associated with a particular form of social and political organization, which the Jesuits strived for: rigid, unchanging, hierarchical, and encompassing all aspects of life.”

For this reason, “the Jesuits reacted with fury to the rise of infinitesimal methods”—which is “drunk” mathematics. “The mathematics of the infinitely small was everything that Euclidean geometry was not. Where geometry began with clear universal principles, the new methods began with a vague and unreliable intuition that objects were made of a multitude of minuscule parts. Most devastatingly, whereas the truths of geometry were incontestable, the results of the method of indivisibles were anything but,” thereby undermining “the Jesuit quest for a single, authortized, and universally accepted truth.”

Thus infinitesimal mathematics was dangerous to the Jesuits not for intrinsic mathematical reasons but because it was associated with diversity of thought unchecked by authority. As one Jesuit leader declared: “Unless mind are contained within certain limits, their excursions into exotic and new doctrines will be infinite, [which would lead to] great confusion and perturbation to the Church.”

One God, one Bible, one Euclid. Set in stone for all eternity. That’s what these guys wanted, and that’s why they liked Euclid. And that’s why, “in a fierce decades-long campaign, the Jesuits worked relentlessly to discredit the doctrine of the infinitely small and deprive its adherents of standing and voice in the mathematical community.” You have to stifle this dangerous new “drunk” mathematics, in which people think for themselves, explore diverse perspectives, and look for new truths (as if there was such a thing!).

So, in summary, mathematics had many possible connotations that could be exploited to various ends. It’s like when someone becomes a celebrity, everyone wants them to endorse their product or sign their petition and so on. A sponsored post on their Instagram is prime real estate. Mathematics had become a celebrity in the 17th century. It had status, for better or for worse. And everyone wanted a piece of it. Coke or Pepsi, PC or Mac—who would get the coveted endorsement of mathematics? Mathematics never sold out or picked a side, but it’s illuminating to see the pitches the PR departments of all these various movements made on its behalf.

]]>**Transcript**

Everybody has seen the painting The School of Athens, the famous Renaissance fresco by Rafael. It shows all the great thinkers of antiquity engaged in lively intellectual activity. Plato and Aristotle are debating the relative merits of the world of ideas and the world of the senses, both gesticulating to emphasize their point. Others are absorbed in other debates and lectures, somebody’s reading, somebody’s writing.

But here’s something most people don’t notice in this painting. There is one and only one person in this entire pantheon who is actually making something. Everybody is thinking, arguing, reading, writing. Except Euclid. Euclid is drawing with his compass. He is producing the subject matter he is studying. He is active with his hands. He’s practically a craftsman among all these philosophers.

In the ancient world, the mathematician is the maker. Geometry is the most hands-on of all the branches of philosophy and higher learning.

Today the cliche is that a math nerd is almost comically feeble in anything having to do with physical action.

But ancient geometry was in the thick of the action. You had to roll up your sleeves to do geometry. Even in theoretical geometry you would constantly draw, construct, work with instruments. It was a short step to engineering. The greatest ancient mathematician, Archimedes, is almost as famous for his feats in engineering. Such as mechanical devices for lifting and moving heavy objects, and for transporting water. Archimedes and other mathematicians were also at the front lines of war, building catapults and many other warfare machines according to precise calculations. They were architects. The Hagia Sophia in Istanbul for example, was designed by a mathematician, Isidore, who had written an appendix to Euclid’s Elements.

In early modern modern times, like the 17th century, this link between mathematics and concrete action was well understood and appreciated.

Francis Bacon was sick of traditional philosophy because “it can talk, but it cannot generate.” This frustration led him to the radical counterproposal: to know is to do. “What in operation is most useful, that in knowledge is most true.” And on the other hand “to study or feign inactive principles of things is the part of those who would sow talk and nourish disputations.” So we have to condemn much traditional philosophy and turn more to action, to doing.

Perhaps the most important difference between ancient mathematics and ancient philosophy is precisely this. That mathematics is active, while philosophy merely “sows talk and nourishes disputations.” Perhaps that is the explanation for why mathematics proved so fruitful, still thousands of years later, both for intricate theory, such as planetary motions, and for practice, such as engineering, navigation, and so on. Try doing that with Aristotle’s doctrine of causes or Plato’s theory of the soul. Those things are great for “sowing disputations” but if doing is the goal then you can’t get much mileage out of them.

Thomas Hobbes, another famous 17th-century philosopher, very much agreed with this analysis. Hobbes famously declared that “Geometry is the only science that it hath pleased God hitherto to bestow on mankind.” How so? What makes geometry different from all other branches of philosophy and science?

Constructions, of course. Hobbes is very explicit about this. “If the first principles contain not the generation of the subject, there can be nothing demonstrated as it ought to be.” This is what makes mathematics different. Its principles contain the generation of the subject: Euclid’s postulates correspond to ruler and compass, and these are tools that generate the figures that geometry is about.

All philosophical and scientific theories are based on some assumptions or axioms. But they are not generative axioms. They are not a recipe for producing everything the theory talks about from nothing.

In this light we can readily appreciate for instance Hobbes’s otherwise peculiar-sounding claim that political philosophy, rather than physics or astronomy, is the field of knowledge most susceptible to mathematical rigour. Here’s how he puts it:

“Of arts, some are demonstrable, others indemonstrable; and demonstrable are those the construction of the subject whereof is in the power of the artist himself, who, in his demonstration, does no more but deduce the consequences of his own operation. The reason whereof is this, that the science of every subject is derived from a precognition of the causes, generation, and construction of the same; and consequently where the causes are known, there is place for demonstration, but not where the causes are to seek for. Geometry therefore is demonstrable, for the lines and figures from which we reason are drawn and described by ourselves; and civil philosophy is demonstrable, because we make the commonwealth ourselves.”

As bizarre as this may sound to modern ears, it makes perfect sense when we keep in mind the all-important role of constructions in classical geometry.

Indeed there are many things that only the person who made it truly understand. At this time, the 17th century, various mechanical devices were becoming more common. Such as pocket watches and all kinds of other machines based on gears and cogwheels and so on. The person who made it knows what all the parts are for, but an outsider cannot see this very easily at all. Today another example might be computer programs. The person who wrote it knows how it works, what it can do, how it could be changed, what might cause it to fail, and so on. It would be very difficult for someone else to get a similar sense of how it all works, even if they had access to the code, or they could pop the hood and look at the gears so to speak. Only the maker truly knows: “maker’s knowledge” is a slogan often repeated in the 17th century.

Hobbes took this idea and built a general philosophy from it. His general philosophical program can be read as a direct generalisation of the constructivist precept to the domain of general philosophy. Here’s how Hobbes defines philosophy: “Philosophy is such knowledge of effects or appearances as we acquire by true [reasoning] from their causes or generation.” This is basically a direct equivalent in more general terms of the principle that constructions are the source of mathematical knowledge and meaning.

Indeed, Hobbes explicitly draws out this parallel: “How the knowledge of any effect may be gotten from the knowledge of the generation thereof, may easily be understood by the example of a circle: for if there be set before us a plain figure, having, as near as may be, the figure of a circle, we cannot possibly perceive by sense whether it be a true circle or no. [But if] it be known that the figure was made by the circumduction of a body whereof one end remained unmoved” then the properties of a circle become evident. You understand a circle because you make it, in other words.

Another way of putting it is that “The subject of Philosophy, or the matter it treats of, is every body of which we can conceive any generation.” Just as, classically, the domain of geometry is the set of all constructible figures.

Concepts that are not constructively defined can easily be contradictory or meaningless: a common problem outside of geometry. As Hobbes says: “senseless and insignificant language cannot be avoided by those that will teach philosophy without having first attained great knowledge in geometry.”

Again, as we have discussed before, anchoring geometrical entities in physical reality is a warrant of consistency. Hobbes makes this point as well. “Nature itself cannot err”; that is to say, physical experiences “are not subject to absurdity.”

It is notable that Hobbes and other 17th-century thinkers who invoked geometry did not have in mind simple school geometry and some superficial remarks in Plato or Aristotle. Rather, they were referring to the rich picture of the geometrical method that emerges from a thorough study of advanced Greek geometry and technical writers. When they call upon geometry they mean not some simplistic idea of axiomatic-deductive method but a rich methodology only conveyed implicitly in the finest ancient works of advanced geometry.

This is why the constructive aspect shines through so clearly. It’s importance is evident if you study the mathematicians and build your idea of philosophy of mathematics from there. You’re not going to learn anything about that by reading Plato and Aristotle.

Hobbes is very clear about this. As he says, his philosophy of geometry is “to an attentive reader versed in the demonstrations of mathematicians without any offensive novelty.” Indeed, one must be “an attentive reader,” because one must draw out the philosophical implications left implicit in these sources. And one must be “versed in the demonstrations of mathematicians,” meaning the technical Greek authors. As Hobbes calls them, those “very skillful masters in the most distant ages: above all in geometry Euclid, Archimedes, Apollonius, Pappus, and others from ancient Greece.” This is why Hobbes, in one of his works, “thought it fit to admonish the reader that he take into his hands the works of Euclid, Archimedes, Apollonius, and others.”

Many other 17th-century philosophers picked up the same themes. Some took it to the epistemological extreme of saying that anything other than concrete, specific experience is strictly unknowable. Gassendi, for instance, did not hesitate to take this leap: “Things not yet created and having no existence, but being merely possible, have no reality and no truth.” “The moment you pass beyond things that are apparent, or fall under the province of the senses and experience, in order to inquire about deeper matters, both mathematics and all other branches of knowledge become completely shrouded in darkness.” Mathematical objects must be “considered in actual things”; indeed, “as soon as numbers and figures are considered abstractly then they are nothing at all.” Those are all quotes from Gassendi, and his point of view makes sense. He merely spells out the consequence of taking concrete construction to be essential to knowledge, just as the mathematical tradition suggests.

Other philosophers agreed too. Vico put it like this: “We are able to demonstrate geometrical propositions because we create them; were it possible for us to supply demonstrations of propositions of physics, we would be capable of creating them ex nihilo as well.” So once again the link between creation and knowledge is all-important, and geometry is the key example of this.

Paolo Sarpi made much the same point: “We know for certain both the existence and the cause of those things which we understand fully how to make [just as] in mathematics someone who composes [that is to say, demonstrates synthetically, in the manner of Euclid] knows because he makes.”

It’s striking how many of these early modern thinkers who were well versed in the Greek tradition seized upon the constructive element as the essence of the more geometrico, “the manner of the geometers.”

But there were of course other perspectives on mathematics as well. A lot of people read too much Aristotle and not enough Archimedes. Then as now, one might add. Anyway, these Aristotelians didn’t like mathematics much, and they tried to undermine its authority.

Here is their main point of attack: Mathematical proofs, such as those in Euclid, show that the theorem is true, but not why it is true. In other words, mathematics does not demonstrate “from causes,” as a true science should, according to Aristotle.

Here’s one typical expression of this argument, from Aristotelian philosopher Pereyra in the 16th century:

“My opinion is that the mathematical disciplines are not proper sciences. To have science is to acquire knowledge of a thing through the cause on account of which the thing is. However, the most perfect kind of demonstration must depend upon those things which are proper to that which is demonstrated; indeed, those things which are accidental and in common are excluded from perfect demonstrations.”

Euclid’s geometry is not a “science” in this sense, according to this point of view. For example, Pereyra, says, consider the theorem that the angle sum of any triangle is two right angles (Euclid’s Proposition 32). “The geometer proves [this theorem] on account of the fact that the external angle which results from extending the side of that triangle is equal to two angles of the same triangle which are opposed to it. Who does not see that this middle is not the cause of the property which is demonstrated? [The external angle] is related in an altogether accidental way to [the angle sum of the triangle]. Indeed, whether the side is produced and the external angle is formed or not, or rather even if we imagine that the production of the one side and the bringing about of the external angle is impossible, nonetheless that property will belong to the triangle; but, what else is the definition of an accident than what may belong or not belong to the thing without its corruption?”

So in other words, Euclid’s proof of the angle sum theorem does not reveal the actual reason why the theorem is true. Instead it proves the result via a non-essential thing, the external angle sticking out from the triangle. This external part was obviously added by the geometer quite gratuitously; it’s not essential to the very nature of the triangle. So it’s a kind of artificial trick to add this extra angle and base the proof on it. Truly explanatory and causal demonstrations should not be based on artificial tricks but on what is truly essential to the situation.

Schopenhauer later ranted against Euclid along similar lines. That’s in the 19th century. These ideas were more important and influential in the 16th century, when Aristotelianism was a dominant philosophy. But it’s fun to quote Schopenhauer anyway, because he expresses the same ideas in a charming way. Here’s what he says:

“Perception is the primary source of all evidence, and the shortest way to this is always the surest, as every interposition of concepts means exposure to many deceptions. If we turn with this conviction to mathematics, as it was established as a science by Euclid, and has remained as a whole to our own day, we cannot help regarding the method it adopts, as strange and indeed perverted. We ask that every logical proof shall be traced back to an origin in perception; but mathematics, on the contrary, is at great pains deliberately to throw away the evidence of perception which is peculiar to it, and always at hand, that it may substitute for it a logical demonstration. This must seem to us like the action of a man who cuts off his legs in order to go on crutches.”

“Instead of giving a thorough insight into the nature of the triangle, [Euclid] sets up certain disconnected arbitrarily chosen propositions concerning the triangle, and gives a logical ground of knowledge of them, through a laborious logical demonstration, based upon the principle of contradiction. We are very much in the position of a man to whom the different effects of an ingenious machine are shown, but from whom its inner connection and construction are withheld. We are compelled by the principle of contradiction to admit that what Euclid demonstrates is true, but we do not comprehend why it is so. We have therefore almost the same uncomfortable feeling that we experience after a juggling trick, and, in fact, most of Euclid’s demonstrations are remarkably like such feats. The truth almost always enters by the back door, for it manifests itself per accidens through some contingent circumstance. Often a reductio ad absurdum shuts all the doors one after another, until only one is left through which we are therefore compelled to enter. Often, as in the proposition of Pythagoras, lines are drawn, we don’t know why, and it afterwards appears that they were traps which close unexpectedly and take prisoner the assent of the astonished learner. The proposition of Pythagoras teaches us a qualitas occulta of the right-angled triangle. In our eyes this method of Euclid in mathematics can appear only as a very brilliant piece of perversity.”

So Schopenhauer agrees with the 16th-century Aristotelians that Euclid’s proofs are not explanatory. Instead they proceed by some kind of trick. Euclid is constantly setting logical mousetraps that force the reader to accept the conclusion even though nothing has truly been explained.

It’s interesting though that Schopenhauer uses the example of a machine that is shown to someone who doesn’t know how it was made and therefore is baffled by it and cannot understand how it works. The people of the constructivist tradition we discussed earlier of course used the same image to prove the opposite point: namely that in geometry we are the makers of the machines we use and precisely for that reason that we have genuine knowledge and understanding of it. The people who looked at it that way were basing themselves on mathematical sources. Schopenhauer and the 16th-century Aristotelian who hated mathematics so much were also the ones who knew the least about it. They had not studied the technical Greek writers like Archimedes, Apollonius, and Pappus. Some of these technical sources had not even been translated into Latin yet at the time the Aristotelians were writing in the 16th century. And by the time of Schopenhauer they had been forgotten again among philosophers.

But these Aristotelian guys in the 16th-century also had further interesting arguments to support their point. For example, consider Euclid’s area theorems for parallelograms and triangles in Propositions 35 and 37. The theorems say that same base and same height implies the same area. The first theorem says this for parallelograms and the other one for triangles. The proof of the second theorem is based on the first one: a triangle is just half a parallelogram, so since we already have the result for parallelograms it follows almost immediately that it is also true for triangles.

But we could just as well have done it the other way around: we could have proved the theorems first for triangles, and the infer the result for parallelograms by saying that parallelograms are basically just double triangles.

Euclid chose to start with the parallelogram and then do the triangle, but this was essentially an arbitrary choice. It doesn’t reflect any causal relation. The two theorems are equivalent. It’s not that one of them is more fundamental and therefore explains or causes the other. Neither of the two theorems is more of a cause than the other. So Euclid’s procedure doesn’t fit Aristotle’s decree that demonstrations should proceed from causes.

These guys, like I said, didn’t keep in mind the whole construction business. They were not aware of that because they had not read much mathematics. Later, Leibniz, who knew both the mathematical and the philosophical traditions very well, argued that the construction perspective solves the problem that the Aristotelians raised. Here’s what Leibniz says:

“[Geometry] does demonstrate from causes. For it demonstrates figures from motion; from the motion of a point a line arises, from the motion of a line a surface, from the motion of a surface a body. Thus the constructions of figures are motions, and the properties of figures, being demonstrated from their constructions, therefore come from motion, and hence, from a cause.”

So basing geometry on constructions imposes a natural order—a causal hierarchy, as it were—on its theorems whence Aristotle’s ideal of demonstrative understanding can be maintained. According to Leibniz anyway.

Let’s have a look at Descartes as well. He also had interesting ideas about what made mathematics such a special type of knowledge, and how its success could be emulated in other fields.

In his Discourse on Method of 1637, Descartes explained his philosophical program and how he arrived at it. In an autobiographical introduction he explains:

“I was most keen on mathematics, because of its certainty and the incontrovertibility of its proofs. Considering that of all those who had up to now sought truth in the sphere of human knowledge, only mathematicians have been able to discover any proofs, that is, any certain and incontrovertible arguments, I did not doubt that I should begin as they had done.”

Those are the words of Descartes, famous for doubting everything; his very method has been called the method of doubt. Yet as he himself says: “I did not doubt” that I should follow the mathematicians.

You just had to extend the mathematical method to other areas as well, to philosophy in general. As Descartes says:

“Believing as I did that its only application was to the mechanical arts, I was astonished that nothing more exalted had been built on such sure and solid foundations.”

Just imagine the amazing things that could be achieved if other fields were as successful as mathematics. This was a common sentiment. Here’s how Hobbes put the same point:

“The geometricians have very admirably performed their part. For whatsoever assistance doth accrue to the life of man, whether from the observation of the heavens, or from the description of the earth, from the notation of times, or from the remotest experiments of navigation; finally, whatsoever things they are in which this present age doth differ from the rude simpleness of antiquity, we must acknowledge to be a debt which we owe merely to geometry. If the moral philosophers had as happily discharged their duty, I know not what could have been added by humane Industry to the completion of that happiness, which is consistent with humane life.”

So the goal of philosophy is to be as good as mathematics. So let’s see what Descartes considers to be the foundations of mathematics. He formulates a method for how to philosophise in general, and he intends for this to be a generalization of the mathematical method.

So you might say his methodological program is part descriptive and part prescriptive. It is descriptive because it describes how geometry works; it’s an analysis meant to capture what made Euclid so great. And at the same time it is prescriptive in that it gives orders as to how one should philosophise. Namely, whatever Euclid did in geometry, that philosophers should do in every field, such as physics, ethics, theology, and so on.

Here’s what Descartes says about the axioms or starting points of a theory. We discussed before whether the axioms should necessarily be obvious. Descartes comes down very firmly on that issue.

“The first [principle of my method] was never to accept anything as true that I did not incontrovertibly know to be so; that is to say, carefully to avoid both prejudice and premature conclusions; and to include nothing in my judgements other than that which presented itself to my mind so clearly and distinctly, that I would have no occasion to doubt it.”

So we should start only from the most obvious things, in other words. Things that are so clear that they cannot be doubted. Things known by immediate intuition, in other words. That’s supposed to correspond to the axioms of Euclid.

So Descartes has a lot of faith in innate intuition. As Descartes says, there are “basic roots of truth implanted in the human mind by nature, which we extinguish in ourselves daily by reading and hearing many varied errors.” So this inner “natural light” is more reliable than book learning.

So we should, Descartes says, “conduct thoughts in a given order, beginning with the simplest and most easily understood objects, and gradually ascending, as it were step by step, to the knowledge of the most complex.” And for the sake of this stepwise process, it is necessary to “divide all the difficulties under examination into as many parts as possible.”

You can see how philosophy is going to look a lot like Euclid if people follow these rules that Descartes lays down.

It is interesting that Descartes also specifically says that one should “posit an order even on those [things] which do not have a natural order or precedence.” This is a kind of reply to the Aristotelian point we mentioned above.

The Aristotelians were arguing that when two theorems are equivalent—such as the areas theorems for triangles and parallelograms—then it is artificial and unscientific to impose a particular order that makes one logically prior to the other, as Euclid does. Because then you haven’t given a causal explanation, as Aristotle says one should.

Descartes turns the tables on this. Instead of criticising Euclid when his method seems to go against philosophical sense, he makes Euclid the boss of philosophy. Whatever Euclid does, that’s good method. So if Euclid imposes an artificial logical order on equivalent theorems, then that’s what one should do in philosophy, Descartes concludes.

It goes against Aristotle—so what? Those people I quoted from the 16th century, a hundred years before Descartes, they thought Aristotle had more authority than Euclid, so they used Aristotle to criticise Euclid. Now, a hundred years later with Descartes, it is the other way around. Descartes would rather use Euclid to criticise Aristotle.

A lot had happened in those hundred years. A lot of new science: Copernicus, Galileo, Kepler, etc. Science had made terrific progress by using Euclid and ignoring Aristotle.

By the time of Descartes, the Aristotelians were dinosaurs. Descartes didn’t pull any punches when making this point: he condemned the Aristotelians as “less knowledgeable than if they had abstained from study.”

This new hierarchy, where mathematics has greater authority than philosophy was soon widely accepted. John Locke, the famous philosopher, put it like this half a century later: “in an age that produces such masters as the great Huygenius and the incomparable Mr. Newton, it is ambition enough to be employed as an under-labourer in clearing the ground a little, and removing some of the rubbish that lies in the way to knowledge.” So philosophy is just an under-labourer to mathematical science. The real geniuses, the real creative forces are mathematicians such as Huygens and Newton. Philosophers take on a subordinated role. The task of the philosopher is to explain to others how to follow the lead of the mathematical sciences. This is why Locke calls himself a mere under-labourer.

So that was the general methodological influence of mathematics on Descartes. But Descartes was not content with merely adopting the Euclidean method in philosophy. He also wants to justify this method; to explain why it is so reliable. He does this in his Principles of Philosophy of 1644.

In the very first sentence of this book, Descartes says: “whoever is searching for truth must, once in his life, doubt all things.” As we just saw, in his earlier work he had said that he did not doubt the method of the mathematicians. Now he’s going to fix this gap.

Let’s say you did doubt the mathematical method, the method of Euclid. According to Descartes, as we saw, the foundations of the method was intuition. Euclid starts from axioms such as “if equals are added to equals, the results will be equal.” Intuitively, these basic truths feel completely undoubtable. We are so convinced that they must be true, even though we cannot prove these things.

You might argue: there will always be something we cannot prove. In a deductive system, one thing is deduced from another, but you have to start somewhere. If I tried to prove Euclid’s axioms, I would have to deduce them from something. Whatever those somethings are, they will become the new axioms. So then they have to be assumed. We can never escape this cycle. We always have to assume something.

Unless. Unless we find axioms that are somehow logically self-justifying. This is the idea of the consequentia mirabilis that we discussed before. Axioms can be self-justifying if it is incoherent to try to refute them. If asserting that the axioms is false actually implies accepting the axiom, then the axiom is self-justifying. That way we can find an end to the problem of infinite regress; the problem of always having to prove everything from something else in a never-ending cycle.

This is going to be Descartes solution. He will give an axiom of that type, and then derive the Euclidean axioms from it. Then he will have closed the loop: there are no loose ends, nothing unjustified, anymore.

Here’s the axiom: I think therefore I am. This is the undeniable truth which cannot be denied because denying it would be contradictory.

Here’s how Descartes puts it: “We can indeed easily suppose that there is no God, no heaven, no material bodies; and yet even that we ourselves have no hands, or feet, in short, no body; yet we do not on that account suppose that we, who are thinking such things, are nothing: for it is contradictory for us to believe that that which thinks, at the very time when it is thinking, does not exist. And, accordingly, this knowledge, I think, therefore I am, is the first and most certain to be acquired by and present itself to anyone who is philosophizing in correct order.”

Ok, so that’s the axiom that cannot be denied because to deny it would be contradictory. How are you supposed to prove Euclid’s axioms from there? That seems difficult. How am I supposed from prove geometrical statements from “I think therefore I am”? Well, Descartes has an answer.

“The knowledge of remaining things [including geometry] depend on a knowledge of God,” because the next things the mind feels certain of are basic mathematical facts, but it cannot trust these judgments unless it knows that its creator is not deceitful. “The mind discovers [in itself] certain common notions [such as the axioms of Euclid], and forms various proofs from these; and as long as it is concentrating on these proofs it is entirely convinced that they are true. Thus, for example, the mind has in itself the ideas of numbers and figures, and also has among its common notions, that if equals are added to equals, the results will be equal, and other similar ones; from which it is easily proved that the three angles of a triangle are equal to two right angles, etc.”

But the mind “does not yet know whether it was perhaps created of such a nature that it errs even in those things which appear most evident to it.” Therefore “the mind sees that it rightly doubts such things, and cannot have any certain knowledge until it has come to know the author of its origin.”

So mathematics depends on intuition, and intuition is something implanted into the mind. So God made us have these intuitions. So justifying our innate intuitions depends on the nature of God.

Here is Descartes’s proof that “a supremely perfect being exists”: “That which is more perfect is not produced by a cause which is less perfect. There cannot be in us the idea or image of anything, of which there does not exist somewhere, some Original, which truly contains all its perfections. And because we in no way find in ourselves those supreme perfections of which we have the idea; from that fact alone we rightly conclude that they exist, or certainly once existed, in something different from us; that is, in God.”

“It follows from this that all the things which we clearly perceive are true, and that the doubts previously listed are removed,” since “God is not the cause of errors,” owing to his perfection, because “the will to deceive certainly never proceeds from anything other than malice, or fear, or weakness; and, consequently, cannot occur in God.” “Thus, Mathematical truths must no longer be mistrusted by us, since they are most manifest.”

So, in summary: Euclid’s axioms are true because we innately feel them to be true, and this intuition was implanted into us by God. Our intuition is reliable because God is not a deceiver because he is a perfect being. God must be perfect, because we have the idea of perfection, and we could only get that idea from actual perfection. Since we can conceive of perfection, there must be perfection, there must be a perfect being, a perfect God. That God has hardwired truths such as the Euclidean axioms into our minds. And they must be right because God wouldn’t be perfect anymore if he tricked us by implanting false beliefs in our minds.

That’s Descartes’s argument. I think it’s interesting how we can tell this entire story as driven almost completely by the analysis of Euclid. This whole thing about God and so on it almost like an afterthought, or a minor stepping-stone. The real goal is to justify the geometrical method or explain why Euclid’s axioms should be believed. All this philosophy and theology stuff—I think therefore I am, the existence of God—those are just supporting characters or secondary concerns. Or at least that’s one way of reading Descartes.

In any case, in Descartes as in the previous philosophers we have discussed today, we have seen the very profound influence of ancient geometry. Euclid was still setting the course for philosophy, two thousand years after his death. All the more reason to study him further.

]]>**Transcript**

Diagrams. What are their role in geometry? Some people like to think that the logic of a geometrical proof doesn’t need the diagram. Mathematics is supposed to be pure and absolute. Diagrams seem connected to the visual, the intuitive, that makes it kind of psychological, and perhaps therefore even subjective.

Certain people don’t like that association one bit, so they try to minimise the role of diagrams. Maybe diagrams are just crutches to help those with weaker minds, whereas a perfect logical reader could follow the proof from the text alone. Some people like to think so. It’s a dogma that fits modern tastes.

But, historically, that interpretation is a pretty poor fit. In some ways, classical geometry appears to have embraced visuality rather than tried to replace it with abstract logic.

There are signs of this attitude in the very language of Greek geometry. The word for proving is the same as the word for constructing: grafein, to draw. To prove something is literally to make it graphic. And a theorem, in ancient Greek, is a diagramma, a diagram. Instead of the Pythagorean Theorem the Greeks would say the Pythagorean Diagram.

Indeed there is always one diagram for each theorem in Greek mathematics. That’s a very rigid rule. In modern mathematics we often find it natural to have several pictures for some proofs, and no pictures at all for many other proofs. Just do what comes natural to explain the particular content. But not the Greeks. One theorem, one picture: this rule was extremely firmly ingrained in their conception of geometry.

And not only in geometry, in fact. Euclid follows this rule slavishly even when he writes about number theory. For example, he proves (Elements VII.30) that if a prime number divides a times b, then it divides either a or b. A very important theorem that is still proved in every modern book on number theory. But no modern book would include a picture for this. It’s just not a visual thing at all, so it makes little sense to draw a picture to go with it.

But Euclid does. The numbers that he is talking about he draws as line segments. The bigger the number the longer the line. But this has little to do with his proof. The proof is not visual. It’s just as abstract as the ones in the modern books. So the diagram doesn’t really do anything. And it’s like that theorem after theorem after theorem: Euclid has these useless diagrams that are basically irrelevant to the content. But he insists on the rule “one theorem, one diagram” even where it doesn’t really seem to serve any purpose.

At least it doesn’t serve any purpose in terms of capturing or visualizing the steps of the proof. Maybe it has other purposes. One purpose could be to signal that number theory is subsumed by geometry. The number 5 really just means a line segment of length 5 units, Euclid seems to be saying with these diagrams. So since Euclid has established the foundations of geometry, and number theory so to speak lives within geometry, then it follows that Euclid has established the foundations for number theory as well. Number theory doesn’t need separate foundations since it is subsumed by geometry. Maybe this is what Euclid is trying to emphasize with his pictures of numbers.

Or maybe Euclid needs pictures because he doesn’t have algebra. A modern proof of theorems like these are very dependent on algebraic notation. If p divides ab, then p divides a or b. In the course of the proof you keep referring to relationships between these number all the time. Suppose p divides ab but not a. Etc., etc. It would be hard to get all that across without algebraic symbols.

If you have a picture you don’t need algebra, because you can point. Instead of the letters a, b, p you have line segments of different lengths that you can point to and say: suppose that one dives that one. You don’t need algebraic symbols or letters, because you are pointing to a picture. The mode of presentation is oral; you have your audience in front of you, and you have drawn the diagram in the sand with a stick, and you point to it as you reason your way through the proof.

You might say: But Euclid does have labels, like A, B, C, etc. So he is referring to entities by letter or label designation, not merely by pointing visually. Well, maybe. But one could argue that that’s not really what Euclid’s A, B, Cs mean.

When Euclid calls things alpha, beta, gamma, it is perhaps inaccurate to translate this as A, B, C. Because it would also mean 1, 2, 3, or first, second, third. The Greeks wrote numbers this way, using the letters of the alphabet. Alpha meant 1, beta meant 2, and so on. So perhaps we shouldn’t think of Euclid’s ABC as algebraic designations. Perhaps it simply means “the first point,” “the second point,” and so on.

This makes it seem a lot closer to the pointing hypothesis. Perhaps the standard way for mathematicians to explain their reasoning was to point to a picture and say “this one,” “that one,” and so on. Then to encode this in writing they used alpha, beta, gamma, to mean “the first one I mentioned,” “the second one I mentioned,” and so on.

If this is right, then the letters in the English version of the Elements are a bit deceptive. They seem more algebraic, more modern, than they really are. From that point of view, diagrams in number theory make some sense.

In fact, in early modern geometry, in the 17th century, you sometimes see people labeling points in diagrams 1, 2, 3 instead of A, B, C. Because they thought this was the right way to translate Greek into Latin. Euclid’s alpha is really a 1, and so on. They were more sensitive to Greek culture back then. Nowadays people have forgotten about that stuff.

Here’s another fun linguistic-cultural perspective on diagrams in Greek geometry. The language in which Euclid describes constructions is quite odd. “Let the circle ABC have been described.” The language of Greek mathematics “makes the author and temporality disappear from a proof,” as one historian has put it. Euclid is not saying that he’s drawing the diagram, and he’s not telling the reader to draw the diagram. He’s just sort of commanding the diagram into existence.

You know the book of Genesis in the Bible: “Let there be light,” God said, and there was light. Euclid uses literally the same kind of construction. It’s exactly the same verb form as in the Ancient Greek version of the Bible. Just as God makes heaven and earth by merely pronouncing that they exist, so Euclid makes geometrical objects appear just ordering them to be. It’s not “I draw” or “you draw” but “let it have been done.”

You could read this as supporting a Platonic conception of mathematics. Euclid is distancing himself from actual drawing. The objects of mathematics just are. They are not something you or I have to make.

But here’s a counter argument to this interpretation. Netz argues that actually Euclid’s grammatical construction merely reflects a purely practical circumstance of the Greek tradition. Namely, that Greek mathematicians had to prepare their diagrams in advance due to technical limitations of the visual media available. Here’s what Nets writes:

“Of the media available to the Greeks none had ease of writing and rewriting. [Standard media were papyri and wax tablets, and, for larger audiences, such as Aristotle’s lectures,] the only practical option was wood painted white. None of these [ways of representing figures] is essentially different from a diagram as it appears in a book. The limitations of the media available suggest the preparation of the diagram prior to the communicative act---a consequence of the inability to erase. This, in fact, is the simple explanation for the use of perfect imperatives [such as] ‘let the point A have been taken’. It reflects nothing more than the fact that, by the time one comes to discuss the diagram, it has already been drawn.”

That’s Netz’s interpretation, and if he’s right then Euclid’s grammatical choice reflects only incidental cultural circumstances and says nothing about philosophical commitments.

So “let it have been done” just means “I did it yesterday”. It doesn’t mean that geometry is set apart from concrete action and that doing has no place in mathematics.

It’s fascinating how the same aspect of the text takes on such a different meaning when cultural context is taken into account, compared to a purely philosophical reading. In fact, let me tell you about another striking aspect of Greek manuscripts which is also like that. Namely, the way diagrams are drawn in manuscripts of Greek geometry.

Diagrams in manuscripts of Greek mathematical treatises are very often very poorly drawn. They are oversimplified and crudely schematic. Ellipses, parabolas, and hyperbolas are represented as pieces of circles and so on. Very poor pictorial accuracy.

Also the simplicity and specificity of the diagrams often obscure important mathematical points. For example, the figure for the Pythagorean Theorem is often drawn in manuscripts with the two legs of the triangle being equal, even though the theorem holds for any right-angle triangle. The diagram thereby gives the misleading impression that the theorem is less general than it really is.

So you might think: aha, clearly the Greeks didn’t care about the diagrams. They are poorly executed, poorly thought through. So diagrams couldn’t have been an important part of geometry then.

Well, not so fast. The diagrams are drawn this way in the manuscripts that exist today. But who wrote these manuscripts, and when? In fact, the oldest manuscript of Euclid’s Elements that exists today is closer to us in time that it is to Euclid. It’s from the Middle Ages. A thousand years ago. That might seem ancient enough, but Euclid lived thirteen hundred years before that.

There was no printing press until the 15th century, so for well over a thousand years the book had to be copied by hand. You had to hire a scribe to write the whole thing out.

Manuscripts are fragile. The Greeks wrote on papyrus. It takes a miracle for a roll of papyrus to survive more than two thousand years. Just think of books from the 19th century, maybe some old book from your grandparents. They are already falling apart, and that was only a hundred years ago. Imagine storing that for twenty times as long. It will fall apart on its own, and that’s not even counting the risk of fires, or floods, or insects, or wars, and so on.

So few documents from Greek times survive to this day, and hardly any of those are mathematical. Only the tiniest little scraps of mathematics from antiquity itself are still around. And they are not enough to say anything about how the Greeks dealt with diagrams.

We only have these later copies. Or better put: a copy of a copy of a copy of a copy and so on. Our oldest manuscript may very well be, who knows, maybe twenty or thirty copying steps away from Euclid’s original.

The state of the diagrams in these manuscripts perhaps says more about the copying and the copyists than it does about Greek geometry. The scribes who copied these manuscripts probably often knew little or no mathematics. They probably had some training as scribes; training in Greek, in writing. Perhaps they mostly copied literary texts or whatever.

So they were probably pretty good at copying text, but not at copying diagrams. It’s pretty straightforward to copy text if you know the language. An A is an A. You can’t really misinterpret it.

Diagrams are a lot more subtle. Often you can only understand what aspects of a diagram are essential by studying the text, the logic of the proof that goes with it. But the scribes would not have done this. They were hired copyists, not research students. They didn’t study the content, they just blindly copied it for a paycheck, like a photocopier.

This is enough to explain why the diagrams are so simplistic. It is natural in such a context of copying that the diagrams gradually degenerate and converge to more simplistic versions. This is the predictable outcome of repeated copying by generations of scribes largely ignorant of mathematical content. For a very simple reason: an ignorant copyist can easily misinterpret a subtle diagram in a simplistic way while going the other way around, toward a more subtle and exact diagram, could only be done by someone with a solid understanding of the mathematical content, who would restore the diagram based on what the text suggests.

For example, in the case of the Pythagorean Theorem, a scribe might get a version of the figure where the two legs look approximately similar and then mistakenly assume that exact equality was intended. He then copies it this way, and specificity is introduced. Now others will keep copying this simplified diagram. No one will restore more generality in the diagram, because that would require revising the figures based on mathematical understanding, which was not the task of copying scribes.

There’s a fun paper on this by Christian Carman in a recent volume of Historia Mathematica. Carman tested this hypothesis with his students. He had them go in a circle and copy a mathematical diagram from one another, like the children’s game Chinese whispers or telephone where you whisper something, then they try to pass it on, and so on. By the time the message has made it full circle it has become something else. It’s the same with diagrams.

You can see also how the specificity aspect emerges from this. The original diagram might show two lines meeting at an angle of, say, 75 degrees. Copying is a bit imperfect, so maybe someone copies it more like 82 degrees. Then the next guy thinks: well, this is probably supposed to be 90 degrees, they just drew it a little bit wrong. So they make it 90 exactly. Then from that point on everybody copies it as 90 degrees. Because exactly 90 degrees looks a lot more intentional than 82. This is why the process almost always goes toward more specificity.

So we cannot conclude anything about ancient philosophy of mathematics from the way diagrams are drawn in the manuscripts. This aspect of the manuscript sources is very likely an artefact of transmission that says nothing about ancient geometry.

So we still don’t know what Euclid thought about diagrams. We know what Plato thought. His opinion was reportedly that mathematicians who “descended to the things of sense” were “corrupters and destroyers of the pure excellence of geometry.” That’s how Plutarch describes Plato’s opinion. So basically an anti-diagram agenda.

But there is no evidence that mathematicians shared these sentiments. On the contrary, the combative way in which this view is presented in the sources clearly show that they were far from a consensus opinion. Plato himself openly puts his view in diametrical contrast with that of the geometers. Here’s what he says in the Republic (VII 527):

“No one with even a little experience of geometry will dispute that this science is entirely the opposite of what is said about it in the accounts of its practitioners. They give ridiculous accounts of it, for they speak like practical men, and all their accounts refer to doing things. They talk of squaring, applying, adding, and the like, whereas the entire subject is pursued for the sake of knowledge [and] for the sake of knowing what always is, not what comes into being and passes away.”

Again, Plutarch reports on the same conflict and makes it crystal clear that Plato’s views on geometrical method was diametrically opposed to that of the leading mathematicians of his day. Here’s what Plutarch says: “Plato himself censured Eudoxus and Archytas and Menaechmus for endeavouring to solve the doubling of the cube by instruments and mechanical constructions.”

So not only is there no evidence that any notable Greek mathematician was a Platonist, but the Platonic sources themselves clearly and openly admit that their view is an ideological extreme that was not widely shared, especially not among mathematicians.

So what’s the alternative? If the mathematicians were not Platonists, what were they? Maybe the didn’t care about philosophy at all. Here’s how Netz puts it:

“Undoubtedly, many mathematicians would simply assume that geometry is about spatial, physical objects, the sort of thing a diagram is. The centrality of the diagram meant that the Greek mathematician would not have to speak up for his ontology. The diagram acted, effectively, as a substitute for ontology. One went directly to diagrams, did the dirty work, and, when asked what the ontology behind it was, one mumbled something about the weather and went back to work.”

That’s what Netz thinks, and it seems consistent with Plato’s rants against the geometers that they would have been disinterested in these questions indeed.

But I think Netz is selling the mathematicians short. I do not believe that Greek mathematicians “simply assumed” these things, and could only “mumble something about the weather” if pressed on the issue. I suspect that, on the contrary, Greek mathematicians had a philosophically sophisticated defence of their ontological stance, based on the operationalist ideas that we discussed before.

Let’s see how this plays out in a concrete mathematical example. From a modern point of view, the right way to do geometry is as a formal axiomatic-deductive system. The Greek tradition has often be interpreted as aspiring toward, but falling short of, this ideal. According to this view, Euclid’s Elements was a brave and admirable attempt at a formal treatment of geometry, especially for its time, but that it contains some fundamental flaws stemming from Euclid’s inability to fully avoid implicit reliance on intuitive and visual assumptions.

Operationalism, by contrast, embraces visual reasoning and keeps abstract logic at arm’s length. This arguably fits the Greek geometrical tradition better than modern formalistic conceptions of geometry. Indeed it is well known that Greek geometry sometimes bases inferences on diagrammatic considerations that are not explicitly formalised.

The most famous example is Proposition 1 of the Elements. In this proposition, the existence of a point of intersection of two circles is tacitly assumed but can arguably not be formally justified from Euclid’s definitions and postulates.

The modern mathematician rejects anything not obtained through logical deduction from formal axioms. The operationalist classical geometer rejects anything not obtained through concretely defined operational procedures. We can formulate the difference between the two points of view in terms of what kind of audience the geometer is trying to convince. If we adopt the modernistic point of view, we can picture the audience of a mathematical proof as a veritable logic-parsing machine. The mathematician feeds in statements, in the form of symbolic strings in a suitable formal language, one by one, and the machine tests whether each statement follow from the one before it based on basic logical inference rules or previously established theorems. This point of view fits very uneasily with classical geometry for a range of reasons, including the use of diagram-based reasoning.

The operationalist point of view, on the other hand, envisions the audience of a mathematical proof differently. A Euclidean proof is addressed at a person with a ruler and compass. This person is every bit as critical as the logic machine of the modernists. He is hell-bent on trying to argue against us at every stage. But our strategy for convincing him to nevertheless concede the truth of our theorems is not by appeals to formal logical inferences. Instead we make him draw things. We build our results up from simple operations with ruler and compasses. In this way we put our critic in a difficult position. He is forced to either agree with us, or to deny a very specific, concrete claim about a very specific, concrete figure that he himself has drawn.

For instance, what is the person with the ruler and compass supposed to say regarding the intersection of the circles in Proposition 1 of the Elements? He just drew the two circles himself on a piece of paper. It would be ridiculous for him to claim that there is no justification for the assumption that they intersect. They clearly intersect right there in front of his eyes, and it was he himself who drew it using tools whose validity he had admitted.

Since operationalism gives absolute primacy to the concretely constructed diagram, the sceptic has no other foothold from which to reject the proof. The logic machine of the modernist paradigm would catch the gap in Proposition 1 at once, and shoot down our proof. But operationalist mathematics is not susceptible to that kind of critique. Geometrical proofs are claims about what happens when you carry out concrete constructions. Constructed diagrams is all there is, so the only way to question a geometrical proof is to question what it says about a concretely constructed diagram. The sceptic cannot hide behind sophistical logic and vague generalities, but is forced to either concede the validity of the proof or deny something so obvious that he will look ridiculous.

The conception of a proof as addressing a sceptic fits the Greek context well. It’s just like a Socratic dialogue. You extracting concessions from a determined opponent in incremental steps. Just as Socrates does in the dialogues of Plato. And just as disputants would aim to do in a stage debate of the kind the Greeks loved.

I think one could argue that the diagrammatic inferences Euclid permits are precisely those that such a sceptic, who has drawn the diagram himself, could not reasonably doubt. This fits well with Kenneth Manders’ observation that Euclid permits diagrammatic inferences only of properties of the diagram that are invariant under minor variations or imperfections in the drawing process.

For example, in Proposition 1 of the Elements, the equality of the legs of the triangle can of course not be established merely by visual inspection of the diagram; rather, these equalities have to be derived from postulates and definitions, as do all exact properties of diagrams in Euclid’s geometry. Indeed, a sceptic could very well question whether such properties hold, despite having just constructed the diagram himself. The equality of the legs is not immediate from the diagram in and of itself, but only follows when we remind ourselves that we used the same radius for both circles and so on. You could draw the diagram without keeping such things in mind. You could not, however, draw the diagram without directly experiencing one circle cutting unequivocally right through the other one.

Operationalism relies on diagrammatic reasoning only in this restricted sense. It attributes foundational status to diagrams in certain respects, but of course it does not go so far as to say that the truth of propositions or veracity of solutions to problems can be verified merely by measurements in a diagram. Of course such things have to be established by rigorous demonstration, which is obviously the main preoccupation of Greek mathematical sources.

What Plato says about inferring geometric truths from diagrams remains true also for operationalists. This is a quote from the Republic (VII 529): “If someone experienced in geometry were to come upon [diagrams] very carefully drawn and worked out, he’d consider them to be very finely executed, but he’d think it ridiculous to examine them seriously in order to find the truth in them about the equal, the double, or any other ratio.”

Indeed, exact properties such as ratios cannot be inferred from diagrams, no matter how carefully drawn, just as Plato says. But the operationalist enterprise does not rely on such epistemic overreach. Instead, its use of diagrammatic reasoning is much more restrictive and limited to essentially qualitative or topological or inexact inferences from diagrams.

So operationalism makes sense of Euclidean practice with regard to diagrammatic reasoning. It eliminates the need to attribute to Euclid a big logical blunder in his very first proof, or the need to denigrate the more visual aspects of Euclid’s reasoning as lowly intuition and an imperfect form of mathematics. Instead it articulates a philosophy of mathematics that incorporates this aspect of Euclidean mathematical practice into a coherent and purposeful whole.

So that’s one way to argue that the so-called logical gap in Euclid’s Proposition 1 is not a gap at all. It’s only a gap if you want geometry to be completely reduced to formal logic. From the point of view of operationalism it is not a gap.

There are other ways to try to save Euclid’s proof. More conservative ways. If you know some modern mathematics it’s a fun game to play to try to read all kinds of things into Euclid’s definitions.

For instance, Euclid’s definition of a circle specifies that it is contained by a single curve, and that it has an inside, and by implication also an outside. In Proposition 1, when you draw the second circle, it is evident that the second circle will have some points inside and some points outside the first circle. So you could argue that, topologically, there’s no way a continuous curve could go from the inside of a closed curve to the outside of it without crossing it. Therefore the existence of the intersection point can be regarded as implied by Euclid’s definition rather than a logical gap.

If you are a modern mathematician you might reply: well, that depends on the underlying field! The argument works for the plane of real numbers, but not if the underlying field is that of rational numbers only. Then indeed the intersection does not exist. So Euclid would have to specify the underlying field before the argument based on inside and outside could work.

Interestingly, one could argue that Euclid sort of does this actually. Because he says in Definition 3 that “the extremities of lines are points.” Now if you wear your modern glasses, you can read this as saying that lines contain their limit points. So the Euclidean plane is a complete metric space. So that rules out the argument based on the rational numbers.

Well, if you know modern mathematics it’s fun to think along these lines, but for my part I vote for the operationalisation reading of Euclid as the more historically plausible way of saving Euclid’s proof of Proposition 1.

Here’s an objection though to the operationalist interpretation. The so-called generality problem. Geometrical theorems are about entire classes of objects---infinite sets of them. For instance, the angle sum of all triangles. Yet all geometrical proofs in the classical tradition are always illustrated with, and reason based on, one particular diagram. The standard way to defend geometrical reasoning against this challenge is to say that geometrical proofs concern only properties that hold generally and do not rely on incidental properties that hold only for the particular diagram. This view was expressed already by Proclus.

Operationalism suggests a very different way of dealing with the generality problem: it denies the premiss that there is such a thing as “all triangles” in the first place. Before you have put your pen on the paper, there is no geometry. There are no lines, no circles, no triangles. We do not make the metaphysical assumption, as the modernists do, that there is some preexisting universe of these things “out there” about which geometry looks for universal truths.

From this point of view, the “problem” of generality ceases to exist. The theorem is not: there is an infinitude of triangles and all of those have angle sum 180 degrees. Instead it is: any triangle has angle sum 180 degrees. Which really means: if you put your ruler down and draw a line segment, then another one, then another one, then the angles of that one triangle has angle sum 180 degrees. The theorem has no other meaning than that. And the proof is not a logical schema talking about an infinite class of objects. Rather, it is a set of instructions for the sceptic to carry out that will convince him, regardless of which triangle he started with, that the theorem is true for that triangle. It is precisely the strength of the insistence on constructions to reduce everything from the abstract to the concrete in this way. We only talk about what we can see and draw and put on the table right in front of us. To do otherwise would be to engage in empty metaphysics, according to operationalism.

Greek geometry is remarkably consistent with such a reading. Indeed, as Netz has observed, Greek mathematical texts never explicitly claim generality beyond the concrete proof based on a particular diagram.

From a modern point of view, any reliance on diagrams in mathematics is inherently problematic, since mathematics is in essence independent of diagrams. On this view, diagrams are merely a secondary representation of mathematics, and furthermore one contaminated by intuition and other limitations. How, then, can diagrammatically based reasoning be a legitimate way of doing mathematics? That is, how could we ever be sure that what is true of diagrams is true of the “actual” content of mathematics? Operationalism does not answer the question but rejects it. There is nothing more “actual” than the diagram.

So the generality problem is dissolved since operationalism rejects the Platonist ontology of mathematics on which it is based. Nothing exists except what the geometer has constructed.

This view re-emerged in modern mathematics for reasons independent of classical geometry. Here’s how famous Dutch intuitionist Brouwer puts it it his dissertation:

“Wheresoever in logic the word all or every is used, this word, in order to make sense, tacitly involves the restriction: insofar as belonging to a mathematical structure which is supposed to be constructed beforehand.”

There is no “all triangles.” There is only “all the triangles you have made.”

To be sure, many who are concerned about the generality problem will feel that operationalism “solves” the problem only by introducing further problems of equal or greater magnitude. For one thing, operationalism implies that “the very nature of meaning itself makes it impossible to get away from the human reference point,” as Bridgman puts it, since nothing exists or has meaning in geometry except through human agency. But operationalism denies that this is a problem, as Platonists would have it.

Regarding the generality problem more specifically, a modern mind may feel that the operationalist solution merely shifts the problem one step over. Even the operationalist is committed to a form of generality, in the sense that the proof of, say, the angle sum theorem must always work for any given triangle. Isn’t the operationalist mathematician still obligated to somehow justify that the proof has this form of generality, which is essentially the original generality problem in slightly different guise?

It is of course true that the proof is intended to be general in this sense, but officially the operationalist mathematician does not need to be committed to having proved that it is. The operationalist mathematician can simply say: I assert that such-and-such a construction will always have such-and-such an outcome; if you want to prove me wrong, feel free to try to come up with a counterexample.

Of course, psychologically the mathematician presenting a proof must be convinced that it will always work, for if a counterexample would be forthcoming he would be exposed as a fool. But this can be left to the discretion of the mathematician’s intuition. Internally, operationalist mathematicians are of course concerned with this kind of generality. But externally, as a reply to sceptical and philosophical challenges to the epistemological status of mathematics, there is no need for them to saddle themselves with the burden of claiming that their proofs themselves have inherent characteristics that strictly ensure such generality. Instead they can restrict themselves to presenting the proof as a challenge to any sceptic: apply these construction and inference steps to any one figure that fulfils the conditions stated, and you will find that you cannot credibly doubt the validity of any step, and hence you will become convinced that the proposition holds for that figure. It is possible, for the operationalist, to maintain that this is what a proof is.

One may well feel that this restrictive view of what a proof is sells mathematics short and fails to account for the nature and status of mathematical knowledge. However that may be, the fact remains that operationalism makes it possible to take such a stance. The restrictive view of the nature of proofs fits naturally with the operationalist conception of mathematical content and meaning, while it is incompatible with a Platonist conception of the nature of a mathematical theorem.

The restrictive view is a scorched-earth defensive position that can be useful when under philosophical attack. Saying that this is the only sense of mathematics one is willing to defend against sceptical attack does not preclude one from holding more expansive, Platonist beliefs in private. But it is a powerful way of cutting off lines of philosophical attack without changing the practice of mathematics substantially.

So, in conclusion, I have argued that Greek mathematicians were prepared to base geometry on actual diagrams. Despite their physicality, despite their links to human action and perception. Greek mathematics went against modern tastes in this respect.

One could argue against this by pointing to the crudeness of diagrams in surviving manuscripts, or the strangely passive language that Euclid and others used to describe constructions of diagrams. But we have seen that those things can better be explained as the result of cultural context rather than philosophy of mathematics.

The modern view that geometry should be studied through abstract reasoning not dependent on the visual and the physical also has ancient support in Plato’s philosophy. But Plato was not a mathematician. In the words of Francis Bacon, when “human learning suffered shipwreck [at the end of classical antiquity], the systems of Aristotle and Plato, like planks of lighter and less solid material, floated on the waves of time and were preserved,” while more mathematically advanced works were lost forever.

To understand ancient mathematics we must look beyond the surface. We must look beyond loudmouths like Plato. We must seek instead the assumptions conveyed implicitly in the way the mathematicians wrote their proofs. Based on this kind of evidence, a diagram-based mathematical practice can be plausibly reconstructed.

]]>**Transcript**

Ancient Greek geometers were obsessed with constructions. Why?

Euclid’s Elements spends almost as much time showing how to draw geometrical figures as it does proving theorems about them. In fact, it seems Euclid thought drawing was a prerequisite for proving. For instance, the first theorem involving squares is the Pythagorean Theorem. In the proposition right before it, Euclid explains in detail how to construct a square by ruler and compass. The same goes for every other geometrical entity ever used in the Elements: first you construct it, and only then can you say anything about it. Without constructions there can be no geometry, Euclid seems to be saying.

And not only Euclid. All the best Greek geometers had their own signature constructions. Three famous construction problems dominated higher geometry for centuries: doubling the cube, trisecting the angle, squaring the circle. The long list of mathematicians who contributed their own distinctive solutions to these problems is a who’s who of everybody who was anybody in ancient geometry.

What fundamental motivations—what philosophy—drove ancient Greek geometers to this fixation with constructions? Why did Greek mathematicians think it was a good idea to spend hundreds of years trying to make an angle the third of another, or a cube twice the volume of another, in dozens of different ways? Why did they so stubbornly bang their heads against the same wall for century upon century? What sin could be so grave that they imposed on themselves such a Sisyphean task?

Why indeed make things at all? And why do so only sometimes? Why meticulously articulate recipes for transferring line segments by ruler and compass, only to then suddenly move entire triangles like it’s nobody’s business in the very next proposition, as Euclid seemingly does? (When he uses superposition to prove triangle congruence in Proposition 4.)

Euclid knew what he was doing, in my opinion. Constructions were a deliberate strategy to guard against fundamental threats to the reliability and rigour of geometry. If our house is built on rotten pillars it’s only a matter of time before it comes crashing down.

Some ancients critics of geometry indeed identified some ominous cracks in its foundations. Remember, the quarrelsome Greeks, they questioned everything with zeal. Some people tried to take down geometry. They were determined to show that it was pseudo-knowledge that was by no means as certain and exact as the mathematicians claimed.

Geometers had to deal with such external attacks. More so than in other centuries and cultures, mathematicians in Ancient Greece were under constant critical-philosophical attack. They had to formulate a defense. And they did, in my opinion. This is where their obsession with constructions comes from.

So what were these philosophical attacks, to which constructions were the answer? There are a number of them. Here’s one:

False diagram fallacies. If you draw diagrams that are slightly off, you can easily fool yourself when doing geometry. There’s a famous example for instance where one proves that any triangle is isosceles. The conclusion is obviously absurd. But it is “proved” in a way that looks just like any other proof in Euclid.

The false proof is made possible by a subtle error in the diagram. The proof involves bisecting one of the angles of the triangle, and then raising the perpendicular bisector of the opposite side. These two lines meet somewhere. That’s drawn in a plausible-looking way in the diagram. The proof then proceeds based on the diagram, just as Euclid does in his proofs.

But the way the diagram was drawn was erroneous. The two lines were drawn as meeting inside the triangle when in fact their true intersection would be outside the triangle. This is a subtle issue that is easy to miss. So when we reasoned based on the diagram we made some hidden assumptions that we were hardly even aware of.

The rest of the proof was typical Euclid-style stuff. So this example shows that a small and subtle mistake in the way we drew a diagram can destroy the certainty of geometrical reasoning. All the other steps of the proof were very carefully justified, just as Euclid always justifies each of his steps. But that was all for nothing since the subtle error in the diagram poisoned the well and destroyed the whole thing.

The Greeks were evidently well aware of this type of problem. Plato mentions it explicitly: “geometrical diagrams have often a slight and invisible flaw in the first part of the process, and are consistently mistaken in the long deductions which follow.” (Cratylus, 436d) Plato is exactly right. “A slight and invisible flaw” at the outset is enough to ruin “the long deductions which follow.”

We even know for a fact that Euclid himself wrote a (now lost) treatise on fallacies in geometry which is likely to have dealt with these kinds of issues. So the Greeks were clearly well aware of this threat to geometrical certainty. What did they conclude from this?

Today, the issue of diagram fallacies is taken to show how dangerous it is to rely on visual and intuitive assumptions. The solution is to purge geometry of any kind of reasoning based on diagrams. In the late 19th century this view was expressed forcefully by leading geometers, and it has remained the mainstream view ever since. “A theorem is only proved when the proof is completely independent of the diagram,” as Hilbert said for example. Instead of relying on pictures, geometry must be made to proceed through purely logical deduction.

But this is not the only possible diagnosis and treatment of the problem with the erroneous proof. Another point of view is to say: the problem is not that the proof relied too much on diagrammatic reasoning, but that it did so too little. The problem is not that the proof is insufficiently divorced from visual considerations, but that it is too divorced from them. The example doesn’t show that diagrams are dangerous even if they are just schematic accompaniments to otherwise logically solid proofs, but rather that diagrams are dangerous when they are merely treated as such.

The solution is not to place less emphasis on diagrams, but more. That is, to demand diagrams to be not merely schematically sketched but in fact precisely constructed according to the most exacting standards and rigorous proofs that these constructions accomplish the configurations in question. This would indeed prevent errors of this type from occurring. No one adhering to this mode of doing geometry would ever find themselves reasoning about false diagrams like the one in the above example.

This diagnosis of the source of error in the false proof above leads immediately to the conclusion that precise constructions of angle bisectors, bisectors of segments, and perpendicular lines are foundationally very important, and that no proof must ever be formulated without constructive recipes for all entities occurring in it having been established beforehand.

And this is exactly what we find in Euclid’s Elements. Without fail, Euclid always meticulously shows how to construct all entities involved in all of his propositions. And all the constructions needed to ensure that we end up with the correct figure rather than the deceptive one in the above example are carefully spelled out as core propositions right at the heart of the Elements: how to bisect an angle (Proposition 9), how to bisect a line segment (Proposition 10), how to raise a perpendicular from a point on a line (Proposition 11), how to drop a perpendicular from a point to a line (Proposition 12).

In other words, right off the bat of the Elements, Euclid carefully explicates precisely the tools needed to solve the false diagram problem mentioned by Plato. Coincidence? I don’t think so. Euclid knows the problem. Euclid knows how to solve it. That’s why he’s obsessed with constructions.

Or rather, it’s one of the reasons. There are other, equally compelling grounds to base one’s geometry on constructions.

Constructions are related to existence issues. It is impossible to conduct a serious axiomatic study of geometry without paying attention to existence issues. For example, do squares exist? Existence is separate from definition. Euclid defines what a square means in his definitions. But that doesn’t mean there are any. You could also define what “unicorn” means. That doesn’t mean unicorns exist.

You might think it’s obvious: of course there are squares, any child realizes that, you can see it with your own eyes. But it’s more subtle than you might think. In fact the existence of squares implies the parallel postulate. Wallis showed this in the 17th century. You could replace Euclid’s parallel postulate with the assumption that you could make a square on a given line segment. Then you could prove Euclid’s parallel postulate and all his other theorems based on that assumption. So it’s no small matter to assume that there are squares.

Therefore, any investigation that aims to elucidate the fundamental assumptions of geometry cannot treat any object whose existence has not first been either proved or explicitly postulated. To do otherwise would be to render the entire enterprise of axiomatic geometry useless and moot, since it would open a back door through which any number of hidden assumptions can creep in. The point of an existence proof for squares, then, would not so much be to establish that there is such a thing as squares, but to ensure that any foundational assumptions involved in supposing the existence of squares have been systematically accounted for.

Another example of this type occurs in Legendre’s attempt to prove by contradiction, using only the first four postulates of Euclid, that the angle sum of a triangle cannot be less than 180 degrees. His proof implicitly assumes that given two intersecting lines, and a point not on those lines, it is possible to draw a line through that point that intersects the two given lines. This assumption does not hold in hyperbolic geometry. Therefore Legendre’s attempted proof is worthless, since the contradiction did not come from the assumption he intended to refute, but from an innocent-seeming existence assumption introduced along the way in his argument.

This shows once again the danger of letting even the most harmless-looking existence or construction assumptions proliferate without explicit control. Inconsistencies can arise from even the most inconspicuous of assumptions. The moral of the story is that the mathematician must stick to a minimalistic set of stringently controlled construction principles, whose consistency should be as unquestionable as possible.

Issues of this nature were recognised in antiquity. Quite possibly, even the specific issue of Legendre’s assumption may have been investigated in works that are no longer extant, such as the lost treatise On Parallel Lines by Archimedes. At any rate, closely related issues emerge explicitly in the treatments of parallels by Simplicius and Al Jawhari.

On a more conceptual level, Aristotle pinpoints the same type of fallacy in the work of some “who suppose that they are constructing parallel straight lines: for they fail to see that they are assuming facts which it is impossible to demonstrate unless parallels exist. So it turns out that those who reason thus merely say that a particular thing is, if it is.” (Prior Analytics, 65a) Indeed. Aristotle is right. Circular assumptions are easy to make, especially with respect to existence issues and subtle foundational questions in the theory of parallels.

Aristotle draws the obvious conclusion that existence issues must be controlled by either explicit postulates or existence proofs. “What is denoted by the first [terms] is assumed; but, as regards their existence, this must be assumed for the principles but proved for the rest,” says Aristotle (Posterior Analytics, 76a). For example, “what a triangle is, the geometer assumes, but that it exists he proves.” That’s Aristotle again (Posterior Analytics, 92b). He’s quite right.

Constructions are a way to ensure existence. Euclid’s first proposition proves that equilateral triangles exist. His 46th proposition proves that squares exist. And so on.

But there’s much more to constructions than merely establishing existence. Construction also establish consistency. That is, it shows that objects are not self-contradictory.

For example, suppose I add to Euclid’s Elements the definition: “A superright triangle is a triangle each of whose angles is a right angle.” Then its angle sum is three right angles by definition, but also two right angles according to a theorem of Euclid’s. So two right angles equal three right angles—an obvious contradiction.

The definition of a superright triangle is disturbingly similar to that of an equilateral or isosceles triangle, and applying Euclid’s theorem to it sounds just like the kind of thing we do in geometry all the time. So it casts doubt on the entire enterprise of geometry. How do we know that the propositions of the Elements are not one or two steps away from leading to contradictions? The geometers must reply with some definitive criterion that explains why none of their theorems are susceptible to this kind of error.

In a way it is clear what the problem is. There are no superright triangles. Hence one can consider the problem solved by ensuring the existence of the objects one speaks of. One way of accomplishing this would be to say: Only constructive definitions, that imply a recipe for making the object defined, are permitted in mathematics. This is clearly not the path taken by Euclid, however. For instance, Euclid defines a square at the outset but only shows how to produce one much later, in Proposition 46, based on substantial previous results.

Another strategy would be to demand that we cannot make a propositional statement about a particular class of objects unless we have first shown beforehand that the class in question is nonempty. Thus the types of inferences made in the false argument are only warranted if supported by suitable existence proofs, and that is why theorems about triangles cannot be applied to superright triangles, but can be applied to equilateral and isosceles triangles, which Euclid indeed proves exist by means of constructions.

But existence is not the only aspect that should be emphasised here. Another important lesson from the superright triangle example is the danger of defining objects through multiple conditions. A superright triangle is defined as: having three sides; having one right angle; another right angle; ant yet another right angle. The first two conditions were fine. It was taking all of them together that was impossible. The more conditions you add, the greater the risk of ending up with an inconsistency.

Another example of this is to say: Let ABC be a triangle such that: one angle is a right angle; the sides next to the right angle have lengths 4 and 7; the third side has length 9. Actually I have taken this example from a 16th-century geometry textbook. But the book messed up. Some of these conditions would have been fine on their own, but all of them taken together are inconsistent. You cannot make a right triangle with those side lengths. Those numbers contradict the Pythagorean Theorem.

Hence defining or introducing an object through a list of specifications of its properties is unacceptable. Doing so would leave the door wide open for possible inconsistency to enter mathematics, and hence ruin the claim to certainty of mathematical reasoning.

A rigorous mathematical theory needs a systematic guarantee that such errors cannot be committed. Constructions are a way to provide such a guarantee. Instead of introducing objects by a list of properties, construction builds it up step by step. Thus properties can no longer be ascribed to an object merely by decree. Rather they must be introduced by a rigorously controlled stepwise process. Each step in this process involves the application of a construction postulate or a demonstrated construction proposition or theorem, which means that assumptions and conditions of validity are carefully monitored and reduced to a few axiomatic principles.

One could argue that the challenge posed by the superright triangle fallacy is not convincingly solved by the insistence on existence proofs. This solution diagnoses the problem as effectively just another variation on the existence issue discussed before, which Aristotle mentioned and so on. But one can readily see it as pointing to a deeper problem. It arguably casts doubt on the credibility of verbal logic altogether. While it is clear that being more careful about existence issues would eliminate the particular problem of the superright triangle, it is not clear whether this is the only problem with relying on verbal logic.

We know for a fact that logical paradoxes and fallacies figured prominently in Greek thought in the classical era. Some of these are clearly relevant to mathematics, such as Zeno’s paradoxes of motion. But there are others.

The liar paradox arguably shows that natural-language propositional logic is incoherent. It shows that verbal logic allows propositional statements to be formulated that are inherently contradictory. “This statement is false” or “I am lying” are examples of such statements. If the statement is true, it follows that it is false. And if the statement is false, then it follow that it is true. So there is no way of assigning a truth value to such a statement without ending up with a contradiction. This kind of thing clearly poses an issue for a logic-based conception of mathematics, not least in connection with proofs by contradiction.

Another example of a paradox discussed in ancient times was that of the horn: What you have not lost, you have; but you have not lost horns; therefore, you have horns. Here again the blind, mechanical application of logical inferences in a quasi-algebraic manner leads to an absurd conclusion. As with the superright triangle fallacy, it is possible to attribute the problem to some specific cause: in this case not so much an existence issue as a certain misleading ambiguity in the first premiss. Furthermore, the fallacy may be regarded as “obvious.” But trying to defuse the paradox in these ways does not solve the core issue exposed by the paradox, namely that “blind” logic, in and of itself, seems to lead to erroneous conclusions.

This multitude of logical paradoxes arguably validates the suspicion that when we supplemented verbal logic with existence proofs we had perhaps not gotten to the bottom of all its problems yet.

It would not have been out of character for the Greeks to have taken radical steps to protect themselves from logical fallacies and paradoxes. The situation may be somewhat comparable to the discovery that the square root of 2 is irrational. This would have been in the very early days of Greek geometry and we don’t know much about it for certain. But a development more or less along the following lines has often been imagined.

In the beginning, the Greeks seem to have blissfully assumed that arithmetic and geometry would always be in natural harmony. The square root of 2 discovery ruined this by showing that natural geometric entities such as the diagonal of a unit square could not be represented by “numbers” (that is to say, rational numbers). The Greeks had burned their fingers and would not make that mistake twice: their response was an extreme foundational purge that eradicated any foundational status of arithmetic and then some. From that moment on, everything is at bottom geometry. Even where a modern mind wishes to see algebra, Euclid and the other Greek mathematicians insist on geometrical formulations with a pedantry bordering on paranoia.

The historical evidence, or absence thereof, of such a “square root of 2 crisis” is a much-debated issue among historians. But the basic point—that Greek mathematicians may very well have gone to great lengths to protect themselves from foundational objections—is plausible. It was a time when the foundations of any subject was constantly under attack from rival philosophers, and people were ready to go to the ends of the earth to rebut such charges.

Extreme action in response to paradoxes that call the bedrock of mathematics into question is a quite plausible scenario. And arguably one with a strong historical precedence in the form of the square root of 2 case. It is squarely within the realm of historical possibility that such a context may have led to the radical proposal of denying any reliance on abstract logic in mathematics and instead founding all of geometry on concretely constructed figures.

So verbal logic is dangerous. It invites paradoxes, like the liar paradox and the paradox of the horns. It has no guard against reasoning about inconsistent objects such as superright triangles. Making constructions, rather than logic, the foundation of geometry solves these problems.

It is suggestive in this connection that Euclid’s proofs are all “purely quantifier free.” That is to say, they never make assertions of the form “there exists” or “for all.” From the point of view of modern mathematics, which is a logic-oriented mathematics, those phrases are fundamental and are used all the time. These phrases are so commonly used that mathematicians do not even have the patience to spell out these two-syllable expressions every time they use them. So they have made up special symbols to abbreviate them. A backwards E and an upside-down A.

I mentioned Hilbert, a leading pioneer of the modernist movement, around 1900. I mentioned that Hilbert wanted to translate all visual information or any inferences based on diagrams into purely logical form. That leads precisely to formulations with those favorite phrases of the mathematician: “these exists” so-and-so; “for all” objects of such-and-such a class, this and that property holds.

That’s the language of modern mathematics. But not of Euclid. He never uses that manner of speaking which is so natural to logic-oriented mathematics. This fits very well with the hypothesis that Greek mathematicians vehemently rejected the notion that their reasoning was based on syllogistic or propositional logic. Instead they relied on constructions. In fact, they did so in part precisely because logic is so problematic.

Or so I have claimed. As usual we cannot know for certain. Euclid didn’t say why he’s so obsessed with constructions. But I think this is a good way of making sense of it.

Ok, so we have seen a number of specific considerations that point toward the foundational importance of constructions. Let’s bring these ideas together into a single philosophy. In fact, there is such a philosophy. I call it operationalism.

Operationalism is a term most closely associated with a 20th-century movement in philosophy of science that grew out of relativity theory and quantum mechanics. But several of its key ideas are much older and more universal. I propose that this rich tradition in philosophy of science was largely foreshadowed in Greek philosophy of geometry. The key commitments and motivations of modern operationalism and related traditions could very plausibly have been precisely mirrored in Greek geometrical thought.

One of the leading modern defenders of operationalism is the Harvard physicist Percy Bridgman, a Nobel Prize winner. Here’s how he formulates the core principle of operationalism: “we mean by any concept nothing more than a set of operations; the concept is synonymous with the corresponding set of operations.”

Bridgman had physical concepts in mind, but it works equally well for geometry. For example, what does “triangle” mean? The operationalist answer is that “triangle” means: the figure obtained when drawing three intersecting lines with a ruler. This diagram is not a drawing of a triangle, or a physical instantiation of the formal concept of a triangle, or in some other way subordinated to or derived from some purer concept of triangle. No, a diagram resulting from these operations simply is what a triangle is. The act of drawing itself is the root meaning of the concept of “triangle.” The act of drawing is the foundational bedrock on which any claim about triangles ultimately rests. When Euclid says “let ABC be a triangle,” he strictly speaking simply means: draw one line, then another, then another (making three points of intersection).

In the same way, what is a line? Take a piece of string and pull the ends; that’s a straight line. What is a circle? Take a piece of string and hold one end fixed and move the other end while keeping the string taut; that’s a circle. What does it mean for two things to be equal? Put one on top of the other; if they align, and neither sticks out beyond the other, then they are equal. As Euclid says in Common Notion 4. What is a right angle? Cut the space on one side of a line into two equal pieces; that’s a right angle. As Euclid says in Definition 10. And so on.

Every statement Euclid makes in the Elements can be read as a statement about operations or the outcome of operations. Not every geometry book is like that. Far from it. Most geometry treatises of later eras do not allow themselves to be interpreted in operationalist terms.

Consider for example the parallel postulate, Postulate 5. This postulate is very convoluted and hard to read the way Euclid states it. It goes like this: “if a straight line falling on two straight lines makes the interior angles on the same side less than two right angles, the two straight lines, if produced indefinitely, meet on that side on which are the angles less than the two right angles.”

Basically, this says that if two straight lines are heading toward each other then they meet. And the postulate includes a criterion for checking whether two lines are heading toward each other or not. Namely: cut across them with a third line, and check the angles it makes. Less than 180 degrees on one sides means the lines are inclines toward each other, so they will eventually meet on that side, says the postulate.

In fact it is clear already from the Elements itself that Euclid could have used a simpler, equivalent statement in place of this complicated thing. Such as: given any line and any point not on this line, there is no more than one parallel to the line through that point.

Why did Euclid opt for his much more convoluted formulation of the postulate? From the point of view of modern mathematics, his choice is strange, as witnessed by the majority of more modern treatments that much prefer the formulation in terms of existence of parallels. But from an operationalist point of view Euclid’s choice makes perfect sense. Euclid’s version of the postulate is purely about operations: if you draw two lines, and discover by an operational test that they stand in such-and-such a relation, then if you extend them such-and-such a thing will happen. Everything is formulated in terms of actions that the geometer performs. The existence formulation, on the other hand, is incompatible with operationalist principles. It only makes sense in some kind of preformationist framework that assumes that all the objects of geometry are already “out there,” independently of any geometer.

Similarly, Euclid doesn’t say “there are infinitely many prime numbers” but rather: if you have a list of prime numbers, you can make a larger list of prime numbers (Elements, Book IX, Proposition 20). This achieves the same thing but without needlessly entangling itself with the quasi-metaphysical assumption that “the set of all prime numbers” is a preexisting entity whose properties we are proving theorems about. There is no need for mathematics to make assumptions of that type. Doing so would only invite attacks from philosophical sceptics.

Operationalism avoids the dubious ontological assumption that the totality of all objects of geometry are somehow already at our disposal. The modern formulation of the parallel postulate assumes that mathematics can, so to speak, survey the totality of all lines through a particular point and make proclamations about this infinite set. Operationalism doesn’t make such an assumption.

We can also read for example the Pythagorean Theorem this way. Again, this theorem does not say that every element of the infinite set of all right-angle triangles has a particular property. Rather, operationally speaking, it says: if you have drawn a right-angled triangle, and if you then draw the square on each of the sides, then the areas of those particular squares are related in such-and-such a way. Until you have drawn a right-angled triangle, the theorem can be said to have no content.

Operationalism cuts away a huge amount of philosophical baggage, yet still allows us to retain virtually all mathematical content. Insofar as mathematical practice needs to be adapted when operationalism is adopted, this is in the form of explicating constructions for all objects dealt with. But, as we have seen, there are in any case strong internal reasons for mathematicians to adhere to this ideal.

Indeed, operationalist geometry is automatically protected from the fallacies discussed above in straightforward ways. The existence and false diagram issues are resolved because they could never arise in strict operationalist practice. And the verbal logic problems do not arise since verbal logic is not accorded any foundational role in operationalist mathematics. Thus operationalism very conveniently cuts off in one fell swoop numerous lines of attack of philosophical scepticism directed at mathematics, without the need for any sacrifices in mathematical content.

Operationalism is related to positivism. Science can only speak about observable facts, according to positivism. So positivism implies a strong adherence to a scientific worldview as the only source of knowledge, and a rejection of other humanistic or philosophical theories or belief systems. Science, positivists say, prudently restricts itself to what is actually knowable, while other forms of philosophy speculate futilely about the ultimate nature of things and all sorts of other concepts that transcend observable reality. On this view, much grandiose philosophising is wrongheaded and even strictly meaningless.

Here’s how Bridgman puts this point. “It is quite possible, even disquietingly easy, to invent expressions or to ask questions that are meaningless. It constitutes a great advance in our critical attitude toward nature to realize that a great many of the questions that we uncritically ask are without meaning. If a specific question has meaning, it must be possible to find operations by which an answer may be given to it.”

You know how all courses have to have “learning goals” these days? You can’t say: “this course is about quadratic equations.” Instead you have to say: “after completing this course, the student will be able to obtain the solutions to equations of the form blah blah blah.” So you must operationalise what it means to succeed in the course. You must state it in terms of what the student can do. Not in terms of just naming the topics.

Greek geometry was like that too. Everything is formulated in terms of doing. It’s not enough to just give names to things: you must make those names meaningful by explaining what you can actually do with them.

Here’s another quote from Bridgman: “Politics, philosophy and religion are full of purely verbal concepts. Such concepts are outside the field of the physicist. Only in this way can the physicist keep his feet on the ground or achieve a satisfactory degree of precision.”

So positivism and operationalism go hand in hand with an “us versus them”—scientists versus philosophers—type of attitude that is as much about rejecting other perspectives as it is about affirming its own principles.

It is possible that this dynamic was directly paralleled in antiquity. Ancient mathematicians would have felt that their geometry was a lot more grounded in reality than even quasi-science such as the four elements theory, not to mention more abstract philosophy such as, say, Aristotle’s doctrine of causes. Ancient mathematicians would have felt that their results were qualitatively different from philosophy in terms of reliability, objectivity, and many other dimensions. They may even have felt that much philosophy was empty gibberish. Perhaps this would have led them to articulate general methodological principles that would “explain” why their form of reasoning and knowledge was superior to that of the philosophers, as many scientists have been inclined to do ever since.

What methodological dicta might Greek mathematicians have seized upon to set their field apart from philosophy? Certainly not anything like the modern identification of mathematics with logic and axiomatic-deductive reasoning. Logic and deduction were already highly prized among Greek philosophers. If anything, they were too obsessed with deductive logic: Zeno’s argument that there can be no such thing as motion is one example among many of extreme faith in abstract deductive reasoning even when it is in blatant conflict with the most basic common sense. So ancient mathematicians could certainly not hope to stand out by their reliance on abstract deductive reasoning.

Axiomatics too was far from the exclusive purview of mathematicians; indeed it is obvious that basing one’s theories on a list of allegedly evident but ultimately unjustified axioms is very convenient for mathematicians and sophists alike. It may even be reasonable to say that the abundance of deductive philosophical systems that were clearly in conflict with one another would rather have been an incentive for the mathematics to insist that, unlike the philosophers, they did not rely on abstract logic.

Operationalism would have been an alternative readily at hand. Constructions had always been a central part of geometry, from the time of the Egyptian “rope-stretchers” whom the Greek identified as the originators of the field. Later theoretical developments, such as the irrationality of the square root of 2, had spoken in favour of taking geometry as the foundational bedrock of all mathematics. It would have been a short and natural step for the mathematicians to tie the foundations of their subject to their already ubiquitous ruler and compass.

To the mathematicians it would have cost little to embrace all-out radical operationalism. Virtually all of mathematics was readily susceptible to being reframed in such a paradigm. It would have been a way of legitimating existing practice that would have necessitated little or no deviation from what they were already doing. Meanwhile, other branches of philosophy stood no chance of founding their teachings on an operationalist basis. So if the mathematicians were looking for a way to set themselves apart from the philosophers—to explain why their field had cumulative progress, universal agreement, and inviolable truths while philosophy had paradoxes and schools in constant disagreement with one another without any prospect of reconciliation—then operationalism would have been the obvious way to go.

Another virtue of positivism that could be held up as distinctive is that it restricts all knowledge claims to the domain of what is actually knowable in a straightforward empirical sense. Failing to adhere to positivism means making statements that are, by their very nature, empirically unverifiable and hence arguably unknowable almost by definition. Unlike most of philosophy, any statement of geometry is readily equated with a claim regarding certain empirical circumstances. Ancient mathematicians had a golden opportunity to highlight this natural attribute of their field as an epistemic virtue. They could pose to head-in-the-clouds philosophers the very difficult challenge of explaining what good a theory is if it has no “cash value” in the real world, in the form of empirically testable claims. And they could stress that geometry, by contrast, has no need to engage in that kind of theorising.

Related to this is the ideal of falsifiability. When the geometers claim that any triangle has an angle sum of two right angles, they are sticking their necks out. If their claim was false, it should be simple enough to find a counterexample. The operationalist formulation of geometry makes it possible to press this point very strongly. The theorem simply means: if you put a ruler down on a piece of paper and draw three intersecting lines, then cut out the three corners and put them point-to-point, then the three pieces fill precisely the angle on one side of a straight line. The very meaning of the theorem directly contains a concrete recipe for testing and potentially falsifying it.

Karl Popper, the philosopher of science, is the name most prominently associated with the philosophy of falsificationism. This was in the first half of the 20th century. To Popper, falsifiability is what set science apart from non-science.

As examples of non-scientific theories, Popper had in mind things like the theories of Marx and Freud, which were influential at that time. These theories had a sort of quasi-scientific appearance. They postulated fundamental laws and used these to explain many phenomena. But according to Popper it was pseudo-science. Because, no matter what phenomena were observed, they could always tell some story about how that fits with their laws.

So these theories pretended to have laws, but they were vague enough to allow many different possible applications, so that almost anything could be construed as consistent with these laws one way or the other. Just as astrological horoscopes in the newspaper make so-called predictions about the future, but in fact they are so vague that they can often be interpreted as having been correct no matter what happens.

This is why Popper emphasized the importance of falsifiability. For a prediction to be scientific, there must be clearly specified condition under which it would be regarded as having failed. The scientist must say: if such-and-such a thing happens, then I was wrong. Before making an experiment or observation, the scientist has already set down those criteria, that is to say, the conditions under which the theory must be regarded as having been falsified.

Non-scientific theories like those of Freud or Marx are not like that, according to Popper. Advocates of those theories use them to “explain” all kinds of things, but they never say: if such-and-such a thing were to happen, then that would prove me wrong and I would give up the theory.

Formulating geometry in terms of constructive operations is a great way of making it scientific in Popper’s sense. It makes the theorems of geometry directly testable. Euclid’s constructions are like lab instructions for carrying out such a hypothesis test. Do the construction and measure for yourself if it came out the way the theorem said.

Euclid’s parallel postulate is something that can be performed and tested in a very concrete way. It says: here’s what going to happen if you draw this kind of configuration.

Alternatives to the parallel postulate are not like that. Instead of the parallel postulate, you could say: Given a line and a point, there is precisely one parallel to the given line through the given point.

How would you test that? It’s in the form of a metaphysical statement, rather than in the form of a falsifiable scientific hypothesis. There is one and only one parallel. It’s like saying: There’s one and only one God. How can you verify that? How could you even prove it wrong? You can’t. Unlike scientific hypotheses, and unlike operationalist geometry, statements of that form do not come with a concrete set of operations one can perform to see if it works or not.

Operationalising geometry makes it falsifiable. It also makes geometry theory-independent. You do not need to accept the definitions and postulates of the mathematicians in order to perform this empirical test. Sceptics who try to criticise mathematics in general terms can thus be confronted with a concrete challenge: regardless of whether you accept any of our assumptions or modes of reasoning, we offer you hundreds upon hundreds of claims of the form: if you perform such-and-such concrete operations, then the outcome will always be one particular way rather than another. Feel free to prove us wrong, the mathematician can say. It would be impossible to meet the challenge and very difficult to try to dismiss it as illegitimate.

The operationalist formulation of mathematical statements is reducible to straightforward recipes whose neutrality and objectivity is very difficult to deny. This is in stark contrast with many philosophical claims, which must often be bought into or rejected wholesale along with an entire theory because all the parts of the theory are interdependent. Even the very meaning of the concepts the theory uses is inherently bound up with the system as a whole.

Operationalism ensures that geometry is not like that. Operationalist geometry is not an entangled holism.

Here’s an analogy for this. Consider a casino. It has roulette and black jack and so on. You play with casino money. Plastic chips that only have meaning and value inside the casino. Once you leave the casino you can’t buy anything for those worthless poker chips.

Non-scientific thought-systems such as philosophy or religion are like the casino. Internally, they have all kinds of intricate laws and explanations for how everything fits together. And it’s easy to get caught up in the system once you buy into it. But to link it to the real world, you have to ask yourself: what’s the actual cash value of this stuff? That is to say, what could I actually do with any of this in the real world, concretely?

Operationalism is “cash value” geometry. It translates everything into real-world operations that anyone can perform. It’s cash money. You can use it directly and it works. It’s not casino money, which only makes sense if you accept the entire premise of the casino with all its internal rules.

Even someone who doesn’t believe in the postulates of Euclid, or doesn’t believe in geometrical proofs, etc. Even such a person can test these things. They can cut the corners off a triangle and see if they fit together the way Euclid says. Or they can draw squares on the sides of a right-angle triangle and see if the areas are equal the way Pythagoras says. Those are scientifically, concretely, real-world testable claims.

Let’s summarise. Operationalism safeguards mathematics against a multitude of plagues. It prevents us from reasoning about entities and concepts that are inconsistent, incoherent, non-existent, or imaginary.

Mathematicians would have had every reason to articulate such a philosophy. Greek antiquity was an age of sceptical philosophical attacks. Mathematics would have found itself under fire, and its enemies were no fools. The logic and rigour of mathematical proofs were by and large hugely impressive. Yet it had a conspicuous Achilles heel: a veritable self-destruct button that could bring the entire edifice crashing down at the slightest trigger. For if there was any way an inconsistency could slip into mathematical reasoning undetected, then everything that followed would immediately be rendered logically worthless. What guarantee do we have that this will never happen, or indeed that it has not already happened?

This vulnerability pertains especially to the way objects are introduced into mathematical discourse. It is safe to say “let ABC be a right-angled triangle,” but if you say “let ABC be a triangle with two right angles,” then you have introduced an inconsistency and all is lost. Then you can prove that 2 is equal to 1, and the entire credibility of mathematics collapses. So geometry needs to systematically guarantee that it could never commit an error of this type. In other words, it needs a meticulous gatekeeping policy that only allows the most carefully vetted entities to enter mathematical discourse.

Constructions are the answer to this problem. By insisting that geometry only speaks of entities that are constructed, the mathematician immediately knocks the legs out under boogeymen examples of inconsistent objects such as the superright triangle.

Constructions also ground mathematics in reality and gives a straightforward account of what geometry is and what geometrical statements mean. This can be used to set geometry apart from empty philosophy, from metaphysics, religion, astrology, all kinds of empty pseudo-science.

Philosophers of science in the 20th century spent a lot of effort trying to formulate the criteria that distinguished science from non-science. One of their answers was falsifiability: scientists bravely specify what would prove them wrong. They say: try this for yourself, and if it doesn’t come out the way I said I promise I will admit that I was wrong and that my theory should be rejected.

They also found that to follow through on this program it was important to translate abstract theoretical notions into observable real-world terms. Instead of merely speaking abstractly about for instance the concept of the force of gravity, it is necessary to translate the meaning of that theory into something doable, testable, such as: if you hang this led weight from this spring, then the spring will extend by so-and-so many centimeters. Things like that is what the concept of gravity comes down to in practical terms. This concreteness is essential to science, and essential to separate science from fancy games with words.

Euclid’s geometry is a perfect fit for all this stuff. It’s almost as if Euclid had read these 20th-century philosophers of science. Maybe Euclid and his friends had many of the same ideas. Maybe they too wanted to set their theory apart and explain why it was superior to other branches of philosophy. The way they based geometry on constructions is a perfect fit for making those kinds of arguments.

So there you go. These are many reasons to ground geometry in constructions. It is not for nothing that all depictions of Euclid shows him with ruler and compass in hand. These are no mere practitioner’s tools. They are in fact essential even to the theoretical foundations of geometry in numerous respects. That is what I have tried to argue.

]]>**Transcript**

Euclid’s Postulate 4 is super weird. It says: “all right angles are equal.” What kind of a postulate is that? 90 degrees equals 90 degrees? A right angle is equal to itself?

Why would you need to state that as an axiom? And if you do need to state it as an axiom, why only right angles? Why wouldn’t you need other axioms starting that various things are equal to themselves? 10 degrees equals 10 degrees, 1+1 equals 2: Why don’t we need axioms like those? What’s so special about right angles? Why do they need to be singled out like that, in their very own postulate?

But Euclid knew exactly what he was doing. His postulate only appears crazy and weird. There’s a way to make sense of it. We must reconstruct the original context and intent of the postulate.

I say: Euclid included this postulate in order to rule out cone points. I will explain what I mean by this. But let’s note first the historical methodology we are using here.

There is nothing in Euclid, and in fact nothing anywhere in any ancient source, that actually says that this was the intent of the postulate. The interpretation that the postulate has to do with cone points is purely a hypothetical reconstruction by historians, formulated thousands of years after Euclid.

Yet the reconstruction is so convincing. It just has to be right. If it’s right, everything fits; everything makes sense. If it’s not right, then we can’t explain the postulate, and we just have to assume that Euclid hadn’t really thought it through all that carefully and just put the postulate down on kind of whim or whatever and it doesn’t mean all that much in the greater scheme of things.

This is the difference between a great text and an average one. Great texts in intellectual history, like Euclid’s, reward reflection. If something seems weird it’s because you haven’t understood it. There’s a reason behind every step of the text. The text is the tip of an iceberg. It’s built on a huge body of supporting thought. This is why the text rewards reflection. The text is not just whatever popped into Euclid’s head. It is the fruit of an intellectual culture where these ideas had been scrutinized and criticized forwards and backwards and inside and out.

This is why you should read great texts like Euclid’s. These are the kinds of texts that, every time you dig into even the weirdest parts you realize that, huh, that’s actually a good point. The more you probe the text, the more compelling it becomes. It’s the mark of a great text that when you scrutinize an apparent weakness, it turns out to be a strength.

Euclid’s right angle postulate is an example of this. It looks silly and weird at first sight, but when we think about it, it opens our eyes to new and unexpected perspectives and insights.

So, “all right angles are equal,” what’s the deal with that? First of all, what does “right angle” mean? Euclid defines it in Definition 10. Draw a line. Consider the space on one side of the line. Cut that space in half with another line. That’s a right angle. A right angle is half the space on one side of a line.

So what does “all right angles are equal” mean? It means: Suppose you have made a bunch of right angles. That is to say, you have drawn various lines and then cut the space on one side in half. So you have a piece of paper full of what looks like a bunch of copies of the letter capital T. There are a bunch of T’s scattered across the paper, at random angles and positions. “All right angles are equal” means: if you cut out one of those T’s and put it on top of one of the other ones, then it fits. All the different T’s align perfectly with each other, as far as angles are concerned.

So the right angle postulate is really a kind of homogeneity postulate. It effectively says that no part of the paper is different than any other. The space on the side of a line is the same anywhere.

A cone is an example of a surface where that is not the case. Like an ice cream cone. The cone is non-homogenous. It has an exceptional point, the cone point, the apex of the cone, which is different from the other points.

Euclid’s postulate is false on the cone. A right angle at the cone point is smaller than a right angle elsewhere.

You can see this if you think about how a cone is made. You can make a cone like this. Start with a circular piece of paper. Then cut out a wedge from it, like a pizza slice. Then grab the two sides of the cut and pull them together. Now you have a cone.

Think about the amount of space around each point. Most points are surrounded by the same amount of space as they were originally, on the paper we started with. 360 degrees’ worth of space, so to speak.

But the cone point is different. It’s surrounded by “less space” or “fewer degrees” than before. The pizza wedge you cut out took away some of the angle sum around this point. Not so for any other point. Even the points along the sides of the cut are still normal. They lost 180 degrees, but then you pasted another 180 degrees right back in. So they are back to normal. But not the cone point. It lost some of its angles and never got them back.

So the right angle postulate is false on the cone because right angles are smaller at the cone point. Since right angles are defined in terms of cutting the space on the side of a line in half, then if there’s less space around some points compared to others, then the right angles there will be smaller too.

I believe this is what Euclid had in mind when he wrote his postulate. We can’t prove that this is what Euclid meant, but it is the most satisfying explanation.

Here’s a little cultural sidelight. The Declaration of Independence of the United States starts in a kind of Euclidean manner. It says: “We hold these truths to be self-evident,” and then it lists a number of “truths” the first of which is “that all men are created equal.”

So the Declaration of Independence has self-evident axioms just like Euclid, and they sound the same too: “all right angles are equal”; “all men are created equal.”

That’s no coincidence. The founding fathers of the United States were obsessed with antiquity. They used the ancient world as a model all the time. As a model for their political system, of course. The senate, for instance, is straight up copied from Rome, and so on with many other things. Euclid was part of that package as well. A very conscious revival of ancient enlightenment.

So the founding fathers of the United States called their axioms “self-evident.” And of course many people have interpreted Euclid that way too. You don’t have to prove the postulates because they are immediately obvious. You can draw a line from any point to any point: yes, of course you can, it’s too simple to even prove, but it’s impossible to doubt. That’s one way to think about Euclid’s postulates. A common way.

But one could argue that it’s a bit more complex. This is suggested even by the word that Euclid uses: postulate. These simple and self-evident starting points are called postulates. But this term doesn’t suggest that these things are self-evident or impossible to doubt. To postulate is more of a demand or a request. So the term doesn’t seem to take assent for granted but rather the opposite: it seems to imply that some people might oppose these statements, no matter how obvious they might seem.

How could anybody deny that you can draw a line from any point to any point? In fact, some people in Ancient Greece did deny this, and they were not crazy; they had some very compelling reasons.

A useful book on this is The Beginnings of Greek Mathematics by Arpad Szabo. I will summarize it for you.

“Mathematics grew out of the more ancient subject of dialectic”—that is to say, philosophical debate. Just as we discussed before, the argumentative Greeks, they loved debating. Two philosophers passionately disagreeing and trying to poke holes in each others’ arguments in a lively disputation before an audience: that was their idea of a good time. Instead of “dinner and a movie” you would go to a philosophy debate.

So that’s “dialectic”—a debate with two warring sides. Terms such as axiom, postulate and many others seem to have originated in this setting. These terms were imported into mathematics from dialectic. Today only their mathematical meaning survives. Therefore to us these terms have rather different connotations than they did for the ancient Greeks. That’s how the idea that axioms or postulates are supposed to be self-evident has become associated with the terms even though that was not the original intent or meaning.

The terms axiom and postulate originally mean something like “concessions which the participants in a discussion have agreed to make.” “We know that the term aitema [=postulate] came from dialectic where it was used to denote a ‘demand’ about which the second partner in a dialogue had reservations.”

“Let us see whether there is any connection between this early meaning of the word and Euclid’s postulates. At first glance, Postulates 1-3 appear to be such simple, self-evident and easily fulfilled ‘demands’ that one is tempted to disregard the literal meaning of their name.” But no.

Euclid’s postulates arguably rely on motion. To draw a straight line from any point to any point: how do you do that? You put a ruler down and trace the line with a pen. The pen is moving: you put it at one point and move it to the second point. Same thing with circles: you draw them with a compass, which is also a moving instrument.

It’s quite possible to deny that such things can be done. In fact, you may have heard about the famous paradoxes of Zeno, which purport to prove that motion is impossible. One of them goes like this.

Suppose I have to walk from A to B. Before I can walk all the way to B, I first have to walk half the way to B. Then, when I’m at the halfway point, before I can get to B I have to walk half of what’s left. And so on. Whatever distance is left, I always first have to go half of it.

But this process never ends. There’s always “another half to go.” So to go from A to B you have to “do an infinite number of things,” so to speak.

You can think of it this way. When I have gone half the way from A to B, I say: one. Then when I have gone half again of what’s left, I say: two. I go half of what’s left: three. And so on. This implies that if in fact I can go all the way from A to B, I will have shouted out all the numbers that exist: one, two, three, four, five, … all of them.

So to say that you can go from A to B is to say that you can count through all the numbers in finite time. But of course you can’t. Nobody has ever counted through all the numbers. So therefore you can’t move either. Motion is impossible. It must be an illusion.

We only think we move. That’s feeble sensory “knowledge,” or so-called knowledge. We discussed before the extreme rationalistic tendency of Greek philosophy: reliance on pure deductive reason at the expense of all other forms of knowledge. Zeno’s paradox is an example of this. The senses say we can move, but deductive “reason” says we cannot.

We discussed before how the stage debate format incentivized philosophers to pick the side of reason in such cases, no matter how extreme and outrageous the conclusion may be. “All is water”, “all is fire”: the crazier the better. Proofs of radically unexpected conclusions is perfect for the stage debate setting.

Zeno’s argument is great way to dazzle an audience and to show how clever you are. Being reasonable and arguing that one can walk from A to B is boring. Who wants to hear that? You won’t become a blockbuster debate star by arguing for the obvious. You gotta have some signature absurdities that you claim to prove.

Zeno also had a second form of his argument that is equally amusing. Here’s how Simplicius describes it:

“The argument is called the Achilles because of the introduction into it of Achilles, who, the argument says, cannot possibly overtake the tortoise he is pursuing. For the overtaker must, before he overtakes the pursued, first come to the point from which the pursued started. But during the time taken by the pursuer to reach this point, the pursued always advances a certain distance; even if this distance is less than that covered by the pursuer, because the pursued is the slower of the two, yet none the less it does advance, for it is not at rest. And again during the time which the pursuer takes to clever this distance which the pursued has advanced, the pursued again covers a certain distance. And so, during every period of time in which the pursuer is covering the distance which the pursued has already advanced, the pursued advances a yet further distance; for even though this distance decreases at each step, yet, since the pursued is also definitely in motion, it does advance some positive distance. And so we arrive at the conclusion that not only will Hector never be overcome by Achilles, but not even the tortoise.”

So that’s another way to prove that motion is impossible. Those who believe in motion believe that Achilles can out-run a tortoise. But that contradicts reason, as we have just seen. Therefore those who believe in motion must be wrong.

Why did Zeno prove the same thing in two ways? Maybe he was just like: Hey guys, I thought of another funny one, it has a tortoise in it, I’m sure you’ll get a kick of it. Or is there more to it than that? Do Zeno’s two forms of the argument differ in substantial respects?

I think they are subtly different. You might say that the Achilles argument assumes the possibility of motion and derives a contradiction. It so to speak plays along with those who believe in motion for a bit, only to then trap them in a paradox.

The other argument—the dichotomy, or half half half argument—doesn’t really need to even presuppose motion at all. It derives the impossibility of motion more from the nature of length. It has more to do with the infinite divisibility of the continuum than with motion as such.

So in that respect the dichotomy argument is more “pure” as it were. Since it doesn’t need to use motion to refute motion.

But on the other hand it is less pure in another respect. It assumes metricity; that is to say, an absolute notion of distance. For the argument to work, it must be possible to talk about the half of something. But half involves quantification. You need to put a number on the full length before you can know what half of it is.

So Zeno’s opponents could say: Your argument doesn’t disprove my beliefs, because although I believe in motion I do not believe in metricity. I do not believe that numerical lengths can be assigned objectively to the paths between points. Therefore the whole business about halfs doesn’t work, and you haven’t really disproved motion after all.

If Zeno’s opponents tried to wiggle out from under the dichotomy argument along those lines, then Zeno could just hit them with the Achilles argument. Because the Achilles story doesn’t involve assigning numerical lengths to anything. It purely about relative positions: the tortoise is in front of Achilles. It doesn’t say by how much. The argument doesn’t need the notion of being in front to be quantifiable. It needs only relative positions. So in that sense the Achilles argument is the purer one.

Well, that’s fun to think about, but let’s get back to our original purpose. I brought up Zeno’s paradoxes because they are related to the issue of whether Euclid’s postulates are obvious or not.

“If we bear [Zeno’s paradoxes] in mind, it is easy to understand why Euclid’s first three postulates had to be laid down. They really are demands (aitemata) and not agreements (homologema); for they postulate motion [such as the motion of a pen that is drawing a circle], and anyone who adhered consistently to [Zeno’s] teaching would not have been able to accept statements of this kind as a basis for further discussion.”

So when Euclid is presenting his postulates, he doesn’t seem to be saying: surely you all agree with these statements; they are clear even without a proof. Instead Euclid seems to mean by postulate: these are assumptions that must be accepted for the sake of argument if we are to do geometry; if you don’t like them, then we just have to agree to disagree.

The same goes for Euclid’s Common Notions. “Our text of Euclid” has a separate heading called common notions, but this was not a well-entrenched term and these principles “obviously bore the name axioma in pre-Euclidean times,” and “the noun axioma, when used as a dialectical term, originally just meant a ‘demand’ or ‘request’.”

Indeed the common notions could be doubted. They “are assertions which are justified by practical experience and, in some cases, directly by sense-perception. [One of them] states that ‘things which coincide with one another are equal to one another’. It can literally be seen that plane figures which coincide are actually equal; hence this axiom is verified by sensory experience.” Therefore the common notions “could not have been accepted by [those] who required that all knowledge be obtained by purely intellectual means and without appealing to the senses.” And there were plenty of people like that. Just as Zeno’s argument implies: an extreme trust in purely intellectual reasoning, even when it goes flatly against even the most basic and immediate experience.

This is why “these principles were originally called demands (axiomata): because the other party in a dialectical debate had reservations about accepting them as a basis for further inquiry or, in other words, because their acceptance could only be demanded.” People would not have agreed that these things were self-evident; that’s why they had to be “demanded”, or postulated.

There’s yet another way to criticize Euclid’s principle that “things which coincide with one another are equal to one another.” Not only does it rely on sense evidence, it is also arguably conceptually incoherent. If “two” things coincide and are equal, doesn’t that mean that they are actually one thing? Does it even make sense “to speak of two things unless they can be distinguished from one another“?

So we see how Euclid’s axioms can be questioned in various ways. The Greeks loved to quarrel. Mathematics was born in this kind of climate. Everybody criticizing everything, trying to poke holes in it.

So that’s why Euclid’s text starts with “demands.” Many later readers were happy to accept them as self-evident, but Ancient Greek geometers could not have expected to get away with that.

So the terminology of “postulates” and “axioms” points to this ancient context. But the meaning of the terms morphed over time. In the very early days, mathematics lived within the dialectical tradition and was a subordinate part of it. But mathematics took on a life of its own and soon outlived dialectic.

Soon “the essentials of [the old] dialectic [context] were no longer very well understood; hence the ancient term axioma acquired a new meaning. Since it had always been used to refer to a group of principles which, from the viewpoint of common sense, were evidently valid, it came now to denote those statements whose truth was ‘accepted as a matter of course’.”

So that American phrase—“we hold these truths to be self-evident”—is perhaps not as Euclidean as Jefferson and those guys thought.

Here’s another interesting aspect of Euclid’s postulates. The first three postulates basically state that lines and circles can be drawn. That is to say, lines and circles can be taken to exist. That’s a primitive assumption of geometry.

Lines and circles are so to speak the Adam and Eve of geometry. In the beginning there are only these two, these male and female generative principles. You couldn’t get very far with just one of them, but together they combine to make rich offsprings. They eventually populate the entire Euclidean universe. Everything that ever happens in Euclid’s world comes from these two parents, the line and the circle.

The line and the circle are also embodied in physical tools: the ruler and the compass. To what extent is that important? Is this physical realizability important to the credibility of these postulates? Or is Euclid merely talking about lines and circles in the abstract, and it’s just a coincidence that they correspond to physical tools?

There is no simple answer. Euclid’s text is ambiguous in this respect. You can read it either way.

Insofar as we can say anything about what Euclid meant in this respect, we must infer it from the technical material later in the text. Euclid never tells us: “here’s my philosophy.” We can only read his proofs and ask ourselves what implicit assumptions appear to be made and what implicit philosophy might have guided the particular choices Euclid makes in technical arguments.

Already Proposition 2 is very interesting in this regard. Euclid shows in Proposition 2 how to transfer a length from one position to another, using only his postulates about line and circle, or ruler and compass.

In other words, somebody has drawn a line segment on a piece of paper, and now you want to draw an equally long line segment somewhere else on the paper.

Euclid accomplishes this by a very elaborate construction. It involves drawing numerous circles and an equilateral triangle. Very elegantly, this leads to exactly what you need: the given segment has been reproduced in the new position, with exact mathematical precision.

That’s all very neat, but it’s also weird, isn’t it? It seems totally out of touch with reality. If a craftsman or engineer or architect would need to transfer a length, surely they would not use Euclid’s absolutely baroque procedure.

First of all you might say: just use a ruler. Measure the given length. It’s so-and-so many centimeters. Then put the ruler wherever you want the length to go, and mark off the same number of centimeters there. Done. No need for drawing a bunch of circles and god knows what else.

Why doesn’t Euclid accept this and save himself some time? Actually it’s not so crazy. In a way you might it’s a mistake to think that length lives in the ruler. Actually, out of the two “parents” ruler and compass, length comes from the DNA of the compass, not the ruler.

We are so used to working with rulers, measuring things with rulers. It’s the prototypical manifestation of length. But think about it. Where do rulers come from? How do you make a ruler? How do you put the centimeter marks on it?

You do it with a compass. You set the compass to a fixed opening, and you mark off the size of that opening repeatedly along the ruler. Can you feel it? You start with a blank ruler, just a straight piece of wood. Now you take your compass and make it so to speak “walk” along the edge of the ruler. Left foot, right foot, left foot, right foot. The places where the compass “stepped” so to speak become the marks of the ruler. So when you use a ruler to measure things, you are really relying on the compass. Length is born from the compass.

This suggests that Euclid is on to something when he involves circles in his proof of Proposition 2. But it still doesn’t explain why it has to be quite so complicated.

A compass can solve the problem directly. Just open the compass to the length you want to move, then lift it and put it back down wherever you want the length to go. The length you wanted in the new position where you wanted it is directly manifested in the form of the distance between the two legs of the compass. Piece of cake. There’s nothing to it. You can move lengths directly with the compass without any hassle.

Euclid acts as if this is not possible. One might say that Euclid behaves as if his compass is “collapsible”: it stays at a fixed opening while drawing a particular circle but as soon as it is lifted from the paper it “collapses,” or closes up, so that the opening to which it was set is lost and cannot be used elsewhere.

Of course there are no collapsible compasses. It’s not a real thing. So you might say: aha! This proves that Euclid is in fact talking about lines and circles abstractly, maybe in the manner of Plato and his world of ideals. From that point of view Euclid’s proof is not problematic. It’s a dazzling intellectual construction. Great stuff. Hopelessly impractical, to be sure, but that’s just all the better of course as far as Plato is concerned.

Meanwhile, if you want to say that Euclid’s postulates correspond to actual rulers and compasses, then you have to bend over backwards and make up stories about “collapsible compasses,” which don’t exist.

So it seems we have a clear winner. Only the abstract, non-physical reading of Euclid makes sense.

But I’m not so sure. Maybe “abstract versus physical” is the wrong lens to use here. We can also make sense of Euclid’s peculiar proof from a different point of view that is independent of this issue of physical versus abstract.

This point of view is: assumption minimalism. Euclid masterfully reveals the minimum assumptions necessary for geometry. Remember: reduce, reduce, reduce. That seems to be Euclid’s mantra. That’s the philosophy of “reading backwards.” If you can avoid an assumption, then you should avoid that assumption.

This kind of minimalism or purism doesn’t depend on whether geometry is physical or abstract. Either way, if something can be proved rather than assumed, then that’s regarded as a win. This kind of reduction is about exploring and clarifying the ultimate foundations of geometry and the bedrock source of geometrical knowledge. It is applicable regardless of whether geometry is physical or abstract.

This perspective of minimalism demands that we do not allow lengths to be merely transferred directly by a compass. Even if we do think physical compasses are somehow important to geometry, we should still pursue this reduction. It is our duty to always reduce.

Just as a chemist reduces molecules to atoms. Of course molecules are great. The best level at which to explain many things is molecules, not atoms. But since they can be reduced, they must be reduced. It is our scientific duty to run the reduction as far as it goes. Of course we still retain the explanatory power of molecules. The reduction to atoms is just a supplement.

Maybe so also in Euclid. Maybe the physical compass should be seen as the operative tool throughout the Elements, just as molecules are the right level of analysis for many chemical phenomena. But even so it makes sense to show up front how it could, in principle, be reduced even further to more basic building blocks. We might say that Euclid does this in Proposition 2.

The fact that one can do away with the assumption that a compass can transfer length is an interesting foundational insight. Since Euclid can prove this, he does. This does not imply that he is opposed to the idea of a non-collapsible compass. One could simply delete Proposition 2 from the Elements and all the rest would still stand verbatim as a treatise about constructions with non-collapsible compasses.

So Proposition 2 can be viewed as an optional exercise in foundational minimalism within a paradigm otherwise fully based on physical compasses. Rather than as evidence of conceptions fundamentally at odds with such a physical point of view.

Analogous situations occur in modern mathematics all the time. For example, open any textbook on abstract algebra and turn to the definition of a group. The definition of a group says that any group has an identity element: anything multiplied by the identity stays the same. As far as this definition is concerned, there could potentially be several identity elements in any given group. However, all textbooks immediately proceed to show that the identity element is in fact unique. Other groups axioms imply that it must be unique.

These textbook authors could have made life easier for themselves by simply making the uniqueness of the identity element part of the definition. Then there would have been no need to prove it a separate theorem. But it is better to keep definitions and axioms as simple and minimalistic as possible, for instance in order to minimise the risk of inconsistency, or because proving properties instead of gratuitously including them in the definition illuminates fundamental relationships.

But note that one cannot infer from this that the uniqueness of the identity is somehow a secondary or less embraced aspect of the group concept. It is proven as a theorem rather than included in the definition solely because of the technical possibility of doing so, not because it was seen as less essential than the definitional group properties. This does not show that the fundamental conception of a group that mathematicians have in mind is ambivalent regarding the uniqueness of the identity. On the contrary, this is arguably a core aspect of the intuitive notion of a group that has, in itself, no less of a claim to being fundamental than the definitional properties. But if one tries to find the smallest set of key properties of a group to take as definitional, then one finds that uniqueness of the identity is a property that can most efficiently be made into a theorem.

In the same way, then, one might argue, Proposition 2 of the Elements does not show that Euclid’s fundamental notion of the circle-drawing constructions and postulates were divorced from a physical compass. It does not prove this any more than a modern textbook proves that the uniqueness of the identity is fundamentally divorced from the group concept.

Just as a modern algebra textbook would have nothing a priori against including uniqueness of the identity in the definition of a group, so Euclid may very well have had nothing a priori against assuming a non-collapsible ruler. Just as the modern algebra textbooks nevertheless arrives at the conclusion that it is better to make the uniqueness of the identity into a theorem because that enables the minimisation of definitional properties overall, so Euclid may very well have decided to assume only a non-collapsible ruler purely for reasons of axiomatic minimalism. If so, it would be a mistake to infer from this proposition that he didn’t care about physical tools like the compass.

Even if you’re not familiar with group theory I’m sure you have encountered a similar aesthetic elsewhere. For example, some people, when they cook pasta, they save a few spoonfuls of the cooking water and toss it into the dish. To make the pasta less dry.

I always thought it’s a little pretentious when TV chefs do this. Obviously you could achieve the same result various other ways. Instead of adding some of the cooking water, you could add other water, oil, make your sauce a bit runnier, etc. I’m sure nobody could tell the difference.

But it’s cool somehow to use the actual cooking water. It makes you feel creative and spontaneous. Almost spiritual: it’s like you’re in synch with the universe like some ancient Indian who lived in harmony with the land. Making use of everything, every part of the pig, even the cooking water. It takes skill and true understanding to use things for something other than their intended purpose. It’s a rock ‘n’ roll move. Anybody can cook the way it says on the tin, but I’m such a creative rebel that I use the very cooking water itself.

Euclid’s Proposition 2 is a bit like that. Of course you could accept the transfer of lengths as a separate assumption, or implied by the compass. But it’s cooler if you can do without it, and instead use what is already at hand in an unexpected new way. Euclid uses the cooking water, so to speak. He uses the assumptions from the postulates that were already necessary anyway. By cleverly combining these, he shows that you don’t need anything else. It’s satisfying in the same way the pasta trick is satisfying.

I think this is enough to explain why Euclid wanted to include Proposition 2. So we don’t need to attribute to Euclid any anti-compass agenda. It’s enough that he thought this was a cool trick.

So the question is still open then whether Euclid meant his postulates to correspond to ruler and compass or not. We will have to keep reading to find out more. Let’s do that.

]]>