Search

Infinitely many Proofs of Pythagoras' Theorem

This post is inspired by an MSE question and the brilliant answers that it received. However, it can be understood independently. Everything here is joint work with Satvik Saha.

Consider the following diagram made of 16 right triangles of sides ab<c and four squares of side ba. This being a tiling of the outer square (of side length 2c), we have
(2c)2=4(ba)2+16×12ab
hence implying
c2=a2+b2
thus proving the Pythagoras' Theorem.

Outer square of side 2c



















This is similar to previously known proofs (eg. see proof #3 and #116 here), but it is hard to say whether this exact tiling appears anywhere before.

Now, consider the following diagram made up of right triangles of sides ab<c, (pink) sqaures of side a, (yellow) squares of side b, and (green) squares of side (ba). The reader is welcome to check that this also gives a completely valid proof of the Pythagoras' Theorem.

Outer square of side 3c



















Some of the readers might have started asking the right question. And the answer is yes, it can be generalised! With the outer sqaure of side nc, we get 4(n2) of the red and yellow squares, n24n+8 of the green squares, and 4(n24n+8) of the triangles. This gives
n2c2=4(n2)(a2+b2)+(n24n+8)(ab)2+2(n24n+8)ab
hence proving the Pythagoras' Theorem. For your viewing pleasure, here's the 16×16 big square. Feel free to print it out and put it on your wall.

Outer square of side 16c

Of course, the keen ones among you would ask whether this is truly a tiling for any values of a and b or is it a coincindence of the value chosen in the diagrams. To clear this doubt, here's an animated version of the diagrams. Notice that the animation also shows that it doesn't matter whether you call the larger side of the triangle a or b.

Outer square of side 6c



















The keener ones would of course ask for a proof that this is indeed a tiling. For that, one only needs to note that the diagram is made of four spiky triangles (made of a layer of blue triangles followed by a layer of yellow and a layer of pink squares followed by a number of layers of blue triangles) - this is almost a triangle except two spikes on two sides. All one needs to convince oneself is that these spiky triangles fit with each other properly and leave a space in the middle which can be nicely filled by the green squares.

Feel free to play with the animation here (p5js code here). Also, feel free to check out my answer in the aforementioned MSE post.

Worst Way to Calculate Area of a Triangle

A useful technique from Complex Analysis helps us to identify whether a real number u is positive or negative by checking a discontinuous integral. We have
12πicic+iesudss={0 if u<012 if u=01 if u>0
for c>0.

We will apply this to calculate the area of the triangle
T(N):={(x,y)R0:ax+byN}
with (a,b)=1. This area is given by
x,y01Naxby0dxdy=x,y012πis=cic+ies(Naxby)dssdxdy=12πis=cic+i(x0easxdx)(y0ebsydy)esNsds=1ab12πis=cic+iesNs3ds
by the above mentioned trick. Notice that the integral with Naxby=0 is zero, and hence was ignored.

Now, moving the contour far to the left we pick up the residue from the pole of order 3 at s=0 and hence the area comes out to be
1abN22=N22ab
which matches with the area calculated by other known methods!

Acknowledgement : I learnt this from lectures by Prof. Andrew Granville for the course Distribution of Prime Numbers.

Infinitely many Proofs of Pythagoras' Theorem

This post is inspired by an MSE question and the brilliant answers that it received. However, it can be understood independently. Everythin...