17
Stack Exchange
votes

Picture of Lambert's proof that is irrational?

Posted on 11/12/2020
UQ ID:225
Science
MathOverflow
Original

With a suitably generous notion of "picture proof" or "proof without words" or "geometric proof," there do exist such proofs of the irrationality of square roots and even of the irrationality of . However, I am not aware of anything that could plausibly be called a geometric proof of the irrationality of . Given that the original definition of was geometric, it seems a pity that there is no such thing. Or maybe there is?

Perhaps one can try to draw pictures to accompany Lambert's irrationality proof. For example, is there a way to draw a picture of the following fact? And if so, is there any way to draw a picture of the fact that such a continued fraction is irrational when and are positive integers?

Keywords & Tags
nt.number-theory
continued-fractions
transcendental-number-theory
irrational-numbers
alternative-proof
Model Answers (9)
Submitted answers from various AI models
Unsolved
No reviews yet

K2-Think

Pending UQ Validation
Needs Human Verification
MBZUAI-IFM
Qwen2.5-32B
Sep 18, 2025, 01:50 PM

Creating a "picture proof" for Lambert's proof of π's irrationality, particularly focusing on his continued fraction for (\tan(x)), involves visually conveying the structure of the continued fraction and its implication for rationality. Here's a conceptual framework:

1. Visualizing Lambert's Continued Fraction for (\tan(x))

Lambert derived the continued fraction:
[ \tan(x) = \cfrac{x...

0 model reviews
0 human reviews
No reviews yet

GPT-5

Failed UQ Validation
Needs Human Verification
OpenAI
Sep 2, 2025, 06:56 AM

Two “picture-able” pieces lie behind Lambert’s proof:

  1. Where Lambert’s continued fraction for tan comes from

There are many ways to derive tan x = x / (1 − x^2/(3 − x^2/(5 − x^2/(7 − …)))) that admit pictures.

  • Riccati/orthogonal-polynomial/J-fraction picture. The function y(x) = tan x is the unique analytic solution of the Riccati differential equation y′ = 1 + y^2 with y(0) = 0. Solutions ...
0 model reviews
0 human reviews
No reviews yet

Claude Opus 4

Failed UQ Validation
Needs Human Verification
Anthropic
Aug 1, 2025, 12:00 AM

This is a fascinating question about visualizing Lambert's proof of π's irrationality. Let me explore what's possible here.

The Challenge

Lambert's proof relies on showing that tan(x) has a specific continued fraction expansion, and that this expansion is irrational for rational non-zero values of x. Since tan(π/4) = 1 is rational, π/4 (and hence π) must be irrational.

Visualizing the Con...

0 model reviews
0 human reviews
No reviews yet

o3 Pro

Failed UQ Validation
Needs Human Verification
OpenAI
Aug 1, 2025, 12:00 AM

Yes—Lambert’s argument can in fact be drawn.
The picture that goes with the proof already appears in his original memoir “Mémoire sur quelques propriétés remarquables des quantités transcendentes circulaires et logarithmiques’’ (Berlin Academy, 1761); a modern redrawing is reproduced, for instance, in

• H. Martini – H. J. Töplitz, “Lambert’s Lemma”, Elemente d. Math. 55 (2000) 19-26
• D...

0 model reviews
0 human reviews
No reviews yet

DeepSeek R1

Failed UQ Validation
Needs Human Verification
DeepSeek
deepseek-ai/DeepSeek-R1
Aug 1, 2025, 12:00 AM
<think> Okay, the user is asking about a geometric proof for the irrationality of π, specifically referencing Lambert's proof and continued fractions. They mention existing geometric proofs for irrationality of square roots and e, and wonder if something similar exists for π.

Hmm, this is a deep question. The user seems mathematically sophisticated—they're familiar with Lambert's proof and conti...

0 model reviews
0 human reviews
No reviews yet

Gemini 2.5 Pro

Failed UQ Validation
Needs Human Verification
Google
Aug 1, 2025, 12:00 AM

This is a fantastic question that gets to the heart of what makes proving π's irrationality so different from proving the irrationality of numbers like √2. While a full "proof without words" in the style of the √2 proof is likely impossible for π via Lambert's proof, we can explore how to visualize the key components of the argument, as you suggest.

Let's break down the two parts of your question...

0 model reviews
0 human reviews
No reviews yet

o3

Failed UQ Validation
Needs Human Verification
OpenAI
Aug 1, 2025, 12:00 AM

Below is, first, a faithful “cartoon’’ of Lambert’s argument that
tan x = x /(1 – x² /(3 – x² /(5 – …))) (1)
is a continued fraction, and second, an equally faithful “cartoon’’ of how the positivity of certain areas forces (1) to be irrational whenever x² is a non–zero rational number.
The drawings really are the whole argument; the little algebra that is left merely s...

0 model reviews
0 human reviews
No reviews yet

o4 mini

Failed UQ Validation
Needs Human Verification
OpenAI
Aug 1, 2025, 12:00 AM

Lambert’s original proof that π is irrational really boils down to two steps:

  1. One shows by elementary calculus that for any real x
  2. One observes that if (x=a/b) is a nonzero rational then the continued‐fraction on the right never terminates, hence...
0 model reviews
0 human reviews
No reviews yet

Claude 3.7 Sonnet

Failed UQ Validation
Needs Human Verification
Anthropic
Aug 1, 2025, 12:00 AM

Picture Proof of π's Irrationality: Lambert's Approach

You're asking about a visual or geometric representation of Lambert's proof that π is irrational, particularly focusing on the continued fraction expansion of tan(a/b).

To my knowledge, there isn't a widely recognized "picture proof" of Lambert's demonstration of π's irrationality. This is indeed a gap in mathematical visualization, especi...

0 model reviews
0 human reviews
Submit Your Model's Answer
Your submission will be validated by our LLM validator and then by human experts

You need to be logged in to submit an answer