Cantor Diagonal Argument Proves Real Numbers Exceed Countable Sets

Numberphile

Georg Cantor's diagonal construction proves that real numbers cannot be arranged into an exhaustive list, establishing that uncountable infinity is strictly larger than the infinite set of integers and fractions.

AI-assisted summary of Infinity is bigger than you think - Numberphile — original source. Check the original for full context.

Source published: 2012-07-06

Key ideas

Infinity is not a concrete number but a mathematical concept of endlessness, within which distinct sizes and tiers of infinity exist. (0:00)
Infinite collections such as integers and rational numbers are mathematically classified as countable because their elements can be systematically ordered into a discrete list using methods like diagonal grid traversal. (0:58)
Cantor's diagonal argument demonstrates that the real numbers are uncountable by systematically constructing a missing decimal that differs from every entry in any putative exhaustive list. (3:40)
Georg Cantor's revolutionary formulation of set theory and varying infinities was met with intense hostility and dismissal from his contemporaries, contributing to severe depression and institutionalization before gaining universal acceptance. (6:54)

Detailed notes

## Background Published on YouTube by the channel Numberphile on July 6, 2012, this video features Dr. James Grime exploring the nature of infinite sets. As noted in the video description, Grime unpacks how some infinities are bigger than others, drawing on the groundbreaking mathematical foundations laid by Georg Cantor.

## TL;DR Dr. James Grime explains that infinity is not a single number but an endless concept, demonstrating that different sizes of infinity exist within mathematics. He illustrates how whole numbers, negative integers, and rational fractions are all 'countable' or listable, using Georg Cantor's diagonal traversal across an infinite grid to show that fractions can be systematically ordered. Moving to the continuous real numbers, Grime employs Cantor's diagonal argument to construct a decimal guaranteed to be missing from any supposedly complete list, proving the reals represent a strictly larger, uncountable infinity. The discussion concludes with the poignant history of Cantor, whose revolutionary ideas were initially rejected by contemporaries as nonsense before gaining universal acclaim.

## Atmosphere & style The presentation carries an intimate, conversational charm driven by pen-and-paper demonstrations and playful banter between Dr. James Grime and off-camera host Brady Haran. The tone moves effortlessly from accessible childhood counting intuitions into mind-bending mathematical rigor as Cantor's diagonal construction unfolds. A somber emotional arc emerges when Grime recounts Cantor's severe depression and tragic alienation by peers, before settling into a warm, celebratory appreciation for his enduring genius.

## Bulletin - [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) Infinity is not a concrete number but a mathematical concept of endlessness, within which distinct sizes and tiers of infinity exist. - [00:58](https://www.youtube.com/watch?v=elvOZm0d4H0&t=58s) Infinite collections such as integers and rational numbers are mathematically classified as countable because their elements can be systematically ordered into a discrete list using methods like diagonal grid traversal. - [03:40](https://www.youtube.com/watch?v=elvOZm0d4H0&t=220s) Cantor's diagonal argument demonstrates that the real numbers are uncountable by systematically constructing a missing decimal that differs from every entry in any putative exhaustive list. - [06:54](https://www.youtube.com/watch?v=elvOZm0d4H0&t=414s) Georg Cantor's revolutionary formulation of set theory and varying infinities was met with intense hostility and dismissal from his contemporaries, contributing to severe depression and institutionalization before gaining universal acceptance.

## Key takeaways - [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) Infinity is an abstract mathematical concept representing endlessness rather than a specific numerical value. - [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) Intuition about infinity often begins in childhood upon realizing that one can perpetually add one to any number without reaching a terminal limit. - [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) A fundamental discovery of modern mathematics is that some infinities are strictly larger than others. - [00:58](https://www.youtube.com/watch?v=elvOZm0d4H0&t=58s) The first level of infinity is known as countable infinity, which James Grime suggests is more intuitively described as 'listable.' - [00:58](https://www.youtube.com/watch?v=elvOZm0d4H0&t=58s) Despite including both positive and negative values alongside zero, the set of all integers has the exact same size of infinity as the natural numbers. - [00:58](https://www.youtube.com/watch?v=elvOZm0d4H0&t=58s) Attempting to list fractions row by row in a grid fails because the infinite sequence of denominators prevents ever moving beyond the first row. - [00:58](https://www.youtube.com/watch?v=elvOZm0d4H0&t=58s) Georg Cantor proved fractions are countable by weaving through a two-dimensional grid along finite diagonals, guaranteeing every rational number is eventually reached. - [03:40](https://www.youtube.com/watch?v=elvOZm0d4H0&t=220s) The continuous real number line includes fractions and irrational numbers such as pi and e, rendering it impossible to step sequentially from one number to the next. - [03:40](https://www.youtube.com/watch?v=elvOZm0d4H0&t=220s) Cantor's diagonal argument sets up a proof by contradiction by assuming an exhaustive list of all real decimals between 0 and 1 can be compiled. - [03:40](https://www.youtube.com/watch?v=elvOZm0d4H0&t=220s) Extracting the n-th digit of the n-th decimal and modifying each digit according to a consistent rule constructs a completely new decimal number. - [03:40](https://www.youtube.com/watch?v=elvOZm0d4H0&t=220s) The newly generated diagonal number is guaranteed to differ from every number on the list by at least one decimal place, proving the original list was incomplete. - [03:40](https://www.youtube.com/watch?v=elvOZm0d4H0&t=220s) Because real numbers cannot be fully listed, they represent an uncountable infinity that is strictly larger than the infinity of the counting numbers. - [06:54](https://www.youtube.com/watch?v=elvOZm0d4H0&t=414s) The diagonal contradiction cannot be resolved by continually appending the new numbers to the list, as the diagonal construction can instantly be rerun to find another missing value. - [06:54](https://www.youtube.com/watch?v=elvOZm0d4H0&t=414s) Cantor faced widespread academic condemnation and personal attacks from contemporary mathematicians, spending significant parts of his later years in mental institutions before his work was recognized as foundational.

## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) Breaking the Numberphile Rule: What Is Infinity?

Dr. James Grime begins by announcing that they are about to break one of the core rules of Numberphile by talking about something that is not actually a number. That subject is infinity. As Grime clarifies, infinity is not a number at all, but rather an idea and a concept representing endlessness and the property of going on forever. He notes that virtually everyone is introduced to the intuition behind infinity at an early age. A child of around five years old might count one, two, three, four, five, and wonder what the single biggest number in existence could be, guessing twenty. As they grow older, they might propose a million, before eventually realizing that the process never terminates because one can always add one more.

While the notion that numbers go on forever is widely understood, Grime reveals that the nature of infinity holds a far more startling mathematical truth: there is not merely one infinity. As James Grime puts it, “Some infinities are bigger than others.”

- Infinity is a mathematical concept of endlessness rather than a fixed numerical value. - The intuitive understanding of infinity stems from the fact that one can always add $1$ to any number without ever reaching a final limit. - Different sizes and varieties of infinity exist within mathematics.

[00:58](https://www.youtube.com/watch?v=elvOZm0d4H0&t=58s) Countable Infinities: Listing Integers and Fractions

Grime introduces the first category of infinity, known mathematically as "countable" infinity. He immediately expresses his distaste for the terminology, noticing camera operator Brady Haran give a skeptical reaction. Grime points out that calling an infinite set countable is counterintuitive, because by definition an endless collection cannot be fully counted. As James Grime explains, “I think it's a terrible name. I prefer to call it listable.” To demonstrate what listable means, Grime starts jotting down the natural numbers: 1, 2, 3, 4. Brady jokingly interrupts to ask if he plans to write down all of them, to which Grime asks how much time they have, and Brady laughs that they have ten minutes. Grime finishes writing 5 and 6, showing that the whole numbers can be systematically arranged into a discrete list.

Next, Grime considers the set of all integers, which includes zero and all negative numbers. One might arrange them as 0, 1, -1, 2, -2, 3, -3, and so on. Even though this set appears twice as large as the positive integers because it incorporates both signs, both sets share the exact same order of infinity because the integers can still be completely listed in sequence. Grime points out that fractions (the rational numbers) can also be listed, though doing so requires a more ingenious strategy. If one writes out fractions in an infinite two-dimensional rectangular grid—placing $1/1, 1/2, 1/3, 1/4 \dots$ across the first row, $2/1, 2/2, 2/3, 2/4 \dots$ across the second, and continuing downward—attempting to list them row by row fails completely. Moving endlessly through the denominators of the first row means one would never reach the second row.

To solve this, Grime demonstrates Georg Cantor's diagonal traversal method. Instead of reading across rows, one weaves across the diagonals of the grid: the first diagonal contains just $1/1$; the second covers $1/2$ and $2/1$; the third covers $1/3$, $2/2$, and $3/1$; followed by the fourth and fifth diagonals. Because every conceivable fraction occupies a definite row and column in the array, every single fraction is guaranteed to appear on one of these finite diagonals, proving that the fractions are listable and therefore have the exact same size of infinity as the counting numbers.

```mermaid flowchart LR accTitle: Cantor Diagonal Traversal accDescr: Sequence showing fractions ordered by successive diagonals to ensure every rational number is reached. subgraph D1["Diagonal 1"] N1["1/1"] end subgraph D2["Diagonal 2"] N2["1/2"] --> N3["2/1"] end subgraph D3["Diagonal 3"] N4["1/3"] --> N5["2/2"] --> N6["3/1"] end D1 --> D2 D2 --> D3 D3 --> DNext["..."] ```

- Natural numbers ($1, 2, 3, 4 \dots$) are countable because they can be enumerated in a sequence. - All integers ($0, 1, -1, 2, -2, 3, -3 \dots$) form the same size of infinity as the natural numbers despite appearing twice as large. - Fractions cannot be listed row by row because each row is infinite. - Traversing a grid of fractions along finite diagonals guarantees that every fraction eventually appears on the list.

[03:40](https://www.youtube.com/watch?v=elvOZm0d4H0&t=220s) The Real Numbers and Cantor's Diagonal Argument

Grime moves from fractions to the continuous real number line, drawing a line that includes zero, positive integers, fractions like $1/3$, and irrational constants such as $\pi$ and $e$. He asks how one could possibly list all the real numbers. If one starts at $0$ and jumps to $1$, $1/2$ has been missed; adding $1/2$ misses $1/4$; and between those lie decimals like $0.237\dots$. In fact, Grime asserts that the real numbers cannot be listed at all. As James Grime declares, “In fact, rather remarkably, I can show you that we can't list them, even though were talking about something so complicated as infinity.” Brady urges him on, joking that they will need an infinite amount of paper for the demonstration.

Grime sets up a proof by contradiction: suppose for a moment that every decimal between $0$ and $1$ could be written down in an exhaustive list. He invents an arbitrary sample list of four decimal expansions: 1. $0.121\dots$ 2. $0.221\dots$ 3. $0.31111129\dots$ 4. $0.00176\dots$

To prove this list can never be complete, Grime extracts the diagonal digits running down the list: the first digit of the first number ($1$), the second digit of the second number ($2$), the third digit of the third number ($1$), the fourth digit of the fourth number ($0$), and so forth, forming an initial number $0.12101\dots$. He then applies a simple transformation rule to build a brand new number: wherever a digit is a $1$, change it to a $2$; if it is a $2$ or any other digit, change it to a $1$. Following this rule, the diagonal sequence transforms into $0.21212\dots$.

Grime highlights that this newly constructed number cannot possibly exist anywhere on the supposedly complete list. It cannot be the first number because it differs in the first decimal place; it cannot be the second number because it differs in the second place; it differs from the third number in the third place, the fourth number in the fourth place, and so on for every entry down to infinity. Because a missing real number can always be constructed, the real numbers are unlistable and uncountable, proving the existence of a genuinely larger type of infinity.

- The real number continuum contains rational fractions and irrational numbers such as $\pi$ and $e$. - Cantor's diagonal argument assumes a complete list of decimals exists, then takes the $n$-th digit of the $n$-th number. - Applying a digit-flipping rule (e.g., $1 \to 2$, anything else $\to 1$) constructs a number guaranteed to differ from every number on the list by at least one decimal digit. - Uncountable infinity represents an order of magnitude strictly greater than countable infinity.

[06:54](https://www.youtube.com/watch?v=elvOZm0d4H0&t=414s) Georg Cantor's Legacy and Tragedy

Brady questions whether the diagonal paradox can be bypassed simply by adopting Grime's procedure as an iterative game: “Surely we could, James, because all we've got to do is keep doing your game and making them and adding them to the list. And if we keep doing that, won't we get there eventually?” Grime counters that the fix does not work, because as soon as that new number is appended to the list, one can immediately run the diagonal construction again to generate yet another number that remains omitted.

Grime explains that this groundbreaking proof was developed by the German mathematician Georg Cantor around the turn of the 20th century. At the time, Cantor's revolutionary ideas about multiple sizes of infinity were met with fierce hostility from his contemporaries. Fellow mathematicians dismissed his work as nonsense and labeled him a charlatan. The intense professional rejection took a heavy emotional and mental toll on Cantor, who spent much of his later life suffering from severe depression and entering and leaving mental institutions, where he ultimately died. Only near the end of his life did the mathematical community finally come to recognize the validity and brilliance of his discoveries. Brady and Grime conclude on a lighter note, with Brady observing that Cantor has now made it onto Numberphile, which Grime calls “the greatest accolade of all.”

- Appending the missing diagonal number back into the list fails to make the reals countable, because a new diagonal number can instantly be constructed from the updated list. - Georg Cantor formulated these foundational set theory concepts in the late 19th and early 20th centuries. - Cantor faced extreme hostility, professional alienation, and accusations of charlatanism from contemporary mathematicians. - Cantor spent substantial portions of his final years in mental institutions before his work received the universal recognition it holds today.

## Notable quotes > [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) Some infinities are bigger than others. — James Grime

## Mentioned Numberphile, James Grime, Georg Cantor, Brady Haran, Minute Physics

## Video description Sometimes infinity is even bigger than you think... Dr James Grime explains with a little help from Georg Cantor. More links & stuff in full description below ↓↓↓

Minute Physics video on this topic http://www.youtube.com/watch?v=A-QoutHCu4o (somewhat more fast-paced... but we did film ours BEFORE his was uploaded, so similarities are coincidental... well actually, no they are not... we are all building upon Cantor's work!!)

NUMBERPHILE Website: http://www.numberphile.com/ Numberphile on Facebook: http://www.facebook.com/numberphile Numberphile tweets: https://twitter.com/numberphile Subscribe: http://bit.ly/Numberphile_Sub

Videos by Brady Haran

Patreon: http://www.patreon.com/numberphile

Brady's videos subreddit: http://www.reddit.com/r/BradyHaran/

Brady's latest videos across all channels: http://www.bradyharanblog.com/

Numberphile T-Shirts: https://teespring.com/stores/numberphile Other merchandise: https://store.dftba.com/collections/numberphile

[Source](https://www.youtube.com/watch?v=elvOZm0d4H0)

中文简报

康托尔对角线论证证明实数集大于可数集格奥尔格·康托尔的对角线构造法证明了实数无法被穷举列出，从而确立了不可数无穷严格大于整数和分数的无限集合。詹姆斯·格赖姆博士解释道，无穷并非单一的数字，而是一个无止境的概念，并证明了数学中存在不同大小的无穷。他阐释了整数、负整数和有理分数如何都是“可数”或可列出的，利用格奥尔格·康托尔在无限网格上的对角线遍历法展示了分数可以被系统化地排序。而在谈及连续的实数时，格赖姆运用康托尔的对角线论证构造出一个必然在任何所谓完整列表中遗漏的小数，从而证明实数代表着一种严格更大的、不可数的无穷。讨论最后讲述了康托尔令人唏嘘的经历，他的革命性思想最初被同代人斥为无稽之谈并遭摒弃，最终才获得了普遍赞誉。无穷并非具体数字，而是一个无止境的数学概念，在其中存在着不同大小和层次的无穷。 (0:00) 像整数和有理数这样的无限集合在数学上被归类为可数集，因为它们的元素可以通过网格对角线遍历等方法被系统地排入一个离散列表中。 (0:58) 康托尔的对角线论证通过系统性地构造出一个与任何假设的穷举列表中每一项都不同的小数，证明了实数是不可数的。 (3:40) 格奥尔格·康托尔关于集合论和不同无穷阶的革命性构想曾遭到同代人的强烈敌意与排斥，这导致他患上严重抑郁并被送入精神病院，之后其理论才获得普遍认可。 (6:54)

Reading summary

Dr. James Grime explains that infinity is not a fixed numerical value but an endless concept, demonstrating Georg Cantor's proof that some infinities are strictly larger than others. The Concept of Listable Infinity (0:58) While infinity represents endlessness rather than a specific number, mathematicians classify different infinities by whether their elements can be arranged into a discrete sequence. Grime notes that although the mathematical term is 'countable,' he prefers the description 'listable.' The natural numbers (1, 2, 3, 4, and onward) serve as the baseline for this order of infinity. Surprisingly, expanding this set to all integers—including zero and negative numbers—does not produce a larger order of infinity. Even though the collection of integers appears twice as large as the positive integers, both sets share the exact same cardinality because integers can still be completely enumerated in an alternating pattern, such as 0, 1, -1, 2, -2, 3, -3. Mapping Rational Numbers via Diagonal Traversal (0:58) Fractions, or rational numbers, also belong to the countable order of infinity, but proving this requires overcoming the limitation of standard two-dimensional arrays. If fractions are organized into an infinite grid where rows share numerators and columns share denominators, proceeding row by row fails because the infinite sequence of denominators prevents one from ever reaching the second row. Georg Cantor resolved this by introducing a diagonal traversal method, weaving sequentially through finite diagonals: the first diagonal covers 1/1; the second covers 1/2 and 2/1; the third covers 1/3, 2/2, and 3/1; and subsequent diagonals continue outward. Because every fraction occupies a definite row and column, each is guaranteed to appear on one of these finite diagonals, proving the rationals are listable. Cantor's Diagonal Argument and Uncountable Reals (3:40) When examining the continuum of real numbers—which incorporates rationals and irrational constants like π and e—the listing method breaks down entirely. Cantor's diagonal argument demonstrates that the real numbers between 0 and 1 cannot be listed, establishing that this infinity is strictly larger than the countable set. If one assumes a complete list of decimals exists, a new number can be constructed by extracting the nth digit of each nth number in the sequence (for example, taking the diagonal digits 1, 2, 1, 0...) and applying a transformation rule, such as turning every 1 into a 2 and every other digit into a 1. The resulting decimal differs from the first number in the first decimal place, the second number in the second decimal place, and so on for every entry down to infinity, proving that an exhaustive list is impossible. Iterative Defense and Cantor's Historical Reception (6:54) Addressing the suggestion that omitted numbers could simply be appended back to the list in an iterative game, Grime points out that any updated list immediately permits the construction of another omitted number using the same diagonal technique. Despite the foundational importance of this discovery, Cantor's revolutionary ideas faced intense professional hostility from contemporaries around the turn of the 20th century. Critics dismissed his findings as nonsense and branded him a charlatan. The professional ostracization contributed to severe depression, leading to Cantor spending substantial portions of his later years in mental institutions before the mathematical community ultimately recognized the validity of his work. Cantor's proof establishes that infinite sets possess distinct, non-equivalent magnitudes. While the proof fundamentally restructured set theory, it met severe contemporary resistance, and full professional acceptance arrived only near the end of Cantor's life.

精炼阅读摘要

数学家 James Grime 在 Numberphile 节目中指出，无穷并非一个具体的数值，而是一个代表无止境的概念；更为惊人的是，数学上的无穷存在不同的“大小”维度，某些无穷在规模上严格大于另一些无穷。可数无穷与“可列表”的数学集合 (0:58) Grime 介绍，第一类无穷被称为“可数无穷”（Countable Infinity），但他更倾向于将其称作“可列表的”（Listable）。自然数集合（1, 2, 3, 4...）是这类无穷的典型代表。尽管全体整数集合（包含正整数、负整数和零）在直观上似乎是自然数的两倍大，但通过交替排列的方式（如 0, 1, -1, 2, -2...），我们同样可以将其排成一个无遗漏的单一序列。这证明了整数集合与自然数集合具有完全相同规模的无穷大。对于分数（有理数）集合，虽然无法直接按行或按列简单罗列，但通过格奥尔格·康托尔（Georg Cantor）设计的“对角线遍历法”，可以将所有分数排布在一个二维网格中，并沿着一条条有限长度的对角线进行蛇形穿梭。由于网格中的每一个分数都对应着确定的行和列，这种遍历方式确保了任何一个分数最终都会出现在这个一维列表中。因此，分数的无穷大尺寸也与自然数完全相同。实数连续统与康托尔对角线证明 (3:40) 与分数不同，实数轴上的数（包括有理数以及像 $\pi$ 和 $e$ 这样的无理数）是无法被完整列表呈现的。Grime 借助康托尔的对角线论证法（Diagonal Argument）对此进行了反证法证明。假设存在一个包含 0 到 1 之间所有实数的完整列表，我们可以通过提取该列表中第 $n$ 个数的第 $n$ 位小数，构建出一个全新的对角线数字序列（例如 $0.12101\dots$）。接着，应用一个简单的转换规则（例如：若某位数字是 1 则改为 2，若是其他数字则改为 1），便能生成一个全新的实数（如 $0.21212\dots$）。这个新生成的数不可能存在于原有的“完整”列表中，因为它与列表中的第一个数在第一位小数上不同，与第二个数在第二位小数上不同，以此类推。这一矛盾证明了实数是无法被穷尽列表的，从而确立了“不可数无穷”（Uncountable Infinity）的存在，其维度严格大于可数无穷。无法规避的无限循环与康托尔的学术悲剧 (6:54) 针对“能否通过不断将新生成的对角线数字添加回列表来最终完成列表”的疑问，Grime 解释称这种尝试注定失败。因为一旦将新数字追加到列表中，我们就可以立即针对更新后的列表重新运行对角线构建程序，从而产生又一个被遗漏的新数字。这是一个永远无法被填补的无限循环。这一颠覆性的理论由德国数学家格奥尔格·康托尔于 19 世纪末至 20 世纪初提出。然而在当时，康托尔关于“多种无穷大”的革命性思想遭到了同时代数学家的强烈抵制与敌意，甚至被斥为骗局。长期的学术孤立和同行排挤对康托尔的精神健康造成了毁灭性打击，导致他晚年饱受严重抑郁症的折磨，频繁出入精神病院，并最终在精神病院中去世。直到他生命接近尾声及去世后，数学界才终于普遍承认并确立了他这一发现的伟大价值。康托尔对无穷大分类的探索彻底改变了现代集合论与数学基础。尽管他的理论在早期引发了巨大的学术争议并导致了个人悲剧，但如今“不同大小的无穷”已成为无可争议的数学基石。

Detailed notes — zh

## Background 该视频由著名科普频道 Numberphile 于 2012 年 7 月 6 日在 YouTube 上发布，视频由布雷迪·哈兰（Brady Haran）拍摄制作。视频简介指出，詹姆斯·格赖姆（James Grime）博士将在格奥尔格·康托尔（Georg Cantor）数学成果的启发下，向观众解释为什么“有些无穷比其他无穷更加庞大”。 ## TL;DR 节目打破了 Numberphile 只讨论数字的惯例，深入探讨了“无穷”这一抽象概念，并阐述了无穷存在不同层级这一惊人事实。格赖姆博士首先解释了自然数、整数以及所有分数（有理数）如何通过交替排列和二维对角线折线法被排成清单，证明它们都属于同一级别的“可数无穷”。随后，他借助著名的康托尔对角线论证法，严谨证明了 0 到 1 之间的实数无法被完整列举，从而揭示了实数是更高维度的“不可数无穷”。最后，节目回顾了集合论奠基人康托尔当年因提出这些超前理念而饱受学界嘲讽、晚年在精神病院度过的悲剧一生及其终获认可的历程。 ## Atmosphere & style 整期节目散发着轻松诙谐却又引人深思的英式科普魅力，格赖姆手持马克笔在棕色牛皮纸上边画边讲，形式质朴且极富亲和力。格赖姆与掌镜人布雷迪之间自然的问答互动趣味横生，布雷迪代表观众提出的直觉疑惑与格赖姆抽丝剥茧的逻辑推演形成了绝妙的张力。在讲述康托尔因不被理解而凄凉晚景的段落中，氛围转入沉肃与感伤，但随即又以幽默的致敬化解，给观众带来一场既烧脑又充满人文温度的数学之旅。 ## Bulletin - [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) 无穷并非一个具体的数值，而是一个代表无限延续的概念，且数学上证明了存在不同大小的无穷。 - [02:08](https://www.youtube.com/watch?v=elvOZm0d4H0&t=128s) 自然数、整数以及所有正分数（有理数）属于同一层级的“可数无穷”，皆可通过特定顺序排成单向清单。 - [03:38](https://www.youtube.com/watch?v=elvOZm0d4H0&t=218s) 康托尔运用著名的对角线论证法证明了实数无法被列出，确立了实数集属于更大层级的“不可数无穷”。 - [06:54](https://www.youtube.com/watch?v=elvOZm0d4H0&t=414s) 数学家格奥尔格·康托尔提出超穷理论时曾遭同行猛烈抨击与孤立，晚年备受精神疾病折磨并在精神病院离世，其理论在晚期及身后才获得学界公认。 ## Key takeaways - [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) 无穷不是具体的数字，而是一个概念，核心特征在于无论数字多大都可以继续“加 1”。 - [00:35](https://www.youtube.com/watch?v=elvOZm0d4H0&t=35s) 詹姆斯·格赖姆指出“可数无穷”这个术语具有误导性，更准确的称谓应是“可列的”（Listable）。 - [01:25](https://www.youtube.com/watch?v=elvOZm0d4H0&t=85s) 整数虽然直观上看似是自然数的两倍，但通过交替列出正负数（0, 1, -1, 2, -2...），证明了二者属于同等大小的无穷。 - [02:08](https://www.youtube.com/watch?v=elvOZm0d4H0&t=128s) 所有正分数可以排列成无限二维阵列，但若采取逐行扫描的方式会永远卡在第一行而无法推进。 - [02:44](https://www.youtube.com/watch?v=elvOZm0d4H0&t=164s) 通过沿倾斜对角线折线遍历的方式，二维分数矩阵可以被成功拉平成一条单向清单，证明有理数是可数的。 - [03:38](https://www.youtube.com/watch?v=elvOZm0d4H0&t=218s) 包含无理数的连续实数轴无法确定相邻元素，在逻辑上根本无法按顺序一一列出。 - [04:25](https://www.youtube.com/watch?v=elvOZm0d4H0&t=265s) 康托尔对角线论证法假设存在一个完整的 0 到 1 之间小数清单，并提取主对角线上的数字构成序列。 - [05:32](https://www.youtube.com/watch?v=elvOZm0d4H0&t=332s) 通过对角线数字替换规则（如 1 变 2，其他变 1），必定能构造出一个与列表中每一项都至少在一位小数上不同的新实数。 - [06:54](https://www.youtube.com/watch?v=elvOZm0d4H0&t=414s) 即使将构造出的新数追加到列表中，重复该方法依旧会生成新的遗漏数字，彻底证明实数集不可数。 - [07:21](https://www.youtube.com/watch?v=elvOZm0d4H0&t=441s) 格奥尔格·康托尔因提出不同阶无穷的理论曾被同时代理论界斥为“江湖骗子”，晚年凄凉，直到生命末期才获得应有的学术认可。 ## Detailed walkthrough ### [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) 无穷的本质与“可数”无穷节目开场时，詹姆斯·格赖姆（James Grime）博士风趣地表示他们要打破 Numberphile 的一项规则——讨论一个根本不是数字的对象，那就是“无穷”（Infinity）。他解释道，无穷并不是一个具体的数，而是一个概念，代表着无边无际、永远延续。即便是小孩子也能直观地理解无穷：当你从 1、2、3 开始数数时，五岁的小朋友可能会觉得 20 是最大的数字，等长大一点又会觉得是一百万；但数字永远没有尽头，因为你总能不断地“加 1”。然而，数学中一个极其惊人的事实是：无穷并不只有一种，有些无穷要比其他无穷更为庞大。正如詹姆斯·格赖姆所断言的那样：“Some infinities are bigger than others.” 第一种无穷被称为“可数无穷”（Countable infinity）。格赖姆直言自己非常讨厌“可数”这个词，当他提到这个术语时，摄像机后的布雷迪·哈兰（Brady Haran）也发出了怀疑的咕哝声。格赖姆解释说，既然无穷是无止境的，人根本不可能把它“数完”，因此称其为可数是个糟糕的名字，他更愿意将其称为“可列的”（Listable）——即我们能否将这些数排成一个清单。格赖姆开始在纸上列举自然数 1、2、3、4，布雷迪在一旁开玩笑问道：“You're not gonna do all of them, are you James?” 格赖姆反问他们有多少时间，布雷迪笑着回答只有 10 分钟。随后格赖姆指出，不仅自然数可以排成列表，包含所有负数的整数集同样可以列出：从 0 开始，依次交替写下 1 和 -1、2 和 -2、3 和 -3。虽然直觉上整数看起来似乎是自然数的两倍之多，但它们其实是同一级别的无穷，因为它们都能够被完整地列出来。 - 无穷不是具体的数值，而是表示事物无止境、无限延续的概念。 - 格赖姆提议用“可列”（Listable）取代具有误导性的术语“可数”（Countable）。 - 自然数列表展示：$1, 2, 3, 4, 5, 6, \dots$ - 整数列表展示：$0, 1, -1, 2, -2, 3, -3, \dots$（证明整数与自然数具有相同大小的无穷势）。 ### [02:08](https://www.youtube.com/watch?v=elvOZm0d4H0&t=128s) 分数的可数性：康托尔对角线折线法比整数更令人惊讶的是，所有的分数（有理数）同样也是可以被完整排成列表的，但这需要运用一些巧妙的构造方法。格赖姆在纸上画出了一个二维的矩形阵列：第一行是分子为 1 的分数（$1/1, 1/2, 1/3, 1/4, 1/7, \dots$），第二行是分子为 2 的分数（$2/1, 2/2, 2/3, 2/4, \dots$），第三行则是分子为 3 的分数（$3/1, 3/2, \dots$），以此类推，构建出一个包含所有正分数的无限二维阵列。如果试图按照逐行读取的方式来将它们排成列表，就会遭遇致命的问题。因为第一行本身就有无穷多个数（$1, 1/2, 1/3, 1/5, 1/6, \dots$），沿着第一行一直数下去将永远无法结束，也就永远无法进入第二行。面对这种排列困境，格赖姆强调：“You can't list them that way.” 解决之道在于沿着对角线方向穿行：第一条对角线只有一个数，接着列出第二条对角线、第三条对角线、第四条对角线。通过这种倾斜的对角线折线扫描，格赖姆保证矩阵中的每一个分数最终都会出现在某一条对角线上，从而成功地将二维阵列拉平成一条能够一一列出的单向清单。 ```mermaid flowchart TD accTitle: 分数对角线排列法流程 accDescr: 展示如何通过二维网格的对角线扫描将所有分数排列成单向序列 subgraph Grid["无限分数阵列"] R1["第一行: 1/1, 1/2, 1/3, 1/4,..."] R2["第二行: 2/1, 2/2, 2/3, 2/4,..."] R3["第三行: 3/1, 3/2, 3/3, 3/4,..."] end subgraph Method["扫描机制"] Fail["逐行扫描: 永远卡在第一行 (不可行)"] Diag["对角线扫描: 沿倾斜方向遍历 (可行)"] end subgraph Result["有序输出列表"] List["第1条对角线 ➔ 第2条对角线 ➔ 第3条对角线... 覆盖全部有理数"] end Grid --> Fail Grid --> Diag Diag --> List ``` ### [03:38](https://www.youtube.com/watch?v=elvOZm0d4H0&t=218s) 实数不可数：康托尔对角线论证在展示了分数的可列性之后，格赖姆在纸上画出了一条连续的数轴，将讨论引向包含所有实数（连续的小数体系）的领域。数轴不仅包含 0、1、2、3 以及 $1/3$ 这样的分数，还包含了 $\pi$、$e$ 等所有无理数。如果尝试列出实数，问题便暴露无遗：从 0 开始之后该写哪个数？写 1 会漏掉 $1/2$；补上 $1/2$ 又漏掉了 $1/4$；补上 $1/4$ 又漏掉了 $0.237\dots$ 无论怎样尝试，都无法确定相邻的下一个数是什么。格赖姆随后借助严谨的逻辑证明指出：实数是根本无法被列出来的。为了证明这一点，格赖姆在纸上演示了著名的“康托尔对角线论证法”（Cantor's diagonal argument）。假设我们能够将 0 到 1 之间的所有小数排成一张完整的清单，列表可能形如下列形式： - 第 1 个数：$0.121\dots$ - 第 2 个数：$0.221\dots$ - 第 3 个数：$0.31111129\dots$ - 第 4 个数：$0.00176\dots$ 格赖姆沿着这条列表的对角线提取各个位置的数字——第一个数字的第 1 位小数、第二个数字的第 2 位小数、第三个数字的第 3 位小数，从而提取出一个数串（例如 $0.12101\dots$）。随后，他设定了一条简单的数字替换规则：如果对角线上的数字是 1，就将其变成 2；如果是 2 或其他任何数字，就将其变成 1。通过这种逐位修改，便构造出了一个崭新的小数。这个新数绝不可能存在于原始列表中：它与第 1 个数在第 1 位小数上不同，与第 2 个数在第 2 位小数上不同，与第 3 个数在第 3 位小数上不同，以此类推。正如格赖姆所总结的：“You've made a number that's not on that list.” 既然无论怎样假设一张完整清单，总能构造出一个不在其中的实数，这就证明了实数集是不可数的（Uncountable / Unlistable），代表了一种更大、更高维度的无穷。 - 假设列出的无穷小数矩阵中，对角线上的元素分别为主对角位置的数字。 - 构造规则：若第 $n$ 个数的第 $n$ 位小数为 1，则新数的第 $n$ 位取 2；否则取 1。 - 矛盾结论：新生成的实数与清单上第 $k$ 个数在第 $k$ 位必不相同，因此它不存在于清单中，原假设“实数可列”不成立。 ### [06:54](https://www.youtube.com/watch?v=elvOZm0d4H0&t=414s) 康托尔的悲剧与迟来的认可面对这一构造证明，布雷迪提出了一个非常直观的疑问：我们难道不能把这个新造出来的数追加到列表的最顶端或末尾，然后继续重复这个造数的游戏，一直加下去难道最终不会把所有数都填满吗？布雷迪质疑道：“And if we keep doing that, won't we get there eventually?” 格赖姆当即回应，只要你把新数加进列表，根据同样的对角线规则，立刻又能构建出一个不在最新列表中的全新数字，这意味着实数永远无法被穷尽。格赖姆介绍说，提出这一深刻洞见的是活跃于 19 世纪末至 20 世纪初的德国数学家格奥尔格·康托尔（Georg Cantor）。然而在当时，康托尔关于“存在不同大小的无穷”的划时代思想并没有受到数学界的赞誉，反而招致了同时代学者的激烈嘲讽与孤立。人们称他是“江湖骗子”（Charlatan），贬斥他的理论纯属无稽之谈。由于遭受同行极其恶劣的对待与沉重的精神打击，可怜的康托尔在晚年饱受精神疾病困扰，频繁进出精神病院，并最终在精神病院中凄凉离世。直到他生命垂危的晚期，他的数学成果才逐渐被学界验证并广泛接受，获得了他应有的学术荣誉。视频最后，布雷迪打趣说现在康托尔登上了 Numberphile，格赖姆随即笑着附和道：“And now he's on Numberphile, the greatest accolade of all.”以此向这位伟大的数学先驱致敬。 ## Notable quotes > [00:00](https://www.youtube.com/watch?v=elvOZm0d4H0&t=0s) Some infinities are bigger than others. — James Grime ## Mentioned Numberphile, James Grime, Brady Haran, Georg Cantor ## Video description Sometimes infinity is even bigger than you think... Dr James Grime explains with a little help from Georg Cantor. More links & stuff in full description below ↓↓↓ Minute Physics video on this topic http://www.youtube.com/watch?v=A-QoutHCu4o (somewhat more fast-paced... but we did film ours BEFORE his was uploaded, so similarities are coincidental... well actually, no they are not... we are all building upon Cantor's work!!) NUMBERPHILE Website: http://www.numberphile.com/ Numberphile on Facebook: http://www.facebook.com/numberphile Numberphile tweets: https://twitter.com/numberphile Subscribe: http://bit.ly/Numberphile_Sub Videos by Brady Haran Patreon: http://www.patreon.com/numberphile Brady's videos subreddit: http://www.reddit.com/r/BradyHaran/ Brady's latest videos across all channels: http://www.bradyharanblog.com/ Sign up for (occasional) emails: http://eepurl.com/YdjL9 Numberphile T-Shirts: https://teespring.com/stores/numberphile Other merchandise: https://store.dftba.com/collections/numberphile [Source](https://www.youtube.com/watch?v=elvOZm0d4H0)

Source quotes

[0:00] James Grime: Some infinities are bigger than others.

Source quotes

[0:00] James Grime: Some infinities are bigger than others.

Original transcript

[0:00] JAMES GRIME: We're going to break a rule. [0:01] We're break one of the rules of Numberphile. [0:03] We're talking about something that isn't a number. [0:06] We're going to talk about infinity. [0:08] So infinity. [0:11] Now like I said, infinity is not a number. [0:16] It's a idea. [0:17] It's a concept. [0:18] It's the idea of being endless, of going on forever. [0:22] I think everyone's familiar with the idea of [0:23] infinity, even kids. [0:25] You start counting 1, 2, 3, 4, 5-- [0:29] you might be five years old, but already you're thinking, [0:32] what's the biggest number I can think of. [0:35] And you go, oooh, it's 20. [0:37] You get a bit older, and you go, maybe it's a million. [0:39] It never ends, does it? 'Cause you can keep adding 1. [0:42] So that's the idea of infinity. [0:44] The numbers go on forever. [0:47] But I'm going to tell you one of the more surprising facts [0:50] about infinity. [0:51] There are different kinds of infinity. [0:54] Some infinities are bigger than others. [0:57] Let's have a look. [0:58] The first type of infinity is called countable. [1:01] And I don't like the name countable. [1:03] And Brady gave me a little bit of a hmm, just then. [1:05] Because if you're talking about infinity, you can't [1:09] count infinity, can you? [1:10] Because it goes on forever. [1:12] I think it's a terrible name. [1:15] I prefer to call it listable. [1:17] Can we list these numbers? [1:19] All right. [1:19] Let's do these simple numbers, 1, 2, 3-- [1:23] BRADY HARAN: You're not gonna do all of them, are you James? [1:25] JAMES GRIME: 4. [1:25] How long have we got? [1:27] BRADY HARAN: (LAUGHING) 10 minutes. [1:28] JAMES GRIME: Right. [1:28] 5, 6-- [1:31] so you can list the whole numbers. [1:33] So this is called countable. [1:35] Listable, I prefer. [1:37] What about the integers? [1:39] All the integers. [1:40] That's all the negative numbers as well. [1:41] So there's 0. [1:42] Let's have that. [1:43] But there's 1 and minus 1, there's 2 and minus 2, there's [1:49] 3, and minus 3. [1:51] Now, that is an infinity as well. [1:54] And in some sense, it's twice as big, because there seems to [1:58] be twice as many numbers. [2:00] But it is infinity as well. [2:02] They're both infinity, and they're both the [2:04] same type of infinity. [2:05] They both can be listed. [2:08] Perhaps more surprisingly, the fractions can [2:10] be listed as well. [2:12] But you have to be a bit clever about this. [2:13] Let's try and list the fractions. [2:15] I'm going to write out a rectangle. [2:17] 1 divided by 1. [2:19] That's a fraction. [2:19] [INAUDIBLE]. [2:20] Let's have 1 divided by 2, 1/3, 1/4, 1/7-- [2:26] OK, that goes on. [2:27] Let's do the next row and have two at the top. [2:29] 2/1, 2/2, 2/3, 2/4. [2:34] Let's do the next one. [2:35] 3/1, 3/2. [2:38] 4/6, 4/7. [2:40] That goes on and we can keep going. [2:42] So here, I've made some sort of an infinite rectangle array [2:46] of fractions. [2:46] Now if I want to make it a list like this, though, If I [2:52] went row by row, you're going to have a problem. [2:54] If you go row by row, I'll go-- [2:57] there's 1, 1/2, 1/3, 1/5, 1/6, 1/7-- and [3:00] I'll keep going forever. [3:02] And I'm never going to reach the second row. [3:04] I can't list them. [3:05] Not that way. [3:06] You can't list them that way. [3:07] You'll never reach the second row. [3:08] This is how you list them. [3:09] Slightly more clever than that. [3:12] You take the diagonal lines. [3:16] Now, I can guarantee that every fraction will appear on [3:21] one of those diagonal lines. [3:23] And you list them diagonal by diagonal. [3:25] So that's the first diagonal. [3:27] Then you list the second diagonal-- there it is. [3:29] Then you list the third diagonal, then you take the [3:31] fourth diagonal, and the fifth. [3:33] So eventually, you are going to do this every fraction. [3:36] Every faction appears on a diagonal, and you're [3:38] going to list them. [3:38] Now, if you take all the numbers, right? [3:41] That's the whole number line. [3:42] Let's try that. [3:43] Look, I'm going to draw it. [3:44] It's a continuous line of numbers. [3:46] These are all your decimals. [3:48] You've got 0 there in the middle, and you'll [3:50] go 1 and 2 and 3. [3:52] But it has a 1/3. [3:53] It will contain pi, and e, and all the [3:57] irrational numbers as well. [3:59] Can you list them? [4:00] How do you list them? [4:01] 0 to start with, and then 1? [4:03] But hang on. [4:04] We've missed a half. [4:05] So we put in the half. [4:06] Hang on, we've missed the quarter. [4:07] We put in the quarter. [4:09] But we've missed 0.237-- [4:12] so how do you list the real numbers? [4:15] It turns out you can't. [4:17] In fact, rather remarkably, I can show you that we can't [4:20] list them, even though were talking about something so [4:23] complicated as infinity. [4:25] BRADY HARAN: Do it, man! [4:26] JAMES GRIME: We need paper. [4:27] BRADY HARAN: We need an infinite amount of [4:28] paper here, I think. [4:30] JAMES GRIME: (LAUGHING) It's a big topic. [4:31] Imagine we could list all the decimals, right? [4:36] We can't, actually. [4:37] But pretend we can. [4:38] What sort of-- [4:39] what would it look like? [4:41] We'll start with all the 0-point decimals. [4:44] Let's pick some decimals. [4:45] 0.121-- [4:47] dot dot dot dot dot. [4:48] Let's pick the next one. [4:49] Let's say the next one is 0.221--. [4:53] Next one, let's do 0.31111129--. [5:00] And let's take another one, here. [5:02] 0.00176--. [5:06] Now I'm going to make a number. [5:09] This is the number I'm going to make. [5:11] I'm going to take the diagonals here. [5:13] I'm going to take this number and this number and this [5:16] number and this number and this number. [5:20] And I am going to write that down. [5:24] So what's that number I've made? [5:25] It's 0.12101-- [5:30] something, something, something. [5:32] Now this is my rule. [5:35] I'm going to make a whole new number from that one. [5:38] This is the number I'm going to make. [5:39] If it has a 1, I'm going to change it to a 2. [5:44] And if it has a 2 or anything else, I will change it to a 1. [5:52] So let's try that. [5:53] So I'm going to turn this into-- [5:55] 0-point. [5:56] So if it has a 1, I'm going to turn it into a 2. [5:59] If it's anything else, I'm going to turn it into a 1. [6:01] So that will be a 1. [6:03] I'm going to change 1 here into a 2. [6:06] I'm going to change that one into a 1. [6:07] I'm going to change that one into a 2-- that was my rule. [6:10] And I'll make something new. [6:12] That does not appear on the list. [6:15] That number is completely different from anything else [6:18] on the list, because it's not the first number, because it's [6:22] different in the first place. [6:24] It's not the second number, because it's different in [6:26] second place. [6:27] It's not the third number, because it's different in the [6:30] third place. [6:31] It's not the fourth number because it's different in the [6:33] fourth place. [6:35] It's not the fifth number, because it's different in the [6:37] fifth place. [6:38] You've made a number that's not on that list. [6:40] And so you can't list all the decimals, in which case it is [6:45] uncountable. [6:46] It is unlistable. [6:47] And that means it's a whole new type of infinity. [6:51] A bigger type of infinity. [6:54] BRADY HARAN: Surely we could, James, because all we've got [6:54] to do is keep doing your game and making them and adding [6:58] them to the list. [6:58] And if we keep doing that, won't we get there eventually? [7:01] JAMES GRIME: But you could then create another number [7:03] that won't be on that list. [7:04] And so the guy who came up with is a German mathematician [7:06] called Cantor. [7:08] Cantor lived 'round about the turn of the 20th century. [7:11] He was ridiculed for this. [7:14] For this idea that there were different types of infinity, [7:17] he was called a charlatan. [7:19] And he was called-- it was nonsense, it was called. [7:22] And poor old Cantor was treated really badly by his [7:27] contemporaries, and he spent a lot of his later life in and [7:32] out of mental institutions, where he died, in the end. [7:37] Near the end of his life, it was recognized. [7:39] It was true. [7:40] It was recognized. [7:41] And he had all the recognition that he deserved. [7:45] BRADY HARAN: And now he's on Numberphile. [7:46] JAMES GRIME: And now he's on Numberphile, the greatest [7:47] accolade of all. [7:49] Georg Cantor.