Proteins are what you get when you string amino acids together, and we need a lot of them. No one really knows, but there may be as many as a million types of protein in the human body, and each one is a little miracle. By all the laws of probability proteins shouldn't exist. To make a protein you need to assemble amino acids (which I am obliged by long tradition to refer to here as "the building blocks of life") in a particular order, in much the same way that you assemble letters in a particular order to spell a word. The problem is that words in the amino acid alphabet are often exceedingly long. To spell collagen, the name of a common type of protein, you need to arrange eight letters in the right order. But to make collagen, you need to arrange 1,055 amino acids in precisely the right sequence. But—and here's an obvious but crucial point一you don't make it. It makes itself, spontaneously, without direction, and this is where the unlikelihoods come in.
你把氨基酸串在一起，就得到了蛋白质。我们需要大量的蛋白质。其实谁也不大清楚，但人体里的蛋白质也许多达100万种之多，每一种都是个小小的奇迹。按照任何概率法则，蛋白质不该存在。若要制造蛋白质，你得把氨基酸（按照悠久的传统，我在这里应当将其称之为“生命的砌块”）按照特定的顺序来排列，就像你拼写一个单词必须把字母按照特定的顺序来排列一样。问题是那些以氨基酸字母组成的单词往往长得不得了。若要拼出“胶原蛋白（collagen）”（一种普通蛋白质的名字）这个名字，你只需要以正确的顺序排列8个字母。若要制造胶原蛋白，你就得以绝对准确的顺序排列1055个氨基酸分子。但是——这是个明显而又关键的问题——你并不制造胶原蛋白。它会自发形成，无须你的指点，不可能性就从这里开始了。

The chances of a 1,055-sequence molecule like collagen spontaneously self-assembling are, frankly, nil. It just isn't going to happen. To grasp what a long shot its existence is, visualize a standard Las Vegas slot machine but broadened greatly—to about ninety feet, to be precise— to accommodate 1,055 spinning wheels instead of the usual three or four, and with twenty symbols on each wheel (one for each common amino acid).
坦率地说，1055个氨基酸分子要排列成一个胶原蛋白这样的分子的概率是零。这种事情完全不可能发生。为了理解它的存在是多么不可能，请你想象一台拉斯韦加斯普通的老虎机，不过要把它大大地扩大一下——说得确切一点，扩大到大27米——以便容纳得下1055个转轮，而不是通常的三四个，每个轮子上有20个符号（第一个代表一种普通氨基酸）