How Do Words Get Their Meaning? 字的意思從哪裡來？

🔊The picture fits “dog” and “table.” It fits less well when we try other words. Nothing can be pointed at for “five,” “because,” “perhaps,” or “hello.” And pointing itself has to be understood. If I point at a rabbit running past and say “gavagai,” you cannot tell from the gesture whether I am naming the animal, its color, its running, the number one, or the act of pointing. A child who learns from pointing must already know a good deal about what kind of thing is likely to be meant.

這個圖像很適合「狗」與「桌子」。換別的字試試看，它就沒那麼適合了。「五」、「因為」、「也許」、「哈囉」，都沒有東西可以指。而且「指」本身也需要被理解。如果我指著一隻跑過去的兔子說「gavagai」，你無法從這個手勢看出：我是在叫那隻動物、牠的顏色、牠的奔跑、數字「一」，還是「指」這個動作的名字。一個從別人的指點裡學習的孩子，必定已經相當知道，對方多半是指哪一類的東西。

🔊John Locke moved meaning indoors. “Words, in their primary or immediate signification, stand for nothing but the ideas in the mind of him that uses them.” I have an idea of gold, I attach the sound “gold” to it, and when I speak, the sound calls up the matching idea in you. The link between sound and idea is “perfectly arbitrary,” which is why there are many languages. This explains how words for absent or imaginary things can have meaning. It leaves a difficulty that Locke saw. I can never look into your mind to check that your idea matches mine. If meaning is a private item, no one can know that anyone else means what he means.

約翰・洛克把意義搬到了室內。「字詞，就其原初或直接的意指而言，代表的不是別的，就是使用它的人心裡的觀念。」我有一個「金子」的觀念；我把「金子」這個聲音繫在它上面；當我說話的時候，那個聲音在你心裡喚起相配的觀念。聲音與觀念之間的連結是「完全任意的」；這就是為什麼有許多種語言。這解釋了：指稱不在場或想像中的事物的字詞，怎麼能有意義。它也留下了一個洛克自己看到的困難。我永遠無法看進你的心裡，去核對你的觀念跟我的相不相配。如果意義是一件私人的東西，就沒有人能知道別人的意思跟自己的一樣。

🔊In 1892 the logician Gottlob Frege showed that meaning cannot simply be the thing named either. The morning star and the evening star are one object, the planet Venus. Yet “the morning star is the evening star” was a discovery of astronomy, while “the morning star is the morning star” tells us nothing. Two names for one thing can differ in meaning. Frege concluded that a name has a sense, a way in which it presents its object, as well as a reference. Sense, he insisted, is public. It is what a whole community grasps, and no one keeps it in his head as a private image.

1892 年，邏輯學家戈特洛布・弗雷格指出：意義也不可能單單就是被叫到名字的那樣東西。晨星與暮星是同一個對象，也就是金星。然而「晨星就是暮星」是天文學的一項發現；而「晨星就是晨星」什麼也沒告訴我們。同一樣東西的兩個名字，意義可以不同。弗雷格的結論是：一個名字除了有「指稱」，還有「意義」，也就是它呈現其對象的方式。他堅持，意義是公共的。它是一整個社群所掌握的東西；並沒有哪一個人把它當作私人的意象，收在自己的腦袋裡。

🔊Ludwig Wittgenstein began his last book by quoting Augustine’s account of learning to speak, and spent the book taking it apart. He asked his readers to stop looking for the thing a word stands for and to look at what people do with it. “For a large class of cases,” he wrote, “the meaning of a word is its use in the language.” Words are like the tools in a toolbox, a hammer, a ruler, a pot of glue, and are as various in their work. To know what “five” means is to be able to count, and to know what “hello” means is to know when to say it.

路德維希・維根斯坦在他最後一本書的開頭，引了奧古斯丁對學說話的那段描述；然後用整本書把它拆開。他請讀者不要再去找一個字「代表」的東西，而去看人們拿它做什麼。他寫道：「在很大一類的情況裡，一個字的意義就是它在語言中的用法。」字詞就像工具箱裡的工具：一把鎚子、一把尺、一罐膠水；它們的工作也同樣五花八門。知道「五」是什麼意思，就是會數數；知道「哈囉」是什麼意思，就是知道什麼時候說它。

🔊Against Locke he offered a small story. Suppose everyone has a box with something in it, which we call a beetle. No one can look into anyone else’s box, and each says that he knows what a beetle is only from his own. It would be quite possible for everyone to have something different in his box, or for the thing to keep changing, or for a box to be empty. The word “beetle” would go on being used all the same. The thing in the box, Wittgenstein says, “has no place in the language-game at all.” It drops out as irrelevant. If the meaning of “pain” were a private object that only I can inspect, the same would follow for pain. Since we plainly do talk about pain and understand one another, the word must get its meaning elsewhere, from the cries, the winces, the comforting, and the shared life in which it is used.

針對洛克，他講了一個小故事。假設每個人都有一個盒子，裡面有某樣東西，我們叫它「甲蟲」。誰也不能看進別人的盒子；而每個人都說，他只從自己的那一隻知道甲蟲是什麼。很有可能每個人的盒子裡是不同的東西；或者那樣東西不停地在變；或者有一個盒子是空的。「甲蟲」這個詞還是會照樣被使用。維根斯坦說，盒子裡的那樣東西「在這個語言遊戲裡根本沒有位置」。它作為不相干的東西，被消掉了。如果「痛」的意義是一個只有我能檢視的私人對象，那麼對痛也會得出同樣的結論。既然我們顯然確實在談論痛、也彼此聽得懂，這個詞的意義必定來自別處：來自叫喊、皺眉、安慰，以及使用它的那個共同的生活。

🔊Chinese philosophers had argued about names for two centuries before the first empire. The logician Gongsun Long defended the claim that “a white horse is not a horse,” on the ground that “horse” names a shape and “white” a color, so that the two terms do not mean the same. Thinkers who wanted language to serve government found such puzzles dangerous. Xunzi replied with the clearest statement of conventionalism in the ancient world: “Names have no fixed rightness. They are fixed by agreement. When the agreement is settled and the custom formed, they are called right, and what departs from the agreement is called wrong.” A name is correct because a community uses it so. For Xunzi it follows that keeping names stable is a public duty, and that those who play with them unsettle everything built on them.

在第一個帝國之前，中國的哲學家已經為「名」爭論了兩個世紀。名家的公孫龍為「白馬非馬」這個主張辯護；理由是「馬」命名的是形，「白」命名的是色，所以這兩個詞的意思並不相同。希望語言為治理服務的思想家，覺得這樣的難題很危險。荀子的回應，是古代世界對約定論最清楚的陳述：「名無固宜，約之以命，約定俗成謂之宜，異於約則謂之不宜。」一個名之所以正確，是因為一個社群這樣用它。對荀子來說，由此可以推出：保持名的穩定是一項公共的責任；而玩弄名的人，動搖了建立在它們之上的一切。

🔊The Daoists drew the opposite lesson from the same fact. If names are our arrangement and nothing more, they cannot reach what was there before any arrangement. The Daodejing opens: “The name that can be named is not the constant name.” Zhuangzi put it as an image. “The fish trap exists for the sake of the fish. Once you have the fish, you can forget the trap. Words exist for the sake of meaning. Once you have the meaning, you can forget the words. Where can I find a man who has forgotten words, so that I can have a word with him?”

道家從同一個事實裡，引出了相反的教訓。如果名只是我們的安排、別無其他，它們就搆不到任何安排之前就已經在那裡的東西。《道德經》開頭就說：「名可名，非常名。」莊子把它說成一個意象：「荃者所以在魚，得魚而忘荃……言者所以在意，得意而忘言。吾安得夫忘言之人而與之言哉？」

🔊“Meaning is use” is closer to a direction for looking than to a theory, and it has costs. If a word means whatever a community does with it, it becomes hard to say how a whole community could use a word wrongly, and yet people once applied “fish” to whales. In the 1970s Saul Kripke and Hilary Putnam argued that some words are tied to the world in a way that use does not capture. “Water” named a certain stuff long before anyone knew its chemistry, and what the word meant all along depended on what that stuff really is. Meanings, as Putnam put it, are not entirely in the head, and on his account they are not entirely in our practices either. The present view is a division of labor. Use and agreement explain most words, and the world itself helps to fix the meaning of some. Almost no one now defends the private image in the mind that Locke began with.

「意義即用法」與其說是一個理論，不如說是一個「往哪裡看」的指示；而它是有代價的。如果一個字的意思，就是一個社群拿它做的任何事；那就很難說明，一整個社群怎麼可能把一個字用錯。然而人們確實曾經把鯨魚叫做「魚」。1970 年代，索爾・克里普克與希拉蕊・普特南主張：有些字與世界連結的方式，是「用法」捕捉不到的。早在任何人懂得它的化學成分之前，「水」就已經是某一種東西的名字了；而這個字一直以來的意思，取決於那種東西實際上是什麼。照普特南的說法，意義並不完全在腦袋裡；而照他的講法，它們也不完全在我們的實踐裡。目前的看法是一種分工。用法與約定解釋了大多數的字；而世界本身，幫忙固定了其中一些字的意義。洛克當初拿來當起點的那個「心裡的私人意象」，現在幾乎沒有人為它辯護了。