《马氏文通》
The Mashi Wentong

许慎解释的是一个一个的字。汉字为什么这么写，他讲得很清楚；可是句子为什么这么说，他没有讲。中国人写了两千多年的文章，却一直没有一本书专门讲：词和词是怎么放在一起的。一直等到一八九八年，一位叫马建忠的学者写出了《马氏文通》。他在法国留过学，看见法语、拉丁语都有厚厚的语法书，就问了一句：汉语为什么没有？于是他读了大量古书，从里面找例子、比较例子，第一次系统地描写了汉语的句子。

What Xu Shen explained was characters, one at a time. Why a character is written as it is, he made very clear; why a sentence is said as it is, he never touched. Chinese people wrote for two thousand years without a single book devoted to how words are put together. Not until 1898, when a scholar named Ma Jianzhong published the 马氏文通. He had studied in France, he had seen the thick grammars that French and Latin possessed, and he asked one question: why does Chinese not have one? So he read enormous quantities of older texts, pulled examples out of them, compared them, and described Chinese sentences systematically for the first time.

他做的第一件事，是给词分类。他先把词分成两大类：一类自己有意思，可以单独回答问题；另一类自己没什么意思，只帮别的词做事，比如“的”“吗”“了”。第一类他又分成五小类，第二类分成四小类，一共九类。每一类叫什么、怎么认，第二部分会一类一类地讲。这里你只要记住那个想法：词是可以分类的，分类的根据是它能站在什么位置、能跟什么词搭在一起。

The first thing he did was sort words into classes. Two big groups to begin with: words that carry meaning on their own and can answer a question by themselves, and words that carry almost no meaning and only do jobs for other words — 的, 吗, 了. He split the first group into five smaller classes and the second into four, nine in all. What each class is called and how to recognise it is Arc B's work. Here you only need the idea: words fall into classes, and what decides the class is where a word can stand and what it can combine with.

有一个小地方很有意思：马建忠说的不是“词”，是“字”——他给每一类起的名字都带一个“字”。那时候还没有人把“字”和“词”分得那么清楚。后来的学者改了过来：我们在这几课里说的“词”，是一块有意思的语言单位，可以是一个字，也可以是两个字。就是这个小小的改动，开始了今天的现代汉语语法。

One small detail is worth noticing. Ma Jianzhong did not talk about 词; he talked about 字, and every class name he coined ended in 字. In 1898 nobody yet drew a firm line between the written syllable and the unit of meaning. Later scholars redrew it, and 词 — a block of meaning that may be one character or two — is what you have been calling a word since lesson two of this level. That tiny change of term is where modern Chinese grammar starts.

《马氏文通》也有毛病。它讲的其实是古书的语言，例子都来自两千年前的文章，不是当时人嘴里说的话；它用的框架又是从拉丁语语法搬过来的，有些地方套在汉语身上并不合适。后来的学者一边用它，一边改它：吕叔湘、朱德熙这些人，用汉语自己的事实重新描写了汉语。所以今天我们说“现代汉语语法”，说的是一百多年的研究，不是一个人的一本书。

The book has real faults. What it describes is the language of old texts, not the speech of 1898 — every example comes from writing two thousand years old. And its framework was carried over from Latin grammar, so parts of it fit Chinese badly. Later scholars used it and corrected it at the same time; people like Lü Shuxiang and Zhu Dexi redescribed Chinese from Chinese facts. So 现代汉语语法, modern Chinese grammar, means a century of research rather than one man's book — and Arc B is that research, arranged for you.

把这两本书放在一起看一分钟。一本讲字，一本讲句子；一本在东汉，一本在清朝末年；中间隔了一千八百年。可是两个人做的事完全一样：收集大量的例子，比较，然后说出它们共同的道理。下一课，你要用这个办法看你自己会说的五个句子。

Hold the two books side by side for a minute. One is about characters, the other about sentences; one belongs to the Eastern Han, the other to the last years of the Qing; eighteen centuries separate them. Yet the two men did the identical thing: gather a great many examples, compare them, and say out loud what they have in common. In the next lesson you do it to five sentences of your own.