Oxford comma (牛津逗号)

牛津逗号（Oxford comma），又称连续逗号（Serial comma），是指如果并列三个或更多的名词，在倒数第二个名词之后添加逗号。例如
France, Italy, and Spain
假如省略这个逗号，就写成
France, Italy and Spain
两种写法都不错，但牛津逗号现在越来越流行，尤其是在美国。《牛津书写风格手册》曾建议使用，因此得名。

为什么牛津逗号值得推荐？因为英语里有同位语短语或词组，它与前面的名词也是用逗号与它隔开。同位语词组由两部分组成，中间用一个逗号。有三个或更多的名词时使用牛津逗号有助于将它与同位语词组区分开来。例如
（1）We invited the rhinoceri, Washington, and Lincoln.（“我们邀请了犀牛、华盛顿和林肯。”）
如果不用牛津逗号
（2）We invited the rhinoceri, Washington and Lincoln.（“我们邀请了犀牛，即华盛顿和林肯”，或“我们邀请了犀牛、华盛顿和林肯。”）
语句（2）有两种解读，如果读者习惯了牛津逗号的书写并且以为作者在他的写作中也遵守这个规则，读者就会理解为“我们邀请了犀牛，即华盛顿和林肯”，就是说将Washington and Lincoln（“华盛顿和林肯”）看作是复数的rhinoceri（“犀牛”）的同位词组而对它具体列举说明（好比I have two sons, John and Jack中的John and Jack）。这种理解并非完全无理，因为有可能犀牛饲养员或动物园的确有两只犀牛，还给它们分别取名为“华盛顿”和“林肯”。

但牛津逗号并不解决所有问题，例如
They went to Oregon with Betty, a maid, and a cook.
无论如何这句话都有两种理解：
“他们和贝蒂、一个女佣和一个厨师一起去了俄勒冈州。”
“他们和贝蒂（一个女佣）和一个厨师一起去了俄勒冈州。”
为避免这种语义模糊性，语句必须改写。一个好的作者或作家必须随时提醒自己，写出的语句是否会被误解。当然，故意造成歧义而产生一种特殊的修辞效果则另当别论。

Shall or will?

When I learned English thirty plus years ago in China, we were taught to say "I shall", "we shall", and "you will", "he/she/it will". That is, "shall" for the first-person verb and "will" for the second- and third-person. But coming to the US, I realize "shall" is rarely said. It is said if the speaker intends to emphasize his (her) point and in that case it's not limited to the first person. I just came across the renowned English language linguist David Crystal's The Cambridge Encyclopedia of the English Language where (on p.224) he says "[m]odern usage does not observe this distinction. Indeed, it may never have existed in the language, but only in the minds of grammarians anxious to impose order on a 'messy' area of usage". Wasn't that enlightening, especially the latter statement! I don't know if the English textbooks in China still recommend that usage. It misled a generation of English learners back then. One coworker of mine who came to the US from southern China in the 1970s still says "I / we shall" today, though this is a fellow that couldn't care less about English, judging by his frequent spelling errors and strong accent in spite of his 40+ years living in the US. Fortunately, no other rules made up by the prescriptive grammarians came into the textbooks we were using, such as the no split infinitive rule (i.e. avoid saying "to fully understand"). Apparently the English educators in China did keep an eye on the actual usage of the language in English-speaking countries.

Phonetic symbols /i/ and /u/ in some dictionaries mislead English learners

Many English dictionaries published in China use /i/ and /u/ symbols to represent the sounds /ɪ/ and /ʊ/ respectively. As a result, some learners mistakenly think the vowel in, for example, bit is just a short version of that in beat.

复旦大学葛传椝主编的《新英汉词典》于1975年首次出版，在随后的拨乱反正、科学的春天和改革开放初期发挥了不可估量的作用。随后多次的修订版增删了许多词汇，并对词汇例句做了很大的改变。但词典在词汇注音上没有变化，它采用的注音方式不经意地对学习者造成了一些负面影响。

《词典》声明“用国际音标注音，采用宽式注音法。音标注在本词后，放在方括号内”。所谓宽（broad）、窄（narrow）式注音并无严格的定义，两者分别相当于说宽泛、严格的注音。[1]但《词典》至少有两个元音的标注是误导的，/i/和/u/。

在《词典》中，/i/是比如bit（“一点”；“比特”）的元音，而/i：/是比如beat（“打”）的元音。[2]从符号看，后者似乎只是前者发音的延长。但实际上，bit中元音的发音部位也与beat的元音不同，正确的音标符号是/ɪ/，口腔中比/i：/的发音部位略低、略靠后、嘴形略大，介于/i/与/e/之间。从维基“次闭前不圆唇元音”网页看，上海话“一”即发此音。（会上海话的网友可帮助验证！）

另外，《词典》标记比如book（“书”）的发音为/buk/，boot为/bu：t/，使学习者以为后者中的元音/u：/与前者元音/u/的唯一区别是音长。但事实上，book的正确发音是/bʊk/，其中/ʊ/比/u/发音部位略低、略靠前、嘴形略大。

如果你读book为/buk/，就会给人留下有外国口音的印象。如果读bit为/bit/，在缺乏上下文的情况下还是会被人以为在说beat。[3]只有借助上下文听者才能听懂你说的是bit，但会感觉你有口音。

用/i/代替/ɪ/、/u/代替/ʊ/注音似乎是许多在中国出版的英语词典的惯例。由于这种惯例可能影响学习者对/ɪ/和/ʊ/发音的掌握，词典编撰者有必要纠正。

注：
[1]注意，所谓宽、窄这种分法并不对应音素、语音转录（phonemic / phonetic transcription）注音的区分，后者有严格的定义。
[2]国际音标中有专门的符号表示长元音，这里用冒号代替以便更好在网页中显示。另外这里用斜杠而不用方括弧，但斜杠在此并不代表音素转录。
[3]在英、美英语中，没有唯一通过音长以区别两个词的例子。但在澳洲英语中存在，如full和fool、pull和pool完全靠元音长短区分，见维基Australian English phonology条。

Study time: vocabulary vs. grammar

Steve Kaufmann is an undisputed celebrity in the language study community. According to Wikipedia, "[a]s of 2023, he has an understanding of 20 languages, to varying degrees". In his blog, he says "Vocabulary is much more important than grammar. The grammar you acquire gradually as you become familiar with the language, with the words, but first of all you need words." This caused much debate in the Facebook "Polyglots (The Community)" group. Most comments disagree with him. To find out whether this disagreement is genuine, I started a poll in the same group.

"For all the languages you're studying, given 10 hours dedicated to vocabulary and grammar, what is the average ratio of time of your study in these two areas? It's true that oftentimes there is overlap. What is polled here is a subjective one. So just give a rough estimate."

After a few days, there are 42 votes. The following is the result, shown as ratio of vocabulary:grammar study time, and percent of the responses

8:2  30%
9:1  20%
10:0 15%
7:3  11%
5:5   9%
4:6   8%
1:9, 3:7, 6:4 2%
2:8   1%

We can see that for instance nearly 1/3 of the language learners spend 8 out of 10 hours studying vocabulary and 2 hours studying grammar, while 1% of the people do exactly the opposite. This result shows that the polyglots taking this poll definitely spend more time studying vocabulary than grammar. If this time distribution implies relative importance, it is clearly consistent with Mr. Kaufmann's opinion that vocabulary is more important than grammar.

This poll is followed by 26 comments. Some interesting findings from them are:

(1) If the learner is a beginner in learning a specific language, he or she spends a significant amount of time studying grammar. The vocabulary:grammar study time ratio could be 5:5 or even lower. But as study progresses, the ratio gradually increases.

(2) This poll is about the learner's current state, averaged over all the languages being studied if multiple. One interesting example is a Portuguese learner who says 10:0 when studying Spanish (no need to study grammar as the two languages are so much alike on that), 8:2 when studying English, and 7:3 when studying Swedish. So I did an average for him, which is (10+8+7):(0+2+3)=25:5=8.33:1.67 or about 8:2. He agreed.

(3) Different languages require different ratios. For example, Chinese is generally considered to demand an extraordinary amount of time on vocabulary but very little time on grammar, unlike say Latin, Ancient Greek, or Sanskrit. Since many languages are studied and polled about, there won't be bias introduced by any specific language. And if a learner is studying multiple languages, he's supposed to enter his average.

(4) Some people say they don't study either because their study is completely immersion. That is unusual for an adult learner. But lack of urgency, prioritizing fun well above everything else, and having a childlike curious mind make this option possible.

Back to Kaufmann. We can reasonably believe that he is at an advanced stage on all or most of the languages he knows. As said above (see (1)), at this stage, the vocabulary:grammar ratio tends to be high, leading him to make that remark. Why do people show their disagreement with him? It's possible that most people have the tendency to misread "X is more important than Y" as "X is important but Y is not". This tendency is especially common when people read an online article about health or medical science. Secondly, people disagree with somebody else by interpreting the latter's words as a universal rule, to make it more criticizable. If one of the languages you're studying takes more time on grammar than vocabulary, even if you're studying multiple languages for which this ratio is averaged to be in favor of vocabulary instead, you still disagree by ignoring the average.