奎因哲学词条：使用与提及

Christian Morgenstern's werewolf betrayed intellectual limitations when he asked the schoolmaster to decline him. We do not decline people, or even werewolves; we decline words. Confusion of things with their names, signs with their objects, is confusion of use and mention.

当克里斯蒂安·摩根斯特恩笔下的狼人请求校长把他变为复数形式时，他暴露了自己的智力缺陷[1]。对于我们来说，人没有复数形式，狼人也没有，只有语词才有复数形式。混淆事物及其名称，混淆符号及其对象，就是混淆（对语言表达式的）使用和提及。

GOOGLE:　Christian Morgenstern's werewolf reveals his intellectual shortcomings when he asks the schoolmaster to turn him into the plural. For us, people don’t have plural form, and neither do werewolves. Only words have plural form. To confuse things and their names, to confuse signs and their objects, is to confuse use and mention (of a linguistic expression).

To mention something we use its name, or some description. In saying that Boston has thirteen councilmen we use the name of the city and thereby mention the city, as I have just done. Little mystery thus far, thanks to the happy circumstance that there is little less like a city than a name. Mentioning cities and other concrete objects is child's play; just use their names.

为了提及某个东西，我们使用它的名称或对它的某个描述。当我们说波士顿有十三名市议员时，我们使用这个城市的名称并因此提及这个城市。这里并没有什么神秘之处，因为一个城市和它的名称之间几乎没有任何相像之处。提及城市以及其他的具体对象是手到擒来的事情；使用它们的名称即可。

GOOGLE:　To mention something, we use its name or some description of it. When we say that Boston has thirteen councilmen, we use the name of the city and thus mention the city. There is no mystery here, as there is almost no resemblance between a city and its name. Mentioning cities and other concrete objects is easy; just use their names.

Caution begins to be in order, however, when we take to mentioning names. To mention a name, like anything else, you use a name of it. Boston is not disyllabic, but 'Boston' is; the quotation serves as a name of the name. A quotation names its insides. It is a name of its own guts.

然而，当我们提及名称时，就开始要小心了。提及一个名称，就像提及任何别的东西一样，你需要使用它的名称。Boston（波士顿）不是双音节的，但“Boston”是双音节的；引语起到了名称的名称的作用。引语命名其里面的东西。引语是其里面的东西的名称。

GOOGLE: However, we need to be careful when we mention names. To mention a name, just like mentioning anything else, you need to use its name. Boston is not disyllabic, but "Boston" is; the quotation functions as a name for a name. A quotation names, or is a name of the thing inside it.

Nor let it be supposed that 'Boston' is a quotation. 'Boston' {232} is just a six-letter word, and contains no quotation marks. To mention the quotation we use its name, a quotation of the quotation. ' 'Boston' ', contains one pair of quotation marks.

请不要认为“Boston”是一个引语，它只是一个含有6个字母的单词，并不包含引号。为了提及引语，我们使用它的名称，即引语的引语。“‘波士顿’”包含了一对引号。[2]

GOOGLE:　Please don't think of "Boston" as a quotation, it's just a six-letter word that doesn't contain quotation marks. To mention a quotation, we use its name, a quotation of a quotation. "'Boston'" contains a pair of quotation marks.

In mathematical writings, where there are no concrete objects such as cities to cleave to, confusion of use and mention is rife. The following passage, from a widely used textbook on the differential calculus, is fairly typical:

The expression D_xyΔxis called the differential of the function and is denoted by dy:

dy=D_xyΔx

The third line of this passage, an equation, is apparently supposed to reproduce the sense of the first two lines. But actually, whereas the equation says that the entities dyand D_xyΔx (whatever these may be) are the same, the preceding two lines say rather that the one is a name of the other. And the first line of the passage involves further difficulties; taken literally it implies that the exhibited expression ' D_xyΔx ' constitutes a name of some other, unexhibited expression which is known as a differential. But all these difficulties can be removed by a slight rephrasing of the passage: drop the first two words and put the first occurrence of 'dy' in quotation marks.[3]*

在数学著作中，由于没有城市等具体对象可供坚守，使用和提及的混淆比比皆是。下面这段出自一本曾广泛被使用的微分学教科书里的话是相当典型的：

表达式D_xyΔx被称为函数的微分，用dy表示：

dy=D_xyΔx

这段话的第二行是一个等式，显然是为了把第一行的意思再说一遍。但实际上，等式说的是dy和D_xyΔx这两个实体（无论它们是什么）是相同的，而第一行说的却是一个实体是另一个实体的名称。而且，这段话的第一行还涉及更多的困难；从字面上理解，它说的是“D_xyΔx”这个写出来的表达式是另一个没有写出来的被称为微分的表达式的名称。但是，只要对这段话稍加改动，就可以消除所有这些困难；去掉最前面的“表达式”三个字，并把第一次出现的“dy”加上引号[4]。

GOOGLE: In mathematical writings, confusion of use and mention abounds as there are no concrete objects such as cities to cling to. The following passage from a once widely used differential calculus textbook is quite typical:

The expression D_xyΔx is called the differential of the function, represented by dy:

dy=D_xyΔx

The second line of this passage is an equation, obviously purport to repeat the meaning of the first line. But in fact, the equation is saying that the two entities dy and D_xyΔx (whatever they are) are the same, while the first line is saying that one entity is the name of the other entity. Moreover, the first line of the passage involves further difficulties; taken literally, it says that the written expression "D_xyΔx" is the name of another unwritten expression called a differential. However, all these difficulties can be eliminated by making a slight change to this passage; remove the first two words "the expression" and put the first occurrence of "dy" in quotation marks.

Mathematicians' carelessness over use and mention has led to philosophical confusions about identity. Thus Leibniz explained identity as a relation between the signs, rather than between the named object and itself: "Eadem sunt quorum unum potest substitui alteri, salva veritate." Various mathematicians, including Whitehead at one point (in Universal Algebra, 1898), were led to look upon equations as relating numbers that are equal but distinct. Count Alfred Korzybski, in Science and Sanity, went so far as to propound a doctrine of nonidentity: {233} identities are always false. Even '1 = l' is false, he wrote, because the two numerals are unlike at least in position. He confused the ink spots with the number, abstract and invisible.

数学家对使用和提及的漫不经心产生了混乱的同一性哲学。莱布尼兹把等同解释为符号之间的关系，而不是被命名的对象和自身之间的关系：“Eadem sunt quorum unum potest substitui alteri, salva veritate.”（如果一个术语可以替代另一个术语而不改变任何陈述的真理性，那么这两个术语是相同的。）包括怀特海在内的很多数学家说等号联结的是相等但有区别的数。阿尔弗雷德•科尔兹布斯基伯爵在他的《科学和理智》一书中甚至提出了一个所谓的非同一性学说：所有等式都是假的。即使“１＝１”也是假的，他写道，因为两个数字至少在位置上是不同的。他将墨渍和数混淆了起来，但这两者不是一回事，后者是抽象的、不可见的。

Mathematicians' carelessness in use and mention has produced a confused philosophy of identity. Leibniz interpreted equality as a relation between signs rather than between the　named object and itself: "Eadem sunt quorum unum potest substitui alteri, salva veritate."（Two terms are the same if one can be substituted for the other without altering the truth of any statement.） Many mathematicians, including Whitehead, say that the equality sign connects equal but distinct numbers. Count Alfred Korzybski even proposed a　doctrine　of nonidentity in his book Science and Sanity: all equations are false. Even "1 = 1" is false, he wrote, because the two numerals are at least different in position. He confused　ink stains with the　number, but they are not the same thing; the latter is abstract and invisible.

Granted, a robust nominalist will have no truck with abstract objects such as numbers. That still does not reduce '1 = l' to an identification of ink spots; it rates it as not identifying anything.

诚然，一个坚定的唯名论者不会接受数之类的抽象对象。但唯名论者不会把“1=1”理解为关于墨渍的语句，而会认为该语句没有谈到任何东西。

GOOGLE: It is true that a committed nominalist would not accept abstract objects such as numbers. But a nominalist would not understand "1=1" as a statement about inkblots, but would think that the statement is about nothing.

Strict observance of the distinction between use and mention of expressions is essential to clear thinking, no matter what one's philosophical position may be regarding the reality of abstract objects that those expressions may purport to name. Clarity of the distinction is indispensable, in particular, to an understanding of the proof of GÖDEL'S THEOREM, and of Tarski's work on TRUTH.

严格遵守表达式的使用和提及之间的区别，对于清晰的思维至关重要，无论一个人对这些表达式旨在命名的抽象对象的实在性持有怎样的哲学立场。特别地，清晰地区分这两者，对于理解哥德尔定理的证明和塔尔斯基在真理问题上所做的工作，是不可或缺的。

GOOGLE: Strict adherence to the distinction between use and mention of expressions is essential to clear thinking, whatever one's philosophical position may be regarding the reality of the abstract objects those expressions purport to name. In particular, a clear distinction between the two is indispensable for understanding the proof of GÖDEL'S THEOREM and Tarski's work on the issue of TRUTH.

Some philosophical diffidence regarding the existence of UNIVERSALS such as numbers has perhaps been a factor in mathematicians' mishandling of use and mention. The difference between a number and the numeral that names it does not vividly obtrude if there is no number, but it should. There being a numeral but no number should surely be difference enough.

对数之类的共相的存在性缺乏哲学信心，可能是数学家对使用和提及的不当处理的一个因素。如果数根本不存在，那么数字和它所命名的数之间的区别，就不会那么显眼，但其实应该是显眼的：数字存在但数却不存在，这肯定是一个足够的差别了。

GOOGLE: A lack of philosophical confidence in the existence of UNIVERSALS such as numbers may be a factor in the poor handling of use and mention by mathematicians. If numbers do not exist at all, then the difference between a numeral and the number it names would not be so obvious, but it should be: it is a sufficient difference that the numeral exists but the number does not.

Anyway the metaphysics of abstract objects is quite beside the point of use versus mention. Be there numbers or be there none, the point is to distinguish between the use of the numeral, '5' say, and the mention of it. We mention '5' when we contrast it with its modern Arabic counterpart, which looks more like '0', or when we explain that the curved lower part of '5' derived from three strokes, making for a five-stroke character. We use '5' when we say there are 5 fingers on a hand, or 5 New England states other than one's own.

无论如何，关于抽象对象的哲学观与使用/提及的区别是无关的。不管数是否存在，这里的要点要把对数字（比如“5”）的使用和对它的提及区别开来。当我们把“5”与它在现代阿拉伯语中的对应者（看起来更像“0”）加以对比；或当我们解释说：“5”的弯曲下部是如何从三笔画演变为五笔画的，我们是在提及“5”这个数字。而当我们说一只手有5个手指，或除了某人自己所在的州之外，新英格兰还有5个州，我们是在使用“5”这个数字。

GOOGLE:　In any case, the philosophical view of abstract objects has nothing to do with use versus mention. Regardless of whether the number exists, the point here is to distinguish between the use of the numeral (such as "5") and the mention of it. When we compare "5" with its modern Arabic counterpart (which looks more like "0"); or when we explain how the curved lower part of "5" has evolved from a three-stroke to a five-stroke, we are mentioning the numeral "5". And when we say that there are five fingers on a hand, or that there are five New England states except one’s own, we're using the numeral "5."

Confusion of use and mention can surface in subtle ways also beyond the bounds of mathematics, as this dialogue illustrates: "Well, pǎtronage, pātronage, as you please." "You seldom get one without the other."

使用和提及的混淆在数学以外的领域也会以微妙的方式出现，下面的对话就是一个例子：“好吧，赞（zǎn）助也行，赞（zān）助也行，随你便。”“你很少会只得到赞（zǎn）助而得不到赞（zān）助，或只得到赞（zān）助而得不到赞（zǎn）助。[5]”

GOOGLE: Confusion of use and mention can arise in subtle ways outside of mathematics, as exemplified by the following conversation: “Well, both pǎtronage and pātronage are all right.” “You rarely get one without the other. "

Quotation is one of two usual ways of naming an expression. The other way, customary in linguistic studies, is putting the {234} expression itself in italics. This way is impractical in treating of mathematical and logical formulas, which are largely in italics to begin with, or in alien alphabets, or in none. This book deals largely with matters where it is usual to name expressions by italics, and largely also with matters where it is usual to name them by quotation. I have wanted to conform to custom on both sides, and yet to maintain some semblance of consistent usage. I have found I can reconcile these three desiderata to some degree by using italics when the main interest lies in the form or history of the expression, and quotation otherwise.

引语是命名表达式的两种常用方法之一。另一种是语言学研究中的惯用方法，即把表达式本身用斜体标出。这种方法在处理数学和逻辑公式时是不实用的，因为这些公式本身大体上就是用斜体来表示的。这本书里我这两种方法都有采用。我希望遵守关于这两种方法的惯例，同时也希望我能在这两种做法中能保持某种表面上的一致性。我发现了一个可以同时满足这三点的折衷方法：如果对某个表达式的兴趣在于其形式和历史，就使用斜体，而在其他情况下则使用引语。

GOOGLE: Quotation is one of two common methods of naming expressions. The other is a customary in linguistic studies, which is to italicize the expression to be discussed. This method is not practical when dealing with mathematical and logical formulas, because the formulas themselves are generally represented in italics. In this book I use both methods. I hope to adhere to conventions regarding both methods, but also hope that I can maintain some semblance of consistency between the two methods. I've found a compromise that satisfies these three points: use italics if the interest in an expression lies in its form and history, and use quotations in other cases.

There is another use of quotation marks, called sneer quotes or scare quotes, that is not meant to switch use to mention. It serves rather to mark the expression as one that the writer is using without recommending. It has the force of use without prejudice, or, in Yiddish, soil mir nicht schuldigen. Italics also of course have uses other than the naming of expressions-notably, as just now, for foreign languages.

引号的另一种用途被称为冷笑引号或恐吓引号，其目的不是从使用转向提及被引的表达式，而仍然是一种使用。但这种使用不涉及引用者的褒贬或立场，或者用意第绪语来说，在这种使用中：soil mir nicht schuldigen（我是无罪的）[6]。斜体也有命名表达式以外的用途，特别地，像我刚才所做的那样，被用来表示被斜体的语词是外语。

GOOGLE: Another use of quotation marks is called sneer quotes or scare quotes, the purpose of which is not to move from use to mention of the quoted expression, and the purpose is still use of it. But this kind of use does not involve the quoter's praise or position, or in Yiddish: soil mir nicht schuldigen (I am innocent). Of course, italics can also be used for purposes other than naming expressions，in particular, as I just did, it is used to indicate that the italicized words belong to a foreign language.

The naming of expressions by quotation proves inadequate when we rise to generalities. Thus the logician wants to say that the sentence 'Tufa floats or tufa does not float', and all sentences like it, are true. Like it? Well, where φ is any sentence, 'φ or not φ' is true. No, this will not do. If the letter 'φ' is being used as a variable ranging over sentences (see VARIABLES), or in effect as a name of an unspecified sentence, then it is grammatically a noun, and 'φ or not φ' is not a sentence at all, much less true. What we want to say is rather that 'φ or not φ' becomes true when the letter 'φ' is supplanted in it by the sentence φ. In 1940 I introduced a notation of quasi-quotation for the purpose: ⌜φ or not φ⌝ is to be understood as the result of supplanting 'φ' by φ in 'φ or not φ'. The general convention is that the quasi-quotation designates what its insides would become if all its Greek letters were supplanted by the expressions they designate. The device has been taken up, for it is needed in some logical studies. I have avoided it in this book; {235} I would simply say, if occasion arose, that 'p or not p' is true where 'p' stands in place of any sentence. I have gone into the matter now only to illustrate that there is more to quotation and its like than at first meets the eye. For still more in that vein see TYPE VERSUS TOKEN.

当我们想概括地谈论表达式的名称时，引语是不够的。例如，逻辑学家想说：语句“石灰华要么漂浮要么不漂浮”，以及所有与之类似的语句，都是真的。与之类似？是不是说：如果φ是任何语句，那么“φ或者并非φ”是真的？不，这是不行的。如果字母“φ”被用做指称语句的变项，即实际上被用做一个未明确指定的语句的名称，那么它在语法上是一个名词。因此，“φ 或者并非 φ”根本就不是一个语句，更不是真语句。我们真正想说的乃是：“φ 或者并非 φ”将会变成真的，如果其中的“φ”被φ所替换的话。1940年，我为此引入了一个我称为准引语的设施：⌜φ or not φ⌝被理解成“φ 或者并非 φ”中的“φ”被φ所替换的结果。这里的一般约定如下。准引语指称的是：引号里面的东西——在其中的所有希腊字母被替换以它们所指称的表达式后——所变成的表达式。这一设施因为在有些逻辑研究中用得上而被采用了。但我在本书中避免了它；如果有需要，我会径直说：“p或者并非p”是真的，无论“p”所在的位置填进什么语句。我谈起准引语这一设施，只是为了表明，引语之类东西远不是表面看上去的那样简单。关于这方面的更多内容，请参阅类型和标记。

GOOGLE:　When we want to talk about the name of an expression in general, quotation is shown to be insufficient. For example, a logician wants to say that the statement "Travertine either floats or it doesn't," and all similar statements, are true. Similar to this statement? Does it mean that if φ is any sentence, then "φ or not φ" is true? No, this will not do. If the letter "φ" is used as a variable for statements, that is, as in effect the name of an unspecified statement, then it is grammatically a noun. Thus "φ or not φ" is not a sentence at all, let alone a true sentence. What we really want to say is: "φ or not φ" becomes true if "φ" is replaced by φ. In 1940, I introduced a facility called quasi-quotation for this purpose: ⌜φ or not φ⌝ is understood as the result of replacing the “φ” in “φ or not φ” with φ. The general convention here is as follows. A quasi-quotation refers the expression of what is inside the quotation marks becomes after all the Greek letters in it are replaced by expressions they refer. This facility has been adopted because it is needed in some logical studies. But I avoid it in this book; if necessary, I will simply say: "p or not p" is true, no matter what sentence is inserted in the place of "p". I mention the facility of quasi-quotation only to show that there is more to things like quotations than what they seem. For more on this, see TYPE VERSUS TOKEN.

[1] 这是德国诗人克里斯蒂安·莫根施特恩（1871——1914）的作品“狼人”中的情节。——译者注

[2] Boston是一个城市。“Boston”是一个（城市的）名称。“‘Boston’”是一个（城市名称的）名称，这是一个引语。“‘“Boston”’”是一个（城市名称的名称的）名称，这是一个引语的引语。——译者注

[3] Here I have transcribed a paragraph from my Mathematical Logic (1940; Harvard, 1951), p. 25.这里，我从我的《数理逻辑》一书抄了一段。

[4] D_xyΔx被称为函数的微分，用“dy”表示：dy=D_xyΔx。【“表达式D_xyΔx……”意味着：D_xyΔx是表达式。但D_xyΔx并不是（微分的）表达式，而是超语言的微分。所以“表达式D_xyΔx……”应该改为“D_xyΔx……”。“用dy表示”意味着：dy是表达式。但dy并不是表达式而是超语言的微分，所以“用dy表示”应该改成“用‘dy’表示”，“dy”才中表达式。】——译者注。

[5] 很显然，“赞（zǎn）助”、“赞（zān）助”，以及“赞助”，这三个不同的词指的是同样的事情。对话前一句中应该是提及了前两个词，而后一句则应该是使用了这两个词。据此，对话可以修改为：“好吧，措辞无所谓，“赞（zǎn）助”也行，“赞（zān）助”也行，随你便。”“嗯，赞（zǎn）助和赞（zān）助是一回事。”——译者注

[6]举例说明：林彪事件后，人们在中国公开出版的报刊中有将近两年没有见过他的名字（1971年10月——1973年8月），这期间他成了“刘少奇一类骗子。”【在这段话中，我们使用引语“‘刘少奇一类骗子’”，目的并不是提及“刘少奇一类骗子”这个短语，而是使用“刘少奇一类骗子”这个短语来提及林彪。但显然，引者并不一定认同林彪是刘少奇一类骗子这一定性。】——译者注。

微信扫码关注该文公众号作者

来源：哲学园

奎因哲学词条：使用与提及

相关新闻