《墨辯》中的認知與事實

English title: Cognition and fact in the Mo-bian

作者：德龍（Donald Sturgeon）
東吳大學哲學系研究生
MA student, Soochow University Department of Philosophy, Taipei

[提要]

　　這篇文章將要嘗試以《墨辯》的文本資料為根據，從語言哲學的角度探討《墨辯》中對認知與事實的看法，以及古代漢語對這些看法所造成的影響。

　　在有關認知與事實的部分，首先會討論《墨辯》如何談論不同類型的知識和不同認識活動。《墨辯》對語言的分析的一個很大的成就在於它較具體地描述了人透過與事物的接觸而獲得的直接認識和透過語言活動或理性思考得來的間接認識之間的關係。這一部分首先會提到《墨辯》中「知」的概念及其有關知識的分類和解釋，然後說明《墨辯》如何使用「類」、「名」、「辭」來解釋直接認識到間接認識的具體過程。依本文的看法，《墨辯》的作者對認知過程的解釋大體上如下：人面對外在世界中的實存之物（實）首先通過感官能力（五路）觀察到事物的某些性質（同異），然後根據這些性質形成有相同性質的概念（類），再把這些概念與被公認的語言符號相聯結（名實耦），因而獲得了用語言符號指稱實存之物的能力（以名舉實），並且通過這些語言符號來傳達思想（以辭抒意）。這樣，《墨辯》說明了透過語言表達的間接認識如何可能。

　　接下來文章會探討《墨辯》有沒有直接討論到事實或真理的概念。在某一種意義上，每當我們說出一個判斷的時候，我們已經有使用到真理這一概念；但有沒有對真理概念本身進行反省是另外一個問題。《墨辯》當中確實有探討跟真理和事實相關的一些概念，因此這一節要對這些概念作分析，並討論《墨辯》中是否有談到一種連接世界和語言的類似於「真理述詞」的概念。《墨辯》很強調判斷一定要根據客觀的標準；這一部分還會通過對《墨辯》中「當」、「可」、「然」、「是」的分析釐清《墨辯》對事實的看法。

　　文章第二部分的主題是古代漢語對《墨辯》中有關認知與事實的理論和分析的可能影響。首先會提到西方漢學家陳漢生的「物質名詞假設」及其與《墨辯》中的認知的關係。如果陳漢生的假設能夠成立的話，它對《墨辯》作者的認知觀念會有很大的意義；但本文將試圖說明陳漢生的這一論證就《墨辯》而言是值得懷疑的。這一節一方面會從《墨辯》的「體」、「兼」觀點分析，一方面用西方分析哲學家蒯因的「根本翻譯」（radical translation）論起。

　　其次，本文將要分析古代漢語文法的靈活性與《墨辯》「名」、「辭」結構之間的關係。在這一節將會說明為什麼《墨辯》的「名」、「辭」結構與古代漢語的特徵密切相關，並且通過對〈小取篇〉的分析，具體地說明古代漢語如何影響了《墨辯》對語言的分析。

　　最後，將要從《墨辯》的立場面對古代中國有沒有真理概念這一問題。這裡又會從陳漢生的主張談起，從《墨辯》的語言理論和語言哲學兩個角度指出陳漢生「中國哲學沒有真理概念」此一論證難以成立的理由。除了上述《墨辯》直接有關真理、事實問題的討論之外，這些理由主要見在其論證的前提，包括「古代中國沒有語句的概念」、「古代漢語只能表達『詞信念』而無法表達『語句信念』」等。

關鍵詞：墨辯、認知、事實、真理、物質名詞

Cognition and fact in the Mo-bian

[Abstract]

This paper will attempt to use the text of the Mo-bian (Mohist dialectical chapters) to discuss Mohist ideas of cognition and fact, and the role the classical Chinese language may have played in influencing these ideas.

The first part of the paper will deal with cognition and fact, and will begin with how the Mo-bian discusses different kinds of knowledge and different kinds of cognition. A considerable achievement of the Mo-bian’s analysis of language was that it gave a relatively concrete picture of how we can progress from direct acquaintance acquired through direct contact with the world, to indirect acquaintance acquired through language or rational thought. This section will firstly discuss the Mo-bian’s conception and classification of zhi (knowledge), and then explain how Mo-bian used the concepts lei (class), ming (term), and ci (expression) to explain in concrete terms the progression from direct acquaintance to indirect acquaintance. This paper will argue that the author of the Mo-bian had the following view of cognition: faced with actual things in the external world (shi), we firstly use our five senses (wu lu) to distinguish properties of these things (tong-yi), then establish classes of things with similar properties, or concepts (lei), then tie these concepts to commonly recognized linguistic symbols, or names (ming shi ou), so acquiring the ability to use names to refer to actual things (yi ming ju shi), finally using complex expressions containing these names to express and transmit thoughts or ideas (yi ci shu yi).

Next I will discuss whether or not the Mo-bian talked directly about the concepts of fact and truth. In some sense, whenever we make a declarative statement, we have made use of the concept of truth; but whether or not we have reflected upon truth as a concept is a separate question. The Mo-bian did indeed discuss concepts related to truth and fact; here I will analyze these terms and discuss whether or not the Mo-bian mentioned a concept connecting language and reality similar to the truth predicate. The Mo-bian stressed that judgments should be made on the basis of objective standards; here I will use an analysis of dang, ke, ran and shi to explain the Mo-bian’s view of fact and truth.

The second part of the paper will discuss possible effects of classical Chinese on the Mohist analysis of cognition and fact. Firstly I will discuss Chad Hansen’s mass noun hypothesis and its relation to cognition in the Mo-bian. If the hypothesis were to hold, it would have great significance for the cognitive stance taken by the author of the Mo-bian; however I will attempt to show that it is doubtful whether the hypothesis can hold for the Mo-bian. This will be done on the one hand by an analysis of ti (part) and jian (whole), and on the other using Quine’s idea of radical translation.

Secondly, the paper will discuss the relation between the flexibility of classical Chinese grammar and the Mo-bian’s ming, ci structural analysis of language. Here I will attempt to show that this structural analysis is closely related to the flexibility of grammar. A detailed analysis of the xiao-qu chapter of the Mo-bian will be used as a concrete example of this.

Finally I will face the question of whether or not ancient China had a concept of truth from the Mohist position. Here I will proceed from a discussion of another of Chad Hansen’s theories, that “Chinese philosophy has no concept of truth”, and give reasons why this theory is problematic from the points of view of both the Mo-bian and philosophy of language. In addition to the previous discussion of the Mo-bian’s views on cognition and fact, these will include problems with Chad Hansen’s assumptions, such as that Chinese philosophy had no concept of a sentence, and that classical Chinese could only express so-called “term-belief” and not “sentential belief”.

Keywords: Mo-bian, Mohist dialectical chapters, cognition, fact, truth, mass nouns