上海直面Rorty：

交互超越主义与新实用主义的交锋

翟振明

（中山大学哲学系）

中国现代外国哲学学会年会暨西方技术文化与后现代哲学学术研讨会大会论文 

美国多年来最当红的哲学家理查德·罗蒂（Richard Rorty）2004年7月17日至18日在上海华东师大的学术活动，主要是在“罗蒂、实用主义和中国哲学”研讨会上与中外学者就他的哲学思想进行研讨交流。就像罗蒂在18日下午的回应报告中所说的那样，笔者是他在这次学术活动中遇到的主要对手，这是因为，我的交互超越主义与他的新实用主义针锋相对、势不两立。早在10年前，我就在我的第一本英文专著《本底抉择与道德理论》[1]中运用交互超越主义的基本原则开拓道德形上学的新领地，为从描述到规范的过渡做出了新的尝试。这次与罗蒂的直接交锋，就是围绕我对罗蒂认识论上的自然主义之内在矛盾的揭露而展开的。直面的交锋主要有两次，第一次是现场生发的质疑与应对，第二次是双方各自事先准备好的陈述与回应。

（一）

17日上午会议开始，罗蒂作了“Trapped between Kant and Dewey: Current Situation of Moral Philosophy”（“困于康德与杜威之间：道德哲学的当前状况”）的主题报告。报告结束其他几位先提了几个问题后，我拿过麦克风告诉大家，我有十多个问题要问罗蒂，但由于时间不允许，只能挑其中三个问。

第一个问题是针对他对道德概念的重新解释而发的。他在报告中说，康德式的道德哲学自以为可以为人们的抉择提供普遍的标准，那是一种错误。哲学家在道德问题上不比小说家或其他人文学者更能提供抉择的标准。道德哲学家能做的，只是把由自己的博学和深思而扩展了的想象力充分发挥，将人们面对的各种可能性展示给他们，从而为他们对自己的“实践认同”（practical identities）的抉择提供更广阔的空间。因此，在罗蒂看来，道德哲学的作用，是在抉择之前展示前景，而不是在抉择中提供指引。我的问题是，罗蒂对道德概念的重新解释，与康德及任何其他人所理解的道德毫无关系，谁会认为，在行动之前看到了更多的前景，不管最后做出怎样的抉择，就是更接近道德的要求呢？罗蒂不是用了“道德”这个词来指称与其本来的所指毫无合理的逻辑关系的东西吗？这样把概念偷换了，怎么能对康德的道德哲学构成任何反驳呢？对我这个问题，罗蒂作了最简单方便的回答：承认他的道德概念与康德的不一样。但是，罗蒂对他的道德概念是否与大家公认的道德概念不一样、采用与康德不同的基本概念是否能构成对康德道德哲学的批驳这些关键问题，没有做出回答。

我的第二个问题，是针对罗蒂对康德prudence/morality (慎思/道德)的两分提出的责难。他说，康德把慎思得出的规范看作有条件的，而把由绝对命令得到的规范看作无条件的，亦即，前者是手段，后者是目的。但是，在罗蒂和杜威看来，这种分法是任意的，从而是不能接受的。所有目的，都是由当前情景决定的，有条件的，无所谓终极目的或终极价值。我反问罗蒂，一切有条件的价值，都是指向它之外的，“有用”的东西，只有相对于它服务的外在目的才会“有用”。但如果这样一直指向外边，就会导致无穷倒退，从而使“有用”的说法变得毫无意义。因此，任何有条件的外在价值，在逻辑上必然要求最终导向某种无条件的自足的价值，像自由、尊严、幸福等，才获得意义。因此，抛弃了无条件的终极价值的概念，也就同时抛弃了有条件的工具价值的概念。罗蒂怎么能够拒斥其一而保留其二呢？罗蒂的回答是，这种相对与绝对的逻辑关系，只要落入其圈套，就必定不能自拔。但是我们可以从一开始就拒绝这种语言游戏，问题也就不会存在了。但是，我要问罗蒂的是，当你说一切目的都是“有条件”的目的时，你不就已经承认了“有条件”和“无条件”的区别，从而已经在这个语言游戏中扮演角色了吗？你怎么能不落入圈套呢？不过当时没机会进一步追问他。

    我的第三个问题，稍微复杂些，引起我和罗蒂几个回合的交锋。这个问题，是罗蒂回答来自美国的倪培民教授的有关康德道德哲学在大学伦理学教学中起的作用时的说法引出的。罗蒂认为，要教会学生懂得人的自由自律的精神、懂得不能把他人只当工具的道理，用援引历史事例的方法要比学习康德道德哲学有效得多。比如说，给学生们讲解黑奴解放、妇女解放等等人的解放(emancipation)运动的过程，要比叫他们学习康德哲学更好。我就问他，按照你的历史自然主义，不存在超历史的标准可以来衡量历史过程中的事件，那么你如何判断哪些历史事件属于“解放”的范例，哪些又不属于呢？罗蒂回答说，我就给学生讲解具体的事例，为何需要告诉他们跨越历史的原则呢？我说，你可以不告诉他们，但在你告诉他们实际事例之前，你在自己的思想中总要先做出区分吧？罗蒂即刻顿住了，好一会才含糊地说，你这是柏拉图式的想法，好像要区分哪些事例属于人的解放的事例，哪些不属于，就必然要应用某种原则才能做到。这种柏拉图式的观点并没有被证明是正确的，因而是可疑的。但是，有机会的话，我还要继续追问罗蒂：不管柏拉图是否正确，也不管你有没有明确地使用某条原则，难道“解放”这个词可以在每个事例中各有不同的含义吗？如果这样，那不就任何历史事件都可以被任意看作是“解放”或相反吗？这样，所谓“人的解放的历史”的说法，不就变得毫无意义吗？你不把希特勒的种族灭绝行为拿出来当作“人的解放”的事例，难道是任意的、没有标准的？无论如何，这里必定预设了某种超越历史的价值标准。其实，会后，吴冠军告诉我，他在电梯里顺着我的问题继续追问罗蒂，罗蒂说，翟的问题问得好，而他的回答却不那么好，他承认自己的答复有点牵强。

（二）

   第二次的交锋，是在罗蒂认真阅读了我提交的会议论文后进行的。我在文中提出“交互超越主义”直接与罗蒂的实用论自然主义相对立。我的文章分三部分：1）讨论为何罗蒂把康德哲学归入表象主义的范畴是完全错误的。康德把知识的普遍有效性归结为理性主体间的范畴体系和感性直觉的共同性，根本就不涉及对物自体的“表象”。2）证明罗蒂的实用论自然主义是不可辩护的。罗蒂只承认因果关系并把理性辩护的过程仅仅看作是因果过程，这样的话，他就不可能为他的哲学进行辩护，因为任何理论辩护都预设了与自然主义对立的交互超越主义原则，因而，罗蒂要为他的自然主义进行理论辩护，一开始就要站在反自然主义的立场上。3）罗蒂在价值规范问题上的历史自然主义，就是从他本人的实用主义立场上看，也会导致社会实践上的危险。罗蒂认为，以历史主义的态度对待实践目标的选择，能使我们相互宽容，获得更多的自由。但是，我在文中阐明，在某种社会条件下，罗蒂式的历史主义态度将会鼓励政治权力的自我膨胀，导致社会的大多数成员丧失自由。相反，如果我们采纳交互超越主义的基本原则，对每一个社会目标的选择，都通过理性主体间寻求超越一己偏见的共识的过程来达到，就更有可能在社会实践中实现人的自由、尊严等这些罗蒂推崇的价值。况且，哲学家的刨根问底，本来就有其自足的价值，并不一定要服务于某个特定的社会目标。最后，我对罗蒂的哲学取向作了一般性的评论。他说哲学家不对命题的真确性负责，只需为人们提供新的词汇来谈论他们的希望、恐惧、抱负、和前景，以此来改善我们语言的描述功能。这样的话，我说，罗蒂可以随意变换他的词汇和他的论证，根本就无需为自己哲学的逻辑一贯性负责。这样，他对自己的新实用主义的任何辩护都是多余的，因此我们对他的任何哲学断言都不必过于认真。

    虽然罗蒂把我当作他此行的主要对手（most unsympathetic）来对待，但由于他的哲学确实在理论上不可辩护，给出的回应比较空洞。他对我文章第一部分提出的问题，即康德与表象主义的问题，没有正面回答，而只是在回答其他两个问题时对此有所暗示。一开始，他说，对于我的整篇文章提出的理论责难，他不知道应该如何从理论上进行辩护。他接着又结合第一次交锋时我提的第二个问题，即工具价值与内在价值的逻辑关系问题，一概说，历史上哲学家们寻找普遍原则的失败，是他认为这种寻找徒劳无益的唯一根据。从柏拉图到康德一直到现代，哲学家们并没有给我们留下什么普遍有效的原则，就算有，人们也不会太当一回事。所以，罗蒂认为，我们应该改变哲学的功能，放弃对超历史的普遍原则的追寻。我回应道，哲学史上没完成的东西，正是我们需要继续追寻的东西；如果已经完成了，我们还需要继续做什么呢？任何其他学科，都不会用过去的不成功作为放弃的理由，为什么哲学就相反？如果我的钱包丢了，找了一会没找着，就可以断定，我本来就没有钱包吗？哲学的历史，相对于人类历史可能的漫长程度，实在不算长，我们为何要轻言放弃？况且，哲学家们是否已找到了普遍原则，也不能以多数人的认同为准。在我看来，康德哲学中的主要部分，就是普遍有效的，尽管没有很多人接受其有效性。

    对于我文章第三部分提出的社会实践问题，罗蒂说，我似乎比他更相信哲学的社会政治功能。他反问我，你认为我们到底采纳实用主义还是交互超越主义，对社会历史的进程会有什么影响吗？真正影响社会进程的，是经济活动本身、社会政治力量内部张力导致的动态平衡、等等，而不是哲学家及其他人文学者们的话语。如果我们要想成为政治活动家，根本无须谈论哲学。他的意思是，我高估了哲学对现实的影响力。我的回应是，不管实际上哲学对现实有多大影响，我的问题是按照罗蒂他自己的实用主义思路提出来的，而不是基于对哲学思想实际上起的作用的猜测。按黄勇的说法，他的哲学实用主义，是不顾逻辑后果，只讲实践的后果的。最后，刘擎的一个问题，把罗蒂问住了：人们不谈论哲学家的问题，也许说明哲学家的思想已经被接受而不是被拒斥，说明哲学已融入人们的生活之中起了很大的作用？比如说，罗蒂所反对的符合论的真理观，是否已被科学家共同体接受，使得科学成为现在这个样子？罗蒂说，这是一个非常有趣的思路，他自己还没认真考虑过。

（三）

    总的来说，我这次与罗蒂的直面交锋，不是只停留在言词之争，而是涉及到了罗蒂哲学思想的核心问题：罗蒂式的实用主义，是否有理论上自我辩护的内在结构？在与罗蒂讨论的同时，我首次在论坛上提出了我已实践多年的“交互超越主义”（Reciprocal Transcendentalism）哲学，现姑且表述如下：交互超越主义哲学,就是在找到两个或两个以上的可言说对象以后，从其中一个的视角审视另一个，然后再反过来，发现过程的互为可逆性和逻辑对等性，从而挖掘出此种对等可逆性的前提条件，达到对可言说的对象之超越；如果这种超越达到了先验层面，在其逻辑极限处被最后表述的，就是哲学家们一直在寻找的作为理解一切的出发点的第一原则。我认为，任何要认真做哲学，决心要“将讲理进行到底”的人，都必须首先接受这种交互超越主义的原则。

这次罗蒂的上海之行由黄勇教授陪同，黄勇读过罗蒂的所有著作，是罗蒂最好的解释者。同时，黄勇也是我的很好的朋友。黄勇和我也偶尔就罗蒂哲学有过讨论，他向与会者介绍了罗蒂的“进步”概念的含义，以及罗蒂对把人的自由作为衡量历史进步的标尺的论证思路。他认为，这个标尺的建立，与跨历史的普遍原则没有关系。但在我看来，对自由的终极价值的论证，只有通过康德的道义论道德哲学的进路才是可能的，而通过功利主义或实用主义的进路是不可能的。而罗蒂，既要反康德，又要承认自由的终极性，难为他了。

------------------------------------------------------- 完 ------------------------------------------------------

附件：我提交的反驳罗蒂的会议文章（可删去这一部分）：

Reciprocal Transcendentalism vs. Rorty’s Pragmatic Naturalism

Zhenming Zhai

Sun Yat-sen University, Guangzhou

    For Rorty, the standard epistemology is representationalist in the sense that it not only attempts to legitimate our claim to knowledge about reality but also to mirror reality by itself.  He takes an antirepresentationalist position because he believes that there is no valid justification for the distinction between the mind that represents and the reality to be represented.  In Rorty’s interpretation, Kant’s epistemology has to be based on the conviction that all the so-called synthetic judgments, a priori as well as a posteriori, represent something that is real.  But a pragmatist like himself has adequate reason to refuse such a conviction.  Accepting Quine’s critique of the “two dogmas,” Rorty rejects the Kantian premise that the mind has it’s own a priori laws distinct from empirical laws of nature.  Thus, taking a naturalist position, he avoids talking about concepts or propositions or judgments, and chooses to speak of vocabularies and discourse that are subject only to causal interpretation like any natural or social facts or events.  As a consequence, he only admits historically conditioned social teleological justification for knowledge claims and, perhaps, normative claims, and declares that Kantian representationalism is impossible and thus has to be replaced by epistemological behaviorism.  In this paper, I will first argue that Kantian transcendentalism is not in any sense representationalist.  I will also argue that the kind of intersubjectivity to which philosophical justification appeals is transcendental, not social. Transcendental intersubjectivity is not representationalist, nor is it naturalist; it is rather the basis for the intelligibility of any argumentative discourse, Rorty’s included. Its acknowledgment, therefore, makes possible a conception of social justification and of cross-cultural dialogue.  If Rorty admits no such transcendental intersubjectivity and insists on his so-called epistemological behaviorism, he would not be able to justify any of his own positions, and thus could not take his own words as having more logical power than a bird’s chirping.  I’ll conclude the paper by arguing that at least part of our philosophical interests are not to be subsumed under any kind of social purpose, while all alleged social purposes should be open to philosophical reflection before they are allowed to lead our social practice.

1.   Kant: Universality Is Not Representational

In one of his recent essays, Rorty discusses his commonality with, and differences from, Habermas.  He thinks that Habermas and himself both distrust metaphysics.  But whereas Habermas thinks that a notion of universal validity free from metaphysics is possible and necessary, he himself thinks that such a notion of universal validity must depend on some kind of metaphysics.  In what sense does Rorty talk about metaphysics?  In my reading of his texts, by metaphysics Rorty must mean something that makes the so-called “representationalist epistemology” possible, namely, a theory of reality which reason as a human faculty is supposedly somehow attuned to in a successful cognitive process.

Curiously, Rorty attributes such kind of metaphysics to Kant repeatedly.  However, Kant does not intend to “mirror” reality in his epistemology.  On the contrary, starting from Kant, the notion of universality has little to do with that of subject-free objectivity, insofar as the concept of the rational being is independent of natural causality.

In his major works, Kant repudiates the assumption that human reason can represent any mind-independent reality, and it is for this reason that Kant interprets the notion of objectivity in terms of universality.  Although Kant still uses the word “objective” here and there, he is clear that objectivity in his sense is nothing more than universality transcendentally determined among all rational beings.  This pertains to the following two points.  Firstly, the universal validity of synthetic judgments does not consist in their alleged state of “mirroring” any reality, but in their connection to the a priori determination of the formal structure of the rational mind.  Secondly, the legislative mind is not a controlling power as interpreted by Rorty, because how the mind works cognitively (pure reason) does not depend on the choice of the will (practical reason) for Kant and therefore is not under the control of the will.  And if the will does not control the way the mind works, the mind, which does the controlling through the will power, can in no way control it either. 

In Kant’s philosophy, if anything can be called reality in the context of epistemology, that would be what he calls the “noumenon,” as opposed to “phenomenon.”  But Kant clearly states that our cognitive faculty has no access to noumenon, and in no way the mind can “represent” it.  As for phenomenon, it is just the totality of experience.  Instead of “mirroring” reality, the mind puts the raw material into a spatio-temporal framework and then categorizes it to form experiences and pass judgments on them.  With regard to natural science, Kant does not think that it “represents” reality, given that only “noumenon” deserves the name of reality.  Instead, he understands scientific knowledge in terms of universal laws.  For Kant, the necessity of scientific laws is based on certain a priori conditions under which we can experience objects in space and time.  But these objects are just part of our experience, not anything to be represented by the experiencing subject.  As we have seen, these a priori conditions are not controlled by the individual mind, but pre-determined and shared by all possible rational cognitive subjects.  Thus, according to Kant, anything that can be viewed as an object is already framed by such an a priori determination, and thus to speak of representing something a priori or a posteriori is a contradiction in term.  In the contemporary context, therefore, we can understand Kant’s notion of universality in terms of transcendental intersubjectivity, insofar as the common structure of all rational minds is understood as intersubjective. 

When Rorty goes further to interpret Habermas’s “communicative rationality,” by means of which universal validity is to be reached, he claims that this kind of rationality is not “a natural human endowment, but a set of social practices.”  Here we witness a false dichotomy between the “natural” and the “social” because rationality transcends both.  There is only causal connection in the realm of the “natural,” and causal plus cultural connection in that of the “social,” but neither the causal nor the cultural can characterize the rational.  Rationality is the agreement between the individual mind’s judgment and its experience structured by the a priori constitution of faculty of understanding, as Kant would put it.  This transcends the natural and the social interpretation, and has to be implicitly admitted by anybody engaged in an argumentative conversation, pragmatic naturalists like Rorty and Quine included.  This is what I call “reciprocal transcendentalism.”  Let us now turn to an analysis of the necessity of this kind of transcendentalism.

2.   Rorty’s Naturalism Undermines His Own Discourse 

For the sake of clarification, let us now suppose that I, a reciprocal transcendentalist, am brought to a society of naturalists, and very soon meet the first naturalist on Mainstreet and talk to him about my philosophical beliefs and the way I act according to these beliefs.  How would he respond? 

Perhaps he would say: "Because there are similarities between your cultural background and mine, we have some common habits, and therefore I can understand what you have said and the way you said it.  But due to the difference between the part of your social life that has contributed to the formation of your philosophical beliefs and the part of my social life that has contributed to mine, you have become a

transcendentalist while I a naturalist."  He may continue to talk about the ways naturalists think and act which they have adopted from their ancestors; or about how their rulers have taught them to think and act so; or about how their religion tells them what their God has revealed.  But as a practitioner of transcendental universalism, I would not be satisfied with what he said, and I would ask: "But who has a better reason for believing what one believes, you or me?"  If he replies: "I do not have a habit of offering any reasons for what I believe, so you keep yours and I keep mine," then he will not have a chance to defend his naturalism.  Since Rorty does seem to have defended his naturalist position, let us tentatively assume that Rorty is not this kind of naturalist.  So let us continue.

No matter what kind of argument he tries to develop, he must suppose that the persuasive force must be based on something that transcends the differences of social

heritages between him and me.  That is, he must assume that the reason he offers must go beyond the causal connections he has exposed.  Why so?  Because he will try to claim that the recognition of the validity of his naturalist belief is required of every subject insofar as the subject lives in that society, and that complying with such a requirement is better than violating it.  But I would ask him: “better” according to what criterion? Then, if he continues the conversation, he has to assume that between him and me there is an intersubjectively recogniable criterion with regard to what counts as "better" as opposed to what as “worse.”  What will be this type of criterion?  It cannot be natural causality, because he knows that he can also find the natural causality for anybody’s using any criterion.  In such a conversation, he is supposed to offer a reason for, instead of presenting the cause of, his adopting a criterion.  Therefore, in order for the dialogue to continue, both parties have to acknowledge the same trans-cultural rationality as the basis for mutual understanding.  In my previous publications, I have called this position on the necessity of transcendental rationality, together with its systematic justification, “reciprocal transcendentalism.”

Rorty’s antipresentationalism is tied up with his conversationalist stand.  He thinks that by getting rid of presentationalism we are no longer responsible for the Truth that mirrors reality, but only responsible for our fellow beings.  For Rorty, mutual respect between different societies relies on open dialogues.  But, if one is a full-fledged naturalist of Rorty’s kind, how can a meaningful dialogue between members of different societies take place?

Certainly, Rorty might reject my reciprocal transcendentalism and insist on his pragmatic naturalism by telling us that he does not intend to use his words to express anything intelligible, but only to agitate us and arouse our certain sentiment.  As he says, vocabularies of both sides, justicatory vocabularies included, are on the same footing with “the anteater's snout or the bowerbird's skill at weaving.”  In view of such kind of naturalism, I just don’t know how justification of any kind is possible at all for Rorty if we adhere at least to the basic requirement of formal logic.  Or, perhaps, he, as an evolutionary naturalist, does not even require himself to be self-consistent insofar as logic is understood by him as no more than another toy being played by some representationalists in the unsettled process of natural selection.  If he thinks that a self-contradictory way of saying things can lead to certain desirable consequences, he would contradict himself as often as he wishes.  In such a case, our philosophical discussion with him would be totally pointless.

After all, we should not take seriously Rorty’s pragmatism, anti- representationalism, historicism, or whatsoever he seems to be advocating, if he does not believe reciprocal transcendentalism.  His naturalism alone would undermine his theoretical discourse altogether.        

3.   The Danger of Pragmatic Naturalism in Social Practice

The meaning of philosophical thinking has long been thought as having its own intrinsic value, apart from the possible consequences of doing philosophy.  When Socrates claimed that “unexamined life is not worth living,” he meant that, among other things, philosophical reflection on socially imposed values before accepting them is a pre-condition for a decent life.  In the Kantian words, human dignity consists in the autonomy of our practical reason, which is a counterforce of heteronomy of natural and social causality.  Therefore, if there are any intrinsic values in human life, understanding the world correctly must be one of them.  Being an intrinsic value, correct understanding of the world is good by itself regardless of what will actually follow from it.  Obviously, Rorty’s pragmatic naturalism cannot accommodate such an understanding of human dignity and the intrinsic value of philosophical thinking, because it does not allow any distinction between autonomy and heteronomy, and between reason and cause.

Rorty starts with a naturalistic epistemology, but when he interprets Habermas’s notion of communicative rationality in terms of social interaction, he extends his naturalism to the realm of normative justification.  That is, in the realm of social practice, the notion of society substitutes that of nature and social causality replaces natural causality.  In such a framework, human persons are viewed as mere slaves of historical contingency with no power of self-determination.  Thus, while thinkers such as Rawls and Habermas go in the Kantian direction for their normative theories, Rorty attepmts to have Hegel speak his post-pragmatist language.

The major problem of defining the meaning of our social life in terms of goal achieving is that it reverses the logic of normative relation of things.  We choose one goal rather than another because we have reason to believe that the chosen one has more intrinsic value, or is more conducive to realizing certain intrinsic value(s).  So we endeavor to achieve a certain goal because we believe it’s more worth pursuing than others.  How can the ultimate value of human life be pursuing a series of goals, if no content of any goals has any intrinsic value?  That is, how can the meaning of our action rest on pursuing something that is meaningless in the first place?  We must find some intrinsic value in the goal apart from the fact it is a goal in order for our goal-pursuing practice to be meaningful.  So we must acknowledge that there are certain criteria by which we can distinguish the good from the bad in the content of a goal, otherwise we would not be able to make sense of our goal-pursuing social life at all.  Therefore, goals are first things that await justification, before they can justify anything else.  So construed, the rationality by which we justify our goals must transcend the historicity of these goals.

In principle, any social teleological assessment must be based on a justified choice of the goals for social practice.  If we believe that protecting the weak and needy is a desirable kind of social practice, but some others believe the opposite, we have a burden of justifying our belief.  Rorty’s appeal to the notion of “social justification” is of no use unless he allows a trans-social justification of the goals set by a society.  More importantly, the action of saving lives or defending liberty, for example, is not taken to serve any social goals separat from the action.  Instead, the action maintains or promotes the intrinsic value of human life directly.  If Communitarianism makes some sense in expaining our social identity, it makes no sense in understanding the normative constraint of our social practice.  Here, Rorty might have taken a wrong side in the Liberal-Communitarian debate. 

But Rorty claims that he is a liberal, and his pragmatism is supposedly tied to his political liberalism.  How could such a connection hold itself?  It could not, unless he gives an adequate account of rational justification in terms of universal validity at the level of transcendental intersubjectivity.

Rorty’s pragmatism may not for the moment lead to any visible destructive consequence in the social setting of North America because Americans do not face drastic changes for the time being, and matters of principle have been taken care of by their Constitution and social structure.  In such a society, one’s believing a wrong philosophy will have little effect on the social life.  You could have converted a great number of people into pragmatic naturalists, and the society as a whole might still operate on some other fundamental principles.  But societies such as China are experiencing constant social-political changes at the basic level, and no fundamental principles have been incorporated into the constitutional system.  Here the major two forces at play are the current political authority and the unruly market.  What would follow if we take a naturalist approach as proposed by Rorty in these societies?

Insofar as social causality is concerned, centralized political power will always tend to expand itself, and create the kind of discourse of “social justification” that does nothing but serve its self-expansion.  The other major force, namely, the market, does not have the capacity of formulating its own goals unless by affecting the political authority’s power structure in a certain way.  In such a case, the so-called “social justification” becomes an inherent component of the coercive power, that it, becomes the self-justification of the self-expansive political power.  Everybody can see the danger of such a notion of justification.

The very function of justification in political philosophy is, however, exactly counter-coercive.  If you still want to understand it as a kind of causality, then justificatory discourse is meant to be a causal power that transcends specific social causality.  Rorty might think that so long as the naturalist discourse is conversational, it cannot be part of the coercive power, even when it’s being used by a totalitarian authority.  This could not be correct, but were it correct, it would certainly have no counter-coercive power, and would play no role in justifying anything whatsoever.

Rorty thinks that his naturalistic vocabularies are more conducive to mutual tolerance between various social groups, but the irony is that the most destructive conflicts are often accompanied by a claim of the correctness of social Darwinism, which is a standard version of naturalism.  On the other hand, anytime social groups in hostility try to make peace to each other, they must at least show a sign of appealing to something that transcends any particular interests of their own group. Such an appeal does not mean that they believe any kind of metaphysics, of course.  It only indicates that reciprocal transcendentalism is a necessary assumption for any inter-social dialogue.

It is unfortunate but not always extremely detrimental that the word “social” has become a catchword in various fashionable thoughts.  But Rorty is not just a regular social ideologist.  He is rather a thinker who has been talking in such a way that he appears to be a judge of all possible thoughts.  He wants to appeal to social standards for measuring rationality itself and leaves only the social standards untouched.  The bottom line is that he believes that no justification is anything but social justification.

Rorty urges us to move away from Kant’s alleged representationalist philosophy and adopts a Hegelian historicism. This means that philosophers should not be concerned with the truth value of any proposition, but with offering new ways to talk about our hopes and fears, our ambitions and prospects, as he says.  Accordingly, philosophical progress is “not a matter of problems being solved, but of descriptions being improved.”(See his “A pragmatist view of contemporary analytic philosophy,” 1999)  I am especially curious about the way Rorty understands the words “description” and “improved” here.  First, how can our discourse have any descriptive function if its only connection to anything else is no more than a causal one?  If a sentence uttered is not categorically different from a yawn in principle, how can the former describe something while the latter doesn’t?  Secondly, if Rorty admits, as he does, that anything can be improved at all, he must also admit that there is a standard by which we can tell the better from the worse across history, as we have seen earlier.  Wouldn’t this defeat his historicism immediately?

After all, discourse, just taken as an act of word-uttering, has little or no causal power, and thus cannot be in any sense justificatory.  Therefore, a pragmatic naturalist notion of social justification is a contradiction in term.  Certainly I realize that Rorty has already confronted around the world many critics who presented arguments similar to what I have presented here, and has responded to them forcefully.  But being forceful is not the same as being correct.  Here again, I suppose Rorty does not care whether or not he is correct, since he rejects the basic logic that warrants the distinction between correctness and incorrectness.  Insofar as his words can arouse verbal or emotional or any kind of responses from his audience, he would be satisfied.  If so, I hope my presentation so far has somehow satisfied him.