Edmund Husserl · 埃德蒙德·胡塞尔
The Origin of Geometry · 几何学的起源
Undated 未注日期
The Origin of Geometry — Reading notes
Apple Notes
Revised 30 September 2024
Reading / 研读:Jacques Derrida · Introduction to Husserl’s Origin of Geometry 胡塞尔《几何学的起源》导论
1
Derrida sees the fundamental problem that concerns late Husserl is the problem of history and its status inside phenomenology. This consists in three aspects: 1. why history needs phenomenology to elucidate itself and its eidetic presuppositions; 2. how history, after being elucidated by phenomenology, can become a genuine science, given that it seems to be concerned only with events “very particular”; 3. how phenomenology is intrinsically “historical.”
What is even more troubling here is that in all Husserl’s earlier works, historicity is never essential to phenomenology or any genuine science elucidated through phenomenology. Earlier, science is to be grounded through the egological, and inside the egological through the grounding of logic, or formal ontology. That is to say, logic and experience alone seem to give us a sufficient genesis for a genuine science.
Yet, through examining a particular science such as geometry, a new horizon of historicity is discovered, which remains concealed in the earlier method but is nonetheless essential for any genuine science. In what sense, then, are we to understand the relation between history and science? Why is historicity concealed in the genesis of logic or formal ontology yet unconcealed by an examination of material ontologies? Is there some inversion of priority occurring in Husserl’s thought?
2
The key issue is, what is a phenomenological reduction in terms of historicity? What is a return inquiry that reduces historically and phenomenologically at the same time?
First, we see that the inquiry is not concerned with parts of or facts about geometry, but rather the whole essence of geometry. Second, we see that it is not concerned with the conception or psychology of the first “geometer”, nor the axioms and the science as a whole, nor all the historical facts relevant to geometry, since all these would presuppose a ready-made factual existence of geometry. Rather, what the return inquiry is up to is exactly to determine the original sense of the instituting of geometry as an essential horizon of possibility.
It is obvious that a return inquiry is to reactivate the origin and remove forgetfulness. Forgetfulness is shown as a matter-of-factness, that is to say, everything geometrical is handed down to us as simple facts, and we do not take full responsibility for clarifying them and do not see the intrinsic (not extrinsic) connection between our work and tradition, but simply build on those stacked up facts.
The problem is, through static phenomenological analyses, we already recover, if only partially, our freedom. Through eidetic reduction, we reduce what was indistinguishably presented as a fact to be really a norm, and through phenomenological reduction, even what is already shown to be a norm has its factuality or givenness further reduced to a norm of our own constitution. What then is the status of a historical analysis?
Kant has to situate concepts in the realm of transcendental idealism, so as to reject any ready-at-handness of either historicist empiricism or unwarranted metaphysics. But in doing this, he forgets that concepts themselves, though “irreal”, might still appear to be ready-made within us. To fully reduce the factuality, therefore, Husserl has to push reduction to the limit, and an account of the constitution of transcendental concepts is necessary, therefore, of the complete phenomenological reduction.
Yet still, a genetic account might very well be necessary, but why history in particular? Is history not the factuality par excellence? But this is because we are here dealing with a material ontology, so the factuality is always already here. That is, we are always already acquainted with a factual science handed down to us, and we are trying to determine the sense of what is handed down to us. This has to be historical, for we can very well comprehend how we now get to understand geometry via a genetic account, yet it is impossible to disregard the already determined sense of geometry imposed onto us. The account therefore has to be more than genetic.
The return inquiry is not dealing with factuality as such, but a factuality already reduced eidetically into an essence. The historical fact is an essential fact, a fact that is invariable both because it is a “possible fact” that can be varied in its realm of essentiality (eidetic reduction), but also because it can be reactivated by later acts and therefore is an irreproducible anchor for all later references (return inquiry). In its own way, therefore, the historical horizon of phenomenology as an account of sciences is both dependent and secondary and independent and primary: it is dependent on formal ontology to give a radical eidetic sense to the totality of its “facts” (that history also applies to any object whatsoever, that factuality itself has to be accounted for in transcendental logic), but is independent insofar as formal ontology itself has to originate in history.
3
The origin in question limits the science as a whole. This origin is, however, not the regulative ideal which posits a pure deductivity or axiomatics, but rather has a more originary sense of truth that determines projectively the scope of what is decidable and what is not decidable for a particular science. Any axiomatic system has to confront the incompleteness theorem, yet exactly what relation the particular kind of undecidability has to what is decidable is predetermined by the originary sense of the science.
This primordial ground, therefore, is a horizon of free activities and projects. This means that activity and passivity have to be intertwined, since passivity is intrinsic to activity as its immanent transcendence: science is historically constituted. It is a tradition handed down like any tradition, but in its sedimentation it is always capable of reactivation of its originary sense. This is what historicity means. It cannot be an ideal object being fulfilled in subsequent additions, for in such cases, the ideal object is closed off and muted and without its content impossible as its genesis implies an active reduction from experience, whilst the empirical additions are discontinuous among themselves. Reactivation is never a logical explication.
4
Husserl, in his study of the origin of geometry, is not greatly concerned with the origin of the geometrical sense, but the “sense of sense”, or more simply, the genesis of the ideal objectivity of the geometrical sense, which is to say, the genesis of sense in general, as sense is equated to the primordial evidence of objectivity.
As Dilthey is concerned with the subjective reception of tradition, Husserl is concerned with the subjective creation of a tradition. How does Subject constitute the Objective Spirit in the first place? This has turned the study from the concrete historical a priori to the a priori historical, or, from the particular fact that institutes the whole transcendental horizon to the general conditions for historicity. In this sense, what is characterized as genesis here is rather the “static structure of genesis.” It is fictive in the sense that no geometrical sense ever “precedes” its “objectivation in history”, though this difference can be laid out structurally. This is not contradicting Husserl’s own projection of genesis and development, however, since no such eidetic structure of historicity is possible without a primary concrete history.
5
The question of the transcendental possibility of geometry or that of science as such is then also a question of the transcendental possibility of a “pure language.”
It is to be noted that such a transcendentally pure language is neither the “eidetic language” nor the “transcendental discourse.” This language is the constituting language which makes any constituted, empirically real language possible. This is the language that first allows an object to be constituted in thought through a “voice”, and then be translated to other subjects in a community of intersubjectivity.
Eidetic language on the other hand is the objectivated language with its pure a priori structure. As an objective eidos, it is not yet phenomenologically reduced to the transcendental subjectivity. Though a transcendentally pure language is necessarily eidetic, since it is beyond any constituted product, yet how such a language constitutes its own eidetic structure without presupposing such a structure remains problematic, as is shown in Chapter 6.
But at least the initial step of reduction is obvious. The degrees of ideality of language are laid out as follows: 1. the ideality of a word as always recognized as that one word, which grounds the workings internal to a language; 2. the ideality of a content (essence as Wesen) as always referring to that empirical content bounded in space and time, which grounds the translatability of any language; 3. the ideality of an eidos as unbounded and isolated, which grounds the possibility of science as such. To say that the ultimate ideality of an eidos is linguistic therefore does not mean that it is relative to any factual language, but that it is ultra-translatable, that it is constituted through a pure and transcendental linguistic intersubjectivity.
The other thing is that this “pure and transcendental language” is nonetheless not a “phenomenological description,” or a transcendental discourse, since it is neither “concrete” nor “reflective.” The transcendental discourse aims to reactivate the transcendental language internal to the factical language, so as to also reactivate itself. But this is immensely difficult, since such an endeavor necessarily needs to go into a circle, as it presupposes that this discourse is itself not merely factical but concretely carried out through the transcendental subject, yet its very concreteness relies upon the employment of a factical, hence natural and naive, language, for instance, German. It therefore has to be infinitely self-critical, hence Heidegger’s argument for a transcendental hermeneutics, which moves within the givenness of the language.
Husserl, however, suspects the facticity of language. Rather, he clings to the possibility of a pure transcendental language, in which all facticity as such can be removed, and sees it as evident that it is fundamentally possible to invent a discourse for phenomenology. In other words, equally as Heidegger suspects the primacy of the ideality of language which has to be phenomenologically clarified in its relation to the subject’s historicity, which clarification (as phenomenological discourse) itself relies upon a presupposition of this ideality, Husserl is not convinced that the linguistic historicity is so fundamentally factical that even phenomenology itself becomes limited in a certain naturalness. Any measure of phenomenology’s success, then, has to be the evidence in its own carrying out the task.
6
Derrida criticizes, or rather clarifies, Husserl’s approach to language on three grounds. First, Husserl privileges “normal adults” to be the exemplary of transcendental subjectivity. This is nonetheless consistent, for transcendental subjectivity, as transcendentally reduced, does not necessarily entail an eidetic “abstraction,” not even a reduction to a fixed “essence.” Just as the eidetically reduced language is still but an objectivated language, not the speech of the speaking subject as such. Rather, the transcendental reduction is to reduce a “fact” to a “norm”, an “eidos” to a “telos.” So just as the eidetic speech is not yet spiritual, so the eidetic subject is not yet concrete.
This expresses Derrida’s second point. Though Husserl is quite clearly only interested in the linguistic constitution of ideal objects such as mathematical ones, of the poles of identity or centers of reference, he is not unaware that at the borders of such objective language, an elusive subjective language related to poetic objects is nonetheless at work. Language as a horizon has to allow for a general division between immanence and transcendence, no matter to what degree accordingly. This is why even when an eidetically fixed language as such is possible, there might still be anomalous discourses as produced teleogically in reference to the linguistic “norm.”
Derrida’s third point is that, since language is primarily the horizon of intersubjectivity, it necessarily involves an inquiry of the “ground” of such linguistic communities. This “ground” is the “Earth” on which we live, which we constantly speak of and point to. To inquire into geometry is to inquire into the possibility of language, but also to inquire into the “earth”, the primal ground for all objects as such. Here the originary sense of geometry is implicitly reactivated: geometry is a living science in relation to the earth and the space we live in, thus is reimplicated in the life-world.
7
The problem of origin and its reactivation is marked by the fundamental tension between eidos and telos. Though an origin is always experienced as a fact, it is nonetheless not an empirical fact, but a fact as such. In other words, an origin can only be infinitely reactivated, thus constituting a pure history as such, only through the fundamental act of infinitization, or idealization. An eidos is reached and then chained to writing, which as a pure writing is ideal and transcendental. Such a writing is a pure act and has nothing to do with any factual preservation of writing, but is merely the “literality” of ideas. However, this literality of eidos is nonetheless immanent to it, since any eidos is not only a given Absolute, but also a corporal and spiritual “signitivity.” In such sense, the true sense, though immortalized, has to express itself through its multiplicity, and is in this sense only a telos, that is, eidos as constituted.
The process of reactivation, therefore, as the reduction of historical passivity to recover the originary activity, is the intensified comprehension of an essence as historically constituted. This is to say, it is not merely an inquiry as to the essence of an object, but building off that, an inquiry as to how that essence is to be factically produced, though such facticity is itself an ideal one.
In the case of geometry, therefore, it is first the recovery of the ideal and true sense of geometry from the half-comprehending symbols. Such a reduction has to go two ways, firstly, it reduces the factical contingency to recover an eidos, what geometry itself is given to be in its clear evidence, and secondly, it in doing so has to also reduce the factical necessity to recover a telos, where its evidence is not only true, but also adequate to the absolute transcendental subject. Though in the second reduction, it seems as if a return to other contingent facts is inevitable, it is really not so if one supposes even the factical grounds a transcendental subject makes use of is in themselves pure and transcendental.
In this sense, one can say that Husserl is James Joyce inverted. Joyce is trying to practice equivocation as much as possible to reach that infinite telos as implicating an eidos, Husserl is trying to reduce equivocation as much as possible to have an eidos as attracting a teleology.
8
Again to inquire into any particular origin is to inquire into a particular historicity that is a reduced factuality. In this sense any external empiricism is invalid, unless it is grounded by a prior understanding of phenomenological reduction and an a priori structure of pure historicity.
But even if we do grant pure historicity it is not necessary to admit historicism. For historicism is predicated under a unity of “historicity”, and historicity is a relativism only under the unification of one historical task.
Therefore we can say there is an order of cultures based on three levels. First level is the general historicity immanent to any culture. Second level is the European culture that exemplifies the idea of pure historicity. Third level is the ultimate clarification of pure historicity not as an idea only exemplified or possessed by Europe, but as constituted by Europe as a telos.
9
Any inquiry to a particular origin presupposes the inquiry to pure originality as such. But more than the inquiry of a pure historicity that constructs the objectivity of geometry, this investigation of the origin of geometry has to at the same time account for the constitution of ideality intrasubjectively.
Yet such a particular constitution cannot be without historically determined factualities providing a basis of material for it, that is, providing norms for it. That means, these materials cannot be simply facts, but have to be grounds of possibilities of a science. They have to be eidetic structural stratums that are both concrete and scientific. A different science is required for this task, i.e. phenomenology. Any objectivism (positivism) or subjectivism (Heidegeerianism) is thereby illegitimate.
But since Husserl also uncovers the layer of subjectivity, how then is the pre-predicative stratum of experience related to the historicity? Is pure historicity that constitutes any science itself constituted more profoundly by a transcendental subjectivity? Yet if any ego is fundamentally saturated by historicity, how is such a historicity again constituted by an ego? Self-constituting constitution that first converts all ideal facts back to norms, then discovers how norms interrelate. That is, first, to reduce all factual histories (including the factuality of the European eidos of pure historicity) to potential historicities, the eidos of pure historicity is therefore itself constituted historically, but still constituted transcendentally. Next, to uncover the structure of all these interrelated norms, namely, these potential historicities, including the pure historicity, itself dependent on other transcendental norms, which norms are dependent on historicity as well. Hence self-constituting constitution.
10
There is a disruption or a turning away from between what is geographical and what is geometrical. The “roundness” and the “straightness”, for example, are geographical essences that are morphological types still inherent in sensible shapes, they are anexact but not inexact. That is to say they are vague yet nonetheless essences subject to a “more or less” of eidetic variations.
The first infinitization of geometry is to institute this ideality beyond sensible shapes. This gives us the “ideality of objects.” Derrida is perceptive of the fact that the second infinitization ushered in by modernity is an ideality of “systems as such”, or systemicity, which however is nonetheless founded upon the first institution of a closed system.
For Husserl, any reactivation is necessarily an infinitization, for a reactivation of sense is nothing other than the reactivation of a transcendental intersubjectivity’s constituting acts. It therefore necessarily implies the ideation of a pure historicity and the idealizing transformation of the object under such a pure historicity (or objectivity theoretically, freedom ethically).
This idealization presupposes the practical and theoretical aspect of philosophy. These two aspects, unlike for Kant if taken at face value in his earlier writings, for Husserl are essentially united. To think is to be free. This attitude is nothing but the pure historicity of a transcendental intersubjectivity. This historicity is not itself again constituted by history, but is only constituted by itself throughout history as the teleology of self-constituting Reason. That is why between morphological types and pure ideas there is always a leap, for types are always only hints or indications to infinity, but infinity itself has to be instituted only by infinity itself. Only in this leap is history possible: otherwise facts are only externally chained together, without immanent and intrinsic relation of infinity saturating the finite.
Husserl and Heidegger are different in that Husserl presses for the de jure primordiality of phenomenology (or possibility) while early Heidegger presses for the de facto primordiality of ontology (or factuality).
In this sense the limit of phenomenology is the phenomenalization of the transcendental subjectivity, or the factical analysis of this subjectivity. It is a certainty of form but not the infinite knowledge of content. Intentionality is the horizon-certainty. This is guaranteed through the essential twofoldness of the reduction: a phenomenological reduction might posit a fundamental constituting ego, making pure activity a fact, while the eidetic reduction can reduce this activity to a constituted possibility; while as the essence the subject can be a pure constituted fact, it again can be reduced to a product involving the act of constitution.
11
The Idea in history is the self-manifestation of the telos of Reason. The problem of ahistoricity, historicity and history all united: the atemporality of Idea is omnitemporality of transcendental historicity, and is the pure happening of a factual history. The instituting of a pure historicity through phenomenology (Husserlian) only emphasizes or purifies the ontological problem of Being and History (Heideggerian), or the factical event of Being. The problem of sense purifies the factuality of all its materiality and presents the problem of a pure fact. Phenomenology is a critical discourse that always reflects upon the primordial Fact and is always delayed in expanding and being situated in the horizon: in its delay (différance) phenomenology is primordial in its teleological unity with ontology, it “hears” the primordial Being and understands Being can only be purified into its primordial “show” when phenomenology institutes itself as pure difference in pure difference. Phenomenology is the horizonal critique (that is, the unfolding of a differentiation) of the ontological question of unprethinkable Being. Ontology can only pose the question of this Being qua the primordial Fact, it cannot answer it, since it does not relate it to anything else; only phenomenology has the right to develop it, since it is late to the fact and is this belatedness. This is why Husserl says that phenomenology legislates to philosophy as such. The transcendental and the eidetic have to coincide at the ultimate level. Reduction is a fruitful method that gives birth to Being as such and grounds ontology, whereas ontology clarifies the fundamental question and limit for phenomenology. Heidegger is the other side to Husserl.
本文由 AI 从英文原文翻译。请以原文为准。
1
德里达认为,晚期胡赛尔所关切的根本问题,是历史及其在现象学内部的地位。这包括三个方面:1. 为什么历史需要现象学来澄清自身及其本质预设;2. 既然历史似乎只关乎“极为特殊”的事件,它在被现象学澄清之后,如何成为真正的科学;3. 现象学如何内在地就是“历史性的”。
这里更令人困扰的是,在胡赛尔此前的全部著作中,历史性从不是现象学或任何经由现象学澄清的真正科学的本质。此前,科学要通过自我论获得奠基,并在自我论内部通过逻辑或形式本体论的奠基获得奠基。也就是说,单凭逻辑与经验,似乎就足以给出真正科学的生成。
但通过考察几何学这样一门特殊科学,一个新的历史性视域被发现了。它在先前的方法中保持遮蔽,却对任何真正的科学都具有本质意义。那么,我们应当在何种意义上理解历史与科学的关系?为什么历史性在逻辑或形式本体论的生成中被遮蔽,却在对质料本体论的考察中被揭示?胡赛尔的思想中,是否发生了某种优先次序的颠倒?
2
关键问题是:就历史性而言,什么是现象学还原?什么样的回溯探问,能够同时进行历史的与现象学的还原?
首先,这一探问关心的不是几何学的部分或有关事实,而是几何学的整体本质。其次,它关心的不是第一位“几何学家”的构想或心理,也不是公理和作为整体的科学,更不是与几何学有关的所有历史事实,因为这些都预设了几何学现成的事实存在。相反,回溯探问所要做的,恰恰是将几何学之创立的原初意义规定为一个可能性的本质视域。
显然,回溯探问要重新激活起源、消除遗忘。遗忘表现为一种理所当然的事实性:几何学的一切作为单纯事实传给我们,我们不承担澄清它们的全部责任,也看不到自己的工作与传统之间内在而非外在的联系,只是在这些堆积的事实之上继续建造。
问题是,通过静态现象学分析,我们已经恢复了自由,哪怕只是部分地恢复。通过本质还原,我们把原先不加区分地呈现为事实的东西,还原为实际上是一种规范;通过现象学还原,即便已经被揭示为规范者,其事实性或被给予性也进一步被还原为我们自身构成的规范。那么,历史分析究竟处于什么地位?
康德必须将概念置于先验观念论的领域,才能拒斥历史主义经验论或无根据的形而上学所主张的任何现成性。但如此,他却忘了,概念本身虽然是“非实在的”,仍可能在我们内部显得是现成的。因此,要彻底还原事实性,胡赛尔必须把还原推进到极限;要完成现象学还原,就必须说明先验概念的构成。
但即使生成性的说明确有必要,为什么偏偏需要历史?历史难道不是最典型的事实性吗?这是因为,我们在这里处理的是质料本体论,所以事实性总已经在此。也就是说,我们总已经熟悉一门传给我们的事实科学,并试图规定这份传承的意义。这必然是历史性的,因为我们固然可以通过生成性的说明理解自己如今如何理解几何学,却不可能忽略几何学已经规定好、并加诸我们的意义。因此,这一说明必须超出生成性。
回溯探问处理的不是事实性本身,而是已经通过本质还原成为本质的事实性。历史事实是一种本质事实,是一种不变的事实:一方面,它是在本质领域内可以被变更的“可能事实”,这是本质还原;另一方面,它可以被后来的活动重新激活,因此是一切后续指涉不可再造的锚点,这是回溯探问。因此,现象学作为对诸科学的说明,其历史视域以自己的方式既是依赖的、次要的,又是独立的、首要的:它依赖形式本体论,才能给其全部“事实”以根本的本质意义,即历史也适用于任何对象,事实性本身必须在先验逻辑中得到说明;但就形式本体论本身必须起源于历史而言,它又是独立的。
3
这里的起源限制着作为整体的科学。然而,这一起源并非设定纯粹演绎性或公理系统的规约性理想,而是具有一种更原初的真理意义,投射性地规定了某门特殊科学中可判定与不可判定的范围。任何公理系统都必须面对不完备定理,但某种特殊的不可判定性与可判定者究竟有什么关系,已由这门科学的原初意义预先规定。
因此,这一原初根基是自由活动与筹划的视域。这意味着主动与被动必须相互交织,因为被动性作为活动的内在超越性,内在于活动:科学是在历史中构成的。它像任何传统一样被传承,但在其沉淀中,始终能够重新激活原初意义。这就是历史性的含义。它不可能是一个由后续添加不断充实的理念对象,因为在那种情况下,理念对象被封闭、被静音,而没有内容便不可能成立,因为它的生成意味着从经验中进行主动的还原;与此同时,经验性的添加彼此不连续。重新激活绝不只是逻辑阐释。
4
胡赛尔研究几何学起源时,主要关心的不是几何学意义的起源,而是“意义之意义”,或更简单地说,几何学意义之理念客观性的生成,也就是一般意义的生成,因为意义被等同于客观性的原初明见性。
狄尔泰关心主体如何接受传统,胡赛尔则关心主体如何创造传统。主体最初如何构成客观精神?这使研究从具体的历史先天转向先天的历史性,或者说,从创立整个先验视域的特殊事实,转向历史性的一般条件。在这个意义上,这里被刻画为生成的,其实是“生成的静态结构”。它是虚构的,因为任何几何学意义都不曾“先于”其“在历史中的客观化”,尽管这种区别可以在结构上展开。但这并不违背胡赛尔自己对生成与发展的筹划,因为没有首要的具体历史,就不可能有任何这样的历史性本质结构。
5
几何学的先验可能性问题,或科学本身的先验可能性问题,因此也是“纯粹语言”的先验可能性问题。
需要注意,这种先验纯粹语言既不是“本质语言”,也不是“先验话语”。它是构成着的语言,使任何已构成的、经验上实在的语言成为可能。这种语言首先让对象通过“声音”在思想中被构成,继而让它被翻译给主体间性社群中的其它主体。
本质语言则是具有纯粹先天结构的客观化语言。作为客观的eidos,它尚未通过现象学还原回到先验主体性。先验纯粹语言必然是本质性的,因为它超出任何已构成的产物;但这种语言如何不预设本质结构,却构成自身的本质结构,仍是一个问题,第6章会表明这一点。
至少,还原的第一步是清楚的。语言的理念性可以分为如下层次:1. 一个词总被认作同一个词的理念性,这奠基了语言内部的运作;2. 一个内容,即作为Wesen的本质,总指涉那个受时空约束的经验内容的理念性,这奠基了任何语言的可翻译性;3. eidos作为不受约束、孤立之物的理念性,这奠基了科学本身的可能性。因此,说eidos的最终理念性是语言性的,并不是说它相对于某种事实语言,而是说它是超可翻译的,通过纯粹而先验的语言主体间性被构成。
另一点是,这种“纯粹而先验的语言”仍不是“现象学描述”或先验话语,因为它既不“具体”,也不“反思”。先验话语要重新激活事实语言内部的先验语言,从而也重新激活自身。但这极为困难,因为这一努力必然进入循环:它预设这套话语自身不只是事实性的,而是由先验主体具体实行;但其具体性又依赖于对某种事实语言的运用,比如德语,因而依赖自然且朴素的语言。因此,它必须无限地自我批评;海德格尔由此主张一种在语言被给予性内部运动的先验诠释学。
但胡赛尔怀疑语言的事实性。他坚持纯粹先验语言的可能性,认为在其中一切事实性本身都可被消除,并将为现象学发明一套话语的根本可能性视作明见的。换句话说,正如海德格尔怀疑语言理念性的首要性,认为必须通过现象学澄清它与主体历史性的关系,而这种作为现象学话语的澄清本身又依赖对理念性的预设,胡赛尔也不相信语言的历史性在根本上如此事实化,以至于现象学本身也受限于某种自然性。因此,衡量现象学是否成功,只能依靠它自身实行任务时的明见性。
6
德里达从三个方面批评,或更确切地说澄清,胡赛尔对语言的处理。首先,胡赛尔赋予“正常成年人”以特权,把他们作为先验主体性的典范。但这是一贯的,因为先验还原后的先验主体性,并不必然意味着本质“抽象”,甚至不意味着还原到固定“本质”。正如经过本质还原的语言仍只是客观化语言,而非言说主体本身的言说。先验还原其实是把“事实”还原为“规范”,把“eidos”还原为“telos”。因此,正如本质性的言说尚不是精神性的,本质性的主体也尚不是具体的。
这表达了德里达的第二点。虽然胡赛尔显然只对数学对象这类理念对象的语言构成,对同一性之极或指涉中心的构成感兴趣,但他并非不知道,在这种客观语言的边缘,一种与诗性对象相关、难以捉摸的主观语言仍在运作。语言作为视域,必须允许内在与超越之间的一般划分,无论相应的程度如何。所以,即便一种本质上固定的语言本身是可能的,仍可能出现参照语言“规范”、依目的论而产生的异常话语。
德里达的第三点是,既然语言首先是主体间性的视域,它就必然涉及对这些语言社群之“根基”的探问。这一根基就是我们生活其上、不断谈论并指向的“大地”。探问几何学,就是探问语言的可能性,也是探问一切对象本身的原初根基“大地”。几何学的原初意义在这里被隐含地重新激活:几何学是一门与大地以及我们居住的空间相关的活的科学,因此重新被牵连进生活世界。
7
起源及其重新激活的问题,以eidos与telos之间的根本张力为标志。虽然起源总被经验为事实,但它不是经验事实,而是事实本身。换句话说,只有通过无限化或理念化这一根本活动,起源才能被无限重新激活,从而构成纯粹历史本身。一个eidos被达到,继而被拴系于书写;作为纯粹书写,它是理念的、先验的。这样的书写是纯粹活动,与任何事实上的文字保存无关,而只是理念的“文字性”。然而,这种eidos的文字性仍内在于eidos,因为任何eidos不仅是一个被给予的绝对者,也是一种身体性与精神性的“符义性”。在这个意义上,真实意义虽被永恒化,仍必须通过其多样性来表达自身;因而它只是一个telos,即被构成的eidos。
因此,重新激活作为对历史被动性的还原,以恢复原初活动,是对一个本质作为历史地构成者的加深理解。也就是说,它不只是对对象本质的探问,而是在此基础上,探问这一本质如何在事实上被生产,尽管这种事实性本身是理念的。
就几何学而言,首先要从半懂不懂的符号中恢复几何学理念的、真实的意义。这种还原必须双向进行:第一,还原事实的偶然性,恢复一个eidos,即几何学在清晰明见性中被给予为何物;第二,在这样做时,也必须还原事实的必然性,恢复一个telos,在其中,其明见性不仅为真,也对绝对先验主体而言是完善的。虽然第二种还原似乎不可避免地回到其它偶然事实,但如果假定先验主体所使用的事实根基本身也是纯粹而先验的,就并非如此。
在这个意义上,可以说胡赛尔是颠倒的詹姆斯·乔伊斯。乔伊斯试图尽可能实行歧义,以达到那个含摄eidos的无限telos;胡赛尔则试图尽可能还原歧义,以获得一个牵引着目的论的eidos。
8
再次,探问任何特殊起源,就是探问一种作为被还原的事实性的特殊历史性。在这个意义上,任何外在经验主义都无效,除非它被先行理解的现象学还原,以及纯粹历史性的先天结构所奠基。
但即使承认纯粹历史性,也不必承认历史主义。因为历史主义是在“历史性”的统一性之下被述说的,只有在一个历史任务的统一之下,历史性才是一种相对主义。
因此,可以说文化有一个以三个层次为基础的秩序。第一层是任何文化内部的一般历史性。第二层是体现纯粹历史性理念的欧洲文化。第三层则是对纯粹历史性的最终澄清:它不再只是被欧洲体现或拥有的理念,而是被欧洲构成为telos的理念。
9
对特殊起源的任何探问,都预设了对纯粹起源性本身的探问。但几何学起源的研究不止要探问构成几何学客观性的纯粹历史性,还必须同时说明理念性在主体内部的构成。
然而,这种特殊构成不可能没有历史地规定的事实性,为它提供质料基础,也就是提供规范。这意味着,这些质料不能只是事实,而必须是科学可能性的根基。它们必须是既具体又科学的本质结构层。完成这一任务,需要另一门科学,即现象学。任何客观主义,即实证主义,或主观主义,即海德格尔主义,都因此不正当。
但既然胡赛尔也揭示了主体性的层次,那么经验的前谓词层与历史性是什么关系?构成任何科学的纯粹历史性,本身是否更深地由先验主体性构成?然而,如果任何自我在根本上都浸透了历史性,这种历史性又如何由自我构成?自构成的构成:先把所有理念事实转回规范,再发现规范如何彼此关联。也就是说,首先把一切事实历史,包括欧洲的纯粹历史性之eidos的事实性,还原为潜在历史性。因此,纯粹历史性的eidos本身是在历史中构成的,却仍是先验地构成的。接着,揭示所有这些相互关联的规范,即这些潜在历史性的结构;其中包括纯粹历史性,它自身依赖其它先验规范,而这些规范也依赖历史性。因此,是自构成的构成。
10
地理性的与几何性的之间,有一种断裂或转离。比如,“圆”与“直”是地理性的本质,是仍内在于感性形状的形态类型;它们是非精确的(anexact),但不是不精确的(inexact)。也就是说,它们虽模糊,却仍是可以经受“或多或少”的本质变更的本质。
几何学的第一次无限化,是在感性形状之外创立这种理念性。这给出“对象的理念性”。德里达敏锐地注意到,现代性带来的第二次无限化,是“系统本身”或系统性的理念性,但它仍奠基于一个封闭系统的首次创立。
对胡赛尔来说,任何重新激活都必然是无限化,因为意义的重新激活,无非是对先验主体间性的构成活动的重新激活。因此,它必然意味着纯粹历史性的理念化,以及对象在这种纯粹历史性之下的理念化转变,或者说,理论上的客观性、伦理上的自由。
这种理念化预设了哲学的实践与理论两个方面。不同于从字面上理解康德早期著作的情形,这两个方面在胡赛尔那里本质上是统一的。思考就是自由。这一态度无非是先验主体间性的纯粹历史性。历史性本身并不是再次被历史构成,而只是在整个历史中,作为自构成理性的目的论,由自身构成。这就是形态类型与纯粹理念之间始终存在跳跃的原因:类型总只是对无限的提示或指示,而无限本身只能由无限自身创立。只有在这一跳跃中,历史才成为可能;否则,事实只会外在地串联,缺少无限浸透有限的内在而本质的关系。
胡赛尔与海德格尔的不同,在于胡赛尔强调现象学或可能性在权利上的原初性,而早期海德格尔强调本体论或事实性在事实上的原初性。
在这个意义上,现象学的界限是先验主体性的现象化,或对此主体性的事实分析。它是形式的确定性,而非关于内容的无限知识。意向性就是视域确定性。这由还原本质上的双重性来保障:现象学还原可以设定一个根本的构成自我,使纯粹活动成为事实,而本质还原则能将这一活动还原为被构成的可能性;主体作为本质,可以是纯粹被构成的事实,但它又可以被还原为涉及构成活动的产物。
11
历史中的理念,是理性之telos的自我显现。非历史性、历史性与历史的问题在此全部统一:理念的无时间性,就是先验历史性的全时间性,也是事实历史的纯粹发生。通过现象学创立纯粹历史性,即胡赛尔式的做法,只是强调或纯化了存在与历史这一本体论问题,即海德格尔式的问题,或存在的事实事件。意义问题把事实性从其一切质料性中纯化出来,呈现纯粹事实的问题。现象学是一套始终反思原初事实的批判话语,在扩展视域、置身视域时始终有所延迟;正是在这种延迟,即延异(différance)中,现象学在与本体论的目的论统一中是原初的。它“听见”原初存在,并理解到:只有当现象学把自身创立为纯粹差异中的纯粹差异时,存在才能被纯化为其原初的“显现”。现象学是对不可预思之存在这一本体论问题的视域性批判,即差异化的展开。本体论只能将这一存在作为原初事实而提出问题,却无法回答,因为它不把这一事实与任何其它东西联系起来;唯有现象学有权展开它,因为现象学迟于事实到来,并且就是这种迟到。这就是胡赛尔说现象学为哲学本身立法的原因。在最终层次,先验与本质必须重合。还原是一种富有成果的方法,使存在本身诞生,并为本体论奠基;本体论则为现象学澄清根本问题与界限。海德格尔是胡赛尔的另一面。