Transcendentalism appeared in the 1930s and 1840s.Transcendentalism is a movement among young intellectuals in the United States, which emphasizes the goodness, creativity, andSelf-developmentPotential.Ralph Waldo Emerson A leader regarded as transcendental.Transcendentalism is idealismepistemologyA form of expression of.withMaterialismTheory of reflectionFundamental opposition.It believes that human knowledge is something that precedes sensory experience and social practice. It is innate, also known asTranscendentalism、idealismTranscendentalism.
Ancient Greek philosopherPlatoIt advocates that outside the real world, there is an ideal world that transcends experience, time and space, and exists forever;People's experience is unable to understand the world of ideas;People's knowledge about the world of ideas is pre-existing in people's minds, and they can be activated through acquired learning.Kant, the classical German philosopher, believed that knowledge should be endowed withuniversalityInevitabilityThe category form of "" is innate to the subject, and it exists before experience.
Transcendentalism cuts off the link between people's knowledge (rational knowledge) and sensory experience and social practice, and inevitably denies that knowledge is the same asobjective worldThe connection between reflection and being reflected, thus turning knowledge intoinherentThe material cannot be separated from the material, and there is no objective coordinate to prove false,objective thingsNot based on human will.Transcendentalism isGenius theoryandHeroic view of historyThe theoretical basis of.
Inspection conditions
Announce
edit
Apriorism
existingLiterature pairtranscendental arguments Most of the discussions on "transcendental idealism" (Strawson [1966], [1959]; Stroud [1968];Korner[1966]；Rorty[1971]）。In fact, such a discussion is still following the impulse of transcendental philosophy, so it is still within the transcendental philosophy.Contrary to this route,Zhao Tingyang[2005] provides a general reconstruction of transcendental argument (hereinafter referred to as RZ).On the basis of this reconstruction, the author criticizes the activity mode of transcendental philosophy itself.
However, the reconfiguration of RZ is not without problems.In fact, yestranscendentalDemonstrativelogic analysisIt gains importance in the following "meta analysis" background: what is the relationship between philosophy as a discourse activity and argumentation activity?For a particular philosophy, what is the task of the discourse it provides to complete, and how does the argument in this philosophy ensure the completion of the task, or ensure its discourseEffectivenessOf——How did philosophy become "discourse" through argument?However, the refactoring provided by RZ cannot fully answer these questions, or even lead to wrong answers.The purpose is to try to give a reconstruction that can better explain these problems, and preliminarily determine the general method of logical reconstruction of philosophical argument. Finally, by giving the "evolution diagram" of transcendental philosophy——
Test conditions (1)+(2) RZ are based on one“Demonstration mode+The binary structure of "inspection conditions" appears.The test condition is "a priori demonstrationcore technology ”。Let's start with this core part.
Apriorism
The test conditions in RZ have the following forms: 1. If p is a necessary interpretation of q, then p must meet the following conditions: if a paradox of p is constructed, then p must form a self refutation and therefore is not true. The reason is that it is necessary to constructpreconditionIt is just p, or the paradox of p, that p is a direct proof of the necessity of p;And 2, assuming that p is the only interpretation of q, then p must satisfy: if you imagine any interpretation r different from p, then r cannot satisfy (1), so it is proved that p is the only legal interpretation.
The first problem shared by these two test conditions is that the expression is too weak.In other words, the scope of application is too small.In fact, these two conditions can be regarded as the limitation of "interpretation of q", that is, if condition (1) is met, p will become "necessary interpretation", while if condition (2) is met, p will become "unique interpretation", and if both conditions are met, it will become "unique legal interpretation".However, such inspection conditions cannot be applied to, for example,DescartesI think the argument.Because the p in Descartes' argument, that is, I think, is not the explanation of anything.
At the same time, within their scope of application, the standards they provide are too strong.For example, consider Kant'sTranscendental deduction [1] In the first place, it cannot meet the requirement of "non p is direct proof of p" in condition (1), and it cannot meet the requirement of "uniqueness" in condition (2).
In fact, Kant's transcendental deduction of category aims to prove that category is "useful".This is not to say that all intuitive miscellaneous must be combined with the category, but that if an intuitive miscellaneous becomes knowledge by combining with the category, then it must be combined with the category.The whole proof is to prove the necessity of category through the necessity of apperception.It has the form [2]:
(1) It is known that if the image wants to become knowledge, it must be accompanied by self thinking/apperception.
(3) Therefore, all appearances are accompanied by the same cogito/apperception.
(4) Because the function of category is apperception.
(5) Therefore, the combination with category is necessary for the representation (already so combined)
But if written strictly, it should have such a form,
DC[4]：
Ling is the realm of appearance,Knowledge area, is apperception, F is category set,
(6) Known, [5]
(7) So, ([2])
Apriorism
(8) Since,, if, meets, then, ([4])
(9) Then,, if, meets, there is,
Descartes
From this strict rewriting, we can see that the transcendental deduction of the intellectual category is not used“ApperceptionOnly through category "(category is the only form of knowledge), so it does not meet the condition (2).In fact, the whole proof only needs "all appearances are accompanied by apperception", and if an intuitive multiplicity is indeed combined with a category, then the combination is "equal to" the combination with apperception (6).
However, condition (2) does not excludeKant All prior deduction.In fact, if transcendental deduction is generally understood as proof of "inevitable combination", then apperception/cogito also has transcendental deduction (7).This is also (6) and (7) in the above DC.Namely (1) - (3) in informal expression:
(1) It is known that knowledge needs to be accompanied by an image of self thinking to become knowledge [8]
(2) Moreover, I have identity (uniqueness condition)
(3) Therefore, all appearances are accompanied by the same cogito/apperception
Here, the two attributes ("all" and "the same") in (3) are obtained from (1) and (2) respectively.In addition, it can be seen that the whole proof is built in the DC, that is, the transcendental proof of the intellectual category needs the conclusion proved here.
Demonstration mode
Announce
edit
Demonstration Mode M
The specific form of "demonstration mode" in RZ is as follows:,
Mode M (9):
(10) If p is a necessary prerequisite for q,
Apriorism
(11) Then, q will become like this because of p;
(12) In fact, q is just like this,
(13) Moreover, it is impossible to imagine a q that is not like this,
(14) Then, p is undoubtedly a prerequisite for q,
(15) P is of course true.
Apriorism
The first problem of Mode M may still be descriptive.In fact, if "necessary precondition" and "prerequisite" refer to the same thing, then since p is already necessary for qpreconditionP is obviously a prerequisite for q.(11) - (13) No need to appear again.Secondly, if we regard (11) as "expansion by definition" of (10) (when "necessary preconditions" and "prerequisites" are not the same thing), then this model is not an argument.In fact, if "q like this" is recorded as f (q), the argument mode M can be rewritten as:
[10 '] It is known that p → f (q) (p makes q like this)
[11 '] Also, q → f (q) ("q is not like this, which is unimaginable" means "if there is q, then q is like this")
[12 '] So, q → p (p is a prerequisite for q)
This is obviously an untenable inference.From the perspective of this inference itself, we will naturally think that it would be better if f (q) → p were in [10 '].However, if this is the case, the whole transcendental argument will become a simple one“AfterpartReasoning ": f (q) → p, q → f (q) Γ q → p, there is nothing new from the perspective of logic.In addition, if p is a category, q is knowledge, and f (q) is "knowledge has unity", then we have seen in the previous section that f (q) → p, that is, "if knowledge has unity, then it is obtained through categories", is actually an overly strong condition.
logic analysis
Announce
edit
Empirical argument
Apriorism
Strictly speaking, all the refactoring work mentioned above is only in the sense of rewriting.That is to say, the argumentation process is recounted in logical language.No analysis has been carried out.In fact, for the logical analysis of an argument, its object should be from the known proposition to the proving propositionconstraint conditionThat is, it needs to ask: Under the given conditions, what conditions need to be met to advance to the conclusion?
Several examples of logical analysis: global conditions
Kant's argument
As has been said before, the transcendental deduction of I think/apperception is a corollary from identity to the inevitability of the combination of appearance and apperception.
And the conditions that guarantee the validity of this inference areKant The first sentence in verse 16, "I think must be able to accompany all appearances".That is, [1] in the DC.We call such a condition "global" condition, [1] "global condition [1]" and UC1 [11].
In fact, it should be noted that in the previous section, when we reconstructed Kant's transcendental deduction, we actually quantified the transcendental argument of I think/apperceptionFirst-order predicate calculusWhen reconstructing the transcendental argument for categoryEquivalency。Compared with M in RZ (M is a common first-order predicate calculus), the key here is obviously "quantization".The global condition is just quantitativeLogical systemIn order to be expressed.
However, this analysis is still preliminary.Because Kant's transcendental deduction of "I think" cannot be regarded as the whole content of his transcendental argument.In order to fully illustrate this point, let's first analyze Descartes' self thinking argument and make some preparations.
Descartes Argument
At first glance,DescartesThe "I think" argument is much simpler than Kant's transcendental deduction.That is,
DA：
If I think is doubtable, then I can doubt I think, and I doubt I think is also in my thinking, so I think is not doubtable
Apriorism
Descartes, however, this argument is not Descartes' complete proof of cogito: just proving that cogito exists does not mean that it is the "Archimedes point" that Descartes needs.In fact, the latter part of the work is throughMeditation on the First PhilosophyThe whole first two meditations are completed through (tentative [12]) doubt to exclude all other beings.In fact, only after this exclusion can Descartes say "I am just a thinking thing" after the second meditation [13].
In this sense, Descartes also needs to carry out some "deduction" of cogito, that is, "proof of rights".Whether Kant or Descartes, the proof of this right is actually the proof of the role of cogito in its transcendental system.In fact, DA can only say that it proves my candidacy as the "Archimedes Point".The real proof comes from the exclusion of "display".
Kant's Re argument
Through the preliminary analysis of Descartes, it is natural to ask Kant a question of "parallelism".If the purpose of transcendental deduction is only to prove the right of a certain base point (for Kant, this right means "the inevitability of combination"), then how did the candidate qualification of this base point appear in Kant?
This brings us back to section.In fact, we have seen from our DC that my transcendental deduction ([1] - [2]) is "embedded" in the whole transcendental deduction of categories.In other words, the transcendental deduction of categories is "through" the transcendental deduction of I think.However, this is only the order in which rewritten refactoring occurs.If we focus on cogito/apperception, then the order is just the opposite: the candidacy of cogito is prompted by categories [14] - it is precisely because categories themselves cannot explain their rights that we introduce cogito/apperception, a "possibility of general connection".However, the real candidate is the identity of my thoughts/apperception.Because Kant explains the understanding from this point: "The understanding is nothing more than the ability to place the multiplicity that gives the appearance under the identity of apperception".It is also the "supreme principle" of all understanding.
In fact, Kant's transcendental deduction can be seen as two imputation processes.The first is to use UC1 to classify representation into apperception in the deduction of cogito/apperception, and the second is to classify category into apperception in the deduction of category.A global condition, [3] in the DC, is still used here, which is recorded as UC2.However, the field here is not the representation field Q, but the category set F.In fact, the significance of this transcendental field (namely the formal field) was really determined in the two classifications.
Thus, for Kant's entire transcendental argument, we can give a reconstruction of such a constraint formula,
LRK：
For a given domain E, e is any element in E, and a priori domain F, f is any element in F. If f satisfies the global constraints UC1 and UC2 for e, then the following inference holds:
If f and e are related to R, then f and e must be related to R
For Kant's transcendental deduction, the relationship R and domain E in the above LRK can be interpreted in two ways.One is to regard R as the so-called "combination" relationship, while domain E is a domain composed of intuitionistic miscellaneous. Alternatively, R can also be regarded as a "part" relationship, while domain E is regarded as a "knowledge domain". At this time, "f and e are related to R", that is, "f is a part of e" [15].However, it should be noted that no matter which interpretation is adopted, the relationship R here is not arbitrary, but determined by UC1 and UC2 (because UC1 and UC2 are actually about this relationship).In this sense, the whole inference is not only bound by it in terms of its establishment or not, but also endowed with meaning by them.
Descartes' Re argument
As has been explained previously, Descartes' classification is not completed in DA but in exclusion activities.This exclusion is actually a global exclusion.In other words, this is another form of global condition, which is just to express whether the form is definite or not.The specific form of this condition can be written by modifying condition (2) in RZ,
UC3: E is the whole existential domain, and a is the only element with attribute r in E.
Note that (2) in RZ also calls (1).Our statement is a "pure" global condition.In addition, with Kant's apriorityArgumentation structureThe difference is that UC3 here is a deduction in itself, not a constraint to guarantee another (deductive) inference.However, like UC1 and UC2, its expression also shows the relationship type between classification point (a) and domain (E).In fact, according to UC3, a is the element of E, in other words, other elements in a and E areJuxtaposition relation——It is not like UC1 and UC2, where f is part of the element in E or has a "combination" relationship with the element in E.It is also in this sense that Kant's transcendental field is the formal field, while Descartes' is not.In other words, for a transcendental philosophy, the global condition in its transcendental argument gives the type of its transcendental field.Moreover, obviously, this type is divided based on the relationship between the prior domain and domain E.
The General Mode of Transcendental Argumentation and Transcendental ArgumentationApriority
In fact, what the whole transcendental philosophy tries to determine is the relationship between the transcendental domain T [16] and domain E.The latter may be called the "basic field".Then, the whole transcendental philosophy can be represented by a diagram:
Apriorism
The so-called "conclusion" is, for example, "all appearances are accompanied by cogito/apperception", "the combination of categories and intuition is inevitable", and "what I am means is cogito".These conclusions, as mentioned above, are all obtained through the "classification method" of transcendental argument.However, different ways of classification have a direct impact on how different transcendental philosophies obtain meaning.
The general mode of transcendental argument/transcendental argument can be written in the following form:
If a satisfies a global condition UC, then in the sense specified by UC, a and E have a certain relationship.A is called the inclusion point.
The difference between different transcendental philosophies lies not only in the different points of incorporation, but also in the different ways of incorporation, that is, different UCs.This is exactly the problem to be analyzed in the next section.
A discussion of transcendental philosophy
In fact, transcendental philosophy, if we start from "affinity", can draw a "pedigree map" (from left to right is the historical order):
Apriorism
Moreover, in fact, the standard of "directness" is also different from the standard of transcending problems in the analysis of transcendental philosophy, which is called "nothing"Externalization”Standards.According to this standard, the relationship would be as follows:
Apriorism
In the simplest terms, it is obviously not monotonous ("pedigree" is not monotonous).In other words, the degree of "non externalization" of transcendental philosophy increases monotonously with time, but its directness is far from that: all people are weaker than Descartes,HusserlIt is also weaker than Hegel.In fact, the so-called directness is concerned with the transcendental philosophy as discourse, more accurately speaking, as discourse activity - in this sense, it is actually just the conclusion it obtains through transcendental argument.The transcendental problem is concerned with the transcendental field as "matter".The most obvious example is that Descartes obtained the greatest directness with the weakest degree of "nothing".In fact, for Descartes, I am a parallel being with other beings.However, his "method" conclusions are directly applied to allhuman activity。The problem of Husserl and Kant is that they have taken the route of "objectification", and as a result, the transcendental field has become a purely formal field.The "method" conclusion thus obtained can only be applied to form - category isobjectivityThe "cogito/apperception" of the form of (therefore object), which is explained by the meaning provided by the category, is therefore limited to the formal field.Hegel, Heidegger and Derrida did not have this problem.But their transcendental field itself is obtained by so-called "repetition".The repetition of Hegel and Heidegger is a kind of historical repetition.
Activity mode
Announce
edit
The mode of activity of transcendental philosophy
The core part of philosophical activities can be said to be argumentation.It always gets conclusions through argument, and then produces "influence" through these conclusions.As for the logical analysis of philosophical activities, the key point lies in the analysis of how the results of argumentation activities work as philosophical discourse.It is at this point that the logical analysis of the philosophical activity itself becomes a handy tool for the analysis of the history of philosophy.
In fact, it is through this method that,Zhao TingyangIn his paperGreek philosophyThe difference between (Kant's) transcendental philosophy and (Kant's) transcendental philosophy is said to be the difference in the method of "determining the basis of knowledge".This general conclusion is certainly true.However, from the point of view of logical analysis of philosophical activity itself, this conclusion is not sufficient.In fact, this difference in methods based on the determination of knowledge is rooted in the different understanding of "foundation".Greek philosophy has always believed that the answer to "what is the basis" is meaningful in itself.However, in the activity mode of transcendental philosophy, the purpose of determining a foundation is to get a conclusion about the foundation, and what is meaningful is not "what is the foundation", but exactly this conclusion (and the reason why this conclusion is meaningful is that this "foundation" is the activity of "the point of incorporation").In this sense, the "reflection" of transcendental philosophy on "foundation" has been a direct discourse activity (althoughDirectnessDifferent degrees), rather than some existing discourse activities (such asKnowledge productionActivity) Provide "authoritative" logistics demonstration.It is because of this that we can understandAutocorrelationWhy is it not the fundamental feature of transcendental argument.In fact, only when the foundation is to become the source of authority does it need to become the "ultimate" thing that can be established by itself without resorting to anything else.For transcendental philosophy, as long as it can prove that its "stop" point can "put other things in the transcendental field into itself", it can complete its role.Because of this, the focus of the transcendental argument is always on the relationship between the "entry point" and other things in the transcendental field.And transcendental philosophy, as an activity, is actually a kind of future philosophyRehearsalThis rehearsal is always beyond the transcendental field itself and "embedded" in people's actual activities.
