Rorty distinguishes two kinds of relationship between conceptions. Conceptions either have a common basis, representing different versions (paradigms) of one and the same type; or they have no such common basis, belong to different types. In the first case - with reference to the common basis - interconceptional argument is possible, in the latter - where such a basis is lacking - it is not.

Within the traditional type of epistemological philosophy, for example, arguments between normal and revolutionary positions are possible. Rorty, drawing support from Kuhn, understands normal philosophy to be the dominant conception at a particular time, which is then confronted by a revolutionary paradigm - which, if successful, subsequently becomes the normal philosophy. Since the normal paradigm and the revolutionary paradigm, with all their other dissimilarities, share the basic premiss that the concern is to render the real as adequately as possible, an argumentative dispute between them is possible with reference to this premiss: the new paradigm propounds reasons as to why the common aim is better attained by its means.(11) In this sense it is possible, for example, for Hegel to argue against Kant, or analytic philosophy against philosophy of consciousness.(12)

In the conflict between epistemological and edifying philosophy, however, it is not paradigms of the same type, but of differing types which confront one another. Edifying philosophy, as opposed to epistemological philosophy, is not only not normal, but even more than revolutionary: it doesn't create a new paradigm within the same basic framework, but rather attempts to leave the old field of play and to engender a new type of philosophy. The elementary premiss of epistemological philosophy - that representing what's real is at stake - is no longer found here. Edifying philosophers, Rorty says, do "something different" to "offering accurate representations of how things are" (M. 370 f). They do not claim that their terms "are the new-found accurate representations of essences" (M. 370). They don't seek "to find objective truth" (M. 377). - As a result of this difference in type argument between such positions is no longer possible.

So much for Rorty's reasoning on the impossibility of argument between typologically different conceptions. I consider these expositions to be sound in principle (later I shall come to speak of certain limitations). Moreover, the argumentative logic of this thesis is not altogether new: "contra principia negantem non est disputandum" was already a proven maxim of philosophical tradition.

To conclude this introductory section by accentuating the positive component once again: Rorty by no means brushes aside argumentative obligations as being burdensome, as is sometimes suggested. He hasn't left his analytic schooling behind him in favour of a cynical relativism. Rather he intends to make clear, with analytic precision, the very standards of possible and impossible argument and to urge that they be taken account of.(13) And he demands strict performative consistency - from himself just as from others.(14) He points out time and time again that philosophers of his school would be performatively contradicting themselves if they were to claim suddenly that their view be the true view, the one corresponding to the essence of things and reality.(15) He criticizes, as examples, Nietzsche and Derrida for having fallen prey to this temptation (cf. C. 8).(16)

3. Problematic consequences for edifying philosophy

Before I come back to Rorty's restriction of the possibility of argument I want to discuss the consequences of Rorty's abstinence from argument for edifying philosophy, the type of philosophy he recommends.

a. Conversation in place of argument

Rorty recommends the transition to this type under the aspect of fruitfulness. Edifying philosophy, he says, will lead to "new, better, more interesting, more fruitful ways of speaking" (M. 360).

I am afraid, however, that abstinence from argument and fruitfulness go together badly. More precisely: it is exactly the exclusion of argument which seems to me to render the remaining practice of edifying philosophy regrettably unfruitful. According to Rorty, it should be a matter of practising "conversation" in place of argument.(17)

But the way Rorty defines conversation reveals a dilemma. He says three things: that edifying philosophers are "conversational partners" for other edifying philosophers (M. 372); that not the discovery of objective truth, but the continuation of conversation constitutes the aim (M. 377, 373); that "keeping a conversation going" in this way represents a "sufficient aim of philosophy" (M. 378).(18) Rorty sketches philosophy as being conversatio perennis.

He does not, however, say what this conversation should actually be about, and what kind of exchange is still possible there.(19) I fear that he would be unable to state this either, for this conversation - in so far as it complies with Rorty's description - is no longer really about anything.

Conversational partners provide one another, at best, with stimulation and a basic encouragement: `have the courage to develop a new, deviant description, one which might even seem mad at first'.(20) But all processes of refutation, falsification, of argumentative discussion and clarifying dispute are eliminated. Hence it is to be expected that this results not, as Rorty says, in "keeping a conversation going" (M. 378), but rather in the opposite: in the silting up of these conversations, in which nothing is really at stake, in peaceable conversational murmurs, in discursive entropy.(21) Rorty's theory of conversations seems better suited to high-brow parlour talk amongst the educated and their inspiriters than to the business of philosophy.

b. Does philosophy mean "just saying something"?

Neither does Rorty shy from obviously minimizing definitions of philosophy. According to him in some circumstances the edifying philosopher "might just be saying something" (M. 371). This is what philosophy is supposed to be from now on. As prime examples Rorty even adduces Heidegger and Wittgenstein (ibid.).

Yet, both of them would turn in their graves, if they had to hear that they had "just said something". Heidegger and Wittgenstein strove life-long with the utmost of exertion for cogent evidence, sound arguments, and reliable alternatives. One can certainly question their value, but not ignore or suppress their clarifying efforts as such. Heidegger and Wittgenstein didn't just talk out of the blue; they weren't the antecedents of the contemporary `That's how I see it, full stop, that's all there is to it, finished; you're welcome to view it differently - so much the better, that way the wealth of human viewpoints grows'.

Heidegger drew attention to the transition which took place with Plato from aletheia to orthotes and saw this as laying the seeds of the "calculative thinking" which has progressed ever since, culminating in the "atomic age" - whose abandonment in favour of "contemplative thinking" was imminent.(22) In all of this his claim was firstly to accomplish argumentatively a relativization of calculative thinking, secondly to develop the concept of a history of being which embraces both calculative and contemplative thinking, and thirdly to present grounds for the distancing from the metaphysical tradition. And when Wittgenstein said that traditional problems of philosophy result from a fundamental misunderstanding and misuse of language,(23) then he was convinced that this could be proved by many points and that this would eventually be one of the reasons why subsequent generations would make the transition to another type of philosophizing. Heidegger and Wittgenstein did not simply play off a new vocabulary against an old one and most certainly did not "just say something", but analyzed and argued insistently.

Rorty could counter that if Heidegger and Wittgenstein thought they had done more than simply play off a new vocabulary against an old one, then they misunderstood themselves. However, it is noteworthy that Rorty doesn't level this accusation, which he does make against Nietzsche and Derrida, at Heidegger and Wittgenstein. For what reason? Heidegger and Wittgenstein did in fact do something other than Nietzsche and Derrida. The latter claim (according to Rorty) to state truthfully how things stand in reality. Since, on the other hand, they declare this to be impossible, they rightly acquire themselves "charges of self-referential inconsistency" (C. 8). Heidegger and Wittgenstein, however, claim only to criticize argumentatively a preceding view. This should indeed be impermissable according to Rorty's demonstration of the impossibility of argument, but once again: Rorty doesn't make this accusation with regard to Heidegger and Wittgenstein. Is it to be possible after all in some cases, in Rorty's own estimation, to argue between heterogeneous conceptions and bring about refutations?

I will now return to the main focus of my considerations, the problem of argument. Using an example I want to show that Rorty himself de facto goes beyond his own restrictive thesis.

4. Rorty's refutation of the epistemological model

In Contingency, Irony and Solidarity Rorty presents a refutation of traditional epistemological discourse, one which to me seems sound. What's wrong is not this refutation, but Rorty's playing-down of its value and penetrative force.

a. Anti-representationalism - beyond realism and anti-realism

Rorty's argument proves that the assumption that our cognition has a representationalistic status is untenable. And not because we have suddenly gained some experience or insight that things in truth behave quite differently and just how things really are, but rather because the representational model contains a conceptual error. It sets out from the assumption that there is first of all a reality, which is then to be represented as accurately as possible in cognition. In this, reality is conceived of as being prior to, outside of, and independent of our reference to it. In the following I would like to call this kind of reality `alpha-reality' for short (in contrast, say, to understandings of reality in the life-world or in individual sciences). Reference to such an alpha-reality is fundamental to the representational model and all its variants, which reach from realism through to antirealism and from criticalism through to agnosticism, and which individually state, with respect to this alpha-reality, that we are able to cognize it authentically, or merely in translation, only approximately or not at all.

Rorty argumentatively demonstrates that the talk of an alpha-reality makes no sense conceptually. Alpha-reality is a pseudo-idea, in respect of which there is neither anything to be known nor to be missed. With this, Rorty's argument lifts the representationalist cognitive model off its hinges at the same time. Naturally you are curious to know how this wonderfully effective argument runs.

b. The basic form of the analytic-philosophical argument: the bind to language games of all reference to reality

The argument is of an analytic philosophical kind. I would like, first of all, to state three of its elements and then to develop it further.

Firstly: truth is a characteristic of sentences, not of objects. Whenever we talk of the true state of objects, we actually mean the truth of corresponding sentences. "This board is black" means that the sentence making the claim is true.

Secondly: sentences belong to vocabularies or language games and are dependent on these for their meaning.(24) The identical-sounding sentence "Peter towered over all" has a different meaning in the anthropometric context than in the historical one.

Now - thirdly - one would, however, like reference to reality independently of the conditions of such language games. Or one would like to make a decision between different descriptions of reality, suggested by different language games, by measuring their degree of adequacy to reality against reality itself. But this cannot be done. It would require language-free access to reality; and there is no such thing. And, even if there were, it would be of no use for the purpose named, since - as a result of its non-linguistic nature - one would be unable to derive from it sentences like "reality as such behaves in such and such a way". Yet it is precisely such sentences that one would need in order, as intended, to speak about reality as such, or to be able to decide between different descriptions thereof.

As such, the boundedness of all reference to reality to language games represents an ultimate limit which cannot be transcended. The idea of a reality-in-itself, an alpha-reality is in stricto sensu impossible, since it can only occur within a language game - but then it is no longer the idea of an alpha-reality, of a reality completely independent of language games. Our talk of reality is always the talk of a "reality-under-a-certain-description" (M. 378).

So much for the basic form of the argument (cf. C. 4 f and 21). In the next step, I would like to discuss an objection, and by doing this get to the core and pure form of the argument.

c. The realistic language game as counter-evidence?

aa. The basic thesis of the realistic language game

Among philosophical language games there are some which make use of reference to an alpha-reality: the language games of the representationalist group. They operate with the idea of a reality which precedes our cognition and exists independently of it. Let us consider, for the sake of brevity, this group's leading paradigm alone, the realistic language game - the result will be apply to the other versions too.

The realistic paradigm states that the task of cognition is to apprehend alpha-reality as exactly as possible, no matter how difficult this might be. Moreover this paradigm insists on the exacting notion that a cognition-independent reality is indispensable for the concept of cognition, since without it cognition would degenerate to mere self-application.

bb. The realistic language game as one amongst others - no chance

of external refutation

What can Rorty make of this counter model, one which - contrary to his analytic philosophical thesis of the impossibility of reference to an alpha-reality - nonetheless operates with such an idea? Now, "conforming to [his] own precepts" (C. 9), Rorty must - as surprising as this might seem at first - recognize this language game. For it is precisely when language games represent an impassable horizon of meaning and argument, as Rorty claims, that one is unable to criticize or refute this language game in the name of another language game or a general insight which overarches language games (cf. C. 8 f). It may well be that other language games get by without the predicate `real', but `real' has an indispensable function for the realistic language game, and one must seek out and discuss the predicate `real' - just as all other predicates too - wherever it makes sense, and within the realistic language game it undoubtedly makes sense.

cc. Immanent criticism of the realistic language game - `alpha-reality' as

an interpretative concept

It is this finding - the requisite recognition of the opposing model precisely under Rorty's premisses - that first makes it necessary to uncover the actual core and penetrating point of Rorty's argument.(25)

Let us consider whether the realistic language game is consistent - whether, above all, it is consistent in its central point, with regard to the use of the expressions `real' and `reality'. These expressions should, as I've said, refer to an alpha-reality, that is to a reality prior to, outside of, and independent of all interpretation. Do they really do this? Obviously not. For these expressions already contain - contrary to their intention and pretension - a particular exposition of the X to which they refer, moreover an exposition which is anything but self-evident.

It is a particular exposition in so far as the truth is claimed of the proposition that this alpha-reality exists prior to, outside of, and independently of our cognition. With this, however, some claims about this reality are already made. That what's concerned here is in no way a self-evident exposition becomes clear as soon as one bears in mind that historically there have been quite different ways of understanding reality, say the ancient thesis of a basic commerciality between reality and cognition, or the apprehension in modernity of an interactionism between cognition and reality. These comparisons show that the talk of an alpha-reality is heavily laden with content - but with this it is elementarily self-contradictory. It claims to talk of a reality prior to, outside of, and independent of each and every interpretation, but by so doing it already undertakes a highly resultful interpretation of this reality. This contradictory trait - the claim that reality, so interpreted, be a reality prior to all interpretation - even constitutes the core of the realistic paradigm. It is just because of this ascribed meaning of being independent of and prior to cognition that the whole emphasis is placed on this reality and that this becomes the measure of all descriptions and any cognition.(26)

To put it another way: the realist does something other than he thinks and claims he is doing. He believes he is pointing to a reality free of interpretation, but in truth he has already unavoidably provided this reality with a certain interpretation. To this extent, the talk of alpha-reality is fundamentally inconsistent. And there is no way of escaping from the dual character of the asserted independence from interpretation and the factual interpretative status - unless one no longer wanted to say anything at all, but one could not then advocate the alpha-thesis, and the whole paradigm simply wouldn't exist.

The realistic position would only be remediable if it were to recognize the interpretative status of the alpha-reality it postulates. It would have to proceed from belief in its reference to an interpretation-free reality, to the understanding that it references a reality interpreted in a certain way (albeit perhaps a very meaningful way). It is only in this modified form that the paradigm would be consistent, whereas in its original form it abrogates itself. (Such a reflected form is found, for example, in Putnam's `internal realism'.) The result of these considerations is not a flat rejection of the realistic paradigm, but rather the call for a conceptually adequate grasp of that position, one familiar to us all and intuitively credible, called realism.

Allow me to add three more observations. First: my development of Rorty's argument goes beyond the standard analytic philosophical objection, according to which we can refer to reality only through propositions, that is within language, so that it is impossible in principle to refer to things transcending language (say, to a reality transcending language) - with Rorty's wording: "[...] no linguistic items represent any nonlinguistic items."(27) This might be right, but is not sufficient. The general condition of sentences - their immanence in language - is one thing. The specific contouring of a certain meaning however - in this case the meaning of the alpha-reality - is something else. The decisive argumentative point is not that the realist can speak of a reality transcending language only within language,(28) but rather that the predicate `reality' used by him inevitably attributes a certain meaning to its referent, whereas it is precisely this which is allegedly not to be the case.

Secondly, I would like to point out once again that this criticism of the realistic language game in no way makes use of a view of how things are in reality. One is not saying that reality is a construction, that it doesn't even exist, or that we are not capable of knowing it. This would indeed be self-contradictory, since the impossibility of such insight is the core of Rorty's argument. Rather, the refutation results simply and solely from an examination of the consistency of the representationalist paradigm.(29)

Thirdly, the argument against the representational model, in spite of its penetrative force, is of course not to be understood as conclusive, one irrefutable for all times. It is quite conceivable that another argument will some day be found which overrides, modifies, or refutes Rorty's argument.(30) One thing can, however, be said safely: that within the framework of those concepts available to us - those developed until now - nothing can be seen to detract from Rorty's argument. In this sense the argument is (for the time being) a valid one, a reliable proposition according with all standards known to us.

d. The cogency and penetrative force of Rorty's argument

What have we established? The consideration of what appears to be its greatest opponent, the realistic language game, results in a confirmation of Rorty's thesis, according to which the idea of a reality exempt from description is untenable. With this, however, Rorty's refutation of the representational model as a whole proves itself to be penetrative. Reality is always - in the realistic language game too - reality-under-a-certain-description. Rorty's argument - which is already found similarly with Wittgenstein, Sellars, Goodman, or Davidson (and which of course, in its content, was already Hegel's argument) - is sound in every respect.(31),(32) Rorty has - at least according to all standards known today - achieved a valid refutation of the representational model.(33)

5. Limits of argument?

Are typologically different conceptions heterogenous in every respect?

However, with a glance at Rorty's conception, this result itself is double-edged. For it might confirm his refutation of the epistemological outset, but it evidently contradicts his thesis about the restricted possibilities of argumentation.

Remember Rorty's declaration that argument and refutation are not possible between typologically different conceptions. But we have just met an example in which Rorty himself convincingly refutes a typologically contrasting position. There must be something wrong with the restriction of argumentation. What?

a. On the design of typologically different conceptions: complete heterogeneity or common features after all?

Rorty is misled by imprecisions and overgeneralizations in his definition of typologically different positions. He stylizes their difference into total diffence. Whereas paradigmatically different conceptions, according to him, by all means have something in common - namely their basis (Locke and Kant, for instance, in spite of all other differences, are to have had the idea of representational cognition in common) - typologically different conceptions (say Kant's transcendental philosophy and Dewey's pragmatism) no longer exhibit a common basis - and hence, Rorty concludes, no longer have anything at all in common. As a result all reciprocal argument between them is excluded.

Now, in so far as conceptions are in fact fully heterogenous, the rejection of argumentative possibility certainly exists rightly. And of course - this constitutes in my eyes the healthy core of Rorty's restriction - all direct basis argument between such positions is impermissible. Wanting to refute the base assumptions of one concept with recourse to the axioms of a typologically different one would be logically incorrect and at best rhetorical.

But it is precisely then that explanation is needed as to how Rorty himself can succeed in refuting a typologically contrasting concept - that of alpha-realism - by way of obviously interconceptual argument.

It was obviously specific linguistic-analytic reflection on the funtioning of concepts (such as is originally alien to realist discourse) which provided the key to refutation here. It is now revealing, however, that Rorty did not simply bring forward axioms of his model against the other approach, but that his refutation succeeded through an examination of immanent consistency of the opposing concept.(34) On the other hand, however, this means that the refutation follows in a manner contradicting both the total difference of typologically varying conceptions which Rorty insinuates and his thesis as to the impossibility of argument which he derives from this. Obviously even conceptions with basic differences are not so totally heterogenous that the impossibility of all kinds of interconceptual argument might, as Rorty claims, result from their dissimilarity.

b. Possibilities of interconceptual argument - right through to fundamental criticism

To put it another way: one must make the concept of interconceptional argument more precise and differentiated. Interconceptional basis-arguments - the refutation of one basis in the name of the content of another - are indeed impossible. Rorty is thusfar right. Interconceptional detail-arguments however - arguments relating to singular assumptions or constituents of a conception - are most certainly possible. This Rorty has overlooked. The fact that fundamentally different conceptions are heterogeneous in their base assumptions in no way means that they must be heterogeneous in all of their elements. Herein lies Rorty's actual error. On account of their different bases he stylized typologically different conceptions into structure heterogenous everywhere and hence (like Lyotard) obscured all intersections and all common features between them. Yet it is obvious that such conceptions also exhibit common features: they have at least logical structures and, equally, general conditions of theory such as coherence and consistency in common. In addition, intersections in content often exist. All this opens up possibilities for argument and criticism even between conceptions with different bases, which on the one side comes from outside, but on the other side - and this is decisive - is internally legitimate and relevant at the same time insofar as it refers to imminent components of the criticized position and takes place in the name of standards which are also recognized by this position.(35)

Such criticism can, moreover, be very helpful in arriving at a more reliable version of the conception in question. It can of course even necessitate the reformulation of leading elements - right through to the reformulation of the base assumption. Rorty's refutation of alpha-realism provided an example of this.

To summarize: since typologically different conceptions, notwithstanding all basic differences, still have some common elements (logical, cross-theoretical, content elements) a whole range of interconceptual argument remains possible - beyond coarse basic arguments, which Rorty, wrongly, considers alone but, rightly, rejects. They can, moreover, extend as far as might be wished, namely right through to (merely non-direct, but, so to speak, indirectly effected) refutation of base assumptions.

6. Departure from claims to truth or differentiation between types

of understanding of reality?

Rorty would presumably not agree with my reconstruction. He would not like to admit that his objections to the representational model amount to a refutation in the strict sense. He would rather have us believe, that he had only played off one vocabulary against another (cf. C. 73). Such restraint is personally honorable, but to me it seems more than problematic philosophically. - With this I come, in conclusion, to the subject of truth - or to Rorty's under-determination thereof.

a. Departure from the adequationist understanding of truth

Rorty rightly avoids the claim that the anti-representationalism he advocates represents the truth, that his "sort of philosophy corresponds to the way things really are" (C. 8). It is exactly such an idea of truth that has had the ground removed from beneath its feet. But does this also mean that truth is no longer to be spoken of at all in the new context? This would only follow if the concept of truth were bound exclusively to the adequational model. But there is no reason for this assumption. If a particular formulation of truth - here, the adequationalist one - proves itself to be untenable, then it is in general replaced by another. The adequationalist concept of truth is not the only one possible: pragmatic, coherence or consensus based etc. concepts of truth are possible in the same way. One can very well make the transition from representational to edifying philosophy - but one should then also clarify which altered contours truth assumes in this transition.

b. Explication of an altered understanding of truth - Rorty's conservatism and its shortfalls

Rorty, however, neglects this task. He equates the departure from the representationalist idea of truth with the departure from truth altogether. He says we should cease to "see truth as a deep matter, as a topic of philosophical interest" (C. 8).(36) This seems to me insufficient. This consequence would result, as I've said, only if the objectivist apprehension of truth were to represent the only possible understanding of truth. It is almost paradoxical that Rorty, on the issue of truth, continues to follow the stipulation of the adequationalist position refuted by him. - It is for this reason alone that he can equate the breakdown of this position with the abandonment of truth altogether - that is, throw out the baby with the bath water.

De facto of course, Rorty cannot help but raise claims to truth. The whole business of his critical explanations shows that even in the context of edifying philosophy one doesn't just say or accept arbitrary things. Besides which, Rorty himself once said the edifying philosopher "agree[s] with Lessing's choice of the infinite striving for truth over `all of truth'" (M. 377). This is something quite different to the straightforward abstinence from truth . On the other hand, Rorty thinks he can spare himself a clarification of his understanding of truth through the sweeping equation of the idea of truth with the adequation theorem. Recently he has been fond of declaring straightforwardly that truth is not an interesting subject for the pragmatist - something which permits no philosophically fruitful work.(37)

c. Argument and truth as persisting constituents of philosophy

Well, I cannot unfurl here the whole subject of truth, but two things can be said with certainty. Firstly, there is sufficient room for a whole series of intermediate positions between Rorty's downright abandonment of claims to truth on the one side and the old concept of absolute objective truth on the other; secondly, things true can obviously still be said, at least in the sense of being more true - of being argumentatively more correct and superior. So there is cause not to abandon the issue of truth, but far more to take it up again in more precise rational conditions.

I wanted to show the extent to which some of Rorty's arguments are better than his declarations and deny his post-philosophical claims. Today still philosophy does not have to silt up in a conversatio perennis, but can continue to be an exciting field of argument and endeavours for truth.

Translation by Andrew Inkpin (Magdeburg)

1. Cf. the reader The Linguistic Turn: Recent Essays in Philosophical Method, edited by Rorty in 1967 (Chicago and London: The University of Chicago Press 1967). Admittedly Rorty had already formulated a detailed critique of analytic philosophy at the end of his introduction. As a result the volume became a point of departure for post-analytic schools of American philosophy.

2. Richard Rorty, Philosophy and the Mirror of Nature (Princeton, N.J.: Princeton University Press 1979); abbreviated to M. with page number in the following.

3. Cf. Rorty & Pragmatism, ed. Herman J. Saatkamp, Jr. (Nashville & London: Vanderbilt University Press 1995).

4. I have discussed Rorty's position in more detail in my Vernunft. Die zeitgenössische Vernunftkritik und das Konzept der transversalen Vernunft (Frankfurt a.M.: Suhrkamp 1995, stw 1996), pp. 211-244.

5. Cf. especially Richard Rorty, Contingency, Irony and Solidarity (Cambridge: Cambridge University Press 1989); abbreviated to C. with page number in the following.

6. Thus Rorty says: "a pragmatist theory about truth [...] says that truth is not the sort of thing one should expect to have a philosophically interesting theory about. For pragmatists, `truth' is just the name of a property which all true statements share. [...] Pragmatists doubt that there is much to be said about this common feature." "Pragmatists think [...] that there is no interesting work to be done in this area" (Richard Rorty, Consequences of Pragmatism, Minneapolis: University of Minnesota Press 1982, pp. xiii and xiv). - Hence one should no longer waste one's efforts on the issue of truth, but work instead towards the establishment of a post-philosophical culture (cf. ibid., pp. xiii f. and xxxvi-xliv).

7. Hence, too, his appeal for a post-philosophical culture.

8. In doing this I will, for reasons of space, try to piece together an account of Rorty's position from several sources in such a way that I give them a clearer or more tenable form from the start, without setting out in detail the steps which lead from analysis of the text through to this distilled form. Rorty's utterances seem to me not always to be consistent. My account hence provides a reconstruction of Rorty's position rather than reproducing the wording of his utterances. The difference between this reconstruction (which, I believe, is completely in Rorty's spirit) and my criticism will, however, remain clear.

9. "This sort of philosophy [...] says things like `try to ignore the apparently futile traditional questions by substituting the following new and possibly more interesting questions.' It does not pretend to have a better candidate for doing the same old things which we did when we spoke in the old way. Rather, it suggests that we might want to stop doing those things and do something else. But it does not argue for this suggestion on the basis of antecedent criteria common to the old and the new language games. For just insofar as the new language really is new, there will be no such criteria." (C. 9)

10. Wittgenstein had already said: "In a certain sense I canvass for one style of thought and against another." (Ludwig Wittgenstein, Vorlesungen und Gespräche über Ästhetik, Psychologie und Religion, Göttingen: Vandenhoeck & Ruprecht 21971, p. 55, no. 37)

11. For the sake of clarity I will go slightly beyond Rorty's statements here. Rorty tends to present paradigmatic differences as being just as incisive as differences in type - at least with regard to their results. Thus he speaks, for example, of incommensurability and irreducibility, without further differentiation, in relation to the difference between revolutionary and normal paradigms within a type as well as in relation to the difference between types (cf. M. 388). Another case of such uncertainty is that Rorty once describes (imprecisely) both edifying and paradigmatically revolutionary philosophers as revolutionary philosophers (M. 369), whereas a little later (precisely and correctly) he contrasts the type of the edifying philosopher with the type of the revolutionary philosopher. The decisive point is that more argument is always possible between paradigmatically differing positions than between typologically differing positions. Rhetoric may always be in play, but in the first case it is not everything, whereas in the latter case it seems that it must be everything.

12. But it must not be forgotten that not every new position is eo ipso right as a new position, rather this is first to be sounded out argumentatively. Also it should not be thought that a dispute once historically decided cannot be taken up again - possibly with a different outcome.

13. Rorty reminds not only systematic, but in particular edifying philosophers too, not to ignore the argumentative preconditions outlined and not to pass off their position as the true one.

14. The argument is naturally directed against systematic philosophers too, who only know their own type of philosophizing and hence, when faced by edifying philosophers, demand from them unjustified arguments - such as would only be possible within the systematic type.

15. Cf. M. 377; C. 8. See also the general criticism of the schema of "seeing" (M. 371).

16. Nietzsche often seems to think that his criticism of belief in reality does in fact state the way reality in truth is: "NB. Appearance, as I understand it, is the real and sole reality of things" (Friedrich Nietzsche, "Nachgelassene Fragmente. Juli 1882 bis Herbst 1885. 2. Teil: Frühjahr 1884 bis Herbst 1885", in: Nietzsche, Sämtliche Werke. Kritische Studienausgabe in 15 Bänden, eds Giorgio Colli and Mazzino Montinari, Munich: Deutscher Taschenbuch Verlag 1980, vol. 11, p. 654). With Derrida it is, in my opinion, different. He does not claim, as Rorty suggests, that there is no reality of the kind claimed by metaphysicians, but that metaphysicians' statements in this repect could only superficially lay claim to such being, that precisely viewed however they would prove forced to attribute this being with a sense other than that they believed. Derrida speaks - like Rorty - not about reality, but about modes of reference to reality (i.e. about the law of `differance' to which all such reference is subject).

17. Within edifying philosophy argument is only possible still within a paradigm - for the purpose of increasing or establishing its consistency - but no longer between alternative paradigms. Although they too have - as previously the paradigms within the epistemological type of philosophy - a common basis, and, according to what has previously been said, this seems to allow the expectation that argument between them should be possible. But this expectation is wrong: the common basis of the versions of edifying philosophy states precisely that no claims regarding an accurate account of reality can be raised, and this does away with the possibility of dispute over right and wrong, and hence the sense of argument. The versions of edifying philosophy can argue neither among one another, nor in relation to the alternative type of epistemological philosophy. For just this reason Rorty consistently suggests another model of interaction: that of "conversation".

18. In this Rorty even sees the "philosophers' moral concern" (M. 394).

19. Rorty attempts to define edifying philosophy as the love of wisdom; yet in this the decisive role is played, of course, not by the love of argument, but "as the practical wisdom necessary to participate in conversation" (M. 372). Rorty, however, provides not a positive, but only a negative definition of this practical wisdom: it is acquired not "in finding the correct vocabulary for representing essence", and it seeks to prevent "conversation from degenerating into inquiry, into a research program" (ibid.).

20. The only precise form in this exchange consists in what Rorty calls "hermeneutics with polemic intent" (M. 365). This is certainly an interesting example of the (legitimate) kind of reference to other conceptions which is still possible now. But it cannot lead to refutations, but only to proposals and suggestions, there is no more argument. Reciprocal hemeneutics goes further than "hermeneutics with polemic intent" - which seeks only to set out "how the other side looks from our own point of view" (M. 364). Rorty undertakes such reciprocal hermeneutics in, for example, "Overcoming the Tradition: Heidegger and Dewey" (in: Consequences of Pragmatism, pp. 37-59): "In what follows, I propose to offer sketches of Dewey as he would presumably look to Heidegger and of Heidegger as he would presumably look to Dewey" (42).

21. Toulmin sarcastically notes: "Reading Rorty's essays, we carry off the image of a party of ex-soldiers disabled in the intellectual wars, sharing, over a glass of wine, memories of `old, forgotten, far-off things, and battles long ago'" (Stephen Toulmin, Cosmopolis. The Hidden Agenda of Modernity, New York: The Free Press 1990, p. 10).

22. Cf. Martin Heidegger, The Principle of Reason (Bloomington & Indianapolis: Indiana University Press 1991).

23. "For philosophical problems arise when language goes on holiday." (Ludwig Wittgenstein, Philosophical Investigations, New York: Macmillan 1958, 19e [38]) "The confusions [...] arise when language is like an engine idling, not when it is doing work." (ibid., 51e [132]) "When we do philosophy we are like savages, primitive people, who hear the expressions of civilized men, put a false interpretation on them, and then draw the queerest conclusions from it." (ibid., 79e [194]) "[...] the clarity that we are aiming at is indeed complete clarity." (ibid., 51e [133]).

24. This was elaborated, in particular, by Lyotard using the distinction between "sentence regimes" (régimes de phrase) and "kinds of discourse" (genres de discours) (Jean-François Lyotard, Le Différend, Paris: Minuit 1983).

25. Which is why - unlike Rorty - I have included the realist objection in the run of the argument.

26. So as not to be misunderstood: I am not saying that this interpretation of independence is - so to speak, objectively - false, I am simply pointing out that - in contrast to its self-understanding - it already contains a certain interpretation, exegesis and understanding of alpha-reality. Without some understanding or other it would be quite impossible even to speak of reality; one could have no idea thereof and would be unable to use the term `reality', which would then also have only some meaning or other - as vague as this may be.

27. This, according to Rorty, is a basic thesis of his "antirepresentationalism". (See Richard Rorty, Objectivity, relativism, and truth, Philosophical Papers, vol. I, New York: Cambridge University Press 1991, p. 2.)

28. This is formulated by Rorty, for example, as follows: "We have to drop the notion of correspondence for sentences as well as for thoughts, and see sentences as connected with other sentences rather than with the world."(M. 371 f.).

29. It doesn't make propositions about reality, but analyzes the status and the possibility of such propositions. Rorty discusses - as does Derrida in my view - not reality, but a mode of reference to reality. An argument does not make a statement about reality as such, but about the conditions, misunderstandings and the correct understanding of all propositions in which we use predicates like `world', `real', `reality' etc.

30. An example of this requisite awareness of provisionality can be found towards the end of the Mirror of Nature when Rorty writes: "It may be that mirror-imagery and `mainstream,' systematic philosophy will be revitalized once again by some revolutionary of genius" (M. 393).

31. In the introduction to his Phenomenology Hegel showed for the first time the contradictions in which the idea of cognition as a "tool" or "medium" entangles itself if it is to serve to grasp the absolute. He explained that the aspects of Being-by-self (An-sich-sein) and Being-for-another (Für-ein-anderes-sein) both fall within consciousness. This was, from the basis of the philosophy of consciousness, a form of expressing the view that - whenever we speak of an alpha-reality - we can do nothing other than conceive this by way of a construct of consciousness (an idea). (Cf. Rorty's reference to this in "Representation, social practise, and truth" [in: Rorty, Objectivity, relativism, and truth, pp. 151-161, p. 158] as well as his portrayal of the young Hegel as a thinker who founded "an ironist tradition within philosophy" whose "practice undermined the possibility of the sort of convergence to truth about which the older Hegel theorized" [C. 79]; it's exactly the Phenomenology which Rorty understands as being "a paradigm of the ironist's ability to exploit the possibilities of massive redescription" [C. 78].) Analytic philosophy has certainly developed this argument with more clarity and sharper conceptual means than Hegel - but the insight as such was already Hegel's. - Furthermore it is nothing short of a piece of objective irony that analytic philosophy thus finally arrives at a realization which, 200 years previously, had already been a basic insight of that very philosopher whom the analysts least of all believe capable of things meaningful. And, on the other hand, this insight has painful consequences for analytic philosophy itself. It necessitates the departure from the very project, for the sake of which analytic philosophy had embarked upon its clarifications. It had set about the "logical analysis of language" precisely in order to make it the sharp instrument which - in place of the vaguenesses of the traditional philosophy of consciousness - ought finally to be suited to the clean and efficient execution of the project of a precise rendering of reality - the old project of representation. A sharpening of these means was indeed attained, but at the same time the untenability of the project was discerned. It is just this which Rorty makes clear in his Mirror of Nature. In this sense he states elsewhere that he intends to argue for the view "that the internal development of analytic philosophy had led that movement to conclusions which contradicted those of its founders, conclusions which sound more like James than Russell, and much more like Dewey than like Frege or Moore" (Richard Rorty, "Comments on Sleeper and Edel", in: Symposium on Rorty's Consequences of Pragmatism. Transactions of the Charles S. Peirce Society, vol. 21, 1985, pp. 39-48, here p. 39 f.).

32. One could indeed ask whether this insight (in constrast to Rorty's construction of relationships in the Mirror of Nature) had not already belonged to the whole endeavour of modern age thinking. Already Descartes's new science aimed not to copy a preexisting reality, but at a construction of reality. Kant apprehended our reality as a construction within the framework of transcendentally given prerequisites (forms of intuition and categories). Spectacularly this led finally to Nietzsche's thesis that the human is above all an animal fingens - a being which creates and is confronted by fictions.

33. Two other ways in which Rorty formulates his rejection of an interpretation-free reference to reality are his rebuttal of assumed "essences" (e.g. M. 367, 370) and of the assumption of an "intrinsic nature" (e.g. C. 4, 8, 9) These are, so to speak, more traditional and thematic versions (relating to the world, humanity, or society) of the subject `reality' and, as such, other varieties of realism. These would also demand access to an entity lying beyond the diverse descriptions.

34. This applies at least, as I've said, for the specified form of the argument presented here. It is not the standard analytic objection, according to which we can only refer to reality within language, but the proof of the inconsistency in content of the talk of an alpha-reality which first constitutes the legitimate and penetrating counter-argument. In my view Rorty has this point in mind, even though he occasionally seems only to argue in a standard analytical manner.

35. Rorty rightly says that the usual expectation that central components of a vocabulary could be shown to be "`inconsistent in their own terms' or that they `deconstruct themselves'. But that can never be shown" (C. 8). If an inconsistency of this kind existed then the vocabulary or paradigm concerned would be so non-functional that it would, in all probability, be unable to establish itself for any duration at all. But Rorty overlooks the other possibility that such untenability can in some cases be recognized from outside and then also more or less demonstrated internally. - Hence there are often good reasons as to why an older position is dropped once a new one has been developed. This must be explained not in a straightforwardly behavioristic manner - as a transition from one habit to another - but in a falsificationist one at the same time.

36. This passage, which is characteristic of Rorty's strategy, amounts to equating the abandonment of the idea of objective truth with a move away from the idea of truth altogether (that is, it neither presents this abandonment as ultimately argumentative, nor allows the ground thus arrived at to appear as a ground on which claims to truth are still possible): "To say that we should drop the idea of truth as out there waiting to be discovered is not to say that we have discovered that, out there, there is no truth. It is to say that our purposes would be served best by ceasing to see truth as a deep matter, as a topic of philosophical interest, or `true' as a term which repays `analysis'. [...] But this claim about relative profitability, in turn, is just the recommendation that we in fact say little about these topics, and see how we get on" (C. 8).

37. See footnote 6.

[Top of page][Wolfgang Welsch Homepage]

Document last updated 22st January 1999