A Problem for the Kantian-style Critique of the Traditional Metaphysics
For centuries, metaphysics was one of the most respected disciplines. During the modern era and especially during the 20th century, the possibility of this field of study became the subject of doubt. Some claimed that even if there is a metaphysics, it is not in fact about the being as being but about our concepts of being. Other critics proclaimed this field to be highly speculative and they treated the metaphysical statements as meaningless. Finally, there was a group of authors contending that our discourse about alleged reality is not only carried out in language, but moreover, it cannot transcend the language. Therefore, they continued that the language should not be viewed as an unproblematic means how to get access to reality but as a socially constructed phenomenon based on the existing power relations. Furthermore, the concept of extra-linguistic reality is highly problematic and so is the metaphysics itself.
Despite all the criticism there are authors who still believe that metaphysics could be a general discipline studying what there is and what is even more important that it could be carried out. They are convinced that most of the objections of the critics are not well founded and they do not really force them to abandon this time-honored discipline.
In this paper I will try to outline some of the objections directed against the traditional metaphysics. Despite the fact that the criticism of the traditional metaphysics has some popularity, the arguments of the critics do seem to have their own problems. Therefore, I will try to show that if one embraces one of the most popular criticisms of the traditional metaphysics (to the effect that metaphysics cannot transcend the conceptual framework and study reality as such), one may end up in espousing a contentious epistemic position (skepticism or infinite regress or dogmatism). I will conclude the paper with a consideration of how to grant the possibility of a certain type of metaphysics and yet to avoid the undesirable consequences.
The structure of the paper should be as follows. The first section will present the criticisms of the traditional metaphysics. The second one will focus on the unacceptable consequences of the Kantian-style objection against the traditional metaphysics and the final one will consist of the concluding remarks concerning the possibility of doing metaphysics. Let me finish the introduction by a brief note on methodology. While the paper is concerned with a metaphysics it approaches the issue from the epistemological point of view. The paper does not engage in doing metaphysics but it tries to analyze the possibility and nature of doing metaphysics, i.e. of a metaphysical knowledge.
1. Criticisms of the Traditional Metaphysics1
At least since Aristotle one group of authors conceives of metaphysics as a discipline studying what there is. In fact, it is not viewed as focusing on a peculiar area of the existing reality; on the contrary, it is supposedly the study of the existing reality in general. In Aristotle’s words (Book IV. of his Metaphysics), it is the study of being as being. I will call this type of metaphysics – studying the extra-linguistic reality in its general aspect – “traditional metaphysics”.
One of the most popular attacks on the traditional conception of metaphysics comes from the Kantians.2 According to them humans approach reality only through some a priori acquired concepts or forms. Kantians claim that people do not get to know reality as it is but only as it appears to them. Therefore, traditional metaphysics trying to transcend the human view of reality and attempting to uncover the reality as such is impossible. It is allegedly futile to speak about what the being is really like, because we humans do not have a direct access to its nature. At most we can study the conceptual frameworks or schemes that enable us to get as close as possible to knowing reality. Thus, on this conception, the only possible metaphysics is the study of the conceptual schemes.
It is possible to illustrate the main point of the Kantians by the following example. Causation is one of the familiar metaphysical topics. According to the manipulation account of causation, a cause-event is a kind of handle by manipulating of which the effect-event is modified as well. In short, a change on the cause side is followed by a change on the effect side.3 Now the interesting questions from the point of view of metaphysics are: Do the adherents of the manipulation account say something about the reality? Do the “causes-as-handles” belong to the repertoire of the world itself? Kantians would probably want to claim that the proponents of this account do not really characterize events as such (their manipulation-friendly nature so to speak) but only our way of thinking and speaking about the reality. They allegedly attempt to point out that in our conceptual framework there is a close link between the concept of cause and manipulation. On the Kantian view, metaphysicians studying causation do not examine extra-linguistic reality itself but they do look for the concepts suitable for the analysis of our notions of cause and effect.
The next criticism comes from empirically-minded philosophers like David Hume and the logical positivists (see e.g. Ayer 1936). According to them, all trustworthy knowledge comes from the sense perception. As a consequence, besides the analytic knowledge, only the statements which are somehow reducible to sense experience deserve to be taken seriously. Traditional metaphysics, on the contrary, tries to uncover what is behind the mere sense experience, what transcends the appearance. But this transcendent reality is behind the reach of our senses and so the metaphysical statements are behind any possible (empirical) control. Therefore, this type of study is unverifiable and highly speculative. On this basis empiricists distinguish between sensible and meaningless discourse. Analytic statements of logic or mathematics and synthetic statements of empirical sciences fall into the former while the metaphysical claims fall into the latter group. They are meaningless because they cannot be verified by experience and experiments. As Rudolf Carnap puts it:
“(Meaningful) statements are divided into the following kinds. First there are statements which are true solely by virtue of their form (“tautologies” according to Wittgenstein; they correspond approximately to Kant’s “analytic judgments”). … Secondly there are the negations of such statements (“contradictions”)… With respect to all other statements the decision about truth or falsehood lies in the protocol sentences. They are therefore (true or false) empirical statements and belong to the domain of empirical science. Any statement one desires to construct which does not fall within these categories becomes automatically meaningless. Since metaphysics does not want to assert analytic propositions, nor to fall within the domain of empirical science, it is compelled to employ words for which no criteria of application are specified and which are therefore devoid of sense …” (Carnap 1959, 76)
All this reasoning results in a final advice – reputable scholars should avoid any speculative and meaningless metaphysics. Or to quote David Hume’s closing paragraph of his Enquiry Concerning Human Understanding:
“When we run over libraries, persuaded of these principles, what havoc must we make? If we take in our hand any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion.” (Hume 1910)
Finally, there is a postmodern criticism of metaphysics. This critique may be articulated as follows. The usual common-sense view has it that people use language as a means for speaking about the world, as a means for representing the reality. The problem is how to conceive of this relation between language and extra-linguistic reality. People cannot step out of the language and grasp reality without the language. We seem to be bound by the language and all we can do is to stay within its limits. To illustrate the point I will quote an adherent of the postmodern philosophy of history discussing the importance of Derrida’s view that “there is nothing outside of the text” for history:
“For while historical representations ostensibly refer to a past outside of themselves, that past, in the very process of becoming historicised (theorised, constructed, interpreted, read, written …) loses its “pastness”, its radical alterity to us, and becomes totally textual, totally “us”. The only way that the past can become capable of being analysed historically is for it to become “historical”. It is in this sense that theoretical textuality/intertextuality thus “goes all the way down” such that the truth of the past per se not only eludes us but makes no sense as a concept. Without “us” the past is nothing; …” (Jenkins 1999, 50)
If we are in such a position does it still make sense to speak about something extra-linguistic? Is it really the case that our language represents something language-independent? What if the language does not mirror some external reality but only follows the rules stipulated by a certain society? What if the language tells us more about our social structure and power relations than about some unreachable language-independent world?
Not only do postmodern critics doubt that there is a way how to step out of the language and how to get to know being as being; in addition, they hesitate to admit the existence of the language-independent reality. If we cannot transcend the language, it does not seem to make sense to speak about something extra-linguistic. It follows that we should abandon doing traditional metaphysics. Since there is no need to assume the existence of an extra-linguistic reality, there is no need for metaphysics to study such a non-existent thing.
2. A Problem for the Kantian-style Objection
Both empiricist and postmodern critiques seem to be based on the Kantian-style objection. The Kantian view asserts that it is impossible to transcend our conceptual scheme (our language) and to attain the knowledge of the reality as it is. So do empiricists argue that all our knowledge is limited to sense data and the statements about the nature of what lies behind the sense data are meaningless speculations. In a similar way, postmodernists exploit the idea that there is no way how to step out of the language. From this they conclude that the idea of extra-linguistic reality is problematic. It is obvious, therefore, that one way to counter the criticisms (and to attempt to defend the traditional metaphysics) is to address the Kantian-style objection.
Michael Loux points that the Kantian view of metaphysics is problematic. If Kantians ban traditional metaphysics on the grounds that the ever present conceptual scheme does not allow us to study being as such, they seem to face the same problem with regard to the conceptual scheme – the object of the study of the Kantian-style metaphysics. Loux puts the argument in a simple and elegant way:
“The central premise in the schemer’s [Kantian-style] argument against traditional metaphysics is the claim that the application of conceptual structures in the representation of things bars us from genuine access to those things; but the defender of traditional metaphysics will point out that we need to employ concepts in our characterization of what the schemer calls a conceptual framework, and they will conclude that, by the schemer’s own principles, that entails that there can be no such thing as characterizing the nature and structure of a conceptual scheme.” (Loux 2002, 10)
This is a nice example of a reductio. Strict and general application of the main Kantian assumption (that the approach via framework bars us from genuine access to the object of our study) leads to the elimination of the Kantian-style metaphysics as well. So Loux’s argument implies that the Kantian-style criticism of the traditional metaphysics leaves us with a comprehensive skepticism in the area of metaphysics. There is no metaphysical knowledge because neither traditional study of being as such nor the new study of conceptual schemes is possible.
There is also a slightly different way to make the same point that the Kantian-style objection against traditional metaphysics is problematic. The second argument stems from a somewhat different emphasis in the formulation of the Kantian view. While the above argument led to the metaphysical skepticism, the second one is aimed to show that there is a danger of infinite regress. (It seems to be possible, however, to interpret this type of infinite regress as a kind of metaphysical skepticism as well.)
Kantians claim that while doing metaphysics we allegedly cannot study reality itself but only the concepts we use to approach it. But since we need a conceptual framework to be able to speak about reality, is it not the case that we need another conceptual framework to be able to approach the previous conceptual framework? Kantians need to be consistent and if they require a framework to approach the reality, there must be another framework for analyzing the framework used for reality. Therefore, Kantians cannot deny the possibility of traditional metaphysics and at the same time allow for a direct analysis of the conceptual framework. Kantian view seems to end up in an infinite regress. To be able to analyze any conceptual framework we must presuppose the existence of a different conceptual framework at a higher level that allows this analysis. Moreover, it seems that at no level can we get to the conceptual framework as it is but only as it is seen from the point of view of another conceptual framework.
Let me outline the presuppositions of the second argument. The whole argument starts with the Kantian view of the situation in the area of metaphysics and knowledge in general. Although, on this account, being as such cannot be studied any more, there is a fitting substitute available. The first premise is a new metaphysics assumption claiming that the conceptual scheme or framework is the proper object of the metaphysical study. Concerning the possibility of knowledge Kantians assert that we can never get to know the being an sich; the only thing we can grasp is the being as it appears to us. The unknowable being as such is so to say mediated by the use of the conceptual framework. The framework or scheme is a necessary instrument for gaining any type of knowledge. Moreover, the conceptual framework does not provide us with knowledge of the genuine being as such but only with the knowledge of being as it appears to us. By the use of the conceptual framework we can gain only, in a certain sense, disturbed or modified “picture” of the object (“object” in a general sense; not limited to a concrete individual). We have reached the second premise, necessary scheme assumption, which states that in order to gain knowledge it is necessary to use a conceptual scheme and that by the use of the scheme we achieve a modified representation of the object.
If we combine these two assumptions, we have an argument that is very similar to the above cited Loux’s objection:
The proper object of the metaphysical study is the conceptual scheme.
(The new metaphysics assumption)
To gain knowledge it is necessary to use a conceptual scheme which, however, modifies the object it mediates.
(The necessary scheme assumption)
Therefore, we achieve only a modified representation of the conceptual scheme.
If the expression “modified representation of the conceptual scheme” is understood as “not the genuine knowledge of the nature of the conceptual scheme” it is obvious that this is basically the first argument showing that Kantian view leads to skepticism in the area of metaphysics. In order to formulate the second argument pointing out to the infinite regress, it is necessary to supplement the argument with an additional assumption.
It is a commonplace to assume that every area of knowledge (be it metaphysics, history or anything else) is studying a certain object. Usually the object of study of a discipline is something distinct from the discipline itself. For instance, history is studying past and not the nature of the historiography itself. Historiography, on the other hand, is the object of study of a different discipline, namely the theory or philosophy of history. The same applies to other disciplines like mathematics, physics etc. So there is no doubt, that there is some kind of hierarchy in our body of knowledge.
But is this hierarchical structure only a contingent matter or are there logical reasons why this should be so? One of the lessons of the liar’s paradox (the statement claiming its own falsity, i.e. “This statement is false”) is that we should be suspicious of self-referentiality. Philosophers familiar with A. Tarski’s solution to the paradox usually point out the need to avoid self-referential statements and to distinguish between various levels of discourse; otherwise, we may end up espousing a contradiction. Careful stratification of different levels is widely recommended not only in the context of language but also in other contexts. Therefore, if there is an object of study O of a peculiar discipline D1, then D1 should study O but not D1 itself. Of course D1 may become the object of a study, but this additional study will be carried out by a discipline D2 located at a higher level than D1. In this way we arrive at the necessary hierarchy assumption claiming that there is necessarily a hierarchy of disciplines (levels of inquiry) in which every discipline (level of inquiry) is about something distinct from itself, namely, higher level inquiry is about a lower level. In other words, no discipline is about (studies) itself.
The second argument might be stated as follows:
The proper object of the metaphysical study is the conceptual scheme.
(The new metaphysics assumption)
To gain knowledge it is necessary to use a conceptual scheme which, however, modifies the object it mediates.
(The necessary scheme assumption)
Every level of inquiry is about (studies) something from a lower level.
(The necessary hierarchy assumption)
Therefore, every conceptual scheme is studied from the point of view of a higher level scheme which means we never get to know the scheme as such but only as it is represented from the point of view of a higher level scheme.
The obvious difference between the two arguments is the fact that the first one does not rule out the possibility that the object of the metaphysical study – the conceptual scheme – is approached from the point of view of the same conceptual scheme. The first argument allows a situation in which the very same conceptual framework is the object of the study and at the same time also “the instrument” used in the study. The second argument, on the other hand, requires a strict hierarchy. With respect to the generality, the first argument covers all the cases of the second one and the additional case of the conceptual scheme figuring as an object and instrument at the same time.
The reason for a separate articulation of the second argument is the fact that it demonstrates the problem of infinite regress more vividly. The first argument might be viewed as underlining that there is no genuine metaphysical knowledge about the conceptual scheme, while the second one is claiming that the only thing we may know about the scheme is how it appears from the point of view of a higher conceptual scheme. One may assume that by uncovering the nature of the higher level scheme we may get closer to a genuine knowledge of what took part in creating our representation of the lower level and in this way we would probably get also a better view of the genuine nature of the lower level. To reveal the nature of the higher level scheme, however, we have to approach it from yet a higher level which means we will not uncover the nature of the higher level scheme but only how it appears from a different level. And such a chase for a proper metaphysical knowledge would have to continue ad infinitum.4
These two arguments imply that the Kantian critique of the traditional metaphysics and the Kantian account of metaphysics as a study of conceptual schemes are problematic. Kantian view leads either to skepticism or to infinite regress. Thus it seems that on this view there is no prospect of doing metaphysics.
Is there a way out of skepticism and infinite regress? The third option is to defend dogmatism. Proponents of this solution would contend that there is a special sort of entities metaphysicians are able to access without any problem. In a certain sense traditional metaphysics might be viewed as a form of dogmatism – it assumes that it is possible to study being as being. It seems, however, that dogmatism is precisely the view Kantians want to criticize. On the other hand, if they stated that the conceptual scheme is the special type of thing that could be studied directly they would seem to embrace the same attitude as a traditional metaphysician, only towards a different type of entity. But why should one accept that it is possible to study directly a conceptual framework but not the being as such? Kantians would need to supply an argument to show that it is plausible to choose just their favorite option. Otherwise, their choice would be entirely ad hoc and in a sense dogmatic.
3. On Doing Metaphysics
One may try to avoid getting trapped into one of the unacceptable positions (skepticism vs. infinite regress vs. dogmatism) by embracing Popper’s solution of the so-called Fries’ trilemma. In his Logic of Scientific Discovery Popper discusses the possibilities of scientific knowledge (mainly within empirical sciences). The traditional view has it that we may claim to possess (scientific) knowledge only after there is some kind of foundation that justifies scientific claims but does not need any further justification itself. Most common proposals of the justificatory foundations are either self-evident truths (dogmatism), or sense perceptions (psychologism). Otherwise, if we try to justify one scientific claim by citing a different one (and this different one by citing yet another one, etc.) we will inevitably end up in an infinite regress of justification. Although it may seem that we are bound to choose from one of these three options (dogmatism vs. psychologism vs. infinite regress), Popper claims that there is a way out of Fries’ trilemma. According to him, one should embrace a critical attitude towards all scientific knowledge. It means that the scientists should not look for the firm foundations and sources of their knowledge. Rather, they should try to come up with their proposals and hypotheses, and criticize them instead. Science should be a critical activity making use of the method of trial and error (Popper 1959).
Is it possible to conceive of the metaphysics in a similar way? I believe that the critical approach is (at the very general level) a common feature not only of the sciences but also of philosophy and metaphysics. One may think of metaphysical theories as nets we throw either at the being as such or at the conceptual schemes.5 Some of these tentative theories may “catch something from the reality” and it is up to us to find out which of them are more successful and which of them are “empty”. How do we distinguish between good and bad theories? We approach all of them with a critical attitude, i.e. we test them. In these tests we may either consider (i) whether they are internally consistent, (ii) whether they are coherent with some other assumptions we want to make, (iii) whether they match with what modern science tells us; or we may compare more theories and assess them (iv) with respect to their simplicity, (v) with respect to the amount of counterexamples they have to face, etc.6 These forms of critical testing ask for elaboration, but intuitively, they show that there is a way how to subject the metaphysics to a criticism and how to distinguish between better and worse metaphysical theories.
In fact, one need only to examine some of the current work on metaphysics to find out that these forms of critical testing are actually applied. In one of his papers, Jonathan Schaffer (2003) questions a common metaphysical assumption that there must be a fundamental level of being. Besides other things, he tries to show that the metaphysics without this assumption is possible (a kind of (i)) and that the modern science does not provide the evidence for fundamentality (a kind of (iii)). In his recent paper, Theodore Sider (2007) argues against monism (there is only one thing—The World) on the grounds that if monism is correct, every property belongs to The World as a whole and not to the particular objects (electrons etc.). And this seems to run against our belief that some properties belong to particular objects and not the whole being (a kind of (ii)).
Now, how does the discussion in this section relate to the Kantian-style criticism of the traditional metaphysics and to the problem of the Kantian objection outlined in the second section? There is a danger which could be easily overlooked, that if the results of these two criticisms are viewed from the perspective of the either-or logic, we may be forced to abandon doing any metaphysics. The either-or logic is based on an everything-or-nothing attitude.7 Either the metaphysics uncovers being as being, or because of the fact that we cannot approach being as such but only as it appears to us, such a metaphysics is impossible. Either the metaphysics correctly characterizes the conceptual schemes, or because of the need to apply some kind of mediating conceptual scheme to their study, even this type of metaphysics cannot be carried out.
The either-or reasoning tries to convince us that either we may grasp the absolutely pure and in no way conceptually tainted subject matter of metaphysics (be it being as being or conceptual framework), or otherwise we learn nothing at all. In my view, this type of reasoning disregards quite plausible option that despite all the conceptual mediation we may learn at least something. The fact that the absolute “ideal” seems to be unattainable should not entail that we finish all our attempts to study being or conceptual frameworks. Embracing a critical attitude may be a way out of this tricky situation. Being aware of the ever present conceptual intermediary between us and the object of our knowledge, we may become cautious and engage in doing a critical metaphysics.
The criticisms of traditional metaphysics outlined in the first section seem to be based on the Kantian view that we cannot learn about the being as such because we must apply our conceptual schemes to its study. Nevertheless, the Kantian approach to metaphysics suffers from a similar problem. If we cannot get to reality as such, either we cannot know the genuine nature of the conceptual scheme or we are able to learn only how the scheme could be seen from the point of view of another conceptual scheme. In each case we lack the pure and final metaphysical knowledge we allegedly looked for. The Kantian-style critique of the traditional metaphysics seems to undermine the Kantian view of metaphysics as well.
All these problems, however, do not necessarily force us to abandon metaphysics as such. The only thing we should dismiss is the uncritical attitude. In case we realize that the metaphysical knowledge is revisable and not certain, we may engage in doing modest but viable critical metaphysics. Or as E. J. Lowe claims, “… metaphysics can indeed be about reality, and can avoid collapse into empirical scientific theory, provided we can learn to be content with the fact that, as far as actuality is concerned, metaphysics cannot provide us with certainties.” (Lowe 1998, 27)8
Ayer, A. J. (1936): Language, Truth and Logic. London, Victor Gollanz.
Carnap, R. (1959): “The Elimination of Metaphysics through Logical Analysis of Language.” In: Ayer, A. J. (ed.): Logical Positivism. New York, Free Press, pp. 60-81.
Carr, B. (1987): Metaphysics: An Introduction. Atlantic Highlands, New Jersey, Humanities Press International.
Collingwood, R. G. (1940): An Essay on Metaphysics. Oxford, Clarendon Press.
Gasking, D. (1955): “Causation and Recipes.” In: Mind 64, pp. 479-487.
Hume, D. (1910): An Enquiry Concerning Human Understanding. Harvard Classics Volume 37, P.F. Collier & Son.
Jenkins, K. (1999): Why History? Ethics and Postmodernity. London, Routledge.
Lorenz, C. (1998): “Can Histories Be True? Narrativism, Positivism, and the ‘Metaphorical Turn.’” In: History and Theory 37, pp. 309-330.
Loux, M. (2002): Metaphysics: A Contemporary Introduction. 2nd edition, New York, Routledge.
Lowe, E. J. (1998): The Possibility of Metaphysics. Oxford, Clarendon Press.
Popper, K. R. (1959): The Logic of Scientific Discovery. London, Hutchinson.
Schaffer, J. (2003): “Is There a Fundamental Level?” In: Nous 37, pp. 498-517.
Sider, T. (2007): “Against Monism.” In: Analysis 67, pp. 1-7.
Volek, P. – Labuda, P. (2006): “On the Role of Different Types of Metaphysics in Science-Religion Dialog.” Available at
von Wright, G. H. (1993): “On the Logic and Epistemology of Causal Relation.” In: Sosa, E. – Tooley, M. (eds.): Causation. Oxford, Oxford University Press, pp. 105-124.
1 This section is inspired by Michael Loux’s presentation of the criticisms in the Introduction to his (Loux 2002). Nonetheless, my analysis of these objections does not necessarily agree with his. In this paper I follow quite common practice to make a distinction between traditional view of metaphysics and metaphysics treated as a study of conceptual schemes. See for instance (Carr 1987) and (Loux 2002).
2 By “Kantian view of metaphysics,” “Kantian-style objection” etc. I refer to the views, objections etc. of the authors inspired by the ideas of I. Kant. I do not attempt to outline the views of I. Kant himself. For instance E. J. Lowe speaks of “neo-Kantianism”; see (Lowe 1998, chapter 1).
3 See for instance (Collingwood 1940, chapter XXXI.), (Gasking 1955) and (von Wright 1993).
4 Infinite regress might be viewed as leading to skepticism of a sort. If there is no final knowledge of conceptual scheme as such possible (only knowledge of how one scheme appears from the point of view of a higher level scheme), then there is no metaphysical knowledge.
5 This metaphor comes from Popper: “Theories are nets cast to catch what we call “the world”: to rationalize, to explain, and to master it. We endeavour to make the mesh ever finer and finer.” (Popper 1959, 59).
6 It seems to me that these ways of testing depend heavily on the conceptual (language) analysis, which is presented as a method of metaphysics in (Volek – Labuda 2006).
7 The either-or logic metaphor is used in (Lorenz 1998).
8 I am indebted to participants of CIS seminar for their comments and suggestions.