PRACTICAL KNOWLEDGE VERSUS KNOWLEDGE AS PRACTICE

Transcription

1 HUMAN AFFAIRS 19, , 2009 DOI: /v PRACTICAL KNOWLEDGE VERSUS KNOWLEDGE AS PRACTICE ISTVÁN DANKA Abstract: The main thesis of this essay is that practice is superior to a theoretical vs. practical distinction. In this sense, every sort of knowledge is essentially practical ; so-called theoretical knowledge is an historically overemphasised borderline example of the practical. Based mostly on Wittgenstein s view, I shall gradually refine an opposition between theoretical and practical knowledge by analysing some related dualisms on an active, processual, communicative and applicative concept of knowledge. Then I will provide some arguments as to why knowledge as a practical matter in this sense should be seen as, both logically and temporally, prior to the distinction. Keywords: theory of knowledge; practical knowledge; theory of action; Anti-Cartesianism; Wittgenstein. One of the greatest challenges to philosophy has always been its self-legitimisation. This has become more pressing in the Information Age, since so-called practical knowledge is of more importance than before. Increasingly, in scientific journals as well as in popular magazines, we read that in the 21 st century, theoretical knowledge, and hence, philosophy, will lose its traditional role. Although we will live in a knowledge society, knowledge will become something other than that which it is, or seems to be, at present. My aim, in this paper, is to show that the view that holds that theoretical knowledge is losing importance is based on a generally accepted, but mistaken concept of knowledge, which explains knowledge as being primarily theoretical. Following Nyíri (1992), I think that the common ground shared by the two most influential philosophers of the 20 th century, Ludwig Wittgenstein and Martin Heidegger, is helpful in addressing this issue. Their position was that all of our knowledge is rooted in practice. If they are correct, I argue that it is not true that the way in which we acquire knowledge has really changed; what has changed is our picture of how knowledge is acquired. Since Plato, the traditional model of how knowledge is constructed has focused on propositional, i.e., theoretical, knowledge. In this model, knowledge has a hierarchical structure with philosophy at the top and practical knowledge at the bottom. Most philosophers paradoxically claimed that the pyramid of knowledge is built from the top down. As Wittgenstein showed, however, theoretical knowledge presupposes practical skills: following a rule, i.e., learning and using language is embedded in practice, and 397

2 cannot be analysed outside practice. It seems that theoretical knowledge is preceded by practice both temporally and logically. Practice is therefore a necessary precondition to acquiring theoretical knowledge. Explaining methods of acquiring knowledge requires a new, practice-oriented approach to knowledge. To say this is not the same as saying that methods for learning/for acquiring knowledge have changed. Rather it has become more obvious that knowledge is not something that fits the traditional characterisation. A new approach should see the pyramid of knowledge built from the ground upwards, i.e. from practice to theory if one is to speak of a pyramid at all. The old conception saw knowledge as a representational, or, as I shall label it, a relational category. The relational theory was, I think, generally accepted before Nietzsche (with the exception of some inferentialists such as Leibniz or Spinoza, as Brandom [2002] emphasises). According to relational theory, knowledge is a relation between two objects (or two radically different sets of objects), namely facts and beliefs, world and mind, reality and language, plurality and unity, object and subject, etc. Explaining knowledge in terms of mediation between two radically different sets of objects has not been carried out up to now. Some philosophers have concluded that knowledge itself does not work. I prefer to claim that our concept of knowledge does not work and that is why we cannot capture the way in which knowledge itself could or should work. A possible rival, and hopefully more successful, approach manages knowledge as an active (non-contemplative), communal-communicative (non-individual), dynamic or processual (non-static), applicative (non-relational), i.e., practical (non-theoretical) matter. Knowledge does not grasp or describe reality, but prescribes some orders or offers advice to help us to interact within the world. Explaining knowledge in such a way entails recognising that theoretical knowledge is not something radically different from practical knowledge, but that it is only a set of tools for more complex tasks. Therefore, if theoretical knowledge is far removed from application then that is the case simply because of the complexity of the (practice-involved) problems that it is to solve. Below I will compare theory-based conceptions of knowledge with practice-oriented ones, gradually refining a position by following the above-mentioned oppositions. Then I will show why such a practice-oriented approach does not run counter to the rival view but is superior to the theory-practice dualism. Finally, I shall say a few words about some further consequences. Following Steup (2001), there are three radically different kinds of knowledge. Something is claimed to be knowledge if it is (1) knowing a place or person, or (2) knowing how to do something, or (3) it is propositional knowledge. It was (3) which most attracted the attention of philosophers. Knowledge of philosophical importance was traditionally formulated as a relation between a subject and the proposition S knows that p. (1) can be reformulated in terms of propositional knowledge if the place or person in question has a definite description (as it does have if it is known in the proper sense of knowing), i.e., a proposition which identifies it. Finally, (2) was mostly left unconsidered, given that it raises purely practical questions which are outside the scope of epistemological enterprise. A significant part of our knowledge was not explained in terms of a theory of knowledge until the last century when philosophers such as Ludwig Wittgenstein, Martin Heidegger and the American pragmatists emphasised those aspects which Gilbert Ryle called knowing how, as opposed to knowing that. 398

3 Propositional knowledge is often explained as a special relation between human beings and the world: S knows something about the world in the form of p. This relation is essentially unidirectional in most of the approaches to the issue: the world affects us and we are passive recipients of that. In philosophical approaches to knowledge and action, what we know about the world was discussed separately from how we act in the world: theories of knowledge and those of action were treated as different fields. In these views, acquiring knowledge, as opposed to human activities in general, is contemplative i.e., passive reception. Knowledge seemed to be a distinct access to the world, a necessary prerequisite for being able to act. The traditional argument was that without knowing how things were, it was impossible to act within the world, in accordance with its rules. In some recent views, following Berkeley s, Kant s and Schopenhauer s insights, the Self (wholly or partially) constitutes what is known. Acquiring knowledge can be understood as a kind of activity. Hence, questions of knowledge can be explained in terms of an action theory. The basic intuition behind these views is that experience of the world does not leave an impression, or, in Locke s words, does not imprint ideas into our mind. Cognition is interaction: the Self reaches the world. We ask questions to which the world responds, or reacts, at least. Even if it were true that without knowing how things were, we were unable to act within the world properly, without knowing how to acquire the ability of acquiring knowledge, we would not be able to acquire knowledge either. To avoid an infinite regress, it should be supposed that acquiring knowledge is a practical matter that is a kind of activity which is related to humans non-propositionally. As a consequence, knowledge acquisition is primarily not an epistemological question. In these views, the theory of action seems to be a more general discipline in which questions of knowledge acquisition is a subfield: they are discussed within a framework of questions concerning human activity in general. There are at least two important reasons for accepting the latter approach. On the one hand, if we avoid the infinite regress, the sceptical problems that continuously crop up within in the history of epistemology would disappear. On the other hand, if questions of propositional knowledge could be explained in terms of actions, the opposition between knowing that and knowing how would become less distinct. It therefore seems that if we accept the view that acquiring (propositional) knowledge is action, then it would be possible to explain the above-mentioned different kinds of knowledge in a single, uniformed framework. One of the main difficulties with explaining knowledge acquisition as action is that propositional knowledge is claimed to have content, whereas it is hard to imagine the same of actions. The content of knowledge is what is actually known; propositional knowledge is a relation (of the knower) to its content. The content of knowledge represents a certain fact, object or state of affairs of the world. Contents are therefore often explained as mental representations. A relation of an agent of an action to the world is explained not as being representational but as being causal. Actions do not have contents; they are mostly explained in terms of reasons and causes, referring to pro-attitudes (e.g. beliefs, desires, and intentions). Explaining beliefs is, nevertheless, a central question pertaining to knowledge acquisition as well. Beliefs are mostly understood in epistemology as mental representations of facts i.e., they are static, fixed entities. Their theoretical role is rooted in Cartesian/Lockean 399

4 ideas : they are knowledge bearers located immediately at the front of the mind. Acquiring knowledge in this view is closely connected to the contemplative view mentioned above: our Mind s Eye looks at the world and what it is able to see are its beliefs, which are ideally exact representations of pieces of reality (for different approaches to the eye of the mind metaphor, see Ryle [1949], Havelock [1962], and Rorty [1979]). This model can still be seen as dynamic; since it is the mind which looks at the world, it should be explained as an active agent of the process of acquisition. Certainly, the role of beliefs should be reconsidered in this model. Understood within the framework of the theory of action, beliefs are not (conceptual and/or perceptual) contents, whose role is representational but theoretical suppositions, which serve as reasons for an action. Action theorists should not need to deny that representational functions can be attributed to beliefs; what they should deny is only that beliefs should be explained as essentially representational. Beliefs are better described for their purposes as psychological events, which are, in some descriptions, causally continuous with the physical reality. If beliefs represented reality, they would be able to do so because they are causally embedded in reality. By rationalising an action, we describe our reasons in terms of pro-attitudes. A description of our pro-attitudes is regularly constituted as a proposition one of the typical forms of a description of an action in terms of reasons is precisely how propositional knowledge is formulated. In such a description, beliefs are content-like. It is not the same as saying that beliefs themselves are content-like; they are, nevertheless, content-like in certain descriptions. In the debates over their status, beliefs were understood as either mental representations or psychological events as either representational or non-representational. That was one of the main reasons why action theoretical approaches seemed to be incoherent with knowledge theoretical ones. In a more general framework, however, the action theorist could say that some propositional attitudes could have a representational function, even if there certainly are some which have no such role; the only restriction is that all propositional attitudes must be causally connected to physical (or social) reality. For some theoretical purposes, it can be fruitful to suppose representational functions: that supposition could explain why we reasonably think that our scientific theories represent how things are. But the reason why this representation works correctly is not that there is a well established one-to-one relation between our beliefs and states of affairs. The reason is that our beliefs are caused by those states and causation guarantees that no serious epistemic misapprehension can occur. On the other hand, causal connection also makes room for misrepresentation: in a fallibilistic approach, it is not a decisive error if something turns out to be misrepresented. It is not an argument for the scepticist either an assurance of our relation to reality is not representational but causal, where representation misses the target, causality maintains the connection. All the same, fallibilism also provides an alternative answer to some questions concerning another allegedly definitive feature of propositional knowledge, namely justifiability. In 20 th century epistemology and the philosophy of science, justifiability becomes a less central question of knowledge. From strong verificationism via confirmationism to falsificationism and beyond, there is a shift of emphasis from knowledge as being justified once and for all to knowledge as being accepted provisionally. In these views, knowledge is accepted 400

5 not because it is claimed to be eternally true but because it seems, based on the available evidence, to be the best approach of the possible alternatives. Justification is a method which establishes a relation between beliefs and truths. It suggests that isolated pieces of beliefs can be confronted with isolated pieces of reality and under ideal circumstances, there is one and only one, observer-independent relation to be drawn. This view is sometimes rejected on the grounds that it is too close to relativism. This is not the case, however. There is a strong objectivist claim behind fallibilistic approaches which relativisms lack. Fallibilists agrees with justificationists on two of their objectivist claims at least. First, they reject that knowledge can be pluralist: under any certain circumstances, there is one and only one correct possibility. Second, they reject that knowledge can be observer-dependent: anyone in a certain situation, if the situation is correctly judged, would be able to acquire one and the same truth. The reason knowledge is not invulnerable to revision is simply because of the temporal character of knowledge acquisition: i.e., under different temporal circumstances we have different expansions of evidence. Fallibilism does not claim either that under ideal circumstances beliefs can still be revised, or that ideal circumstances do not exist, or even that they cannot be reached; it only claims that what ideal circumstances are cannot be fixed once and for all, since our belief on this is also not invulnerable to revision. Concluding from this then, the claim that there is no point in speaking about ideal circumstances is precisely a consequence of (antirealist) justificationism; namely that inference supposes that that which cannot be justified cannot be true. What then, if not justification, guarantees the truth of our beliefs? Fallibilists do not have to say that justification plays no role in rationalising what counts as knowledge. Their claim is rather that justification, in the sense of confronting pieces of beliefs with pieces of reality, is not a definitive feature of knowledge. Our hypotheses certainly need to be tested on reality; the difference lies in the method of testing. The sort of fallibilisms I prefer to follow claim that a method of testing can be best elaborated if we reject the old individualistic framework of how to acquire knowledge. This framework is based on the assumption that even if the individual is able to misrepresent reality, it would be, at the very least, highly improbable that each member of a community could misrepresent it in the same way. The scepticist may reply that even if it were improbable, it would not be impossible. But according to the above-mentioned two features of objectivity, this misrepresentation would be objective if there was one and only one alternative that claimed to be correct observer-independently. In other words, either intersubjective agreement is sufficient for objectivity, or it should be claimed that there is a non-intersubjective way of objective knowledge. By claiming the latter, however, scepticists are arguing against themselves. If we refer back to the non-representational picture of knowledge acquisition, we are causally connected to the world which guarantees that there is no ontological gap between humans and non-human reality. If there were supposedly an epistemic gap at the level of representation, intersubjective criteria would be more successful than individualistic ones. If everyone misrepresented reality in the same way then the method of misrepresentation would be causally involved in human nature securing the connection with reality. If misrepresentation were carried out in pluralistic 401

6 ways, reasons and evidence could serve as weapons for the fight to the finish that cannot be elaborated in an individualistic framework in which communication cannot play any role in justification. In an individualistic approach there is no way of positioning whole sets of reasons and evidence against each other but only beliefs against reality i.e., there is no way of explaining knowledge tests in an epistemologically homogeneous framework. Conceding that epistemically heterogeneous sets of entities are included in explanations of knowledge tests always assists the skepticist. The traditional, rather odd solution to this was coherentism, according to which only beliefs can justify beliefs. In a communal/communicative approach like Wittgenstein s, beliefs of others play this role, whereby the subject is not enclosed within a prison of its eventual fictions. At the representational level, beliefs (as mental representations) of the members of a community are connected to each other in a (more or less) coherent set. At the non-representational level, beliefs (as psychological processes) of the members of a community are connected to each other via social causation. No one-to-one correspondence is required between the causal and inferential connections among beliefs; but if one were not to hold at least most of the fundamental beliefs of a community, one would lose one s causal relations as well, i.e., one would not count as a member of that community. There is a strong objection to the above-mentioned argument. We can assume that there is an agreement in the community, and that the members are causally connected to physical reality. But if we accept that there are two levels of explanation with no one-to-one relation, pieces of beliefs are not only revisable but also never completely reliable. We never know whether our representation of the world is correct; what we know is only that if we were to be mistaken then it would be the world that would be causing the mistake. The answer to this objection is that pieces of beliefs should not correspond to reality, and hence knowledge should not be a relation between pieces of beliefs and pieces of reality. It is sufficient if a theoretical framework based on our set of beliefs is applicable to interact with the world under certain circumstances. Therefore, instead of empirical justification or axiomatic proofs, it is applicability which plays the central role in knowledge tests. It should be explained why and how this approach differs from the traditional coherentist point of view claimed to be an odd solution above. In coherence theories the only guarantee of the truth of a belief is that it fits into a coherent whole of our beliefs i.e., knowledge should be understood as relations between beliefs. Objections to this view mostly counter that coherency can be well established in several, even contradictory sets. Therefore, either reality itself is plural, which seems absurd, or one and the same reality can be fully described in different, even contradictory ways. That would mean that reality is self-contradictory, which is no less absurd. Hence, so far as there are other alternatives, an explanation in which coherence is the only criterion should be avoided. On the other hand, as I alluded to above concerning the more or less coherent nature of beliefs, there are in fact no coherent sets of beliefs (of communities, but probably also very few individuals could boast fully coherent views). It seems that coherence in a strong, holistic sense is an exaggerated criterion; otherwise we would have to say that none of us know anything. Ideally, a relatively high degree of coherence would be required, but actually it should be realised that sometimes even a non-coherent set of beliefs e.g. a conjunction of rival theories in science describes one or other aspects of the world very well. If the above-mentioned are correct, it is important to avoid introducing two epistemically different kinds of entities into our explanation. Therefore 402

7 coherentism is right in that no single, isolated pieces of beliefs should be confronted with reality. But it is mistaken in claiming that coherence among beliefs is either a sufficient or necessary condition of knowledge. A restricted holism labelled contextualism, in which coherence is a criterion only within certain contexts, is a good starting point for introducing the rival theory. But pure contextualism would permit relativism. That is the reason why a criterion of applicability is required. No single belief should be confronted with reality but sets of beliefs should be applied to serve as the reasons for an act. That is that there is no confrontation, no one-to-one relation not even a relation between sets of beliefs and states of affairs. Our sets of beliefs certainly represent states of affairs but representation should be seen, as I suggested, as insignificant in our relation to the world, even if it is significant in rationalising that relation. Our sets of beliefs are caused by states of affairs, and by action, they also cause certain changes in states of affairs. In this view, a certain set of beliefs exists in a knowledge-relation with reality if it causes the appropriate (wanted, forecasted, etc.) changes. Our relation to the world is explained causally, which means that there is a one-way, asymmetric relation by which the world affects us. If we are able to cause appropriate effects in the world, then our relation can be seen as a symmetric one. If we ask the appropriate questions, the world responses with good answers. If not, our preconceptions should be revised. This view is a kind of pragmatism since success plays an important role in an explanation of knowledge. It is not a standard pragmatism, however. Truth and success are not the same; a successful application does not necessarily contain a true description of facts. What this view suggests rather is that truth as a correct description of states of affairs plays a less significant role in explanations of knowledge than it was supposed by many theories of knowledge. Knowledge is not a description of the world; if knowledge is to be explained as a relation to the world at all, it certainly should not be seen as a subject s fixed, representative reception of objective reality. Knowledge rather serves as advice on how to interact; it is applicative, and therefore a dynamic and flexible, processual activity of a community within the world. I mentioned that the view developed above can be understood as a pragmatist approach to knowledge (even if not to truth). It is certainly pragmatist in the sense that it claims that in deciding which one is better of two rival theories, there are two cases: either there are practical consequences of our choice or there are only pure theoretical consequences. In the latter case, pragmatism claims, there is no real difference. So-called empirical underdetermination of theories has nothing to do with this view: if two theories have the same effects on human activities, then it is more or less all the same which one is chosen. It is also far from relativism: the model claims that there is one and only one reality, and there is also one and only one correct relation to reality. But as far as knowledge is not a representation of reality but a tool of producing certain effects, differences in representation that do not have different causal effects in activity are neither here nor there. This claim is not the same as saying that practical knowledge is more important than theoretical knowledge. My point is rather that it is appropriate for every sort of knowledge to be explained as practical, whereas it is certainly inappropriate to explain some sort of knowledge in terms of theoretical knowledge. Before turning toward an explication of this claim, two further remarks are relevant here. First, it should be noticed that there is a further distinction which seems to be appropriate 403

8 at first sight to characterising a difference between theoretical and practical knowledge, namely an opposition between abstract and concrete knowledge (Bloor 2004, 113). It is not by accident that I have not mentioned that distinction. On the one hand, I think that such a distinction is not bipolar but that there are different levels of abstraction. Hence, a difference between the abstract and concrete is rather gradual. This would not be a reason for its inapplicability, however, since a theory-practice dichotomy could probably also be explained as gradual difference. But on the other hand, I do not think that practical knowledge would be concrete and theoretical knowledge would be abstract. On the contrary: practical knowledge, so far as it contains the most general human skills, is certainly not concrete, even if its applications are often concrete. My skill of reading books is a general knowledge, applicable in countless situations, even if it is manifested in my specifically reading a book. In this sense, however, theoretical knowledge is concrete as well. No knowledge can be manifested in non-concrete situations, simply because a non-concrete situation makes no sense. On the one hand, it seems therefore, that both practical and theoretical knowledge are at a certain level of generality. On the other hand, it also seems that its applications are concrete. (It also means, of course, that even if knowledge is applicative, it is not the same as if knowledge were pure application.) Second, I mentioned that a turn from theoretical to practical knowledge should focus on processes rather than pieces of entities. Schatzki (2001, 7) claims that [w]hereas philosophers and social investigators once cited mental entities such as beliefs, desires, emotions, and purposes, practice theorists instead highlight embodied capacities such as know-how, skills, tacit understanding, and dispositions. Above I tried to refine a theory of knowledge by mosaics which rejected cognition as contemplation, beliefs as static, (primarily) representational pieces, acquisition as individual activity, and knowledge as a fixed relation between beliefs and states of affairs step by step. The above-mentioned outlines of what beliefs and other pro-attitudes are seem to better fit into a picture that the practice theorist requires. One further important step should not be prevented, however. This step deals with the direction of an explanatory strategy of knowledge. As I mentioned, a traditional order of explanation is grounded in individualistic frameworks, seeing pieces of knowledge as isolated, atomistic entities. An explanation of knowledge as practice cannot begin with individuals and isolated pieces our explanations of practice are totally incompatible with such atomistic views. An obvious starting point would be to begin with sums of individuals called communities and sums of pieces of beliefs called sets of beliefs. This approach would be misleading, however, and would be unable to avoid the problems that occurred in atomistic approaches. As I mentioned, the view presented here is closer to holisms (restricted to contexts) which deal with the totality of individuals and that of beliefs. The distinction between sum of and totality of is that an explanation of a totality focuses on complexes, on interrelations among pieces rather than on pieces themselves. That is also a reason why causality is the central issue of this view, not the selves of individuals, the contents of beliefs or states of affairs. The explanatory strategy applied here is therefore a top-down account of totalities, as opposed to a bottom-up approach to sums. Action theoretical approaches miss their target unless an anti-atomist turn is followed up: a change from speaking of isolated actions to speaking of practice as a whole. In agreement with Herbert Dreyfus, I think that background practices 404

9 [...] are the condition of the possibility of all rule-like activity (Dreyfus 1979, 57). It is with the broadest totality of practice that our explanation should begin; and finally we hope to arrive at the smallest pieces of activities as well as (both theoretical and practical) knowledge. A reversed direction of explanation would presuppose that individuals and atomistic pieces of beliefs should still be explained in a framework that is precisely attempting to by-pass. Now back to the point that all kinds of knowledge can be explained as practical but some sorts of knowledge cannot be explained as theoretical. First it should be noticed that not all philosophers dealing with practical knowledge emphasise only the differences between practical and theoretical knowledge. While Ryle s purpose was to characterise knowing how by distinguishing it from knowing that, he also claimed that theorising is one practice amongst others (Ryle 1949, 26, italics added). For Wittgenstein in the 1940s, practice was a prerequisite of propositional knowledge (von Wright [1982, 178], Nyíri [1992, 48]) but this insight of his is certainly not too far from the moral of the final paragraph of his early Tractatus as well. According to a growing set of interpretations following Apel s view (1967), there is also a parallel between Wittgenstein s analysis of practices and Heidegger s notion of being-in-the-world. The latter s influence on Ryle is also worth noticing (see Murray [1978]). It is also widely held that Ryle, as every philosopher in the UK in the 1940s, was hardly unaware of Wittgenstein s manuscripts entitled Blue and Brown Books (Nyíri 1992, 95). Both Heidegger and Wittgenstein emphasise that our relation to the world is not an opposition of two isolated sets of entities, as is supposed in the Cartesian tradition. Humans are still within the world; there is no clear gap between our Selves and our physical environment or more extremely, as Descartes thought, between our soul and body. Using the terminology suggested above, human activities are causally connected to other events in the world. Understanding our role in the world appropriately does not suppose that we must represent the way things are exactly; it seems to be sufficient if we are able to operate things using our causal power. Wittgenstein s account of following a rule directly depends upon his views on practice (the decisive passages are: [MS 117, 145], [MS 129, 121], [MS 142, 49], [MS 180a, 36v], [TS 220, 42-43], [TS 227a, 145], [TS 227b, 48], [TS 228, 78], [TS 239, 42]). I suggested above that at a non-representational level, causal relations warrant our connection to the world. At the representational level, however, causal relations often cannot put it into words i.e., the connection cannot be formulated propositionally. Wittgenstein s solution is that one can only recline upon the practice of a language game; one has to accept that what others in one s community do is appropriate. To avoid an infinite regress, reasoning shall be stopped somewhere. That is where the realm of practical knowledge begins, which lies outside the domain of reflection or reasoning and presupposes a subject to whose make-up traits other than mental essentially belong (Nyíri 1992, 47). Practical knowledge is temporally prior to theoretical knowledge. First one should acquire in practical terms how to relate oneself to the world (via conforming to a community). Acquiring theoretical knowledge can only be a further step. Moreover, one joins one s community by birth, knowledge acquisition without entrance to the practice is impossible. It seems therefore that practical knowledge is also logically prior to theoretical knowledge. 405

10 What has been done so far, however, has only inverted the dichotomy between theory and practice. Traditionally it was thought that theory is, logically at least, prior to practice; the above claims that the opposite is the case. But I shall say something more here. My claim is that practice is above and beyond a practical vs. theoretical opposition. In the traditional opposition, theoretical knowledge was prior to the practical in the sense that for an appropriate action, an appropriate representation of the world is required. It would be hard to deny, however, that humans actually do act, even without having a full, appropriate representation of the world. It would be also hard to deny that humans act mostly in an appropriate way in practical terms, at least. Therefore, theoretical knowledge is not a strong presupposition of activities in general it is supposedly a presupposition of only certain sorts of activities. On the other hand, even if theoretical knowledge in the traditional view is prior to practical knowledge, there is a clear gap between them. The explanation starts with the theory, yet an understanding of practice never seems to be developed in the same framework. As opposed to that, an analysis of practice offers a framework in which both theoretical and practical knowledge can be explained. As was suggested, knowledge as applicative is a way of having an appropriate effect on reality. Practical knowledge as applicative can easily be understood in this way: being able to cycle means that if one tried to cycle (under certain circumstances), one would cycle. Theoretical knowledge as applicative can also be explained similarly: it is in some aspects also a way of having (albeit more complex) appropriate effects on reality as well as on a scientific community. This is not the same as saying that theoretical knowledge is nothing more than having appropriate effects. For certain purposes, it is better to explain theories as representations of the world. But so far as theorising is an activity, and therefore theories are not only representations, there is a possibility of continuity between the theoretical and practical, via a unified explanatory framework, focused on an analysis of practice prior to both sorts of knowledge. A unified explanatory framework is at least aesthetically more elegant and economically more remunerative than a dual framework. Not to mention that a dual framework lacks an explanation of the relations between its isolated subjects, and therefore a further explanation is required as to why and how practical actions depend upon theoretical acquisition. In theory-based models, an explanation of practical knowledge is often missing, or manifestations of practical knowledge are claimed to be a set of uninteresting exceptions at best. On the other hand, even if in theory-based approaches theory is not a strong presupposition of practice, in practice-oriented views, practice is a strong presupposition of both practical and theoretical knowledge. A common practice, acquired directly by living in a community, is a necessary precondition of both human actions and theories. The reason is precisely that without theoretical certainty, the only source of guarantee can be practice: following something that has hitherto been working quite effectively. There is of course a real alternative which helps defend the priority of theoretical knowledge. This doctrine claims that the sort of knowledge labelled as being above the theoretical is something that is still given by our birth (or even logically prior to it): a claim that the most fundamental part of theoretical knowledge has been established well before; reality is represented approximately correctly, and all that has to be done is polish the surface, i.e. by creating theories which explain in more detail what has essentially been captured. I do not think that I shall argue against this view here. The present question is more fundamental 406

11 than whether or not this view would provide any illuminating insights. The question is rather whether we are interested in adhering to the old conceptual frameworks which have provided so much of theoretical importance but in practice offer ineffective work for philosophers, or are we satisfied with the conclusion that philosophy, like any other theoretical discipline, should and could produce practical applications grounded in practice. 1 References Apel, K.-O. Wittgenstein und Heidegger. Die Frage nach dem Sinn von Sein und der Sinnlosigkeitsverdacht gegen alle Metaphysik. Philosophische Rundschau 75, 56-94, Brandom, R. Tales of the Mighty Dead. Cambridge Mass.: Harvard University Press, Bloor, D. Ludwig Wittgenstein and Edmund Burke. In T. Demeter (Ed.). Essays on Wittgenstein and Austrian Philosophy. Amsterdam: Rodopi, Dreyfus, H. What Computers Can t Do: A Critique of Artificial Reason. New York: Harper & Row, 1979 (1 st ed: 1972). Havelock, E. Preface to Plato. Cambridge, Mass.: Harvard University Press, Murray, M. Heidegger and Ryle: Two Versions of Phenomenology. In M. Murray (Ed.). New Haven: Yale University Press, Nyíri, J. C. Tradition and Individuality. Dordrecht: Kluwer, Rorty, R. Philosophy and the Mirror of Nature. Princeton: Princeton University Press, Ryle, G. The Concept of Mind. London: Hutchinson s University Library, Schatzki, T. R. Introduction: Practice Theory. In T.R. Schatzki, K.K. Cetina, and E. von Savigny (Eds.). The Practice Turn in Contemporary Theory. London-New York: Routledge, Steup, M. The Analysis of Knowledge. In E. N. Zalta (Ed.). The Stanford Encyclopedia of Philosophy (Spring 2006 Edition). URL = < von Wright, G. H. Wittgenstein. Oxford: Blackwell, Wittgenstein, L. Nachlass. The Bergen Electronic Edition. Oxford: Oxford University Press, Subject Centre for Philosophical and Religious Studies, Higher Education Academy, UK and Department of Philosophy, University of Leeds, UK danka.istvan@gmail.com 1 Acknowledgement: I am grateful to Teodore R. Schatzki for valuable criticism of a previous version of this paper. 407