What's good about naturalistic epistemology


from issue 5/2012, pp. 18-31

“Naturalism or non-naturalism?” David Chalmers and David Bourget recently asked 931 professional English-speaking philosophers this question. The result: 49.8% of those questioned professed naturalism and 25.8% non-naturalism; the other participants in the survey abstained (see http: // philpapers. org / surveys / results.pl). This result shows two things: On the one hand, it confirms the impression that naturalism is the dominant philosophical worldview today, at least at English-speaking universities; on the other hand, it also makes it clear that not a few contemporary philosophers take decidedly anti-naturalistic positions. The fundamental debate about naturalism, it seems, is far from over. A look at the various branches of philosophy confirms this impression. In the philosophy of mind, metaethics, philosophy of mathematics and epistemology in particular, intense controversies are still waged over naturalistic and non-naturalistic approaches. But what do we actually understand by “naturalistic approaches”? What variants of naturalism are there, and what speaks for or against them?

Methodological and ontological naturalism

There is a first, fundamental difference between the methodological and the ontological (or metaphysical) Naturalism. Methodological naturalism is a metaphilosophical thesis that states that philosophical questions can ultimately only be answered using empirical methods (in the broad sense) and that philosophy should therefore be more closely based on the model of the natural sciences. The methodological naturalists thus turn against the project of a “first philosophy” - an a priori foundation of the natural sciences - but also against the method of conceptual analysis. In their view, philosophy is not about clarifying our concepts. Rather, it is a matter of examining the phenomena themselves and developing empirically supported, synthetic theories that adequately describe these phenomena (cf. [1], Section 2.1).

An example of the implementation of the methodological-naturalistic program is the naturalized epistemology (cf. [4], [3], [2]). Supporters of this approach deny that questions about knowledge and justification can be answered with traditional philosophical methods, e.g. by checking and systematizing conceptual intuitions. They consider knowledge and justification to be natural phenomena whose constitutive properties, just like the constitutive properties of light, water or mammalian blood, can only be determined empirically (cf. [3], Chapter 2). Whether epistemology thus becomes a chapter of descriptive psychology (cf. [4], p. 83) or retains its status as an independent normative discipline (cf. [3], p. 26) is a matter of dispute among representatives of naturalized epistemology .

At the center of this report, however, is not methodological but ontological naturalism. This is a thesis about the nature of the world that ranges from methodological assumptions to conceptual analysis and A prioriKnowledge is independent, even if it is often represented together with methodological-naturalistic conceptions.

Ontological Naturalism: What Are Natural Entities?

The ontological naturalist maintains that all entities that exist - all items, properties, facts and events - natural Entities are. There is nothing beyond natural reality, nothing beyond natural reality. In this general form, however, the naturalism thesis is too imprecise to be philosophically interesting. The naturalist must therefore also have a Naturalness criterion formulate - a condition that an entity must meet in order to be considered "natural".

Many authors try to specify the thesis of ontological naturalism by characterizing "natural entities" as objects of an ideal scientific theory. For example, Paul Moser and David Yandell write: “Ontological naturalism takes various forms. We will understand such naturalism in terms of this core view: every real entity either consists of or is somehow ontologically grounded in the objects countenanced by the hypothetically completed empirical sciences [...]. “([28], p. 4; cf. . Also [13], p. 14) According to this definition, the ontological naturalist allows two classes of entities: (i) the fundamental natural entities postulated by the ideal (complete and correct) scientific theory of our world - that is, presumably individual things like electrons, DNA molecules and platypus, properties like mass, pH and overall fitness, etc. -, and (ii) the derivative natural entities, which are composed of the fundamental entities or, to put it more carefully, in one ontological dependency relationship are related to these entities. The objects of the second category are not themselves postulates of the natural sciences, but are considered "natural" because they cannot exist independently of the fundamental natural entities. Their existence is therefore compatible with the basic naturalistic idea that there is nothing beyond or beyond the natural world. Uncontroversial examples of derivative natural entities are paper clips, telephone booths and power lines, as well as their characteristic properties; ambitious naturalists also include items such as nations, symphonies, and exchange rates in this category.

The only question is: Does the thesis formulated by Moser and Yandell really capture what philosophers usually understand by naturalism? Let us assume that interactionist substance dualism is true and that every person is a non-spatial Cartesian subject who acts by causally interacting with his body. Under these circumstances, it would be entirely conceivable that scientists would find empirical evidence for the interaction of mind and body - for example, by showing that events regularly occur in the brain that have no sufficient physical cause (i.e. are physically indeterminate), and that ours intentional body movements are always caused by neuronal events of this kind. In addition, there may be data on out-of-body experiences that confirm that people who have such experiences actually leave their bodies. B. Data showing that after out-of-body experiences people have knowledge that they could only have acquired by leaving their body. Under these circumstances, it would obviously be rational for neuroscientists to accept a substance-dualistic theory of mind ([34], p. 168; similar examples can be found in [5], section 1). Such considerations show that the naturalistic thesis as formulated by Moser and Yandell is compatible with the existence of non-corporeal souls - a result that consistent advocates of ontological naturalism cannot accept. Non-corporeal souls are a paradigmatic example of entities that exist beyond the natural world. A thesis worthy of the name “naturalism” must therefore rule out the existence of such entities.

In more general terms: a formulation of the naturalism thesis is only adequate if its truth coincides with existence more paradigmatic non-naturalEntities is incompatible. These entities include Cartesian subjects, but also divine beings, ontologically fundamental qualia (as postulated e.g. by David Chalmers in [11]) and ontologically fundamental moral facts (as assumed e.g. by G.E. Moore in [27]). From a genuinely naturalistic thesis it must follow that entities of this kind do not exist - this seems to be largely agreed among contemporary philosophers.

So what could an adequate naturalism thesis be? A second formulation that great facie is more promising than the first, operates with the concept of paradigmatic natural entities. This does not mean the postulates of ideal scientific theories, but entities that de facto Are the subject of current scientific research. Individual things such as electrons, DNA molecules and platypus can be cited as examples, as well as their characteristic physical, chemical and biological properties. After this second suggestion, the following applies: All entities that exist in our world belong of the same kind like paradigmatic natural objects (cf. [6], p. 223f). Clause (ii) of the first definition can be dispensed with here, since the criterion of equality of species can be interpreted in such a way that derivative natural entities - paper clips, telephone booths, high-voltage lines, etc. - are also included.

But this formulation of the naturalism thesis is also problematic. First The paradigmatic natural objects form a very heterogeneous class; therefore it remains rather vague which entities meet the criterion of species equality and can thus be described as "natural". Secondlyand this is the point, this version of naturalism is a dialectically unstable position. On the one hand, the thesis allows that there are non-physical properties that are ontologically fundamental, i.e. are not in any ontological relationship to physical facts. (Such properties are also referred to as "emergent".) Let us assume that emergent biological properties are involved in DNA replication - properties of the DNA or the enzymes involved that do not depend on the chemical-physical properties of these molecules (atomic structure , Conformation, polarization, etc.) and therefore cannot be explained by them. The existence of such properties would be compatible with the second variant of the naturalism thesis, since biological properties are among the paradigmatic natural entities. At the same time, the representatives of this variant must insist that their thesis emerges with the existence of psychological Properties is incompatible - otherwise the adequacy condition developed above for formulations of the naturalism thesis would be violated. So you have to assume that psychological properties should not be included in the list of paradigmatic natural entities, nor can they be counted in the same way as those entities that are on the list. However, this is not a stable position. The same arguments against emergent psychological properties can also be used against emergent chemical and biological properties - a point made in the section What are the arguments in favor of physicalistic naturalism? taken up again and carried out further.

From the natural to the physical

Contemporary ontological naturalists often go a step further for this reason: they deny that there are emergent properties - or other emergent entities - at any level of reality. Ontologically fundamental, so their thesis, are solely those physical Entities. Ontological naturalism almost inevitably leads to physicalistic naturalism or Physicalism.

This variant of ontological naturalism states that there are only two types of entities, namely (i) the fundamental natural entities that can be equated with physical entities - individual things like electrons, quarks and neutrinos, properties like mass, charge and spin, etc. -, and (ii) the derivative natural entities that are in an ontological dependency relation to the physical entities. The second category in this case includes not only paper clips, telephone boxes and other artifacts, but also all chemical, biological and psychological entities.

This formulation of naturalism first raises the question of how the meaning of the term “physical” can be specified in more detail. Here it makes sense to use the strategy that other naturalists use to define “natural” and simply characterize as “physical” those entities that of the same kind belong like a certain selection of paradigmatic physical objects (electrons, quarks, neutrinos) and properties (mass, charge, spin) (cf. [16], p. 7; [29], p. 419f). This selection is far less heterogeneous than the class of paradigmatic natural entities; it is therefore much clearer than in the case of the second definition of naturalism which entities meet the criterion of species equality and which do not. The existence certain The degree of vagueness that remains is undeniable. But that is not a problem for the physicalistic naturalist, since the paradigmatic the unnatural entities mentioned above (Cartesian souls, ontologically fundamental moral properties, etc.), quite certain according to the criterion of species equality Not belong to the physical entities. (The only problem arises in connection with proto-panpsychism - a theory according to which fundamental physical properties are also proto-mental properties. This approach should not be classified as “naturalistic”, so strictly speaking it is necessary to support the thesis of the to supplement physicalistic naturalism with the addition that the fundamental physical entities are non-mental; see [30], p. 421 and [29], p. 25. To simplify the presentation, however, this addition should be ignored in the following the position of proto-panpsychism is not discussed further, as it differs fundamentally from all other variants of non-naturalism and brings with it special problems; cf. the contributions in [37].)

The central question that still needs to be clarified concerns clause (ii) of the new definition: How can the “ontological dependency relationship” be characterized, which we are talking about here (as in the first formulation of the naturalism thesis)?

The physicalistic naturalism as a supervenience thesis?

According to physicalistic naturalism, higher-level entities must have a certain dependency relationship with the fundamental physical entities in order to be considered "natural". An obvious candidate for this relationship is the relation of Superveniencewhich has been discussed intensively in the last few decades (cf. [17], [25]). If, in order to avoid formal complications, we concentrate on the ontological category of facts, the naturalistic thesis would then be as follows: All facts are physical or supervise over physical facts. Since facts of all kinds trivially supervise over themselves (as the following explanations of the concept of supervenience make clear), this formulation can be simplified still further: All facts supervise over physical facts. This is a concise thesis - but is it also viable?

Supervenience is usually understood as a dependency relation between two 'levels' or 'levels' of reality: One level supervises over another precisely when there can be no differences on the higher ("supervening") level without there being also on the lower (“subsidizing”) level differences exist. This basic idea can, however, be specified in very different ways, which has led to a large number of supervenience definitions. For our discussion the relation is the global supervenience crucial. This relation exists (roughly speaking) between two levels of reality if and only if there is no possible world that is completely identical to our world in terms of the subsidizing level, but differs from our world in terms of the supervening level. Mental facts, for example, supervise globally over physical facts if and only if there is no (minimal) physical duplicate of our world in the set of possible worlds that differs from our world in terms of mental facts (cf. [16], p. 12) .
If the "ontological dependency relation" in clause (ii) of the definition of naturalism is understood as a global supervenience relation, then ontological naturalism is equivalent to the following thesis:

(GSP) All facts globally supervise the physical facts.

But can (GSP) - the thesis of global supervenience of all facts on the physical - can actually be equated with naturalism? At first glance, this seems plausible; but on closer inspection it turns out that such an equation is untenable. After all, the global supervenience thesis only states that all facts correspond in a certain way with physical facts covariate; on the character, the nature it says nothing of the covariating facts (cf. [17], p. 148). The assumption that there are paradigmatic non-natural facts that globally supervene over physical facts is therefore completely coherent (cf. [15], p. 24f).

This can be demonstrated very well using the example of non-naturalistic moral realism. The non-naturalistic realist postulates primitive, non-natural moral facts, but at the same time asserts that these facts supervise over descriptive facts. When there is a moral difference between two actions H and H* there, then, the plausible assumption, must be H and H* also distinguish in a non-moral sense: in their consequences, their psychological causes or any other descriptive property. But now speaks great facie nothing against the fact that the descriptive facts over which the moral facts superven are entirely natural. If the non-naturalistic moral realist accepts this, then his position is compatible with (GSP). But it would be wrong to say that this would make him a naturalist.

So the thesis of global supervenience should not be identified with naturalism - the truth of (GSP) is just necessary, but not sufficient for the truth of the naturalism thesis. In order to be able to formulate his position adequately, the physicalistic naturalist has to fall back on other theoretical resources.

Variants of naturalism

The following question now arises: What condition must facts that globally supervise over physical facts, additionally still meet in order to be considered natural facts? In the current discussion three different answers to this question are advocated, and each of these answers forms the basis for an independent variant of the naturalism thesis. The main features of these variants will be presented in the next subsections.

Analytical naturalism

The analytical naturalism is also under the terms "analytical physicalism" and "A priori-Physicalism "known; His representatives include Frank Jackson, Terry Horgan, David Lewis and (with restrictions) David Chalmers (cf. [12], [15], [16], [21]). Analytical naturalists believe that a analytical There is a relationship between the higher levels of description - everyday language, biology, psychology, etc. - and the basic, microphysical level of description. Their central thesis is slightly simplified: All true propositions can in principle be derived from a complete physical description of the world a priori be derived.

An often cited example is the statement "Water = H2O", which contains an expression with "Water" that belongs to a higher level of description. According to the Analytical Naturalist, “water = H2O” can be derived from a complete physical description of the world a priori can be derived because (i) the physical description contains numerous statements about the behavior of H2O molecules ("Accumulations of H2O molecules are liquid, translucent, freeze at 0 ° C, etc.") and (ii) it is an analytical truth is that water behaves in a certain way ("Water is the substance that is liquid and translucent, freezes when it is cold, etc."). It follows logically that water is H2O. (This representation of the derivation is greatly simplified; for various complications cf. [12].)

A great advantage of analytical naturalism is that its representatives can very well make plausible why their position as naturalistic Consider: If the existence of higher level facts follows conceptually from the existence of fundamental physical facts, then it seems clear that higher level facts are not entities that exist beyond natural reality or in any way go beyond that reality.

The analytical naturalist can rightly claim one reductiveExplanation to bid for higher level facts. However, he also commits to some controversial assumptions. Among other things, it assumes that all terms of the higher levels of description - such as "water", "warmth" or "opinion" - unite descriptive Own salary. Descriptive content is the prerequisite for the existence of substantial analytical truths, without which it would not be possible, sentences of higher levels of description a priori can be derived from a complete physical description of the world - as the above example of the derivation of "water = H2O" shows. This descriptivistic conception of terms is controversial. Overall, however, it is clear that analytical naturalism is an extremely attractive theoretical option for proponents of a naturalistic worldview.

Synthetic-reductive naturalism

However, many naturalists do not believe that all true propositions can be derived from a complete physical description of the world. They believe that there is a synthetic relationship, not an analytical one, between physical facts and facts on a higher level; therefore their position as "Synthetic Naturalism" (or "A posteriori-Naturalism "/"A posteriori-Physicalism "). Well-known representatives of this view are Ned Block, Richard Boyd, Hilary Kornblith, William Lycan and Robert Stalnaker (cf. [9], [10], [19], [24]).

The most important challenge for synthetic naturalists is to develop a plausible naturalness criterion for higher-level entities - a criterion that demands more than mere supervenience over the physical and less than conceptual deductions from the complete physical description of the world. The response of the Synthetic Naturalists to this challenge has been inconsistent. Essentially, two different versions of the synthetic approach can be distinguished - a reductive and a non-reductive version.

The Synthetic-reductive naturalism according to higher level entities are natural if and only if they coincide with physical entities identical are. Since the talk of "higher level entities" is misleading in this context, this suggestion can also be formulated as follows: Terms higher levels of description are acceptable if they refer to (possibly relatively complex) physical entities. For example, the existence of water is compatible with the synthetic-reductive thesis, since "water" refers to H2O.

But how can it be that “water” refers to H2O, although not propositions about water a priori can be derived from propositions about H2O? Synthetic reductive naturalists usually answer this question by responding to the causal theory of reference refer (cf. [20], [32]). According to this theory, proper names and terms for natural species (such as "water") refer not because they have a certain descriptive content, but because they are in a certain causal relation to their referent (cf. [10]) or because they are connected via a causal chain of communication acts with a "baptismal act" with which the reference of the expression was established (cf. [20], pp. 91-97). On the basis of this semantic theory, the synthetic naturalist can uphold the identity thesis and at the same time deny that the analytic relationships exist that are necessary for a A priori-Directability of sentences of higher levels are necessary.

The synthetic non-reductive naturalism

However, a serious objection has been raised against synthetic-reductive naturalism: the objection of the Multi-realizability of the properties of higher levels. This objection states, in a nutshell, that biological, psychological, and sociological characteristics cannot be identified with physical characteristics because they are multi-realizable are, d. H. because the individual instances of these properties are realized (or constituted) by completely different physical properties. So are z. B. Pain states in different people realized through different neuronal states; and far greater differences exist between the pain realizers of humans, reptiles, and octopuses. In more physical Regarding, it seems, these states don't have very much in common; an identification of the pain facts with physical facts is thus great facie locked out. There are indeed critics of this line of argument (cf. [18], [35]), but most contemporary philosophers consider the objection of multi-feasibility to be convincing.

This also applies to numerous followers of the synthetic-naturalistic approach who, for this reason, have a synthetic-non-reductive Defend version of naturalism. Synthetic-non-reductive naturalists reject - like the reductionists - the thesis of the A priori- derivability, and justify this with the causal theory of reference. However, they deny that higher-level terms refer to physical entities, and instead claim that such terms (often) refer to irreducible higher level entities To refer - to entities that supervise the microphysical facts, but cannot be identified with them.

The question now is: what speaks in favor of this position as a variant of the naturalism understand? What makes the irreducible higher level entities "natural" entities? The synthetic-non-reductive naturalist will answer first that the entities postulated by him, in contrast to the qualia of the property dualist or the moral facts of the non-naturalistic realist, are not ontologically fundamental, but are realized by physical entities. But this answer is only really satisfactory if it is supplemented by an analysis of the realization relationship - or at least by an informal explication of this relationship.

This supplement poses considerable problems for synthetic non-reductive naturalists. After all, there are some interesting explication suggestions, such as B. the (sketchy) characterization of the realization relation developed by Derk Pereboom and Hilary Kornblith (cf. [19], p. 130ff). According to this, properties of higher levels are realized through physical properties, if we can give a "constitutive explanation" on the physical level for each individual case of an explanation on the higher level, such as "He cries because he is in pain" - an explanation, which describes how the physical realizer of pain causes the physical realizer of 'screaming behavior' under the given conditions. According to Pereboom and Kornblith, this explanation does not pose a threat to non-reductionism because it cannot be generalized (each individual case requires a new "Constitutive explanation"), but it anchors properties and processes of the higher levels in the physical world and thereby makes them naturalistically acceptable.

In summary, it can be said that synthetic naturalism, despite the difficulties with which it undoubtedly has to struggle, is one of the naturalist's options worth discussing.

What are the arguments in favor of naturalism?

All variants of physicalistic naturalism are genuine naturalistic theses - theses that are incompatible with the existence of paradigmatic non-natural entities. The crucial question now is whether it is good Arguments for accepting one of these theses. Why should we believe that there are no non-natural entities? What speaks for the assumption that no facts exist that neither in a conceptual derivability relationship, still stand in the identity or realization relation to the physical facts?

In this section a general argument against unnatural facts is to be sketched (cf. [34]; special versions of this argument can be found in [7], [14], pp. 3-10, [36], p. 33- 38). The aim is to show that every form of non-naturalism is confronted with an almost insoluble dilemma - that Interaction dilemma. This dilemma arises because the non-naturalist has to answer the question of whether the non-natural facts postulated by him (a) act on physical facts or (b) have no physical effects. Either option, as we will see, creates serious problems.

Let us first consider the first horn of the dilemma. From assumption (a) it follows that non-natural and physical events exist ipso facto have non-physical causes. (These events are hereinafter referred to as “connecting events”.) Now it would undoubtedly be absurd to assume that every connecting event additionally has a sufficient physical cause for its unnatural cause. This would mean that a large class of events is systematically overdetermined, in each case by two entirely different, ontologically independent causes. An unexplained parallelism of this kind must be ruled out for methodological reasons (cf. [34], p. 182). Therefore, the non-naturalist who chooses Horn (a) is bound to the thesis that the connecting events are physical events, their occurrence - or more precisely, their occurrenceprobability - is not determined solely by physical conditions. He must therefore take the position that the principle of the causal closeness of physics is wrong (for a precise formulation of this principle cf. [31], p. 59f).

However, this position is empirically implausible. All special sciences, from chemistry to biology to psychology and sociology, work today with the basic assumption that the laws of physics also apply without exception in their area of ​​investigation, and so far no phenomena have been discovered that fundamentally call this assumption into question (cf. . [26], p. 160ff; [31], p. 55ff). Anyone who rejects the principle of causal unity is contradicting a well-established thesis of the empirical sciences. This is not a compelling objection to Horn (a), but it is a very strong one - especially since many non-naturalists claim that their arguments are directed against naturalism only, not against the natural sciences.

This brings us to option (b), the second horn of the dilemma. A non-naturalist who does not want to give up the principle of causal closeness in physics can deny that non-natural facts have physical effects. That means: He can either do one Epiphenomenalism represent and claim that the unnatural facts postulated by him are indeed produced by physical events, but do not have an effect on physical reality, or he can do one isolationist Defend the position that the unnatural facts in question have neither physical causes nor physical effects. With regard to qualia, most non-naturalists prefer the first option, with regard to normative and mathematical facts the second.

It is well known that positions of this kind lead to problems. Let us consider the example of qualia epiphenomenalism for which David Chalmers in The Conscious Mind argued. As Chalmers himself expressly admits, his theory is faced with a fundamental difficulty: that Problem of explanatory irrelevance (cf. [11], pp. 172-209). Obviously, if qualia epiphenomenalism is true, then qualia are irrelevant to the explanation of physical events.This also applies to those physical events over which our opinions and linguistic expressions supervise. It seems to follow from this that qualia are explanatory irrelevant for our expressions and opinions - also for our expressions and opinions about qualia. My statement "I feel stabbing pain in my left hand" can be fully explained by natural - and thus ultimately physical - causes, as can my opinion that I now have a red perception. Qualia play no role in the occurrence of these facts. This sounds extremely strange, and (at least) three questions immediately arise: (1) How can we successfully respond to Qualia in words and thoughts to refer towhen qualia have no causal or explanatory relationship to these words and thoughts? How is one in particular direct Can you refer to Qualia ("So this is how it feels to see red")? (2) How can we Knowledge or justified beliefs have about qualia when qualia are explanatory irrelevant to those beliefs? (3) Given the explanatory irrelevance of qualia, how does one explain why we have so many true opinions have about them? The qualia epiphenomenalist does not seem to have plausible answers to any of these questions.

Yet with such questions is everyone Faced non-naturalist who chooses option (b). Non-natural facts without physical effects are necessarily explanatory irrelevant to our opinions and expressions about them; It does not matter whether the facts are mental, normative or mathematical (cf. [7] on mathematical and [14], pp. 3-10, on moral facts). The second horn of the dilemma thus leads in any case to the problem of explanatory irrelevance.

Against this criticism of epiphenomenalist and isolationist positions, the objection is often that it inadmissibly presupposes causal theories of reference, knowledge or justification (cf. [33], [22], pp. 108-115, [11], p. 193) -196). This is incorrect for two reasons. First is only one in questions (1) and (2) Explanation demanded how reference to or knowledge of explanatory irrelevant entities is possible. This demand cannot be avoided simply by rejecting causal theories of knowledge and reference. (A simple analogy: someone who claims to have detailed knowledge of what is happening on distant extrasolar planets cannot simply answer the question "How can you know all this without having causal contact with these planets?" reject the causal theory of knowledge, this question is unjustified ".) Secondly This is the case with question (3) - which is not about reference, knowledge or justification, but about truth goes - quite obviously that no illegitimate preconditions are made (cf. [23], p. 71).

The non-naturalist is faced with a real dilemma: Both theoretical alternatives available to him - assumptions (a) and (b) - are subject to grave objections. Physicalistic naturalists, on the other hand, are faced with no analogous problem. They assume that there is a close relationship between higher-level and physical facts - realization, identity, or A priori- derivability -, therefore you can defend the causal-explanatory relevance of these facts without having to give up the principle of the causal closeness of physics. (Only in the case of non-reductive-synthetic variants do certain problems arise in this regard, but they are not insoluble; cf. [8], p. 327f). So there are good reasons in favor of physicalistic naturalism.

Objections to Naturalism

In the course of the history of philosophy, however, numerous objections to naturalism have also been formulated. Most of these objections take the following general form: “X is an apparently real phenomenon; X is not in any ontological dependency relation to the natural facts and is therefore not 'naturalizable'; so naturalism is wrong ”. Candidates for X are, among other things, (i) phenomenal states of consciousness or qualia, ie mental states that “feel a certain way”, (ii) intentional states, ie mental states that “deal with something” / one have representational content, (iii) normative facts such as facts of rationality and morality, (iv) modal facts, ie facts about necessity and possibility, (v) free will, and (vi) mathematical facts.

In order to make the thesis of the non-naturalisability of these phenomena plausible, non-naturalists have developed a number of complex arguments that cannot be dealt with in the context of this report. In general, however, it can be said that each of these arguments is controversial and that in each case the naturalist has several answer options. First the naturalist can try to develop a theory that X is real natural Phenomenon treated - be it in the form of an analytical, a synthetic-reductive or a synthetic-non-reductive naturalism. In order to be successful with this strategy, however, he must refute the arguments of the non-naturalists mentioned above. Secondly he can represent a non-cognitivistic theory with regard to X, according to which our opinions and statements about X do not have the function of describing reality, but rather serve a completely different purpose. Examples of this approach are expressivist theories of morality or quasi-realistic theories of modality. However, proponents of these theories must explain, among other things, why we usually feel this way seemsas if our opinions and statements about X had a descriptive function. Third the naturalist has the option of advocating a theory of error about X, i. H. a theory according to which X - contrary to our everyday beliefs - is not a real phenomenon. Examples of this are error theories of morality, but also eliminativistic approaches in the philosophy of mind, positions skeptical of freedom of will and fictionalistic theories of mathematics. Error theorists, however, need to formulate strong arguments for their position in order to make the rejection of common sense plausible, and they also need to explain why errors about the existence of X are so persistent and widespread.

In summary, it can be said that the integration of phenomena (i) to (vi) is a difficult, but certainly not a hopeless undertaking - especially since the history of science shows that some apparently emergent phenomena, such as the property of being alive, can be let naturalize.


Naturalism or non-naturalism? Even a brief overview of the debate makes it clear that this question is being discussed today with impressive acuteness and great argumentative finesse. Most of the participants in the discussion are not concerned with simply positioning themselves in one of the two camps, but rather with a critical and constructive examination of the philosophical opponent. Even if a consensus is far from in sight, the results of the debate so far are already of great value for philosophy: They show which fundamentally different conceptions of the world and of man are possible, and what speaks for and against these conceptions.


Peter Schulte is assistant at the Institute for Philosophy at the University of Erlangen-Nuremberg. He has published on the topic: Plea for a physicalistic naturalism, magazine for philosophical research 64, 2010 and ends and means in a natural world. Instrumental Rationality as a Problem for Naturalism? Paderborn: mentis, 2010.

Literature on the subject

General introductions:
[1] Papineau, David (2009): Naturalism, in E. Zalta (ed.): The Stanford Encyclopedia of Philosophy (Spring 2009 Edition), URL =

On methodological naturalism:
[2] Kitcher, Philip (1992): The Naturalists Return, Philosophical Review 101, pp. 53-114.
[3] Kornblith, Hilary (2002): Knowledge and its Place in Nature, Oxford: Oxford University Press.
[4] Quine, W.V.O. (1969): Epistemology Naturalized, in W.V.O. Quine: Ontological Relativity and Other Essays. New York: Columbia University Press, pp. 69-90.

On ontological naturalism:
[5] Beckermann, Ansgar (unpublished): Comments on Naturalism, URL =
http: //phillister.ub.uni-bielefeld. de / publication / 1199
[6] Beckermann, Ansgar (2001): Naturalism and Freedom. Reply to the comments by Geert Keil, Jasper Liptow and Gerson Reuter, Allgemeine Zeitschrift für Philosophie 36.2, pp. 217-237.
[7] Benacerraf, Paul (1973): Mathematical Truth, Journal of Philosophy 70, pp. 661-679.
[8] Bennett, Karen (2007): Mental Causation, Philosophy Compass 2, pp. 316-337.
[9] Block, Ned; Stalnaker, Robert (1999): Conceptual Analysis, Dualism, and the Explanatory Gap, Philosophical Review 108, pp. 1-46.
[10] Boyd, Richard N. (1988): How to Be a Moral Realist, in G. Sayre-McCord (ed.): Essays on Moral Realism, Ithaca: Cornell University Press, pp. 181-228.
[11] Chalmers, David (1996): The Conscious Mind, Oxford: Oxford University Press.
[12] Chalmers, David; Jackson, Frank (2001): Conceptual Analysis and Reductive Explanation, Philosophical Review 110, pp. 315-361.
[13] Goetz, Stewart; Taliaferro, Charles (2008): Naturalism, Grand Rapids, MI: Eerdmans.
[14] Harman, Gilbert (1977): The Nature of Morality, Oxford: Oxford University Press.
[15] Horgan, Terry (1983): Supervenience and Cosmic Hermeneutics, Southern Journal of Philosophy 22 (Supplement), pp. 19-43.
[16] Jackson, Frank (1998): From Metaphysics to Ethics. A Defense of Conceptual Analysis, Oxford: Clarendon Press.
[17] Kim, Jaegwon (1990/1993): Supervenience as a Philosophical Concept, in J. Kim (1993): Supervenience and Mind, Cambridge: Cambridge University Press, pp. 131-160.
[18] Kim, Jaegwon (1992/1993): Multiple Realization and the Metaphysics of Reduction, in J. Kim (1993): Supervenience and Mind, Cambridge: Cambridge University Press, pp. 309-335.
[19] Kornblith, Hilary; Pereboom, Derk (1991): The Metaphysics of Irreducibility, Philosophical Studies 63, pp. 125-145.
[20] Kripke, Saul A. (1972/1980): Naming and Necessity, Cambridge, MA: Harvard University Press.
[21] Lewis, David (1983/1999): New Work for a Theory of Universals, in J. Kim, E. Sosa (eds.) (1999): Metaphysics. To Anthology, Malden, MA: Blackwell, pp. 209-232.
[22] Lewis, David (1986): On the Plurality of Worlds, Malden, MA: Blackwell.
[23] Liggins, David (2010): Epistemological Objections to Platonism, Philosophy Compass 5, pp. 67-77.
[24] Lycan, William (1986): Moral Facts and Moral Knowledge, Southern Journal of Philosophy 24 (Supplement), pp. 79-94.
[25] McLaughlin, Brian; Bennett, Karen (2008): Supervenience, in E. Zalta (ed.), The Stanford Encyclopedia of Philosophy (Fall 2008 Edition), URL = .
[26] Melnyk, Andrew (2003): Some Evidence for Physicalism, in S. Walter, H.-D. Heckmann (ed.): Physicalism and Mental Causation, Exeter: Imprint Academic, pp. 155-172.
[27] Moore, G.E. (1903/1988): Principia Ethica, Amherst: Prometheus Books.
[28] Moser, Paul K .; Yandell, David (2000): Farewell to Philosophical Naturalism, in W.L. Craig, J.P. Moreland (ed.), Naturalism. A Critical Analysis, London: Routledge, pp. 3-23.
[29] Nimtz, Christian (2009): “Physical” and multi-realizability, or: two problems for physicalism solved, in J. Michel, M. Backmann (ed.): Physicalism, free will, artificial intelligence, Paderborn: mentis, p 17-36.
[30] Nimtz, Christian; Schütte, Michael (2003): On Physicalism, Physical Properties, and Panpsychism, Dialectica 57, 413-422.
[31] Papineau, David (2009): The Causal Closure of the Physical and Naturalism, in B. McLaughlin, A. Beckermann, S. Walter (eds.): The Oxford Handbook of Philosophy of Mind, Oxford: Oxford University Press, Pp. 53-65.
[32] Putnam, Hilary (1975): The Meaning of Meaning ’, in K. Gunderson (ed.), Language, Mind and Knowledge, Minneapolis, 131-193.
[33] Rosen, Gideon (2001): Nominalism, Naturalism, Epistemic Relativism, Philosophical Perspectives 15, pp. 69-91.
[34] Schulte, Peter (2010): Plea for a physicalistic naturalism, magazine for philosophical research 64, pp. 165-189.
[35] Shapiro, Lawrence (2000): Multiple Realizations, Journal of Philosophy 97, pp. 635-654.
[36] Tye, Michael (2009): Consciousness Revisited, Cambridge, MA: MIT Press.
[37] Journal of Counsciousness Studies 13, No. 10-11, 2006.