HABERMAS ON COMPATIBILISM AND ONTOLOGICAL MONISM Some problems

Philosophical Explorations, Vol. 10, No. 1, March 2007 HABERMAS ON COMPATIBILISM AND ONTOLOGICAL MONISM Some problems Michael Quante In a first step, I disentangle the issues of scientism and of compatiblism versus incompatibilism. I then analyse Habermas refutation of compatibilism and argue that his refutation of the compatibilism defended by Harry Frankfurt is not successful, since Habermas has shown neither that compatibilists have to defend scientism nor that compatibilism is committed to the claim that agents themselves must hold a compatibilist interpretation of their actions. In a third step, I discuss Habermas broad conception of nature and ask whether nature in this broad sense can be dealt with entirely within the observer perspective or not. I show that, in this respect, Habermas overall position is in need of further clarifications, which may also influence his treatment of compatibilism. KEYWORDS compatibilism; naturalism; determinism; scientism; nature; monism The starting point of Habermas s essay ( The Language Game of Responsible Agency and the Problem of Free Will ) is a debate that has been going on in Germany over the last five years. Several so-called brain-scientists have claimed that recent discoveries in neuroscience demonstrate that free will is an illusion. This, they claim, ought to have consequences for our daily practices of ascribing and taking responsibility (in fact it has been claimed that our system of criminal law has to be revised). Due to the authority that science has in our culture these pronouncements have not only been hotly debated among scientists and philosophers but have also gained the notice of a wider public as well. In the following I begin by disentangling the issue of scientism (as a philosophical position) from the classical problem of free will and determinism (Section 1). Then I will distinguish several theses and positions one can hold concerning the problem whether free will and determinism are compatible or not (Section 2). Finally, I try to show that the position Habermas defends in this essay faces several problems (Section 3). Habermas s overall argument has the following structure: philosophers who claim that free will and determinism are compatible can use one of three strategies. Habermas rejects all of them so compatibilism of free will and determinism is rejected completely. In order to assess whether Habermas s argument succeeds, we have to distinguish several compatibilist and incompatibilist claims that figure in Habermas s arguments. ISSN 1386-9795 print/1741-5918 online/07/010059-10 # 2007 Taylor & Francis DOI: 10.1080/13869790601170151

60 MICHAEL QUANTE Having made these distinctions in Section 2, I go on to argue, in Section 3, that Habermas s first argument against compatibilism is unsuccessful. 1 Either he has not shown that determinism is incompatible with our practice of ascribing and taking responsibility (what I call the incompleteness problem ) or his argument has implications that threaten to make his position inconsistent (what I call the inconsistency problem ). 2 As I will try to show at the end of my paper, this threatens to make Habermas s conception of ontological monism instable. Whether it is really inconsistent or not will depend on further clarifications Habermas needs to provide so that we are able to assess whether he has succeeded in reconciling ontological monism with the epistemic dualism of observer and participant perspective. 1. Scientism, Determinism and Compatibilism: A Shortcoming of the German Debate We can understand scientism as Habermas seems to do as subscribing to the following two theses: T-1 Everything there is can be explained fully within a nomological system of causal laws formulated exclusively in scientific vocabulary (as part of a scientific theory). T-2 To use Sellars s language, the scientific image (including the epistemology of the observer-perspective, the methodology of deductive-nomological explanation and the ontological commitments of science) dominates the manifest image (including the epistemology of the participant-perspective, the methodology of everyday action explanation and the ontology of persons and intentional states). The first thesis excludes every kind of ontological dualism and allows only for a kind of aspect-dualism or a dualism of stances (or perspectives). And the second thesis makes clear that in those cases where statements formulated from the manifest-image stance are incompatible with statements that are both formulated from the scientific-image stance and taken to be true according to best scientific standards, the former statements (and our practices depending on them) have to be given up. As I understand him, Habermas takes scientism to be committed to a third thesis: T-3 The ultimate causal laws formulated in science are (or will be) deterministic. If we keep the current German debate in mind, Habermas s starting point suggests itself, for that debate has centred on the traditional philosophical question concerning the compatibility (or incompatibility) of determinism with free will, that is, with our practice of ascribing and taking responsibility. But if we put aside the particular circumstances of that debate, it is not really fair to burden scientism with this third thesis, for at least three reasons, each of which underscores the importance of disentangling determinism and scientism. Only the first reason amounts to an objection to Habermas s arguments so far, insofar as he seems to hold that scientism is committed to determinism. The second and third reasons pose the question of whether determinism must be committed to scientism. 3 (i) If science were ultimately to demonstrate that we live in an indeterministic world, it should still be possible to defend scientism, as the philosophical position according to which no cognitive enterprises other than science will tell us the truth about our universe

HABERMAS ON COMPATIBILISM AND ONTOLOGICAL MONISM 61 (ii) (iii) and our place in it. It would be unfair to scientism to commit this position per definition to the view that our universe is of the deterministic kind. To put it the other way round: scientism would be a viable philosophical option in an indeterministic world, too. At least it should be as viable in such a world as it is in a deterministic world. There is a second reason not to follow Habermas in combining the question of scientism and the quarrel between compatibilists and incompatibilists in the way he does. In the history of philosophy the question concerning the relation between free will and determinism has been discussed on the basis of many different versions of determinism. Traditionally questions about the compatibility of human freedom and the existence of God or the problem of logical determinism have been very important. Clearly, the most plausible current candidates for holding the view that our universe is determined have to include the claim that there are deterministic laws of nature. But on the conceptual level that we address if we ask questions about compatibility, we should leave conceptual space for alternative ways of understanding the problem of free will and determinism. Narrowing down the quarrel between compatibilists and incompatibilists to the issue of deterministic laws of nature assumes that causal determinism is the only variant of determination that challenges our practice of taking and ascribing responsibility. There are at least two problems with building this assumption into scientism. On the one hand, it ignores the possibility that determinism (or fatalism) might be an integral part of religious (rather than scientific) beliefs that, if widely accepted, could threaten our practice of taking and ascribing responsibility. And on the other hand, the restriction of causality and causal explanations to efficient causality and nomological explanation comes down to the claim that the conception of causality and causal explanations in science has no alternatives. In debating the merits of scientism, we should avoid such restrictions and leave open the possibility that causality and causal explanations can be understood in a broader sense. The need for more conceptual space here can be seen from the following fact as well: the question whether free will and determinism are compatible can also be a philosophical challenge for philosophers who are not committed to scientism, understood in terms of T-1 and T-2 alone. This is evident if we take into account philosophers who do not understand determinism in terms of deterministic laws of nature. But this claim can be made plausible even if we accept this further premise. Neither compatibilists nor incompatibilists are committed to taking a stance on T-3 at all. Rather, they defend theses regarding relations between our daily practice of ascribing and taking responsibility, on the one hand, and the truth of determinism, on the other hand. They are not committed to defending or denying the truth of T-3 itself. For sure most compatibilists are motivated by their belief in determinism and many incompatibilists are motivated by their belief that we live in an indeterministic world. But to spell out the relation between determinism (broadly understood) and our practice of ascribing and taking responsibility might be a valuable philosophical project even if one is silent on T- 3 since one can hope thereby to uncover important features of our practice of ascribing and taking responsibility. The problems surrounding free will are complicated, and debates dealing with them are complex (or even muddled). To keep things as simple as possible here, I will identify determinism with T-3 and will ignore other aspects of the dispute between compatibilists

62 MICHAEL QUANTE and incompatibilists more broadly construed. But for the reasons given above I will not subscribe to the thesis that scientism has to include thesis T-3. 2. Which Compatibilism? Habermas himself warns us that we have to be aware of how we use the term compatibilism : A note here on the term compatibilism : since I want to suggest an affirmative answer to the question that I raised in the title, I am also a compatibilist of sorts. The human mind, with its complementary, interlocking epistemic perspectives, is part of the universe of nature. Where I depart from the compatibilist mainstream, however, is in rejecting the scientistic thesis that this universe is adequately characterized as the object domain of the established nomological sciences (on the model of contemporary physics). 4 For the purposes of my paper, this remark is of the utmost importance. Firstly, we can see here that Habermas ascribes thesis T-1 to scientism. Secondly, Habermas admits that there might be different sorts of compatibilism in play. Since this is indeed the case we should eliminate the ambiguity of the notion compatibilism by making explicit which claims of compatibility or incompatibility are under scrutiny. Thirdly, Habermas states that his own overall aim (expressed in the title of his paper) can be characterized as compatibilist in some sense. In this section I will try to make explicit which compatibilism- and which incompatibilism-claims are involved in Habermas s own position. Before I can start another point has to be addressed briefly since, fourthly, one might disagree with my claim that Habermas confounds the problem of scientism and the quarrel concerning free will and determinism: Habermas himself seems to agree that his rejection applies only to one particular variant of compatibilism. Although this is correct, it is no argument against my claims, since Habermas s counter-proposal does not affect the relation between scientism and determinism but only the relation between compatibilism and the notion of nature. Since it might be that the established nomological sciences will ultimately settle on indeterministic laws, Habermas s refutation of the scientistic concept of nature implies neither that this scientistic notion implies determinism nor implies that the broad notion of nature Habermas has in mind implies indeterminism. We simply have to keep apart the question of determinism/indeterminism and the question of nature. Habermas s claim is that we forego the basic scientistic assumption that nature, as it is conceived in nomological science, extends to the whole of what we encounter, in one way or the other, as nature (p. 38). This broad notion of nature refuses to restrict nature to the meaning it has in nomological science as such, and not necessarily to forms of science that are committed to deterministic laws. In the title of his paper Habermas himself refers to our daily practice of ascribing and taking responsibility as the language game of responsible agency. In the following I will use this phrase to formulate different compatibilist theses (CT) and incompatibilist theses (ICT) that need to be distinguished. I will start with those that normally are at stake in the free will debate: CT-1 ICT-1 The language game of responsible agency is compatible with determinism. The language game of responsible agency is incompatible with determinism.

HABERMAS ON COMPATIBILISM AND ONTOLOGICAL MONISM 63 Although these claims normally are the starting points in this debate it is important to distinguish them from two other theses: CT-2 The language game of responsible agency is compatible with our belief in the truth of determinism. ICT-2 The language game of responsible agency is incompatible with our belief in the truth of determinism. Distinguishing these four theses exploits the difference between the epistemic level and the ontological level. It is evident that one can hold both CT-1 and ICT-2 without being inconsistent, regardless whether or not we count that position as a variant of compatibilism, or not. But it must be clear which theses are involved. It might be objected that distinguishing between the epistemic level and the ontological level introduces a distinction incompatible with Habermas s arguments. But this is not the case, since Habermas uses this distinction in his arguments himself. In the passage quoted above he admits to be some sort of compatibilist and says that his particular thesis regarding compatibility is expressed in the title of his paper. Although it is posed as a question it is evident that Habermas has in mind the following claim: CT-H Epistemic dualism is compatible with ontological monism. We should notice that the distinction between the epistemic and the ontological level is used by Habermas himself. To understand CT-H we need to know what epistemic dualism and ontological monism mean. For the purpose of this section we do not have to be very precise concerning ontological monism and can take it as the claim that mind is a part of natural history (p. 39). For Habermas s ontological monism it is important to keep in mind his broad notion of nature. He claims that from other modes of encountering nature, we do get other concepts, for example, the concepts of natural evolution and natural history (p. 39). And he warns us against a premature answer to the question of how the nature of natural history, broadly understood, differs from the nature of the natural sciences (p. 39). As we will see below, offering no answer to this question might cause trouble for Habermas s own arguments in the end but we can ignore this for the moment. Though we have to emphasize once more that the distinction Habermas wants to establish here lies across the determinism indeterminism-question. Neither does the restricted conception of nature science holds need to be deterministic nor must the broad conception Habermas has in mind be indeterministic. To sum up: ontological monism should be understood as the claim that mind can be taken as a part of nature (in the broad sense). Throughout his paper Habermas distinguishes between the observer perspective and the participant perspective, the former being the epistemological and methodological stance of science (and scientism), the latter being the stance we take in our daily practice of ascribing and taking responsibility. The talk of epistemic dualism implies a special incompatibility thesis (ICT-H) Habermas defends in his essay: ICT-H The observer perspective and the participant perspective are incompatible. Both perspectives are incompatible in the sense that there is no getting around a dualism of epistemic perspectives that must interlock in order to make it possible for the mind, situated as it is within the world, to get an orienting overview of its own situation (p. 35). Thus the participant perspective cannot be reduced to the observer perspective

64 MICHAEL QUANTE and there cannot be a third perspective that fuses them into a new unified perspective. Both perspectives are irreducible (thus dualism) and necessary for getting a complete overview of mind s place in nature. This means that we cannot take both stances at once, but have to switch from one perspective into the other. The philosophical challenge is to develop a plausible theory of how this epistemological interlock might work. Habermas claims that this epistemological incompatibility (i.e. ICT-H) is compatible with ontological monism (i.e. CT-H), if we presuppose a broad notion of nature. 5 3. Habermas s Rejection of the First Strategy of Compatibilism: A Problem Habermas discusses compatibilism as one of two strategies for showing that the observer and the participant perspective can, in principle, be reconciled while acknowledging that we can and must understand ourselves as part of nature. Compatibilism is the option that looks for a conceptual route (p. 26) out of the problem. The other strategy Habermas discusses is the ontological one, which, according to Habermas, has two variants: reductive and non-reductive materialism. 6 Since the alternative strategy is characterized as ontological it is plausible to understand Habermas s characterization of the compatibilists strategy as a conceptual route as including the distinction of epistemological perspectives. This interpretation is well founded since Habermas s discussion of the three strategies of compatibilism itself focuses on epistemological questions. Starting from the classical problem of how to reconcile determinism with our practice of ascribing and taking responsibility, Habermas discusses the classical compatibilist solutions developed by G. E. Moore and Harry Frankfurt, who aim to show in different ways that our practice does not presuppose the existence of alternate possibilities. Habermas takes this as the first of three typical variations (p. 27) of compatibilism. And he rejects this strategy because they blur the transition from the participant perspective to the observer perspective (p. 27). According to Habermas the solutions Moore and Frankfurt offer are not satisfying: The compatibility of determinism and a sense of freedom thus remains at best a compatibilist truth about agents (p. 28). Frankfurt s scenarios aim to show that the existence of alternate possibilities is not a necessary condition for our ascribing and taking responsibility. Therefore CT-1 is true. In these cases it is also true that determined agents do not know that they have no alternatives. We can, as Habermas seems to suggest, take this feature as expressing ICT-2: for an agent taking the participant perspective it is impossible to believe that there are no alternatives open to him. 7 The compatibilist s solution thus includes an incompatibilist thesis which excludes belief in the truth of compatibilism. On Habermas s interpretation of Frankfurt s account, then, compatibilism is committed to the conjunction of theses CT-1 and ICT-2, which combine a compatibilist and an incompatibilist claim. (I will call this CON in the following.) And Habermas s objection to CON is that this can be at best a compatibilist truth about agents rather than a truth for them. The qualification at best indicates that one might deny that CON is a compatibilist position at all, since it includes an incompatibilist claim, too. Since this labelling is not important for my own point, I admit that we can qualify CON as a compatibilist position. And Habermas has good reasons to do the same, since he qualifies his own position as compatibilism of some sort although it contains at least implicitly an incompatibility

HABERMAS ON COMPATIBILISM AND ONTOLOGICAL MONISM 65 claim. But does Habermas have good reasons to claim that CON is not a sufficient answer to the compatibility problem? To see things more clearly we have to introduce one more distinction. Up to this point we have focused on three theses: CT-2 The language game of responsible agency is compatible with our belief in the truth of determinism. ICT-2 The language game of responsible agency is incompatible with our belief in the truth of determinism. CON CT-2 and ICT-2 are both true. Now it is time to ask what our means here? For present purposes two interpretations must be distinguished. On one reading, our refers to us as philosophers; on the other reading, our refers to us as actors who take the participant perspective. Thus we get four theses: CT-2.1 The language game of responsible agency is compatible with our believing (as philosophers) the truth of determinism. CT-2.2 The language game of responsible agency is compatible with our believing (as actors taking the participant perspective) the truth of determinism. ICT-2.1 The language game of responsible agency is incompatible with our believing (as philosophers) the truth of determinism. ICT-2.2 The language game of responsible agency is incompatible with our believing (as actors taking the participant perspective) the truth of determinism. Given that a deterministic interpretation of the mind can be given only in the observer perspective, ICT-2.2 is plausible. And, given his premises, Habermas is committed to it. But we should notice two points. First, Habermas s refutation of CON depends on his reading ICT-2 as ICT-2.2. Since no argument is given for this claim, compatibilists can reply to Habermas s objection that CT-2.1 is all they need and that this thesis is compatible with ICT-2.2. Second, for his refutation, Habermas needs only ICT-H, since the incompatibility of observer perspective and participant perspective is doing all the work here. Determinism is important only because it connects the compatibilist s argument to the observer perspective. As far as I can see, there are two options. Habermas can accept that determinism implies the observer perspective, 8 in which case we need further arguments, since only one variant of determinism is considered in this essay. Alternatively, he can refrain from committing himself to the thesis that determinism implies the observer perspective, in which case his objection is not a rejection of CT-1 at all. Therefore Habermas faces the incompleteness problem here anyway. Since his strategy against compatibilism and against reductive or non-reductive materialism requires rejecting each of the three compatibilist strategies, this incompleteness is a serious threat to Habermas s overall argument. With respect to the classical problem of free will and determinism this means that Habermas s critical diagnosis so far does not provide good reasons for the claim that determinism (in any form) and the language game of responsible agency are incompatible. His argument works, if I have got it right, only for those conceptions of determinism that imply that determinism requires the observer perspective. The incompatibilist work Habermas

66 MICHAEL QUANTE is doing in his rejection of compatibilism is done by his ICT-H, which is more general than ICT-1 and ICT-2. Habermas himself states that other forms of determinism that are operative in developmental psychology, sociology or economics cannot really threaten the language game of responsible agency since the participant perspective is still implicitly included there. Thus Habermas seems to see no problem in taking these forms of self-explanation as compatible with our practice of ascribing and taking responsibility. But the problem vanishes only if determinism takes the specific form of nomological determinism that is conceptually tied exclusively to the observer perspective. Habermas s overall position, stated above as CT-H, implies ICT-H since an irreducible dualism of observer and participant perspective is presupposed. And CT-H includes an ontological claim since ontological monism is an ontological thesis about the mind. Although I cannot discuss Habermas s conception of nature and ontological monism here in detail, there seem to be two options: either ontological monism is tied to the observer perspective or it is not. Given that he defends a broad notion of nature and a conception of natural history in this essay it seems plausible to start with the assumption that Habermas is not committed to the claim that the truth of ontological monism has to be spelled out entirely within the observer perspective. If the conception of natural history or evolutionary theory Habermas has in mind allows that nature in the broad sense is not exclusively the domain of the observer perspective, three options are open: (i) Habermas could claim that we can develop a unified perspective in which nature (in this broader sense) can be grasped. This option seems to be blocked by ICT-H and by Habermas s idea that nature can be fully understood only in terms of an interlocking of the observer and participant perspectives. Moreover, the conceptual framework Habermas presupposes in his arguments would thereby dissolve, since such a unified perspective would overcome scientism as it is defined here (and would require an as yet unknown form of science). As a result, this option cannot be very attractive for Habermas. (ii) Alternatively, he could claim that ontological monism can be understood entirely from the participant perspective. This would be a strong claim and would render meaningless the problem Habermas is dealing with in this essay. If we reject the thesis that the mind can be described from the observer perspective at all we would not even need a compatibilism of the sort Habermas offers us. (iii) If Habermas sticks to the more restricted claims that ontological monism does not entail the exclusive authority of the observer perspective, he leaves a third alternative open. We could distinguish between the participant perspective of common sense and a scientific participant perspective which is the epistemological and methodological stance presupposed by evolutionary theory and which is needed to tell the natural history of the mind. This seems to be the most plausible strategy if one accepts that we need a broad notion of nature. But as far as I can see it will not help Habermas since now we have to reconsider his incompatibility thesis ICT-H. Either the scientific participant perspective is the unified perspective that overcomes the dualism of the participant perspective of common sense and observer perspective, in which case (as stated above) the overall conceptual framework underlying Habermas s arguments erodes. Or the scientific participant perspective is taken to be compatible with the observer perspective, in which case Habermas either loses the incompatibilist part of his overall position or ends up with the view that the participant perspective of common sense is incompatible with the scientific participant perspective, a view that would make it hard to see how ICT-H (read that way) could be reconciled with

HABERMAS ON COMPATIBILISM AND ONTOLOGICAL MONISM 67 ontological monism. Alternatively, we could claim that the scientific participant perspective is incompatible with the observer perspective. This would mean that the incompatibility of perspectives is anchored within ontological monism in such a way that the story about mind s place in nature that is told from the scientific participant perspective would be incompatible with the story about mind s place in nature that is told from the observer perspective. 9 The broad notion of nature would then be in danger of becoming an inconsistent concept and its inner tension might give rise to ontological dualism, on the one hand, and scientism, on the other hand. So it seems that any attempt to avoid treating ontological monism as committed to the observer perspective seems to make Habermas s arguments either pointless or even inconsistent. Let us take for granted, then, that ontological monism is entirely tied to the observer perspective. If we accept this, we can see that Habermas s arguments against the first compatibilist strategy dissolve, since we get the result that CT-H can be a truth about us as agents but not a truth for us as agents. As philosophers, we can accept the truth of CT-H and CON or not (we do not have to decide this issue; it is sufficient to see that both stand or fall together). As actors from within the participant perspective of common sense, we cannot hold the truth neither of CON nor of CT-H. Either Habermas accepts that compatibilism as a philosophical thesis is established even if it cannot be part of our self-understanding as agents. In this case the compatibilism of Frankfurt is as plausible as the compatibilism of Habermas. Or he demands that a satisfying compatibilist account must be such that it can be integrated into the participant perspective of common sense, in which case then neither CON nor CT-H is satisfying. So my overall conclusion is that either Habermas does not have a convincing refutation of compatibilism or cannot defend his own compatibilist solution. A short concluding remark: the difficulties I find in Habermas s overall position and the objections formulated in this comment are restricted to his rejection of compatibilism. They should neither be taken as objections against Habermas s arguments against scientism nor as a repudiation of his broad concept of nature. My arguments are compatible with the positions Habermas defends on these issues. I am in deep agreement with the thesis that nature in its totality cannot be reduced to what can be conceived by science (neither deterministically nor indeterministically conceived) but this has to be examined thoroughly, as part of a philosophical enterprise geared toward answering the question of how to put the different perspectives together. If we want to follow this path, as I think we should, we must not burden our enterprise with the claim that the solution we will find as philosophers must be integrated directly into our common sense participant perspective. ACKNOWLEDGEMENTS Thanks to Joel Anderson and Marcus Willaschek for many helpful comments. NOTES 1. Due to the overall structure of his argument this implies that Habermas has not refuted compatibilism since the first strategy still is open for compatibilists. 2. As we will see it is difficult to decide this since Habermas s position defended in this essay can be spelled out in different ways.

68 MICHAEL QUANTE 3. It is not clear whether Habermas would make this claim (as far as I can see his arguments do not commit him to hold that determinism always implies scientism); thanks to Marcus Willaschek who prompted these clarifications. 4. Habermas, Note 30. 5. Joel Anderson has objected that ICT-H is not a claim of incompatibility but only of incommensurability since Habermas does not deny that both kinds of descriptions can deliver true propositions. Although I agree with Anderson in this and speak of epistemic incompatibility therefore (having a broader notion of incompatibility in mind), I think it is better to characterize ICT-H as an incompatibility claim. My reason is that in Habermas s analysis of the criminal law discourse (Section 2 of his essay) he seems to presuppose that both perspectives can interfere. This could not be the case if incommensurability is all that is at stake here (see Note 8 for a similar worry concerning Habermas s ontological monism). 6. If I understand the overall argument of Habermas s essay correctly these two are the only strategies within the paradigm of scientism. 7. A hidden premise operative here is that determinism is incompatible with alternate possibilities. Since my objection does not depend on this aspect of Habermas s arguments I will not discuss this premise and will accept that a deterministic interpretation of the mind can be had only within the observer perspective. If determinism and the existence of alternate possibilities are compatible thesis ICT-2 becomes implausible. 8. It is important to see that determinism is not thereby committed to scientism, since no commitment to T-2 has been undertaken so far. 9. Saying that both stories do not deal with the same ontological realm would introduce that kind of dualism Habermas rejects in committing himself to ontological monism. Michael Quante, Philosophisches Seminar, Universität zu Köln, Albertus Magnus Platz, 50923 Köln, Germany. E-mail: mquante@uni-koeln.de