The indispensability of the manifest image

Abstract

It is very contentious whether the features of the manifest image have a place in the world as it is described by natural science. For the advocates of strict (or scientific) naturalism, this is a serious problem, which has been labelled ‘placement problem’. In this light, some of them try to show that those features are reducible to scientifically acceptable ones. Others, instead, argue that the features of the manifest image are mere illusions and, consequently, have to be eliminated from our ontology. In brief, the two options that are open to strict naturalists for solving the placement problem are ontological reductionism and eliminativism. Other advocates of naturalist philosophy, however, claim that both these strategies fail and, consequently, opt for ‘mysterianism’, the view according to which we cannot give up the recalcitrant features of the manifest image even if we are not able to understand the ways (which certainly exist) in which they could be reduced to the scientific features. Mysterianism has the merit of facing the difficulties that whoever wants to explain reductively, or explain away, the features of the manifest image encounters. It is also a defeatist philosophical view, though, since it considers the most important philosophical problems as unsolvable mysteries. For this reason, I argue that mysterianism can also be taken as a reductio of strict naturalism, given its presumption that all phenomena are either explainable by the natural sciences or to be rejected as illusory. In this article, it is argued that the failures of reductionism, eliminativism and mysterianism should teach us that both the scientific image and the manifest image of the world are essential and mutually irreducible but not incompatible with each other. To support this claim, in the second part of the article, the case of free will is discussed.

Keywords

eliminativism free will manifest image mysterianism naturalism

1. Strict naturalism and the placement problem

Wilfrid Sellars famously claimed that the conceptual tension between the ‘manifest image’ (the world as conceived when one employs the normative categories originating from common sense) and the ‘scientific image’ (the world as it is understood by the natural sciences) could be solved by a ‘stereoscopic vision’, in which the two images were fused into one.¹ It has been highly debated, and still is, whether and how one can attain such a project, and it is also uncertain whether Sellars himself stayed consistent with it. He famously wrote, for example, that ‘Speaking as a philosopher, I am quite prepared to say that the common sense world of physical objects in Space and Time is unreal – that is, that there are no such things’.² In this perspective, our ontology should be entirely defined by the scientific view of the world, whatever this implies for the plausibility of the stereoscopic project. Be that as it may, in the contemporary English-speaking philosophical world, the thesis that the natural sciences (and possibly physics alone) have a monopoly on questions regarding ontology is arguably a majority view; yet philosophers who defend this view have different ideas about how it should be interpreted. More specifically, if is undisputed that prima facie the features that compose the manifest image are very different from the scientific features (because they are characterized by a peculiar normative and/or intentional and/or phenomenological and/or abstract character), it is very contentious whether they actually have a place in the natural world, and if so which one. A clear-cut presentation of this problem has been offered by John Searle:

How can we square a conception of ourselves as mindful, meaning-creating, free, rational, etc. agents with a universe that consists entirely of mindless, meaningless, unfree, nonrational, brute physical particles?³

Huw Price has named this puzzle ‘placement problem’. The philosophers who take the natural sciences as the arbiters of all ontological matters (let’s call them ‘strict naturalists’) have two main options regarding the features of the manifest image. First, they may try to show that those features are reducible to scientifically acceptable ones. Second, strict naturalists may argue that the features of the manifest image are mere illusions – and in this case, they have to show that these features can be satisfyingly eliminated from our ontology. In brief, the two options that are open to strict naturalists for solving the placement problem are ontological reductionism and eliminativism. Let’s consider them in turn.

2. Reductionism and eliminativism

Ontological reductionism has been a familiar trend in philosophy for several decades.⁴ To cite only some examples: in philosophy of mind, the mind–brain type-identity theories (in both the original Australian version and the species-specific version more recently offered by Jaegwon Kim and others)⁵; William Lycan’s attempted the reduction of consciousness and qualitative mental content⁶; in philosophy of mathematics, Penelope Maddy’s naturalism⁷; in ethics, the many attempts to reduce moral properties to natural properties⁸; in esthetics, the remarkable recent success of neuroesthetics, a very trendy reductionist approach⁹ – and the list could go on and on.

Not a few other philosophers, however, are sceptical about these reductionist attempts and would agree with a harsh judgement concerning them that Tyler Burge gave some years ago:

The flood of projects…that attempt to fit mental causation or mental ontology into a ‘naturalistic picture of the world’ strike me as having more in common with political or religious ideology than with a philosophy that maintains perspective on the difference between what is known and what is speculated.¹⁰

Hilary Putnam was even harsher when he claimed that, ‘None of these ontological reductions gets believed by anyone except the proponent of the account and one or two of his friends and/or students’.¹¹ If Putnam has a point here (as I think he does), one can say that, even if reductionism is still very popular as a general philosophical stance, when one looks at its concrete applications in specific philosophical areas, one notices that the enthusiasm is much less marked. Unsurprisingly, then, a growing number of strict naturalists, instead of embracing reductionism, take the second route available to them, claiming that the features of the manifest image that are recalcitrant to reduction are indeed mere illusions that should simply be eliminated from our ontology.

Alex Rosenberg, one of the most vocal proponents of this view, recently wrote:

Science forces upon us a very disillusioned ‘take’ on reality. It forces us to say ‘No’ in response to many questions to which most everyone hopes the answers are ‘Yes’. These are the questions about purpose in nature, the meaning of life, the grounds of morality, the significance of consciousness, the character of thought, the freedom of the will, the limits of human self-understanding, and the trajectory of human history.¹²

Versions of eliminativism have been offered in most philosophical fields. According to the eliminativists, if the features of the manifest image that are resistant to reduction were real, they would be ‘queer’ entities, to use John Mackie’s famous definition¹³ – that is, they would be wholly unfit to occupy a place in the natural world. Consequently, if we want to stay faithful to principles of naturalistic philosophy, we should think of them as illusions or fictions and eliminate them from our ontology.

Moral philosophy is one of the fields in which eliminativism has flourished the most. The placement problem in this field has been clearly presented by Simon Blackburn: ‘The problem is one of finding room for ethics, or placing ethics within the disenchanted, non-ethical order which we inhabit, and of which we are a part’.¹⁴ Mackie – who sees no hope in the reductionist approach – offered the paradigmatic framework for eliminativism in this field: ‘If there were objective values, then they would be entities or qualities or relations of a very strange sort, utterly different from anything else in the universe’.¹⁵ In this light, values should be eliminated by our ontology, and, relatedly, no genuine explanations should presuppose their reality. More specifically, Mackie defends the particular form of eliminativism called ‘ethical fictionalism’, which reconciles cognitivism and antirealism. In his view, ‘thick’ ethical concepts such as ‘cruel’ are nothing more than natural concepts, but when one applies to them ‘thin’ moral concepts such as ‘wrong’ or ‘bad’ – whose would-be referents are ontologically unacceptable – one gets statements that are, at the same time, ‘truth apt’ and necessarily false. Besides Mackie’s fictionalism, other eliminativist projects are common today in moral philosophy, such as sentimentalism, quasi-realism, emotivism and biological naturalism.¹⁶

A similar fictionalist strategy was adopted by Hartry Field with regard to mathematical entities.¹⁷ Field view offers a clear-cut response to Paul Benacerraf’s famous puzzle,¹⁸ according to which it is utterly mysterious how we could know anything about mathematics, if the mathematical entities were abstract and consequently unable to participate in any causal interaction with us. In this perspective, Field writes, ‘What my anti-realism involves is a disbelief in mathematics. Or at least, it involves a disbelief in mathematics if mathematics is taken at face value’.¹⁹ The consequences of this view are striking, but Field is ready to accept them:

The sense in which 2 + 2 = 4 is true is pretty much the same as the sense in which ‘Oliver Twist lived in London’ is true: the latter is true only in the sense that it is true according to a certain well-known story, and the former is true only in that it is true according to standard mathematics.²⁰

With regard to the mind–body problem, influential versions of eliminativism have been offered by Paul Churchland, Patricia Churchland, and Stephen Stich.²¹ According to these philosophers, a mature science of human thought and action will have nothing to do with the common-sense items postulated by intentional psychology, which are nothing more than pseudoscientific relics such as the concept of ether. In this light, concepts such as ‘belief’ or ‘desire’ do not correspond to anything in reality and the features that would allegedly correspond to them should not have any place in our ontology. Intentional psychology, then, is

A false and radically misleading conception of the causes of human behavior and the nature of cognitive activity. On this view, folk psychology is not just an incomplete representation of our inner natures; it is an outright misrepresentation of our internal states and activities.²²

Other ambitious projects aim at showing that the ontological elimination of consciousness²³ and the self is the best theoretical option we have.²⁴ However, proposing a solution of the placement problem by appealing to the eliminativist strategy is simple; defending that strategy in a convincing way is a different matter. As John Erman wrote once:

It seems that the attempt to locate human agents in nature either fails in a manner that reflects a limitation on what science can tell us about ourselves, or else it succeeds at the expense of undermining our cherished notion that we are free and autonomous agents.²⁵

Accepting eliminativism in a consequential way, we should in fact give up the idea that we are free, autonomous and responsible; that our words have meaning; that qualia have some reality; that abstract entities may exist; that some actions can be objectively bad and so on. Unsurprisingly, some advocates of strict naturalism are pessimistic about the perspectives of both reductionism and eliminativism concerning the features of the manifest image. These philosophers defend mysterianism, that is, the view that we cannot give up the recalcitrant features even if we are not able to understand the ways (which certainly exist) in which they could be reduced to the scientific features.²⁶ The inspiration for this view comes from Chomsky’s famous distinction between ‘problems’ and ‘mysteries’²⁷: the former are questions that we are in principle able to solve (such as Goldbach’s conjecture or the possibility that there is a ninth planet in the solar system); the latter are questions that, because of the limitations of the human species, we are and always will be unable to solve. And how to locate the recalcitrant features of the manifest image in the natural world will always be a mystery for us – exactly like proving Pythagoras’s theorem would be a mystery for the species canis lupus familiaris.

Mysterianism has the merit of facing the difficulties that whoever wants to explain reductively, or explain away, the features of the manifest image encounters. It is also a defeatist philosophical view, though, since it considers the most important philosophical problems as unsolvable mysteries (in Problems of Philosophy, McGinn mentions as mysteries consciousness, the self, meaning, free will, the a priori, and truth: what is left for philosophers to do?). For this reason, mysterianism can also be taken as a reductio of strict naturalism, given its presumption that all phenomena are either explainable by the natural sciences or to be rejected as illusory.²⁸

Let me appeal to a specific case study, that of free will, to make this point clearer.

3. The case of free will

Free will has been rejected as a mere illusion by many contemporary thinkers, including Galen Strawson, Derk Pereboom, Saul Smilansky, Daniel Wegner and Sam Harris.²⁹ A recent book edited by Gregg Caruso, appropriately titled Exploring the Illusions of Free Will and Moral Responsibility, offers an excellent survey of the eliminativist positions regarding free will.³⁰

Free will is an essential component of the manifest image. Several other key features of that image hinge on it, including responsibility, retribution, the dignity of life and arguably all the notions connected with the idea of agency. How is it, then, that so many philosophers have abandoned Sellars’s ideal of a stereoscopic vision and taken the eliminativist route with regard to free will?

Some (such as Galen Strawson and Pereboom) have been convinced by conceptual arguments that free will is impossible whether the world is deterministic or indeterministic; others (such as Wegner and Harris) have been convinced by the evidence coming from social psychology, neuroscience and genetics. This is not the place to discuss these lines of argument (even if, in my view, they can be contested).³¹ Instead, what I want to do is to show how disruptive the abandonment of the idea of free will would be, especially with respect to our theory of punishment, notwithstanding the declarations to the contrary of many eliminativists, such as Joshua Greene and Jonathan Cohen, who wrote:

At this time, the law deals firmly but mercifully with individuals whose behaviour is obviously the product of forces that are ultimately beyond their control. Someday, the law may treat all convicted criminals this way. That is, humanely.³²

Greene and Cohen’s view is that the abandonment of the idea of free will should generate the abandonment of retributivism – that they take to be an obscurantist view of punishment, rooted into obsolete notions such as desert and guilt. On this perspective, the only acceptable way of looking at punishment is purely utilitarian: we should punish people when it is useful to punish them – that is when the general happiness of society is increased (or the general suffering is decreased) because someone is punished. This view is shared by many other philosophers and scientists, such as Adrian Raine, M.L. Corrado, Michael Gazzaniga and Derk Pereboom.³³

In my view, however, a society in which punishment was distributed only on utilitarian grounds would be very far from being ideal. To understand why, one should notice that the retributivist ideal is constituted by two components. The first component is positive and states that all individuals who are guilty deserve to be punished, The second component is negative and it states that only the individuals who are guilty deserve to be punished but not too severely.³⁴ Both of these components hinge on the notion of merit, which in turn presupposes the notion of moral responsibility and, consequently, the notion of free will. Agents who freely perform some wrongdoings are morally responsible for them; consequently, those agents (and no others) deserve to be punished for their wrongdoings.

The positive component of retributivism motivates the strictness of that view, since it states that justice requires the punishment of everybody who is guilty – independently from the actual practical consequences of the punishment. The negative component of retributivism, instead, acts as a safeguard for justice for two reasons. First, it bars scapegoating, since it states that nobody who does not deserve to be punished should be punished, even if such punishment would increase public utility. Second, the negative component of retributivism bars excessive punishment – that is, it refuses too severe and inhumane punishment (such as torture), even when it would produce social benefits. One can imagine, for example, a situation in which torturing terrorists until they confess their plans may produce a net benefit for society. However, today any civilized juridical mind would (or should)³⁵ deny that torture is an unacceptable form of punishment, since nobody ever deserves to be punished in that way – which is just another way of stating the negative component of retribution. Thus, pace Greene, Cohen & Co., it seems that the notion of ‘desert’ (with its conceptual correlates, such as ‘responsibility’, ‘guilt’ and ‘innocent’) cannot be abandoned without compromising our ideal of justice.

An advocate of a purely utilitarian conception of punishment could respond to this objection by abandoning standard ‘act utilitarianism’ – according to which actions are moral as long as they maximize general utility (or minimize general suffering) – and embracing ‘rule utilitarianism’, the view that the morality of an action is determined by its compliance with the norms that maximize overall happiness.³⁶

Some reflection, however, shows that even rule utilitarianism cannot solve the problems of scapegoating and excessive punishment.³⁷ We can in fact conceive of cases in which the general utility would be maximized by accepting a rule that envisions the possibility, in determinate conditions, of punishing an innocent person or dispensing an excessive punishment. A clear example in this sense is offered by the practice of ‘decimation’, which was very common during World War I among the French, British and Italian armies, while it was almost unknown among the German and Austrian armies.³⁸ This practice, which originated in ancient Rome, consisted in trying and (almost unavoidably) executing some soldiers, randomly chosen among the troops of a company that, as a whole, was supposed not to have shown enough courage against the enemy. The aim of this form of punishment was to offer an unforgettable warning to the fellow soldiers of the executed ones. Considering the random selection of the soldiers to be executed and the high number of decimations, the possibility that some of the executed soldiers were actually not guilty of cowardice was of course very high (as masterfully shown by Stanley Kubrick in his Paths of Glory, which tells a true story that took place on the French front in 1916). Therefore, it is plausible – and in any case it can be granted for the sake of the argument – that the practice of decimation contributed to the victory of Britain, France and Italy. If so, that practice did in fact maximize the collective utility of the nations that applied it (what could be more useful for a nation than winning a conflict as terrible as World War I?). And consequently should we comply with the norm that in such extreme conditions decimations are required, as a norm-utilitarian may claim? Or, to put it differently, does its (arguable) utility prove that the practice of decimation was just and morally acceptable? Once again, the civilized juridical mind should have no doubt in answering negatively to that question.

To summarize, a problem that neither action utilitarianism nor rule utilitarianism is able to solve is how to articulate the ideal of justice in terms of general utility. In fact, if one assumed a purely utilitarian theory of justice, one could not rule out the possibility of having to accept practices that are intuitively immoral, such as scapegoating or excessive punishment. There is, however, a much better alternative, which does not require that one accepts the obsolete idea of positive retribution (‘Each person who deserves to be punished, should be punished’). This alternative was developed by two giants of the Anglo-Saxon juridical thought of the second half of the 20th century: John Rawls and HLA Hart.³⁹ Hart, in particular, presented the most convincing treatment of the issue. On the one hand, he accepted the idea that punishment can be justified only on a utilitarian basis: we should punish only the persons who are useful to punish (a view that implies the refusal of the positive component of retributivism). On the other hand, however, Hart also accepted the negative component of retributivism as a constraint in the distribution of justice: nobody should be punished who does not deserve to be punished and nobody should be punished in an excessive way.

The two morals of this story should be obvious. First, contrary to the bald statements of the advocates of eliminativism, abandoning the idea that we sometimes act freely, and in those cases – and only in those cases – we are responsible for our actions, and in case should inexorably be punished, would generate a monstrous conception of justice. Second, and more generally, eliminativism is a stance that is very easy to state, much less to defend. In fact, if one takes it to its extreme consequence, the view presents a world in which nobody should desire to live.

Unsurprisingly, some philosophers, unhappy with this conclusion, prefer to defend mysterianism regarding free will. Peter van Inwagen, for example, writes:

The problem of free will is so evidently impossible of solution that I find very attractive a suggestion that has been made by Noam Chomsky (and which was developed by Colin McGinn in his recent book The Problems of Philosophy) that there is something about our biology, something about the ways of thinking that are ‘hardwired’ into our brain, that renders it impossible for us human beings to dispel the mystery of metaphysical freedom.⁴⁰

Van Inwagen sees no way of locating free will in the natural world, and at the same time, he knows that we should not presume that we can consign free will to the dustbin of history. However, also in this case, one wonders why, instead of giving up any hope to understand free will, we should not give up the idea that we should understand it in the same way in which we understand the scientific world.⁴¹ Could not it be that free will belongs to the ‘space of reason’, meaning that it is autonomous from, and irreducible to, the natural world, but not as incompatible with it?

More generally, it seems that if one really wants to pursue Sellars’s ideal of a stereoscopic vision, one should give up the idea of the priority of the scientific image over the manifest image.⁴² As the failures of reductionism, eliminativism and mysterianism should teach us, these two images are both essential and mutually irreducible but not incompatible with each other.