Something is rotten in the state of our knowledge. Science imposes a gulf between nature and the life-world which is invisible to both scientists and philosophers.

(By 'nature' I mean the world as it would be if man passed out of existence; by 'life-world' I mean the perceptual world in which we live our everyday lives. The world in which we meet other people, talk with them, work with them, quarrel with them, but also the world in which we come across and work on things. Nature, I here take for granted, exists independently of man, whereas, of course, our life-world does not.)

In our philosophically non-reflective lives, we all of us take it for granted that we often perceive a man-independent nature. In our life-world (and, I think, in the life-world of most - probably all - cultures) nature is part of the life-world. Modern perceptual psychology, however, has since long implicitly taught that we do not directly perceive nature. One might think that one of the aims of perceptual psychology is to explain the mechanisms we use to perceive the world, but perceptual psychology puts forward theories which tell us that in our perceptions we cannot be in contact with nature. Of course, nature is regarded as one kind of cause of our perceptions, but such causes are regarded as wholly external to our perceptual acts. Our life-world subscribes to direct realism, our science subscribes to indirect (representative) realism. This is not acceptable; especially not since science nowadays is part of the life-world, too. Our life-world is incoherent. Something philosophical has to be done.

When modern philosophy emerged, both Descartes and Locke sketched the outlines of what was, some centuries later, to become the specialized sciences of perceptual psychology and sensory psychophysics. According to their story, ordinary veridical perception consists of a causal chain starting in the thing to be perceived, then passing through space to our body, into our body, and into our sensory organs. By different mechanisms the causal process is assumed to proceed through the body and in the head, in order to end somewhere in the brain. Here, at last, the perception itself is said to occur. Hence a perceptual act in which a thing is perceived is necessarily distinct from this thing itself, both spatially and temporally. The thing and the corresponding perception must be spatially distinct since the thing and the brain are in different places, and they must be temporally distinct since the causal process takes time. Furthermore, things and perceptions are categorially different. Things are material but perceptions are mental. The Cartesian-Lockean philosophy of perception implies that nature is wholly outside our life-world. Locke wrote as follows:

Descartes made the same point in the following way:

Locke remained on the abstract level of the considerations presented in these quotations, but Descartes tried to fill in the concrete details of the causal process. He put forward many hypotheses about different kinds of particles moving around in our body. In particular, he thought that there are some extremely small material particles, misleadingly called animal spirits, which are able to connect the sensory organs and the brain. There is no philosophical reason for learning about this detailed picture today. It was presented mainly in L´homme, which was published after his death. However, in figure 1, three drawings from that book are reproduced.⁴

Both Descartes and Locke were ontological dualists, although, epistemologically, Descartes was a rationalist and Locke an empiricist. Let us look at the relationships between their ontologies and their epistemologies. For an empiricist, an ontological dualism between mind and matter creates an insurmountable epistemological problem. Bertrand Russell, for one, has made this point forcefully:

Most post-Lockean British empiricists, from Berkeley to Russell, have freed themselves from Locke's inconsistency by becoming ontological idealists or phenomenalists. Although rationalists do not have exactly the same epistemological problem, they have one which is structurally similar. How does reason, which resides in mind, come to know anything about material particulars? Descartes was of the opinion that in order to silence all doubts about the existence of the external world, a proof for the existence of God was needed. Most of the great rationalist continental thinkers who followed Descartes, became, like Leibniz, idealists. Kant, of course, retained the thing in itself but made it unknowable. In my opinion, no materialist thinker has so far really solved the Cartesian-Lockean problem of perception. A metaphysical realist should be able to connect nature and the life-world.

Modern physics and perceptual psychology has discovered a lot about all the material processes which are necessary for veridical perception, and have shown that Descartes' detailed hypotheses were false. Broadly speaking, collisions between material particles have been replaced by electromagnetic interaction and chemical reactions, and animal spirits have been replaced by synapses and neurons, but nonetheless the abstract picture is the same. It implies an ontological gulf where nature and life-world are kept apart. Perceptual psychologists take the existence of a man-independent nature for granted, but, according to their theories, no part of nature can be a real part of a perceptual act.

It is possible to distinguish between three different paradigms or perspectives "that inform contemporary investigations of perception".⁶ One, "the inference and empiricist perspective", is closely connected with Locke, although its major figure is Hermann von Helmholtz. According to this perspective, perception consists of two parts, sensation and interpretation; and interpretation is regarded as built up only out of earlier sensations. Association of ideas is the main explanatory principle. A distinction between sensation and interpretation is also to be found in "the Gestalt perspective", but here interpretation is seen as stemming from innate ideas and from mind's own creative ability. Descartes, Kant, and Gestalt psychologists are among those who should be placed in this paradigm. However, differences notwithstanding, both the empiricist and the Gestalt perspective imply the dualism between nature and life-world which I have described.

The third theoretical perspective, which I would like to call the Gibsonian perspective,⁷ denies that veridical perception can be split up into two parts, sensation and interpretation. The founding father, J.J. Gibson,⁸ distinguishes between sensory receptors which respond to stimulus energy and perceptual systems which respond to stimulus information. The perceptual (visual, auditive, etc.) systems are assumed to be active, to interact with each other, and to be able to respond without any process of interpretation to stimulus information. The concept of stimulus information is one of Gibson's theoretical creations. According to Gibson, if one takes into consideration all light, reflected as well as non-reflected, it is possible to demonstrate that this ambient light contains structures and invariants which contain information about the environment. These structures and invariants can remain the same even when frequencies and intensities of the light change. The properties which sensory psychophysics has studied, e.g. frequencies and and intensities of electromagnetic radiation, is given a very subordinate role. The receptors in the eyes cannot discover such invariants, but, Gibson claims, the visual perceptual system can. He also assumes that our perceptions often give us correct information about nature. Sometimes, as in the following quotation, he even seems to be a direct realist in the sense that I use this term.

In spite of this quotation, and in spite of all the philosophical advantages which I think Gibson's perspective has compared with the other ones, he is not a real direct realist. Stimulus information is assumed to travel by means of electromagnetic radiation from the things perceived to the perceiving persons. The uptake of stimulus information is made at the surface of the body. Even in Gibson's theory, a thing perceived seems to be regarded as being wholly external to the perceptual act in which it is perceived. Gibson has never discussed the problems which I shall try to highlight below under the headings connection at a distance, x-ray perception, backward perception, and the changeful limits of our ego. Gibson might be called an epistemological realist but not an ontological realist. He says that veridical perception contains no process of interpretation and that it give us direct knowledge of the world, but he does not say that a veridical perceptual act contains parts of that which is perceived.

If we abstract from the differences between Descartes, Locke, and competing perspectives within modern perceptual psychology, we find a common core which can be illustrated as in figure 2 (where there are two persons who perceive the same tree). The point of the picture is that the Cartesian-Lockean heritage is monadological in its ontological import. Every mind is closed within itself. It is numerically distinct both from all other minds and from all external things and states of affairs which are said to be perceived in veridical perception. In contradistinction to Leibniz's monadology, however, the monadology of perceptual psychology has a materialistic basis. Perceptual acts are assumed to have material causes, and different minds are assumed to be connected with different material bodies. Bodies can directly interact with each other, but the minds cannot. When two people look into each other's eyes, the situation can be pictured as in figure 3. According to perceptual psychology, if you look into your beloved one's eyes, you will really not see her eyes. You will only see eyes in your own mind which are partially caused by her eyes. If she whispers sweet words, you will only hear words which exist in your own mind. A sad story, I would say. Ontological narcissism, as it may be called, is the necessary consequence of such a theory.

If we look at 20th century philosophy, there is not much awareness of the problem I have sketched. There are mainly two reasons for this neglect: the focus on logic and language within analytic philosophy and the 'bracketing' of the sciences within the phenomenological movement. Since we owe the concept of life-world to Husserl and Scheler, I shall first comment on the phenomenological movement.

The central concept of phenomenology is that of intentionality. Perceptions and thoughts are the main examples of of intentional acts. In both perceptions and thoughts we are directed at something; sometimes at something existing, sometimes at something non-existing. The aim of phenomenology is to study phenomena as they are given in our intentional acts. In order do to this accurately, according to (the middle late) Husserl, we have to perform the epoché, i.e. we have to suspend judgements. In particular, we have to bracket, or put within parentheses, all scientific explanations of how the phenomena to be studied are caused. Such explanations, it is claimed, refer to entities which are external to the phenomena in question. Also, we should bracket the question whether anything transcends intentionality. In this way the whole Cartesian-Lockean problem is simply put within parentheses. It was not to be dealt with, neither epistemologically nor ontologically, by phenomenologists.

Later on Husserl himself introduced the methods of transcendental and eidetic reductions, and other phenomenologists made other changes. But there is one thing all these changes within the phenomenological tradition have in common, the Cartesian-Lockean problem is pushed aside. According to the phenomenologists, when two people in the life-world perceive the same tree, they do perceive the same tree, and when two people look into each other's eyes, they do see each other's eyes. In the life-world we are direct realists, and most phenomenologists seem to rest content with knowing this. The method of epoché (and similar procedures) gives us the nature of the life-world but not the real man-independent nature which natural scientists think they are studying, it gives us the intersubjectivity of the life-world but not real intersubjectivity rooted in man-independent nature. Not even so-called realist phenomenologists (e.g. Roman Ingarden and John Wild) have taken the implications of natural science seriously. This is very clear in a recent book in this tradition with the telling title Back to 'Things in Themselves'¹⁰. Here, the traditional epoché of phenomenology is very explicitly thrown away. The author argues for the existence of objective knowledge of a mind-independent world, but he does not discuss the Cartesian-Lockean problem of perception in spite of the fact that he writes that "Phenomenology proper is so far from being opposed to causal explanations of things that it even calls for them."¹¹

The Cartesian-Lockean view emphasizes that there is a causal chain directed from the perceived thing to the perceiving person, whereas 'the life-world view' emphasizes that there is intentionality directed from the perceiving person to the perceived thing. If both kinds of directedness are represented by an arrow, the arrows will point in diametrically opposed directions (see figure 4). It can be noted that many theories of perception in antiquity really assumed that in vision there is a causal process which has the same direction as visual intentionality. Something was assumed to emanate from the eye and go from the eye to the thing perceived. John Burnet, in his famous Early Greek Philosophy, says that "what is characteristic of Greek theories of vision as a whole, /is/ the attempt to combine the view of vision as a radiation proceeding from the eye with that which attributes it to an image reflected in the eye."¹² He also writes:

Phenomenology, in particular the technique of the epoché, has in my opinion taught us a lot about the content of various perceptions. However, if we want a whole world-view we cannot in principle suspend our judgements about presumed causal explanations and try to be free from ontological commitments. The method of epoché has done its job and we have to face the conflict between the explanations of perceptual psychology and the findings of phenomenological philosophy. Something has to change somewhere. But let us first take a quick look at analytic philosophy.

Analytic philosophy has contained two main sub-traditions. One with Gottlob Frege and Bertrand Russell as the great ones among the founding fathers, and the other with G.E. Moore in a similar role. In the first one, philosophy came to be identified with conceptual analysis, logic and the construction of artificial languages; in the second sub-tradition, philosophy was in some way or other to be confined within the limits of ordinary language. However, as is often the case, the founding fathers were somewhat atypical; at least in the beginning. Russell was a real metaphysician who (now and then) advocated phenomenalism, and in the paper which triggered off Moore's career, The Refutation of Idealism,¹⁴ Moore defended direct realism without any appeals to common sense or to ordinary language. Later on, of course, he changed strategy and tried to make common sense the final arbiter in both epistemology and ontology. Founding fathers apart, the important thing now is that in both these sub-traditions natural science, realistically conceived, was put aside and philosophy was restricted to language. Each in their own way, like the phenomenological movement, put realist science within parenthesis.

In the Russellian line of analytic philosophy, science was highly esteemed, but it was claimed that philosophers as philosophers could not say anything about the world. Scientists were accorded a monopoly on claims about the world; philosophers could only indirectly be of help in the attempt to get knowledge about the external world. Philosophers could analyse the concepts of science but no more. This is the so-called underlabourer conception of philosophy. Logical positivism, which belonged to this sub-tradition, turned most scientific theories into instrumentalist theories by means of their principle of verifiability. Ontological problems were claimed to be literally meaningless. This means that it is impossible to discuss what I have called the Cartesian-Lockean heritage.

In the ordinary language tradition science was given no prominence at all. Moore's classic A Defence of Common Sense¹⁵ could just as well have been called A Defence of the Life-World. Common sense is for Moore more secure than science. Later on, Gilbert Ryle and the so-called Oxford Philosophy, explicitly claimed that the central aim of philosophy is to analyse ordinary language; or, to use Ryle's phrase, "determine the logical geography of concepts" and "rectify the logical geography of the knowledge which we already possess".¹⁶ In Ryle's The Concept of Mind this move amounts to almost exactly the same thing as the epoché of the phenomenological movement. Before he mounts his attack on the Cartesian ghost in the machine he says that: "It will be argued here that the central principles of the doctrine are unsound and conflict with the whole body of what we know about minds when we are not speculating about them."¹⁷

It was also within Oxford philosophy that the life-world conception of agency first entered analytic philosophy. I think there is a simple reason why ordinary language philosophy is so close to the phenomenological philosophy of the life-world. Without perception there is no language, and ordinary language is permeated by ordinary perception. In everyday life, we say we simply see a tree or see another person because that is the way the world is perceptually presented to us. At Royaumaunt in France 1958, Oxford philosophers met phenomenological philosophers at a conference. Ryle read a paper called Phenomenology versus 'The Concept of Mind'.¹⁸ In the discussion which followed both Herman van Bréda and Maurice Merleau-Ponty stressed that there are strong similarities between Ryle's views and those of the phenomenological tradition. Van Bréda said:

Merleau-Ponty made in the discussion it clear that he had worked with Ryle's book,²⁰ and he started his contribution as follows:

So much for phenomenology and analytic philosophy in relation to science and the Cartesian-Lockean problem. Similar things can be said also about the philosophy of later Wittgenstein and German hermeneutics. In my opinion, these traditions have taught us a lot about ordinary language, scientific language, perception and intentionality in general, but they are all wrong in denying the Cartesian-Lockean problem admission to philosophy.

Phenomenology has taught us that reductive materialism is obviously false. We all intermittently have intentional acts (veridical and illusory perceptions, images, acts of imagination, dreams, thoughts, etc.). This is impossible to deny. It is a truth as secure as any scientific truth; and intentionality cannot possibly be reduced to the kind of categories which make up natural facts in general or neurophysiological facts in particular. Ordinary materialist properties (like having shape, mass and electromagnetic properties) and relations (like being larger, being at a distance from, and being caused by) do not have the kind of directedness which constitutes intentionality.²² Not even a velocity has the same kind of directedness as thinking and perceiving have, even though a velocity "points" in space.²³ I agree completely with John Searle when, in his book Intentionality, he writes:

Intentionality cannot be denied. Let us now investigate the peculiarities of that intentional phenomenon we ordinarily call veridical perception, and let us see whether veridical perception can be interpreted in a way which solves at least some of the problems which perceptual psychology poses for it. The first thing to be noted is that veridical perception contains what might be called connection at a distance. In the perception of a tree (see figure 2), the perceiving person is by intentionality connected with the tree despite the spatial distance between them. In veridical perception there is, so to speak, a 'hop' in space. A real material unit like a stone (i.e. not an aggregate like a number of stones spread out on the ground) is usually thought of as being compact in space, i.e. all its parts are contiguous in space. Veridical perception, however, is different. When a person perceives a thing, the person is at a particular place in space and the thing at another. The visual act connects the person and the thing without 'filling out' the space between them.

In physics, action at a distance has ever since Newton been looked upon with suspicion. The gravitational forces of Newtonian mechanics looked mysterious even to Newton himself. How can the sun momentarily affect the earth which is eight light minutes away? How is this distance bridged? Causality means contiguity between cause and effect, but the concept of action at a distance denies this contiguity. Action at a distance is an occult relation. Like telepathy and telekinetics it connects cause and effect by a 'hop' in space.

Independently of whether or not action at a distance is possible, the actual existence of connection at a distance by means of intentionality is undeniable. It reoccurs in every everyday perception. Nor has anyone really denied it. But most philosophers have been content to say that it must be a mental phenomena, as though this classification would make it more easy to comprehend.

In our life-world there are colours, shapes, and a lot of other properties which are taken to be given facts. Whatever their ontological status is, they do exist. Similarly with connection at a distance. It is a fact that entities which are spatially apart can be directly connected. Whether these entities and the connection itself should be regarded as material, as spiritual, or as something else, may be open for discussion, but not that this kind of connection exists.

In what follows I shall assume that life-world veridical perception often really is veridical; I shall assume that direct realism is true. And by direct realism I do not mean that form which reduces intentionality to material states and dispositions and which ought to have another name.²⁵ By direct realism I mean the view that in acts of real intentionality we can be in direct contact with material things and states of affairs.

In veridical perception we reach out towards persons, things, and states of affairs in the world, and sometimes we become connected with them. How, then, one might ask, is this peculiar connection at a distance related to the material things which occupy the intervening space?

According to all our knowledge, be it science or common sense, usually we see through the air. Sometimes, of course, we see through glass or water. The general conclusion is that in visual veridical perception we perceive through material things. Connection at a distance is here not connection across empty space; it is connection across a materially filled space. Superman is said to have x-ray vision. He can see through material buildings. My point is that, apart from perception in outer space, all visual veridical perception can be called x-ray vision.

Michael Polanyi has drawn attention to facts about perceiving which amount to a similar conclusion with regard to the tactual sense, too.²⁶ Often, tactually, we perceive through material things which do not belong to our body. When we write with a pen we are not aware of the hand touching the pen. Instead, we are aware of the pen's touching of the paper. Similarly, when we use a hammer we feel the hammer against the nail, or even the nail against the wood, but not the hand against the hammer. When we are skiing we do not feel our feet against the ski but the ski against the ground.

If we only could be in direct contact with that part of nature which is at the the limits of our body, then we should only see the air just in front of our eyes, and we should only feel what the hands and feets actually meet. This, however, is simply not the case. Often we do, to summarzie the last two sections, perceive through material things to natural facts which are at a distance from us.

Phenomenology tells us that if in veridical perception we can perceive nature, then we do it by means of connection at a distance and x-ray perception. Let us now repeat what today's science tells us. It says that there is some kind of energy flow from a perceived thing to the body of the perceiving person, and it also says that it takes some time for the energy to move from the thing to the person in question. Therefore, we have to accept that intentionality which is connection at a spatial distance is also connection at a temporal distance. When we perceive a thing, we do not perceive it as it is now. We perceive it the the way it was structured when the relevant energy (according to Gibson: electromagnetic radiation with stimulus information) left the thing. This view, by the way, is not strictly confined to science. There are life-world situations which display the phenomena. Thunder comes after the lightning, and when one sees a man hammering at a great distance the blows are heard after they have been seen, and this difference is perceived.

The difficulty with intentionality through time, in comparison with instantaneous intentionality, is that it implies that one is in direct contact with states of affairs which no longer exist. Connection at a spatial distance can be rather difficult to accept, but, as first presented, it is nevertheless a relation between a perceiving person and a perceived state of affairs which exist simultaneously. Temporal connection at a distance, on the other hand, connects a presently perceiving person with an earlier existing state of affairs. The farther away the perceived fact is, the more obvious is this temporal relation. Perception of distant stars is the paradigm example. We can see stars which no longer exists. If this really is veridical perception, then veridical perception is backward perception.

I have maintained that most 20th century philosophers have, in different ways, avoided the problem of direct realism. One of the exceptions to the rule is Roderick Chisholm. He has written the following:

To my mind, Chisholm does not take 'the problem of star perception' seriously enough. I agree with him that "the paradox arises only because we tend to assume, until we are taught otherwise, that any event or state of affairs we perceive must exist or occur simultaneously with our perception of it". But there is more to it than Chisholm recognizes. His presumed solution, i.e. the similarity he finds between perception and causality, does not exist. That, today, we can receive light from a non-existing star is of course no mystery. Such light has an existence which is independent of the star which emitted it a long time ago. There is a causal chain through time, but each part in such a chain exists only at one moment, and there is no direct connection between two parts which are not contiguous in time. In veridical perception it is different. Chisholm has no clear grasp of the concept of connection at a distance. If direct realism is true in relation to things around us, then we perceive backwards in time, but it is then an extremely short time interval, almost infinitesimal, which we bridge. However, if direct realism is true even for star perception, then we can perceive backwards over a huge time interval. Mostly, we do in one and the same perception perceive things at different distances from us. This means that veridical perceptions of the world are extended backwards in time. This is what we have to teach ourselves if we want to be direct realists, but Chisholm doesn't even hint at this. As far as I know, there is only one philosopher who has stated this feature of perception clearly, and that is Samuel Alexander at the beginning of this century, but, for some reason, he mentioned it only in a footnote.²⁸

In the former section I claimed that connection at a distance, which is a feature of veridical perception, is something other than action at a distance, which (if it is at all possible, which I think it isn't) is a kind of causality among material entities. Similarly, backward perception is a feature of veridical perception which must not be conflated with backward causation, which (if it is at all possible, which I think it isn't) is a kind of causality among material entities. In the next section we shall see that the existence of these features means that the limits of our ego are undetermined and very changeable.

What I have said about veridical perception has repercussions for our conception of the human ego. The history of philosophy contains several different ontologies of the ego which I find false in every important respect. I have already mentioned one such ontology, reductive materialism. Another easily refuted metaphysics is the one which takes intentionality on the one hand, and materialist categories on the other, to be merely different ways of apprehending the same phenomena. Such a Spinozist multiple aspect theory is reflexively inconsistent. The phrase 'different ways of apprehending' presupposes the category of intentionality. An act of apprehension is an intentional act, which means that a Spinozist view amounts to saying that intentionality and materialist categories relate to the same substance apprehended in different intentional acts. The last use of 'intentional act' cannot be replaced by any phrase contaning concepts which only refer to materialist or other non-intentional categories, which means that 'the intentional aspect' as a whole cannot possibly be equivalent with 'the materialist aspect'.

The falsity of ontologies which, like those of Descartes, Locke, Leibniz and Berkeley, regard the ego as a spiritual substance, is not as total as that of reductive materialism. Such substance ontologies of the ego do not deny the existence of irreducibly mental phenomena, but they lack a clear conception of intentionality. They have not noted a certain feature of intentionality, the non-substantiality of the ego, which Heidegger and Sartre²⁹ brought to our attention.

Heidegger characterises man by saying, among other things, that man's situation in the world is one of "Geworfenheit" (= thrownness). Man is in a sense throwned, flung, or cast out into the world. Sartre describes the essence of man as a Nothingness. Both descriptions, at bottom, focus attention on the fact that the directedness of intentionality is in many, perhaps most, intentional acts not apparent at all in the acts themselves. When we are fascinated by something we are observing, we are just 'thrown out' in the object; we do not perceive ourselves perceiving. We are aware only of the perceived object. The same goes for perceptions accompanying concentrated actions. All there is is the action. We are lost in it. Following Sartre, I shall "call such a consciousness: consciousness in the first degree, or unreflected consciousness".³⁰

The directedness of intentionality has two poles. One may speak of a 'to-pole' (an intentional object or correlate) and a 'from-pole'. If, now, in a reflective intentional act (i.e. consciousness in the second degree), we try to make the from-pole of an earlier act (i.e. the presumed ego) into the to-pole of this later reflective act, what do we find? When we make an earlier intentional act itself into an intentional correlate, i.e. the to-pole of the present intentional act, we find between the 'from-pole' and the 'to-pole' of the unreflected consciousness nothing similar to a relation between things. In the latter case - think for instance of a perception of the relation of being larger than - we perceive two things and a relation between them. In an unreflected intentional act directed towards nature we perceive things and states of affairs but no relation between ourselves and the intentional correlates in question. Therefore, it is adequate to say that the to-pole is something, whereas the from-pole of intentionality is empty. The ego which is assumed to exist in the from-pole seems to be thrown out into the to-pole or to be a kind of nothingness, or, better, emptiness.

It should be noted that when a reflective (second degree) intentional act is directed at the from-pole of an unreflected (first degree) act and discovers the corresponding emptiness, the reflective act is not directed at its own from-pole. This from-pole has the same emptiness, but in order to see it a third order act which has the second order act as its intentional correlate is needed.

That the from-poles of intentional acts are empty does not mean that they do not exist. Their existence may be compared to (but not identified with) that of the void (in non-relational conceptions of space). Where there are no things (or fields) in space there is void, but void is not nothing. It is empty space. The from-pole of intentionality is empty but nonetheless it is something. It belongs to reality.

The emptiness or nothingness of the ego is, it is important to note, only an emptiness on the level of intentionality itself. As stressed by another phenomenologist, Maurice Merleau-Ponty, intentionality so to speak radiates from our body.³¹ When in reflective acts we are looking for the from-pole of intentionality, we will always find a material body. Moreover, our own material body. In this sense we find something, but not what we are looking for. We look for our soul but we find our body. We find a substratum for intentionality. Perceptions are perspectival and they refer back to our body. Only an Aristotelian account of the soul can comprehend this feature of intentionality. Intentional acts presuppose for their existence something which is not part of themselves, namely a body with a nervous system and a brain. The ego, therefore, is neither only a Nothingness nor only a "Geworfenheit". It is a complex unity with both a body and intentionality; the body is a substratum upon which intentional acts are emergent properties.

Looking upon Heidegger and Sartre from this Aristotelian point of view, Heidegger puts too much emphasis on the to-pole and Sartre puts too much stress on the emptiness of the from-pole of the ego. For Heidegger the ego seems to be merely "Geworfenheit", merely the indwelling in the perceived objects and facts; and for Sartre the ego seems to be nothing. The ego is not a pure "Geworfenheit", nor is it merely Nothingness, but neither is it a spiritual or material substance. It is intentionality fused with a body.

Since having perceptions (and having the capability of having perceptions) is part of the essence of human egos, egos cannot have the kind of spatial limits which material things have. Material things are enclosed in a space volume, whereas the ego in veridical perceptions reach out from the body into the world at a distance. Intentionality is not the only kind of phenomena which deviates from the things-with-properties or substance-accidence scheme. Relations like 'being larger than' and 'being more circular than' are not confined to a compact space volume either. Instances of such relations exist in the scattered particulars they relate. The point I want to make, however, is not that intentional acts are some kind of relations. In my view, on the contrary, intentionality is a category distinct from both external, internal and grounded relations.³² Reality contains several categories which, from the perspective of a substance-accidence scheme, have peculiar spatial limits. Therefore, one should not look for the spatial limits of the ego the same way one looks for the spatial limits of a material thing and its (monadic) properties.

Some things are easy to think, some things are hard. The spatial limits of material objects are easy to think, the spatial limits of the ego is hard to think. Material objects have rather well defined spatial limits even if the limit, like that of a shrinking balloon, is rapidly changing. But intentionality behaves in a different way. It is not limited by other material objects nor by other intentional acts. In veridical perception the ego is fused with natural facts, and the spatial limits of the ego are the spatial limits of its intentional acts. Wherever an intentional act turns the non-perceiving of material things into perceiving, the spatial limit arises. The ego is not spatially confined to its body in spite of the fact that its intentional acts are existentially dependent upon the body. Intentionality makes the ego spatially undetermined. Normally, the limits of our ego are changing. At one moment we are looking at states of affairs close to us, and at the next moment we are looking at more distant states of affairs.

The common sense distinction between oneself and external things is easily turned into false philosophical ontologies of the ego, where an inner-outer distinction is wrongly made identical with a distinction between subjective (= mind-dependent) and objective (mind-independent) phenomena. In relation to material things, the distinction between being inner and being outer is clear. What is within the spatial limit of the thing is inner, and that which is outside the limit is outer. But since intentionality does not have the same kind of spatial limits, we are not allowed to think the inner-outer distinction of the whole ego in the same way as we can think the distinction in relation to its body.

Usually, we perceive pains as located within our body. The same is true of our heart beats, of tired muscles, of nervous stomachs, and of other similar phenomena. In cases like these, the intentional correlate is both mind-dependent and inside our body. Subjectivity and inwardness here go together. We perceive ordinary things, plants and animals as located outside our body. Here, objectivity and outwardness go together; the intentional correlates are both mind-independent and outside our body. However, when we perceive the colours of the things in question, these intentional correlates are both mind-dependent and outside our body. We have subjectivity and outwardness together. The same is true in visual illusions and hallucinations as well as in tactual illusions like the "phantom pains" of amputated legs. If all mind-dependent intentional correlates are said to be inner, in contradistinction to mind-independent correlates which always are outer, then 'inner' looses its original contrast with 'outside the body', and, consequently, has to take on a completely new meaning. A meaning which has to turn all intentional correlates into inner entities, and we are back into the ontologies of idealism or dualism. In such ontologies nature can never be perceived.

In some intentional acts (like dreams and imaginations) the intentional correlate (the dream and imagination, respectively) is wholly mind-dependent, whereas in other acts (like veridical perception) some parts of the intentional correlate (for instance a material thing) are mind-independent and some parts (for instance colours) are mind-dependent. In the latter cases there is a fusion of mind-dependent and mind-independent parts. Our egos are not spatially confined within the limits of our bodies, this is the ontological truth to remember.

The ontology of the ego now sketched contradicts ordinary psychological (and Humean) projection conceptions in which secondary qualities, like colours and other life-world qualities, are projected onto nature. When, with a projector, we project a picture on a screen, there is a picture inside the projector which by means of a light beam is copied on the screen. However, most life-world qualities do not first exist inside our head in order, later, by means of a 'perceptual beam', to be copied out in the world. Their primary existence is outside our body. The projection metaphor is adequate only in those cases of writing and drawing when we know in advance what to write and what to draw. Here, first there are thoughts and then there are corresponding outward-oriented perceptions.

Just as the feature of connection at a distance implies that our egos are not spatially confined within the limits of our bodies, so the feature of backward perception implies that our present egos are not temporally confined within the limits of the present. If we are in direct contact with past states of affairs, part of the ego must be extended into the past. The time limits of our egos are, just like the spatial limits, undetermined and changeful. However curious this may seem, it should be noted that, according to the claims put forward here, the ego is never in veridical perception extended into the future. Also, I want to repeat, the closer to the perceiving body a perceived state of affairs is, the closer it is to the present. Our body only exists actually in the present, although it can preserve its identity through time. The body is never extended in time outside the present, only some acts of intentionality are.

If we want to claim that, in our life-world, we do really perceive parts of nature, then we have to accept that veridical perception is x-ray perception, backward perception and connection at a distance; we also have to accept that our ego has no determinate spatial and temporal limits. Peculiar features. Perhaps they are too odd to believe in. But, we have to ask, on what grounds are we to take the decision to accept them or reject them? Since most philosophers nowadays are fallibilists in epistemology, the problem is not that I cannot prove that in (so-called) veridical perception we are in contact with nature. The only thing a fallibilist can do is to estimate the reasonableness of this view in comparison with other possible alternatives. Of course, one should choose the most credible view; or, as in this case, the least incredible alternative. In order to reject veridical perception, we have to find an ontological alternative which is cheaper. What, then, are the main alternatives to real veridical perception? 

For many thinkers reductive materialism gives us the truth about perception, but, as is clear from my earlier remarks on this view, I regard reductive materialism as the most incredible of all views. It simply denies the obvious fact of connection at distance. Therefore, to my mind, the main alternative is the claim that connection at a distance is possible only in a completely mental sphere. In idealist conceptions and in indirect realism there is no need to postulate either x-ray perception or backward perception. If all intentional acts are wholly mental, then it is tautologically true that x-ray perception (i.e. perception through material things) is impossible. Backward perception, on the other hand, is not ruled out in principle, but there seems to be no reason at all to claim that there is a temporal distance between the perceiving person and that which is perceived. Note, though, that connection at a distance is not explained away. It is merely maintained that connection at a distance is wholly mental. There is no denial of my claim that the existence of connection at a distance is an indisputable fact.

If we dismiss idealism, be it old-fashioned or linguistic, then 'the mental alternative', taken together with perceptual psychology, implies a monadology. We are back in some kind of Cartesian-Lockean ontology. Every person is confined within his own mental world. The only connection which exists between people and between people and things are causal relations in the material part of reality. If we are not to abandon completely the belief that our life-worlds have something in common, we are forced in this ontology, as was Leibniz in his, to postulate a predetermined harmony among all the numerically different mental spheres. Fundamentally, we each live in our own mental world, but the worlds have great similarities with one another. Ontologically, we are as beings with a consciousness, in contradistinction to clumps of pure matter, completely and helplessly isolated from one another. My life-world is only mine and your life-world is only yours; even when our bodies, which are part of nature, are as close to each other as they can be. If life-world and nature is kept apart, the life-world breaks apart, too.

As far as I can see, our fallibilist ontological choice today consists, to put it sharply, in either accepting a monadology or accepting a direct realism wich contains the peculiarities of connection at distance, x-ray perception, backward perception and undetermined limits of the ego. In my opinion, the monadological alternative is more incredible than direct realism with its implications.

What is most difficult to accept in the kind of naive realism, or life-world realism, which I am advocating, is of course that part which is not directly in keeping with genuine naive realism, namely the view that one perceives across or through time, not at one particular moment. But, I want to stress once again, 'through time' is always connected with 'through space'. The farther we get in time, the farther we have to get in space. Normally, when in one and the same instant we are looking at several things whose spatial distance to us varies, we perceive them as simultaneous. According to the view I have argued for, this must be wrong. Space cannot in veridical perception open itself towards more distant states of affairs without time's also opening itself - and vice versa. It is an illusion that ordinary veridical perceptions are momentary in time. If this is accepted, then it is not too difficult to accustom oneself to the view that one can perceive backwards through time.

The problem of perception is merely one of a number of ontological problems which need to be solved before we get a stable philosophical connection between life-world and man-independent nature. Apart from the problem of perception, there are also the problem of agency (or of mental causation) and the connected problem of how to reconcile (to speak with Kant) causality of freedom with causality of nature. In our life-world we take it for granted both that our will can affect our body and that our actions are not wholly predetermined by natural laws, be they deterministic or statistical; indeterminism is something other than freedom.

I shall not discuss the problems of agency and determinism here, merely mention that they, like the problem of perception, can be viewed from a new perspective now that fallibilism reigns supreme in epistemology. Fallibilism means that neither ontologists nor scientists can speak with the voice of the one who has recourse to absolute truth. Philosophers can contest science and scientists can contest philosophy. A philosopher who contests science, need not and should not, regard himself as a 'master-scientist' standing above science; and a scientist who contests philosophy, need not and should not, take recourse to scientistic views. This means, among a lot of other things, that the reductive views of the natural scientists should not have too much authority in ontology. Science cannot in and of itself show that the life-world experience of agency is false, and that, therefore, the concept of freedom should be wholly replaced by that of causality of nature.

The existence of agency implies that the future cannot be wholly pre-determined. Something must be open. This means that veridical perception through time into the future is impossible. Forward perception conflicts with agency, but backward perception does not. And I have only argued for backward perception.

I do think that we are philosophically entitled to say that we, in our life-world, can be in contact with nature and with each other. However, since I regard my ontological views as fallible, I shall end by letting Nature herself speak. I can hear her saying: "To be perceived, or not to be perceived: that is the question." She is looking at a skull. A philosopher's skull. She wants it alive again.