Indexicalism

This is the first version of what I want to submit to the call of Open Philosophy:

Towards an indexical paradoxico-metaphysics


1. Paradoxico-Metaphysics

Jon Cogburn introduces the idea of a paradoxico-metaphysics as follows:

Metaphysics aims to give a maximally general account of what reality is like such that we encounter the phenomena that we do. But what if we encounter phenomena […] that seem to entail that metaphysics is impossible? Then the task of metaphysics is to give a maximally general account of what reality is like such that metaphysics is impossible. If the project sounds metaphysical, that is because it is.1

The impossibility of metaphysics requires an explanation and if any such explanation cannot stop short of making metaphysical assumptions or providing metaphysical arguments, paradox looms in the horizon. If we bite the bullet, we engage in paradoxico-metaphysics, the endeavor to provide a metaphysical explanation for the impossibility of metaphysics. That metaphysics is impossible ought to teach us something maximally general about reality. Additionally, reality is or could be such that we can find out that no metaphysics is possible. The way forward for metaphysics would then be to become a continued effort in self-destructing.

Cogburn understands paradoxico-metaphysics as amounting to a claim that metaphysics is itself necessary and impossible. Further, he understands that metaphysics ought to somehow tackle totalities as it addresses maximally general accounts of reality but also should acknowledge that those totalities are often themselves paradoxical. By enclosing everything they fail at least to enclose themselves. Cogburn understands that the only totalities that can concern metaphysics are the ones that turn out to be inconsistent. He writes:

We cannot help but to treat the totalities of most interest to us as if they constitute the world, but in doing so we think of such totalities as both closed and transcendent, leading to contradiction. And for the paradoxico-metaphysician the world itself is the inconsistent totality that generates these contradictions when we interact with it. If our metaphysics is contradictory, that is because it accurately reflects reality.2

Inconsistent totalities are the staple of a paradoxico-metaphysician's diet. To envisage a maximally general account of what there is leads to paradox if something in what there is makes it impossible for a maximally general account of what there is to be given. Perhaps the paradoxical element can be accommodated if we consider that 'metaphysics' is being used here in related but slightly different senses. One sense is about the totality of what exists while the other is simply about what exists. The word points on the one hand to an enclosed totality and on the other to its transcendence through a question of what makes it enclosed. The two senses of 'metaphysics' are entangled enough, however, to make it intelligible that there is a metaphysical reason why metaphysics is impossible – a paradoxico-metaphysical reason.

If there are limits to any grasp of totality, these limits are no less metaphysical than the effort to reach a totality. Paradoxico-metaphysics, viewed close to Kant's critique of traditional metaphysics, could be seen as a claim that any transcendental philosophy is inextricably committed to metaphysics. To reform metaphysics and append it with a transcendental story is therefore not enough for reality should be one way or another for the transcendental story to be the case. There should be something in reality that makes it impossible for phenomena to transcended the reach of empirical judgments. Also, there should be something that makes it possible to draw the limits of what is accessible – of what is knowable or thinkable. Jacques Derrida has hinted that the effort of recoiling from our empirical judgments to a transcendental sphere is only distinguishable from the attempt to go beyond our empirical judgments if we focus on the direction of the path one walks.3 Without that, he notices that the ultratranscendental resembles the precritical. In other words, the recoil to explain why metaphysics is impossible ends up eventually committing to metaphysical assumptions. If this is so there are metaphysical reasons why metaphysics cannot be done; it is necessary to recoil into metaphysics to show that metaphysics is impossible. Metaphysics invites a foray into paradox.

There are at least as many ways into paradoxico-metaphysics as there are critiques of the metaphysical endeavor. In fact, whenever an attempt at departure from metaphysics is met with its felt inescapability, the options of either rejecting metaphysics or abandoning its criticism could easily seem inappropriate. The alternative is to try to embrace both criticism and metaphysics, to hold together the suspicion against the endeavor and a reoriented effort to accomplish it. Kant was right in betting on a renewed sense of metaphysics as much as he was wrong in gauging that this new sense has no bearings on the old one. This paper sketches and begin to argue in favor of a specific paradoxico-metaphysics: indexicalism. As a paradoxico-metaphysics, indexicalism is self-destructing; like an attempt to provide a maximally general account of reality, it explains how self-destruction is brought about.

2. Indexicalism

Pointing is a usual practice – perhaps too usual to attract much metaphysical attention. Not much has been said about the fact that reality is such that it can be pointed. Some elements in language relate to the practice of pointing: we talk about this and that or about this hand and that mountain. We can proceed this way if we are in the midst of things, in media res to use the latin phrase. We are in a place where we can point at the hand and the mountain, we are amongst them. Our though can therefore be about what is around us, it can be de re, that is, about the specific things around me and dependent on my position around them.4 De re thought requires a relation to what is thought. We can point at things – and think in a demonstrative, de re, way about then – because we are in relation with them, we don't contemplate reality from nowhere. We must share with what we point some common feature of reality: for example, it is plausible to say that our pointing of things requires an insertion in the same space and time where what is pointed is or that our pointing of an event is as much an event as the event we point.

Resorting to pointing expressions – this and that – is often taken to be sufficient to present what is around but not more than a starting point in the business of offering an image of the furniture of the universe.5 An answer to that somehow grander question would involve stepping beyond our current position and seeing things from nowhere. It is precisely the idea of such a nowhere that can afford to be a place for viewing – and the most impartial, the closer to how things really are and therefore the most commendable place for viewing – that indexicalism attempts to challenge. Instead of a view from nowhere, it proposes a situated, local viewing point. The claim is that the grander question concerning the furniture of the universe can indeed be answered without stepping out of where one stands.

There is a line of argument – call it substantivism – according to which the claim that the universe is made of this and that is at best the beginning of an account of what there is without our point of view – it is at most like a vague assertion about everything. According to substantivism, it is incomplete and imprecise: there is more to the universe than this and that and while it is (arguably) true that the universe is made of whatever is denoted by 'this' and 'that', it is not made by this and that themselves. The external world, or the Great Outdoors, cannot depend so much on the window from which we see it. An account of what the universe is like should rather involve something like a description of whatever 'this' and 'that' denote – at least a partial and tentative description of the hand and the mountain being pointing at. Substantivism favors nouns and not demonstratives when it comes to give an account of the universe; there are hands and mountains and, as a consequence and only derivatively, this hand and that mountain. An account of the universe should not be given in terms of de re beliefs aiming at de re truths, but rather in de dicto terms: it should be free of the inextricably positioned character of any talk involving 'this' and 'that'.

The perils of positioned talk, the substantivist line continues, are manyfold. I can miss the point of how things are “out there” by confusing them with what is around – confusing the furniture of the universe with my own surroundings, maybe my own perceptions or perhaps my own way to relate to what is outside. The perils go by names such as idealism which is different from subjectivism which are both different from correlationism. These three different perils point at some sort of recoil from the idea that there is something external – or that there is something external accessible to us, the Great Outdoors. The peril of idealism is that anything external is replaced by an ersatz inside me. The peril of subjectivism is to take anything external as being, to some extent, at the mercy of my subjective activity. The peril of correlationism is to have to accept that only a correlation with what is external is available to me.6 In order to make sure that what is out there is properly considered, the substantivist recipe is that any positioned talk must be exorcised in favor of an impersonal view from nowhere that is the only way to account for how things are “out there”.7 To be sure, subjects, positions and point of view could be considered as they are part of reality by being part of the furniture of the universe in a substantivist approach, but they would have to be treated in a de dicto, impersonal way, in the language of substantives. One could describe subjects in terms of their capacities, or in terms of their qualities or relations, with substantives. To describe a position, a perspective or a point of view, according to substantivism, is neither to endorse it nor to accept any other. Substantivism comes in many flavours: it can be object-oriented and assume there are objects out there independently of any point of view, it can assume an interdependence between what exists, it can endorse a metaphysics of substances and substrata that display necessity and a permanent character. Although an ontology of events like the one of the early Whitehead8 seem farther from substantivism, it can endorse it as long as a non-indexed description of the events that constitute the world can be given.

Indexicalism is the claim that indexicals are the furniture of the universe. As we can perhaps already see, they form a paradoxical furniture. The claim entails that the universe is deictic or indexical and therefore demonstratives rather than substantives are best equipped to deal with it and, ultimately, to provide a description of it. It is a rejection of substantivism: although nouns can be used to provide useful accounts of great part of what exists, they provide no more than a façon de parler. The thesis is that the starting points of any metaphysical account of the universe are not substantives (therefore, neither substances, nor actual entities; neither objects (or subjects) nor material items; neither neutrinos nor forces but rather demonstratives like this, that, in, out, same, other, here, there, etc.) There is an irreducibly deictic character in the furniture of the universe; its elements include what is around and what is away, what is internal and what is external, what is farther, outer and different. As a consequence, it is not only thought or language that are arguably mostly de re; reality is itself indexical; further, perhaps, to be is to be indexed.

However maybe there is indexicality even beyond being. One can find a possible precursor of indexicalism in Plato's Stranger in his Sophist where five μέγιστα γένη (greater kinds) are posited: Same, Other, Rest, Movement and Being. Being appears surrounded by four indexical kinds that are external to it but affect it from without. The Stranger promotes the parricide of Parmenides by holding that not being could indeed be if it is understood as other than being; nothingness is not a greater kind but a product of Other applied to Being. It is as if the Stranger comes from outside to provide Being with an address: it is only in relation to what is other than being that being is. Nothing precedes these ultimate five kinds, there is nothing substantive underneath them. They are, the five of them, equally ultimate. Albeit perhaps as old as Plato, indexicalism is far from a standard view in metaphysics. This is arguably because it steps close to paradox.

Indexicalism has indeed an air of paradox. If the furniture of the universe is deictic, both the idea of furniture – composed of pieces that are more or less detachable – and the idea of universe – composing one big thing with no external borders – become weird. One way to begin to come to terms with this is to take indexicalism as both a metaphysical doctrine and a way to argue that metaphysics is impossible (or ought to be severely bound). It is the latter because it attempts to show that nothing substantive can be said about what there is – at least nothing in terms of irreducible substantives. It is only from a positioned place, and not from no place at all, that the universe can become accessible and its pieces of furniture considered. If the metaphysical project involves giving a general account of everything at once, the project has to be abandoned or severely revised. But the project is doomed not because there are limits to our access to what there is – it is not the fault of the wanna-be metaphysician. The fault is that there is no such thing as a metaphysical picture in this sense – a general account of everything at once – to be attained. This is why indexicalism is also a metaphysical doctrine: it explains why this metaphysical picture cannot be attained and therefore why the metaphysical project is impossible. It is a metaphysics that deals with the impossibility of metaphysics. Analogously, it is a metaphysics that responds to the questions of why and how metaphysics should be overcome: there is no metaphysics of pointers unless metaphysics itself is a pointing. Indexicalism holds that any enclosed realm described by substantives has something beyond it – 'beyond' is a constituent of reality. It is becoming visible why indexicalism is a paradoxico-metaphysics.

2.1 The Other

Levinas is an important character in my indexicalism plot I'm putting together because he focused his philosophical attention on what he understood as an irreducibly indexical element: the Other that he treated respectfully with a capital first letter. The Other, as in Plato's Sophist, is external to being and, as such, provides some sort of a constrain on it from without. Levinas has championed a struggle against postulating a transcending-free totality. At first glance, he had an ethical problem with totality: it is unfair and violent to the Other, for it converts the Other into part of a totality, something that can be eventually neutralized, something substantive. In fact, his ethical project can be described as the effort to rescue the Other – an indexical in an asymmetrical relation to the same – from substantives. To describe what is other in terms of substantives is to make it less other, to bring it to a common ground that is a step towards totality.

Levinas compares the existence of the Other as such with the ontological argument: it is in the nature of the Other to exist as something exterior.9 This is the metaphysical import of his effort: to clearly disentangle exteriority from totality. It is an error to try to achieve the exterior through a view from nowhere that sees everything while being in no place; it is a violence to subject the Great Outdoors to a total image that makes it less outdoors (and maybe less great). Interestingly, a view from nowhere is precisely what is rejected both by Parmenides in his exorcism of nothingness beyond being and by Plato's Stranger's parricide that insists that nothingness is intelligible but only as being completely other. Levinas' commitment to the Other is a commitment to exteriority that precisely distances him from a quest for totality. In this sense, Levinas will appear in my construction of indexicalism as someone who provides a key to distinguish it from many forms of idealism, subjectivism and correlationism. To be confined in an environment of indexicals – of this and that and the like – is not to be enclosed away from anything external precisely because the indexical environment includes the beyond, the other, the across the pale. Levinas' strategy of comparing his defense of the exteriority of the Other with the ontological argument intends to show that while indexicals are inside an environment, their role is to point at something exterior.

In the spirit of Levinas' attention to exteriority rather than totality, indexicalism privileges the metaphysical aim at attending to what is outer. Stepping out of subjectivity – or of our correlation, of our own ideas – is to be understood as a step towards what is other and if it is a step towards what is other, it can described in indexical terms. In contrast, substantivism attempts to see what is exterior – as much as what is interior – no longer as an Outdoors but rather as a piece in a totality that can be seen from inside a purportedly reachable position that is non-situated and that would provide a view from nowhere. Indexicalism holds that such view is not attainable and if it were it would do no more than including what is exterior in an all-inclusive total view.

Indexicalism as spouses a thoroughly situated metaphysics. That means that positions where indexicals apply are ultimate building blocks – corresponding to me, here, now. Indexicals, as de re talk, makes sense only in the middle of things, from a position, from somewhere. There are indexicals only if there are positions, positions that in their turn cannot be fully described in substantives. A position can be understood as composed primarily of a border that distinguishes what is inside it and what is outside. I will call this positions interiorities and they are composed by deictic operations that can be seen like the ultimate particles. Each interiority is in what I call an indexical environment, the space mapped out by deictic operations. To be sure, interiorities are not substantive realities that precede the demonstrative talk but only make sense – only have their borders drawn – through the deictic operations. As such, interiorities are units that cannot be contemplated, or considered, from nowhere – they are positions mapped by demonstratives and, as such, they are partly determined from outside, from other positions. In an important sense, interiorities are not closed for they are also drawn by indexicals that point outwards; not in terms of substantive descriptions of what is outside but rather in terms of demonstratives that fix positions without providing a substantive determination of what it is. Indexical operations, as we will see shortly, fix references and it is through this fixation that they establish positions (and compose interiorities). Each interiority includes therefore an exterior, each one is composed both by a here and an away, by a now and a later, by an around and an elsewhere, by a same and an other. If we see these operations as ultimately constituting particles, they draw borders that are relative to other positions; interiorities are always positions with respect to other interiorities. They have an other, an exterior, embedded in them. Exteriority is possible because there are interiorities but those are not enclosed units, but rather they are hostages of what is exterior. Jean Wahl, considering Hegelian dialectics and how it leads to totality, contrasts its result with its process that is bound to draw on the fragmentary in each step. In a sense, indexicalism is a way to never betray exteriority by falling for the attractions of totality; each position in an indexical map has what Wahl calls un contact nu et aveugle avec l'Autre.10

2.2 Interiorities and totality

Interiorities are therefore intimately marked by the exterior but form units that can be referred to through demonstratives. Those units of position, according to indexicalism, are what makes predication possible. They are existents that have to be present to carry the burden of concrete existence: nothing exists deprived of any position among concrete things, every concrete existence requires an interiority to rest on. No predication can be done without a position to be predicated. It is the position, the interiority, that is predicated and, in that sense, it is similar to what Aristotle called, in the Organon, a τοδε τι or an ουσία προτε.11 It is what supports predication and something that is not in itself any predicate (or a cluster or bundle of predicates). It is a concrete position that is ultimately predicated; “being mortal” applies to an interiority that is Socrates, delimited by deictic operations. The interiorities thus delimited can also be compared with substrata, something that can make two indiscernibles – two subjects of the same predications – not identical. If two particular entities have different substrata, they can have the same qualities and relations but still be different. We will see that the deictic operations that compose an interiority can require predicates, but once a position is fixed, it becomes somehow independent of the predicates used in its composition. The predication of something, the “is P” of a predication, requires a subject. Levinas remarks that even “being Spinoza's God”, the most impersonal entity he can conceive, is tied to an entity that carries the predication.12 The subject of a predication is something that is fixed by deictic operations, it is an interiority. According to indexicalism, these operations and the mapping of positions precede any predication – before any S is P, before S being anything, it is constituted as a position by deictic operations, composed by demonstratives such as this, there, same and other. Clearly again, as we will see, these demonstratives are often implicit in predicates and substantives. But indexicalism claims that deictic operations underly any predication.

The notion of interiority can be compared with Whitehead's notion of locus standi. In his ontology of events, Whitehead feels the need to introduce more or less fixed point to make sense of the idea of goings-on. These more or less fixed point is the percipient event – itself also an event with a duration associated to it. This is described in terms of a locus standi which is “represented in thought by the concept of 'here'”.13 The percipient event's duration associated to its 'here' – Whitehead calls this relation cogredience – that can only be conceived within a time-system and that is thoroughly situated. This 'here', the locus standi that enables an actual entity capable of perceiving events – while being amid events as events cannot be perceived by seeing them go on – is a place in the world. The percipient event doesn't have a structure that is more relevant than its location; it is a location from which events take place – the restful point of view from which movement makes sense. This locus standi is an interiority. Whitehead's need for loci standi shows that interiorities are situated, as they are demarcated by deictic operations. The percipient event is what provides the foothold of any perception in the world, perception requires a point of view, a standing location. Situated interiorities, drawn by deictic operations and ascribing a position for predication, are a point of departure for indexicalism.

Levinas gives a central importance to (his notion of) interiority. He acknowledges that it is a Cartesian inheritance that makes him start out with interiority while bearing in mind that there ought to be something else that makes it possible – including a responsibility for the Other that lies underneath any spontaneity within interiority. In Totalité et Infini he analyses interiority as what provides a break in totality.14 Interiority has a structure that enable genuine exteriority which is understood to conflict with totality.

There are several senses of totality that emerge from this contrast with exteriority. One of them, is the absence of interiority. If there are no interiority, one cannot make sense of the asymmetrical distinction between same and other and, arguably, of any indexical reasoning or, more generally, of any deictic operation that cannot be just a shorthand for a relation between predicates. In any sense of interiority, there is the external element that contrasts with what is inside: the external world beyond the res cogitans, the other units of action in Leibniz, what is beyond an ego and its intentional acts. It can be said that the indexicalist interiority is a minimum denominator that is required for an interiority to be such: the capacity to occupy a position that draws a distinction between inner and outer through deictic operations. An interiority has to have an exterior – and in that sense we can say that a totality cannot be an interiority. Leibniz' attempt to conciliate interiorities filled with predicates and the totality of a possible world assume worlds themselves have no interiority. (The same can be said about the totality formed by all possible worlds, that Leibniz presents in the metaphor of the palace of Palas towards the end of the Theodicy15 – here each one of the infinitely many rooms is a possible world.) Indexicalism postulates that there is no being that is not indexical – if it is so, there is no such thing as a totality.

Another important sense of totality in the contrast with exteriority has to do with asymmetry. Levinas considers that my relation with the other cannot be that of symmetry because if it is so, I'm already seeing both myself and the other from nowhere. I am not the Other for the Other because it is for me that the Other is Other. This is partly shown, for Levinas, by considering how absurd and criminal it is to demand of the others what I demand of myself.16 The refusal of totality is the refusal to understand demands as lingering from nowhere – this is why there is no moral code that can replace the ethical stance in Levinas. In any case, asymmetry is an important feature of indexicalism: it is not that there is the mountain-for-me, as opposed to the mountain-in-itself that could be only viewed from nowhere, and me-for-the-mountain but rather that there is a mountain-for-me and the other which is the mountain, seen from my point of view as external. The other is always as situated as an interiority – there is no external world independently of a viewpoint. Conflating exteriority with totality leads to the idea that realism about the external world has to entail the postulation of a totality; if there is a totality, there is no genuine exteriority. To consider that the internal is the external for the external is to engage in something akin to what John McDowell labeled a “sideways-on picture”: a picture that would separate conceptual operations on one side and the world at the other.17 In this case, the sideways-on picture would also produce a totality by putting the deictic operations on one side and the world at the other – or the indexicals by means of which I occupy a position and the Other who is seen stereoscopically from a different (and reciprocal) set of indexicals. We can alternate between the points of view of two interiorities, but the stereoscopic view would require that we attain both at the same time – and this is the sideways-on picture. In other words, we assume that there is a description of the external that is not to be done in terms of anyone's indexicals – just like McDowell intends to exorcise the idea that the world is alien to our conceptual operations. As a consequence of exorcizing this picture, we have to be careful with third-person descriptions; if they are indexical-free, they slip in a sideways-on picture. If they are themselves from a position, the predicates they are using to describe the others are supported by the deictic operations that distinguish between the describing I and the others being described.

Totality can also spell an appeal to a common denominator or a neutral predicate that would apply both to what is internal and what is external. This sense of totality is expressed by the idea that the Other is another I, that the Other is an alter-ego. Edmond Husserl, in his fifth Cartesian Meditations, attempts to show that the phenomenological endeavor he had described in his first four meditations entails no solipsism.18 The difficulty lies in that the point of departure of the endeavor is the subject of intentional acts and, if it is so, any other subject would be a product of these intentional acts, and therefore not independent realities on their own. Husserl then proposes to dispel the fear of solipsism by coupling his phenomenology with a (neo-)monadology that enables him to consider the other subjects as a reflected image of my own ego, source of all intentional acts; this reflected image compose an alter-ego. The alter-ego is just like me, is a specular image of myself and my own intentional acts. Levinas is explicit in his debt with Husserl and his phenomenological method that allowed him to engage in a thoroughly first-personal way of thinking in philosophy. Husserl is right in starting out in the first person and is also right in worrying about the danger of solipsism. However, the idea of alter-ego seems to him as just the wrong path for, as he puts it, the Other as Other “n'est pas l'alter-ego, c'est ce que moi je ne suis pas ».19 The problem with the alter-ego is that it takes my deictic operations, that are implicit in my intentional acts, as a predicate that can be projected in what is external. Levinas' rejection of the alter-ego could seem like an extreme movement, after all if the external deserves respect, it is reasonable to conceive it as an ego, as an interiority just as myself. Levinas' dissatisfaction with the idea of the Other as an alter-ego arise, at least, from two lines of thought both of them illustrative of an indexicalist way of thinking. First, a rejection of the totality that symmetry brings – the alter-ego is a specular image of the ego that replaces exteriority by more of the same. Second, a rejection of the idea that the Other stands in the same foot as me. The Other, deserving the respect of a capital initial, is not like me, not familiar, but precisely what is external and stranger to me; this stranger character is brought to the fore as I acknowledge the Other as another. My intentional acts are grounded in the deictic operations that shape my interiority and these include the distinction between same and other through which the borders of my interiority are drawn. Levinas prefers to the idea that the Other is to be seen primarily as like me – the idea of symmetry – the idea that the Other is superior, asymmetrical and external to what I am. To be sure, I can take into consideration what I know about myself to understand the Other, but not because my effort is to step in the Other's shoes but rather because I understand the external as such only when I let the Other step on my shoes. Respect for the external is not to project me on the Other, but rather to let myself be infected by the external, by the stranger. This is what Levinas calls substitution, replacing oneself by another – acting not only towards the Other, but for the Other.20

A final sense of totality to be rejected by the indexicalist is the idea that a complete description of the external world is possible. As I said, this is where indexicalism looks like a case concerning the impossibility of metaphysics. Here again, Levinas helps in his insistence that the Other is never fully captured by me. Every description is not only from the viewpoint of an interiority, but also incomplete. Indexicalism holds that being is indexical and therefore that nothing exists but as this, that or the like from a positioned point of view. There's more to what is there than what I can contemplate and describe. Something is this for me and that for another position – or same for some and other for others. The impossibility of a complete description – which would be like a view from nowhere – is a consequence both of anything being more than what it is for an interiority and of all existence being indexical. The same and the other are never only the same and the other and no indexical will exhaust what anything is – because they can be something else pointed from elsewhere. But couldn't the different situated description of an object – this desk-for-me and this shelter-for-the-cat – be merged together? Granted, this desk-for-me cannot be exhausted neither by a description oriented by my deictic operations nor by any other situated description. What about an aggregate of different situated descriptions? A first problem with such an aggregate totality is whether one can aggregate all the possibly infinitely many situated viewpoints to form something that could be a candidate to a complete description. Even if this aggregate is conceivable, there would be no such thing as what is described by is since being is always indexical – a cubist view from nowhere is no less a view from nowhere. A cubist totality is still non-situated and it provides no complete description because it extracts out the deictic operations that compose being if being is being indexical. From this indexicalist point of view, that cannot be a complete description and a cubist aggregate, apart from being hardly complete, would provide no suitable description.

2.3 Demonstratives and proper names

John Perry has advocated the idea that indexicals are essential in that they cannot be fully replaced and they play a role that cannot be played without them.21 His thesis is that beliefs and belief dynamics – and therefore elements of thought and speech – cannot be entirely understood without an appeal to indexicals. In particular, they cannot be properly accounted in the model of propositional attitudes according to which a belief is a relation between a subject and a proposition that can be otherwise stated as an independent relatum in the relation. When I come to believe that I'm the one making a mess with paper packaged sugar in the supermarket, or when I come to believe that the meeting I'm scheduled to go at noon is now, the proposition and the subject remain the same while by belief state has clearly changed. The essentiality of indexicals, for Perry, implies that beliefs are positioned and a great deal of what we think depends on de re content. Indexicals, he argues, challenge the very idea that there are propositions constructed in a de dicto manner that mediate a relation between a subject and the object of her belief. We cannot do away with the de re nature of belief and Perry asks the question, relevant for an argument in favor of indexicalism, of whether we can instead discard the de dicto element in beliefs. Doing so, he says, will imply that we “would think of belief as a system of relations of various degrees between persons and other objects”.22 A belief would a relation to something that cannot be described but in the middle of things, so that there is a difference between not believing that I am making a mess while believing that someone is and coming to believe that this someone is me – or coming to believe that noon is now. Perry suggests that “I stand in the relation, believing to be making a mess, to myself”.23 Clearly, there are many ways to stand in such a relation because there are different indexical words that would amount to the same – the relation is essentially de re. Then he concludes that “de dicto belief might be seen as merely an illusion, engendered by the implicit nature of much indexicality”.24 The essential indexical, that Perry takes to be a problem for a model of belief involving ultimately only de dicto content, is also a problem for any metaphysics that assumes its mission can be accomplished leaving indexicals aside. The indexicalist claims that if beliefs – and thoughts and speech – are unavoidably de re, such is the nature of reality, it cannot be understood without indexicals. Parts of the universe pointing at other parts of the universe is what gives existence to both of these parts of the world. Reality is such that everything in it is in the middle of it. This is why the essential indexical is a component of indexicalism. Further, if a metaphysics is something akin to an articulated class of beliefs – that provide a general account of reality – it cannot pursue these beliefs understanding them to be ultimately de dicto and ignoring the essential indexical.

Not only Perry's essential indexical, but all the transformation in the philosophy of language brought about by works on direct reference by Keith Donnellan, David Kaplan, Howard Wettstein, Hillary Putnam and Saul Kripke, among others, contribute to a fleshed out indexicalism.25 Wettstein summarizes what he called the revolution in the philosophy of language brought about by direct reference in a motto: linguistic contact without cognitive contact.26 The idea is that one can point at something through linguistic expressions – demonstratives, proper names, natural kind terms and even definite descriptions – without having a correct description of what is pointed. In other words, one can appropriately refer without possessing a truthful description of what is referred – or, still, denoting dispenses with de dicto contents. If it is so, one can successfully denote x without being able to position x in a view from nowhere. Kripke, embracing an analysis of names recommended by John Stuart Mill, argues that proper names are not identical with any definite description associated with them.27 Socrates was a philosopher in the actual world, but there should be a different possible world in which the same person was a sailor; as a consequence, one can refer to Socrates through his name without knowing anything about him, believing, for example, he was a fat miner from Potosí. Kripke's ideas on proper names followed from his work on the semantics of modal logic where he introduced an idea of possible worlds significantly different from Leibniz's. It is crucial for Kripke that the same Socrates could be in different possible worlds – that he has a trans-world identity, preserved in different possible worlds. The use of the name 'Socrates' in suitable circumstances refer to the individual Socrates that happens to have been a philosopher in the actual world. That individual is detached from what is true of him in the actual world and indeed of any definite description – and, in that sense, individuation precede description.

This precedence is revealing: what makes Socrates an individual is that he was first denoted by a name, 'Socrates'. It is the pointing of something through a proper name – a deictic operation – that makes Socrates an individual, with a trans-world identity. That deictic operation identify something independently of any description. Kripke understands that “Phosphorus is the morning star” fixes the reference of the name 'phosphorus'. The sentence – which contains explicitly indexical expressions such as 'morning' – can be false as a description of what phosphorus is. It could be discovered that a celestial body other than Venus shines first in the early morning in Earth's skies. But once fixed the reference of 'phosphorus', it follows that necessarily, yet not known a priori, phosphorus is hesperus. Kripke uses an example from Putnam to clarify the notion of reference-fixing, which is of great importance for metaphysical indexicalism. It can be found out that cats are robots sent from extraterrestrials to spy on human domestic life. In this case, the sentence “cats are animals” are false and still fix the reference of cats. Now, if descriptions that fix reference had to be true, the consequence of the discovery would be that cats don't exist – Kripke, and Putnam, understand that cats are already individuated for precisely because of the false sentence the discovery that cats are robots have any content. The discovery is about cats, identified by the false sentence describing them as animals. The false sentence manage to individuate something so that it fixes the reference of 'cats'. In other words, it is not the truth of a description that denotes. It is not the substantive content of a description that determines what the sentence is about; rather, it is the deictic operations carried on by the description that do the trick. The lesson for indexicalism is that reference-fixing, a linguistic deictic operation, is independent of the truth of a description – and the substantives in the description have an underlying deictic role. Indexicalism holds that being is being indexical and that deictic operations like reference-fixing is what brings something (individuated) to existence.28 If there is more than one reference-fixing procedure for one term, or more than one set of deictic operations for one interiority, then a convergence is discovered. To discover that phosphorus is hesperus, for example, is to find out something about a single position from different viewpoints. To discover such convergence is not to engage in a view from nowhere, but just to find out that one position can be viewed otherwise. I can then spot the morning star while aware that it is also the evening star – and that I could see the same thing later in the day from another viewpoint. Sameness, or otherness and exteriority, implies no totality.

Kripke's lesson concerning how descriptions do a reference-fixing job is reinforced by earlier work by Donnellan on definite descriptors.29 Donnellan distinguished two uses of definite descriptions, an attributive one and a referential one. In the first use, but not the second, something is attributed to what is described. In the referential use, “the last planet in the solar system” can refer to Pluto even if it is no longer considered a true planet. The reference is already fixed, so the truth of the description becomes irrelevant, unless attributive usages are also at stake. Indexicalism, together with any approach that considers de re content to be primary, holds that a referential use – sometimes irrelevant – underlies any attributive use of a definite description. Attributive use of a definite description cannot be fully reduced to referential use for a description has a truth-values that determines what is satisfied by them. Still, that definite descriptions have an underlying referential usage show that they perform deictic operations. They also point at what Perry calls “the implicit nature of much indexicality”.30 Expressions like “the last planet” and “the solar system” are respectively an implicit demonstrative – 'last' is a position pointed from center of the system – and a proper name. Terms like 'last', 'far', 'faster' or 'smooth' perform implicit indexical work; they are relative to the position from where the words are uttered. The first modes of Aenesidemus show how things are far from some perspectives while close from others, sweet for some tastes while bitter for others, small when seen from afar and bigger than the sun when seen from nearby and so on.31 Many expressions seem to carry no indexical content – seeming purely substantive or de dicto – because of a tacit common point of view shared by many language users; so we say that the road is smooth (assuming that one will drive and not craw on it), that the marzipan is sweet (assuming that one will not mix it with sour cream or eat it after consuming mashmellow), that the sky is blue (assuming the ordinary day-time lighting). Several similar expressions are less than explicitly indexical because a tacit point of view in the middle of things is assumed – a point of reference or a fixed standard for measurement is implicitly established. Arguably other deictic operations could be hidden by a shared and tacit point of view. Prototypical substantives like 'mountain', 'book', 'sky' or 'liquid' could be regarded as hardly indexical because a point of view is also tacitly assumed; a mountain can be a shelter, a book can be a source of food, the sky could includes the Earth viewed from Mars and a drop of liquid can be a solid surface for some microorganisms. From where I utter these words, the position being assumed is so granted that it seems irrelevant.

Kaplan extends the Kripkean disentanglement between reference and descriptions by deflating further the substantive character of any demonstrative expression.32 He concocts an expression – demonstrative that, dthat – that forces a referential use of definite descriptions. It is as if he's attempting to show how descriptions can be stripped of all its attributive employment. There is, he claims, a demonstrative, indexical structure behind the interplay of descriptions. Kaplan's analysis of demonstratives have several important elements for the indexicalism I'm constructing. While Kripke takes reference-fixing something that is still attached to a pre-existing time framework, Kaplan is more willing to free demonstratives from such a (substantive) constraint. Kripke considers that there are some essential properties of what is referred that depend on the history of the introduction of the denoting expression. So, he says a table denoted by 'this table' cannot be the same while loosing its historically defined essential properties:

though we can imagine making a table out of another block of wood or even from ice, identical in appearance with this one, and though we could have put it in this very position in the room, it seems to me that this is not to imagine this table as made of wood or ice, but rather it is to imagine another table.33

This table could not be this table while being made of a different material – its material constitition is an essential property. Kaplan, in contrast, is open to the possibility that demonstrative expressions like 'this table' to have minimal substantive features in its content. If this table was made of ice or another block of wood, it would still be this table – the table denoted by the demonstrative expression. Kaplan distinguishes between the content and the character of an expression: the former is what it expresses and the latter how it does – through me saying 'I' or someone else saying my name. In the case of indexical expressions, content is just its reference. If this table could have been made of a different material, the reference of the expression 'this table' would still be the table in the position that is pointed. Essential properties that make this table what it is are substantive properties that go beyond the deictic operations that shape the border of the table denoted by the demonstrative expression. The position pointed at by the expression would be the same, the deictic operation would be oblivious to any (possible) substantive essential properties of this table.

2.4 Tense

The tension between substantives and indexicals appear also in discussions concerning the nature of time and its relation to change that followed from John McTaggart's defense of the unreality of time.34 McTaggart's despair concerning the reality of time stems from his conviction that a B-series of substantive, de dicto, markers – such as 20th century, 2018, 30th of January – is not enough to fully understand the passing of time. He deemed that we need also to resort to an A-series of indexical markers such as 'last century', 'this year' or 'today'. In an A-series we can think of past, present, future and the passing of time; it is through the A-series that we have a sense of tense. In contrast, a B-series gives us no sense of now, it provides us with a calendar viewed from nowhere, a handless clock. In order to understand change and events that take place in the passing time – they will happen, are happening and have happened – McTaggart thought we need the indexical series. The sense of genuine change and the existence of events require a de re sense of time that a B-series alone cannot provide. He challenged the idea that a B-series is sufficient for time to be what it is; the substantive series can present myself down the mountain in the morning, myself up the mountain at noon and myself back down in the evening – but it cannot make sense of the change between the states and therefore of the event of climbing the mountain. Without events, it could be hard to account for trans-temporal identity for there is no change between one state of an existent to the other; a world without events is a world without a present state and therefore every moment is simultaneous because no moment elapses and makes room to any other. A handless clock can only represent the passing of time if we also understand that, say, one o'clock has a past and a future. McTaggart, however, understands the A-series as unacceptable as an account of time: it makes reality positioned, relative or incoherent. It cannot be that a reality has a tense, say it is in the present and not in the past and it cannot be relative to a time or at all times at the same time. McTaggart understands reality as forming some sort of totality – neutral with respect to any point of view, absolute and not relative to a position, coherent and not having inconsistent states because they take place in different moments. It cannot be real only that I am in the top of the mountain, it cannot be real now that I am in the top of the mountain and real later that I'm not and it cannot be real that I am and I am not on the top of the mountain. If reality is to form a neutral, absolute and coherent reality, the B-series would be suitable, but it cannot fully account for temporal phenomena. McTaggart then concluded that time – and the A-series it is required for it – cannot be real.

Realist responses to McTaggart involved two paths: the one of trying to conciliate genuine change and events with the B-series and the one trying to show that time can be real even if it requires the A-series. The first path assumes that his picture of reality as a totality is right and time has to be understood within this framework if it is to be real. The second path endorses a different picture of reality in order to show that, even if incorrigibly indexical, time can be real. Kit Fine has studied with some details the alternatives in the second path.35 He considers how to drop each one of the assumption McTaggart makes about reality – each at a time. If we reject that reality is neutral, we can assume that it is in the present; only present states are real states. If we reject that reality is absolute, we can assume it is one way in the present and another way in the future. If we reject that it is coherent, we can say that the past and the present are somehow real although inconsistent – which wouldn't allow for anyone to state, in the present, anything inconsistent.36 Fine favors the last two alternatives over the first one which gives up the idea that reality should be neutral. The problem of withdrawing the requisite of neutrality – and embracing something like presentism – is the ultimately unjustified choice of one tense over the others. The third option – dropping coherence – is worked out by Fine in terms of an incoherent totality: to preserve totality at the cost of incoherence. It is an interesting option, akin to a cubism of tense perspectives.

It is, still, a view from nowhere. If we privilege exteriority and exorcise the quest for totality, the passing of time before the present appears like what comes from outdoors. Indexicalism would reject the single tense alternative. Indexicalist realism conceives reality as having an alien component, the time that will come: reality is many tensed and is in itself positioned. There is no unifying view, tensed or otherwise, that would bring together all moments of time. The present is not the only tense but it is the indexical starting point that carries the weight of the past and faces what is beyond in the future. From an indexicalist point of view, the B-series is intelligible only because it has an underlying deictic structure akin to the A-series; in most if not all B-markers of time, there is a load of implicit indexicality – from 'BC' to 12:00pm, from cenozoic to spring. When we consider the interiorities with respect to which these markers constitute themselves as positions, we suspect that the B-series is the one that is dispensable for the understanding of time. What the A-markers bring to the picture is the crucial sense of 'now': no matter its duration in the time of ants or in the time of stars, 'now' is a threshold. It is the image of me facing beyond, of what is after my border.

2.5 Object-oriented

Graham Harman has been proposing a philosophy oriented to objects. He contends that objects relate to each other in a way that resembles the descriptions Heidegger gives of the physis and of the Nähe: they make themselves present as much as they withdraw. They are not simply what is made present to us, but there is a realm of their own where their presence escapes any perception or sensual quality and takes place of their own accord. He also takes the distinction between the vorhanden and the zuhanden that Heidegger introduces much earlier to be equivalent to that of positionality and proximity.37 While the latter reveals a coupling of objects independently of any framework of making something apparent, the former is like a broken tool that invokes redoing what had taken place by introducing a theory or a practice. The contrast is similar to that between the world where physis takes place on the one hand and the artifact that replaces worldly relations by a relations within a cartography, within a prefigured space. Importantly, both theory and practice requires positionality, a thetic framework that arranges objects so that they are placed in a conscription. There are the available sensual objects, and by that Harman means objects available to any other object, in contact with any other objects and there are the withdrawn real objects present but not available. The distinction between real and sensual objects come together with that between real and sensual qualities to form a quadruple structure that is common to any actual existent.38 The four elements are independent from each other in the sense that they can be perceived on their own; together they compose any object and, a fortiori, anything that exists. The presence of real objects (and qualities) commits Harman to the idea of a limitation: nothing is capable of grasping reality as it is. Access is finite, there is an element associated to proximity – and to what is zuhanden – which escapes any experience, any sensual contact. Real objects (and qualities) are situated, they make themselves present not because they are available, but because they are in proximity. A central feature of Harman's account is that he conceives of real objects and qualities as noticeable in connection to the inevitably partial exposition of their sensual correlates; they escape like noumena, except they are like things-in-themselves beyond the access of any other object. There is a hidden, withdrawing, escaping in-itself in each object and it is not accessible by any other object. Harman makes the Kantian transcendental distinction between phenomena and things-in-themselves into a general feature of any object. There is an ultimate component of any existent that is never shown, and this element beyond the reach of any access is all pervasive as “[e]ven inanimate objects fail to grasp each other as they are in themselves”, he writes.39 “[F]initude”, as a consequence, “is not just a local specter haunting the human subject, but a structural feature of relations in general including non-human ones”. 40

Harman's object-oriented ontology is adamant in its commitment to the rejection of the bifurcation between the human and the non-human realities. Just like indexicalism, his approach attempts to displace the human from the center-stage in philosophy – his is a general account of the object that would make no exception for the human subject. Further, Harman incorporates the idea that reality is intrinsically tied to exteriority: the real object is the Other to perception, it is what escapes any attempt to access, a left-over that cannot be assimilated to any sameness.41 The real object is present while unaccessible, like the Other that makes itself present in a deictic operation. There are elements of exteriority to the very (quadruple) structure of any object, any object is in itself other to anything else – this otherness is precisely the in-itself that escapes the access of anything else. Harman cashes out exteriority in terms of real objects that escape contact and makes itself present while being alien – it is as if a horizon where looms the unaccessible were always part of any territory. The Other has a mapped place in his quadruple structure of any object. However, by doing that Harman turns the Other into a substantive; it is as if there were a position in the cartography of the object where otherness lies, a position that can be described, albeit not accessed, through a substantive description independent of any deixis. His objects can be described as having absolute withdrawn dimensions, irrespective of any other object that attempts to access it and therefore independent of any deictic operation. Real objects are like Heideggerian proximity in that they are coupled and yet not viewed or intended by any situated act which would position it and remove it from their proximal state. But by understanding proximity independently of any deixis, Harman positions the real object independently of its situation; it is an unaccessible feature viewed from nowhere. Once exteriority is viewed as reality that can be viewed from nowhere, it stops being exterior in the sense that it is not exterior to anything in particular.

As a consequence, there are senses in which Harman's object-oriented ontology turns exteriority into totality. Surely, Harman rejects the idea that a complete description of the external world is possible. Real objects (and qualities) are inexorably elusive and makes sure that no complete description can be given, not even a description from the view from nowhere. Secrecy and inaccessibility are features of the furniture of the universe. Similarly, Harman's clearly ontology makes room for interiority, as each object has an interior life independent of any connection with the totality formed by everything. Each object is capable of perceiving as much as it is available to the perception of the others. His object-oriented ontology is committed to the interiority of what exists and of perception as a source of making the world what it is through agency. Harman postulates the Other, understood in terms of real object, as a genuine feature of the furniture of the universe. Still, if totality is understood in terms of the postulation of neutral terms – and through a structural symmetry between the different existents – Harman's object-oriented ontology, by turning the Other into a substantive real object, ends up cashing out exteriority in terms of totality. The withdrawn structure present in the other object is symmetrically postulated in any object – any object is like an alter-ego, like the image of the same object. The quadruple structure is therefore a common denominator or a neutral predication applied to any object that can be seen from nowhere: the world is a world of objects that have withdrawn parts. His commitment to the (paradoxico-metaphysical) desideratum of exteriority – in the form of a real object that escapes access – is at least mitigated by some adherence to the idea of totality in these senses. Harman's objects are viewed from nowhere and analyzed in terms of a neutral common structure where object enjoy symmetrical relations between each other. As a consequence, real objects remove the deictic sting of the Other by flattening everything with substantives.

One could say that the same can be said of indexicalism. If interiorities are surrounded by an indexical environment – that is endowed with a beyond, with an outer – they are endowed with a structure that is common to each of them. Just like with objects, each interiority could become the mirror image of the other. Indexical environments are nevertheless not structures. Indexicalism refrains from a view from nowhere: an interiority is always the point of view from which metaphysics is done. The other interiority has a deictic environment which is nothing substantive, it is just something we point. The indexical character of reality makes no ultimate description possible, unless it is no more than a pointing at what is beyond. Levinas often restricts himself to the first-person and to the view of the world provided by a single interiority. In doing that, however, I believe he is hinting towards the indexical character of any description. Harman's objects present us with a structure that is independent of our standing, they can be described no matter whether they are internal or external. Harman's object-orientation perhaps illustrate a more general tension between the external and the real; a tension expressed in taking the external world to be an independent reality – independent of being external. Often the very idea of reality appears as hypostasis of the deictic exteriority, it is no more than a tamed surrogate of otherness. Reality, in this cases, appear as totality dressed as exteriority. Indexicalism, on the other hand, has an unreachable beyond not encompassed by a system (a system of objects) but by the very indexical nature of reality.

In contrast with Harman, Tristan Garcia proposes an object-oriented philosophy that postulates no real object. He introduces his position in a contrast with two alternatives, a view according to which a thing is defined in terms of a substratum independent of any of its predicates (events, states of affairs, qualities and relations) and a view which understands the being of a thing as distributed throughout a bundle of its predicates. While in the former promoted the hypostasis of the peculiarity of a thing – the thing-in-itself – the latter understands that peculiarity in terms of the discernible differences with anything else. What makes a thing what it is is therefore either its (indiscernible) ingredient that constitutes an in-itself independently of anything else or the vector of predicates. The contrast between these two models is a contrast between the postulation of an in-itself – like a real object – on the one hand, and the vectors of being coming from elsewhere. In Immaterialism, Harman contrasts his object-oriented ontology with Latour's actor-network theory using the case of the VOC (the Verenigde Oost-Indische Compagnie, the Dutch East India Company). While the approach espoused by Latour understood the VOC in terms of its actions and relations, its operations, alliances, characters and sponsors, Harman's object-orientation postulates a real VOC underlying all actions and relations that the company was involved. Harman hence defends the virtue of the first approach considered by Garcia against the second, endorsed by Latour. He argues that a real object – an in-itself – is what best explains the trajectory of a thing and the history of the VOC in particular. Garcia rejects both views on the VOC and on things in general – he advocates a third view that postulates a peculiarity of a thing as such without appealing to an in-itself. He agrees with the second view when it comes to depart from the postulation of an in-itself: a thing has to have something outside it, it is nothing but in contrast with something other than itself. But he also agrees that a thing is not only its contrast with everything else – it is not only what is discernible. He then introduces a third view that contrasts with the other two in a way that is specially interesting for indexicalism. His view is that a thing is like a territory and what makes it what it is is its borders that divide what is inside the thing and what the thing is inside. The border is a contrast between the external and the internal that traces the peculiarity of the thing. To be sure, the border itself is drawn by the predicates in the vectors that makes something discernible – it is the bundle of predicates that distinguishes the thing from anything else. But the border is not only what draws it; like in a national state: the border between Germany and France is to a great extent independent from the history of the wars that place it where it is. The border itself is what makes the thing what it is: a thing, Garcia writes, is like a difference between content and container. It is like an immaterial bag. Anything could be the content of a bag, but not the bag itself – there is no itself, there are only contrasts.42

That independence between the drawing of the border and the border itself is what Garcia calls de-determination.43 It follows from a crucial distinction between objects placed among others and things that only placed within their containing universe. Although Garcia in Forme et objet presents things before he presents objects,44 Cogburn, in his book on Garcia, reckons that the best exposition of his system would be to start with objects and then move to things.45 Objects are determined by their predications, their relations, their qualities, their potencialities, the events they are part. From an objective point of view, there are objects and their predicates that determine objects; there are objects giving an identity to others. From a formal point of view, there are things and the universe around them; things are defined by their borders drawn by objects and predicates, but they are de-determined, they carry on independently of the history of their borders. I understand this movement of de-determination in terms analogous to those used by Kripke and other theorists of direct reference to distinguish between a true description and a fixation of reference. Consider the cats in the example by Putnam I mentioned above. The borders of the kind 'cat' – considered as a thing – were determined by descriptions like “cats are animals”; such description can be false and still fix the reference of the kind 'cat'. This is because when the word cat is introduced, it carves up the thing 'cat' and distinguishes it from anything else in the universe that contains it. The cat is considered, from a formal point of view, independent of the truth of the descriptions that determined its borders; a thing is therefore determined in order to be what it is and de-determined to be considered formally within its containing universe. There is a formal reality independent of the objects that bring being to a thing, this formal reality is not an in-itself, but a difference between the internal and the external. Similarly, in order to fix the reference of something it is not enough to make sure there is a difference – say, between what is and what is not a cat. To fix a reference is not to point at a substratum that subsists in itself or to a first substance in the way Aristotle conceived of his ουσία προτε. Importantly, there is nothing substantive that makes a thing what it is; a thing is what results from the de-determination that is made possible through the effort of determining objects. A thing is an object removed from the realm of objects and placed on its own in an environment composed only of its containing universe. De-determination produces a difference, and not anything in-itself: it produces a border. The border is not only a line drawn, it fixes a region that is circumvented by what is discernible; the description of cats as animals not only state something in its attributive use but also does the referential job of pointing at parts of the world. Every determination comes with a de-determination – there are no object that cannot be viewed as a thing.

Garcia's object-oriented philosophy, and its notion of de-determination, are closer to indexicalism than Harman's. In particular, it is through other objects that things become what they are, there is nothing substantial underlying them. Further, his criticism of the in-itself and of compactness – being without borders – are part of a thorough exorcism of the assumptions of totality in metaphysics. The very idea of totality entails a content with no containing universe, totalities have no border, they cannot be any thing. The universe associated to a thing, in contrast, is no totality for it is both infinite and indefinite; in fact a universe is as relative to a border as its circumvented thing. There is no universe but a universe for a thing. Garcia's system falls nevertheless short of indexicalism. Though the formal characterization of a thing – and its universe – appeals to nothing substantive, it is still not indexical. While things are de-determined and therefore relative to the position they are made to occupy among objects, things are understood independently of these objects. It is as if de-determination deprived objects not only of their subservience to the truth of descriptions but also of the deictic character of the borders that constitute things. The formal image of a thing and its universe preserves an element of the substantive; Garcia's de-determination is still somehow de dicto for the operations that keep the border in place are removed from any formal consideration. Although the border between France and Germany is independent of the history of the two countries, it is only with respect to countries and citizens (and merchandise, taxes, the jurisprudence of laws etc) that the border exists. Borders are not substantive; they cannot subsist without deictic operations, operations that take the form of this and that, here and there, same and other. Garcia defines things in terms of their borders, but some of the indexical character of borders that enable de-determination – and reference independently of the truth of any description – end up dismissed as the formal account of things obnubilates their constitutive deictic operations. Still, the formal account enables the rest of the world to be contrasted with the thing in terms of its universe, infinite and indefinite. From the point of view of the thing, the formal account makes explicit its relation to the surroundings, a relation where exteriority is of central importance.

A full-fledged indexicalism would make use of de-determination as a way to show that objects are themselves reliant on deixis. Indexicalism is therefore a form of undermining objects, in the words of Harman.46 Objects are constituted by their friction with other objects; indexicalism undermines them by considering their relative position – it is a position-oriented metaphysics. Indexicalism views de-determination as an operation of reducing what is substantive to something demonstrative – or rather to make explicit what is indexical in what appears to be self-standing. In contrast with object-oriented approaches, indexicalism places exteriority in the furniture of the universe. It is not about objects forming a totality – or about a thing and its undetermined surroundings – but about what can be pointed as external, as outer, as in the outdoors. Indexicalism is a metaphysics presented from the point of view of the border – of a situated border.

3. Towards an indexicalist paradoxico-metaphysics

Because indexicalism posits exteriority – the Other, the outer, beyond, outdoors – as a main constituent of the furniture of the universe, it could seem to be not about the external world, but about what makes the external world external. This brings part of the air of paradox around it. It could seem that indexicalism is just placing the external within its categories but not addressing what is there, in the Great Outdoors. If we reject the claim that the Great Outdoors form a totality and that itself risks exorcizing exteriority, what is outer can only come into the picture as such, as the outer, as what is external from a point of view (which is mine or ours). But then it seems like indexicalism is providing its own critique of the metaphysical endeavor – and it is. It is showing that there is no ultimately substantive account of reality in general. For indexicalism, this entails that no maximally general account of reality can be given. However this is so precisely because the only maximally general account of reality that can be given is in terms of indexicals. This is a metaphysics that would posit something ultimately beyond its reach – there is always something beyond the pale, no matter where the line is drawn. Further, the line is drawn in different places from different interiorities – yet there is no end to exteriority. This is paradoxical – and at the same breath, it is metaphysical.

Indexicalism is a way to present reality from within – and from within there is always something outer, something beyond. There is an open horizon to metaphysics – and this, in a sense, means that any metaphysical view, including indexicalism, is restricted or severely bounded. The maximally general view of reality is one that constrains a maximally general view. Further, it is not something substantive about reality that constrains it, it is its indexical nature. On the one hand, indexicalism holds that there is nothing but indexicals – situated interiorities and their deictic operations – and the drawbacks of many past metaphysics are due to a failure to avoid undermining indexicals through substantives. On the other hand, it holds that no metaphysical view – not even indexicalism – can provide a complete account of reality precisely because there are irreducible beyond it. It is a total view that precludes totality. Or rather, tt is a recipe to avoid totality that includes postulating something about everything. In other words, it is a recipe for paradox. But arguably, it reflects how things are.