Total Pageviews

Sunday, 30 December 2012

Empiricism and the philosophy of thought

Deleuze finds in empiricism a mysticism of the concept (in the opening of Difference et Répétition). Empiricism claims to be able to see concepts in things – it is as if something else provides thought content for the thinker, the heterological in us, as Derrida says in Violence et Métaphysique. Empiricism holds that experience enjoys some degree of independence from the spontaneity of thought by escaping its spinning. In order to do that, experience needs to be experience as in the sense that its contents have to be available to thought. The long lasting suspicion against empiricism is that the whatever content experience provides is somehow not earnestly earned (it cannot be used in justification, it cannot be self-standing or it smuggles in elements of spontaneity – or our own sovereignty). The content of experience is some sort of (unacceptable) given. The kernel of this kind of criticism is the assumption that there is such thing as a content of the perceptual experience. McDowell's attempt to rehabilitate (minimal) empiricism is in the right direction because it makes content relative to the conceptual capacities of the thinker. In fact, response-dependence theories of secondary qualities favor a defense of empiricism that is not a defense of a fixed content for experience. The merit of empiricism is that it allows for experience to see "something else". So, instead of saying that experience gives us (crystalized) physical objects, empiricism holds that we capture some sort of sense data that can be explored in different ways. Now, empiricism doesn't need to commit to sense data as the content of perceptual experience - experience can always provide a different content.

Experience can be thought as a rough capture. It can be entrained to be in different rhythms, catching different signals, repeating different patterns. This is the thrust of Sellars' message: the plasticity of experience. Paul Churchland's Scientific Realism and the Plasticity of Mind explores the point in interesting directions. There is a fine line between saying experience can provide any content and saying it makes no difference (no epistemological or transcendental difference). The difference experience makes is not in providing a given content but rather in eroding thought from within. Our senses are fully integrated in thought, but empiricism has that they are not always docile and subservient. It is about the rebellion of the senses.

Wednesday, 19 December 2012

Rhythms, rules, regularities (and Anerkennung)

I feel a tiny bit more confident to go back to the hairy issue of normativity and recognition. I used to favor the idea that an act of recognition was no more than a special case of what Latour, for example, calls alliances. But alliances still sounded as if they involved themselves a degree of recognition - of the alliance being made. I thought maybe we can do better if we think in terms of rhythms, modulations, transmission and reception. In terms of capture, to use the vocabulary of a recent post here. We can start considering our thinking about induction and the quasi-inductive nature of the considerations on rule-following that Wittgenstein inaugurated.

Induction is a form of entrainement. It is about creation of expectations that depend on the entrained insertion in a cycle as cycles have built-in expectations. So, we can only think about the whereabouts of the sun tomorrow if we have a sense of rhythm - enough of a sense of rhythm to be sure when tomorrow is. If all our cycles are at the same time put into question, we can no longer figure out what is put into question. Doubts about the future makes sense only if there is a clock, a cycle or an expected rhythm we're tied to. Induction is about habits - habit-making is entrainement and therefore production of further cycles. Analogously, to learn to follow a rule (say, +2) is to follow a rhythm that is being shown. It cannot be done if we cannot associate the rhythm to be learned to others that are already available to us (i clocks or cycles we're tied to). I believe this is the grain of truth in naturalist approaches like Millikan's. There should be something in the pupil that prepares him to the learning - and I guess this has to do with the cycles the pupil already harbors. The 996-998-1000-1004 pupil of section 185 (of the PU) captures the wrong rhythm, nothing to do but to repeat the entrainement having in mind that the pupil's receptors could be such that they cannot mimic this rhythm as what is captured is something (slightly, as Brandom says) different. Concept-acquisition, and rule-following, is a way to explore important features of the rhythm of the world (of the many real patterns to be explored by different matrixes of difference and indifference). It makes recognition possible. But this story can be told in terms of capture.

The struggle for recognition is the struggle to be captured. The state is a giant antenna (and a giant entrainer) and politics is about making things received - turning them visible. But there are rhythms that skirt around the giant antenna and entrain around anyways. Concepts are things we exploit if we can grasp them - and it is an issue of being prepared to grasp, like having them in the deliverances of our senses. Additionally, somethings need to be recognized to affect others - their signal has to be captured, otherwise they don't entrain. There are things, however, like the malaria mosquito that defends the Amazon (as I mentioned in my recent paper on the register and the registrar), that broadcast signals that are modulated and transmitted without any act of recognition.

Tuesday, 18 December 2012

Seeing qualitons as qualia

Another paper of my collaboration with Eros is out. It is in this interesting Open Journal System. They republished papers from the Kirchberg Wittgenstein Symposium.
Seeing qualitons as qualia

The paper is a dialogue between Wittgenstein texts and the idea that mental tropes can be seen as qualia and somehow explored in concept-acquisition.

See our first paper on tropes and qualia here.


Deleuze's remark towards the end of Le Pli that Whitehead inaugurated a move from clausure to capture is very interesting. Ontologies of clausure have to do with closed individuals - substantialism as in the postulation of closed monads that subsist in time and substatism according to which individuated units are independent of their actual properties. (To be sure, Leibniz held obviously the first but not the second.) Ontologies of capture postulate signals, broadcasts, transmission, modulators, antennas, satellites, connections and all those things that can contaminate each other and display vulnerability to the way they receive and to how they are received. Tarde prefigured this move towards capture in his analysis of societies: social populations are tied together in a way that they are modulated by other individuals of the same kind - like cellular automata. Molecules are highly social, birds are very social, humans are less social because they capture signals from bacteria, the elements, their inner ecology etc - and these signals affect their rhythms. Planets are social to planets and large masses because few things outside the society of planets and large masses can affect them.

Capture is about rhythms - about entrainement and heterochrony. The more social a population is, less entrained by other, extra-social elements. Repetition gives rise to regularity and if what is repeated is tuned to the same signals, a regularity is maintained. I remember our house in Hove street, Brighton, where lived 4 women and only Mar was taking the pill. Mar entrained the menstrual cycle of all the others while she was entrained by extra-social elements. All this entrainement happen simultaneously. Simultaneity is the nature of chronos, the time of present: things happen at the same time in chronos, the hormonal processes, the digestive processes, the emotional processes, the social processes all at the same and this means all synchronizing each other. Entrainement is the mark of chronos - the mark of the density of the present. Now, events are entrained by the co-existing rhythms but also by the capacities of matter that can be thought in terms of its folds, the folding capacity of each thing. Entrainement is enabled by the capacity of something to fold in a way that the rhythm can be somehow acquired and modulation is possible. Folds and beats. Differences both in folds and in beats can always configure an overall difference - the folding capacities of my body enable my rhythmic entrainement and the folds I'm capable of doing. You could dance like me but have different folds responding to different beats.

Heterochrony, on the other hand, is the introduction of a different rhythm that will change this ontological jam session. This is where I find urges. But difference is tied to repetition - heterochrony comes from entrainement. I tried to explain the constitutive drift of repetition in my introductory class to Deleuze's D&R through Chinese Whisper. We do no more than repeat - and the drift is there. There are too many extra-social elements in the reception and in the transmission. These extra-social elements can be seen as micro-urges to which we are tuned. In fact, it is not about urges that (passive voice!) get recognized but rather about who capture them as such. All receptors are modulated in a matrix of differences and indifferences - a tick can be oblivious to a quake etc. The urge that is effective is the urge that is captured. Shocks and catastrophes are urges that everyone (in a population) captures. Something that intervened in the repetition and that comes from a synchronic process that is happening at the same time. Less sociality (in Tarde's sense) means greater vulnerability to urges. In a rigid society, nothing is taken as a disturbance because receptors are oblivious to where the disturbance could come (Tough Guys Don't Dance). To dance is to expose your antennas - to show what tunes you.

Wednesday, 12 December 2012

Anarcheology and the virtual

Been launching our book on anarcheology of Heraclitus` polemos. Our notion of anarcheology is associated to thought without archés - it is connected with freeing philosophy from the facts of its history. Zouzi Chebbi once connected anarcheology to his remark that to the south of the Maghreb there is no distinction between fact and version. The jounalist who interviewed us about the book defined it as a mix between anarchy and archeology. We meant to make a gesture concerning the original Heraclitus in the book: philosophy is not a slave of the genuine, authentic Heraclitus. Aharon Link recently pointed out to me that anarcheology is also understood as the militant study of the absence of archy, of government. In fact, this multiplicity of meanings is welcome, it has to do with the intimate connection between ontology and politics, for instance, in the term arché.

But I thought anarcheology, in our sense, can be linked to an understanding of virtual history. Or rather, virtual bibliography. A history of philosophy that could have happened but didn't. An attractor that didn't become a trajectory, something like this. Because thinking deals in the virtual - in what could have happened but didn't. There
are thoughts that could have put forward and could have had consequences of all kinds but didn't become actual. Virtuality has a resemblance to what is at stake in the principle of sufficient reason. Thinking is virtual in the sense that it is there but it depends on everything else - its actualization is not un-hinged, it is not unconditional. Anarcheology is an archeology of the virtual - we extract from what can be thought about the polemos, for example, text. Deleuze, thinking about the several ways to formulate problems, comes up with the idea of an empiricism of the virtual. If such idea makes sense, we can take anarcheology to be the realm of discovery within the vritual.

Aharon Link on me (version ALPHA)

So Aharon writes about me... This is a first version of his abstract for the Performance Philosophy conference in Guilford, next April.


In this paper, I will use the question of doing philosophy through
performative acts for investigating how, through crossing of seemingly
un-related elements, new entities of questioning and practices are
Diogenes was a human that was metaphorically crossed species with canine
to become “Diogenes the Dog”. I will argue that such a crossing helped
delineate who he is and how the practice of his philosophy is done, and
that the way in which Diogenes did philosophy resembles contemporary
Performance Art.
Instead of being concerned with the performance of language, speech and
speaking, we have in Diogenes a performer of actions and activities. It is
plausible to imagine a performance artist living inside a bath, wanking in
parliament (though perhaps will be arrested), or plucking a chicken and
declaring it human. Indeed the choice of performative acts in themselves
is a constant crossing between that which is “Life” and that which is
“Performance”. That very crossing - life reflecting upon a performance
reflecting upon life etc - is what I think to be a constant energiser for
using performative elements in questioning what we know, don’t know and
have no idea about..

Dressed as a bride, the philosopher Hilan Bensusan used his crossdressing
practice to bring a murdered performer Pippa Bacca, back to life and roam
the streets of Istanbul. While a performative act, such as “Brides for
Peace”, which Bacca, performed with tragic outcomes was questioning the
lines and links between life and performance questioning what we imagine
possible - Hilan’s activity was was of a more philosophical nature,
questioning the knowledge we think we have.

Another Performance Art/Act link of Bensusan’s philosophical lexicon is
The interest in Heracletian frgments as an ongoing anarchological
performative practice. Like Beuys, Bensusan makes use of
archeological-like fragments to invent time. Acts of performative nature
are practiced to cross, at will, between invented and shared realities of
the past. Through fragments which invoke an injection of cross questions
about what might question about, what they know not, and would have liked
to know of the object/fragment.
This way, both Beuys & Bensusan use fragments to create effects of
possible information crossings that tease desire within audiences.

Like any cross, there is a point-time of contact/collision of two
different inquiry trajectories. That point-time, I will argue, helps to
light up the different trajectories. Whereas the philosophical
performative inquiry is focused on How of thinking is questioned, the
artistic sensitivity is of how the sensations of imagination are
questioned. Hence, in my view, we have different, yet constantly crossable
trajectories which feed one another.

Through the examination of Bensusan’s extensive work on cross dressing,
cross speaking, cross translation, and indeed cross speciasation - which
takes us back to the post human origins of cynicism and the unknowns such
practices question - I will illustrate how performative, embodiment acts
that do cross practices, both generate new possible lives/information and
simultaneously remain at the prison cells of their propagator’s minds.

Friday, 7 December 2012

Correlations, response-dependence and the metaphysics of subjectivity

The idea of response-dependence was introduced (by Philip Pettit and Mark Johnston back in the 1990s) as an account of secondary qualities. It is a McDowellian idea: we, and our circumstances, have to be apt and ready in order for something to be grasped by us. One way of presenting the idea of response-dependent secondary qualities is to say that those qualities are in such a way that they are fashioned for some beings and for some conditions. They are somehow tuned in some frequency and cannot be captured otherwise. I need to be prepared to grasp smoothness, my environment has to help me so that I meet what it takes to get the signal. It is an issue of transmission - how good the signal is broadcasted and how it is received. A quality - or a bunch of qualities with or without substrata - is a transmission, a message that is sent towards the appropriate antennas. We can think of a correlation between the signal - the object - and its receiver so that it is a matter of fact that both are coupled. The signal is sent, whoever captures it exploits it. It helps to think of the Gibson's vocabulary of affordances. Objects put forward several different affordances and some devices are, as a matter of fact, tuned to exploit them. Secondary qualities abound like attributes of a Spinozan substance but we cannot acknowledge more than a handful of them. Primary qualities, on the other hand, are more universally captured - and it is, perhaps, a matter of degree how spread is the reception of a quality.

Response-dependence accounts are tailored to answer to skeptical challenges such as Aenesidemus' modes. First mode: the appearances of things differ according to the animal that perceives them - well, fix a thing (an affordance) for each animal or somehow establish that some animals are good at capturing some things and others not. Second mode: the appearances of things differ depending on which human subject receives them - well, either say that different people perceive different things or establish that only some humans can perceive some things (the conceptually apt ones, the virtuous ones etc). Third: different sense perceive different things - well, redness is best perceived by the eyes etc. Four: different circumstances convey different appearances - well, some circumstances can be established to be better than others. Etc. For each skeptical variation we add either a respective plurality or rather a condition of transmission associated to each thing. We end up either with an abundant ontology - a rain forest one - or with an ontology of transmitters where to be is to be a frequency broadcasted. To be is to afford. The modes of the Pirrhonist give the impression that such an ontology is made fit to answer skeptical challenges. (Whether it can answer the spirit of these challenges is something else.)

To what extent such an account is under the spell of the correlation - and to what extent it is a variety of metaphysics of subjectivity in Meillassoux's terms? It does accept that what we perceive depends on us - that what we can manage to know is relative to our correlation to what there is, to affordances. It also takes seriously the idea that it is factual that we perceive somethings and not others. The effect of the correlation is contingent - nothing prefigures which receivers will be tuned to which messages (to each affordances). Signals (or affordances) are absoutes and some of them we can grasp. (To be sure, we can do that with the aid of some favors form the world, we grasp them by epistemic luck - but here again we can sweep this skeptical tome aside.) Maybe it is a form of realism that takes correlation seriously (in the two features that Meillassoux insits: that correlation are primary and that they are factual). On the other hand, however, affordances are such that they make correlations necessary - even if they leave it up for grabs which subject would match each object.

Thursday, 29 November 2012

In defense of a Heraclitean Aenesidemus

I'm preparing the final version of my text on neopirrhonism and the ontology of doubts. In other words, about the whereabouts of Aenesidemus' thought. I guess there is a way to think about the epistemology of the polemos that makes it somehow less vulnerable to Sextus' charge of dogmatism (in PH 210-12). Such an epistemology of the polemos could start out with a new anarcheological fragment of Heraclitus (in my recent book Heráclito - Exercícios de Anarqueologia, São Paulo, Idéias e Letras, 2012). It reads more or less like this:
131. There is polemos where we don't expect, not only in weapons but also in the surprise produced by polemos itself, in the temptation for polemos and in the knowledge we acquire of it.
The message: knowledge of polemos is itself ridden with polemos. It is not that we are contemplating the workings of the doubts from outside, like a dogmatic who holds a conviction as something that stands beyond the waves of doubt. An ontology of doubt informed by 131 doesn't place doubt in a privileged ontological position which makes it invulnerable to doubt itself. On the other hand, such a position is better than sheer pirrhonism as we find in Sextus because it doesn't shy away from a speculative leap but rather engages in it while exorcising any commitment to convictions.

The move can be expressed as follows. The merit of an ontology of doubts is that of going beyond our doubts and suspension of judgement. The reverse image (an initial form of ontology of doubt) is to posit an absolute doubt - as opposed to doubts that are locked within a correlation. But absolute doubt is dogmatic - Sextus charge. The way out is to find a way to reach absolute doubt while avoiding this charge. If we put together the initial form of an ontology of doubts - the world is (also) made of doubts - with the pirrhonist stance concerning our knowledge we reach a position in which nothing - not even our knowledge of doubts - is doubt-free (our knowledge of polemos is itself polemos, as the 131 puts it). The stance of knowledge is not outside the scope of what there is and if what there is is doubts, there is no apophantic, dogmatic knowledge not even of doubts themselves.

Wednesday, 21 November 2012

Deferring and coupling

Thinking a bit about infinite regresses in justification. My argument in a paper I published recently in this blog (otherwise unpublished and five years old...!) was that when I say I have a good reason (a good justification, a reliable process etc) I am somehow deferring and coupling, that is, deferring to a reason that I endorse and coupling my belief to an existing and accepted chain of reasons. The issue, of course, is whether I can buy the justification of my belief by coupling it to an infinite chain of reasons or by deferring to an infinite process. I don't know. But it is interesting to bite the bullett and claim that there is nothing else to justification than good deferral and good coupling. That is, deferral to a commonly accepted reason and coupling to a commonly accepted (infinite) chain. After all, justification always makes appeal to accepted reasons. These reasons can be out there in chains and processes and to justify could be no more than to accommodate a belief to them. If I am a good detector of red - because the chain that starts with me being a good detector of detectors of red and goes on to me being a good detector of detectors of detectors of red etc is out there to be used -, I can say I'm justified in spotting something red.

Tuesday, 20 November 2012

Aenesidemus of Knossos and the ontology of doubts

I am in a small conference in honor of Porchat, a Brazilian Neopyrronist. I’m proposing (in my paper today) a reading of Neopyrronism. I maintain that common sense claims (together with some commonly accepted scientific theories or commonly known philosophical disputes) can be accepted by the sceptic only in order to fuel his exercise of doubting. The Neopyrronist takes something like the opposite of a speculative flight. She accepts a basis of science or common sense in order to doubt further. The goal is to maximize doubts. The basis (the Neopyrronist phenomenon) is contingent, it can be anything because any determination can enable further exercises of doubt to take off. Instead of flying towards broader vistas, the Neopyrronist strives to gain further and deeper doubts. She aims to apply her strategies for epokhé further and, in that sense, harbors an atitude where doubts are more worth seeking than belief. She craves to make more use of her weapons that include her doubting machines - say, the modes of Aenesidemus and Agrippa which enables her to suspend the judgement over all kinds of issues. She engages in a flight towards the heights of suspended judgement. It is a speculative flight, but towards a proliferated landscape of doubts and uncertainties. This Neopyrronist is a kind of the opposite of the knowledge seeker - or rather the devil of any knowledge seeker: she aims at spreading consistent and well-argued ignorance.

I contrasted this version of Neopyrronism with my ontology of doubts. Surely, the sceptic can challenge the ontologist of doubt by questioning the belief that there are doubts out there. The sceptic suspends judgement concerning the ontological status of doubts (whether they are out there, in here or in a special substance, a res cogitans). The sceptic is noncommittal concerning doubts and determinations and
takes the ontologist of doubt to work with a fixed determination, a fact of the matter concerning doubts. (The supposed reply of the ontologist of doubt is that there would be no alternative to either an ontology of doubt and an ontology of fact and, facing this tertium non datur, it is better to embrace the former). The sceptic then calls the ontologist of doubt to a battle in open field where the nature of doubts (and determinations) is also there, up for grabs. Both act as if they should look for doubts - the ontologist because she seeks knowledge, the sceptic because she seeks to dismantle all sorts of knowledge claims. The Neopyrronist makes use of some determination to unveil further doubts.

Concerning the ontology of doubts, I was told that Aenesidemus of Knossos (one of the classical sources of sceptic modes, mentioned by Sextus) moved on to follow Heraclitus and the polemos. Aenesidemus became then somehow an ontologist of doubts. Maybe then he is the missing link between the sceptic challenge on justification and the Heraclitean challenge on determination. (See photos above)

My new reading of Neopyrronism

Today I presented this paper in honor to professor Oswaldo Porchat, in an event in São Paulo celebrating his 80th birthday. The paper was given in Portuguese. In the next post I kind of summarize the argument in English.

O lugar da atitude neopirrônica

Dúvida, determinação e ontologia
Um dos procedimentos mais comuns para exorcizar a dúvida generalizada é a de insistir que toda dúvida tem pressupostos. A insistência é de que qualquer suspenção de juízo requereria que se assumisse algum ônus de prova. A dúvida, essa seria a ideia, tem também que pagar o custo do transporte – e o pagamento é feito em convicções. Este procedimento aparece de diferentes formas – na forma de indicações das condições de possibilidade do pensamento empírico, na forma de análise dos procedimentos subjacentes à atividade de duvidar, na forma de considerações acerca da presunção de verdade de uma parte considerável de uma massa crítica de crenças inteligíveis. Em todos os casos, o procedimento tenta imputar a quem sustenta uma dúvida generalizada um conjunto de convicções subjacentes. A dúvida tem custos: não é possível permanecer na atitude de dúvida sem que haja algum pilar que segure o juízo suspenso. Sem nenhum pilar, o juízo se espatifa no chão.

Uma atitude sem sustentação ou não tem credenciais epistêmcias – como ensinam os ataques céticos à atitude de convicção – e, portanto, estaria vulnerável a ser considerada ilegítima ou se sustenta a si mesma. Há de fato um outro procedimento também comum para exorcizar a dúvida generalizada que é o de encontrar auto-evidências: encontrar convicções que se auto-sustentem. Este é um procedimento que se dirige sobretudo a dúvida generalizada que se instaura a partir do que por vezes se chama de argumento da ilusão: a inferência que vai da premissa de que cada conteúdo de convicção pode ser objeto de dúvida para a conclusão de que todos os conteúdos de convicção podem ser objeto de dúvida. O procedimento tenta mostrar que há pelo menos um conteúdo de convicção que não pode ser duvidado. A auto-evidência é uma convicção que, de alguma forma, se desprende do conteúdo ele mesmo. Quanto a tal conteúdo, não há atitude legítima senão a da convicção. Este procedimento também aparece de diferentes formas – a forma do conteúdo que de alguma forma se impõe por si mesmo, a forma de uma análise da convicção que torna ela mesma refém de algum conteúdo, a forma do apelo a alguma forma de intersubjetividade relevante. Em todos os casos, a estratégia é a de mostrar que há alguma coisa que se sustenta sem sustentação – no caso, alguma convicção que se sustenta sem convicções (ou qualquer outra coisa) que a sustentem. Há um auto-sustentável. Ora, eis aqui um caminho aberto ao cético: usar uma variante do segundo procedimento para neutralizar o primeiro. Se há algum auto-sustentável, alguma atitude que não dependa de sustentação, esta atitude pode ser a de dúvida. Ou seja, o primeiro procedimento insiste que a dúvida depende da convicção para se sustentar (o argumento da ilusão seria portanto uma inferência inválida já que a conclusão – de que o tudo pode ser objeto de dúvida – não poderia ter credenciais já que dúvida depende de convicções) e o segundo procura apresentar convicções auto-sustentáveis. O cético pode tentar então apresentar dúvidas que sejam auto-sustentáveis, que se desprendam diretamente de seus conteúdos. Isto equivaleria a dizer não apenas que a dúvida é a atitude em algum sentido natural a ser tomada com respeito a alguns conteúdos (ou a todos), mas também que ela não precisa se sustentar em coisa alguma.

Um tal caminho parece um tanto impopular nos tempos atuais de ressurgimento da metafísica. Uma das ideias motrizes que embalam este ressurgimento é a de que suposições acerca de como são as coisas são inescapáveis. Esta ideia motriz, claramente, é uma das formas (metafísicas) do primeiro procedimento: insistir que sempre há que haver convicções subjacentes. Heil & Martin (1999), no texto que deu nome à virada ontológica, argumentam que pressupostos ontológicos estão sempre presentes, mesmo nas mais extremas formas de antirrealismo. Ou seja, sempre há adesão a alguma convicção com respeito a como são as coisas. A insinuação é a de que a metafísica – enquanto conjunto de suposições acerca de como são as coisas – é inevitável. Alguma convicção metafísica, ainda que ilegítima do ponto de vista epistêmico, é inescapável. Há que haver alguma convicção sobre as coisas elas mesmas – ainda que seja sobre a substância pensante, sobre os ingredientes da linguagem ou sobre a comunicação humana. A afirmação mesma de uma impossibilidade se assenta sobre algum absoluto. Sobre alguma coisa, a virada ontológica reafirma, deve haver convicção (ainda que ilegítima).

Também a virada especulativa – outro elemento do ressurgimento da metafísica – conclama a filosofia a se engajar em voos para além daquilo que a prudência epistêmica, pelo menos se tomada em sua forma mais simples, recomendaria. É que só a especulação expande horizontes para colocar o pensamento em outras paragens, já que ele está sempre localizado. A especulação permite que aquilo que tenha parecido uma dimensão particular de nós mesmos seja considerada como uma dimensão do mundo mesmo. O mundo não é diferente da res cogitans, diria Whitehead, nem diferente das gambiarras que precisamos fazer para traduzir, diz Latour e nem diferente da facticidade da correlação entre nós e o mundo, diz Meillassoux. A especulação pede que abandonemos os nichos que são sui generis em favor de generalidades mais amplas. Aqui também a metafísica é inevitável e a diferença é entre uma metafísica mínima ou modesta e uma metafísica que resulte do esforço especulativo. Enquanto o esforço cético é muitas vezes o de minimizar compromissos – se precisamos nos comprometer com alguma coisa, tentemos nos comprometer com nada –, o esforço especulativo é o esforço oposto. Já que a metafísica é inevitável, façamo-na com ousadia sem atinar para as demandas da parcimônia a não ser quando ela conduz os conteúdos de nossos trajetos especulativos.

Ontologia da dúvida
Recentemente, no encontro do GT Ceticismo na ANPOF, apresentei uma leitura perversa, realista, especulativa e metafísica de alguns argumentos céticos tradicionais. Tratava-se de formular uma ontologia da dúvida. A ideia inicial era defender que o mundo ele mesmo continha uma diaphonia, uma polêmica, uma ausência de determinação. Na sua primeira formulação, a ontologia da dúvida substitui fatos determinados por dúvidas proliferadas: o mundo ele mesmo contém estados de dúvida. Trata-se de uma resposta negativa à segunda pergunta que Wittgenstein faz no trecho da seção 352 das Investigações , após considerar o tema de se 7777 aparece na expansão de π:
“Deus vê – mas nós não sabemos”. Mas o que isso quer dizer? Nós usamos uma imagem: a imagem de uma série visível em que uma pessoa vê o todo e a outra não. A lei do terceiro excluído diz aqui: deve ser deste jeito ou daquele. Então ela realmente não diz nada […] mas nos dá uma imagem. E o problema deve ser agora: a realidade está de acordo com esta imagem ou não?
A resposta negativa aponta para uma realidade em que não há determinações. Ou seja, não há fatos que sustentem com conteúdos a atitude de asseverá-los (de afirma-los, de suportá-los etc.), mas antes a relação do mundo com os conteúdos é de hesitação, de dúvida, de polêmica. Não há determinação factual no mundo, o mundo é ele mesmo composto de dúvidas. Não há fazedores de verdade para asseverações, mas apenas fazedores de verdade para expressões de dúvida. A manobra é a de realocar os argumentos céticos na ontologia: ao invés de argumentos sobre o nosso acesso ao mundo, apresenta-los como argumentos sobre as determinações no mundo. Ao invés de fatos que servem de tribunal para nossas hesitações, dúvidas que podem torná-las verdadeiras. A ontologia da dúvida apresenta uma imagem do mundo que inclui as aparências enganosas das coisas, a indeterminação (ou plasticidade) dos constituintes do mundo, a polifonia dos relatos, a incerteza acerca de qualquer medida e outros ingredientes que os céticos alojaram nas nossas tentativas de ter acesso ao mundo. A imagem que aparece é a de um mundo que abriga a hesitação, a pluralidade, o dissenso já que faz a verdade de dúvidas e não de afirmações. Nossas dúvidas revelam algo sobre o mundo, mais do que nossas certezas e o ceticismo se torna uma variante de uma realismo: o realismo sobre a articulação das dúvidas.

O primeiro procedimento – aquele de insistir que as dúvidas não se sustentam sozinhas – aparece também como uma crítica a esta primeira formulação da ontologia da dúvida. O mundo não pode ser composto apenas por dúvidas porque as dúvidas precisam de convicções que as sustentem – no mundo também atitudes de dúvida só seriam compatíveis com alguma determinação. Uma segunda formulação da ontologia da dúvida portanto deixaria espaço, no mundo, tanto para dúvidas quanto para determinações. A dinâmica das dúvidas, transformada em ontologia, requereria um conjunto de proposições-dobradiças (hinge propositions); ou, talvez melhor dizendo, estados-de-coisa-dobradiças. Para usar a terminologia do Da Certeza, as dobradiças permitem que o movimento da dúvida tenha lugar. Estas dobradiças não precisariam ser elas mesmas fixas ou determinadas, elas poderiam ser mutáveis servindo apenas de ponto de apoio para o movimento da hesitação. Similarmente, para que uma convicção seja mantida (held fast), algum outro conteúdo deve ser alvo de dúvida – para que um fato seja preservado, alguma indeterminação deve ter lugar. Para que “o mar é salgado” seja mantida como determinação, o sabor de cada um de seus componentes não pode ser determinado. Determinações dependem de anomalias. Em suma, de acordo com esta segunda formulação, o tabuleiro de dúvidas e convicções da parte de nossas atitudes reflete um tabuleiro de indeterminações e dobradiças da parte do mundo. O tabuleiro está no mundo: não é nem que dúvidas e convicções estejam em nossas cabeças nem que fatos estejam no mundo e dúvidas em nosso limitado acesso a eles. O mundo ele mesmo é composto por fazedores de verdade de nossas asseverações e de nossas hesitações. Não é preciso que algo tenha a forma de uma determinação para ter a chancela da realidade.

Uma ontologia da dúvida traz a tona um conjunto de questões sobre a natureza e a epistemologia da atitude de duvidar. É possível duvidar de uma ontologia da dúvida. Trata-se, claro, de um tipo de dogmatismo: as coisas são de tal maneira que há dúvida no mundo – indeterminações e não fatos. A réplica metafísica é a de que a ontologia da dúvida ocupa o mesmo espaço lógico da ontologia do fato – ou das determinações – e, por alguma razão associada à inevitabilidade da metafísica, este espaço tem que estar ocupado. Ou seja, pode-se tentar forçar o cético com respeito a uma ontologia da dúvida a endossar uma ontologia do fato já que trata-se, também aqui, de um tertium non datur. E o cético pode resistir. Pode dizer que não precisa assumir ontologia alguma e que o primeiro procedimento, que forçou o ontólogo da dúvida a admitir determinações em sua ontologia, não o forçaria a nada mais que assumir alguma determinação. O cético então oferece um embate em campo aberto ao ontólogo da dúvida. Nenhum apelo a determinações – inclusive a determinações com respeito à existência de fatos ou de dúvidas (de determinações ou de indeterminações) no mundo – está resguardado do debate (e do combate). E, em campo aberto, nada protege o ontólogo da dúvida em sua suposição de que alguma metafísica tem que ser assumida de uma vez por todas. O cético simplesmente insiste que aceita determinações sempre en passant, sempre de forma transitória – como se apenas procurasse um terreno para pousar seu equipamento de dúvida, suas dobradiças são sempre provisórias. Se o ontólogo da dúvida apresenta o tabuleiro – com dúvidas e dobradiças móveis – e afirma que o tabuleiro está no mundo, o cético apenas atua neste tabuleiro insistindo que é nele que se decide se o tabuleiro está no mundo ou alhures. O cético se recusa a morder a bala da exigência de alguma metafísica e, assim, se move por todos os lados entre dúvidas e determinações – se aceita que dúvidas requerem determinações.

Apesar da possível fragilidade da ontologia da dúvida diante do cético em campo aberto, penso que ela ensina algumas coisas sobre como a atitude cética pode encontrar sua forma mais robusta e, em particular, sobre como entender o lugar da atitude cética em suas variedades. Um elemento interessante introduzido pela ontologia da dúvida é que a dúvida, como elemento do mundo, precisa ser descoberta. Ou seja, ainda que há casos em que a hesitação nos aparece como uma impressão incontornável, a ontologia da dúvida (em ambas as formulações) entende que conhecer o mundo é descobrir suas dúvidas. E assim coincide com uma das obrigações céticas – se para o cético, digamos, os modos de Enesidemo e de Agripa são práticas em direção à recomendável suspensão do juízo (e caminhos para a ataraxia), para o ontólogo da dúvida tais modos são métodos de descoberta. A dúvida precisa ser descoberta. Um ceticismo informado pela ontologia da dúvida pode evitar a postulação de uma determinação acerca de que o mundo é feito de dúvidas e, ao mesmo tempo, endossar a desiderato da descoberta da dúvida. Esta descoberta requer uma atuação no tabuleiro das dúvidas e das dobradiças. Se houver alguma dúvida auto-sustentável, é possível acessá-la sem apelo a determinações, mas se a dúvida precisa de determinações para se sustentar, talvez ela precise, para ser descoberta, de ser vista desde alguma determinação. No tabuleiro, a determinação pode ser uma isca para capturar uma dúvida – é apenas assumindo certos conteúdos como se foram fatos que certas dúvidas se descortinam.

Podemos contemplar um ceticismo informado pela ontologia da dúvida considerando as quatro possibilidades abertas pela proposta de uma tal ontologia:
1. Antirealismo quanto a dúvidas e antirrealismo quanto a fatos (dúvidas e determinações na cabeça): antirrealismo (clássico).
2. Realismo quanto a fatos (determinação no mundo, dúvida na cabeça).
3. Realismo quanto a dúvidas (dúvidas no mundo, determinação na cabeça): ontologia da dúvida da primeira formulação.
4. Realismo quanto a dúvidas e realismo quanto a fatos (dúvidas e determinações no mundo): ontologia da dúvida da segunda formulação.
A atitude cética não pode endossar nenhuma das quatro possibilidades. Em particular, ainda que endosse a ideia de que dúvidas são descobertas, ela não endossa a determinação que envolve realismo ou antirrealismo quanto a dúvidas. A atitude cética suspende o juízo sobre o estatuto ontológico da dúvida – como faz com o estatuto ontológico dos fatos. Nem está fora de discussão (e do escopo da dúvida) que a dúvida está no mundo nem que não está. A atitude cética transborda para o debate entre as quatro posições acima e se estende para a natureza mesma do tabuleiro que agrega determinações e dúvidas. O ceticismo não é nem antirrealismo quanto a fatos e nem realismo quanto a dúvidas – ele é disputa em campo aberto.

Neopirronismo revisitado e a proliferação das dúvidas
Estas observações levam a uma certa maneira de entender a atitude pirrônica e, em particular, da atitude recomendada por Porchat em seu neopirronismo. A atitude neopirrônica é uma atitude de rejeição das convicções – e, portanto, das determinações inclusive daquelas que dizem respeito a existência de dúvidas no mundo. Trata-se de exorcizar convicções de maneira sistemática e deixar com que dúvidas se proliferem. Não há a busca da convicção, a busca pela crença fixa, nem sequer de que as coisas sejam tais que tudo é feito de dúvida. A atitude neopirrônica é crucialmente distinta da atitude de quem procura maximizar convicções – é uma atitude de maximizar dúvidas. Nisso ela coincide com a sanha de conhecimento no cenário de uma ontologia da dúvida: ter acesso ao mundo é ter acesso às suas dúvidas. O neopirrônico toma a dúvida não como ingrediente do mundo, mas tem a mesma atitude – busca a dúvida, tenta encontrá-la por toda parte e não mede esforços para suspender o juízo acerca de qualquer conteúdo de convicção. A atitude neopirrônica não é aquela de quem se conforma em não saber, mas aquela de quem busca a tranquilidade na inquietude das convicções. O não-saber neopirrônico não é um não saber determinado, mas antes é aquele que se arrisca por meio das convicções para descobrir mais dúvidas. É neste sentido que ele se apresenta ao combate em campo aberto: dúvidas podem ser dirimidas a qualquer momento, mas ele atua esperando que elas se propaguem indefinidamente. O neopirrônico não é indiferente a convicções e dúvidas, ele prefere as últimas e combate para prolifera-las.

Eis aqui como o neopirrônico pode justificar uma posição urbana – na tipologia de Michael Frede. Trata-se de exorcizar convicções. Para isso, é preciso eventualmente pousar em determinações se dúvidas requerem dobradiças. Ou seja, se é preciso se apoiar em determinações para o exercício da dúvida, algumas convicções devem ser provisoriamente adotadas. Trata-se, aqui, de convicções transitórias – ou, antes, de conteúdos meramente aceitos ou tomados como fenômenos, como aquilo que aparece naquele momento. Tais conteúdos, objetos de aceitação, estão a serviço da descoberta de dúvidas. Os fenômenos não são aceitos por razões associadas (apenas) a circunstância do neopirrônico – seu ambiente, sua corporeidade etc –, mas são aceitos porque eles propiciam incursões de dúvida. Eles propiciam um espaço de determinação onde pousar a metralhadora giratória da dúvida cética. Não é que ela defende nada, mas a cada momento ela protege uma parte para poder mirar em outra. Aceitar alguns conteúdos é como usar uma escada provisória para atingir alguma altura – depois da altura atingida (das dúvidas descobertas), a escada pode ser deixada de lado ainda que outras escadas sejam necessárias para continuar o percurso. O fenômeno talvez seja nada mais do que aquilo que se necessita para fazer mover a dinâmica da dúvida – nada mais do que uma dobradiça contextual e transitória. Assim, a aceitação neopirrônica da visão comum – e ainda mais de partes da ciência e da filosofia – não é mais do que a aceitação de uma mola propulsora para galgar mais dúvidas. O neopirrônico encontra subsídios nas convicções dos outros para proliferar mais dúvidas – não é que ele precise da visão comum para alguma coisa qualquer que não seja a busca de um apoio para mais suspensão de juízo. A aceitação – temporária – de alguma determinação não é mais do que uma estratégia no esforço de maximizar a suspensão do juízo.

Em minhas discussões com Porchat, muitas vezes apresentei-lhe o argumento de que a adoção, por parte do neopirrônico, de uma visão comum do mundo (em detrimento de outra qualquer), ainda que somente sob a forma de uma aceitação, era arbitrária. Aquelas determinações eram escolhidas para estarem fora do foco da suspensão do juízo sem nenhum bom critério. Tenho a impressão, que pode estar errada, de que ele me respondia por vezes, entre outras coisas, que não importava esta arbitrariedade. Penso que se entendermos aquilo que o neopirrônico aceita como estratégico para seu exercício de maximizar dúvidas, entendemos porque em certo sentido a arbitrariedade de fato não importa. A escolha da visão comum do mundo para a adoção do neopirrônico é contingente. Outra visão poderia ter sido escolhida – e até mesmo nenhuma visão tomada como consistente. É o dogmático – aquele que mantém a atitude proposicional da convicção – que aponta para a necessidade da adoção de um conteúdo. O neopirrônico, em contraste, se movimenta em ambientes de contingências doxásticas. A visão comum do mundo não é, para o neopirrônico, apenas flexível e histórica, ela é adotada contingentemente, nada a baliza, nos a aceitamos de modo fático. Se quisermos apresentar o movimento cético em passos, começamos aceitando que o exercício da dúvida depende, a cada momento, de uma base de determinação. Logo em seguida, precisamos adotar alguma determinação – uma determinação qualquer – para que o exercício da dúvida saia do chão. O neopirrônico pode adotar qualquer crença – podemos aqui incluir a visão comum do mundo, mas também as teorias científicas comumente aceitas já que qualquer pista de decolagem nos permite proliferar dúvidas. Diferentes teorias científicas, portanto, seriam bem-vindas pistas de decolagem já que elas permitem que acessemos (ou descobrimos) outras dúvidas. O juízo acerca de se há mais dúvidas lá fora para serem descortinadas ou se elas são construídas pelo exercício cético (pela aplicação dos modos de Enesidemo, de Agripa etc). O neopirrônico se apoia em determinações para alcançar dúvidas. É o dogmático – acerca da visão comum do mundo – que considera que ela é carrega necessidade epistêmica. O fenômeno que aparece ao cético – fático, flexível, mas associado às circunstâncias do neopirrônico – é um fenômeno qualquer.

O neopirrônico faz, de acordo com a maneira de entender sua posição que proponho, uma espécie de avesso de uma ontologia especulativa. Mesmo de uma ontologia da dúvida. Ao invés de tomar uma visão comum (ou uma ciência compartilhada, ou um conjunto dissonante de filosofias) como um pista de decolagem para uma imagem determinada do mundo, o neopirrônico a utiliza para proliferar a dúvida. Não é a existência da dúvida que é projetada mundo a fora – como faz a ontologia da dúvida – mas, antes, é a própria dúvida que é proliferada pelo neopirrônico. Também a ontologia da dúvida assente a uma determinação como se ela tivesse necessidade epistêmica: a saber, que há dúvidas no mundo. O neopirrônico tal como eu o construo, por outro lado, não se compromete com nada – sua adoção de algumas determinações é contingente, fática. Ele não está interessado em descobrir determinações a partir das dúvidas (nem sequer acerca da existência de dúvidas, como na empreitada de encontrar uma substância duvidante ou de construir uma ontologia da dúvida). Ele quer alcançar dúvidas a partir das determinações. A visão comum surge então não mais do que como uma pista de decolagem para um voo neopirrônico que, como os voos especulativos, procura atingir alturas. Porém são alturas de juízos suspensos.

An old paper on infinitism and Eduardo Barrio's reply

Some five odd years ago I wrote a bit on infinitism comparing infinite sequences of justifications and truth conditions. The paper never got published but Eduardo Barrio
presented a response to it in Spanish to the Curitiba meeting of the Brazilian group
of Pyrrhonists. I publish both texts below while wondering what to do with all this discussion.

Recursive infinite sequences of justifications

Infinite sequences of justifications have been often dismissed as a somehow inadequate way to justify beliefs. That justification requires infinite sequences can be argued by an argument (A) around the following lines (from 1-3 to 4):
1. A belief is justified only if a justified belief is a reason for it.
2. There are justified beliefs.
3. The proper ancestral of the reason-relation is irreflexive.
4. There is an infinite sequence of justified beliefs each of which is a reason for its predecessor.

Attempts to resist 4 (and A) are motivated by taking 4 to be unacceptable. The argument can then be countered by some variety of foundationalism (rejecting 1), some variety of coherentism (rejecting 3) or some variety of skepticism (rejecting 2).
Recently, infinitism has been regarded with more sympathetic eyes (Black 1996, 2003, Klein 1999). The groudwork done has been to establish that the most frequent criticisms to the idea of a justifying infinite sequence are not compelling. (For example, that we cannot have an infinite number of justified beliefs or that our mind cannot hold infinite sequences.) This shows that the dismissal of infinte sequences of justifications has been too hasty. To complete the job of rendering 4 more plausible, one may think that an example of a reasonable infinite sequence of justification has to be given. Now, here is an example of what I take to be an acceptable (and in fact of the sort that is quite present in everyday life) infinite sequence of justifications:

S justifiedly believes that 'x is red' because
S justifiedly believes she knows what is red because
S justifiedly believes she knows what is a knower of what is red because
S justifiedly believes she knows what is a knower of a knower of what is red because

Or, put in other, perhaps more easily readable way:
S justifiedly believes that ‘x is red’ because
S is a good (or reliable) detector of red because
S is a good (or reliable) detector of good (or reliable) detectors of red because
S is a good (or reliable) detector of good (or reliable) detectors of good (or reliable) detectors of red because

Each belief in the sequence is justified by the next one and yet every belief has to be present if S is to justifiedly belief that 'x is red'. The justiifcation of one belief requires the justification of all the beliefs in the sequence. If this is an infinite sequence of justifications invoked to justify an observational report, it is reasonable to consider that we make use of infinite sequences of justifications all the time and that most cases of justification seem to invoke an infinite regress. Infinite sequences of justifications seem to be not only possible but abundant.

Infinite sequences of justfications like the one above bear resemblence to what is a commonplace about truth that, in its turn, is related to what Tarski called the material condition for adequacy in a theory of truth. It is generally accepted that
if x is red then
‘x is red’ is true then
‘ ‘x is red’ is true’ is true then
‘ ‘ ‘x is red’ is true’ is true’ is true then

Maintaining something (or that something is true) is often thought of as maintaining an infinite set of claims. The example of infinite sequence of justifications above can be visible from this commonplace about truth if we consider what we do when we establish that something is true. We can say that we do so by establishing the truth of an infinite sequence of claims:
S establishes that x is red by
establishing that ‘x is red’ is true and she does that by
establishing that ‘ ‘x is red’ is true’ is true and she does that by
establishing that ‘ ‘ ‘x is red’ is true’ is true’ is true and she does that by

My example of an infinite sequence of justification is readily presentable in a recursive manner. I believe this is a good candidate necessary condition for adequate infinite sequences. Klein (1999) considers the objection that there is an element of arbitrariness in infinitism as we can place any claim in an infinite chain of justifications. This can be countered by saying that not all infinite sequences of justifications are adequate justifiers of a claim: only some infinite sequences of justifications are adequate. The next step is to present necessary and sufficient conditions for a sequence to be adequate––which is, of course, an enormous and maybe impossible task that would possibly amount to solve all the epistemological questions at once. I cannot present those conditions but I conjecture that a good necessary condition for an infinite sequence to be adequate is that it is presentable in a recursive manner.

Recursively expressible infinite sequences of justifications can help us deal with the old objection to infinitism that takes infinite sequences to be infinite deferral of justification––justification is never presented, it is at best only promised. Presented in a recursive manner, the infinite sequence is all there and the justification is in front of is––infinte sequences of justification are understood as actual (Cantorian) infinute sequences. The justification is therefore provided by the sequence and nothing is left for an infinite deferral; if all the justification needed for the belief is in the infinite sequence, one needs no infinite time or infinite number of steps to attain it. Justification, then, is not provisional but rather complete. Recursive infinite sequences, however, can be more complex and can involve more interesting (or at least more surprising) justifications but they have the advantage of coming to view in a finite number of steps; namely the recursive clauses.

A Note on Internalism

It is reasonable to assume that we often infer using infinite sets of premises. Consider what is often said about the famous (Carrollian) infinite regress requirement for Modus Ponens:

Premise 1. If p then q
Premise 2. p
Premise 3. If 1 and 2 then q
Premise 4. If 1, 2 and 3 then q
Conclusion. q

We claim that a conclusion can be drawn from 1 and 2 only because we take the meaning of the words––the connectives––to be established somewhere else and not in the argument. The meaning of these words constitute what makes the rule of inference an effective constraint on what we think. This constraint is often thought as coming from an external source––external to the argument itself. This appeal to external sources for constraint is made unnecessary if we assume that Modus Ponens, for instance, involves an infinite set of premises.

The analogy I want to draw is that, in a similar way, an infinite sequence of justification provide all justification needed and could satisfy all the (epistemological) internalist cravings for reasons. There is nothing that is not available to the thinker in the process of justification––everything is there, open to her view and scrutinizable in the form of the recursive clauses. There is, therefore, no need to appeal to true reliable reporters, to truths about the world or to whatever could capture the (epistemological) externalist fancy. Infinitism can prove to be the ultimate resource for an internalist epistemology.

Black, O. (1996) Infinite Regress Arguments and Infinite Regresses, Acta Analytica 16/17, 95-124.
Black, O. (2003) Infinite Regresses, Infinite Beliefs, Proceedings of the 26th International Wittgenstein Symposium, 40-41.
Klein, P. (1999) Human Knowledge and the Infinite Regress of Reasons, Philosophical Perspectives, 13, 297-325.

Escepticismo y cadenas infinitas de justificación
Eduardo Alejandro Barrio
Universidad de Buenos Aires
Conicet – Gaf

En este trabajo intento mostrar algunos problemas con la tesis según la cual hay secuencias recursivas infinitas de justificación. Tal tesis ha sido defendida por Hilan Bensusan como una respuesta a la cuestión de encontrar plausibilidad a la tesis según la cual hay casos en los cuales podemos obtener justificación para una de nuestras creencias por medio de una cadena infinita de rezones. Por ejemplo, de acuerdo con el enfoque defendido por Hilan, es razonable decir que cuando determinamos que estamos justificados en creer que algo es rojo, también determinamos ciertas propiedades, recursivamente expresables, de un conjunto infinito de otras creencias. Señalare que la analogía que Hilan propone entre la caracterización tarskiana del predicado veritativo y la generación de una secuencia ascendente de aplicaciones de predicados y la caracterización de secuencias recursivas infinitas de justificación no nos ayuda a bloquear la desconfianza escéptica según la cual todo regreso nos deja sin justificación.

Los inferencialistas en epistemologia sostienen que la justificación de las creencias de un sujeto está siempre asociada a un proceso inferencial. En este sentido, una creencia estaría justificada sólo si otra creencia justificada fuera una razón para ella. Es decir, el inferencialista justificacional defiende la idea de que es una condición para que una creencia esté justificada, el que esta pueda ser inferida de otra creencia en las mismas condiciones.

Ahora bien, el defensor de esta perspectiva, parece tener a mano dos opciones: o bien aceptar que hay cadenas circulares de justificación o bien aceptar que hay cadenas infinitas. La aceptación de la primera alternativa, rechazando la segunda al mismo tiempo, produce algún tipo de coherentismo justificacional. La no aceptación de ninguna de las dos opciones o bien obliga a buscar algún tipo de justificación no inferencial de corte fundacionista o bien conduce a alguna forma de escepticismo.

Es bien conocido que tanto la tesis de que hay cadenas circulares de justificación como la de que hay cadenas infinitas ha recibido numerosas críticas. Usualmente, se dice que la primera opción comete algún tipo de petición de principio, mientras que la segunda algún tipo de regreso. Al adoptar una perspectiva según la cual, hay procesos circulares de justificación, quizás porque ese conjunto de creencias tenga algún tipo de propiedad especial (por ejemplo, coherencia global) estaríamos evitando comprometernos con una secuencia infinita de pasos justificacionales. Pero, simultaneamente estaríamos cayendo en algún tipo de presuposición inapropiada: lo que queremos justificar, estaría presuponíendose como justificado desde el inicio del proceso. Al adoptar una perspectiva infininitista, en cambio, estaríamos evitando ese problema. Al mismo tiempo, a favor de ese enfoque y a diferencia del fundacionismo, no tendríamos necesidad de comprometernos con hay razones últimas. Toda razón para una creencia necesita de otra razón. Sin embargo, el infinitista ha recibido numerosas críticas. Entre ellas, se destaca la dificultad de explicar cómo un sujeto epistémico con capacidades finitas podría expresar una secuencia justificatoria infinita. Esto es, un punto importante que este enfoque debe resolver es el de ofrecer una explicación de cómo con recursos limitados hay disponible al sujeto una cadena infinita que le permite disponer de algún tipo de justificación.


En “Recursive Infinite Sequences of Justifications”, Hilan Bensusan se ha tomado en serio el infinitismo acerca de la justificación. Según Hilan, es posible que la estructura de una cadena de razones sea infinita y sin repeticiones. Según su perspectiva, el infinitismo puede brindar una aceptable explicación de las creencias justificadas, a condición de que tomemos algunos recaudos sobre la estructura de ese tipo de cadenas. Claro que Hilan no está solo en esta empresa: lo acompañan Peter Klein y Oliver Black y por supuesto, el número de críticos de esta posición es bastante elevado. Sólo para mencionar a algunos, el trabajo de John Williams y Robert Audi.

Si el fundacionismo es inaceptable porque nos pide que aceptemos una razón arbitraria como base de nuestras creencias justificadas y el coherentismo tambíen lo es porque al dar una razón para una creencia esperamos que esa creencia no participe en el proceso por el cual brindamos una justificación a esa razón, el infinitismo se convierte en un rival a vencer por parte del escepticismo. Quizás por eso, en esta dirección, Sexto Empírico sostiene:

El modo de razonar basado en un regreso al infinito es ese a través del cual afirmamos que lo aducido como prueba de un asunto propuesto necesita siempre de una prueba adicional, y esto, una y otra vez, hasta el infinito. Por eso, la consecuencia de este proceso es la suspención ya que no poseemos un punto de partida para nuestro argumento. (Sexto Empírico Outlines of Pyrrhonism, PH 1, 166).

Claro está que, como ya hemos dicho, y más allá de la crítica de Sexto Empírico, uno de los aspectos que el infinitismo tiene que atender es el de la existencia de algún tipo de vínculo epistémico entre el sujeto y la razón de su creencia de manera tal de que esta última esté en algún sentido disponible al mismo. Es la disponibilidad al S poseedor de la creencia de la cadena infinita de justificación lo que haría plausible esta explicación. A grandes rasgos, podríamos decir que el que una razón esté disponible a S quiere decir que un sujeto racional suficientemente informado aceptaría r como una razón para la creencia. Es bien conocido que en este punto, la posición defendida por Hilan, tal como el mismo concede, ha sido criticada: si la mente humana es finita, no puede ser el caso que una creencia esté justificada sólo si hay una cadena infinita de razones. En esta dirección, Williams, por ejemplo, sostiene:

El regreso propuesto de la justificación de las creencias de S requeriría que S tenga un vínculo epistémico con un número infinito de creencias. Esto es psicológicamente, si no lógicamente, imposible. Si un ser humano puede creer un número infinito de cosas, no hay ninguna razón como para que no pueda saber un número infinito de ellas. Pero ambas posibilidades contradicen el sentido común de que la mente humana es finita. Sólo Dios podría contemplar un número infinito de creencias. Pero, seguramente Dios no es el único que tiene creencias justificadas.

Hilan podría replicar que si bien es imposible que seres con mentes finitas puedan concientemente creer un número infinito de proposiciones, un tratamiento un poco más sofisticado del problema, recurriría a un concepto disposicional de creencia. Si el infinitismo requiere que haya un conjunto infinito de proposiciones que esté disponible al S, entonces S debe tener al disposición a formar una creencia para cada uno de los miembros de ese conjunto. No se requeriría que de hecho tenga formada todas y cada una de esas creencias.

No obstante, y aún recurriendo al concepto de disposición, la disponibilidad puede ponerse en dudas. En primer lugar, a menos que las razones no sean expresables lingüisticamente, no habría vocabulario suficiente como para expresar cada una de las razones dentro de la cadena infinita. Y aún cuando sólo se requiera que el S tenga la disposición a aceptar rn como una razón potencial, si no tuviéramos vocabulario suficiente como para expresar esa razón, parece poner en duda la posibilidad de que la misma sea accesible a S. Sólo si aceptáramos razones inexpresables potencialmente disponibles al S, tal explicación tendría plausibillidad. En segundo lugar, y tal como el propio Hilan acepta, el hecho de que sea posible construir una cadena infinita no es suficiente para que esa cadena sea de hecho una cadena justificatoria del último eslabón de la cadena. En esta dirección, hay una reductio, que ha sido ofrecida por Post que muestra que para cualquier creencia contingente, es posible construir una cadena infinita. Para ver el punto, sea p el contenido de creencia contingente y usemos el Modus Ponens de la siguiente manera:

…, r & (r → (q & (q→ p), p))), q & (q→ p), p

Pero, entonces, toda creencia contingente estaría justificada, lo cual haría que el infinitismo sea inaceptable. Claro que la mera existencia de una cadena infinita de justificación no tiene por que ser una condición suficiente. No toda cadena que tenga una propiedad estructural hace a una creencia empírica justificada.


Hilan propone que la estructura se las secuencias infinitas de justificación sea expresable de manera recursiva y presenta una analogía con el modo en el que Tarski presenta el núcleo del funcionamiento del predicado veritativo. De esta manera, secuencias infinitas recuersivamente expresables de justificación estarían disponibles a los sujetos en un sentido análogo al que les está disponible una secuencia infinita de ascensos semánticos. En sus propias palabras “lo que es un lugar común acerca de la verdad y está vinculado con lo que Tarski llama condición de adecuación en la teoría de la verdad. “ Claro está que en el caso del predicado veritativo, la secuencia estaría integrada por eslabones del tipo
Si x es rojo entonces ‘x es rojo’ es verdadera, entonces ‘ ‘x es rojo’ es verdadera’ es verdadera, entonces ‘ ‘ ‘x es rojo’ es verdadera’ es verdadera’ es verdadera, entonces…

Todos estructurados de un eslabón a otro, ascendiendo semánticamente. Sostener que algo es verdadero es pensado como sostener una secuencia infinita de este tipo. Y de acuerdo a Hilan, no se necesitaría ningún tiempo infinito ni ningún vocabulario especial para recorrer o expresar toda la secuencia. “La justificación no sería providencial sino completa”.
En el caso de la justificación, la secuencia estaría integrada por eslabones del siguiente tipo:
Si una persona S tiene una justificación para x, entonces hay alguna razón r1, disponible en principio a S, para aceptar x; y hay alguna razón r2 disponible a S para aceptar r1. y hay alguna razón r3 disponible a S para aceptar r2. etc.
Pero, nótese que para que la analogía funcione, tal como sucede con el caso del predicado veritativo, haría falta suponer que las propiedades epistémicas involucradas en este último caso son aritmetizables. Sólo así tendríamos la seguridad necesaria para que la secuencia generara sea recursiva Aún así, adviértase que ninguna prueba, ninguna secuencia o cadena epistémica, es de hecho infinita. Lo que garantiza que exista un procedimiento recursivo es que para cualquier elemento de un conjunto infinito se pueda determinar si ese elemento cumple una determinada condición. Esto es, por ejemplo, si tal como se muestra a través de la numeración de Gödel, la demostrabilidad dentro de la aritmética es aritmetizable, esto quiere decir que para cualquiera de las infinitas formulas de la aritmética existe un procedimiento aritmético que permite determinar si se trata de una fórmula demostrable. Esto, sin embargo, no quiere decir que la estructura recursiva garantiza la existencia de pruebas de longitud infinita. Por el contrario, los límites de la axiomatización están dados por la suposición de que la longitud de las pruebas de una fórmula demostrable es siempre finito, aún cuando exista un procedimiento recursivo para determinar para cualquiera de los infinitos casos de aplicación del predicado “x es demostrable”, si se trata o no de una fórmula que cumple esa condición.
Más aún, el teorema de Tarski muestra una limitación para expresar el predicado veritativo dentro de un mismo lenguaje: ninguna teoría suficientemente fuerte como para expresar la aritmética, esto es, ninguna teoría cuyas fórmulas tengan propiedades recursivamente expresables, puede contener su propio predicado veritativo. En este sentido, el resultado de Tarski implica una limitación sobre el alcance de la idea de autorepresentación. No todas las propiedades de una lenguaje se pueden aritmetizar, esto es, se pueden expresar recursivamente. En particular, no hay manera de representar recursivamente dentro del lenguaje natural, el predicado veritativo de ese lenguaje. Casos como los del mentiroso o sus afines muestran que hay restricciones en las cadenas semánticas generadas a través de los ascensos correspondientes a través de la condición de adecuación de Tarski. Interesante es el caso del ascenso generado por la oración
Esta oración es verdadera. Entonces “Esta oración es verdadera” es verdadera. Entonces, ““Esta oración es verdadera” es verdadera” es verdadera. Entonces, etc…
Si la primera es verdadera, cada uno de los eslabones es verdadero. Pero si es falsa, cada uno de los eslabones será falso. Esto es, cada uno de los eslabones de la cadena podría ser parte de la extensión del predicado veritativo del lenguaje, si su predecesor ha formado parte de la extensión. Claro, esto genera una secuencia infinita de aplicaciones del predicado veritativo. Pero, esta secuencia cuya estructura queda enumerada recursivamente, nada explica respecto de las condiciones veritativas de la oración inicial. Por eso, aun en el caso de que llegaramos a aceptar con Hilan la generación de secuencias epistémicas con similares características, ¿por qué ellas brindarían al sujeto algún tipo de garantía epistémica sobre su creencia?
La propuesta de Hilan según la cual existen secuencias recursivas infinitas de justificaciones parece tener problemas. No es claro que las propiedades epistémicas sean aritmetizables. Y esta es una condición para que las condiciones de ese tipo puedan estructurarse recursivamente. Aún cuando se pudiera garantizar tal cosa, lo que la recursividad posibilita es la aplicación de una característica a infinitos objetos (en este caso, creencias), pero no que la longitud de la secuencia de justificaciones sea infinita. Una prueba de longitud infinita no es parte de lo que es recursivo. Por otra parte, la analogía con la caracterización tarskiana de verdad no ayuda demasiado. El teorema de Tarski muestra que los conceptos semánticos no son completamente aritmétizables. Esto es, no son completamente caracterizables por técnicas recursivas. Particularmente, en un lenguaje que permita la autoreferencia, donde es posible expresar autoaplicaciones de conceptos semánticos, la aplicación de la condición de adecuación tarskiana conduce a contradición. Por último, aunque se generara una secuencia de longitud infinita, sin generar contradición, tal como sucede con la afirmación “Esta oración es verdadera”, no queda claro en qué sentido esa cadena ofrecería al sujeto algún tipo de garantía epistémica sobre lo que se cree.

Monday, 12 November 2012

Urges and virtuality

The notion of attractor - associated to vector fields - is a central element of the way Delanda presents Deleuze's virtuality. In fact, it is a central element to be considered in a process ontology: actual entities create fields of attraction that would have something to do, I believe, with what Whitehead calls nexus. Attractors guide trajectories - this guiding would entail some sort of instantiation only if the rest of the world is subtracted. Virtuality has to do with the rest of the world - just like any synthetic a posteriori judgment can be taken to be virtual because it depends on the rest of the world. The pattern, of course, is familiar to process philosophers: to bring to ontology what is presented as an epistemological feature. (Contrast this
with some kinds of anti-realist move that would do the opposite movement.) Here, modeling can fail to be implemented because it points at the virtual - at attractors - and therefore predictions can fail. Other parts of the world can interfere. Virtuality is about attraction. What I have once called (in collaborative works both with Manuel and Tomas) a ceteris paribus device - roughly the lie brought up by a law of physics concocted in a lab - is virtual. Models, in general, can be thought as attractors and not as mental projections.

I was thinking of urges in terms of constitutive elements of the world. Urges are pressure points - the subsistence of anything (a community, a society of molecules, a population or a middle-sized object) depends on variables in the ecology of urges. An urge can be resisted and can be countered, but such maneuver involve deviating from an attractor. Sometimes an urge is created (producing economic catastrophes like shocks in the sense of Naomi Klein would be an example) so that other urges subsume or get diluted. The idea here is that urges can fight with other urges - urges claim their hold on trajectories, on what occurs. To change the world is not to change trajectories, but to change attractors and if it is so, it is to act on urges. Make something become urgent. Surely, the effect of urges would depend on the rest of the world - the other urges. Urges are virtual. They shape things by acting on the space of attractors. But can an ontology of urges be lean enough to claim that all attractors are, somehow and at the end of the day, urges?

Sunday, 28 October 2012

Flat ontology and plot metaphysics

Thrilling conversation with André Arnaut on the way back from Curitiba. We were considering Viveiros de Castro's claim that ontographies are either antropocentric or antropomorphic. There is, according to the claim, a tertium non datur connecting anthropocentrism and anthropomorphism. To be sure, Brassier's alternative (as well as Meillassoux's factualism) is not easily accommodated in either horn of the dilemma. No matter whether the dilemma holds, it is interesting to consider the paths from one pole to the other and vice-versa. It is the road between correlationism and the metaphysics of subjectivity and back – for instance, between an enchanted nature and a world stripped of all secondary qualities and filled with our (anthropocentric) projections or between a transcendental idealism and a Whiteheadian view where there is no centrality of one form of (anthropomorphic) actual entity. The anthropomorphic pole – favored by Viveiros – sees some element of humanity everywhere and, in this sense, Meillassoux seems quite right in taking the metaphysics of subjectivity as an ontology where to be is to be a correlate. As general formulae, both sides of Viveiros opposition, though, seem to be portraying a metaphysics constituted by a project that can be to some relevant extent presented with no appeal to its instantiation. It doesn't matter how the (human) center is implemented, it doesn't matter how the (human) form is achieved. If we see things this way, we can eventually feel that process philosophy, for example, is entirely about a plan of transcendence, with no appeal to how things contact and contaminate each other in a plan d'immanence.

Tomas Cardoso and me have defined a plot metaphysics (as opposed to a landscape metaphysics) as one where there is room for a plan d'immanence. That is, it cannot be presented as a project because its instantiation makes a crucial difference. If there is no genuine room for a plan d'immanence in process philosophy, it falls short of being a plot metaphysics. I believe there is (at least) two ways of presenting process philosophy: roughly, as an anthropomorphic set of pieces of furniture and as an openness to processes. The latter but not the former is a plot metaphysics. We moved then to flat ontology – as we believe flat ontology is a way to understand, among other things, what is at stake in process philosophy. Again, we agreed that flat ontology is some sort of project that deflates the importance of implementation. It is as if it takes all things to be equal in ontological status so that the domain of one thing over the others becomes is somehow played down. Things are not up for grabs because of their ultimately democratic nature. There is at least an element of the plan of transcendence that is prior and untouched by the nitty-gritty whereabouts of instantiation. The plan somehow supersedes the plane.

André is working on whether the speculative turn is committed to some sort of primacy of the theoretic over the practical. He conjectures that the exorcism of anthropocentrism (and of humanist paradigms in general) stimulated philosophers to be less careful to the relevance of the practice. One way to put it is to say that speculation paved the way to a renewed and revamped primacy of the theoretic by stimulating varieties of landscape metaphysics. Flat ontology, under these lights, is a (democratic) general theory with no attention to the effects of the practical. If we see things this way, an ontology of objects, an ontology of alliances and even perhaps an ontology of matter stop short of a full-fledged plot metaphysics. The move from flat ontology to plot metaphysics is one where no (ontological) theory can fill the blanks that need to be left for practice, for some non-human notice of practice, a practice of things.

Monday, 22 October 2012

Accelerationismusstreit 1

Thinking about Srnicek's sort of defence of accelerationism against the charge of collaboration with neoliberalism. In a sense, he makes a good counterpoint to Ben Noys suspicions about the politics of an accelerationism ontology. Srnicek claims that neoliberalism is in fact a regime of concentration of powers that protects trusts and corporations against the free flow of liquidity. The discourse of free flow coincides with accelerationism only superficially because the effect of the neoliberal gesture was crucially to open way for more concentration of powers as if to create and protect some feudal territories within the space of capitalism. Srnicek praises relative accelerationism as a political weapon to add liquidity in solidified structures - and therefore as a tool for the political left. He is less committal concerning absolute accelerationism, a position like Nick Land's according to which the flows will dismantle everything including humanity itself. He thinks that while the former form of accelerationism is political and human, the latter is speculative and inhuman. He then goes on to defend a strategic use of relative accelerationism against the financial and corporate feuds.

I take accelerationism as not committed to capital flows and even less to a capitalist machine. This is the only way we can say the Anti-Oedipus is an accelerationist book - which I believe it is (and a provocative one at that). I tend to think that accelerationism - I don't see much point in Srnicek's distinction between the absolute and the relative variety - presses for further acceleration no matter what gets on its way and ultimately capitalism (and capital flows) will find themselves in its way. Srnicek then sheds light to an interesting phenomenon: neoliberalism is an advanced version of capitalism and, as such, it has to create concentrations of power. Noys, for one, holds that capitalism can survive the end of nuclear family and, he hints, even of individuals. I'm not convinced of either, and I fail completely to see how it can carry on without individuals. Capitalism is concentrative. It works by bringing flows somewhere. Accelerationism could be ultimately showing that centrifugal flows are faster and then capitalism will be on its way. Accelerationism (as a typical branch of Marxism) has learned with capitalism, but it is not committed to it. Anything faster will deterritorialize the concentration of power that capitalism harbours (that depends on individuals, as money don't flow from my left hand to my right hand). Yes, neoliberalism is about trusts and corporations but that seems to me another chapter of the history of capitalism - an accelerated history, but a concentrative one.

Friday, 12 October 2012

Markus Gabriel and the modally open horizon

Last week in Porto Alegre I talked to Markus Gabriel about his notion of fields of sense. He defines existence as presence in a field of sense. The inspiration is Fregean (and, although not admittedly, the flavour is Meinongian) but he adds an element of pluralism that could be understood in terms of existential pluralism (in the vein of Souriau´s different modes of existence, for example). A field of sense encompasses domains of discourse or thought, perspectives, modes of presentation such that each work of fiction or each folk ontology have a corresponding field. The notion inherites the ambiguities and the vagueness of the notion of "Sinn" in Frege. But Gabriel holds that senses are somehow de re, they are in the world and in that sense he could side with McDowell interpreting Sinne as a way to determine reference (and not necessarily a method, a criterion, or a description and not even an explicit mode of presentation). Strangely, though, he wants to have no room for denotation. This is interesting from a perspectivist point of view, but I believe it also makes explicit a drawback of some forms of perspectivisms.

If to exist is to be in a field of sense, existing things are attached to a field. They don´t enjoy what I called in this blog the open horizon of life, which is primarily modal. You could think if it in less modal terms if it looks convenient: existing things cannot escape from their field of sense without collapsing. I made my point in the conversation using Kripke´s examples. I think the thrust of Kripke´s criticism of Frege and Russell is not only that descriptions cannot determine denotation, but that a certain element of independence in what is denoted is to be recommeded. As I said in this blog (previous posts about the Millnong project), this independence doesn´t have to be connected to existing among concreta or with an attachment to reference per se. A Meinongian could claim that there are objects independent of whether they belong in an ontology and associate these objects with something other than a description (or a mode of presentation). One could be Meinongian and yet respect the right for an open horizon for each existing thing (and therefore add a Millian element to her Meinonianism - Millnong). In a sense, it is an intuition I find in process philosophy in general, things are not locked in their scopes - they couple, they are taken, they flow. Of course Gabriel could accommodate these transitions by multiplying fields of sense. I find compartimentation has to have a limit. The solution is rather something like Souriau´s surexistence - the crossroads of existences. Different fields of existence bump into each other in the streets. This is what I take to be the crucial Deleuzian legacy in ontology: there are ontological streets - the plan d´immanence. It is not necessarily composed by concreta or of concreta only, but ontologies ought to have streets so that the addresses can be found (and lost).

Sunday, 7 October 2012

Transcendental Surrealism

Last night for no simple reason to explain I remembered transcendental surrealism. I believe it was because somehow I was toying with the interdiction Noli Me Legere in the way Blanchot deals with it in Après le coup. In any case, in my thesis, years ago, I presented briefly what I understand now as an embrionary form of process philosophy I called transcendental surrealism. Surrealism came from Leplin. He claimed that a scientific surrealism was an alternative way to understand the predictive successes of science - alternative to a realism that posits the approximate correspondence truth of scientfic theories in terms of the denotation of its major terms (and an ontology built accordingly). Leplin´s surrealism held that things are as if theories of mature sciences with predictive successes were true. As if, he claimed, was enough. My transcendental surrealism was about other cognitive devices - that I called other inductive biases. I wrote that each bias had an "as if" associated to it - and that was based on my research with machines that perform induction due to their biases. Transcendental surrealism was about the plurality of biases and the plurality of appropriate as ifs. Quite Leibnizian in spirit and I believe, those as if, are very close to waht Whitehead calls subjective forms.

I had a conversation with Eros in Porto Alegre about dispositions versus projections. The background was our common interest in Goodman´s work on counterfactuals. My point could be put in terms of transcendental surrealism: each bias sees things in a way and interact with what they see. The bias for green, instead of grue, makes us act according to a projection. The same for the tick, the bee or the grain of sugar. Now, there is no need to postulate dispositions if we just appeal to the horizon of sight (the horizon of prehension) of a bias. This horizon is enough to enable the bias to be inclined to act one way or another. Other things can intervene, as there is a whole world affecting this seen realm, a whole world that is not perceived clearly - the extensive continuum, in Whitehead. So, whatever happens to the grain of sugar has only to do with the actual relations it holds. If we want to make sense of dispositional predicates, we just need to appeal to what its bias prehends in its corresponding subjective form (it sees apple juice as water, for instance). Is this story appealing to as much internal features as the dispositional story? I believe it is not because it doesn´t have to appeal to dark quarters (say, of the grain of sugar) where it has somehow the rest of the world lined up.

Saturday, 6 October 2012

Factual thought, populated thought

Been to the conference on the ontological turn in Porto Alegre organized by Rodrigo Nunes and his MaterialistS friends. Good discussions with Markus Gabriel, Eduardo Viveiros de Castro, Benjamim Noys, Stephen Shaviro, Eduardo Luft, Caroline Marin and many others. Here is the text of my comment of Shaviro's interesting paper called Uncorrelated Thought.

I believe one of the major attraction of Meillassoux's work lies in its diagnostic power. He makes us see the efforts of philosophy in the last centuries as fitting a general, albeit claustrophobic, pattern. Correlationism, with its two poles of inextricable dependence – thinking and being, has exerted its various charms on the efforts to relate to the world in ways that involve thought. His diagnosis is presented in broad terms: no matter how thought is conceived, no matter how being takes place. His diagnosis provokes many effects. Among them, several ways to deny its bite and several ways to wage heroic battles against the confinement of the correlationist walls. In most of these cases, Meillassoux's gestures is an invitation to speculation. As Shaviro appropriately points out, “philosophy must necessarily take the risk of engaging in some sort of speculation, if it ever hopes to approach the great outdoors. […] it must seek to evade the conditions that are imposed upon thought by the very nature of thought itself” (p. 2). The diagnosis is therefore an invite to the risk involved in speculating. My main aim in this comment is to expand a bit on Shaviro's remarks on thought, in the direction of pointing out that speculation is not only required to approach in a non-correlationist fashion the great outdoors, but also to find alternatives to correlationism in the indoor quarters – within the realms of thought.

Meillassoux's diagnosis gives rise to a geography of alternative routes and positions. We can think of his geography in terms of four poles. One pole is certainly that of correlationism – strong and weak – according to which we cannot reach beyond the correlation of thought and being (weak) and not even think beyond this pale (strong). A second pole is that of a metaphysics that ignores correlationism altogether: a position that hasn't engaged with the correlationist gambit at all and manage to proceed as if the specter of the correlation didn't loom in its shadow. A third position is what we can label the speculative factualism embraced by Meillassoux himself in his Aprés la finitude – a position according to which the correlation points in the direction of the factuality of all things. And the fourth pole, crucial for the articulation of the third, a pole that encompasses several seemingly different positions and the one that I will be primarily concerned here. This is the pole of what he labeled the metaphysics of subjectivity and lies in the roots of what he now calls subjectalism. I sometimes prefer to call this pole the metaphysics of correlation, for it starts out by an effort to focus on the correlation itself – either to make it irredeemably part of the world itself or by multiplying it by finding a plurality of correlations in the world. This fourth pole – that Meillassoux sees as the wrong way out of correlationism – involves most varieties of vitalism, of process philosophy together with varieties of absolute idealism and phenomenology. It lumps together disparate positions such as those embraced by Schelling, Hegel, Bergson, Whitehead, Deleuze and Latour. It is, to be clear, a speculative attempt to escape the trap of correlationism – and as such it responds to the diagnosis in the way Meillassoux recommends. He in fact acknowledged that the metaphysics of subjectivity takes the advisable speculative flight that is nonetheless headed in the wrong direction.

Meillassoux's main problem with the metaphysics of subjectivity has to do with factuality – something that is itself central to his own preferred alternative, that of the third pole of speculative factualism. He claims that the metaphysicians of subjectivity fail to see an important part of the bite of correlationism: that the correlation is not absolute. The metaphysics of subjectivity fails to do justice to the facticity of the correlation. The correlation is a matter of fact, it is neither a structural component of things nor an inescapable dimension of subjectivity. The metaphysics of subjectivity takes the correlation to be either all that there is to be accessed – there is nothing absolute beyond the correlation as things are irreducibly for-us and not in-themselves – or the very dimension of what there is even beyond the correlation we are supposed to be confined to – and therefore there are other correlations beyond ours. The former is the metaphysics of subjectivity of Hegel or, maybe, of some varieties of the linguistic turn. The latter is the one of Whitehead and vitalism, of process philosophy and actor-network ontologies. Meillassoux's complaint concerning all those positions is that they make correlation absolute – either by stating that there is nothing beyond a correlation or by multiplying correlations so that there is nothing that is outside a correlation. Clearly, the sense of 'correlation' here has changed at least slightly: it is not about a confinement we find ourselves within but rather a structure in the world.

Now, let's consider the charge Meillassoux has against this pole. He claims correlations are factual while this pole make them absolute. But why is it important that correlations are factual? I guess it is because it is a matter of fact that something is found within the boundaries of a correlation. It so happened that we find ourselves in a correlation. This could be because we could have been intelligible intellects (as Kant considers in his third Critique) that would see things through as we can think them but it so happens that we're not. But it could also be that we are in this particular correlation, and we could have been in others. We could be in a different correlation altogether. There are two senses of factuality here. But, it seems to me, that the important bit is that we could be outside this correlation. Whether or not what there is beyond this correlation is a bunch of other correlations or something entirely different is up for speculation. Clearly, following Meillassoux's line of argument, to say that there is something beyond this correlation is to engage in a speculative flight. The predicament of correlationism is that we are locked in this correlation.

But here is the blind spot in Meillassoux's diagnosis that Shaviro begins to uncover. Thought, he claims, is not only an intentionality device. Thought is something. He envisages “[a] thought that is nonrelational, or 'autistic'.” It is, for him, a deflationary panpsychism. The uncorrelated thought is already available. I would go further in debunking the assumption that we are confined in a correlation (this correlation) when we think. For what is we? Meillassoux is hostage to an anthropological sleep. Thought is something – and as such it is made of parts, it is coupled with other things and it is, at any moment, a factual assemblage of tools, resources, continuities, granularities and, there we go, (maybe) correlations. Thought is always populated. In fact, it is a population and quite a heterogeneous one. To say that it responds basically to an “us” is already to set the stage for a single, confining correlation. The panpsychist answer could be that thought itself is full of thoughts – full of psyches. If we take the kind of metaphysics of subjectivity that multiplies subjectivity, a debunking reply to correlationism becomes then available: the plurality of subjectivities – the ecology of subjectivities – doesn't stop short of the inner quarters inside this correlation. In other words, the line available to the now-called subjectalist is to proliferate correlations. Then, there is no confinement because there are just too many correlations.

This was my comment. An image of thought that presents it as something other than an intentional unity would itself harm correlationism. Thought is full of foreign elements, of small machines, of roads, of fragments of operation systems, of small animals, of flows. Speculative psychology? Yes, as part of a speculative psychology of the universe, as Nietzsche hailed in the Genealogy. Surely, this is all speculation - just as Meillassoux's factualism. However, there is something else here. There is maybe a pointer to something like an alien phenomenology of the mind. If consciousness is a chorus of voices, we can hear all sorts of haunting voices coming from the great outdoors inside...

Monday, 1 October 2012

A (Quinean) flat ontology of thought

Last week in my metaphysics classes I was on about physical intentionality and virtuality. I tried to explain the contrast in terms of what is internal and what is external - dispositions being internal and virtuality external. Incidentally, or not quise so, connectedness through dispositions leans towards internal relations and the kind of connectedness of things that Schaffer finds indicative of monism. Dispositions place tendencies (or inclinations) inside things - this is why Molnar´s use of Brentano´s notion of intentionality seems so suitable. Dispositions are properties and, only as such, they can be such that water (to which soluble sugar is disposed) could prefigured inside sugar. It does look like mental intentionality. It points towards the world inside - towards the dark quarters of a monad. Surely, dispositions depend on the rest of the world to be triggered, they can find antidotes, they can never be actualised and something can carry finkish ones for ever in its wake. On the other hand, if there is nothing inside things, they act only through external alliances. Nothing internal - things are possible because they are virtual. It is like an occasionalism where everything needs to interfere for sugar to disolve in water.

Harman, in his book on Latour, makes a fruitful comparison between occasionalism and Humean suspicion of necessary connections. Hume´s suspicion concerning what is going to take place tomorrow can be taken as deriving from atomism concerning events. The rise of the sun tomorrow depends on too many things - including everything that sponsors the coming to being of tomorrow. There is an interesting connection between being and thinking here. I cannot think: I don´t know of anything concerning tomorrow, because if I think that, tomorrow itself loses its meaning. One can remark that both thinking and being work in critical masses. It is a Davidsonian remark - and one based on a Davidsonian strategy to overcome correlationism. As such, it is a Quinean remark: all beliefs move in a flat surface, a flat ontology of thought. Everything is outside, nothing is like a meaning locked inside the dark quarters. My belief concerning tomorrow cannot, on its own, find a reference - it is, in this sense, not a thought. The point is not really the critical mass remark, but rather a process ontology of thought. Thought deals in virtualities. If there is nothing inside me about tomorrow (no meaning, no intentional internal object) then my beliefs about tomorrow depend on tomorrow (and what makes tomorrow happen). And some of my beliefs about tomorrow have to stand - otherwise, there is no meaning of tomorrow to which I can recoil.