Total Pageviews

Tuesday, 24 January 2017

Broad anti-necessitarianism as a condition for process philosophy

This is the draft of the text that evolved from my participation in the ABC+ Process Philosophy conference in Liverpool last year:

1. Introduction: Process, Agency and Contingency
Process philosophy is a general name for metaphysical outlooks that focus away from what is stable and take what appears fixed to be what primarily demand explanation. Therefore, the main characters of a process philosophy plot are more likely to be interactions than substances, becomings than the underlying structures of being, constituted movements than constituted things. In the twentieth century, process philosophy was championed by Alfred Whitehead's philosophy of the organism, by Etienne Souriau's claims concerning how existents continuously bring about existence, by Henri Bergson's attention to durations, and by Gilbert Simondon's ideas of meta-stability and of individuals explained away by processes of individuation.1 More recently, process philosophy has been placed in the foreground of several philosophical discussions. It has received increasing political attention in the context of the growing concern about the non-human in ecological debates where Isabelle Stengers has championed the idea of a cosmopolitics influenced by Whitehead's ideas and Bruno Latour has been developing an actant ontology that owes much to Whitehead, Souriau and Simondon.2 It has received analytic treatment, especially due to Johanna Seibt's process mereology of activities and dynamics.3 It has also propelled the so called speculative turn, on the one hand influencing object-oriented philosophy and a renewed interest in the philosophy of nature while on the other hand being taken as a symptom of the age of the correlate that Meillassoux diagnosed and criticized in philosophy since Kant.4

It is difficult consider briefly all variations of process philosophy – and the metaphysical developments they have prompted. There are important tensions between, say, Whitehead's agent-based metaphysics and Simondon's concerns about how agents (among other individuals) become individualized or between Seibt's emphasis on dynamics and Latour's networks formed and maintained by an indefinite and ever-changing number of actants. Process philosophies are possibly held together more by family resemblance than by shared common features. Still some general features of process approaches follow from their general drive against claims that stability is the metaphysically prevalent and underlying feature of everything. Such claims often make novelty either hard to explain or dismissed as a mere appearance. In contrast, process approaches give ontological priority to change or to movement in the form of events or becomings, activities or unrepeatable particulars and can understand novelty either as a feature of whatever exists or as a consequence of another processual feature that is itself susceptible to novelty. In any case, novelty matters for process philosophy for it pictures the universe as something in the making.

An important general feature of process philosophy has to do with necessity and contingency. In order to maintain that reality is ultimately processual, one should not postulate an over-arching single necessity. A necessity of this kind could take the form, for example, of a set of laws, or an ultimate substantive reality, or a ruling supreme monarch. It turns the universe into a “barren tautological absolute” in the expressive phrase of Whitehead.5 If every process, in the form of change, movement or activity, is explained away in terms of an over-arching necessity, the ultimate nature of reality cannot be processual. Process philosophy requires genuine contingency in the world, if it is to take novelty as an ontological feature. In Being Up For Grabs, I argued that contingency follows from the absence of a single, over-arching necessity – if there is more than one necessity, genuine contingency follows suit.6 The relation of opposition between contingency and necessity, I argued, is that the former is the plural of the latter. There I explored how contingency could be understood as being the centerpiece of a metaphysical image while not being the ultimate feature of everything. The contrast is with an image that makes contingency into a over-arching principle independent of anything else. Meillassoux's speculative materialism posits a principle of facticity that transcends everything and is itself necessary – not even the advent of God could change the (necessarily) contingent nature of anything concrete. As a consequence, contingency becomes some sort of barren tautological absolute where no novelty could ever have any consequence.7 In contrast, I propose in the book a process philosophical approach to contingency (tied together to views of contingency based on the insufficiency of reason and on the timing of events). Contingency is presented as central but not as the building block of everything and no determining principle overrides novelty. The image of contingency that emerges from the book is one where agency is crucial. In the present work I contend that agency is not only indispensable to understand contingency but also the shaping force of any genuine process philosophy that avoids collapsing into the postulation of an over-arching necessity.

2. The Process of Reality: Peirce and Anti-necessitarianism
Charles Sanders Peirce endeavored to counter what he calls necessitarianism – the claim that “the state of things existing at any time, together with some immutable laws, completely determine the state of things in every other time”.8 Necessitarianism is understood therefore as the thesis that everything follows by necessity from what was set once and for all given some immutable laws. In other words, necessitarianism is the view that reality was produced (by necessity) in one stroke. According to it, all that there is can be condensed in an arguably finite but certainly close to recursively enumerable set of laws and initial states that is in important senses smaller than the world that, in its turn, depends on it. In the Laplacean formulation of the 19th century that Peirce addresses, “the instantaneous state of a system of particles” is defined by a number that “remains the same at all times” and therefore “the intrinsic complexity of the system is the same at all times”.9 In any form, necessitarianism is the view that there is no genuine diversity, novelty or surprise and Peirce endeavors to oppose it. This is what he responds to his opponent:

[y]ou think all the arbitrary specifications of the universe were introduced in one dose, in the beginning, if there was a beginning, and the variety and complication of nature has always been just as much as it is now. But I, for my part, think that the diversification, the specification, has been continuously taking place.10

He then proceeds to provide reasons to his claim which include the advantages of postulating “pure spontaneity of life as a character of the universe”11 that explains both the irregularities and the uniformity of nature. He believes it can be inferred from “broad and ubiquitous facts” that “there is probably in nature some agencyby which the complexity and diversity of things can be increased”.12 He contends that facing the irregularity and the diversity of the universe, the necessitarian could either deny that they are genuine and hope to show that ultimately everything is regular and explainable in terms of a single over-arching necessity or, failing that, to posit unexplainable irregular elements. The latter resource is what drove the Epicurists to posit clinamina, those swerves in the determined and necessary orbit of any atom so that novelty (and diversity) can take place. Peirce, as many before him including Leibniz in his controversy with Pierre Bayle,13 understands as a drawback of the system the postulation of unexplainable events like a random swerve. In contrast, his hypothesis according to which variety comes from spontaneity can provide enough explanation for both regularities and novelties, for both order and exception.

Peirce's opposition to necessitarianism and his effort to spell out an alternative reflects the kind of atmosphere where the need for process philosophy is felt. Equally, for example, when John Dewey argues that human experience has to be conceived in terms akin to a nature which is wistful and pathetic, turbulent and passionate or Jean Wahl holds that there is always some sort of conflict in the concrete the gates for a process-based approach of reality are open.14 The drive towards process philosophy can arise from the idea that reality doesn't spring from a single plan (and its unexplainable exceptions).15 This is why Peirce's remarks on necessity seem like a good starting point. In fact, necessitarianism can be formulated in a more general way in terms of a word inherited from the Greeks, arché, which are present in words like hierarchy, like archaic, and like archeology and points both to command and to commencement. Peirce formulated necessitarianism in terms of a fixed set of laws and a class of initial conditions. We can understand it in broader terms as the claim that there is only one arché – one command and one commencement – and therefore everything else is the effect of that ultimately fixed structure. Such a single arché entails what can be labelled broad necessitarianism, even if it is taken to be behind activities, becomings, changes, movements or transformation. In fact, even if such an arché is something like Meillassoux's principle of facticity – that makes everything necessarily contingent – it entails necessitarianism if nothing is beyond its scope. Or rather, any single arché entails necessitarianism if nothing is beyond its scope and is not ungoverned; that is, nothing escapes the arché with a reason to do so because if a swerve is not a sheer irregularity it is a product of another command and a new commencement. Peirce pointed out that necessitarianism can accommodate sheer irregularity – through doctrines such as Epicurist's clinamina. In fact, it can further accommodate a world of nothing but sheer irregularities or random moves like Meillassoux's hyper-chaos where everything is (necessarily) contingent because such world has no more than one arché, even though it ensures no more than an an-arché. There is no genuine, irreducible process if everything is under a single, over-arching principle; any true process philosophy requires broad anti-necessitarianism – the claim that broad necessitarianism is false. Indeed I contend that anti-necessitarianism is at least a necessary condition for a bona fide process philosophy.

If this is so, broad anti-necessitarianism sheds light into how process approaches orient themselves in some metaphysical debates. Although not all variety of process philosophy position themselves in the same manner, there are general tendencies that follow from the rejection of broad necessitarianism. A first debate concerns priority, in the sense Schaffer and others have recently understood in the discussion about priority monism and its alternatives.16 Priority monism holds that everything concrete is a part of a whole which logically pre-exists its parts – it doesn't claim that only the whole exists but rather that it is prior to anything else (its parts). Priority pluralism, on the other hand, takes concrete parts to be (logically) prior and the whole to be a matter of composition driven necessarily by the prior parts. While the latter posit something akin to atoms, the former would be comfortable with an all-encompassing cosmos. Process philosophy, conceived as broad anti-necessitarianism, couples poorly with priority monism as it postulates an over-arching cosmos prior to any other commencement (or command) but it doesn't agree very well either with priority pluralism as the command (and the commencement) of the atoms is also set once and for all. In fact, a process philosophy would best embrace priority nihilism, the rejection of both priority monism and priority pluralism. According to priority nihilism, there is no ultimate parts that rule what is concrete and no ultimate whole that shape things either – there is no priority as nothing in the universe is established in just one dose. Priority nihilism is a way to make sure no prior wholes or parts rule everything else through a priority that ensures determination.

A second debate where process philosophy position can be illuminated by its commitment to (broad) anti-necessitarianism concerns the necessary existence of everything. Timothy Williamson has coined the term necessitism for the thesis that everything is necessarily something – and called contingentism its denial.17 These are modal theses about ontology in the sense of whether what exists exist necessarily. If necessitism is right, whatever exists somehow has to be something, although not necessarily the same thing – a glass could be an elephant, or another glass or a dream but necessarily something. In terms of possible worlds, something present in one possible world is present in all of them; trans-world identity is a consequence of necessitism and nothing exists in just one world. Now, because the debate around necessitism is about (modal) ontology, it has no obvious implications for the one on necessitarianism – and therefore there could be a necessitist (non-necessitarian) process philosophy. However, necessitism is best construed as committed to an ontology where what exists arises from a single (logical) stroke; no other entity or interaction interferes in that unique commencement (and command). To be sure, process philosophy could accept the thesis that some things necessarily exists or that all things are capable to exist necessarily but these are already contingentist claims. Just like with priority nihilism, I contend that a genuine process philosophy is best coupled with some variety of contingentism. Moreover, a process philosophy reflection on both issues suggest that, further from broad anti-necessitarianism, process philosophy is committed to the claim that no arché goes unchallenged in all instances (by another arché). In other words, it suggests that no necessity runs within its (concrete) scope without interference of other necessities for the establishment itself of a specific scope for each necessity is a stroke of an over-arching arché. Further, it is not enough that there are (unexplainable) irregularities for that would entail Epicurean necessitarianism; rather, process philosophy requires some sort of multiplicity of archés.

Priority nihilism, contingentism and wide-spread contingency, together with a multiplicity of archés, suggest a similarity with Leibniz. In fact, Leibniz is often mentioned as a source for some (neo-monadological) versions of process philosophy, and this link will be explored below. But Leibniz maintains that his units of agency (his monads) are multiple, compose a world while being interdependent18 and are wordly in the (contingentist) sense that nothing exists in more than one possible world. (Leibniz has that Adam is a sinner and in a possible world where there is no original sin, there is no Adam.19) Leibniz, nevertheless, fails short of holding a process philosophy in an illuminating way. Although he holds that everything is contingent – because there are infinitely many equally possible worlds and no necessity but a wise decision from God selected one possible world among all – he also holds that there is a determination since God chose a world as a whole and did it in one dose, once and for all. Leibniz's system (and his controversies with Arnaud and Bayle20) illustrates how contingency is sometimes understood as indetermination – which is distinct from what takes place in one among many possible worlds. Leibniz insists that while everything in the actual world is contingent because there other equally possible worlds that happened not to be chosen to become actual, what happens is determined by the choice of world once made. Therefore he holds no necessitarianism as there is no necessity behind what is determined and further posits many archés that act spontaneously even though their choices are already determined. However, Leibniz is not a broad anti-necessitarian in the sense that there is an over-arching decision that made the actual world as it is from one stroke – once and for all. The present time where actual things happen is a follow-up from the time of choice when God wisely chose one among infinite many possible worlds and that time of choice itself somehow had to follow a time of contemplation where all the possible worlds were constituted by necessary rules of compossibility. This way these possible worlds are constituted are described by Leibniz through the image of the Palas palace with an infinite number of rooms each of them being like a possible world and each of them containing only elements that stick together (that are, as a matter of reason, compossible). The appeal to these three consecutive times – albeit the first is not itself a time if God needed no time to contemplate all possible worlds – distances Leibniz from broad anti-necessitarianism as there is an over-arching decision taken in a time that precedes the present time.21 The precedence of a stroke of a creator leaving no room for a present time interference from another arché makes Leibniz a broad necessitarian and therefore not a process philosopher. Interestingly, though, he holds both that archés are multiple and that (non-broad) necessitarianism is false.

3. Agency in the World
Process philosophy, understood as committed to broad anti-necessitarianism, is to a great extent about agency. It is therefore important to have a characterization of it but I will limit myself to three rough features. First, it is connected to arché: an agency is a commencement and a command in contrast with anything that follows a determination, a rule or whatever has been previously established. In that sense, exercises in agency are acts that are both non-determined and determining. Second, it requires no intentionality: an unintended act can equally well be an instance of agency if it commences or commands something – spontaneity doesn't have to be intended. Third, agency springs from contingency and generates something that is akin to a determination; further what is there to be determined has to be itself contingent. The logical space of agency is that of something being determined, being chosen, being rules, being founded or being originated. To be sure, what is determined is not necessarily necessary – for it could be otherwise. In fact, what determines, while determining, is never under a necessity because if it were necessary it would not be itself determining anything – not commencing anything – but merely being determined to determine.

Now, if process philosophy is committed to broad anti-necessitarianism, it is committed to the existence of agency in the world (and, in fact, to the existence of more than one agency). If there is no agency in the world – or no agency among concrete things – there is no process philosophy: images of the world without agency lead to broad necessitarianism and, indeed often to (strict) necessitarianism. If everything is determined by necessity there is no agency; but also if everything is (necessarily) contingent there is no room for agency and broad necessitarianism is the case. A principle that assures contingency everywhere doesn't prevent God from coming to exist, but does prevent any (other) arché to come to exist. Meillassoux's principle of facticity, like Leibniz's system as discussed above, falls short of broad anti-necessitarianism even if it is not (strictly) necessitarian. In both cases, determination by necessity and ensured ubiquity of contingency, the concrete is not genuinely productive and nothing in it commands or commences.

On the other hand, the presence of agency in the world can take many formats. To begin with, it can be ubiquitous or exceptional. Peirce's suggestion against necessitarianism was to spread spontaneity throughout the universe so that both stability and irregularities could be explained through them. Process philosophy doesn't have to assume it is ubiquitous, but it often finds agency beyond the human and not only in Gods. Apart from ubiquity, the presence of agency in the world raises the issue of whether or not there are individuated agents. As a consequence, process philosophies can either postulate agents or conceive agency as independent from individuals. There could be agencies responsible for processes that are themselves what give rise to individuals. Individuals are then conceived as the result of an agency or of an interplay of agencies that are themselves not arising from individual agents. The idea of agency without agents is present in several process-oriented claims: Gilbert Simondon's idea that individuals respond to continuous processes of individuation, Deleuze and Guattari's conception of agencements as separated from singularities that prompted Foucault's diagnosis that the individual is a product of power, Karen Barad's agential realism that posits agency inside individuals in their intra-actions (as opposed to inter-actions).22 In all those cases, individuals are not themselves archés. A universe without agency is different from a universe without agents; in the former nothing has the power to affect the course of things, in the latter no individual thing can do it although the course of things can be continuously changed by forces, drives, movements or intra-actions and those changes can include the generation of individuals. In both cases, no agent, qua agent, affect the course of things or promote further diversification or specification in the universe. However, a universe with agency (with or without agents) is a universe that is not made in one stroke but is can be revamped and reconfigured to its utmost structure.

In contrast with a world with agent-less agency, we can envisage a world made by independent, ready-made agents that concentrate all the agency in the world. These individual agent affect the course of things are endowed with agency from the beginning and are individuals before any agency can take place. If we compare views of the universe with views of human society, the image of individual agents can be paired with that of a contractualist society of individuals who build social institutions by making deals. In both cases, the priority in agency is in the individual who independently carries on an action and every collective chance is an effect of the intervention of one or more agents. In this image, it is hard to imagine genuine relations and genuine becomings as both this things boil down to an individual character who promote it; individuation is itself independent of any relation and becoming. Such image is close to that postulated by Graham Harman in his object-oriented ontology where independently individuated objects are the ultimate constituents of the world through their capacity to crack deals, to engage in network and to make room for further, equally independently individuated, objects.23 Harman himself contrasts his view with that of authors like Simondon, Bergson and Stengers on the one hand and Whitehead and Latour on the other by saying he has no room for becomings like the former nor for relations like the latter.24 The concentration of agency in independent individual agents place agents themselves mostly somewhere immune to process, and therefore substantially out of reach to the archés that shape the course of things.

A departure from the idea of individuated agents is provided by process philosophers of a more monadological persuasion – such as Whitehead and Latour. Their position can be seen as in between that of agent-less agents and that of a world of individual agents. Here agents are not individuated through a substratum but rather occupy a place among the various interdependent agents around them. Whitehead holds that agents – his actual entities – are both primary and dependent on the other actual entities on which they act. It is agency that makes an agent what it is while agents are the primitive basic elements of the ontology – Whitehead's ontological principle has that no actual entities, no reason. The individual is therefore a product of power but with the proviso that power itself stems from other individuals. This monadological (or neo-monadological) position shares with Leibniz the postulation of units of action that are all different from each other and all part of the general configuration of things. Like in Leibniz, these units are defined in terms of each other and their mutual compossibility is crucial for their identity. Unlike in Leibniz, no global selection of a series of units of action takes place in one stroke and therefore nothing is determined once and for all – these (neo-)monadologies are process philosophies as they break with the broad necessitarianism held by Leibniz. Each unit of action is driven in its agency by the co-existing other units and none of them has no effect in any other. Monadological process philosophy posits agency as tied to individuals, yet relations between them make individuals what they are to an extent where there is no priority either of the individuals or of the inter-relational network connecting all of them. In Whitehead, actual entities are not substances and do not remain beyond the completion of an act of prehension – which can be roughly understood in terms of perception.25 Further, in Latour there are an indefinite number of monads as the unity of agency can only be counted with respect to a test of resistance – the distinction between a monad and a network of them is relative to the circumstance of their subsisting that is always dependent on a measurement of forces.26 The distinction between agent-driven and agent-free agency is therefore somehow blurred as individuals can become non-substantial, temporary and indefinite. The difference between the two ways of understanding agency seems dim if we take into account that the individuals in a monadological approach to agency are fully embedded in a process that continuously constitute them. The two approaches split only in the sense that the inder-dependent agents approach is monadological in its roots – agency comes from units that are somehow already individuated.

I have presented four positions concerning agency in the universe. They can be ordered from the one which posits no agency to the one that posits fully independent units as sources of agency – respectively the first and third positions that appeared above. In that sequence of positions, the first one finds no room for agency in the universe and embraces a necessitarianism at least in the broad sense. The second posits agency that changes the course of things in the universe but no agents as units that can be deemed as its source. The third one posits those agents but they are interdependent, relational and non-substantial – they mingle with the exercises of agency that continuously take place and give rise to new units of action. Finally, the last position ascribes agency to independent agents whose individuation are immune to the interference of any agency – they are individuated through something like a substratum that is not affected by what happens to them. The ordered four positions can be viewed on a table:

No agency
Agency without agents
Agency with interdependent agents
Agency with independent agents
Necessitarianism (or broad necessitarianism)
Anti-necessitarianism (and broad anti-necessitarianism)
Anti-necessitarianism (and broad anti-necessitarianism)

Process philosophy
Process philosophy

The second and the third position afford process philosophies; maybe there is a continuum of positions between 2 and 3. Both are broad anti-necessitarian, in contrast with 1. Position 4 is anti-necessitarian and perhaps even broadly anti-necessitarian but it is perhaps a less fertile soil for process philosophy as the individuation of units of agency is immune to every arché. In fact, Harman himself when contrasting his object-oriented ontology with positions like Whitehead's and Simondon's stresses that there is room for process in those positions but not in his.27 In any case, positions 2 and 3 gather the conditions for process philosophies as they posit agency in the universe and understand that no unit of agency is constituted independently from the effects of the others.

4. Conclusion: Process and Immunization
If process philosophy springs from the rejection of broad necessitarianism, it entails that the universe is never established once and for all – it is in this sense incomplete and therefore falls short of a “barren tautological absolute”. Therefore further action can always contribute to the shape of things – what is concrete in the universe is open to further agency, to the effects of other commencements (and other commandments) as nothing is established once and for all. The concrete is therefore what is always subject to the effect of other archés – therefore it is up for grabs. Different process philosophies put forward accounts of how this different agencies relate and how they affect what is concrete. Process philosophy has to do with a universe conceived as open to new commencements. Considering the 4 above positions about the presence of agency and agents in the universe, we appreciate how commencements affect other sources of commencement. Position 2 rejects units of agency and therefore agency itself depends on the processes within the universe. Position 3 posit units but they are interdependent and respond to the processes within the universe in their individuation. Position 4, in contrast, considers agents immune to the processes – as a consequence, there is a limitation in scope and therefore in intensity of what is open to be affected by new commencements. Process-approaches committed to position 4 have less space for process because less is up for grabs. The scope for process is therefore bound by individuals that are immune to agency. Apart from requiring broad anti-necessitarianism, process approaches need what is up for grabs. If there is something concrete that is immune to agency, process is bound. The advantages of positions 2 and 3 over 4 (and 1) is that they posit nothing that is immune to agency. Immunity contrasts with process because it contrasts with what can be affected, it contrasts with dependency; what is immune is independent – just like the whole for priority monism or the atoms for priority pluralism. However the amount of immune elements among concreta is a matter of degree; this is why there are ontological assumptions that are make more room to process approaches. When something is immune to further command (or to further commencement), there is a necessity (or an arché) that reigns unchallenged, even if only in its restricted scope.28 To be sure, one can put together a process approach to some areas of what is concrete – for example, to human mind or human societies as does approaches that understand humans as exceptions to a necessitarian rule29 – and a (broad) necessitarian approach somewhere else. In that case, what is immune to further command is precisely what determines the limits of process.

To conclude, few words about the limits of agency. To propose a process philosophy is to propose an image of a process that constitutes the universe by introducing (more than one) commencement and commandment. That image could privilege the search for satisfactory novel increments to the complexity of the world, or the thrive for adequate response to demands or the aggregation of existing elements or forces in new constructs. In any case, the image provides an account of the constituting processes by presenting some limitations embedded in them – we could invoke here a motto: definire est eliminare. Whitehead sets up to address this limitations considering that process cannot be understood unless a framework is provided and that involves, for example, a non-processual notion of spatial relations and a notion of potentiality.30 Importantly, however, these defining limitations of process are not conceived as determining an area of what is concrete in which there is an immunity to any process. These limitations purport to do no more than explain how process takes place positing a structure to them. These explaining structures can be metaphysically very different from each other and they are best if they not only postulate genuine processes reinventing the universe beyond any initial setting but also show that they imply no area of what is concrete that is immune to such reinvention. If broad anti-necessitarianism is a necessary condition for process philosophy, minimizing concrete immunity is perhaps its aim. In any case, it is at its best when it explains genuine novelty without preventing it. In other words, when it gives an account of process that does not subsume the effects of all the archés to come.

No comments:

Post a Comment