Towards a Theory of Dynamicity: Foundations for a Non-Vacuous Process Metaphysics

Abstract

Process metaphysics seeks to provide a novel foundation for metaphysical explanations of entities in both scientific inquiry and everyday experience. It aims to better explain ongoing phenomena—moving, raining, and the like—by analysing them as fundamental processes (FP), that is, dynamic entities not further reducible. Crucially, I argue, this analysis and the ultimate value of process metaphysical explanation hinge on an understanding of what dynamicity is; without one, the central thesis concerning fundamental processes remains vacuous. The paper examines metametaphysically what an account of dynamicity should provide, defending three desiderata: (1) difference-making: it must draw an informative, not merely stipulative, distinction between dynamic and static entities; (2) explanatory power: it must provide the conceptual resources to yield explanatory claims about dynamic entities and apply broadly. On the basis of these desiderata, I argue that prominent accounts of dynamicity, that is the mereological and modal account, prove unsatisfactory or miss their mark. The paper concludes by developing and defending an account of dynamicity as temporal forward-directedness, thereby linking process metaphysics to realist theories of time.

1. Introduction

The guiding conviction of process metaphysics is that any comprehensive understanding of reality must begin with the dynamic nature of being.1 Accordingly, process metaphysics strongly opposes any attempt to reduce dynamicity, or in any other way derive dynamicity from non-dynamic, i.e., static categories. While constituting a highly heterogeneous paradigm (cf. e.g., [2,3,4,5,6,7,8].), the opposition between dynamic and static entities accompanied by an emphasis on the former, is pervasive throughout the process literature. This emphasis typically has two sides: one negative claim, (a), that dynamicity is not to be further reduced, and one positive claim, (b), that in some (possibly all) cases, static entities are derivative of dynamic entities. To illustrate, consider the following two examples, one drawn from everyday experience and the other from scientific inquiry.

(1) A common, everyday phenomenon is rain. Rain is ongoing, has a beginning and end in time, is spatially spread across a region, and occurs in degrees (light, mild, and heavy rain). Rain involves the moving of raindrops, water molecules. We reference the ongoing rain, by expressions such as “it is raining.” Now, process metaphysicians stress that rain should primarily be understood from its dynamic character and that said dynamic character should not be reduced to non-dynamic entities (a). Therefore, according to process metaphysics, rain should not be understood as a mere collection of water molecules that occupy different places at different times. Rather, rain is a complex, relational orchestration, coordinating (“synthesizing” [9] [p. 158]) different aspects of reality. In contrast to logico-grammatical appearance, namely that “it rains” suggests that there is an underlying subject to the ongoing process, process metaphysicians insist that rain is a classical example of what they have termed ‘subjectless’ [10] [p. 48] (‘absolute’ [11] [p. 387], ‘unowned’ [12] [p. 12]) processes, that is, ongoings that are not the activity of an underlying object, thing, or substance [13] [p. 162].

(2) The second example stems from the study of hydrodynamics, that is, fluids and gases under the influence of forces, pressure, and temperature, where a canonical example of hydrodynamic (in)stability is the Bénard convection cell.2 The Bénard convection cell is a structure that arises when a horizontally spread liquid is heated from below and cooled from above (typically by plates) (cf. [15] [p. 71]). If the temperature difference is low, the heat flow occurs through conduction and the fluid remains static (cf. [15] [p. 71]), yet ‘[i]f the difference in the temperature $\Delta$T = (T_h− T_c) is increased, there is a point at which a well-organized pattern of convection cells emerges. The threshold value of $\Delta$T depends on the fluid properties, such as the thermal expansion coefficient and viscosity. What is remarkable about this familiar convection pattern is that it emerges entirely out of chaotic motion associated with heat. Furthermore, the fluid’s organized convection pattern now serves a ‘function’: it increases the rate of heat flow. This is an example in which the energy low drives a system to an organized state which in turn increases the energy flow.’ ([15] [p. 71]) Compare the following (Figure 1):

Process metaphysicians have taken the Bénard cell example (and similar structures) to provide ground to the idea that the second aspect of the emphasis on the fundamental dynamicity of reality lies in understanding that seemingly static entities are derivative of fundamental dynamic entities (b). Namely, according to a processual understanding, the (relative) stability of the cell [here to be understood as a static phenomenon] is to be found in structures of underlying, more fundamental activity—the stable (or static) is derivative of the dynamic, the processual.3 Although apparently a stable thing, an individual convection cell, the Bénard cell’s persistence and relative resistance to small temperature changes ‘emerges entirely out of chaotic motion associated with heat’ [cf. quote above], that is, a coordination of underlying processes.4

Let us bracket the question of whether only process metaphysicians can provide a plausible metaphysical analysis of these examples, that is, whether these examples speak in favour of process metaphysics. Before addressing this, one must ask what the processual explanation amounts to, in contrast to a non-processual analysis. To claim that the entities in question must (fundamentally) be understood as processes is only metaphysically interesting if one understands what a process is in contrast to a substance or other static categories; that is, if one underpins the claim with a meaningful understanding of dynamicity.

By way of comparison, when confronted with popular science writing, one sometimes encounters people insisting that everything fundamentally is energy and that we must revise our worldview/metaphysics accordingly. However, before we can even assess whether this claim is true, we need to clarify what ‘energy’ means here—the technical concept from physics (a conserved quantity associated with symmetries), some quasi-vitalist force, or something else entirely. Without this clarification, the claim remains worse than false; it remains empty.

In this paper, I argue that process metaphysics is committed to a particular understanding of dynamicity, namely, as temporal forward-directedness. The argument for this claim proceeds in three steps. First, I argue that process metaphysics must have some understanding of dynamicity, lest its central thesis—the fundamentality of dynamic entities—be void (Section 2). I then provide and argue for some general criteria against which any theory of dynamicity should be evaluated (Section 3). The third step of the argument consists in evaluating alternative theories of dynamicity against these criteria (Section 4) and arguing that dynamicity as temporal forward-directedness outperforms the considered competitors (Section 5).

2. Process Metaphysics: The Fundamentality of Dynamicity

‘Process philosophy’ is an umbrella term covering a wide variety of views. While originating (in its modern form) in the work of Alfred North Whitehead [17] [p. 3], it has since been adopted, modified, and presented in various forms.5 Thus, although many views have at one point been said to fall under ‘process philosophy’, there is little commonality among them. To defend a strongly coherent, unifying label of ‘process philosophy’, much comparative work would need to be undertaken. Instead of doing so, I focus on ‘process metaphysics’ and provide a substantive, cohesive definition for that label.6 I believe that the ideas presented here are what people usually have in mind when pursuing process philosophy more generally, albeit sometimes in non-metaphysical terms or implicitly. Ultimately, however, my aim is to provide a coherent, fruitful metaphysical view, regardless of precisely which authors will belong to it.

I have argued elsewhere [18] [pp. 53–58] that a particularly fruitful definition of process metaphysics lies in characterising it as comprising those views which adhere to the following claim (the fundamentality of processes (FP)):

FP. Some processes are absolutely fundamental [18] [p. 53].

This claim amounts to holding that there is no deeper building block of reality, no more fundamental level below processes to which the latter reduce. One example of such an alleged reduction is the so-called ‘at-at theory’ of change (cf. e.g., [4] [p. 171]), according to which phenomena of flux can be explained in terms of a series of property instantiations. At a minimum, process metaphysics denies this analysis of dynamic phenomena (sometimes called ‘flux’ (cf. e.g., [10] [p. 57]; [18] [p. 37]))—yet it equally denies any other theory that reduces processes to other types of entities.

I hold that (FP) is but a minimal definition of what process metaphysics encapsulates, in that all process metaphysicians should agree upon it, yet many such metaphysical theories will go considerably further than (FP).7 Process foundationalists hold that processes and only processes are absolutely fundamental, i.e., that no other entities can exist at the most fundamental level.8 Even stronger, process monism is the view according to which all of reality only consists of processes; there are no other entities other than processes.9

I contend that any such processual view requires a supplementary account of the nature of dynamicity. There are three steps to my argument. First, in all processual views, including the minimal view, there is an underlying supposition that featuring processes at the fundamental (or even all) levels of reality is explanatorily useful. In other words, process metaphysicians hold that introducing processes into metaphysics without reducing them yields some explanatory gain: something can now be explained, or better explained, than it could be by alternatives lacking fundamental processes. Without delving into the arguments for such a gain, it is almost trivial to observe that unless such a gain is achieved, one need not bother with process metaphysics. If the existing alternatives are already sufficiently powerful in their explanatory capacity, then process metaphysics is entirely unmotivated.

Second, a claim about what is fundamental is only explanatorily useful if we understand the entity in question sufficiently well. Moreover, one can only argue against views that do not hold the entity in question to be fundamental if there is a significant, well-understood difference between the entity that one wishes to place at the fundamental level and those posited by one’s competitors. What, then, is that difference in the case of processes? What informs the claim to the fundamentality of processes (FP)? It is, arguably, the dynamicity of processes [Seibt, Galton, Bickhard, Rescher, Whitehead]. Processes differ from other categories—substances, properties, or states—in that they are dynamic rather than static.

Third, even the characterisation of processes as the dynamic entities is not sufficient to render (FP) a meaningful metaphysical thesis—that is, unless dynamicity is further specified. Without doing so, the process metaphysician may just as well have said that what sets processes apart from other entities (as required by the second step of the argument) is that they are processual—which clearly begs the question. It is necessary to provide a positive characterisation of the central feature of processes, dynamicity, that does not already presuppose an understanding of what it is to be a process. In other words, the concept of dynamicity must be elucidated in terms that are independent of the category it is meant to characterise; otherwise, the purported explanation is circular.10

In summary, process metaphysics stakes its explanatory ambitions on the fundamentality of dynamic entities; yet without a substantive account of dynamicity, such ambitions remain unfulfilled.11

3. Desiderata for a Theory of Dynamicity

In the previous section, I argued that the prospects of process metaphysics are grim without a robust understanding of dynamicity. The following sections discuss the influential accounts and ultimately present and defend my preferred view. Before turning to that task, however, the present section aims to make explicit what is demanded of metaphysical theories of dynamicity, thereby laying out two central metametaphysical criteria against which any such theory can be evaluated.

Thus far, the literature on dynamicity does not feature agreed-upon desiderata for dynamicity. Yet without such desiderata, there is a danger that different authors will want quite different things from the concept of dynamicity without making transparent what they are after. Therefore, independently of the accounts in question, the formulation of such desiderata provides a framework against which one can first settle what the sought-after concept is meant to achieve. Only once this framework is in place can competing accounts be evaluated fairly. For the game concerning the best concept of dynamicity to be fair, it is crucial that such a list be relatively neutral and not already metaphysically loaded. Otherwise, the disagreement will presumably simply shift to the level of desiderata, for which one would again have to provide a framework in which to adjudicate which set of desiderata is best—and so on ad infinitum.

While there is no explicit consensus in the literature, Seibt has prefaced many of her texts with a methodological section from which one can clearly extract some demands on the concept of dynamicity.12 Similarly, Whitehead’s Process and Reality features an insightful preface on the endeavour of speculative metaphysics, from which one can derive some criteria [28]. While the forthcoming criteria draw on these resources, they are further inspired by Carnap’s general thoughts on the explication of concepts and by the lively metametaphysical debates on the desiderata for grounding.

3.1. Difference-Making

As argued in Section 2, the central demand for any process metaphysics on dynamicity is that it sufficiently informs the thesis (FP). It is because of dynamicity that processes differ from other entities. Thus, first and foremost, a theory of dynamicity must elucidate the distinction between dynamic and static entities.

Difference-Making. A theory of dynamicity should inform in a meaningful way the difference between dynamic entities and static entities (cf. [18] [p. 105]).

Arguably, any definition of dynamicity will make some difference between dynamic and static entities. Suppose that ‘being dynamic’ is identified with the property F. Then, any entity that has F is dynamic, and any entity that lacks F is static. Therefore, regardless of what F is, one can draw a distinction between dynamic and static entities, and consequently, any theory of dynamicity does have some difference between dynamic and static entities as a consequence.13

For that reason, the desideratum of difference-making does not simply require drawing any distinction between dynamic and static entities, but rather ‘informing in a meaningful way’ such a difference.14 One important requirement for doing so is to track the examples in Section 1. These serve as paradigmatic cases; the theory must clarify what it is about them that leads us to judge certain phenomena as dynamic and others as static. More specifically, rain and the underlying chaotic movement within the Bénard cell should, in any adequate theory of dynamicity, unambiguously qualify as dynamic phenomena, whereas the liquid at rest or the stabilised whole of the Bénard cell should qualify as static. A theory that fails to vindicate these intuitive classifications lacks extensional adequacy.

Now, introducing a metaphysical concept need not require that in each and every case where we intuitively (or colloquially) believe the phenomenon to be ‘dynamic’, it does in fact turn out to be so. There may be cases where the introduction of the proper concept requires some revision.15 Equally, there may be cases of phenomena that are intuitively thought to be non-dynamic (i.e., static) that, on the proposed account, turn out to be dynamic.16

In this sense, difference-making is akin to Carnap’s criterion of similarity (cf. [30] [pp. 5–6]). The desideratum is to have a concept that is extensionally adequate, in that the concept captures paradigmatic cases. Otherwise, we would be unable to understand the concept as a concept of dynamicity. On the other hand, significant revisionism is permitted, for ultimately the metaphysical theory is only useful if it provides a new perspective. The balance to be struck, then, is one between conserving our pre-theoretical judgements sufficiently to retain grip on the phenomenon and understanding the theory while permitting enough revision to yield novel theoretical insight.

3.2. Explanatory Power

The second desideratum concerns metaphysical fruitfulness. To claim that (some) dynamic entities are fundamental is worthwhile only if this yields explanatory power, a significant part of which must stem from dynamicity itself (cf. Section 2 ). So, secondly:

Explanatory-Power. A theory of dynamicity should be explanatorily powerful, i.e., (i) provide the necessary concepts and interconnections to yield explanatory rather than merely descriptive claims about dynamic entities, (ii) apply equally across a plurality of domains, and (iii) conform to the characterisations already present in the process framework.

The three aspects that compose explanatory power—(i) to (iii)—are interwoven. Aspect (i) captures the idea that the desired concept of dynamicity should be of genuine service to the process metaphysical view. Specifically, given the explanatorily powerful concept of dynamicity, process metaphysicians should be able to explain (metaphysically) dynamic phenomena, how they hang together, and how they relate to one another. Beyond merely attaching the label ‘dynamic’, the desired concept should inform the metaphysical structure and provide grounds for further metaphysical explanations.

The second aspect, (ii), encapsulates the idea that process philosophy (and thereby process metaphysics) is meant to have ‘full systematic scope’ [1]. As previously mentioned, process metaphysicians attempt to understand biological life, the nature of activity, the fundamental entities of physics, the nature of time, and much more. A theory of dynamicity that applies broadly across these different domains and remains informative is highly valuable, as such a broad application would indicate that it truly captures the essential characteristic of processes. The plurality of domains also covers immediate, everyday experience, and thus, this aspect of Explanatory-Power pays tribute to Whitehead’s idea that process metaphysics should integrate both scientific explanations and (even more importantly for him) immediate experience, for ‘[t]he elucidation of immediate experience is the sole justification for any thought [28] [p. 4].’

Third, the process literature already comprises influential and well-developed metaphysical views on various subjects. A concept of dynamicity that cannot accommodate at least some of these views, or is outright incompatible with them, faces the burden of providing numerous new explanations at once and potentially setting back process-philosophical progress. In contrast, a concept that coheres with the existing literature and conforms to and informs previous work (iii) possesses considerably greater explanatory power.

Ultimately, all three aspects underscore the same demand, namely, that the relevant concept of dynamicity be of explanatory use in process metaphysics. These aspects merely emphasise different perspectives on the explanatory use: (i) how much genuine explanation, rather than mere description, is underwritten by the concept; (ii) how broadly these explanations apply; and (iii) how well the concept coheres with existing explanations.

3.3. Other Desiderata

For present purposes it will suffice to settle on these two desiderata. On the basis of them, one is already in a position to criticise and evaluate influential accounts on the nature of dynamicity. Yet, one might wonder whether these two form an exhaustive set of desiderata. Certainly, other theoretical virtues are standardly invoked in metaphysics, including parsimony, internal consistency, elegance, unification, and fit with established science. I do not deny that these virtues matter; indeed, some are implicitly subsumed under the desiderata already articulated. Internal consistency, for example, is a minimal requirement for any theory and need not be listed separately. Elegance, while desirable, is notoriously difficult to define and risks collapsing into mere aesthetic preferences.

Three further candidate desiderata merit a brief discussion. First, Coherence with Scientific Practice: a theory of dynamicity should, ideally, mesh with how scientists characterise dynamic phenomena in physics, biology, and the cognitive sciences. This consideration overlaps with explanatory power (ii), but it specifically emphasises that metaphysical theorising should not proceed in isolation from empirical enquiry. A theory of dynamicity that conflicts with or fails to illuminate scientific descriptions of dynamic systems counts as a disadvantage.

Second, fecundity: a theory of dynamicity should prove fruitful for further philosophical enquiry. A concept that opens new avenues of research, generates novel questions, and integrates with adjacent debates (e.g., in the philosophy of time, causation, and emergence) is preferable to one that remains isolated from them. Fecundity is difficult to assess in advance, but it provides a forward-looking criterion for evaluating theories once they are articulated.

Third, parsimony: For any two process metaphysical theories with equal explanatory power, one should prefer that theory which minimises the introduction of novel fundamental ideological commitments.

While often cited as a theoretical virtue, the nature of parsimony—whether it should count as a virtue and, if so, how to formulate it—remains controversial (cf. e.g., [31] [p. 3890]). Thus, the ‘taste for desert landscapes’ must be motivated beyond mere aesthetic appeal.17

Following [31], I believe that apart from quantitative considerations, i.e., how many entities a theory posits, how many of them are fundamental [33], or how many types thereof are demanded, there is the question of ‘ideological commitments [31] [p. 3894]’. Ideological parsimony concerns not merely how many things a theory posits, but how many primitive, undefined notions it employs. A process metaphysical theory which minimises those ideological commitments, while remaining explanatorily powerful, is ideologically more parsimonious than a theory that introduces more novel primitives.

These supplementary desiderata are not given separate treatment in what follows, partly because they are subsumed under or closely related to the three main desiderata, and partly to keep the evaluative framework manageable. Nonetheless, they inform my assessment of the candidate theories in Section 4.

4. Theories of Dynamicity

Employing the criteria developed in the previous section, this section discusses two prominent accounts of what dynamicity could be. Surveying the literature on process metaphysics and other dynamic enterprises may reveal more, such as a linguistic theory of dynamicity (cf. endnote 6) or primitive/brute views on dynamicity (cf. endnote 6). Yet, in the field of process metaphysics, the most influential views are certainly the mereological and modal accounts of dynamicity—and, of course, the temporally directed account, which I will defend in the upcoming section. I will discuss and criticise each of the two accounts in turn before working out my preferred proposal more positively.

4.1. Mereological Theories of Dynamicity

In her comprehensive work on process philosophy, Seibt proposed understanding dynamicity in terms of self-partedness. She is a prominent proponent of a mereological account of dynamicity.18 On such an account, dynamicity consists of particular structures of mereity (cf. for example, [23] [p. 12]), that is, dynamic entities display a specific “mereological signature” [26] [p. 486].19

Seibt’s account is inspired (and partly motivated) by a reading of Aristotle’s Metaphysics $\mathsf{\Phi }$, 6 on the (double-)distinction between kinesis and energeia on the one hand and dynamis and energeia on the other hand. According to her understanding, Aristotle draws the distinction with the help of a ‘completeness test’ [23] [p. 6].

This difference in completedness is explicated by means of differences in the inferential role of sentences expressing ‘cases’ of energeia and kinesis. […] [S]tatements about kineseis abide by the pattern

[8.] (subject, Greek verb in present tense) implies the falsity of (subject, Greek verb in the Perfect)

while statements about energeia do not govern such implications—the present tense statement is compatible with the associated statement in the perfect. […] [S]o for statements about energeiai it holds:

[9.] (subject, Greek verb in present tense) implies (subject, Greek verb in the Perfect) [23] [p. 8].

For example, ‘Aristotle is building the house’ implies that ‘Aristotle has built the house’ is false, whereas ‘Aristotle is seeing’ implies that ‘Aristotle has seen’ is true. Therefore, we find that for different activity words, different so-called ‘aspectual inferences’ (cf. [23] [pp. 9–10]) are validated by them.20

Now, importantly, these ‘aspectual inferences dovetail with mereological properties’ [23] [p. 11]. Namely, Seibt argues that to build a proper ontological account, aspectual inferences are licensed by particular mereological properties of the processual entity that the verb corresponds to. In ‘Aristotle is seeing’, the verb corresponds to a process, namely seeing, which has a specific mereological structure differing from the process of building a house. Seeing is ‘self-contained’ [automerious] [23] [p. 12], meaning that some (or even all) spatial and temporal parts of the process are again regions where the entire process occurs.

I want to suggest […] a standard, implicit definition of ongoingness of “dynamicity.” […] I suggest that we take the feature of strict automerity as an inferential constraint that, in combination with additional restrictions within a formal system, all and only ongoing or “dynamic” entities fulfill. The general strategy of my proposal can be summarized into the following four claims: (i) dynamicity or dynamic being is ongoingness by expressing itself—to be is to go on; (ii) self-expression is reflected in a characteristic structure in the description of an entity; (iii) for example, if we describe an entity E in terms of a partition that specifies what is “part of being that entity” (i.e., we specify its spatial, function, material etc, parts within one partition using a generic part relation), the fact that E is dynamic will be reflected in the partition in the form of certain distinctive patterns representing self-expression; and (iv) self-expression is mereologically reflected in strict automerity, which generates self-similar patterns within a partition [23] [pp. 18–19].

Dynamicity, in short, is self-expression, is strict automereity. As Seibt notes, this requires moving beyond classical mereology, where strict self-parthood is impossible. Instead, she proposes to move to a ‘levelled mereology’ (cf. e.g., [6] [p. 141]), dropping either the anti-symmetry or transitivity requirement of classical mereology and, thus, allowing for self-parthood [23] [p. 19].

Other process philosophers have adopted similar understandings of dynamicity. One can find in the process literature the widespread idea that processes (dynamic entities) behave like “mass-terms” (gold, water, etc.)—where again parts of mass-terms are of the same kind (“homomereous”), cf. e.g., refs. [3,35] building on [19,21,22]. In the following, I will label all of these views “mereological” accounts of dynamicity. Mereological accounts claim that dynamicity lies in (is identical to) particular structures of parthood, namely like-parthood (homomereity) or self-parthood (automereity).

The two possibilities of like-parthood and self-parthood allow for mereological accounts of dynamicity to come in different shapes. Some of these accounts claim that minimal homomereity suffices for an entity to be dynamic in at least some sense. Seibt has expressed this in other texts. With such a wide understanding, even entities traditionally understood as substances, such as a cup or a table, are minimally self- or like-parted and, thus, categorised as ‘type-5 processes’ [26] [p. 488], underpinning process monism (i.e., the thesis that there are only processes, cf. endnote 6).

Yet, from the passages above, one could also motivate a stricter understanding of dynamicity that only classifies those entities as dynamic which are strictly automerious. Either way, processes are classified along their ‘mereological signature’ (cf. for example, [26] [p. 487], [23] [p. 12], [6] [pp. 142–143]) and I acknowledge that Seibt’s analyses of different types of dynamicity (cf. for example, [6] [p. 141]), the applications she has provided in the philosophy of cognition (cf. [36]), biology (cf. [37]), field theory (cf. [27]), etc., are proof of the explanatory power of such an account.

Therefore, the mereological account scores high with regard to explanatory power, for (i) it provides new explanations, which are (ii) applicable across a plurality of domains, and (iii) it offers a rich classification and nuanced differentiations of processes.

However, I believe that ultimately the account is unsatisfactory as an account of what dynamicity is, for it is (a) neither necessary nor (b) sufficient for dynamicity to display the mereological signature in question. To see (a), consider the examples of the physical processes of spin inversion and alpha decay. Regardless of whether one holds a wide mereological account, according to which everything is dynamical, or a more narrow account, these two should undoubtedly be classified as dynamical. If that is so, then, according to the mereological account, they should, at the very minimum, be like-parted (or in the stronger account, self-parted).

Now, regarding the parts of spin inversion, we may wonder whether the temporal parts of spin inversion are also processes of spin inversion. Similarly, are the temporal parts of alpha decay also processes of alpha decay? This seems highly unlikely. If some temporal part, $p_1$, of the spin inversion, s, was already a spin inversion, then what happens in $(s-p_1)$—the rest of s outside of $p_1$—cannot be a spin inversion. If so, we would have (at minimum) two spin inversions. However, this means that there are parts of s which are not the same as s and are not even of the same kind (either $p_1$ or $(s-p_1)$ are not spin inversions). This is similarly the case for alpha decay.

It is equally implausible to think that the processes in question are spatially like- (or even self-) parted.21 These processes are hardly spatially localised at all. The spatial region in which they occur typically does not have spatial sub-regions where the process occurs as a whole—one cannot, in other words, pin down more concretely where within that region they take place.

However, both spin inversion and radioactive decay should clearly qualify as proper ‘dynamic’ processes, thus, demonstrating that the criterion does not express a necessary condition.

On the other hand, (b) there seem to be many entities we would not want to qualify as “dynamic” which exhibit the mereological structure proposed to characterise dynamicity. Take a piece of rock—it is spatially like-parted; at least some spatial parts of it will again be pieces of rock. Equally, we may say that the rock on Tuesday is “like” or even identical to the rock on Wednesday. If we understand the rock on a 4-dimensionalist, perdurantist account, then its temporal parts are at the very least like the rock as a whole. Thus, it seems that the rock is also temporally like-parted. Yet, without already buying into process monism, we would not want to qualify rocks as dynamic entities. Moreover, even if we did, it seems that attributing these mereological structures to the rock does not plausibly help us see how “dynamic” the thing is; it seems entirely unrelated to that question.

It must be noted that beyond these worries, the move away from classical mereology towards levelled mereology comes at a cost. If we first have to understand mereology on fundamentally different premises to see how that new mereology sheds light on the idea of dynamicity, one would need independent arguments for such a new mereology if one wanted to defend process metaphysics on that basis. This is a theoretical cost; one has to revise another piece of the metaphysical picture for the theory to work. Some revisionaries may be happy to do it, but if there are other accounts of dynamicity that do not have that entry requirement, they are less theoretically costly.

4.2. Modal Theories of Dynamicity

Besides the mereological view on dynamicity, there is a strong history of understanding dynamicity in modal terms. Aristotle’s distinctions in Metaphysics $\mathsf{\Phi }$ equally motivate an account according to which dynamicity is the actualisation of a potency. In the following, I will label those accounts “modal theories of dynamicity”.

The mereological account focused on the internal, mereological structure of dynamic entities; the modal account, by contrast, locates dynamicity in directedness. One prominent such modal theory can be found in the current debates revolving around power metaphysics. In the philosophy of science, there is a heated debate on the nature of laws where neo-Humean accounts clash with dispositional, power-metaphysical positions: “Unlike neo-Humeanism, the powers metaphysic paints the world as dynamic: it treats action as coming from within; it puts the oomph back in causes; and treats possibility as genuine and bounded.” [39] [pp. 3–4] While the (neo-)Humean picture merely posits “that all there is to the world is a vast mosaic of local matters of particular fact, just one little thing and then another.[…]. At those points, we have local qualities: perfectly natural intrinsic properties which need nothing bigger than a point at which to be instantiated.” [40] [p. ix–x], the powers metaphysicians hold that reality cannot merely consist of those point properties, which are entirely loose and separate. Rather, there must also be powerful properties—properties which specify what a thing “could be in the future, or what it might do, should it find itself in the right (or wrong!) circumstances.” [39] [p. 46]

However, there is some debate on whether (and more importantly how) power metaphysics is actually in the business of advocating a dynamic position. While power metaphysicians have at various points invoked the notion of dynamicity (cf. e.g., [41] [p. 171], [39] [p. 3], [42] [endnote 8] 22), what that notion is and how central it is to the powers paradigm remains a matter of debate. Roughly, there seem to be two prominent positions on that question—(1) activism (or non-passivism) and (2) passivism.

[…], if a powers metaphysic gives a central place to activity (or activities) and/or treats activity as fundamental, it is dynamic; if the status of activities is somehow diminished, then the metaphysic is not dynamic. ‘Activists’ (as I shall call them) are those powers theorists for whom activity plays a central or fundamental role in their metaphysics. ‘Passivists’, on the other hand, treat activities as ontologically derivative [34] [pp. 2462–63].

With regard to that question, there is a strong parallel to the process metaphysical debate. Namely, insofar as power metaphysicians oppose the picture of ‘local matters of particular fact, just one little thing and then another’ (cf. quote above), the project clearly aligns with process metaphysics and qualifies as dynamical in that precise sense. One example of such an understanding of powers metaphysics is Groff (cf. [43]). The way they describe the nature of activity is strongly reminiscent of (if not outright identical to) how process metaphysicians would understand activities.23

The best way that I know of to get at what I mean by activity is to picture a child’s flipbook, or otherwise a reel of film. When the still frames of either are viewed in rapid succession, it looks as if the entities represented in them are moving around, doing things. By ‘activity,’ I mean that aspect of the world in virtue of which real animation differs, if it does, from flip-book or cartoon animation, or from moving pictures. [43] [p. 9883]

In other words, activists aim at avoiding what was previously labelled the at-at theory of change, which reduces phenomena of flux to a mere succession of states (cf. Section 2, page 3). Such a theory is also a theory of merely local (in the temporal sense) things, one and then another. Process and activist power metaphysicians will equally insist that such a story leaves much to be desired.

However, Williams and Groff also point out that activism is only one possibility among the power metaphysicians. Williams contends that “activism presently lacks a satisfactory model for the metaphysics of powers” [34] [p. 2476] and expresses doubt that one will eventually be delivered to make sense of activity, thus, described. In the meantime, he proposes to remain a passivist—i.e., to hold that powers are a genuine and worthwhile alternative to the Humean categorical picture, but not to require fundamental dynamicity or activity (cf. [34] [p. 2476]).

The present paper is not concerned with settling this dispute in the powers literature. Rather, what emerges seems to be one power-metaphysical position that is closely aligned with process metaphysics,24 namely activism, and a second more sceptical of fundamental dynamicity, namely passivism. Insofar as passivism challenges the idea that we can make sense of fundamental dynamicity, it directly opposes the central claim of process metaphysics. If so, it may be an interesting position in its own right, but will arguably be of little value in underpinning the process metaphysical conception of dynamicity. So, as a first caveat to the upcoming discussion, it should be noted that I want to focus here on those positions which may shed some light on the nature of dynamicity and examine whether their definition is applicable to process metaphysics (which may in fact represent the majority of powers views, cf. [43]) [pp. 9889–9890]. My goal is not to claim or defend that powers metaphysics relies on dynamicity or necessarily has to have a view on it—but rather to critically evaluate what notion of dynamicity one does (sometimes) find in the powers literature and to see how that relates to the project of process metaphysics.

A second caveat is in order before diving deeper into the modal analysis of dynamicity. Powers are properties. Thus, even if some criterion of dynamicity were to be found in the powers literature, such a criterion will be tailor-made for properties, i.e., characterise what it means for a property to be dynamic. While that is true, there are at least two strategies available that may yield a criterion for processes regardless. First, the characterisation of dynamicity offered in the powers literature, or other modal theories of dynamicity, may be such that, although primarily intended for properties, it is equally applicable to other categories, or, at the very least, requires only slight modification to be applicable. Second, one could introduce an explicit connection between the dynamicity of the involved properties and the processes or substances involved. Such a “bridge principle” may for instance state that an entity is weakly/strongly dynamic iff some/all of its properties are dynamic (or even more strongly: it is constituted by only such properties). In the discussion of the modal understanding of dynamicity, one will, thus, have to explore multiple routes for transferring such understanding to processes. With these two caveats in mind, let us now turn to the substantive question of what account of dynamicity the modal framework, and the powers literature in particular, can offer.

On the activist understanding of power metaphysics, one important aspect of powers is that they are drivers of change—that which brings about the change are powerful properties doing something, for example, exerting a force, heating, etc. Powers do something (cf. [43] [p. 9884]). Yet, how are they doing that; what is that dynamicity constituted by? A question raised similarly by Marmodoro:

Powers are posited as engines of change in the world: dynamic entities that literally ’power-up’ everything that happens. But what is it for a power to be dynamic? My point is that claiming that it is a power’s dynamism that powers-up change in the world gives a name to what we are looking for, but does not explain it at all. It looks as if it is a bit of magic that somehow makes the world an eventful place of changes. This is the black box problem of powers [46] [p. 4].

In the powers literature one finds different attempts at making sense of that “dynamism”. In her article, Marmodoro discusses the fact that “some philosophers have tried to account for the dynamism of powers in terms of natural teleology—a conception of the world that we owe to Aristotle” [46] [p. 5]. Yet, as Marmodoro argues, the teleological approach, while illuminating, does not fully resolve the black box problem: specifying what a power is directed towards (its telos) does not yet explain what constitutes the power’s own dynamism—what it is about the power itself that enables it to bring about its manifestation in the first place. For this reason, Marmodoro turns to an alternative framework in which the internal complexity of powers is conceived in terms of directedness: “An alternative approach to the explanation of the nature of powers is the one according to which the internal complexity of powers is conceived as the ‘directedness’ of powers: a power (of type) P is directed towards its manifestation/exercise (of type) M. So, what is directedness?” [46] [p. 7]

In another article [42], Marmodoro discusses the prominent influences (Molnar, Psillos) on the essence of ‘pure’ powers:

Powers are entities that are in a state of “readiness for action.” […] U.T. Place proposed […] that we think of intentionality as the mark, not of the mental, but of the dispositional (1996: 92). At the present time, his influence is widening in the circle of power-ontologists who define a power in terms of its directionality toward its manifestation. George Molnar for example writes, “having a direction to a particular manifestation is constitutive of the power property” (2003: 60). In the case of pure powers, in John Heil’s words, “all there is to [such] a property is its contribution to the dispositionalities of its possessors” (2003: 97).

Directedness captures the point of the comparison between intentionality and powers. Just as an intentional state is directed toward something beyond itself, so a power is directed toward its manifestation. Just as in the case of an intentional state, what it points to need not exist; similarly in the case of a power, what it points to need not come about, since a power may never be manifested. A vase has the power of fragility that is directed toward breaking even when the vase never breaks, or an electric charge has the power to attract or repel even if no charged particle ever comes close enough to be attracted or repelled. On the hypothesis that powers are pure, there is nothing more to fragility than the vase’s disposition to breaking under certain circumstances, and nothing more to charge than the particle’s disposition to attract or repel [42] [p. 29].

Powers are essentially directed entities, namely directed towards their manifestations. They are directed regardless of whether the manifestation occurs, that is, actualises. Consequently, powers are intimately tied to modality; the manifestations are possibilities that are brought about by the powers under the right circumstances (for a similar characterisation of dispositions, cf. Vetter in [47] [p. 129]).

From the alleged dynamicity of powers and their characterisation by their directedness towards a manifestation, one can attempt to distil an account of dynamicity, namely a modal account of dynamicity:

Modal account of dynamicity Dynamicity is the ‘actualisation of a potency’. That is, a property [generalised: an entity] is dynamic iff it is modally directed.

The modal account of dynamicity allows for a clear-cut distinction between dynamic and static entities. States and categorial properties, for instance, are not inherently directed towards potentiality; thus, they are static on the modal account. On the other hand, it will classify dispositions (powers) as dynamic. Returning to the examples from the beginning, the macroscopic convection cell is not a directed entity and, thus, static, whereas the movements that make up the cell are directed entities—namely, each movement is directed at occupying some place.

Furthermore, the modal account informs the distinction and provides (further) metaphysical structure as to why some entities are dynamic and others are not. However, it is unclear whether the modal account can identify the ‘right’ entities as dynamic and static. Again, one can argue that modal directedness is neither necessary nor sufficient for dynamicity. To see that it is not necessary, take again an unquestionably (at least for the process metaphysician) dynamic phenomenon—rain. Now, given the modal account of dynamicity and the caveat that it only properly holds for properties: (a) rain must itself be a modally directed entity, (b) rain must have only properties which are modally directed, or (c) rain’s dynamicity is in virtue of having at least some dynamic properties.

With regard to (a), it seems unmotivated and implausible that ‘rain’ itself is a modally directed entity. After all, what would the manifestation of that entity be? Instead, one could think that rain is itself a manifestation of some disposition in the weather system—but that would get us no further in understanding rain’s dynamicity on the current criterion. So, turning to (b), rain may exhibit properties which are modally directed, e.g., the rain is “nurturing” (to plants and animals), which could be said to be modally directed at bio-availability; the rain is “erosive”, which could be understood to be modally directed at eroding material surfaces; or the rain is “obstructive”, modally directed at bad sight, for instance. Similarly, on (c), it could also exhibit just one such modally directed property, ultimately characteristic of its dynamicity, say, e.g., that the rain is “redistributive”, i.e., that it has the property which is directed at displacing large amounts of water. Both (b) and (c) claim that what the rain does, its ongoing dynamicity, lies in some property which is modally directed. While those properties do explain what the rain does, namely the effect that it will have, it hardly seems convincing that they explain why or how the rain does anything at all. That is to say, they do not seem to underpin its dynamicity, its doing something, but only seem to specify that if it were active, what the rain would be doing.25

Second, to see that the modal criterion is not sufficient, take the example of fragility, which is a modally directed property. According to the modal criterion of dynamicity, fragility must be classified as a dynamic entity. Now, suppose we were to ground our process-metaphysical understanding on the modal account of dynamicity. Then, entities which have fragility as a property would qualify as weakly (potentially even strongly) dynamic and, thus, as processes. Take a porcelain vase which is fragile—it, therefore, has the property of fragility. On our current understanding, the porcelain vase is weakly dynamic in virtue of its fragility—but does that seem plausible? The vase itself does not seem to count as a good example of a dynamic entity at all. The only context in which one could think of the vase as dynamic is when it is actually breaking—that is, when the fragility is manifesting. However, the mere property of fragility does not constitute such manifesting—the vase is also fragile when it is not breaking. Thus, something other than the modal directedness involved in fragility underpins the actual dynamicity—the modal directedness by itself is not sufficient for it. As argued in the introduction of the difference-making desideratum, some leeway is allowed for revisionist metaphysics, but this may stretch intuitions too far. Overall, the modal account provides an informative, well-founded distinction, but it is doubtful whether it is the dynamic-static distinction that we are after.

Beyond that, the modal account provides the necessary concepts and interconnections on the basis of which explanations, instead of mere descriptions, can be generated. Powers metaphysicians engage with the sciences and many areas of philosophy seriously. There are interesting debates on the interconnection between powers and the philosophy of time [48,49,50], the nature of laws [51], whether powers can ground modality [52], powers in the philosophy of biology [41], etc. For that reason, the modal account does seem explanatorily powerful. On the other hand, however, there is again the worry that many of these debates will have little to do with dynamicity. Therefore, while power metaphysics may be a fruitful and explanatorily powerful theory, it may not be one that is really targeted at dynamicity.

Finally, the modal account comes at a steep price. Namely, one has to buy into the fundamental reality of local modalities—each of the powers brings with it a possibility (a potency if you will). Neo-Aristotelians will point out that modality plays a fundamental role in their conception of metaphysics (cf. [53]). Then, it does not come at an extra price to understand dynamicity in terms of modality; yet, such ‘metaphysical expenditure’ [1] [Section 5] must be justified to those outside the neo-Aristotelian conception.

The modal account points out something important about dynamicity, which the upcoming section will pick up on, namely directedness. Overall, however, the modal account does not really characterise the nature of dynamicity as much as give rise to a powers metaphysics, which comes apart from dynamic processes (see the examples ‘rain’ and ‘fragility’ discussed before).

5. Dynamicity as Temporal Forward Directedness

Having criticised influential accounts of the nature of dynamicity, this section aims to develop and defend my preferred understanding of dynamicity over these alternatives. I hold that dynamicity is temporal forward-directedness (TFD).26 To unpack this notion, I will begin by (briefly) discussing the interconnection between processes and realist views on time (Section 5.1) and the reality of temporal direction (Section 5.2). Then, the section elaborates on the peculiar directedness of processes (Section 5.3) and how it differs from the directedness of powers discussed in the previous section. The section concludes by defending the (TFD) account by appealing to the desiderata developed above (Section 5.4).

5.1. Dynamicity and Temporality

It is hardly surprising that processes, and thereby their defining feature of dynamicity, are related to time. As explained in the previous sections, process philosophy attempts to provide novel explanations for phenomena of flux. Flux phenomena occur over time; they are not ‘fully’ present at any single moment. Colloquially, they are part of those phenomena that one cannot photograph but would have to film.27 While a photo may suggest that it is raining, it is only in the video that one can see the activity of rain falling. Similarly, the Bénard cell is a structure constituted by movement, yet movement—infamously, as Zeno’s arrow paradox illustrates—is never found at a time point but only across a time span.

This intimate connection between dynamicity and temporality goes deeper than mere intuition and deserves a more explicit exploration.28 I shall argue that the fundamentality of processes, (FP), commits the process philosopher to the reality of time, (RT).29

Consider any fundamental process p. Whatever else it may be, p is a dynamic entity: in virtue of its occurring, one state of affairs gives way to another. Growth transforms a state of being smaller into a state of being larger; motion transforms one spatial configuration into a different one. Every process, thus, exhibits a minimal internal complexity—an initial state, possibly intermediate states, and a final state—and thereby establishes a dynamical order: the ordering of these states as determined by the process itself.

I contend that this dynamical order is, in fact, a temporal order. To see why, suppose it were not. Then, the initial and final states would bear no temporal relation to one another. But then these states contradict one another! If so, we face the classical contradictions of the analysis of change: the growing entity is both smaller and larger at once. The standard strategies for resolving these contradictions—relating incompatible properties to different times—are unavailable precisely because we have denied any temporal relation between the states. Proposing that the states obtain timelessly yields the same contradiction without remedy.

In his influential argument for the unreality of time, McTaggart lends indirect support to this conclusion.

The reality of the A series, then, leads to a contradiction, and must be rejected. Additionally, since we have seen that change and time require the A series, the reality of change and time must be rejected. Therefore, so must the reality of the B series, since that requires time. Nothing is really present, past, or future. Nothing is really earlier or later than anything else or temporally simultaneous with it. Nothing really changes. Additionally, nothing is really in time. Whenever we perceive anything in time which is the only way in which, in our present experience, we do perceive things we are perceiving it more or less as it really is not. The problems connected with this illusory perception will be considered in Book VI [57] [p. 333] (my italics).

In the course of his argument, McTaggart identifies the reality of time with the reality of change: if time is unreal, then ‘nothing really changes’ and all phenomena of flux are illusory [57] [p. 333]. If so, then McTaggart agrees that his view, according to which time is unreal, is actually incompatible with the existence of dynamical entities. And while McTaggart’s argument for the unreality of time has received much critical discussion, it is rather uncontroversial that, if successful, the unreality of time does indeed entail the unreality of change and the illusion of flux. By contraposition, anyone committed to the reality of change (here: the reality of dynamic entities)—as the process philosopher manifestly is—is thereby committed to the reality of time.

So, the dynamical order established by fundamental processes is a temporal order. Additionally, because this order is grounded in real, fundamental processes, it is itself real and objective. Thus: if there exist fundamental processes, then time must be real.

Note that this implication does not depend on a specific understanding of dynamicity. Rather, it entirely relies on the very central thesis of process metaphysics, that some processes are fundamental. If so, then process metaphysics requires the reality of time, regardless of whether the notion of dynamicity involved is also a temporal notion or not.

5.2. On the Direction of Time

Similarly, one can argue that process metaphysics is committed to the reality of the direction of time.30 To see this, one needs again to consider the alternative. Besides the A- and B-theories of time, there are so-called adirectionalist views on time, which hold that time is real (otherwise see the argument above) but not objectively directed, cf. e.g., Farr [58] who has recently built upon ideas of Reichenbach and Boltzmann. On such views, there is no objective, mind-independent direction of time; the apparent asymmetry between past and future is ultimately a matter of convention [58] [p. 11].

My claim that dynamicity is essentially tied to the direction of time stems from the observation that dynamic entities do not merely exist in time but relate to time in a particular manner. They are oriented towards what comes next. This orientation presupposes an objective distinction between what lies ahead and what lies behind.

The realist accounts of time—that is, A- and B-theories—are both committed to the reality of temporal direction.31 On an A-theory, the temporal arrow is pointed towards the future. On a B-theory, the temporal arrow is pointed towards the later. Bracketing the question of whether the A- or B-theory of time is correct, let us simply call the direction of time (whatever it may be) forwards.

My claim, then, is that dynamic entities are forward-directed. Consider, for example, the process of inflating and the process of deflating. I claim that these are two different processes. To say that inflating and deflating (objectively) are two different processes has the consequence that the balloon inflating is different from the balloon deflating. Suppose that process metaphysics holds and that processes are fundamental entities. Then it is a fundamental fact that the balloon is inflating (say) and not deflating. If this is so, then by the very existence of the process of inflating, the temporal direction is fixed—time must run in that direction such that at later times the balloon is larger than at earlier times. The temporal direction cannot be the other way around, for then, in the presence of a process of inflation, later times would find the balloon smaller than at earlier times, which is clearly wrong. If this is the case and the existence of processes is objective, then the temporal direction must also be objective. For if it is objectively the case that the balloon is inflating, time has no freedom left but to go in its (normal) direction.

Adirectionalists of time, like Farr, believe the exact opposite. To them

the description of our universe in the opposite sense of time […] is not describing another universe, or how [our universe] might be but is not, but it is describing the very same thing [59] [p. 327]

Thus, on adirectionalism, there is no objective matter of whether there is a process of inflating or a process of deflating. Instead, there (really) is some undirected happening that can be equivalently described as one or the other, depending on your conventions. These cannot be processes. I would argue that if fundamental processes were these undirected happenings, process metaphysics would lose substantial explanatory purchase on what processes even are. Dynamic entities are not merely extended in time, ordered by a non-directed relation of in-betweenness (cf. [56] [p. 474]), but are going with time. Therefore, it is much more plausible to hold that processes are directed entities and that their direction aligns with time.

5.3. Dynamicity as Temporal Forward Directedness (TFD) Stated

The last ingredient to be added beyond the reality of time and the direction of time is that of directedness.32 Directed entities (like powers) are inherently oriented beyond their present state.33 This directedness finds expression in the fact that it does not suffice to describe such entities merely as they are; rather, in their description one invariably has to include entities which are not yet—their telos. In the directedness of powers, this ‘not yet’ is understood modally—namely, the telos is ‘not yet actualised, but merely possible’. Unlike powers, however, the directedness of dynamic entities is temporally forwards (either spelled out as ‘to the future’ if you are an A-theorist, or ‘to the later’ if you are a B-theorist, cf. subsection before). So, the ‘not yet’ means that the telos is ‘not yet present’ or ‘not contemporaneous with the ongoing process’.

In the case of powers, numerous authors have pointed out that for a power to be modally directed, it is not necessary for the telos to ever actually be realised: ‘a vase has the power of fragility that is directed towards breaking even when the vase never breaks, or an electric charge has the power to attract or repel even if no charged particle ever comes close enough to be attracted or repelled’ [42] [p. 29]. Similarly, the telos of a dynamic entity need not ever actually come into existence. You may pull on a rope, which is directed at moving a boulder attached to it, without the boulder ever moving. Thus, there is pulling directed at future/later moving, but moving never comes about.34

Dynamicity as Temporal forward directedness (TFD) Dynamicity is an inherent ‘drive’ towards the relative future/later. That is, an entity is dynamic iff it is temporally forward-directed.

I claim that this forward-directedness is constitutive of dynamicity. Thus, the dynamicity of pulling is that it aims to bring something about in its relative future, the dynamicity of moving that it brings about new positions, and the dynamicity of growing that it urges to give rise to new size. Processes are what they do (cf. [28] [p. 23], [26] [p. 484])—it is essential to them how they will affect the world around them, an idea which, I believe, is expressed in the conception of their central characteristic, dynamicity, as lying in their forward-directedness.

One might object that (TFD) so characterised lacks genuine content. After all, a Russellian who denies that there is any such thing as flux and who accepts the at-at account of motion—according to which the arrow’s motion simply consists in the arrow being at A at t, at B at $t^{\prime }$, and at all intervening places at intervening times—could seemingly agree that motion is ‘temporally forward-directed’. If so, then (TFD) would be trivially satisfiable even by those who reject dynamicity altogether, and the real explanatory burden would fall on some further, tacitly teleological or powers-theoretic notion.

The crucial rejoinder, however, is that on the Russellian picture, the motion is not a fundamental entity at all: it reduces without remainder to a collection of undirected states obtaining at different times. Each such state—the arrow’s being located at a given point—is in itself entirely non-directed; there is nothing intrinsic to any individual state that privileges one temporal orientation over another. Precisely this is what makes the ‘anti-direction’ redescription available: the same states could, in principle, feature as the reductive base of a motion in the opposite direction, for nothing in the states themselves determines a ‘from’ or a ‘to’. (TFD), by contrast, does not merely ascribe temporal forward-directedness to a pattern of undirected states. Rather, it holds that dynamicity is temporal forward-directedness as a constitutive feature of a fundamental entity.

The process of the arrow’s flight, on this view, is irreducible to states-at-times; it is a genuinely directed particular whose orientation towards its relative future is part of what makes it the very entity it is. A Russellian can borrow the language of forward-directedness but not its ontological substance, for what (TFD) demands is that the directedness be primitive and ineliminable—not a mere redescription of an underlying directionless mosaic.35

5.4. Dynamicity as Temporal Forward Directedness (TFD) Defended

I hold that understanding dynamicity as (TFD) surpasses its alternatives—here, the mereological and modal accounts—on all three criteria developed earlier. It makes better sense of the difference between static and dynamic entities, is explanatorily more powerful, and is more parsimonious than its competitors.

Starting with the first desideratum: difference-making. On the (TFD) account, static entities are, by negation of (TFD), either (a) undirected, (b) directed but not temporally directed, or (c) temporally backwards-directed. In the discussion of the modal account, it was already shown how, for example, the macroscopic Bénard cell qualifies as something undirected (a) and, therefore, static—which is exactly the desired result. Similarly, states—a’s being F—are undirected entities.

The crucial test case, however, is (b)—entities that are directed but not temporally directed. It is here that the (TFD) account most clearly parts company with the modal account discussed in the previous section. On the modal account, an entity qualifies as dynamic in virtue of it (or some property of it) being directed modally. Powers such as fragility or electrical charge are paradigmatic: they point beyond their current state towards their possible manifestation. Yet modal directedness alone does not establish dynamicity in the sense relevant to process metaphysics. A latent disposition—fragility in an undisturbed vase, say—is modally directed at breaking, but it is not temporally forward-directed. It does not, as it were, drive towards bringing about a future state; it merely encodes a modal profile, a space of possible manifestations. On the (TFD) account, such latent dispositions are accordingly classified as static, for their directedness is exhausted by modal orientation and lacks the temporal dimension constitutive of genuine dynamicity.

That this distinction is not merely an artefact of the present framework finds independent support from within powers metaphysics itself. Donati and Gozzano, for instance, two of the authors who explicitly defend the thesis that dispositional properties are dynamic [44] [p. 15]—do so precisely by supplementing modal directedness with temporal forward-directedness. Their thought experiment involving ‘pastgility’—a hypothetical property structurally identical to fragility save that its manifestation is temporally backwards-directed—is instructive. As they argue, temporal orientation is part of what individuates a disposition as the very disposition it is: fragility is essentially future-oriented, and pastgility, being backward-looking, fails to qualify as a genuine disposition at all [44] [p. 23]. Accordingly, they conclude that dispositional dynamism requires not merely modal directedness but, in addition, (i) temporal extension, (ii) necessary change, and (iii) future orientedness [44] [p. 24]. This amounts to a significant concession from within the powers camp: even those who locate dynamicity squarely in the dispositional nature of properties find themselves compelled to invoke precisely the kind of temporal forward-directedness that (TFD) places at the heart of its analysis. The modal criterion, taken on its own, proves insufficient.

The (TFD) account, however, goes beyond what even this enriched powers account offers, for it can explain why fragility—at least so long as the vase is not actively breaking—is a static property, namely in the sense of (b). Only when the hammer hits the vase and the power is ‘activated’ does it act dynamically—and (TFD) makes perfect sense of this, for it is precisely then that the fragility drives towards bringing about a broken vase in time. Yet fragility without activation is modally directed only at the breaking of the vase, which, according to (TFD), renders it static.36

Lastly, entities like memories may be temporally directed, but backwards (c). If so, then memories qualify as static entities on the (TFD) account. I believe that this is the desired outcome. Memories, in being backwards-directed, do not bring about anything; they are not actively doing something. One may believe that this has little to do with their direction and more to do with memories being potentially only mental entities. After all, plans, hopes, and dreams can be temporally forward-directed entities without them doing anything either. In contrast, I would hold that real plans (hopes/dreams), that is, plans in action, are doing something. How do we evaluate whether they are doing something? By evaluating whether they will bring about new behaviours in the future. In contrast, memories can be as real, truthful, and authentic as they come; they are never doing or bringing about anything, simply because they are directed backwards. Therefore, I believe that (TFD) substantiates the distinction between dynamic and static entities in a very fruitful way.

(TFD) enables genuine, explanatory claims about dynamic phenomena beyond mere redescription. Consider, for example, the question of why certain processes can be interrupted or accelerated, whereas others cannot. On the (TFD) account, this is explained by the nature of the telos towards which the process is directed. A process of building a house can be interrupted because the telos—the built house—remains unrealised. To stop (interrupt) building means to halt the activity—means temporarily to refrain from bringing about the future built house. Yet one can resume, for once one begins again the activity aimed at bringing about that precise telos, one resumes the very same activity. By contrast, an activity like walking has no such telos that can be picked up later; to stop walking is simply to cease the process, not to interrupt it. In this way, (TFD) satisfies aspect (i) of explanatory power by providing the conceptual resources for substantive metaphysical explanation.

Turning to aspect (ii), (TFD) is broadly applicable across domains. In physics, processes such as radioactive decay or field excitations can be characterised as temporally forward-directed: decay is directed at the emission of particles, excitation at the propagation of energy. In biology, metabolic processes are directed at the maintenance of organismic structure (that is, the future recurrence of the same structure). In the cognitive sciences, processes of attention, reasoning, and perception could plausibly be understood as directed at their characteristic outcomes—focus, inference, and representation. Thus, (TFD) applies with equal force across various domains that process metaphysics aspires to illuminate, satisfying aspect (ii).

Since (TFD) infuses the process with a telos, it allows for a rich typology of processes (the one presented here is inspired by Seibt, cf. for example, ref. [23] [p. 8]), namely, with regard to what the telos is. Processes can be autotelic, that is, directed at future/later instances of themselves, the classical activities (e.g., raining is directed at further raining, walking at further walking). They can be allotelic, that is, directed at entities other than themselves, typically developments (e.g., building a house is directed at the built house, running a race directed at crossing the finish line, etc.). This undergirds well-established differentiations in the process literature. Thus, the understanding of dynamicity in light of (TFD) checks all three aspects of explanatory power.

Since I hold that process metaphysics is committed to the reality of time and its direction, the (TFD) account does not introduce novel notions that other process metaphysical theories would not equally require. In that sense, it is perfectly parsimonious. The account introduces nothing beyond that to which the process metaphysician is already committed, regardless of their preferred theory of dynamicity. Then, (TFD) comes at no additional cost, for anyone drawn to process metaphysics must already have priced in temporal realities. Ultimately, (TFD) exceeds expectations for a theory of dynamicity—it illuminates what it means to be dynamic, it undergirds explanations building on the dynamic, and comes at no additional cost—which, I believe, constitutes a very strong case in its favour.

6. Conclusions

This paper set out to examine critically the central thesis of process metaphysics: that (some) processes are fundamental (FP). I argued that such a thesis is worthwhile only if accompanied by an understanding of what processes are in contrast to other entities. Since processes are primarily characterised by dynamicity, this raises the question of what dynamicity is. The process metaphysical literature has not settled on an agreed-upon characterisation, and indeed, understanding dynamicity remains one of the paradigm’s central challenges (cf. [1] [section 5]).

The paper attempted to make progress on this challenge in three steps. First, I articulated what is required of an account of dynamicity, arguing for two crucial desiderata: difference-making and explanatory power. Second, I criticised two influential accounts of dynamicity. The mereological account, I argued, proves unsatisfactory because it cannot provide (proper) grounds for distinguishing dynamic from static entities. The modal account, by contrast, does draw a distinction—but not between dynamic and static entities in the required sense of process metaphysics. Third, I developed and defended the account of dynamicity as temporal forward-directedness, arguing that it outperforms its competitors on the basis of these desiderata.

While these three steps hang together, each can hopefully contribute independently to making progress on the challenge. Readers unconvinced by my preferred account may nonetheless find value in the desiderata or in the arguments against the alternatives. Furthermore, I hope the arguments offered here will help connect process metaphysics more closely to adjacent areas—particularly the metaphysics of time and the powers literature. I believe that these interconnections will prove fruitful for further research and benefit process metaphysics to find broader acknowledgement.