Diachronic Constitution

Abstract

It is often argued that constitution and causation are different kinds of dependence relations. Some have argued for a distinction between constitutive explanation of causal capacities that explain what a system would do in specific situations from causal or etiological explanations that explain why an event such as a change in the property of a system happened. In what follows we argue against the claim that causation and constitution are always distinct metaphysical relations. This paper develops a temporal account of constitution. We call this species of constitution, diachronic constitution. We show how diachronic constitution is a consequence of a common type of causation in the empirical sciences: continuous reciprocal causation, a variety of causal production instantiated in complex dynamical systems. Hence, this paper seeks to establish that constitution does not only resemble causation in certain respects. We argue for the stronger claim that constitution can be analysed in terms of causation understood as production of change. Temporalizing the constitution relation is neither as remarkable nor as problematic as it might initially seem. The idea of diachronic constitution will appear almost inevitable in the context of complex dynamical systems, given local interactions between microscale and macroscale states in such systems.

Keywords:
Diachronic Constitution; Material Constitution; Mechanistic Constitution; Metaphysical Dependence; Reciprocal Causation; Dynamical Systems

1. Introduction

A standard assumption in analytic metaphysics is that grounding relations such as constitution (or composition) are distinct from causation. Some have distinguished between (i) constitutive explanation of causal capacities that explain what a system would do in specific situations and (ii) causal or etiological explanations that explain why an event such as a change in the property of a system happened (Salmon 1984; Cummins 1975, 1983; Craver 2007; Ylikoski 2013). Others have argued for a distinction between grounding and causation. Grounding is the relation that connects little things, at lower fundamental levels of reality, to the bigger things they make-up synchronically at a moment in time. Causation, by contrast, is argued to be a distinct kind of relation from constitution, in part because it unfolds over time.¹ The diachronicity of causation and the synchronicity of grounding has been argued to point to “deep metaphysical differences” between grounding and causation (Bernstein 2016). The notion of grounding is admittedly controversial (Wilson 2014). However, let us assume that grounding exists², and constitution (along with composition, realisation and supervenience) are instances of grounding relations.³ Any argument for distinguishing grounding from causation will entail distinguishing constitution (as a species of grounding) from causation.

In what follows we argue against the claim that causation and constitution are always distinct metaphysical relations. We will be concerned with the question of how smaller things produce or make up larger things. Schaffer (2016) points out: “it is apt to use causative verbs like ‘making’ in glossing grounding relations of which constitution is an instance. Likewise, it is apt to invoke general notions like ‘dependence’ in glossing both causal and grounding relations.” (2016, p. 54) Constitution (in common with grounding) also shares key features with causation regarding explanation (Dasgupta 2017). We might ask for instance: Why is the water boiling? One answer might be that I am heating the water to steam some vegetables. A different answer would have to do with thermal instability in the fluid dynamics. Dasgupta (2017) notes that the first kind of answer is a causal answer, in the sense that it tracks a causal history of events. The second answer is an example of constitutive explanation: it explains why the molecular base of the water makes it the case that water has reached a boiling point.

We restrict our discussion of causation and constitution to the production of ordered, coherent and coordinated behaviours in complex dynamical systems. We use the term complex dynamical systems to refer to systems that exhibit interdependent changes over time in both their constituent parts, and in the system’s large-scale, macroscopic organization (Prigogine & Stengers 1984; Kelso 1995; Haken 1996; Clark 1997; Cilliers 1998; Juarrero 1999, 2023). Complex dynamical systems cover a very broad range of systems from tornados to fires to nervous systems to societies. There is much more to be said about how to define complex dynamical systems (see Ladyman & Wiesner 2020), but this informal gloss will serve our purposes for now. In these systems, constitution refers to an intrinsically temporal relation of dependence between macroscopic structures and the microscopic elements that compose the system. Micro- and macroscale dynamics are distinguished by the timescales over which their dynamics unfold, in the sense that microscopic dynamics are fast while change in macroscopic behaviours is, by comparison, slower. Crucially, a diachronic view of constitution makes it possible to account for how processes at the microscopic scale of a system can make up or produce macroscopic processes at the organisational scale of the whole dynamical system.⁴

We will argue that (1) the concept of continuous and reciprocal causality (Clark 1997; Wheeler 2005) can be used to analyse the constitution of dynamical systems. It follows from (1) that the constitution of dynamical systems is (2) diachronic, not synchronic. We will argue, based on (1) and (2), that (3) constitution and causation cannot always be distinguished. We should instead conceive of constitution as a diachronic relation - a process of continuous reciprocal causation - at least for complex dynamical systems. We will therefore challenge the widely defended view that constitution (and grounding more generally) is a synchronic relation between levels of reality, whereas causation is a diachronic relation between things (events, processes, etc.) over time (see, e.g. Salmon 1984; Craver 2007; Craver & Bechtel 2007; Ylikoski 2013; and see Leuridan & Lodewyckx 2021 for a dissenting voice). We will argue by contrast that when the concepts of constitution and causation are applied in the context of complex dynamical systems, the constitution relation is revealed to be a special kind of causal relation.

The remainder of our paper is divided into four sections. Section 1 introduces one of the central concepts of our paper: continuous reciprocal causation. We start from an understanding of causation as production (Salmon 1984; Ingthorrson 2002, 2024; Bogen 2008). Such a conception of causation already brings causation and constitution much closer together. Section 2 turns to the discussion of constitution in metaphysics. There we see how constitution is standardly analysed based on two assumptions that: constitution is relation that holds between an object and its material substrate at an instance time; and constitution is a relation of asymmetric dependence that allows constitution to be distinguished from other metaphysical relations such as coincidence or identity. Section 3 explains how these two assumptions make good sense when the concept of constitution is applied to material objects. In section 4 we show how the three assumptions do not apply to the constitution of dynamical systems. We make the further argument that material constitution is not the right metaphysical relation for understanding how such systems become organised over time through the interaction of their parts. Section 4 concludes our paper and defends our key claim that continuous reciprocal causation accounts for the constitution of ordered macroscopic pattern of behaviour in dynamical systems.

2. Continuous Reciprocal Causality

We use the term continuous and reciprocal causation to refer to a relation of causal production between the constituent parts of a dynamical system and the coherent dynamics of this system as a whole.⁵ Consider the example of a crowd behaviour to illustrate this conception of causality (Clark 1997, p. 107-8). Suppose there is a sudden loud noise in the middle of a busy square in a city. In response to the sound, the individuals in the crowd all suddenly rush in a particular direction to safely escape what they fear could be an explosion. The direction of the movement of the crowd behaves as a collective variable that acts as a constraint on the behaviour of the individuals. At the same time, the emergence of this collective variable depends on the movements of the individuals that make up the crowd and their interactions. The causal interaction of the individuals is productive of the coordinated ordered patterned behaviour of the crowd - a collective property of the group that in turn constrains the behaviours of the individuals that make up the crowd. We will henceforth reserve the term ‘continuous reciprocal causality’ to refer to the production of coherence and order in dynamical systems at macroscopic levels of organisation out of the interaction of the microscopic, constituent elements of the system.

The causal relation between the movement of the crowd as a collective pattern of coordinated behaviour, and the movement of each of the individuals that make up the crowd is continuous and reciprocal. We describe the causal production of this coordinated collective behaviour as ‘continuous’ because the relata of this type of causal relation are processes that possess intrinsically temporal properties. The causal production of ordered behaviour is best described using terms such as ‘phases’, ‘frequencies’, ‘oscillations’, ‘historical path dependence’, ‘periodicities’ and ‘cycling rates’ that span a period of time and unfold at different speeds. We describe this type of causality as ‘reciprocal’ because of the mutual, bidirectional relation of influence that holds between the parts of the dynamical system and the coherent dynamics of the system considered as a whole. In the example of crowd behaviours, the coordinated pattern of movement is the product of the interaction of the individuals that make up the crowd. At the same time, this coordinated pattern of movement behaves as what Alicia Juarrero (2023) has called an ‘enabling constraint’ that loops back down to lock together the behaviour of what would otherwise be independent elements, caught up in the coordinated collective behaviour. The result of this reciprocal bidirectional causal influence is that coherent patterns of interdependence temporarily form, dissolve and reform that hold together the elements of a dynamical system.

In the metaphysics of causation, the production of change is often understood on the model of efficient causation. Consider Salmon’s (1984) example of a baseball that breaks a window. The baseball is struck by the bat with a certain force and momentum that is conserved and propagated as the ball crashes into the window. It is the conservation of the ball’s momentum that produces the broken window. Production here is understood on an agent-patient model where the baseball is the agent that, when it strikes the window with a certain force and speed, produces the breaking of the glass. The effect - the broken window - is the change produced by the cause. Such an analysis of production in terms of efficient causation implies a clear distinction between constitution and causation. Given such an understanding of production, we can distinguish phenomena that cause the breaking of the glass - for example, the baseball’s force and momentum - and phenomena that are constitutive of the baseball’s causal powers - for example, its solidity and mass and the microstructural properties that underlie these properties. We can then explain the baseball’s causal powers mechanistically in terms of the parts of the baseball, their activities, and the organisation of those parts.

Efficient causation is not the right model for understanding production of order and coherence in complex dynamical systems. First, note that if production is understood on the model of efficient causation, then it is a unidirectional agent-patient relation whereby an external cause produces change in an object, property, event or fact that is the passive recipient of this change. On this understanding of causation, to say an object (property, event or fact) C is the cause of an object (property, event or fact) E is to say that E’s occurrence depends on C such that if C had not occurred, E would not have occurred (Lewis 1973). We can think of causation on this model in terms of ‘difference making’ where a cause C is what makes the difference to E’s occurrence. This difference-making conception of causation is unidirectional. But as Ingthorsson (2002, 2024) has noted, such a unidirectional view of causation does not square with classical physics in which change is described as reciprocal. When the baseball shatters the glass in the window, the baseball is also stopped in its tracks and falls to the ground. Newton’s third law of motion states that whenever an entity brings about a change in an object, the object will exert an influence of the same magnitude but in the opposite direction (Ingthorsson 2024, p. 130). Ingthorsson has characterised this reciprocal action described by Newton’s third law of motion in terms of a single process of mutual influence between otherwise distinct objects.⁶ There are, he argues, no patients in a causal relation that undergo change but do not themselves exert any causal influence. Change in a causal interaction is always bidirectional and mutual in the sense that entities involved in an interaction influence each other simultaneously and to the same magnitude (Ingthorsson 2024, p. 136).

In cases of continuous reciprocal causation, we see this same symmetrical and bidirectional relation of influence. However, the relation of mutual influence is different from that observed in instances of Newton’s third law. Enabling constraints exist at the spatial and temporal scale of the whole system, linking and coupling otherwise distinct and separately behaving elements of the system. In our earlier example, the coordinated crowd behaviour is produced by the interaction of the individuals that make up the crowd. However, the crowd behaviour also loops back down to exert an influence on how each of the individuals in the crowd behaves. Thus, the relationship between the crowd and the individuals that make it up is bidirectional and mutual. It would be incorrect to characterise this relationship in terms of individual agents and patients. The individuals in the crowd do not become patients when they begin to move as one because of the coherent and coordinated behaviour of the crowd. They remain agent’s that contribute to the dynamic of the crowd behaviour. Nor would it be correct to characterise the crowd as the patient in this interaction. To do so would miss how the crowd acts as a collective variable influencing the behaviour of the individuals in such a way that they begin to behave as an interdependent group and no longer as separate and independent individuals.

Efficient causation holds between relata that are wholly distinct in the sense that they do not overlap spatially and temporally. Lewis (2000) argues that “C and E must be distinct events - and distinct not only in the sense of non-identity but also in the sense of non-overlap and non-implication” (cited by Craver & Bechtel 2007, p. 552). We call the claim that production is a relation between wholly distinct events the ‘distinctness assumption’. The distinctness assumption coheres naturally with the view of causation we have argued against as a relation between an agent that acts as a cause to produce change in an otherwise inert patient. The cause is here construed as an event which is productive of change, while the effect is thought of as the distinct event which the cause is responsible for producing. We have argued against an understanding of the production of coherence and order in dynamical systems on the grounds that causal influence in such systems is bidirectional, symmetric and mutual.

It is worth briefly noting that not all philosophers writing about causation have endorsed the distinctness assumption. Friend (2019), for instance, explicitly argues that causes and effects are not wholly distinct existents. He defends the part-whole causation thesis that certain kinds of causes can be a part of their effects (e.g., a population of cancer cells causes a person to have cancer). Huemer & Kovitz (2003) have also challenged the distinctness assumptions, developing what they call the simultaneous theory of causation, which holds that “causes always occur simultaneously with their immediate effects”. They contrast the simultaneous theory of causation with the more standard sequential theory, which they trace back to Hume. The sequential theory claims that causes are temporally discrete instantaneous events that precede their effects. Causal laws, Huemer and Kovitz contend, typically have the form: “Temporally extended action e occurs simultaneously with temporally extended cause c.” (2003, p. 556) The simultaneous theory of causation shares with our notion of continuous reciprocal causality a view of causally relevant factors as changing continuously at different rates over a given time interval based on the interaction of these factors. The simultaneous theory of causation calls into question the distinctness assumption in much the same way as we will do later in this paper.⁷

It may be objected that continuous and reciprocal causation does not by itself establish the non-distinctness of cause and effect. Recall the earlier quote from Lewis, which we used to illustrate the distinctness assumption. Lewis wrote:

“C and E must be distinct events - and distinct not only in the sense of nonidentity but also in the sense of nonoverlap and nonimplication.” (2000, p. 78; quote in Friend 2019, p. 5071).

Lewis specifies three conditions for ontological distinctness of cause and effect: non-identity, non-implication, and non-overlap. Let's start with the non-overlap condition. We interpret this to mean that cause and effect do not share parts in the sense that none of the parts of the events of cause and effect occupy the same spatial and temporal locations. This condition is clearly not satisfied in the crowd example in which the interactions between individuals give rise to coherent patterns of crowd behaviour that in turn exert an influence on the behaviour of individuals. For causal influence to be continuous and reciprocal, there must be spatial and temporal overlap between the individuals and the crowd they compose. We set aside the question of whether the other two conditions are satisfied (we think they are), since spatial and temporal overlap is already sufficient to establish the non-distinctness of cause and effect in continuous reciprocal causation.

We saw earlier that in cases of efficient causation it is possible to make a sharp distinction between causal explanation that might answer questions such as why did the window break, and constitutive explanations that account for an object’s causal capacities - in this case the power of the baseball to break the window. We will argue that if order and coherence is constituted in dynamical systems through continuous reciprocal causal interactions, such a distinction between causation and constitution breaks down. Before we make this argument, we consider first how philosophers have tended to understand constitution by starting from examples of material objects. We will then set about showing that the standard analyses of what we will call ‘material constitution’ fail when applied to complex dynamical systems.

3. The Metaphysics of Constitution

Many philosophers have approached questions about the nature of constitution by reflecting upon objects such as Michelangelo’s statue David and the marble Piece from which the statue was sculpted, or pieces of paper and dollar bills, or human bodies and human persons.⁸ The constitution relation is typically argued to be distinct from a relation of identity. Wasserman (2004), for instance, argues that David and Piece are not identical since they differ in their de re modal properties. Piece cannot survive losing any of its parts. David however can lose an arm and continue to exist. Moreover, Piece exists prior to David in the sense that Piece would have existed as a portion of matter. Wasserman (2004) provides two conditions on material constitution that are important for the discussion to come. The first condition he identifies is what he calls ‘spatial coincidence’: “x constitutes y at t only if x and y have the same spatial location at t” (p.694). The second condition he describes he calls ‘material coincidence’: “x constitutes y at t only if x and y have all the same parts at t” (p.694).

What we wish to highlight about these two conditions is that Wasserman presents both conditions as obtaining at a temporal location t. These coincidence relations imply that material constitution is a relation that holds synchronically at a point or interval in time. We will call this the “synchronic assumption”. It claims that the constitution relation holds between an object x and the material substrate it is made from at a temporal instant or interval t. If the synchronic assumption holds, an argument can be made for distinguishing causation from constitution. If causation picks out a relation between wholly distinct events, as many have argued, there can be no spatial and temporal overlap between causes and effects.⁹ We have of course argued against such a view of causation in the previous section.

The synchronic assumption can also be challenged. Kirchhoff (2015) argued for a diachronic understanding of constitution. However, he keeps in play a distinction between causation and constitution. We will argue by contrast that constitution can and should, in the right kinds of circumstances, be understood as a causal relation. Leuridan and Lodewyckx (2021) (building on Kirchhoff 2015) argue for a very similar position to the one we defend in this paper. They argue that the relata of the constitution relation are intrinsically temporal processes. They use the mutual manipulation criteria, as proposed by Craver (2007)¹⁰, to determine whether a process is constitutively relevant to the behaviour of a whole mechanism. They reject the distinction between interlevel and etiological or causal interventions and argue on these grounds that constitution is both diachronic and causal, whilst also leaving the door open to constitution in some cases being synchronic and non-causal.¹¹

To distinguish material constitution from mere coincidence, it is argued that material constitution is an irreflexive and asymmetric relation, whereas coincidence is both symmetric and reflexive (Wasserman 2004, p. 694; italics in original).¹² Material constitution is held to be an irreflexive relation insofar as no object can be said to constitutes itself. Material constitution is an asymmetric relation because while Piece may constitute David the reverse is not the case. The David statue does not constitute Piece - the existence of the latter predates that of David. We will call the claim that the parts constitute the whole, but the whole does not constitute the parts, the ‘asymmetry assumption’.

Barnes (2018) has questioned the asymmetry assumption in making the case for what she calls ‘symmetric dependence’. One example she provides is the relation between WWII and the evacuation of Dunkirk. It seems plausible that WWII would not have been the same war without the evacuation of Dunkirk. Similarly, the evacuation of Dunkirk would not have been what it was unless it were part of WWII. That is, what it is to be the evacuation of Dunkirk is to be part of WWII. Of course, this will ultimately depend on one’s account of the individuation of events. However, at first glance, it seems a strong case can be made that the part-whole constitution relation for events is an example of symmetric dependence. If Barnes is correct, ontological dependence relations need not conform to the asymmetry assumption: some cases of ontological dependence may be asymmetric, others symmetric.

Material constitution is thus standardly conceptualised based on two assumptions: the synchronic and asymmetry assumptions. Thus Ylikoski (2013) writes: “Constitution is a synchronous and asymmetric relation between relata that cannot be conceived as independent existences”. He goes on to add “causation and constitution are not to be confused. Causal and constitutive explanations have different explananda, and they track different kinds of dependency” (2013, p. 2). We will argue in the next section that both the synchronic and the asymmetric assumptions fail in the case of the constitution of coherence and order in dynamical systems. We contend that the constitution relation between parts and whole is a reciprocal causal relation that unfolds continuously in real time. To describe change as continuous at macro and micro-scales is to say that mathematically change should be understood as having the structure of a continuum. Here we agree with Huemer & Kovitz (2003) when they write: “On this understanding of time and change, temporally extended events are not conceived as being ‘built up’ from some smallest units. Rather, every event is already a temporally extended whole, which can be divided into indefinitely many parts, each of which is itself a temporally extended event.” (Huemer & Kovitz 2003, p. 560)

4. Why material constitution does not fit self-organising dynamical systems

We will make use of another common example of a dynamical system in this section: the example of the Rayleigh-Bénard convection system. Bénard cells form when fluid is heated between two planes in a gravitational field. The formation of Bénard cells depends on several things such as the type of fluid, its depth, and the temperature gradient. The latter is central. If the temperature gradient is below a certain value (a function of the Rayleigh number), then the fluid will remain stable despite its natural tendency to move given its viscosity and thermal diffusivity. However, when the temperature gradient exceeds a certain critical value, then thermal instability occurs. As Chemero and Silberstein (2008) put it: as the temperature gradient reaches its critical value, “there is a breakup of the stable conductive state and large-scale rotating structures resembling a series of parallel cylinders called Bénard cells are eventually produced.” (2008, p. 20) All of this is to say that one starts to see fluctuations in the density of the fluid given a specific temperature threshold from which large-scale structures start to arise - a convection roll.¹³

The Rayleigh-Bénard convection system exemplifies what we have been calling continuous reciprocal causality between dynamics at different timescales of activity. The dynamics at the macroscale of the Bénard cell unfolds over slower timescales than dynamics at the microscale.¹⁴ The macroscale dynamics of the system - convection or conduction - is therefore a consequence of the behaviour of the microscopic parts of the system; it depends for instance on the dissipation of energy by these parts. At the same time, the macroscale dynamics of the system constrains and enslaves the microscale dynamics limiting the degrees of freedom of the molecules of which the Bénard cell is composed.

We will argue next that the metaphysical relation of material constitution fails to apply to the constitution of Bénard cells. Consider first the synchronic assumption. It is not plausible to ask whether the formation of Bénard cells is such that its constitutive elements constitute the ensemble behaviour (the observed rolling motion of the liquid) at an atemporal instant t. This is because the formation of Bénard cells is inherently temporal, or processual. At the spatial scale of molecular dynamics, time is continuous, in the precise sense that temporality cannot be broken down into discrete quanta (Huemer & Kovitz 2003).

If one were to claim that the system in question is in a ‘state’, X, at a particular point in time, this would be an approximation - it would “boil down to saying that the average of the system’s states during that period of time was X.” (Spivey 2007, p. 30) Processes can be described at some synchronic point in time only by taking an average of the states the process goes through over time. To see this more clearly, consider what Ladyman & Ross (2007) say about another example of chemical constitution: the relation between water and H₂0 molecules. Under the synchronic model of constitution, if H₂O composes or constitutes water at some point in time t, or at each temporal stage over t ₁ … t _n , then both the hydrogen molecules and the oxygen molecule must be entirely present at that point in time or at each moment over t ₁ … t _n . According to Ladyman & Ross, this synchronic view is however mistaken. Water is constituted “by oxygen and hydrogen in various polymeric forms, such as (H₂O)₂, (H₂O)₃), and so on, that are constantly forming, dissipating, and reforming over short time periods in such a way as to give rise to the familiar properties of the macroscopic kind water.” (2007, p. 21; italics added) Because water is constituted in a complex dynamical system it “makes no sense to imagine it having its familiar properties synchronically.” (Ross & Ladyman 2010, p. 160; italics added) To claim that a dynamical process has the properties it has at some point in time is an abstraction; it is an abstraction that should not be mistaken for how the properties of dynamical processes manifest themselves over time. In other words, from the fact that a system can be described synchronically, it does not follow that it has any of its special characteristics synchronically.

Ylikoski (2013) suggests that questions of constitution abstract away from such details. He says:

The constitutive questions abstract away from the behavior and orchestrated

activities of the parts, and ask how the system has a capacity for this kind of

behavior ... One could say that the question only addresses a synchronous time-slice of the system. Of course, this is a heavy abstraction, but it helps in articulating an important explanatory question. (2013, p. 4; italics added)

We agree with Ylikoski that it can be unproblematic to say that a system is in a particular state, X at time t, if one is making such a claim for epistemic, explanatory purposes to measure the approximate values of a system at some measured point in time. Such a measure may well answer “important explanatory questions” (Ylikoski 2013, p. 4). It is equivalent to saying that a system’s states over some periods of time were on average, X. There is no reason to think that this epistemic claim is unhelpful. As Spivey (2007) reminds us: “This kind of coarse averaging measurement is often a practical necessity in science but should not be mistaken as genuine evidence for the system actually resting in a discrete stable state.” (2007, p. 30)

However, such talk of synchronic states of a dynamical system becomes a problem if one takes such an epistemic claim to imply that the microscale processes that constitute the dynamical structure of interest do this constituting work instantaneously at some discrete point in time. Water is not constituted by H₂O at any given instant in time t. It is diachronically constituted by oxygen and hydrogen forming different polymeric forms over time. Hence, the claim that talk of constitution is a form of abstract explaining of a system at a synchronic instant t is only meaningful if taken as an epistemic principle of approximation.

Unlike the material notion of constitution, which is best-suited to an substance-based ontology, the rolling motion of Bénard cells, and the constitution of order and coordinated behaviour in dynamical systems more generally, are best captured by a diachronic view of constitution, within a broadly speaking process-based ontology (Seibt 2009; Hofweber & Velleman 2011; Rock 2021; Nicholson & Dupre 2018).¹⁵ Process-based ontologies are proposed as an alternative to substance ontologies that start from self-subsisting entities and their properties. Process-based ontologies highlight an important shift away from focusing on the material and spatial composition of the whole by its parts in synchronic constitution relations. Instead, a process-based ontology aims to make sense of practices of individuating dynamical systems in terms of their behaviour over time in particular contexts. Diachronic constitution therefore shifts our focus away from material elements and their co-location (without denying that this is a part of constitution) to a perspective of dynamical functioning of whole complex non-linear systems.¹⁶

We turn next to the applicability of the asymmetry assumption to dynamical systems. At first glance it may seem like this assumption is satisfied by our example of the rolling motion of Bénard cells. The molecules of which the fluid is composed are prior to, or more fundamental than, the rolling motion. The rolling motion is the constructed; the non-fundamental. In contrast, the molecules are the elements that construct; they are the fundamentals. As Barnes (2018) observes: “Dependence is the kind of relation that explains the connection between the fundamental and the derivative - it takes us from the derivative (the dependent) to the fundamental (the independent). Any relation that plays this role must be asymmetric. And so dependence must be asymmetric.” (2018, pp. 54-55)

This view of dependence is however problematic in a range of dynamical systems such as the Rayleigh-Bénard convection system. Note that once formed, the Bénard cell modifies the configurational degrees of freedom of the molecules such that some possible motions are no longer available (Chemero & Silberstein 2008). If we look at the system over time, the relation between the individual molecules and the macroscopic structures they form is not one of asymmetric dependence but is more akin to what Barnes (2018) calls symmetric dependence. The macroscopic dynamics depend on the microscopic dynamics, and vice-versa, and this dependence occurs simultaneously between the microscopic and the macroscopic over time. Macroscale dynamics are constituted out of microscale dynamics that are in turn, in a circular fashion, constrained and enslaved by the very same macroscale dynamics they constitute. Now, one might object to this that the macroscopic behaviour of the system is clearly asymmetrically dependent on the individual molecules. We agree. But this is only the case if considered atemporally. Allow time to unfold and the micro- and macro-scopic dynamics will interlock.

We can therefore conclude that material and diachronic constitution are fundamentally different metaphysical relations. We summarise the differences in table 2.

Moreover, causation and constitution need not be thought of as distinct metaphysical relations since the argument for such a conclusion fail when considering the constitution of dynamical systems. Diachronic constitution tracks continuous and reciprocal causal dependence in dynamical systems over time. Constitution in non-linear dynamical systems just is a kind of causation: continuous reciprocal causation.

5. Concluding remarks

We have provided an analysis of the constitution of order and coherence in complex dynamical systems in terms of continuous reciprocal causation. Causation is often taken to be a relation between wholly distinct events. We have argued that cause and effect in dynamical systems overlap in space and time and stand in complex circular relationship such that the distinctness assumption fails. Constitution in turn has typically been analysed based on two assumptions we have called the synchronic and asymmetry assumptions. The synchronic assumption claims that constitution is a synchronic relation that obtains at an instance in time. Constitution is typically distinguished from identity and from mere coincidence on the grounds that it is a relation of asymmetric dependence unlike these other species of metaphysical relation. We have argued that both assumptions fail to apply to complex dynamical systems, and therefore so also does the notion of material constitution. The relation of dependence between the microscopic parts of a dynamical system and its macroscopic properties as a whole system is symmetrical. This symmetry follows from the enslaving relation that holds between microscopic and macroscopic properties. To explain the constitution of macroscopic properties in such systems we have argued constitution is best understood diachronically in terms of continuous reciprocal causation.

Why think that continuous reciprocal causation is a diachronic form of constitution? We provide four reasons:

1. Continuous reciprocal causality is productive and explanatory. Production plays a core role in constitution, as it signifies ‘bringing something into existence’ (even synchronically as in the case of Piece and David). In a similar fashion, continuous reciprocal causation explains how microscopic behaviour makes up the macroscopic behaviour of the system under investigation, and through enslavement, vice versa.

2. In cases of material constitution, the relata of the constituting relation stand in a relation of spatial and material coincidence of microscopic and macroscopic activity. These relata are taken to endure in the sense of being wholly present at distinct instances in time. In cases of continuous reciprocal causation, the macroscopic and the microscopic co-evolve simultaneously over different timescales. Thus, they exhibit spatial and temporal overlap along similar lines to what you observe in cases of material constitution by virtue of the coincidence relations. However, insofar as the microscopic elements are processual, they do not endure, they perdure (Hofweber & Velleman 2011). This is to say, that each microscopic element as a process is not wholly present at different instances in time. Moreover, the work that the microscopic elements perform in constituting some macroscopic phenomenon of interest is also performed continuously over time.

3. Diachronic constitution is a relation of non-identity. It is typically assumed that if X (or the Xs) constitutes Y, and if X (or the Xs) are not identical with Y, then the constitution relation between X (or the Xs) and Y is asymmetrical. Diachronic constitution is not an asymmetric relation of dependence. Yet, this does not imply that it is an identity relation, for micro- and macroscale dynamics unfold over different time-scales.

4. Diachronic constitution can be expressed in metaphysical dependence conditionals: if the molecules did not bond in a particular way, the ensemble dynamics would not come to exist. Material constitution can also be described in terms of such metaphysical dependence conditionals: if Piece had not been formed in such-and-such a way, David would not have existed. Thus, we take such conditional expressing relations of metaphysical dependence to provide a fourth reason for treating continuous reciprocal causality as a species of constitution.

We have seen above how other philosophers have defended the idea of diachronic causal constitution. Leuridan and Lodewyckx (2021) for instance have provided a well-worked out defence of such a view. Interestingly, a central example they discuss in their paper of diachronic causal constitution is perception-motor coupling. They draw upon Vernon et al. (2015) who use perception-motor coupling as an example of what we have been calling continuous reciprocal causality (though they use the phrase ‘circular causality’).

Leuridan and Lodewyckx depict diachronic causal constitution as an interlevel relation that inherits its diachronicity from the relata, which (following Kirchhoff (2015)) they take to be intrinsically temporal. Their account of constitution is one in which lower-level processes (for instance long-term potentiation of neurons) constitute some higher-level phenomenon of interest (in their example, the spatial memory of a mouse navigating a maze).

There are two key assumptions at play in their account of diachronic causal constitution we wish to end by questioning. The first is that there are genuine ontological levels, e.g. the spatial memory capacity of the mouse is at a higher ontological level of reality than the long-term potentiation of its hippocampal cells.¹⁷ Second, Leuridan and Lodewyckx assume that the constitution relation can be described as an interlevel vertical relation between these lower- and higher-levels, albeit a relation that unfolds over time.

Our diachronic view of constitution puts pressure on both assumptions rejecting the view that dynamical systems can be comprised of entities that exist on different (some higher, some lower) ontological levels. In dynamical systems such as the Rayleigh-Bénard convection system, the physical properties of a system can be understood in terms of fast microscopic variables, whose collective dynamics yields slow macroscopic quantities. The distinction between microscopic and macroscopic dynamics is a consequence of their respective (fast and slow) temporal scales. However, a consequence of the enabling constraints that form between these different timescales is that the processes that unfold over these different timescales cannot be thought of as existing at different lower and higher levels of reality. Microscale and macroscale processes instead stand in a relation of dynamic entanglement we have described in terms of continuous reciprocal causation. This is a view that finds a natural home in systems thinking across different disciplines such as biology, ecology and systems neuroscience (see e.g., Engel et al. 2001; Pessoa 2023), and in certain areas of contemporary philosophy of cognitive science (see e.g., Clark 1997; Juarrero 1999; Thompson 2007; Chemero 2009).

References

• Anderson, M.L., Richardson, M.J., & Chemero, A. (2012). Eroding the boundaries of cognition: implications of embodiment. Topics in Cognitive Science 4(4): 717-30.
• Baker, L.R. (1997). Why constitution is not identity. Journal of Philosophy 94,599-621.
• Barnes, E. (2018). Symmetric dependence. In R. Bliss and G. Priest (eds), Reality and its Structure: Essays in Fundamentality. Oxford: Oxford University Press.
• Bennett, K. (2017). Making Things Up. New York, USA: Oxford University Press.
• Bernstein, S. (2016). Grounding is not causation. Philosophical Perspectives, 30, 21-38.
• Bogen, J. (2008). Causally Productive Activities. Studies in History and Philosophy of Science 39: 112-23.
• Bunge, M. (1959). Causality. Cambridge, MA: MIT Press.
• Cartwright, N. (2016). Contingency and the order of nature. Studies in History and Philosophy of Biological and Biomedical Sciences, 58, 56-63.
• Chemero, A. (2009). Radical Embodied Cognitive Science. Cambridge, MA: MIT Press .
• Chemero, A., and Silberstein Chemero , M. (2008). Defending extended cognition. In B. Love,K. McRae, and V. Sloutsky (eds.), Proceedings of the 30th Annual Meeting of the Cognitive Science Society. Austin, Texas: Cognitive Science Society.
• Cilliers, P. (1998). Complexity and Postmodernism: Understanding Complex Systems. London: Routledge.
• Clark, A. (1997). Being There: Putting Brain and World Together Again. Cambridge, MA: MIT Press .
• Craver, C. F. (2007). Explaining the Brain: Mechanisms and the Mosaic Unity of Neuroscience. Oxford: Oxford University Press .
• Craver, C.F., and Bechtel, W. (2007). Top-down causation without top-down causes. Biology and Philosophy, 22, 547-563.
• Cummins, R. (1983). The Nature of Psychological Explanation. Cambridge, MA: Bradford/MIT Press.
• Cummins, R. (1975). Functional analysis. The Journal of Philosophy , 72, 741-765.
• Dasgupta, S. (2017). Constitutive explanation. Philosophical Issues: A Supplement to Nous, 27, 74-97.
• Engel, A., Fries, P., and Singer, W. (2001). Dynamic predictions: Oscillations and synchrony in top-down processing. Nature Reviews Neuroscience, 2, 704-716.
• Eronen, M. (2013). No levels, no problems: Downward causation in neuroscience. Philosophy of Science, 80(5), 1042-1052.
• Fine, K. (2003). The non-identity of a thing and its matter. Mind, 112, 195-234.
• Friend, T. (2019). Can parts cause their wholes? Synthese, 196, 5061-5082.
• Gillett, C. (2007). Understanding the new reductionism: The metaphysics of science and compositional reduction. The Journal of Philosophy , 193-216.
• Ginzburg, V.L., and Landau, L.D. (1950) On the theory of superconductivity. Journal of Experimental and Theoretical Physics, 20, p. 1064.
• Godfrey-Smith, P. (2010). Causal pluralism. In H. Beebee, C. Hitchcock, and P. Menzies, (Eds.), Oxford Handbook of Causation. Oxford: Oxford University Press , pp.326-337.
• Haken, H. (1996). Slaving principles revisited. Physica D: Nonlinear Phenomena. 97: 95-103.
• Haken, H. (1983). Synergetics: An Introduction. Non-equilibrium phase transition and self-organisation in physics, chemistry and biology, 3rd ed., Berlin: Springer.
• Hall, E. (2004). Two Concepts of Causation. In J. Collins, E. Hall, and L. Paul, (eds.) Causation and Counterfactuals. Cambridge MA: MIT Press, pp. 225-276.
• Hobson, A., and Friston, K. (2014). Consciousness, dreams and inference: The Cartesian theatre revisited. Journal of Consciousness Studies, 21(1-2), 6-32.
• Hofweber, T., and Velleman, D. (2011). How to endure. The Philosophical Quarterly, 61(242), 37-57.
• Horgan, T. (1993). From supervenience to superdupervenience: Meeting the demands of a material world. Mind , 102(408), 555-586.
• Huemer, M., and Kovitz, B. (2003). Causation as simultaneous and continuous. The Philosophical Quarterly , 53(213), 556-565.
• Ingthorsson, R.D. (2020). A causal theory of constitution and persistence. In C. Kantzian & M. Szlatkowski (eds.), Jonathan Lowe and Ontology. London: Routledge .
• Ingthorsson, R.D. (2024). Causal efficacy: a comparison of rival views. In Y. Shan. (Ed.) Alternative Approaches to Causation: Beyond Difference-Making and Mechanism. Oxford, UK: Oxford University Press.
• Ingthorsson, R. D. (2021). A Powerful Particulars View of Causation, New York: Routledge.
• Ingthorsson, R.D. (2002). Causal production as interaction. Metaphysica 3(1): 87-119.
• Juarrero, A. (2023). Context Changes Everything: How Constraints Create Coherence. Cambridge, MA: MIT Press
• Juarrero, A. (1999). Dynamics in Action: Intentional Behaviour as a Complex System. Cambridge, MA: MIT Press .
• Kaiser, M., & Krickel, B. (2017). The metaphysics of constitutive mechanistic phenomena. The British Journal for the Philosophy of Science , 68(3), 745-779.
• Kelso, J.A.S. (1995). Dynamic Patterns: Self-Organisation of the Brain and Behaviour. Cambridge, MA: MIT Press .
• Kim, J. (1998). Mind in a Physical World. Cambridge, MA: The MIT Press.
• Kirchhoff, M.D., and Kiverstein, J. (2019). Extended Consciousness and Predictive Processing: A Third-Wave View. Oxford: Routledge.
• Kirchhoff, M.D. (2017). From mutual manipulation to cognitive extension: challenges and implications. Phenomenology and the Cognitive Sciences, 16, 863-878.
• Kirchhoff, M. D. (2015). Extended Cognition & the Causal-Constitutive Fallacy: In Search for a Diachronic and Dynamical Conception of Constitution, Philosophy and Phenomenological Research, 90(2), 320-360.
• Krickel, B. (2017). Making sense of interlevel causation in mechanisms from a metaphysical perspective. Journal for General Philosophy of Science , 48(3), 453-468
• Ladyman, J. & Wiesner, K. 2020. What is a Complex System? New Haven, CT: Yale University Press.
• Ladyman, J. (2012). Science, metaphysics and method. Philosophical Studies, 160(1), 31-51.
• Ladyman, J., & Ross, D. (2007). Every Thing Must Go: Metaphysics Naturalized. Oxford: Oxford University Press .
• Leuridan, B, and Lodewyckx, T. (2020). Diachronic causal constitutive relations. Synthese , 1-31, https://doi.org/10.1007/s11229-020-02616-0
  » https://doi.org/https://doi.org/10.1007/s11229-020-02616-0
• Leuridan, B. (2012). Three problems for the mutual manipulability account of constitutive relevance in mechanisms. The British Journal for the Philosophy of Science . 63(2): 399-427.
• Lewis, D. (1986). On the Plurality of Worlds. Oxford: Blackwell.
• Lewis, D. K. (1973). Causation. Journal of Philosophy , 70: 556-67.
• Lowe, E. J. (1995). Coinciding objects: In defense of the ‘Standard Account’. Analysis, 55, 171-178.
• Lowe, E. J. (1996). Subjects of Experience. Cambridge: Cambridge University Press.
• Lowe, E. J. (1989) Kinds of Being. Oxford: Basil Blackwell.
• Lowe, E. J. (1983). ‘‘Instantiation, Identity and Constitution,’’ Philosophical Studies 44, 45-59.
• Machamer, P., Darden, L., & Craver, C. (2000). Thinking about mechanisms. Philosophy of Science , 67(1), 1-25.
• Nicholson, D.J & Dupre, J. Everything Flows: Towards a Processual Philosophy of Biology. Oxford, UK: Oxford University Press .
• Pessoa, L. (2023). The Entangled Brain: How Perception, Cognition and Emotion Are Woven Together. Cambridge, MA: MIT Press .
• Potochnik, A. (2021). Our world isn't organized into levels. In D. Brooks, J. DiFrisco & W.C. Wimsatt (Eds.), Levels of Organization in Biology. Cambridge, USA: MIT Press.
• Potochnik, A., & McGill, B. (2012). The limitations of hierarchical organization. Philosophy of Science ,79(1), 120-140.
• Prigogine, I. & Stengers, I. (1984). Order out of Chaos: Man’s New Dialogue with Nature. Toronto: Bantam.
• Piccinini, G. (2022). Neurocognitive Mechanisms: Explaining Biological Cognition. Oxford: Oxford University Press .
• Rock, T. (2021). Dynamic Realism: Uncovering the Reality of Becoming Through Phenomenology and Process Philosophy. Edinburgh, Scotland: Edinburgh University Press.
• Ross, D., & Ladyman, J. (2010). The alleged coupling-constitution fallacy and the mature sciences. In R. Menary (ed.), The Extended Mind (pp. 155-166). Cambridge, MA: The MIT Press .
• Salmon, W. (1998). Causality and Explanation. Oxford: Oxford University Press .
• Salmon, W. (1984). Scientific Explanation and the Causal Structure of the World. Princeton, NJ: Princeton University Press.
• Schaffer, J. (2016). Grounding in the image of causation. Philosophical Studies , 173, 49-100.
• Shoemaker, S. (1999). Self, body, and coincidence. Aristotelian Society Supplementary Volume 73 (1):287-306.
• Seibt, J. (2009). Forms of emergent interaction in general process theory. Synthese , 166, 479-512.
• Simons, P. (1985). Coincidence of things of a kind. Mind , 94, 70-75.
• Spivey, M. (2007). The Continuity of Mind . New York: Oxford University Press.
• Thomson, J.J. (1998). The Statue and the Clay. Nous 32, 149-173.
• Ylikoski, P. (2013). Causal and constitutive explanation compared. Erkenntnis, 1-21. Doi: 10.1007/s10670-013-9513-9.
  » https://doi.org/10.1007/s10670-013-9513-9
• Thompson, E. (2007). Mind in Life. Harvard: Harvard University Press.
• Vernon, D., Lowe, R., Thill, S., & Ziemke, T. (2015). Embodied cognition and circular causality: On the role of constitutive autonomy in the reciprocal coupling of perception and action. Frontiers in Psychology, 6(1660), 1-9.
• Wasserman, R. (2004). The constitution question. Nous , 38(4), 693-710.
• Wheeler, M. (2005). Reconstructing the Cognitive World: The Next Step. Cambridge, MA: MIT Press .
• Wiggins, D. (1980) Sameness and Substance. Cambridge, MA: Harvard University Press.
• Wiggins, D. (1968). On being in the same place at the same time. Philosophical Review 77, 90-95.
• Wilson, A. (2008). Metaphysical causation. Nous , 52(4), 723-751.
• Wilson, J. (2014). No work for a theory of grounding. Inquiry 57: 535-79.
• Woodward, J. (2003). Making Things Happen: A Theory of Causal Explanation. Oxford: Oxford University Press .

• 1
  See Bernstein 2016 for a discussion of a host of other differences some of which will be discussed below.
• 2
  We will not enter directly into this debate. When the concept of grounding is used to track how certain things are produced or made up by other things, the argument of this paper for construing constitution as a diachronic relation of dependence applies (mutatis mutandis) to grounding. In other words, grounding as applied to dynamical systems should be thought of as a diachronic metaphysical relation.
• 3
  Bennett (2017) groups this family of metaphysical relations under the category of what they call ‘building relations’
• 4
  Some have argued that the constitution relation cannot take processes as relata (e.g., Ylikoski 2013 but cf. Machamer, Darden & Craver 2000). We argue later in this section that this claim is mistaken. The elements of which dynamical systems are composed are processes understood as “individuals which are (1) concrete or spatio-temporally occurrent but do not necessarily occur in a determinate bounded region, (2) more or less specific or determinate, (3) more or less spatially/and or temporally recurrent, and (4) more or less complex” (Seibt 2009, p. 495).
• 5
  We will assume in what follows that causation in dynamical systems is best analysed in terms of production. Hall (2004) has argued that the understanding of causation as production relies on a distinct concept of causation from a concept of causation that is analysed in terms of relations of counterfactual dependence between events. We will not take sides in this debate except to say that we tend to agree with Godfrey-Smith (2010) when he writes that “it seems unlikely that all the physical (and perhaps non-physical) connections that are recognized as causal by ordinary criteria will have similarities in their intrinsic make-up and structure” (p.8). We focus on the example of the concept of causation needed to make sense of the emergence of form or structure in complex dynamical systems. We do not claim that the account we give of production in terms of continuous reciprocal causation generalizes to other examples of causal production. Indeed, our main concern in this paper is not with the nature of causation but with the constitution relation between big things (i.e. complex dynamical systems) and the smaller things that make them up (the elements of such systems and their organisation).
• 6
  Here he is following Bunge (1959) who writes “physical action and reaction are, then, two aspects of a single phenomenon of reciprocal action” (p.153).
• 7
  What is not in the simultaneous theory of causation of Huemer and Kovitz is the claim that change in dynamical systems can be analysed as a reciprocal relation between parts and wholes.
• 8
  For discussions of material constitution see, Simons 1985; Lowe 1983, 1989, 1996; Thomson 1998; Wiggins 1968, 1980; Shoemaker 1999; Baker 2000, and Fine 2003, Wasserman 2004, among others.
• 9
  Notice that such an argument can be made without presupposing that causation is a diachronic relation. Even if the simultaneous theory of causation of Huemer & Kovitz (2003) is true, it will still follow that causation and material constitution can be argued to be distinct metaphysical relations. Huemer and Kovitz do not question that causes and effects are wholly distinct events even if they happen simultaneously.
• 10
  The idea of mutual manipulation as a test for mutual manipulation is articulated in detail in Craver (2007). The basic idea is as follows: some X is part of a system or phenomenon S if an intervention on a value of X changes a value of S and vice-versa. In Craver (2007), mutual manipulation is used as a way of testing forconstitutive relevance, where constitution is taken to be synchronic and distinct from causation. However, and following in the general tradition of interventionist accounts of causation (see e.g., Woodward 2003), mutual manipulation can also be used as a test forcausal relevance. Indeed, both Leuridan (2012) and Kirchhoff (2017) have argued that insofar the mutual manipulation criterion picks out a relation of relevance it picks out causally relevant relations between X and S.
• 11
  Leuridan and Lodewyckx (2021) focus on mechanistic explanation between parts of a mechanism and the phenomenon to be explained by identification of such a mechanism. We do not consider mechanistic explanations but consider as our example of diachronic constitution, the emergence of structure in dynamical systems. Leuridan and Lodewyckx (2021) argue that constitution is a diachronic, inter-level relation between parts and whole, where parts and wholes exist on different ontological levels. We will argue later in our concluding comments that complex systems cannot be decomposed ontologically in this kind of way. We develop this argument in more detail in section 4. Finally, we do not take a stand here on the question of whether the mutual manipulation criterion can be used to determine both causal and constitutive relevance.
• 12
  Nor is the constitution relation equivalent to identity since identity (like coincidence) also has the formal properties of being reflexive and symmetrical. One reason these differences are welcomed by constitution theorists is because material constitution thereby provides a core concept for a non-reductive materialist metaphysics (see e.g. Rudder-Baker 1997).
• 13
  One might be wondering whether the Rayleigh-Bénard convection system is a special kind of system, thereby limiting the intended generality of our argument. Our arguments generalise to complex dynamical systems across the board, including the human central nervous system, though we will not attempt to make such an argument in what remains of this paper. For example, Engel et al. (2001) review models of cognitive activity that demonstrate large-scale neuronal dynamics exerting a significant influence on local neuronal activity by ‘enslaving’ local processing elements. The models they review demonstrate how neuronal activity at local levels can be influenced by activity at larger spatial and temporal scales in the brain. This is an instance of the kind of part-whole co-dependence we will defend in the remainder of our paper.
• 14
  This is an observation which coincides with theorems in the physical sciences such as the slaving principle in physics and thermodynamics (Haken 1983). That is, the enslavement of microscale (molecular) by macroscale (ensemble) dynamics is an inevitable result of the delineation of temporal scales, something one can observe in all dynamical systems (Ginzburg & Landau 1950; Hobson & Friston 2014).
• 15
  Should we be interpreted as arguing against material constitution and in favour of replacing this concept with our diachronic view of constitution? We will not attempt to make such an argument in this paper. However, we do wish to note the following since an anonymous reviewer pressed us on this important question When we consider the relation between David and Piece, we are consideringobservable entities- a piece of clay and a statue. Were one to zoom in on the microscopic composition of Piece, one would expect it to have a similar dynamic profile to the unobservable microscopic dynamics of the convection roll system. We agree with the reviewer, our diachronic view of constitution has broader implications for the discussion of the constitution of entities as such. Yet, a full treatment of this kind of implication - i.e. an outright defence of process metaphysics - is a larger project that we hope to return to in future work.
• 16
  Our diachronic view of constitution shares certain similarities with what Piccinini (2022) calls ‘diachronic composition’. Piccinini (2022) targets the idea that a synchronic notion of composition implies strict identity between parts and whole, which he takes to be problematic when applying talk of compositional relations to complex systems such as brains. We agree. Diachronic constitution, as we shall elaborate on a little later, is a relation of non-identity. However, Piccinini holds on to several other assumptions about composition that we identify as deeply problematic when addressing the causation-constitution distinction in complex systems: e.g., the asymmetry assumption.
• 17
  For a discussion of the concept of levels of organization in the context of mechanisms see Craver (2007), Craver & Bechtel (2007); Kaiser & Krickel 2017; Krickel 2017; Leuridan & Lodewyckx 2021. For critiques of the mechanistic account of levels of organisation see Potochnik & McGill 2012); Potochnik 2021and Eronon 2013.

• Article info
  CDD: 501
• Funding:
  No funding agency.

Publication Dates

• Publication in this collection
  21 Oct 2024
• Date of issue
  2024

History

• Received
  12 July 2023
• Reviewed
  05 May 2024
• Accepted
  02 July 2024