On Plato’s Precosmos (Ti. 52d2–53c3)

1 Introduction

In a celebrated paper, G. H. von Wright suggested that the philosophical problem he was dealing with, namely how to consistently grasp the relation between time, change, and the world, has its origins in Aristotle and Augustine.^[1] He was right, but omitted from his (rather short) list of original sources for this problem a name: Plato. The reason why he did not include Plato is, in my view, that in 1969 no philosophically strong explanation of Plato’s conception of precosmos – as opposed to our cosmos, and as distinguished from the latter by an event, or a (divine) action – had been produced yet: in fact, Plato provides some unique evidence for an extreme case, one in which we have ‘generation’ without cosmic time and the generation of time and cosmic ‘generation’ coincide.

Not by chance, the clearest description of a precosmic world, namely Plato’s, is usually regarded as somewhat fanciful and contradictory, and there is the risk that the only attempt to produce such an account may be reduced to good myth-writing. At the same time, also those standard readings of the precosmos which aim to take it seriously eventually lead to puzzling and contradictory views: as I will suggest, this applies to the idea that the precosmos is populated by ‘simples’ of the elements which are not perfectly arranged to one another, or by proto-geometric objects lacking stereometric perfection, or even by products of a random association of precosmic objects. All these readings, moreover, either reject or minimise the idea that the precosmos already participates in the intelligibles and do not take seriously the following crucial and explicit assumptions: the Receptacle is absolutely amorphon in itself; there is generation before the generation of the world.

My overall aim is to provide a fresh reading of Timaeus 52d2–53c3 according to which Plato’s precosmos is not the fanciful piece of writing scholars have usually taken it or made it out, to be. More specifically, I shall argue that it can be consistently conceived of if and only if: the Receptacle’s is intrinsically amorphon and it participates in the Paradigm also before the generation of the world (Section 2); there is full precosmic participation in the Paradigm and precosmic traces are randomly complex bodies (Section 3); precosmic generation is the mechanism according to which precosmic bodies undergo continous coming to be and perishing (Section 4). This will lead me to consider the precosmos not merely as an imperfect and imaginative anticipation of the ‘cosmos’: rather, Plato’s precosmos, far from being a somewhat approximate image, is indeed a possible world, whose description represents the robust philosophical instrument introduced by Plato in order to justify the beneficial nature of the Demiurge’s creative activity (hence the need for an intelligent craftsman).

Before I present my argument, let me briefly clarify three premises. First, I shall not commit to either a sempiternalist or a temporalist reading of Plato’s cosmogony: the puzzle of the precosmos affects both readings; consequently, both would profit from a new solution.^[2] Hence, my reading will prove effective regardless of a decision about this issue, though it will eventually strengthen the case for a temporalist reading. Second, I will not assume the eikos logos to be intrinsically fanciful and epistemologically weak. Not only, as widely known, Timaeus repeatedly refers to his own speech as a logos and not as a mythos; rather, it is possible to show that, at least when the causal power of the intelligible is at stake, the speech is presented as bearer of true and rigourous claims.^[3] Finally (and consistently), I reject the idea that the section of Timaeus’ tale devoted to the Receptacle is programmatically inconsistent:^[4] I would opt for this interpretation only if an explanation taking seriously Plato’s account of the precosmos were not available at all – Plato was not forced to introduce such a detailed description of the precosmos, which must therefore have a specific import.

2 The Conditions for the Precosmos

Let me first quote our key passage in full (Timaeus 52d2–53c3):^[5]

Let this, then, be given as the tale summed according to my judgment: that there are Being, Space, *Generation* – three distinct things, even before the Heaven *had generation*. Now the nurse of *generation*, being made watery and fiery and receiving the *characters* of earth and air, and qualified by all the other affections that go with these, had every sort of diverse appearance to the sight; but because it was filled with powers that were neither alike nor evenly balanced, there was no equipoise in any region of it; but it was everywhere swayed unevenly and shaken by these things, and by its motion shook them in turn. And they, being thus moved, were perpetually being separated and carried in different directions; just as when things are shaken and winnowed by means of winnowing-baskets and other instruments for cleaning corn, the dense and heavy things go one way, while the rare and light are carried to another place and settle there. In the same way at that time the four kinds were shaken by the Recipient, which itself was in motion like an instrument for shaking, and it separated the most unlike kinds farthest apart from one another, and thrust the most alike closest together; whereby the different kinds came to have different regions, even before the ordered whole consisting of them came to be. Before that, all these kinds were without proportion or measure. Fire, water, earth, and air possessed indeed some *traces* of their own nature, but were altogether in such a condition as we should expect for anything when *God* is absent from it. Such being their nature at the time when the ordering of the universe was taken in hand, the god then began by giving them a distinct configuration by means of *forms* and numbers. That the god framed them with the greatest possible perfection, which they had not before, must be taken, above all, as a principle we constantly assert; what I must now attempt to explain to you is the distinct formation of each and their origin. The account will be unfamiliar; but you are schooled in those branches of learning which my explanations require, and so will follow me.

In very general terms, the Timaean precosmos is characterised by disorderly motion and the presence of traces (ichnē) of the elements, all this being embedded in the precosmic Receptacle.^[6] I shall get back to the specific meaning of such ‘traces’ in the next sections, but for the time being it is enough to state that this characterisation implies that precosmic objects are, on the one hand, related to the elements and, on the other, imperfect with respect to them. (To be clear, with the phrase ‘precosmic objects’ I refer to individual precosmic sensible particulars, as it will emerge in due course.) Apparently, a sempiternalist reading might argue for an economical explanation: by hinting at an imperfect precosmic condition, Plato would be stressing that a (fully realised) contact with the intelligible realm is necessary in order for the world to be perfect.^[7] On the temporalist side, Gregory Vlastos suggested that precosmic objects represent a sort of limit for the existence of the Receptacle in the precosmos: since something devoid of form would not exist at all, Plato has to concede that precosmic objects intrinsically possess the minimal ‘amount’ of qualification allowing them to exist.^[8] Probably Vlastos is detecting a crucial philosophical question here – that of the ontological status of precosmic objects – but his interpretation tells almost nothing about their actual status. The temporalist solutions which have been advanced over time paint the following overall picture. If one takes strictly the claim that the elements are produced by the Demiurge by means of forms and numbers (53b4–5) and the imperfect status of precosmic interactions among them, precosmic objects are ‘simples’ of the elements (to use Gill’s word) which are not perfectly arranged to one another, or proto-geometric objects which do not possess the stereometric perfection which the Demiurge will provide, or even products of a random association of precosmic objects.^[9]

Both interpretations are not puzzling with respect to the production of precosmic motion. Indeed, the traces, if conceived as precosmic bodies, can produce shaking:^[10] the Receptacle is a necessary ‘crafting stuff’ for the Demiurge, but also entails the capacity to enact mechanical and necessary causation, hence to enable demiurgic activity to be exploited effectively through persuasion.^[11] The need to read in this way precosmic motion is dictated above all by the fact that the Receptacle, being in itself devoid of any soul, cannot be provided with any intrinsic motion. This is indeed a renown aspect of Timaeus’ account, which however has to be taken more seriously than it usually is, for it is based on a more general premiss: the Receptacle itself is intrinsically devoid of anything which could alter its ‘receptivity’. As I shall stress in what follows, this eventually leads to the positing of precosmic participation.

In a vexed passage (49a6–50b5) Timaeus distinguishes instances of the elements – each of which is a τοιοῦτον – from the Receptacle, which can be taken to be a distinct τοῦτο or τόδε.^[12] Timaeus explicitly states that the reason why one can take the Receptacle to be a τοῦτο is that it always has the same nature and power, from which it never departs in any way (ἐκ γὰρ τῆς ἑαυτῆς τὸ παράπαν οὐκ ἐξίσταται δυνάμεως), and which consists in always receiving everything (δέχεταί τε γὰρ ἀεὶ τὰ πάντα) as an ἐκμαγεῖον (50b6–c6). The explicit characterisation of the Receptacle that follows (50d2–51b2) better explains the conditions allowing it to be such. As is widely known, here the Receptacle is compared to a mother lacking any kind of intrinsic configuration (i.e., ἄμορφον). As Timaeus carefully explains, this is the crucial requirement in order for the Receptacle to accomplish its function, for otherwise the assimilation of sensible properties to related forms would be poorly realised (50e3: κακῶς ἂν ἀφομοιοῖ). So here and in the following lines (51a7–b2) Timaeus explicitly and unhesitatingly (cf. οὐ ψευσόμεθα at 51b2) describes the Receptacle as something invisible, having no form in itself, and intrinsically suited to participating in the intelligibles. These aspects are moreover reciprocally intertwined: the Receptacle is radically devoid of forms, properties or shapes in itself, and this allows it to exploit its distinct power and nature (which enable it to be a specific ontological kind on its own), consisting in effectively enacting the receptiveness of the properties of forms. Moreover, since these are the intrinsic nature and power of the Receptacle (i.e., they are independent from any cosmogonic aspect), the Receptacle’s receptiveness cannot at any point be simply inchoative. From all this it follows that the Receptacle is intrinsically receptive in the sense that at any stage, either before or after the generation of the world, it will receive properties and produce no active interference in their reception.

All this provides solid grounds not only for excluding any intrinsic motion in the Receptacle, as scholars have already emphasised, but also for a more radical conclusion. Interpretations of precosmic objects implying that their features depend just on the Receptacle are to be rejected, for the Receptacle lacks any form in itself: saying that precosmic objects are not as perfect as ‘cosmic’ ones is not enough to allow them to populate the precosmic Receptacle, for the latter is constitutionally incapable of making them emerge from and by itself. A fortiori, precosmic objects cannot be imperfect anticipations of the elements or imperfect geometrical figures already present in the Receptacle as such: this would require the Receptacle to be already ‘geometrically shaped’, hence not to be ἄμορφον in itself.

This leads to the question of how we could really conceive of any kind of precosmic object, if any contact with the intelligible were excluded. Now, if being receptive coincides with the Receptacle’s nature, precosmic participation is to be considered the most effective option: there is no reason why the Receptacle should be receptive of forms only after the world’s generation. On the other hand, this does not imply that the interaction in question always spontaneously produces perfect sensible particulars (i.e., bodies provided with a specific set of properties and a mathematical (micro-)structure): sensible particulars populating the generated world are specific complexes of properties depending on forms, but nothing implies that precosmic objects have to be as perfectly formed as those populating the generated world. Accordingly, the Receptacle must absolutely lack any feature belonging to the ‘properties’ it receives, but nothing implies that at any time the presence of a set of properties, each reproduced properly, will be arranged in such a way as to produce a well-shaped sensible particular (from the elements on). On the positive side, therefore, focusing on the specific relation between the Receptacle and forms is an opportunity for Timaeus to explore what these can or cannot achieve by themselves, that is regardless of the Demiurge.^[13]

We are now faced with the following scenario. First, the Receptacle radically lacks any kind of shape and configuration, which it must therefore draw from some other factor. Second, the configuration of the Receptacle is not bound to result in a well-shaped sensible particular, but only has as its outcome the qualification of portions of the Receptacle with a set of properties. Hence, in the next section I set out to show that Plato’s account leaves room for the presence of randomly complex and not well-shaped precosmic sensible particulars, which are what Timaeus eventually is referring to in his puzzling description of ‘traces’ and that derive from a full, though non-teleological, precosmic participation of the Receptacle in the intelligible Paradigm.

3 Traces and Forms

At the basis of standard interpretations of the precosmos lies the puzzle of what the ‘traces’ (ichnē) of the elements actually are. In very general terms, a good starting point is provided by Verity Harte’s suggestion:^[14] the traces of X cannot be incomplete and imperfect with respect to X as parts of X, but only as something sharing certain features with X, while not having all the features that belong to X and being unable to completely exercise the proper functions of X. However – and here I disagree with Verity Harte – this must not necessarily be understood in terms of a reduction of features and functions: I would submit that they are not simpler than the cosmic elements of which they are traces, but are in fact randomly complex with respect to them, at least from some points of view.

Let us start from the word ichnos, which is often used by Plato in idiomatic phrases such as ἴχνη μεταθεῖν (‘to follow the traces’). There are (at least) two passages in the corpus which reveal quite an unexpected sense of the phrase. In the Statesman (301b–e), after discussing the truest constitution, and assuming that the true king is nowhere to be found in nature and history, the Eleatic Stranger suggests that the Young Socrates move on to the question of laying down laws by imitating the truest constitution and following its ichnē. To this a renowned description of historical constitutions follows, in which the Stranger embarks on an enquiry in order to evaluate what form is closest to the truest constitution. Now, the Stranger’s invitation to pursue the ichnē cannot just coincide with the aspiration to consider the truest constitution, for in this case he and the Young Socrates would have followed the truest constitution itself. The point is rather that, by knowing the truest constitution, one can grasp the properties that it shares with historical constitutions, which are themselves ichnē of the truest one. This is confirmed at the beginning of the Sophist’s sixth division, where the Stranger pushes Theaetetus towards a new attempt to define the sophist by following a certain ichnos of him (226b2). Since five definitions have already been produced, the next ichnos will be just one among many possible ichnē, each revealing and sharing certain properties of the object of which it is an ichnos. At the same time, however, the following division (226b1–231b9), famously leading to the Sophistic of noble lineage, will feature several species and subspecies of the separative art, whose progressive selection will eventually provide a new definition of Sophistic. Now, if this definition, albeit provisional, is what the Stranger is looking for at the moment, then the ichnos of the sophist does not coincide with the final definition, but with each definitional step leading to the sophist himself. Not by chance, each subspecies of the separative art will share some of the properties belonging to the sophistic art, but it will be imperfect inasmuch as it is redundant and impure, and not inasmuch as it is simpler and incomplete.

This is then what ichnos may mean in general for Plato, and such a conception is far from absurd.^[15] Let us consider the most usual case of ichnos, that of a trace of an animal. A trace will reveal some of the features of the animal, for instance its weight, its overall dimension, and even the overall power which the animal can express, whether it is healthy, and so on. At the same time, the reason why the ichnos is imperfect with respect to the animal is, on the one hand, that it lacks many of the properties, or powers, of the animal and, on the other, that it has some other features on its own as an ichnos. Hence, the ichnos as such will be a very different and specific sensible object, and will just suggest the presence of properties in the animal, and just with respect to some of them. The points of contact can of course be more or less strict. For instance, in some cases the trace of an animal just reveals properties of the animal, while in other cases it is itself a bearer of these properties – as in the case of olfactory traces. Moreover, it may also happen that traces of an animal are contaminated by other factors, independent from the animal. In brief, an object Y is an ichnos of X since it reveals some properties of X, which it may also have as its properties, allowing an observer to recognise Y as the ichnos of X, but lacks some of the properties belonging to X and is the bearer of a set of properties which do not belong to X.^[16]

We can now get back to the precosmos passage in order to verify whether the ichnē can be understood as randomly complex traces (in the sense I will be clarifying), and at what conditions. First, precosmic objects must be bodies. Of course, they will not have a perfect mathematical structure, since only the Demiurge’s intervention will ensure that the world is correctly designed through numbers (53b4–5).^[17] Nonetheless, they will be tridimensional, for this is the fundamental spatial feature allowing a body to be such (cf. 31b4–6). The features of the Receptacle as such are sufficient in order to allow for this.^[18] But such configurations must also exert certain powers, which are what the precosmic Receptacle is full of (52e2: τὸ […] δυνάμεων […] ἐμπίμπλασθαι). Here two questions are in order: first, what the cause of these powers might be; second, which the powers involved here are. To the first question I have already indirectly answered in the preceding section: there is no possible cause of properties for the precosmic Receptacle apart from an external one, and the sole possible external cause for the precosmic configuration of the Receptacle are forms. This is also explicitly declared by Timaeus at the beginning of our passage: the reason why the Receptacle is ‘watery and fiery’ can only be its participation in forms, and the presence in it of properties belonging to air and earth is expressed by referring to the fact that already at this stage the Receptacle has the related character (52d5–6: τὴν δὲ δὴ γενέσεως τιθήνην ὑγραινομένην καὶ πυρουμένην καὶ τὰς γῆς τε καὶ ἀέρος μορφὰς δεχομένην).^[19] In brief, if portions of the precosmic Receptacle have the properties which will belong to the cosmic elements, then the Receptacle must already participate in the forms which will be the causes for the reception of these specific properties by the Receptacle in its cosmic configuration. Something of the sort has already been quite cautiously suggested by scholars, either explicitly or implicitly, but only in the sense that this participation is imperfect inasmuch as it produces isolated and minimal “simples” of the elements, in the way Mary Louise Gill has argued in a seminal paper.^[20] The preceding analysis of the notion of ichnos, however, has indicated that Timaeus’ image in no way implies that precosmic traces are imperfect because they are simpler with respect to the cosmic elements, while it is entirely possible that traces are randomly complex with respect to the cosmic elements. This is the claim I will now support by providing two kinds of arguments, namely textual and philosophical ones. I am aware that the text itself does not exclude in principle other readings (such as the traditional ones), but the philosophical analysis will support the idea that my reading is consistent with the text, produces a better philosophical sense and ‘charitably’ saves Plato from a number of inconsistencies.

First, precosmic bodies share two features also characterising cosmic objects, that is weight and density.^[21] This is explicitly stated by Timaeus in the precosmos passage through the comparison of the winnowing-basket, thanks to which (53a1–2) “the dense and heavy things (τὰ μὲν πυκνὰ καὶ βαρέα) go one way, while the rare and light (τὰ δὲ μανὰ καὶ κοῦφα) are carried to another place and settle there”. Timaeus has very good reasons to make this point, for it is a necessary condition in order to maintain a ‘mechanical’ explanation of precosmic motion of the traces. However, this also implies being tridimensional and solid for any precosmic object, and also that there is no reason to regard the traces as inchoative from the point of view of their physical status: precosmic bodies can act and produce affections related to weight and density. Moreover, if precosmic elements had these features but were merely simples, there would be no justification for the fact that the elements will never become isolated from one another and will never stop causing the disorderly shaking of the Receptacle (as Timaeus states at 52e5–6). Plato is well aware that, in general, like is naturally attracted by like and will tend to join it.^[22] Interestingly, this is a mechanical aspect of physical interactions and is not determined by either the Demiurge or the world soul. Rather, the world soul is the means by which the Demiurge avoids a risky consequence of this principle, that is the reciprocal isolation of the elements in the generated world, eventually leading to rest: while each elementary body will tend to join the relative place in which it finds its most appropriate seat and where, as a consequence, bodies of the same kind are located, the revolutions of the world soul produce a sort of centripetal push allowing all bodies to move throughout the universe (58a1–c4).^[23] This is particularly important: if neither the Demiurge nor the world soul are responsible for this physical principle, it must also apply to precosmic bodies.

This point granted, if one identifies precosmic bodies with imperfect simple bodies, it follows that the shaking would eventually result in homogeneity, whereby all similar traces would be gathered together, and this would eventually determine the end of the lack of homogeneity, and hence bring an end to precosmic motion and shaking.^[24] Timaeus’ explicit description of the precosmos testifies against this possibility: the precosmic Receptacle is characterised by dissimilarity in all its regions (52e2–4 and 53a4–6) and the precosmic bodies it comprises are always separate from one another and always move in different directions (52e5–6). All this strongly encourages us to abandon not only the idea that precosmic traces are simples of the cosmic elements, but also the assumption that in the precosmos there is just a set of four precosmic traces of the elements: if this was the case, whatever these precosmic traces may be, they would in any case be arranged as a limited and ‘stable’ set, and hence would eventually stop moving. As a matter of fact, if it is the revolution which ensures that homogeneity in the generated world is never reached, then the presence of four ‘sets’ of elements (also in the precosmos) does not in itself exclude homogeneity: one could envisage a scenario in which all precosmic elements are set apart from one another (the fiery with the fiery, the watery with the watery, etc.), producing four homogeneous (pre)cosmic ‘zones’.

The same conclusion can be reached by considering the issue from a different point of view. There is no reason why one should assume either that there is a one-to-one relation between a form and a portion of the Receptacle, or that it is necessary for a portion of the Receptacle to participate in a set of forms which produce a well-shaped sensible particular. On the one hand, a sensible property is still something different from a sensible particular – i.e., a sensible body – since sensible particulars have (or are, depending on the specific view one has of this issue) a set of properties, hence participate in a set of forms.^[25] At the same time, Timaeus never implies that the Receptacle’s participation in forms, if considered in itself, produces a well shaped sensible particular. The Receptacle acquires a multiplicity of properties, but this does not lead in itself to the generation of well-shaped sensible particulars: the only thing we know about sensible particulars produced in this way (i.e., simply by the participation of the Receptacle in forms) is that they are qualified bodies.

This is entirely consistent with a substantial difference between eidetic relations and the set of properties characterising a sensible particular. In effect, eidetic relations are in themselves far more wide-ranging than any complex of properties characterising a single sensible particular. Take the case of a human being. Each human being bears at every time a set of properties, depending on her participation in a specific set of forms. However, a single human being will never be the bearer of all the properties which they may potentially bear as a sensible particular participating in the form ‘human being’, since intelligible relations of the form ‘human being’ with other forms simultaneously encompass all possible empirical combinations of properties.^[26]

All this suggests that, if one admits that there is some precosmic participation of the Receptacle in forms (as one has to, as we have seen), it is also necessary that this participation not be limited to the forms of the elements: every portion of the Receptacle takes on a complex set of properties, though this set will be different from any set of properties characterising ‘cosmic’ sensible particulars – and this is the reason why precosmic bodies are imperfect with respect to cosmic ones.^[27] This reading also explains in what sense Timaeus not only affirms that the precosmic Receptacle encompasses the specific powers which will belong to the elements – that is, being fiery for fire, etc. – but also says that the Receptacle is “qualified by all the other affections (πάθη) that go with these” (52d6–e1).^[28] This means that each portion of the Receptacle is not just fiery, watery, etc., but the bearer of a multiplicity of properties which in the cosmic scenario follow the elemental powers, are not distinctive of elements in themselves, and cannot be taken to be rationally ‘selected’ for any kind of precosmic body.

One might object that the precosmic Receptacle is not in such a condition as to express the properties which some forms are causes of, or that, if the elements are not properly instantiated, a fortiori it will be impossible for the form, say, ‘Boar’, to be instantiated (for a boar or any other composite body of a living being is ultimately made up of the elements).^[29] Still, counter-objections may be raised. First, even conceding the point, there is an entire set of forms which, at least in principle, do not require any specific living being in order to be instantiated: potentially, all forms of qualities, quantities, or possibly relatives can be instantiated regardless of the existence of living beings, and to this one should add, of course, the greatest kinds. In principle a precosmic tridimensional body, imperfect as it may be, can be imperfectly qualified, or imperfectly quantified,^[30] just as it is imperfectly fiery or watery, hot or cold, etc. Second, the fact remains that in principle nothing would prevent any form from exerting causation on the Receptacle, and the fact that the instantiation of a form does not produce a perfectly shaped image does not imply that in such conditions a form is not instantiated at all (even though we cannot clearly conceive of what the imperfect result would be). As a matter of fact, participation in F is a necessary condition and a cause for something to be f, but Timaeus himself shows that participation is not enough for a full realisation of a perfect sensible particular: indeed, one needs secondary causes (46c–d) and a rational and beneficial orientation, namely that of the Demiurge and/or the lower gods.^[31]

It is also important to stress that no specific assumption about the causation and range of forms is required in order to make this point. I am not assuming any specific view concerning the range of forms at issue: it is enough to state that, whatever this range is, there is no reason to restrict precosmic participation to the forms of the elements. Of course, according to my reading traces would be more ‘randomly complex’ if one takes the range to be particularly wide, according to some explicit – though not necessarily reliable – statements occurring in the dialogues.^[32] However, my point is a formal one: if there is a form – whatever there may be forms of – it must already be active in the precosmos and an object of participation for the Receptacle. Second, that forms must be causally active in themselves directly depends on a serious consideration of the Receptacle’s being in itself amorphon: as we have already seen, the only reason why the precosmic Receptacle may have any qualification is some kind of participation in forms – the only alternative is that no trace at all could even be mentioned. All this is strengthened by two further remarks. First, quite simply, the Demiurge is not said to be the cause of forms being the Paradigm for the world; rather, the Demiurge allows the world to be the best possible image of the Paradigm.^[33] Second, quite obviously, taking forms to be causally active only under certain conditions would compromise their stability: unless one takes their causal power as extrinsic and subject to activation of any sort, their being paradigmatic causes has to be an intrinsic feature of them (just as, if the Receptacle’s nature implies the receptiveness of forms, it is meaningless to assume that its receptiveness ‘begins’ at a certain point).

We now have the following scenario. All forms are instantiated in the precosmos, and each portion of the Receptacle presents a multiplicity of properties dependent on intraeidetic communication and not on those properties which can be instantiated in a ‘cosmic’ sensible particular. Hence, Timaean traces are tridimensional bodies, provided with weight and density; they are not simple instances of a single form and must be different from one another in order to ensure precosmic shaking – more specifically, they are the bearers of properties characterising the elements plus an indeterminate multiplicity of other properties. This is the reason why it is impossible for the precosmic Receptacle to be homogeneous, and hence to ever rest. Though each body can be the bearer of the properties (and hence of the powers) which will characterise the elements, it is never just this. In this way, those precosmic sensible particulars that are most similar to one another because of shared elemental properties (being fiery, watery, etc.) will be attracted to one another (53a4–6: τὰ μὲν ἀνομοιότατα πλεῖστον αὐτὰ ἀφ′αὑτῶν ὁρίζειν, τὰ δὲ ὁμοιότατα μάλιστα εἰς ταὐτὸν συνωθεῖν),^[34] but their difference with respect to other properties will preserve their heterogeneity, hence their motion. And, as we have seen, such a description confirms the sense of ichnos I have outlined: the very idea of a trace may refer to a randomly complex object, which is also the bearer of some of the properties belonging to the object of which a trace is the trace.^[35] Asking according to what ratio each portion of the Receptacle would participate in one set of forms instead of another would lead to quite a straightforward reply: there is no ratio, for there is no precosmic intellect – i.e., no Demiurge – at this stage. I shall get back to this point in the last section.

4 Precosmic Genesis

If my analysis is sound, Plato’s precosmos should also be provided with some mechanism regulating the interactions between precosmic bodies: explaining this will further lead me to account for the puzzling claim that there is genesis (provisionally: ‘generation’) also before the world ‘had generation’ (52d3–4: ὄν τε καὶ χώραν καὶ γένεσιν εἶναι, τρία τριχῇ, καὶ πρὶν οὐρανὸν γενέσθαι). Some lines below (53a6–8) Timaeus iterates the claim and clarifies it: before the orderly generation of the world, all the elements were already present (διὸ δὴ καὶ χώραν ταῦτα ἄλλα ἄλλην ἴσχειν, πρὶν καὶ τὸ πᾶν ἐξ αὐτῶν διακοσμηθὲν γενέσθαι), though their condition was characterised by the absence of measure and reason (καὶ τὸ μὲν δὴ πρὸ τούτου πάντα ταῦτ′ εἶχεν ἀλόγως καὶ ἀμέτρως). This last aspect can well be explained on the basis of my previous analysis: there is no ontological limitation as to the participation of the Receptacle and forms in the precosmos, so that all forms – and, a fortiori, all the forms of the elements – are instantiated in the precosmic Receptacle. The serious puzzle concerns the actual meaning of precosmic genesis, and it proves even more serious since our section is directly linked to the previous distinction of the three kinds within the generated cosmos (since the tripartite distinction of the kinds rephrases 50c7–d2: ἐν δ’ οὖν τῷ παρόντι χρὴ γένη διανοηθῆναι τριττά, τὸ μὲν γιγνόμενον, τὸ δ′ ἐν ᾧ γίγνεται, τὸ δ′ ὅθεν ἀφομοιούμενον φύεται τὸ γιγνόμενον). The key to solving the puzzle consistently is to explain the references to precosmic and cosmic genesis, and also of the genesis of the world (for the world γενέσθαι), without ascribing different meanings to related terms and verbs.

The combined use of genesis to refer to the internal dynamics of the world (i.e., to the fact that objects in the world ‘have generation’, γίγνονται) and of the verb γίγνομαι to refer to the world characterises a crucial passage of Timaeus’ proem (27d6–29c2). Here Timaeus distinguishes “that which always is (what it is), having no genesis”, and “that which has genesis (τὸ γιγνόμενον), though never actually being (what it is)” (27d6–28a1).^[36] A straightforward reading for both genesis and to gignomenon would appear to be ‘becoming’, in the sense of ‘undergoing change’ for an object, and this would not necessarily imply any ‘generation’ of the object itself. However, this would produce two damaging argumentative inconsistencies.^[37] First, in the explanation which immediately follows (28a2–4), Timaeus specifies that the sensible realm is γιγνόμενον and ἀπολλύμενον, ‘destroyed’: this statement is meaningful only provided that γίγνομαι refers here not to becoming, but to the generation and corruption of each sensible particular. Second, the famous statement that the world γέγονεν, ‘is generated’ (or ‘came to be’), which occurs some lines below (28b6–7), is based on the fact that the world is subject to belief and sense perception, and in turn this is said in the aforementioned passage (namely at 28a2–3) to be a specific feature of τὸ γιγνόμενον. Hence, the only way to maintain a degree of continuity in the overall argument for the generation of the world is to take γένεσις, γιγνόμενον and γέγονεν as referring to generation (and to the related corruption, at least potentially) and not to becoming. Interestingly, we find quite a strict parallel for this idea in the passage which immediately precedes the description of the precosmos. After a new introduction of the intelligible realm, described as ἀγέννητον (ungenerated) and ἀνώλεθρον (indestructible) (52a1–2), Timaeus characterises the sensible realm by associating its being γιγνόμενον with each sensible’s being γιγνόμενον and in turn subject to destruction/destroyed (ἀπολλύμενον) (52a4–7). This too only makes sense if and only if Timaeus is not referring to becoming, but to ‘generation’ in quite a strict sense.^[38]

Hence, when Timaeus sums up the three kinds (being, Receptacle, generation) at the beginning of the precosmos passage and states that all three were there even before the generation of the world, he must be referring to the fact that even before the generation of the world there was a set of precosmic sensible particulars subject to generation and perishing. But we can now state what the real reasons are why all three kinds must have been there. The Receptacle in itself cannot account for the precosmos, since the Receptacle is absolutely unqualified in itself. This is the reason why also being – i.e., forms – must be considered to be part of the precosmos, namely as the causes for any property to be instantiated in any portion of the precosmic Receptacle. But both being and the Receptacle in themselves ‘work’ exactly in the same way both before and after the world’s generation, and this kind of participation produces precosmic ‘generation’ and perishing.

Now, any view taking precosmic objects as simples of the elements would strongly diminish the import of Timaeus’ claim, for in this case precosmic generation would only be similar to cosmic generation by analogy. Whatever view of cosmic generation one has, what makes it different from mere becoming is the continuous modification of the set of properties a sensible particular is at any moment. However, if a precosmic object were a simple of fire, water, etc., such a transformation could not occur, for it would always be the same simple (say, a portion partaking of Fire, etc.); furthermore, we would be led back to the puzzle of homogeneity pointed out above.^[39] My reading, on the contrary, makes it possible to seriously account for the robust claim of precosmic generation. If there is full participation in forms, each precosmic particular will participate in a set of forms; this participation will in principle make a precosmic sensible particular f differ not only from any other, but also from the precosmic sensible particular t from which f had generation and the precosmic particular w which f will ‘become’ at its perishing. Clearly, this explanation preserves both unevenness, hence precosmic shaking, and the symmetry between being, the Receptacle and generation both before and after the production of the world.^[40]

5 Plato’s Precosmos: How It Works, and Why It Works as It Does

My view, if it is accepted, entails a long series of consequences, whose full consideration would lead me beyond the scope of this paper. However, it is worthy lingering for a moment on its implications with respect to the quarrel about the nature of Plato’s cosmogony. In principle, my interpretation proves economical and applicable from both a temporalist and a sempiternalist perspective, because what is actually missing in the precosmos is the rationality of an organizing teleological principle. On the one hand, a sempiternalist reader could say that the representation of the ‘precosmos’ is designed to highlight the importance of the world soul, which might in this case be conceived as an immanent organising principle. On the other, a temporalist reader might insist that in this framework Plato can really highlight the need for a divine craftsman: while the Receptacle is intrinsically receptive and forms are intrinsically paradigmatic causes, in order to achieve the kind of participation that can make the world perfect, a divine generative intellect is necessary. This is capable not only of providing the elements with their perfect geometric structure, but also – and above all – of dictating a principio the rational standard determining whether, when and why specific sets of forms can be instantiated. Indeed, as we have seen, what precosmic participation lacks is a ratio ensuring, either directly or indirectly, that participation be accomplished in such a way that proper, well-ordered sensible particulars will be produced, and the Demiurge is precisely the cause introducing this order.

However, a further and major pay-off can be identified. It seems to me that Plato had envisaged the following problem: although the generative action of the Demiurge ensures the teleological orientation and perfection of the world as a whole, this is not enough to clear the Demiurge of all responsibility with respect to the low ontological status of sensible particulars in terms of persistence and identity. This is something which one should be concerned about: if the Demiurge were just what is needed for participation to ‘begin’, at most sensible particulars, including human beings, turn out to be continuously subject to generation and perishing, to the point of being entirely devoid of any identity across time and space.^[41] Our world, perfect it may be as a whole, is populated by shadowy particulars, and may even be described as shadowy itself. This is indeed a constitutive feature of Plato’s ontology, but the cosmological narrative of the Timaeus might risk making the Demiurge directly responsible for this. One might say that “this is the most the Demiurge can do”, but the unqualified nature of the Receptacle would in any case make him responsible for the generation of generation and its mechanisms. The situation is even worse if, as is widely acknowledged nowadays, the Receptacle positively ‘offers’ itself to the Demiurge and allows the best production to be accomplished.^[42] In brief, Plato would put himself in the position to make his beneficial god directly responsible for both the positive and the negative features characterising the sensible particulars’ being. However, a robust account of the precosmos clears the Demiurge of all responsibilities in this respect, and establishes him as an absolutely beneficial cause, for it restricts the range of ‘products’, or outcomes of production, specifically depending on the Demiurge to those which can only be regarded as being beautiful, beneficial and positively oriented.

Let us return to our boar. If the Demiurge were responsible for the participation of the Receptacle in forms, then he would also be responsible, either directly or indirectly, for making any sensible boar merely a shadowy image of related forms. The fact that boars live in a perfect world would be a sort of cosmologic consolation for a depressive ontological condition that in any case depends on the Demiurge. Everything changes if, on the contrary, the Demiurge must manage the already given status of sensible particulars, and a powerful strategy to establish this scenario is to consistently consider participation mechanisms and their implications to be active in the precosmos. In this way, the imperfection of cosmic particulars may be seen to be intrinsic to them, insofar as it exists before and independently of any action of the Demiurge: all the Demiurge does here is to accomplish the best possible reconfiguration of this radical imperfection by establishing a standard of rationality and order according to which participation takes place. At these conditions, therefore, the Demiurge’s contribution to the world is exclusively beneficial, for it is actually limited to the determination of the teleological orientation of already existing dynamics of participation on the one hand, and of generation and perishing on the other. Let us return to our boar for the very last. The Demiurge is not responsible for the fact that it is a sensible particular and a body, or for the fact that, as a sensible particular, it is a qualified portion of the Receptacle; nor is he responsible for its ontological status as a sensible particular. On the contrary, the Demiurge has the merit of establishing in principle the conditions allowing the boar to be a specific sensible particular, provided in each case with certain qualities, constituted by mathematically ordered elementary bodies, and playing its specific role in a perfect world, despite its limited ontological status. Briefly put, the Demiurge is the best possible causes of the best possible kind of generation (29a5–6).