(Forthcoming in the volume on 'Embodiment' in the Oxford Philosophical Concepts series.)
At its most capacious, 'atomism' can be taken to refer to any theory according to which the world consists in elementary units, which are not further analyzable into something more elementary. 'Units' should be understood here not to refer to particles, but to any physical or metaphysical, quantitatively or qualitatively basic principle from which the perceptible world derives. The most familiar variety of atomism, the variety that can still be called 'atomism' even in the most restricted sense, takes the basic units as elementary particles that are (i) physical; (ii) ungenerable and incorruptible; and (ii) qualitatively neutral or lacking in 'secondary' qualities, as the color, smell, etc., of compound bodies emerge from the arrangements of the atoms constituting them.
If we take atomism of this latter variety as the standard model, which we might call 'Democritean', then we may begin to construct a typology of atomisms by plotting them in relation to criteria (i)-(iii). At some remove from the Democritean model we find Leibniz's theory of monads, according to which the basic units of reality are indeed ungenerable and incorruptible, and are responsible for the qualitative variety of the phenomenal world without themselves bearing these qualities, and yet according to which, by contrast with Democritean atomism, these requisites are not the physically indivisible constituents of the phenomenal world, in the way bricks are the constituents of a house, but rather are the metaphysical requisites of the physical world. Leibniz calls them variously 'atoms of substance' or 'metaphysical atoms'. Commentators have tended to take this doctrine, and Leibniz's labels for it, as amounting to a rejection of atomism, but on a certain more capacious view of what atomism might be, we could instead take Leibniz's claim that he is committed to metaphysical rather than physical atoms as commitment to a variety of atomism.
Quite a bit further still from the Democritean model we find the momentary quality atoms of the 7th-century CE Buddhist philosopher Dharmakīrti, or of his predecessor, the 6th-century philosopher Dignāga. These are atoms only in the sense that they are the basic units of reality, not constituted by anything else, and not comprehensible in terms of something more fundamental. Buddhist philosophers posited these as the fundamental ingredients of the world in the course of a denial of any more stable or enduring world of real individual substances.
One implication of any version of atomism might be thought to be that there can be no composite substances, which is the same thing as to say that there can be no embodied selves, and that anything composed out of atoms will remain only an arrangement of atoms, an ens per aggregatum. An important development in subsequent Indian philosophy however, as I will describe in this chapter, would be an attempt from within an orthodox school of what would later be called ‘Hindu’ philosophy to argue, against the Buddhists, that embodied selves can exist within an atomic universe. The Nyāya-Vaiśeṣika school worked out a coherent account of how the world can consist both in atoms and at the same time, in contrast to the Buddhist denial of the self, in selves that are necessarily embodied in complex arrangements of atoms.
Two principles that would seem to be incompatible with atomism are called in the Nyāya-Vaiśeṣika school of Indian philosophy avayavin or 'whole', and sāmānya or 'universal'. According to D. N. Shastri, these are the "two chief synthetic or unifying principles" at work in this school. Yet the principle attainment of at least the Vaiśeṣika branch of this school was to spell out an atomistic theory of the natural world, and an important question arises as to how any atomistic doctrine could possibly accommodate real wholes, let alone universals universals. In at least this regard, the Dharmakīrtian variety of atomism is rather more coherent, or at least predictable, to the extent that it takes only dharmas or discrete moments to exist, and denies any causal connection between them or synthesis of them. Such synthesis can exist only in our thought, according to the Buddhists. Nyāya-Vaiśeṣika, by contrast, wants to have both atoms as well as real, embodied, composite substances.
Here, Leibniz's metaphysical atomism shows up as a middle road, as he evidently believes that, as a result of the perceptual activity of the simple substances from which compound bodies result, these compound bodies may be elevated to the level of corporeal substances, which is to say true unities per se, grounded metaphysically in an embodied self, or, as Leibniz puts it, something analogous to the moi. At the same time, Leibniz utterly denies the reality of universals, which by contrast Nyāya-Vaiśeṣika strongly defended. In this chapter I would like to go some distance towards illuminating the philosophical justification of the Nyāya-Vaiśeṣika defense of ‘substance atomism’, which simultaneously reduces reality to elementary particles, yet holds onto the reality of the substances, and even of the universal kinds, constituted by these particles. I would like, in particular, to show how the coherence of the Nyāya-Vaiśeṣika account of embodied wholes is grounded in, and inseparable from, this school’s parallel theory of universals—a prima facie outlandish theory according to which the universal is literally inherent in the individual whole in a manner comparable to the inherence of water in a pot.
Both Nyāya-Vaiśeṣika atomism and Leibnizian corporeal-substance theory might be accused of seeking to have their cake and eat it too. But this accusation is wrong in both cases, since in both cases it is precisely the unifying function of the moi or soul, or Leibniz, or the mind (manas) for Nyāya-Vaiśeṣika that transforms the thing in question from a bodily aggregate to an embodied substance. This comparative approach may help us to gain insight into the parameters of, and the range of possibilities in, thinking about the problem of embodiment. Ordinarily (as in, e.g., Epicureanism, Gassendism, Buddhism, among others), atomism is taken to hold that there simply are no indissoluble, transcendent selves behind or prior to the collections of atoms. But this is not the case in the varieties of atomism we are considering here. Our examples show, in fact, that on a certain understanding of atomism, one which has emerged independently on different continents and in different centuries, the very purpose of invoking atoms as the basic ingredients of reality is not at all to explain away composite wholes, but rather to give a coherent account of the nature of their composition, an account of a reality that is analyzable down to an atomic level, but that nonetheless consists in real, complex, composite, embodied selves.
 D. N. Shastri, Critique of Indian Realism: The Philosophy of Nyāya-Vaiśeṣika and Its Conflict with the Buddhist Dignāga School, Delhi, Bharatiya Vidya Prakashan, 1997 , 306.