Beginnings of Jaina Ontology and Its Models

The paper analyses the beginnings of systematic ontology in Jainism, which appears to have began after first century CE, albeit certain ontology-relevant terminology in a nascent form was present earlier. A clear expression of systematic ontological reflection is the existence of models that organize ideas and categories in a more consistent conceptual scheme. Jainism follows similar developments that had earlier taken shape in in the early Buddhist Abhidharma, proto-Sāṁkhya-Yoga and proto-Nyāya-Vaiśeṣika. In addition, the paper argues that the models, five in total, can be used as a methodological tool to distinguish various historical layers in early Jaina writings.


Introduction
The sources on the early history of Jainism and its philosophy are extremely scarce and scattered. Jaina writings of any relevance to the research on the history of its earliest thought and pre-classical development comprise (1) the canonical literature (āgama) of the Ś vetāmbaras codified during the third council (vācanā, or 'recitation the date of the extant redactions but due to their strongly compilatory character are difficult to discern.
The stray bits and pieces of information that are contained in these sources suffer from a number of deficiencies when it comes to establishing an approximate picture or the moment of gradual emergence and an early development of Jaina philosophy. The sources are extremely fragmentary, often difficult to date and never organized in any coherent model to the extent that it is practically impossible to speak of any early Jaina philosophy cultivated as a rational enquiry of coherent nature, based on rationally adduced arguments that are also falsifiable, etc. Instead, as it appears, we usually deal with assorted ideas that should rather be classified as early Jaina thought, being inherently linked to monastic discipline and salvific goal, not necessarily a reasoned systematic philosophy.
Truly philosophical issues, ontological models, epistemological matters or logical principles are never discussed in any systematic manner in the first centuries after the emergence of Jainism. 5 When such ideas first occur, they do not form any consistent system yet. Rather, they are ad hoc solutions to particular questions or consist in asceticism-related terminology which would later be elaborated in a philosophical direction. Further, we should not confuse the occurrence-in a particular passage-of certain nomenclature, such as jīva, ajīva, karman, jñāna, loka etc., that only later grew into technical terminology as building blocks of particular ontology or epistemology, with genuine philosophical reflection.
The situation is comparable to the case of the Vedic and early Upaniṣadic thought that contains a number important terms and concepts foundational for the subsequent emergence of Indian philosophy as a reasoned, rational, systematic, coherent, consistent, comprehensive reflection on the reality, that is, on all that is there, reflection supported by reason and some kind of evidence and grounded in an continuous attempt to explain why and how the reality is there, and how we can know it. The sheer presence in the Vedas or Upaniṣads of certain terminology pregnant with future philosophic richness does not turn these texts into philosophical treatises or their thought into philosophy, similarly as the presence of particular nomenclature and conceptions embedded in the myths retold by 'the collective author' Homer does not make the mythological contents of the Iliad and the Odyssey genuine philosophy. To speak of Vedic philosophy or of the philosophy of the Upaniṣads is as problematic (and perhaps even nonsensical) as to speak of the oxymoronic 'philosophy of Zen'. Accordingly, to read the eidetic and philosophically matured semantics into such ancient textual strata would be a methodological flaw. 6 This is what I mean when I speak of an absence of 'truly philosophical issues', and by implication of 'truly ontological' and 'truly epistemological' ones. Under genuinely philosophical approach, I understand rational, coherent and systematic enquiry concerning the structure of the world, its essence, causal background, means of knowing it, etc., with implied attempts to rationally justify one's claims or refute counterclaims with arguments that refer to facts or events observable to anyone, that can be verified and rationally rejected and that are formulated in a manner that consciously avoids contradictions, etc. To maintain that the world is a huge doughnut of blazing fire with a secured flat surface in the middle inhabitable by humans, invisible ghosts and fairies because one has heard so in a mountain cave from the archangel Gabriel does not constitute a philosophy, not only because not everybody is on speaking terms with the archangel and because such claims are not intersubjectively verifiable either. To counter a view at a certain time or in a particular geographical region, the people did not develop philosophy, one may not apply an all-too-easy argument that they in fact did but it is only due to our lack of transmitted information, not to the actual absence of their philosophical reflection, that we cannot prove it beyond doubt. By analogy, the 'lack of evidence' could also be used to support the claim that the ancient Indians flew a man to the Moon, but the evidence has somehow not survived. To meaningfully speak of any evidence for the existence of philosophy or for the Moon flight at an early period, we do need either some pieces of relevant information in the form of transmitted fragments or debris (of texts or of the spaceship) that the people engaged in the enterprise or at least a (more mature and complex) developmental stage attested at a later point of time that would allow us to infer that it must have been preceded by an earlier phase required for the latter stage to emerge in such a developed form. In the context of Jaina philosophy, we do not have any such evidence. The earliest traceable, extant textual fragments do not engage in matters of ontology and epistemology (unlike morality and soteriology). Once these issues are discussed, the enquiry seems to enter a rather nascent form, and the reader is presented with particular claims and schemes as a matter of belief, mostly with no justification and argument.
I would argue that one of palpable evidences for an emerging genuine philosophical system at a historical early stage is the existence of particular models that attempt to consistently portray certain structures postulated to exist either out there, as ontological frameworks, or within, as epistemological theories. Such models attempt to organise what is there in a reasoned, coherent, consistent and comprehensive manner as an explanatory device of the world structure. In addition, they propose particular structures of the world, of all that exists and cannot exist, that consist of certain basic components that are not reducible to other elements. Such a move is an application of an important law of parsimony, or philosophical razor, that requires one to balance various arguments for and against a particular category, component or structure. This is also a philosophically important step.
The aim of this paper is to provide a systematic overview of such basic ontological models that are traceable in Jaina thought from its early stages (first century CE) till the fourth/fifth century CE, viz. till the beginnings of the classical period of the Jaina philosophy, which commences with Umāsvāmin's groundbreaking Treatise on Reality (Tattvârtha-sūtra, TS; around 350/400 CE). In addition, I attempt to provide a time frame for each of these models when they seem to emerge for the first time in Jaina literature. In total, as we shall see, we can distinguish five such ontological models, with minor variants. In most of them, we find similar items, but they are organised differently under different headings depending on the character of the model.
The importance of such a survey of the earliest ontological models is that they present a development of basic concepts of Jaina ontology, consistently dualistic from the earliest detectable times, being rooted in the dichotomy of jīva-ajīva. An additional advantage of such an overview is that it can serve as a useful methodological tool to date other texts in which these models occur, or to check for their internal cohesion, to determine various interpolations or historical layers coexisting in texts that should rather be considered compilations than authored by one and the same person, albeit the the application of this method is not always as straightforward. As I demonstrate elsewhere 7 , the usability of this method can be instantiated with the case of 'Kundakunda', who should be considered, as I argue, a 'collective thinker' to whom a range of textual fragments of different periods were ascribed, complied and redacted under one umbrella name. 8 A natural worry with this approach may concern the problem of the dating of the models in question that are in turn used to date other texts; this could lead to a kind of vicious circle that the models themselves, being difficult to date, serve as a standard to date other texts, likewise difficult to date. Indeed, some of these model are traceable to the canonical literature which is in itself difficult to date, also because the canonical texts consist of various layers belonging to different historical strata. Fortunately, the models, as I would argue, can be dated on the basis of relative chronology, with points of reference for their dating completely independent of the texts that would subsequently be dated on their basis, and therefore such an approach does not involve any kind of circularity. Further, I argue that the models, at least in some cases, can be dated on the basis external to the history of Jainism inasmuch as they reflect certain philosophical ideas that can be observed in other philosophical systems in India at a particular point of time relevant to the emergence of these models or passages in which these models occur.
What is characteristic of all these models is that all of them are historically rooted in quintessentially soteriological ideas; however, with the exception of Model 2 (see below), they present a relatively soteriology-independent representation of ontological categories within the universe structure. Prior to the first such models within Jainism, we find at least two or three consistent attempts to develop, in a systematic way, a model of basic reals that would be both comprehensive and exhaustive, that is, that would postulate all necessary and required components to satisfactorily explain the complexities of the world and its phenomena and as well as incorporate all existent 9 components of the universe. Prior to c. 100 CE, similar attempts were undertaken by (1) the early Abhidharma tradition of Buddhism, which organised all the phenomena into basic constituents (dharma) and the five aggregates (pañca-skandha), by (2) the proto-Sāṁ khya in the form of Sāṁ khya-Yoga of the period 100 BCE-100 CE, known from the Mokṣa-dharma-parvan or from Aśvaghoṣa's Buddha-carita, that distinguished various sets of the reals (tattva) ranging from seventeen to thirty, 10 and most probably (3) a kind of proto-Nyāya-Vaiśeṣika known as the 'yoga', being one of the three methodological currents of Ā nvīkṣikī and mentioned in Kauṫilya's Artha-śāstra, 11 which was composed, compiled and redacted between 100 BCE and 100 CE. 12 Like these earlier ontological models, also Jaina classifications were attempts made in the same direction to present all structural elements of the universe that are either explanatorily necessary or factually existent, 13 even though not required to understand the world structure and its functioning.
The aspiration to determine a set of basic units of the world was noticeable much earlier in Indian thought as abundantly attested in the earliest Upaniṣads, when various models were tested. For instance, the Br had-āraṇyakôpaniṣad offers a few competing models, just to mention two of them (1) BĀ U 2.5.1-14 enumerates earth, water, fire, wind or air (vāyu), the sun, space directions (diś; or the space?), the moon, lightning, thunder, the space (ākāśa; or aether?), moral law (dharma), truth (satya), humanity (mānuṣa) and the self (ātman), all known as the vital juices (madhu); (2) BĀ U 3.2.13 mentions: the human being (puruṣa), speech, fire, wind, breath, the sight, the sun, the mind, the moon, space directions, the hearing, the earth, the body, the space, the self, plants, body hair, trees, head hair, water, blood and semen as the elements of the visible, tangible world, with the invisible brahman implied as what is behind all this. Similarly, the Aitareyôpaniṣad speaks of the five gross elements (mahā-bhūta), i.e. the earth, wind, the space, waters and lights, their combinations, various categories living beings, gods and ultimately brahman (AU 3.3), and the Chāndogyôpaniṣad itemizes a hierarchy of entities on several occasions (e.g., ChU 7.2.1;7.4.2;7.6.1;etc.). Other examples of attempts to classify the reals that make up the world are numerous. Much of such early Brahmanic 10 Cf. Larson (1979, pp. 95-134, esp. 104 ff.). 11 It is the famous passage which specifies that 'the system of methodological enquiry comprises proto-Sāṁ khya, proto-Nyāya-Vaiśeṣika and the materialists' (AŚ 1.2.10: sāṁkhyaṁ yogo lokāyataṁ cêty ānvīkṣikī); see: Balcerowicz (2012). 12 The standard six categories of substance (dravya), quality (guṇa), movement (karman), the universal (sāmānya), the individuator/particular (viśeṣa) and inherence, well-known from the Vaiśeṣika-sūtra, may have already been in existence in the proto-Nyāya-Vaiśeṣika current, which was closely affiliated to the medical tradition of the Caraka-saṁhitā, some portions of which pre-date the Vaiśeṣika-sūtra. There, we find the six Vaiśeṣika categories regularly mentioned in the text as important classifying devices as well as enumerated, CS 1 1.1.9 = CS 2 /CS 3 1.1.28-29: maharṣayas te dadr śur yathā-vaj-jñāna-cakṣuṣā / sāmānyaṁ ca viśeṣaṁ ca guṇān dravyāṇi karma ca // 28 // samavāyaṁ ca taj jñātvā tantrôktaṁ vidhim āsthitāḥ / lebhire paramaṁ śarma jīvitaṁ cāâpy anitvaram // 29 // [* CS 1 : tena rṣayas.] 13 Explanatorily necessary are those elements of a model of the universe that are not necessarily directly perceptible or inferable but postulated to exist because the explanatory model (whether in philosophy, science, or even theology, etc., or any other rationally constructed system) or the proposed mechanics of the world requires it for some (better or worse justified) reason as indispensable for the completeness of the model, such as for instance the Higgs boson first postulated by Peter Higgs, or -in the case of Jainism -the principles of movement and rest (dharma, adharma) to explain the phenomenon of motion/action/ change and immobility/inactivity. Factually existent are those elements which are directly observable or inferable in a way which is not seriously questionable (for a sceptic, this would still remain insufficient). speculations are focused on ritual-related items and very frequently ritual components, including Vedic mantras, assume cosmic dimension and become constituent elements of the world (BĀ U 1. 1.1;2.5.1;3.2.13;3.9;ChU 1.1.2;1.3.7;1.6;etc.). Similar attempts are also found in non-Brahmanical traditions, including the Buddhist model of the five skandhas and a plethora of dharmas. In the account of the Sāmañña-phala-sutta, we find the materialist Ajita Keśakambalin (Pāli: Ajita Kesakambala) who explains the self and consciousness as consisting, or resulting from a particular composition, of the four elements (cātum-mahābhūtiko), and thereby seems to reduce the reality to the earth, water, fire and wind (DN 2.23,p. 55). In the same text, Ā jīvikism-affiliated Kakuda Kātyāyana (Pāli: Pakudha Kaccāyana) 14 speaks of seven bodies (kāya) that make up the world, namely the earth, water, fire, wind, pleasure (sukha), pain (dukkha) and living element (jīva) (DN 2.26,p. 56).
Conspicuously, in the Sāmañña-phala-sutta account of the Jaina teacher Nirgrantha Jñātrputra (Pāli: Nigaṅt˙ha Nātaputta), or Vardhamāna Mahāvīra, we find no trace of any philosophical reflection whatsoever that would comprise ontological or epistemological elements even in a nascent form. The teacher is depicted as exclusively preoccupied with asceticism and ascetic restraint p. 57). Similarly, his teacher and co-founder of Ā jīvikism, 15 Maskarin Gośāla devotes no attention philosophical issues, in particular to epistemology or ontology, and instead is concerned with karmic retribution, asceticism and rebirth. He does speak of cosmological issues though, namely he provides complex enumerations of rebirth classes and rebirth loci, but such classifications are not concerned with ontology per se but rather with religious cosmology and various possibilities of rebirth depending on one's past deeds in view of karmic retributions . As such, they should be treated as direct extensions of moral-salvific speculations: they arrange particular loci of future birth to correspond to various kinds of conduct in order to illustrate a moral principle of karmic retribution, according to which laudable behaviour involves auspicious rebirth and evil deeds incur miserable fate. One should not confuse a cosmological model contained in Gosāla's speculation, which describes various types and hierarchies of living beings, potential rebirths and possible localities, or world strata in which one may be reborn depending on one's karman, with an ontological model or a framework that presupposes some kind of ontological reflection. It is true that such a cosmological model may serve, and indeed later on in history of Ā jīvikism and Jainism did serve as a foundation for their ontology, but an existence of particular cosmology does not necessarily presuppose any existent ontology. Such a cosmology, in the form attested at this early stage in Ā jīvikism, but also in Jainism, had a direct relevance for their monastic code of conduct, morality and salvific doctrine alone. Unlike ritual-centred Upaniṣadic ontology, Jaina ontological speculations developed against the backdrop of salvific processes, ascetic practices, karmic bonds and retribution, moral responsibility etc., for instance with moral terms such dharma and adharma assuming an ontological dimension.
A closer analysis of canonical material may demonstrate that an emergence of Jaina ontology, probably around the first century CE, came in steps, and was not an instantaneous discovery of a more or less complete system. 16 First, out of concepts related to the theory of salvific path and models explaining the workings of karmic retribution, some concepts of irreducible 'reals' as ontological categories developed that later merged into more consistent explanatory models. This came hand in hand with the emergence of epistemological models that aimed at explaining and justifying the emerging ontological concepts.
Even a cursory glance at the Canon in Six Parts (Ṡ KhĀ ), with its repetitive character, which already organizes the doctrinal contents in a slightly more systematic manner, unlike the more incoherent early Ś vetāmbara canonical works (due to a gradual process of their composition and compilation) 17 , shows these dependencies. The text begins with the mention of certain basic categories of living beings and of certain aspects important in their analysis: [2] Thus, with the purpose of investigation stages regarding these fourteen categories of living beings, the following precisely fourteen stages regarding them should be known. The point of departure is a classification of living beings into fourteen categories (jīva-samāsa) and their respective analysis proceeds via investigation stages, or factors (mārgaṇā-sthāna), which constitute a pre-anuyoga-dvāra device, or means of analysis well known from The Lecture of Explanations (Viy) or from the tradition of the Treatise on Reality (TS). 19 All of these fourteen stages (ṭhāṇa, sthāna), elaborated in the subsequent sūtras, are primarily of relevance to ascetic practice, which should lead to spiritual upliftment on the path to liberation. Based on these fourteen investigation factors is a hierarchy, or the fourteen stages of living beings (jīva-sthāna), comparable to a later classification of the fourteen stages of virtue (guṇa-sthāna) 20 and itself a result of a simplification of earlier lists mentioning 16 A useful preliminary study on the subject was done already by Dixit (1971). 17 I distinguish between the two processes, even though they may overlap. A gradual process of composition indicates that some portions of the final version of text were gradually extended and elaborated by subsequent authors, who themselves wrote new portions inserted within earlier layers. A gradual process of compilation means that different blocks of earlier texts, that functioned independently, in their own right, were brought together into a new larger text. 18 Ś KhĀ 1.1.2-4: [2] etto imesiṁ co'ddassaṇhaṁ jīva-samāsāṇaṁ maggaṇa-ṭṭhadāe tattha imāṇi co'ddassa cêva ṭṭhāṇāṇi ṇādavvāṇi bhavaṁti.
[4] gai iṁdie kāe joge vede kasāe ṇāṇe saṁjame daṁsaṇe lessā bhaviya sammatta saṇṇi ākarae cêdi. 19 See Dixit (1971, pp. 18-21). twenty items. 21 Neither this nor other early lists include typically ontology-related terminology, except for jīva. However, we may detect some kind of epistemological model behind the above scheme: the fourteen stages of analysis serve as methodological devices to examine a particular subject, namely the living beings that stand in the centre of Jaina ethics. Nowhere in the whole text do we find a similar analysis, clad in a shape of a model, of basic structural elements of the world independent of the salvific focus that could be considered a systematic ontological theory. One should not take, for example, the systematics of the animate world found in the text (ṠKhĀ 1.1.24 ff.) as an instance of such an ontological theory: the systematics serves to demonstrate the importance of the fourteen investigation stages and moral competence that is reflected in the hierarchy of the animate world.
The above is an illustration that a systematic organisation of epistemological devices into models or schemes 22 developed slightly earlier than similar ontological models, even though certain ontology was already presupposed or taken for granted, such as the distinction into loka (the inhabitable world space) and aloka (the uninhabitable rest of the space). In this sense we may speak of certain priority of epistemological reflection over ontological analysis which later finds an interesting parallel in most philosophical treatises in India in which epistemology precedes ontology. 23 A research on the earliest Jaina philosophy is seriously hampered by the way early Jaina writings were collected and codified. A typical feature of the Ś vetāmbara canon, but largely the same remark applies to Digambara canonical writings, is that -similar to the Pāli canon-the contents was largely reworked, rewritten, readjusted, standardised and unified in terms of terminology, expressions etc. One of numerous evidences for that is that we find exactly the same blocks of texts, whole passages, that were simply copied-and-pasted in various canonical works, and in the process the original, older readings were effaced and replaced with newer readings that conformed to the standardised doctrine at the moment of the codification of the canon. In the redaction process, also changes were introduced to the language, and very frequently the earlier Ardhamāgadhī was replaced or admixed with Jaina Mahārāṣṫrī (Ś vetāmbara) or Jaina Śaurasenī (Digambara). A large number, if not most, of such traces that attest to the fact that the original readings were 'updated' and 'modernized' to later standards date back to the period of final redactions of the Śvetāmbara canon, which took place between the fourth and fifth centuries, in particular in early 4th century CE at two concurrent and competing councils in Mathurā and Valabhī, convoked by Nāgārjuna-sūri and Skandila-sūri (Khaṅḋila), respectively, both producing divergent versions and readings, and finally at the third council of Valabhī (c. 450-480 CE), meant to 20 See: Glasenapp (1942Glasenapp ( , pp. 75-92), (1999, Tatia (1951, pp. 268-280), JSK II: 245-247. 21 See Dixit (1971, pp. 14-16). 22 On the development of epistemological models see: Balcerowicz (2016c). 23 Most philosophical texts begin with the discussion of the epistemic means to know and justify main claims of the particular system, or with a pramāṇa section. It is only after the proper instruments of knowing are established that the enquiry moves to other issues, such as Dharma, liberation, suffering etc. preserve and retrieve the canonical books (pustakârohaṇa). 24 In the process, 'not only were the works written formerly at Mathurā and Valabhī again written and codified, but some more were written.' 25 The emendation, interpolation and update processes, which easily allowed for changes, were facilitated also by the fact that, prior to that moment, there was no rigidly established version of the main texts that would be preserved in writing, for 'those saints who kept Mss. were denounced' and in addition 'penances were prescribed to those who wrote even one letter.' 26 The copy-and-paste practice, which concerned either a recurrent group or list of connected terms or a paragraph (ālāvaga; Skt. ālāpa / ālāpaka) or descriptive passages (vaṇṇaā, Skt. varṇaka), was standardised in a way described by Kapadia (2000, pp. 59-60): 'Several vaṇṇaās which were occurring in more than one Āgama were written out at full length only once, and then they were not reproduced ad verbatim, a second time, but only a reference was made to them by writing the word vaṇṇaā, by indicating their source, by alluding to a parallel person or an object, by mentioning the words occurring in the beginning and the end or by writing the word jīva (sic; rather jāva-P.B.), a stenographic symbol. The ālāvagas, too, were similarly treated.' A good and traceable example of the process is provided by the versions of canonical writings redacted under the supervision of Nāgārjuna, the socalled Nāgārjunīya readings, but also Skāndilīya variants, obliterated at the third council of Valabhī. The Nāgārjunīyas are now altogether absent in the main canonical corpus except stray traces to be occasionally found in some commentaries. For instance, in his commentary Uttarâdhyayana-ṭīkā, Ś ānti-sūri occasionally makes use of the Nāgārjunīya recension. 27 Since the same passages, or blocks of texts, were stamped over earlier passages in various works, this makes it practically impossible to trace the actual development of terminology, not to mention the evolution of the concepts behind them. Much of what we can do is to attempt to reconstruct certain earlier stages overwritten with later increments and 'improvements', and the key to such a reconstruction can be for instance the evidence of incomplete obliteration of earlier stages, that is such cases in which textual revisions of particular passages are not complete, and some earlier terminology is still palpable. This is what undertake in this paper as well.
As I argue below, in early Jainism, we can distinguish five basic ontological models (with some variants) meant to enumerate fundamental ontological categories with which Jainism explained the complexity of the world as a background for its soteriological goal (mokṣa). Interestingly, the subsequent historical developments did not add much new to these models. They all are dualistic in their basic structure and divide all entities into two main types: living elements (jīva) and lifeless elements (ajīva). This distinction into jīva-ajīva seems to be the earliest and most basic ontological division, and it is practically omnipresent, found in all strata of early (and of course, later) Jaina writings. It served as the fertile ground for further developments in to main directions: the primarily ontological one (Models 1, and 3-24 On the redaction process, see Kapadia (2000, pp. 55-63). Cf. also Dundas (2002, pp. 70-73). 25 Kapadia (2000, p. 58). 26 Kapadia (2000, p. 56). 27 Chatterjee (2000, p. 230). 5*) and the primarily soteriological-moral one (Model 2), all of which appear to have branched off this original stem. This root character is the main reason, beside the ubiquity of both terms, why I consider the jīva-ajīva distinction as historically primary, earlier than another distinction jīva-pudgala (see below). I shall briefly describe them in the chronological order.
Model 1, in three reconstructable variants, due to is greatest simplicity and the least number of elements which were later expanded, has to be postulated as structurally the earliest explanatory structure, attested in oldest available Jaina sources, albeit it is nowhere listed exactly in its full and complete form as such, except for occasional references to its single components. The model simply enlists all the elements, without providing any justification, and reflects the earliest attested ontology of the Jainas, but does certainly not date back to the very beginnings of Jainism. It's fullest version could be suggested as follows: The above model is, I admit, nowhere mentioned explicitly in early Jaina writings in this precise form, but its existence has to be postulated for several reasons. One of them is that Model 3 (see below) presupposes it, the other is that some traces of it can unearthed from otherwise strongly unified (in terms of standardised and repeatable expressions, passages and descriptions) canonical corpus of Jaina writings. It also seems that this model successfully merged later on with Model 3, which introduces a similar categorisation, but based on the idea of the extensive entities (atthi-kāya / asti-kāya). Since, as I argue, the asti-kāya scheme (Model 3) became prevalent and dominant before the final redaction of the canon at the third council of Valabhī, the copy-and-paste redaction technique obliterated most original cases (but fortunately not all) in which the elements of Model 1 feature on their own (sc. jīva, dharma etc.), viz., without the subsequent appendage of ºasti-kāya (sc. jīvâsti-kāya, dharmâsti-kāya etc.), characteristic of Model 3. Model 1.2 serves as an explanatory background for a passage found in the canonical Book of Interpretative Positions (Ṭhāṇ'aṁga-sutta). The book itself is practically impossible to date because, by its nature, it classifies various entities numerically in an increasing order, a manner similar to the Buddhist canonical Aṅguttara-nikāya, or the Gradual Collection, in which each successive classification of items is increased by one. Due to its structure, new textual layers could easily be added at any time and earlier readings modified, any particular portion of the text can hardly be assigned to any particular time frame. The text as a whole certainly belongs to the final canonical stage of the fifth century. 28 However, its particular portions are certainly centuries earlier. The passage in question lists important constituent elements of all that exists, and provides some reason why these elements exist: [4] It has never happened nor happens nor will happen this way that living elements (jīva) will become lifeless elements (ajīva), and that lifeless elements will become living elements -also in such a manner the continuity of the world as one has been propounded.
[6] It has never happened nor happens nor will happen this way that the [inhabitable] world space (loka) will become the [uninhabitable] non-world space (aloka) or that the [uninhabitable] non-world space (aloka) will become the [inhabitable] world space (loka) -also in such a manner the continuity of the world as one has been propounded. [7] It has never happened nor happens nor will happen this way that the [inhabitable] world space (loka) will penetrate into the [uninhabitable] non-world space (aloka) or that the [uninhabitable] non-world space (aloka) will penetrate into the [inhabitable] world space (loka) -also in such a manner the continuity of the world as one has been propounded. [9] As long as the stretch for movement of living elements (jīva) and matter particles (pudgala) exists, so long there exists the [inhabitable] world space (loka); as long as the [inhabitable] world space exists, so long there exist the stretch for movement of living elements and matter particles -also in such a manner the continuity of the world as one has been propounded.
1 Ṫ hāṅ 1,2 : etaṁ. 2 Ṫ hāṅ 1,2 : evaṁ pegā. 3 Ṫ hāṅ 1,2 : gati-paritāne. 4 Ṫ hāṅ 1,2 : lukkhattāte. verses 9-10 explain, what keeps the two kinds of space, the inhabitable world (loka) and the uninhabitable non-world (aloka) apart is the same principle that is responsible for the movement of living elements (jīva) and matter particles (pudgala) within the inhabitable world (loka). Thus, the text provides some explanation for the role of dharma and adharma, (without the terms themselves being mentioned explicitly) understood as ontological principles that facilitate movement and rest of jīva and pudgala within the realm of the inhabitable world. 31 Model 1.2 goes back to the period the 2nd-3rd century CE, i.e. to the time when most probably, first, the expressions dharma (dhammo) and adharma (ahammo) adopt their additional, kinetic meanings, beside the standard of righteousness (dharma) and unrighteousness (adharma), namely the principle of motion and the principle of rest; and second, when the term pudgala is adopted by the Jainas in the particular Jaina meaning of 'matter' or 'matter particles'. It seems that the model was subsequently absorbed into later models or accordingly redrafted and assimilated ('updated') with new revisions of texts, which may be the reason why it is no longer preserved in its full-fledged form, but can easily be deduced from texts.
The model is also preserved in an old portion of the Sayings of the Seers (Isibhāsiyāiṁ), a passage formally ascribed to Pārśva, which contains a reference to three basic elements only: the living beings (jīva), lifeless elements (ajīva), matter (pudgala), and perhaps also space, implied by the term 'world' (loka). Neither the passage in question nor the whole text of The Sayings of the Seers know the dyad of dharma and adharma in their latter kinetic meanings. However, the principles of motion and rest may there be indirectly implied -as in the above case of the Ṭhāṇ'aṁga-sutta -through the idea of the upward and downward movements (gati: ūrdhva-gamin, adho-gamin) of living beings and matter particles: [1] What is this world? [The world] is the living element (jīva), the lifeless element (ajīva This and similar passages could be the source of the later concept of the twin kinetic principles, more pronounced in the above quoted Ṭhāṇ'aṁga-sutta section than in the Isi-bhāsiyāiṁ. I fail to find any traces of the terms dharma and adharma in their kinetic meaning which may be older than the beginnings of the common era. It appears that Model 1.2 may have had its predecessor (Model 1.1) at some stage (see below): 31 This model is alluded to also elsewhere with the sequence of jīvā ya poggalā ya, e.g. in Ṫ hāṅ 1  The term pudgala-Pāli puggala or puṁgala-is well attested in the early strata of Buddhist writings in the sense of 'person', 'personality', provisional individual entity embedded in the saṁsāra and correlated with the existence of the aggregates (skandha), or a transitional, phenomenal basis of consciousness to which the attributes of a person are ascribed. 34 It is distinguished from the idea (postulated by other systems) of an eternal self, permanent soul, or ātman, being the perduring, individual, conscious substratum that transmigrates and may become liberated at some point, the existence of which Buddhism vehemently denied from its beginnings. As Tedesco (1947, p. 172) rightly points out, 'the term no doubt belonged already to the Buddhistic Proto-Canon, and probably had there the form *puggala-.' As such, it should therefore be considered mainly of Buddhist, rather than of, generally speaking, Ś ramaṅic provenance, for there is no evidence for it in early Jaina texts at all (unlike the term ajīva). There is hardly any reception of the term in non-Buddhist systems of thought for centuries either. 33 Cf. Ohira (1994, p. 113 § 257): 'The word jīva distinguished from ajīva occurs decisively in the third canonical stage, where ajīva denotes pudgala on the whole. The usage of ajīva in the sense of asti-kāyas minus jīva and addhā-samaya arises in the fourth-fifth canonical stages.' 34 See, e.g., Collins (1982, pp. 160-165). Cf. also ChÂ u (1999, pp. 130-143).
The term pudgala is uniquely attested in the Caraka-saṁhitā (Śarīra-sthāna) in a strongly Sāṁkhya-influenced passage that dates to the period ca. 100-200 CE 35 and that collects various synonyms to designate 'the soul', or the inhabitant of the body, with a longer list of synonyms to follow, among which we also find pudgala, a clear reminiscence of the Buddhist term in this particular context. 36 This can be the time when the term leaves the strictly Buddhist domain and gradually infiltrates other systems. It is most plausible that the Caraka-saṁhitā evidence coincides with the time when also the Jainas gradually adopted the term in a new peculiar meaning. Therefore, we may assume that the Jainas may have adopted the term pudgala in a meaning similar to the Buddhist one around the same time.
An evidence for the adaptation process is provided by a fragment from the Viyāha-pannatti, which in all probability belongs to the first two centuries of the common era. 37 This is a crucial transitional passage-before pudgala ultimately assumes its other meanings of 'matter', or 'material stuff', 'atomic matter', and, later on, becomes 'matter' as one of the dravyas, 'substances' incorporated into Model 5. The passage in question, which clearly displays the Buddhist influence and marks a stage when the term is adopted by the Jainas, concerns the relevant question what the difference between the living jīva and pudgala (which I translate below as 'stuff') is: [59. Indrabhūti Gautama asks:] Sir, is the living being a stuff-possessor (pudgalin) or the stuff (pudgala)? [Mahāvīra replies:] Oh, Gautama, just as [a person] with an umbrella is an umbrella-possessor, [a person] with a stick -a stick-possessor, [a person] with a pot -a pot-possessor, [a person] with a cloth -a cloth-possessor, [a person] with a hand -a hand-possessor, exactly in the same manner, Gautama, in dependence on (with respect to) the sense of hearing, the sense of vision, the sense of smell, the sense of taste [and] the sense of touch, [the soul (jīva) is] a stuff-possessor (pudgalin), [and] in dependence on (with respect to) the soul (jīva) [itself, the soul is] the stuff (pudgala). … Gautama, in this sense, in dependence on (with respect to) the soul (jīva) [itself, the soul] is not a stuff-possessor, [but it] is the stuff [itself]. … [61. Gautama asks:] Sir, is the liberated living being (siddha) a stuffpossessor (pudgalin) or the stuff (pudgala)? [Mahāvīra replies:] Oh, Gautama, the liberated living being is not a stuff-possessor (pudgalin, poggali) or the 35 Cf. Meulenbeld (1999: 'The version of Sāṁkhya as found in the Carakasaṁhitā is earlier than that of the Sāṁkhyakārikā, which belongs to about the period A.D. 350-450, and may belong to the period of Pañcaśikha and the Ṣaṣṭitantra, i.e. about 100 B.C.-A.D. 200.' 36 Car 1,2 4.4.8 = Car 3 4.4.7: hetuḥ kāraṇaṁ nimittam akṣaraṁ kārtā mantā veditā boddhā draṣṭā veditā boddhā draṣṭā dhātā brahmā viśvakarmā puruṣaḥ prabhavo 'vyayo nityo guṇī grahaṇaṁ pradhānam avyaktaṁ jīvo jñaḥ pudgalaś cetanāvān vibhur ātmā cêndriyâtmā cântarâtmā cêti. 37 Ohira (1994, p. 58) places it in the third canonical stage, or between 1st cent. BCE/1st cent. CE and 3rd cent. CE. stuff (pudgala, poggala). Gautama, in dependence on (with respect to) the soul (jīva) [itself, the soul] is [the stuff itself]. 38 The reply indicates that the soul always -whether in an embodied state within the saṁsāra or as a liberated being -remains 'the stuff' (pudgala), or a concrete individual, endowed with its separate identity, individual history and consciousness. 39 This is precisely the meaning which closely approximates the Buddhist meaning of the term. At the same time, the embodied soul, unlike the liberated one, is a stuff-possessor (pudgalin), that is, endowed with material stuff, of which its body is composed: the soul is the owner of the matter, or material body.
Once both (1) pudgala is established in its meaning as 'matter' and (2) the term for atoms, paramâṇu, is adopted, it is only later that these two ideas are combined into the concept of 'atomic matter' or 'matter that consists of atoms' (paramâṇu-poggala). The idea of matter per se does not logically require the idea of atoms, as instantiated by the tradition of Sāṁ khya-Yoga, which knows the former, but not the latter, so these two should be clearly distinguished, and their combination into one unified idea is not logically required. The designation for atoms, paramâṇu, enters Jaina ontology most probably some time in the second-third centuries CE under the influence of the tradition of Vaiśeṣika/Ā nvīkṣikī atomism, which knows the idea and the term, as attested in the Vaiśeṣika-sūtra (VS(C) 4.1.7), and in the Caraka-saṁhitā (Car 1 4.7.17 = Car 2 4.7.17 = Car 3 4.7.18). With this, I do not intend to say that the Jainas adopted the very idea of atoms. They must have known it before, because we have some palpable evidence that its twin-system of Ā jīvikism developed this idea very early, as attested for instance in the Sāmañña-phala-sutta, and it is most likely that the Jainas used this concept as well. 45 The Jainas merely adopted the designation itself, a term which was a standard at that time, as they did with another term, pramāṇa, and many others. The combination of the two conceptually rather distinct concepts, matter (pudgala) and atoms (paramâṇu), happens in probably the third century CE, and thereafter matter is primarily understood of atomic nature, and the standard term is paramâṇu-poggala/ paramâṇu-pudgala, as attested on numerous occasions. 46 What follows is the division of matter (pudgala) into atoms (paramâṇu) and, as a result, into parts (deśa) and, technically speaking, spatial units (pradeśa) is introduced. Parts are natural divisions of any entity, but a novel technical term is pradeśa, or spatial unit, and this marks a new development in the understanding of the structure of matter.
We can detect, as I suggest, the following development: at a certain stage (the beginnings of the common era, which may coincide with the external evidence of the Caraka-saṁhitā), the Buddhist term pudgala enters Jaina vocabulary still retaining its Buddhist meaning of 'the person', 'an embodied individual self'. In the next step, the meaning of 'the person' is relinquished, because of its semantic overlap with two other terms: jīva and ātman (attā, āyā), and pudgala comes to refer to either (1) 'the flesh' or 'material stuff' of the body alone, or (2) to the body itself, being the material part (pudgala) of the individual (pudgala), who is at the same time the stuff-possessor (pudgalin). 47 It will occur only later, after second/third 44 Similar meaning of pudgala is found also in Viy 2 1.6.77, Viy 2 2.5.127, 19.7.739 = Viy 3 1.6.26 [1], 19.7.2. 45 Balcerowicz (2016b, pp. 126-136), chapter '3.2.1.6. Atomism and Elements' in Ā jīvikism. 46 For instance Viy 1.10. 1, 2.10, 5.7-8, 6.7, 8.2, 8.9-10, 9.1, 12.4, 12.7, 12.10, 14.4, 17.7, 16.8, 18.4, 18.6, 18.8, 18.10, 20.2-3, 25.3-4, not to mention other canonical books. Ohira (1994) assigns these passages to the third canonical stage, or '1st century B.C./1st century A.D.-3rd century A.D.' 47 See Schubring (2000, p. 132, § 59): 'poggala, the word for atom, applied in its basic meaning as a "concrete body" in Viy 176b, whereas in 240 it means the part of an individual.' century, under the influence of Vaiśeṣika ontology, especially atomism, that pudgala will assume its final, technical meaning of 'matter', esp. 'material atom'. Model 1.3, structurally speaking, should be understood to be a simplified version of -and therefore later than-Model 1.2, when it was discovered that the category of ajīva became redundant, since it overlapped with the matter, space, principle of motion and principle of rest. Certainly, it is possible to argue the other way round, namely that Model 1.2 is an extended version of Model 1.3, and should be considered later. Such hypotheses, however, cannot explain two transitions: the first transition is from Model 1.1, which already contains the category of ajīva, to Model 1.3 without the term (whereas Model 1.2 would have to add it again); and the second transition, from Model 1.2 (with ajīva as a meta-category comprising its four subdivisions) to Model 3, that of the five extensive entities (atthi-kāya / asti-kāya), which normally does not include the extensive entity of the lifeless element (*ajīv'atthi-kāya / *ajīvâsti-kāya). My hypothesis does not have this weakness.
Further, when we assume that Model 1.3 was conceptualised around third century under the influence of Vaiśeṣika ontology, we can retrogressively date Models 1.1 and 1.2 as earlier. Accordingly, we can recapitulate the above discussion that Model 1, with its three variants developed over the period of roughly three centuries, namely, between the first and third centuries CE.
The section is repeated both in Viy 2 11.10.510 = Viy 3 11.10.15-16 and in Viy 2 20.2.781 = Viy 3 20.2.2 with an explicit mention of the previous section of Viy 2.10. of motion (dharma) and principle of rest (adharma). Again, this corresponds to Model 1.2. Another useful evidence is provide by Digambara writings. The Canon in Six Parts, in a chapter On dialectical ways of analysis of bondage (Baṁdhaṇâṇiyogaddāra; Ṡ KhĀ 5.6) describes the elements involved in the process of bondage (baṁdhana), which is an alternative description for the impact of karman (kammavibhāsā), and-in a sequence of sūtras-mentions the following elements relevant for Model 3, however with conspicuous omissions: jīva (Ṡ KhĀ 5.6. 7,[13][14][15][16][17][18][19], ajīva (Ṡ KhĀ 5.6.7, 20-23), dhamm'atthiyā adhamm'atthiyā āgās'atthiyā (Ṡ KhĀ 5.6.30-31) and poggalā (Ṡ KhĀ 5.6.34). As earlier, we can also here observe that the three elements dhamm'atthiyā adhamm'atthiyā āgās'atthiyā were reworked whereas the remaining ones-viz. jīva, ajīva and poggala-were not. It was possible to change dhamma in this passage (but often in other, similar passages) to dhamm'atthiyā, adhamma to adhamm'atthiyā and āgāsa to āgās'atthiyā because all the three occur exclusively as main terms, i.e. never occur in compounds, whereas jīva and ajīva frequently also feature as members of a compound (e.g. jīva-bhāva-baṁdho, ajīvabhāva-baṁdho) which prevents an easy supplementation with ºatthiyā, automatically appended at the end, after the respective term. Also the commentator Vīrasena is aware of this absence in the case of the remaining standard asti-kāyas. 53 The actual model behind this enumeration is Model 1.2.
A very similar discussion is repeated (copied-and-pasted) in another section of the Viyāha-pannatti (Viy 2 10.1.475 = Viy 3 10.1.8-9) with respect to cardinal directions (disa/diś), an equivalent expression for 'space' (ākāśa). As in the preceding paragraph on the composition of ākāśa/āgāsa, also here the elements found in the space are mentioned as: jīva and ajīva, with their spatial elements, or parts (deśa) and spatial units (pradeśa): jīva, jīva-desa, jīva-padesā, and ajīva, ajīvadesa, ajīva-padesā. The latter is then subdivided into the perceptible (rūvī/rūpin) matter (poggala), or matter-stuffs in the form of atoms (paramâṇu-poggalā), and the imperceptible elements (arūvī/arūpin) that bifurcate into the extensive entity of the principle of motion (dhamm'atthi-kāe) and the extensive entity of the principle of rest (adhamm'atthi-kāe), both with their detailed subdivisions into parts and spacial units (desā/deśa and padesā/pradeśa). As in the case discussed above, also here phenomenal time (Pkt.: addhā-samae, Skt. *adhvan-samaya/addhā-samaya) is appended at the very end of the with no elaboration (unlike all the remaining items), which is an evidence that the passage was reworked at a later stage, when the redactor felt obliged to include time, recognised as a separate substance (dravya) in the fifth century CE. 54 The ontological scheme behind this enumeration is again Model 1.2. This section is, and rightly so, classified by Ohira (1994) as belonging to the fifth canonical stage, or 4th/5th century, before (as I argue) the framework of substance (dravya), characteristic of Model 5, was added, however-as we can clearly see-the passage partly preserves earlier terminology.
To contrast these incomplete enumerations with a similar standard list of the extensive entities found regularly elsewhere, we see that, on numerous other parallel occasions, parts (deśa) and spatial units (pradeśa) of all the extensive entities are discussed, except phenomenal time (addhā-samaya), and these passages specify the items in extenso as asti-kāyas, viz. dhamm'atthi-kāya-desā adhamm'atthi-kāya-desā āgās'atthi-kāya-desā jīv'atthi-kāya-desā poggal'atthi-kāya-desā or dhamm'atthikāya-paesā adhamm'atthi-kāya-paesā āgās'atthi-kāya-paesā jīv'atthi-kāya-paesā poggal'atthi-kāya-paesā. 55 A typical instance is provided by a Viyāha-pannatti passage that-like the few passages mentioned above that partly preserve earlier terminology-also discusses the structure of the space and of the world, and specifies that there are five extensive entities, and each of them is listed as atthikāe. 56 These instances offer the enumerations when the update process was completely successful, that is, all terms in the sequence were appended with ºatthikāya, unlike the few stray passages mentioned before, which preserve earlier readings.
Another indirect evidence is provided by later literature which occasionally lists the asti-kāyas in a manner corresponding to this model, as does Nemicandra who first defines the extensive entities (DS 24), and then enumerates them without the ºatthi-kāya element: There are practically two variants of this model: the shorter with seven items, and the extended one with the addition of two more, merit and demerit. The model rests on the same dichotomy of jīva-ajīva as variants 1.1 and 1.2 of Model 1. Indeed, like all other Jaina models, also this one is clearly dualistic, however in this particular case all the categories characterise a linear sequence of complex developments that relate the two basic categories of sentient jīva and insentient ajīva, an 'evolution' which begins with establishing a relation between the two in the form of the influx of karman and ends with a complete liberation of the soul from the fetters of all lifeless elements. Two aspects call here for special attention. First, we can observe certain analogy to the Sāṁ khya-Yoga dualistic model, known from the earliest developmental stages of this system. In it, the relation between two principles of conscious/sentient puruṣa (a principle of consciousness) and insentient prakr ti (a principle of matterand-activity) triggers a linear process of evolution through a number of stages, such as → objective consciousness (buddhi), or the awareness of objects other than consciousness, → subjective consciousness (ahaṁkāra), or the awareness of consciousness itself, → the inner instrument (manas) that coordinates sensory and other data and → the elements (mahā-bhūta) that make up all the graspable objects, with some modifications at various stages. Once the link between the two main principles is removed, liberation follows, known as kevala or kaivalya, that is separation, exclusivity or detachment of the two principles. Similarly in Jainism, structurally the model is quite similar in the sense that all other categories that ensue from the linkage between the sentient jīva and the insentient ajīva represent a gradual step-by-step evolution till the moment of final separation of the two, which is liberation. The precondition for saṁsāra is the inter-linkage of jīva and ajīva, the course of saṁsāra is determined by the subsequent categories and the end of saṁsāra, known as liberation, is precisely a complete disengagement, or cessation of any links between both principles.
Second, it seems probably no coincidence that the liberated soul is known in Jainism as kevalin, viz. the one who has accomplished 'the absolute isolation' or 'unique (perfect) knowledge', known as kevala, being also a foundational term for the tradition of Sāṁ khya-Yoga and epitomised later as 'pure, unique knowledge' (SK 64cd: viśuddhaṁ kevalam … jñānam). In the proto-Sāṁ khya stage, 58 as attested in the layers of the Mokṣa-dharma-parvan, we find the expression of kevalaṁ jñānam on a number of occasions. 59 Both these models are structurally very similar: both begin with two separate, irreducible categories of the sentient and the insentient, and-through a sequence of processes that are experienced in a series of rebirths in the saṁsāra due to a linkage between the two-culminate in the state of their separation, which defines what liberation is.
These two structural affinities may point to an important historical dependence of Jainism on proto-Sāṁ khya-Yoga. The roots of this model most probably go back to the beginnings of Jainism, inasmuch as the model includes some technical terminology attested in earliest portions of Jaina texts (jīva and ajīva). However, in the developed form as textually attested, the model assumed its final form much later, perhaps 1st-2nd centuries CE or even slightly later. A prior existence of a linear model of evolution of secondary categories other than the pair of primary ones in proto-Sāṁ khya-Yoga may have helped organise various items that related jīva and ajīva in Jainism into a similar 'evolution' model. The model combines the concepts that often circulated independently, such as the ultimate goal of existence, viz. liberation (mokṣa), the pairs jīva-ajīva, puṇya-pāpa or āsrava-saṁvara linked to the ideas of karman, karmic bondage (bandha) and its eradication (nirjarā). At some point of time, all these elements were brought together into one model, which gradually evolved to comprise either seven 60 or nine elements 61 .
This model clearly derives from elements of the ascetic practice that should finally lead to liberation, and contains terminology related to the salvific path. The particular elements that jointly make up this explanatory scheme seem early and individually often occur in historically early passages of the canon. Occasionally, they do occur jointly in one text, albeit not necessarily listed in a linear sequence. For instance, they are found in the ninth book, Lesson of Mahākāsava (Mahākāsav'ajjhayaṇaṁ), of the Isi-bhāsiyāiṁ. It is a relatively early text (first century BCE/first century CE?), containing no technicalities, no specialised terminology, based on very simple images, devoted to the ascetic path, as described by Mahākāśyapa, and taking karman in an early, popular, non-technical meaning of 'deed, act that brings about future retribution' or 'results of former deeds that are to 58 Cf. Larson (1979, p. 72 take fruition', not yet linked to the specialised idea of matter (pudgala), or karmic matter. The verses practically contain all nine standard elements of Model 2, some of them mentioned a few times, except that (1) āsrava/āsava, 62 or influx of karman, is conveyed with terms upādāna and ādāna, 'accumulation/acquisition of karman' (v. 9), (2) mokṣa is likewise not mentioned by name directly but is described as 'the highest eternal and unchangeable state' (v. 31), and (3) ajīva is not directly mentioned. The most relevant verses read as follows: [4] Control (saṁvara) and eradication (nirjarā) [lead to] the destruction of of merit (puṇya) and demerit (pāpa). One should properly practice control and eradication in every way. … [9]  The term upādāna (uvāyāṇa) is well known from Buddhist context, for instance as the ninth element, 'clinging, grasping', of doctrine of dependent origination (pratītya-samutpāda). Perhaps, upādāna was the earlier term to denote the accumulation of karmic results that was subsequently replaced with purely Jaina technical term āsava/āsrava.
In a later section of the Sūya-gaḍaṁga, and in a passage of the identical wording found in the Viyāha-pannatti, the model becomes a part of the edifying knowledge of the best among the laypeople, who become the followers of Jaina ascetics, who have understood [what] living beings and lifeless elements [are, and] who have grasped merit and demerit, who are expert as regards the influx and control (stoppage) [of karmic matter], the experience [of the results karmic matter], eradication [of karmic matter], the subject of actions (i.e. the self), the bondage and liberation. 65 As Ohira (1982: 168-169, § 435) notes, the Jaina layperson was expected to know elements of this model at the final canonical stage (latter half of 4th-5th centuries CE), which indicates that the model was well established in the whole Jaina community by that time to the extent that even the lay people were expected to know it.
It seems that all earliest references to Model 2 mention the categories of merit (puṇya) and demerit (pāpa), which means that the seven-link version is a case of later simplification of the model, in which these two items were found redundant and removed.
Model 3 is quite conspicuous due to an addition of a novel and unique concept of extensive entities (asti-kāya), which become a descriptive framework for all the categories. In it, the living beings are introduced side by side with expressly enumerated sub-varieties of lifeless elements (ajīva), without the term ajīva being explicitly mentioned. extensive entities (atthi-kāya / asti-kāya) as 1. principle of motion (dhamma°/ dharmâ°) 2. principle of rest (adhamma°/ adharmâ°) 3. space (āyāsa°/ ākāśâ°) 4. living beings (jīvâ°) 5. matter (puggala'°/ pudgalâ°) The model has already been discussed above in the context of the reconstruction of the variants of Model 1, which-as I argue-were replaced with it in the fifth century, after Model 3 set the standard. The main feature of Model 3 is the addition of the term (and idea) of atthi-kāya (S. asti-kāya), or 'extensive entities'. The popularity of the model in the fifth century contributed to the fact that, at the time of the final redaction of the Ś vetāmbara canon in Valabhī, it overrode the previous versions of Model 1 with a simple addition of ºatthi-kāya to all the items of the list. Model 3 often serves to project the a new main ontological opposition: jīv'atthi-kāya and poggal'atthi-kāya, 66 which for a while replaces the former division of jīvaajīva, with the idea of ajīva, and consequently ajīv'atthi-kāya/ajīvâsti-kāya, no longer required.
The term atthi-kāya occurs in the Sayings of the Seers (Isi-bhāsiyāiṁ) only once, in an interpolated portion that in all probability belongs to the youngest layers of the text, perhaps 4th century CE. Incidentally, the thirty-first hymn, in which the interpolated idea occurs, contains one of the oldest historical layers of the whole collection, going back perhaps even to Pārśva himself. Both the complexity and the way the terms are introduced (se jahā nāmate) may suggest that the concept of atthikāya was perceived to be a novelty: It never happens that the world did not exist, it never happens that the world does not exist, and it never happens that the world will not exist. [The world] existed, exists and will exist. The world is unending, eternal, everlasting, indestructible, perpetual, permanent. And so are the so-called five extensive entities: it never happens that they did not exist \etc. … up to [ [they are] permanent. And thus is the world: it never happens that it did not exist \etc. … up to [ [they are] permanent. 67 Clearly, the italicised mention of the five extensive entities is appended at the very end of the hymn, which had originally been completed with the enumeration of the 66 E.g. Viy 2.7,2.10. qualities of the eternal world ('unending, etc.'): once the atthi-kāya interpolation is removed, the text not only retains its integrity but turns out even more consistent. 68 This whole passage, including the much later interpolated part, can be regarded as a genuine beginning of Jaina ontology, i.e. Jaina reflection on the actual structure of the being, not simply on the structural layers of the universe (loka-aloka, etc.), which rather belong to the sphere of cosmology.
Apart from a large number of references to asti-kāyas in the Ś vetāmbara canon, it contains numerous detailed descriptions and discussion of all the five, 69 that is in this particular sequence: dhamm'atthi-kāya, adhamm'atthi-kāya, āgās'atthi-kāya, jīv'atthi-kāya and poggal'atthi-kāya. The idea of the five extensive entities is also well known to the Digambara authors of The Canon in Six Parts (Chakkhaṁḍâgame), which occasionally mentions some elements of this model (without extensively enumerating them in a list), such as for instance dhamm'atthiyā (dharmâsti-kāya), adhamm'atthiyā (adharmâsti-kāya), āgās'atthiyā (ākāśâsti-kāya) (Ṡ KhĀ 5.3.30), and which also applies these ideas in its analysis of the world.
The category of 'extensive entity' (Pkt. atthi-kāya, Skt. asti-kāya), or 'the body which is there', does not occur in any other philosophical system in India and is specific to Jainism. It was an attempt to classify basic structural elements of being that are not further reducible to any other category, such as ajīva ('lifeless element'). 'Extensive entity' is a generic term which covers the quadruplet of the principle of motion, the principle of rest, space and matter. Also this structural element, i.e. irreducibility of the five extensive entities, similar to the ontological categories in the Vaiśeṣika system or elementary constituents of reality (dhamma, dharma) in the Abhidharma Buddhism, attests to the concept of the five extensive entities being a later development.
With Model 3-with which Jaina philosophers conceived of a generic abstract idea, or meta-category (asti-kāya), to cover the actual categories as components of the world (present also in variants of Model 1), which was a philosophically vital step of abstraction-their ontology began to develop more seriously and independently of other philosophical systems and external influences. Generally, it seems, the beginnings of Jaina ontology, also including Model 1, were not significantly dependent on other systems, and this conclusion is buttressed for instance by the particular usage of such terms as dhamma/dharma and adhamma/adharma adopted by the Jainas as principles of motion and rest, quite distinct from general understanding of these terms in ancient India. The only dependence on other ontologies is conspicuously detectable in Model 2, which apparently adopted certain 'evolution' scheme from proto-Sāṁ khya-Yoga, with its contents remaining independent of external influence.
matter, being one of the extensive entities, happens to be mentioned as a kind of substance (davva, dravya), which seems again to be a later interpolation introduced by the compilers in order to update the earlier classification to the standards of the mid-fifth century when the canon was finally redacted and the concept of substance and the corresponding term dravya found their way in Jaina ontology (see below). This attests to the historical antecedence of the extensive entities featuring in Model 4 with respect to the later Model 5 (based on the idea of dravya) and to the introduction of the scheme of substances into Jainism at a later point of time.
The same passage of the Lecture of Explanations contains also a unique explanation of this novel and unusual term asti-kāya: Then this venerable [Indrabhūti] Gautama said to these representatives of other systems as follows: "Verily, we do not say that existence (asti-bhāva) does not exist (nâsti); we do not say that non-existence (nāsti-bhāva) exists (asti). O, Dear to Gods, we say that every existence (asti-bhāva) exists (asti); we say that every non-existence (nāsti-bhāva) does not exist (nâsti). 72 The short passage is directly appended to the discussion of the asti-kāyas, and the two equivocal terms-asti-bhāva and nāsti-bhāva-are meant to explain the nature of the asti-kāyas: these are the truly existent entities (bhāva) of which existence can genuinely be predicated of (asti-bhāva). The opposite of the asti-kāyas is that which is absolutely non-existent, entities (bhāva) that do not exist at all and of which nonexistence is predicated of (nāsti-bhāva). This explanation rests on the ideal of the correspondence theory of truth: what exists has to be spoken of as existent, what does not exist is necessarily to be declared non-existent, and such is a true speech. What is indirectly implied is that the set of the asti-kāyas, being the existent entities, exhausts all that exists (asti-bhāva).
A observation of similar kind is found in other works, for instance in the Dravyasaṁgraha of Nemicandra, who also relates the name asti-kāya to existence, but its historical value is much lower inasmuch as this kind of etymologising does not reflect the historical genesis of the term asti-kāya at all, and therefore such kinds of explanations should not be retrospectively projected onto the historical beginnings of the term: Since these exist, the supreme victors (jina) say that these exist like bodies (kāya) that have numerous parts, and because of that these are extensive entities. 73 A separate question is what the origins of the term could have been. I have no answer to this. This would probably require much more material for a reliable historical and semantic analysis how this term could have originated. What is 72 Viy 2 7.10.377 = Viy 3 7.10.6 [2]: tae ṇaṁ se bhagavaṁ goyame te anna-uttie evaṁ vayāsīno khalu devâṅuppiyā atthi-bhāvaṁ natthi tti vayāmo, natthi-bhāvaṁ atthi tti vayāmo. amhe ṇaṁ devâṇuppiyā savvaṁ atthi-bhāvaṁ atthī ti vayāmo, savvaṁ natthi-bhāvaṁ natthī ti vayāmo. 73 DS 24: saṁti jado teṇ'ede atthîti bhaṇaṁti jiṇa-varā jamhā / kāyā iva bahu-desā tamhā kāyā ya atthi-kāyā ya // relevant in this context is, however, that the earliest Digambara sources provide an alternative reading of the terms ending with ºatthiya (not with ºatthi-kāya), namely dhamm'atthiya, adhamm'atthiya, āgās'atthiya, as well as jīv'atthiya and poggal'atthiya, and such a reading occurs also with the terms being primary members of compounds. 74 These could be Sanskritised either as dharmâstika or dharmârthika, etc., but certainly not dharmâsti-kāya, etc. Such an alternative reading should also be taken into consideration in any further analysis. Somehow, the terms in this form structurally resemble terminology related to the epistemological realm, namely the two viewpoints (naya): davv'atthiya and pajjāv'atthiya, Sanskritised either as dravyârthika-paryāyârthika (substance-and mode-expressive viewpoints) or as dravyâstika-paryāyâvastika (substantial / attributive view points), of much later provenance. Later Digambara tradition consistently reads atthi-kāya (alternatively metri causa: atthī-kāya), as attested for instance in Nemicandra's works. 75 Model 3* is found in the first Jaina Sanskrit work, in Umāsvāmin's Tattvârtha-sūtra (TS 5.1), composed around 350-400 CE. I clearly differentiate between Umāsvāmin (c. 350-400), the author of the Sūtra, and Umāsvāti (c. 400-450), the author of the Tattvârthâdhigama-bhāṣya (TBh), both of the Gupta period. 76 It is clearly a later version or an elaboration of Model 3, and we can call it Model 3*. Its distinct feature is that it completely leaves out living beings (jīva) and seems to suggest that these were no longer classifiable as extensive entities (jīvâsti-kāya) by that time, and instead the model focuses on lifeless extensive entities (ajīvâsti-kāya): as living being (jīvâsti-kāya). However it does not, because, first, the idea of living extensive element (jīvâsti-kāya) is nowhere mentioned by name or alluded to in the whole Tattvârtha-sūtra, and second, it is excluded by the next aphorism (TS 5.2; see below), which explicitly includes living beings (jīva) among substances (dravya), but not among extensive entities. The Tattvârtha-sūtra classification is reflected also in the later strata of the Ś vetāmbara canon. An appendage to the same passage of the Lecture of Explanations quoted earlier in the context of Model 3 introduces the idea of ajīvâstikāya: 'Thus, [Mahāvīra] Jñātrputra teaches that there are four extensive entities which are lifeless extensive entities.' 78 A transitory scheme resembling Model 3* is also presented in a later portion of the The Later Chapters (Uttarajjhayaṇa-sutta, Uttar 36.5-6), appended at the end of the whole work, after two verses specifying the well-known and much earlier established dualistic system of living beings (jīva) and lifeless elements (ajīva) as constituting the world (loka): [1] Now, hear from me, with focused mind, about the division into living beings (jīva) and lifeless elements (ajīva), after having known which, a monk succeeds in self-discipline (samyama). [2] Living beings (jīva) as well as lifeless elements (ajīva) are called "this world". The space (ākāśa) which is the place for [only] lifeless elements (ajīva) is called the non-world (aloka). … [5-6] (a) The extensive entity as the principle of motion (dharmâsti-kāya) alongside (b) its parts and (c) its spatial units, (d) [the extensive entity as] the principle of rest (adharmâsti-kāya) alongside (e) its parts and (f) its spatial units, (g) space (ākāśa) alongside its (h) parts and (i) its spatial units as well as (j) phenomenal time divided into units (addhā-samaya) -[these are] the ten imperceptible entities. 79 The model is conspicuous for at least two features. First, it confirms our understanding of the Tattvârtha-sūtra, in which the concept of extensive entity (asti-kāya) was by that time limited to lifeless elements (ajīva), because all varieties of extensive entities were presented as subcategories of lifeless elements. Second, the classification leaves material extensive entity (pudgalâsti-kaya) out, and introduces in its place a category of time under a technical term of addhā-samaya, or 'phenomenal time', i.e. not an abstract time as a substance but the time which is directly felt and consciously experienced as a temporal flow divisible into time units (moments). 78 Viy 2 7.10.377 = Viy 3 7.10.3: tattha ṇaṁ Ṇāyaputte cattāri atthi-kāe ajīva-kāe paṇnaveti. 79 Uttar 36.1-2, 5-6: jīvâjīva-vibhattiṁ suṇeha me ega-maṇā io / jaṁ jāṇiūṇa bhikkhū sammaṁ jayai saṁjame // 1 // jīvā cêva ajīvā ya esa loe viyāhie / ajīva-desam āgāse aloge se viyāhie // 2 // dhamm'atthikāe tad-dese tap-paese ya āhie / ahamme tassa dese ya tap-paese ya āhie // 5 // āgāse tassa dese ya tap-paese ya āhie // addhā-samae cêva arūvī dasahā bhave // 6 // Model 4 seems similar to Model 1, except that it, first, does not explicitly mention lifeless elements (ajīva) in its structure, but instead immediately sets out to enumerate their sub-categories alone, and second, introduces time under the generic term kāla, not the specific addhā-samaya ('phenomenal time'). One might argue that this may actually represent even and older scheme than Model 1, inasmuch as its form seems simpler. However, since it contains an important addition to the model, time (kāla), which belongs to much later historical strata than Model 1, we should rather assume that it is a later classification in a simplified version that merely happens to resemble the much earlier Model 1.

world components
Demonstration of All Things (Sarvârtha-siddhi, SSi). Model 5 lists five items, excluding time, whereas it's other version, Model 5*, adds time, known as kāla, not addhā-samaya ('phenomenal time'), as the sixth substance. substances (dravya) 1. living beings (jīva) 2. principle of motion (dharma) 3. principle of rest (adharma) 4. space (ākāśa) 5. matter (pudgala) [6. time (kāla)] The idea of substance (davva, dravya) does not occur even once in the Sayings of the Seers (Isi-bhāsiyāiṁ) or in oldest portions of canonical texts, which means that early Jaina ontology did not know the concept of substance and the term for it (davva, dravya) was gradually introduced into the early Jaina ontology later, under the influence of the early Vaiśeṣika and the Ā nvīkṣikī system. In the case of the Tattvârtha-sūtra, it is imposed on or appended to the earlier Model 3 of the extensive entities (asti-kāya) as an innovation.
Whether we accept the Ś vetāmbara cautious reading: 'According to some, also time is [a substance]' (TS 1 /TS 2 (TBh) 38: kālaś cêty eke), or the Digambara reading: 'Also time is [a substance]' (TS 3 (SSi) 39: kālaś ca), both demonstrate that already at the time of the final redaction of the text there existed an ontological model of substances which included also time, albeit its ontological status as a separate substance was not universally accepted. This in turn attests to the model's nascent phase around 400 CE, ergo the model did not exist before. The specification of time as substance (TS 38 or 39) is separated from the list of substances (TS 1-2 or 1-3) by three dozens of sūtras which may indicate that the insertion of time to the original text of Tattvârtha-sūtra happened at the redaction stage by the commentators (TBh, SSi) who integrated the new idea in the body of the sūtras. Model 5, which was adopted from the Vaiśeṣika ontology and which enlists substances, set the standard for future Jaina ontology. Its version, Model 5*, is found also in the canon, for instance in a Viyāha-pannatti passage that lists all the six substances, which comprise the five asti-kāya and the phenomenal time (addhā-samaya). 81 Interestingly, the new classification isintroduced into Jaina works by way of imposing iton a pre-existingmodel(usuallytheclassificationofextensiveentities,theasti-kāyas), typically by appending the new classificatory term 'substance' to an earlier section of a particular text and linking both models (say, the older model of extensive entities and the new one of substances) through one or more terms, such as 'matter'. For instance, the description of an earlier model, e.g. Model 3 or Model 3* of extensive entities (asti-kāya), concludes with 'matter' (pudgala), and then the text continues to explain that matter (as the last element of the older model) is, in addition, classifiable as one of substances, which are now all incorporated into the new Model 5. This is done also in Sanskrit works, precisely in the Treatise on the Reality (Tattvârtha-sūtra; TS 5.1-2) around 350-400 CE, in which Model 3* is extended to Model 5. First, four lifeless extensive entities (ajīvâsti-kāya) are enumerated (Model 3*; TS 5.1: ajīvā-kāyā dharmâdharmâkāśa-pudgalāḥ). Then (TS 1 / TS 2 5.2: dravyāṇi jīvāś ca; or TS 3 5.2-3 in SSi reading), these are equated with substances and the new scheme of substances (dravya) is said to also include living beings (jīva) that but no longer classified as extensive entities (asti-kāya). Subsequently, between 540-600 CE, in his commentary, Sarvârtha-siddhi, Pūjyapāda Devanandin explicitly adds time (kāla) as one more substance in its right place, i.e. where the substances are first discussed at the beginning of Chapter 5, and not towards its end (SSi 5.3: evam etāni vakṣyamāṇena kālena saha ṣaḍ dravyāṇi bhavanti). The introduction of time into the Treatise on the Reality (TS) happened, it seems, probably later, and did not belong to the original model found there.
What transpires from a comparison of all these models (see Recapitulation below) is that practically from the very beginnings of Jaina-consistently dualistic-ontology, based on the duality of living beings (jīva) and lifeless elements (ajīva), we deal with practically six basic categories-jīva, ajīva, dharma, adharma, ākāśa and pudgalawhich are sometimes extended with kāla (Models 4, 5*), sometimes abridged. A rather different trajectory is taken by Model 2. The main difference between all these models, which are almost identical, lies practically in how and, most importantly, under what class heading (tattva, asti-kāya, dravya) these basic categories are organised and classified. As I have tried to demonstrate, the minor changes, reflected in the class headings, developed over the period of approximately four centuries, perhaps even less, and came in a manner in which most of the actual ontological contents of a previous model was retained but a new heading was superimposed on it. It is, therefore, these class headings that crucially reflect historical developmental stages of basic structure as reflected in Model 1, as earlier demonstrated, and a reference to these headings alone (not to the components of a given model, which recur over centuries) can be treated as historical marker. The only exception to the rule is Model 2, which developed parallelly to all other models, and retained, as it seems, its original structure throughout the centuries. Like in the case of the gradual transformations of (the variants of) Model 1 to more complex models 3, 4 and 5, it is not impossible that Model 2 also had its earlier versions that may be grouped into a sequence of such gradual transitions from an earlier version to what we eventually classify as Model 2.
The actual application of these models can provide us with a useful method to determine the cohesion and integrity of a particular text or a passage and also its historical time reference. For instance, a mention of the classification of substances (Model 5/5*) in a particular passage sets an earliest possible time limit for it as ca. 400 CE. Further, a text which refers to extensive entities and is oblivious to the concept of substance cannot be later that ca. 400 CE. It is, of course, possible that two or more models can co-occur in one text. The best example is the Tattvârtha-sūtra itself, which lists Model 2, a variant (3*) of Model 3, and Model 5. In such cases, the latest model is the terminus post quem for the work. Further, the method can help us determine that a particular text is a compilation of various historical layers in case we find in it a continuous passage referring only to extensive entities with no mention of substances, and another passage basing its ontology on the concept of substance alone. Also the occurrence of time in a classification can be treated as a historical indicator.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creativecommons.org/licenses/by/4.0/.