Indian religion begins with the problem of suffering (dukkha). In Buddhism, suffering is thought to be caused by craving (taṇhā) and ignorance (avijjā). This has led Erich Frauwallner to suggest that there are two distinct methods in Buddhist soteriology. The first method states that craving is the cause of suffering and must be overcome by the eightfold path. The second states that ignorance is the cause of suffering and must be overcome by knowledge of dependent-origination (paṭicca-samuppāda). Put in another way, is the cause of suffering, taṇhā, to be overcome by calm, or is the cause of suffering, avijjā, to be overcome by insight? Does the Buddhist path aim at overcoming desire or a lack of knowledge? The tradition itself suggests that craving and ignorance are the causes of suffering and that they are overcome by calm and insight.
One approach to this question is to consider the nature of truth in Buddhism. I would like to examine whether Buddhism aims for a true knowledge of the way the world is, or whether it considers that what is true is of value. In overcoming ignorance and craving does Buddhism propose a liberating knowledge, or is what is true equated with its value? Do the early teachings state ‘what is’, or how one ought to act? Is its understanding of truth comparable to a correspondence or pragmatic theory of truth? Is it concerned with how things are or how one ought to act? For example, is dependent-origination a description of reality or a prescription for detachment?
Truth, fact and value
Donald Lopez has suggested that the study of religion is biased because it claims that to understand religions we should primarily understand religious beliefs. He argues that the category of ‘belief’ has been superimposed upon religion by scholarship, and suggests that this discourse, stemming from a preoccupation with belief in Christianity, has affected our portrayal of religion:
‘The accumulated weight of this discourse has resulted in the generally unquestioned assumption that adherents of a given religion, any religion, understand that adherence in terms of belief. Indeed, belief (rather than rituals, for example) seems to have been the pivot around which Christians have told their own history. And with the dominance of Christian Europe in the nineteenth century, Christians have described what came to be known as ‘world religions’ from the perspective of belief. Scholars of religion and anthropologists have almost invariably defined religion in terms of belief or perhaps beliefs and practices, those deeds motivated by belief. And through complicated patterns of influence, the representatives of non-Christian religions have come to speak of themselves in terms of belief.’
As Lopez goes on to suggest, this leads to the assumption that ‘religion is above all a set of truth claims, a system of belief’. Belief is given this peculiar place in the understanding of religious traditions. People are held to believe in doctrines as statements of certain propositions. This point has also been noted by W.C. Smith:
‘The peculiarity of the place given to belief in Christian history is a monumental matter, whose importance and relative uniqueness must be appreciated. So characteristic has it been that unsuspecting Westerners have […] been liable to ask about a religious group other than their own […] ‘What do they believe?’ as though this were the primary question, and certainly were a legitimate one.’
In Buddhist Studies Robert Buswell and Robert Gimello have considered this. They suggest that there is a tendency in religious studies to focus upon ‘certain cardinal concepts or model propositions to which adherents of particular traditions are believed to give their intellectual assent’. In an attempt to surmount this problem Buswell and Gimello characterise the Buddhist tradition as a path, not a creed, which has important implications for an understanding of Buddhist doctrines. They evoke the well known image of the Buddha as a physician whose doctrines heal, guide and show the way, rather than as a theorist. Buswell and Gimello define the path (mārga) as Buddhist soteriology, and as the setting in which its doctrines and patterns of practice should be understood. They argue that the Buddhist mārga ‘reticulates all the various strands of its religious endeavour — its moral values, ritual observances, theoretical doctrines, and contemplative exercises — to form a complex but unified network of practices tending to liberation’. It is in this setting that the doctrines of Buddhism are best understood.
I draw attention to these issues because in this article I am concerned with how we have knowledge of religious doctrines and with the nature of Buddhist doctrines. I do not wish to define Buddhist doctrine in a way that places upon it ideas that were absent in its original setting. The point made by Lopez, Buswell and Gimello is that religious doctrines are not necessarily, and perhaps not essentially, propositional. One implication of this may be that religious doctrines may not only propose what is, but also how one should act.
To understand the nature of religious doctrine we must understand its soteriological role. A religious doctrine works in a particular way and indeed may not be propositional in certain religions. In After Virtue, Alasdair MacIntyre has suggested that a dichotomy between ‘is’ and ‘ought’, between fact and value within religion, is a modern phenomenon. Doctrines are of a complex nature, they can state what ‘is’ and ‘is not’, ‘ought’ and ‘ought not’. Indeed, MacIntyre argues that, until modern times, the distinction between is and ought, fact and value was not made. Western thought may then make a distinction between thought and action, between fact and value that was not made in India. This point has been made by Paul Williams:
‘In the Indian context it would have been axiomatic that liberation comes from discerning how things actually are, the true nature of things. That seeing things how they are has soteriological benefits would have been expected, and is just another way of articulating the ‘is’ and ‘ought’ dimension of Indian Dharma. The ‘ought’ (pragmatic benefit) is never cut adrift from the ‘is’ (cognitive factual truth). Otherwise it would follow that the Buddha might be able to benefit beings (and thus bring them to enlightenment) even without seeing things the way they really are at all. And that is not Buddhism.’
In distinguishing between ‘is’ and ‘ought’, between ‘fact’ and ‘value’, we clearly have distinctions similar to those between correspondence and pragmatic theories of truth. Does the truth of a doctrine depend upon it stating how things are, applied to a real state of affairs, or is a doctrine true because of its pragmatic or practical value? How does Indian thought, and the Buddhist tradition in particular, classify ‘knowledge’ (ñāna), which it is the aim of ‘wisdom’ (paññā) to attain and understand, as right or wrong? Is it because it states how things are, or because that knowledge has a soteriological role?
Two similar themes are taken up by commentators of religious doctrine. William A. Christian argues that doctrines can be an assertion of how things are, or they can be practical in that they propose a course of action, they propose what one should do. Christian himself argues that religious doctrines carry a force that non-religious doctrines do not, the force of ‘recommending courses of inward and outward action’. He then states that religious doctrines probably do not have the force of ‘proposals for belief’. For him religious doctrines are non-propositional. This would be in distinction to the argument of Lopez in which religious doctrines are about belief in propositions; Jesus was resurrected, there is only one God, the Jews are the chosen people, the Buddha did end the cycle of dukkha.
This distinction between what ‘is’ and ‘ought’, between ‘fact’ and ‘value’ has been noted in the early Buddhist suttas. In the Abhayarājakumāra-sutta (M I 392-396) it is stated that the Buddha would only make statements that are ‘true’ or ‘correct’ (bhūtam/tacchaṃ) and ‘beneficial’ (atthasaṃhitaṃ), not statements that are true, correct and unbeneficial. Two terms that reflect these ideas are ‘cognitive’ and ‘affective’. The ‘truth’ (saccaṃ) that is cognitive reflects what is, or has ‘come to be’ (bhūtaṃ), hence the familiar phrase ‘seeing thing as they are’ (yathābhūtaṃ dassanaṃ). Truth corresponds with the actual state of things. The commentaries often gloss ‘right-view’ (sammā-diṭṭhi) as yāthāva-diṭṭhika, ‘seeing things as they are’. Are these terms to be understood as expressing the idea that a proposition is intended to reflect the true state of things in the world? Is it comparable to a correspondence theory of truth? Jayatilleke holds that the phrase ‘seeing thing as they are’ should be taken as meaning ‘in accordance with fact’ or ‘correspondence with fact’. According to his interpretation yathābhūta-dassana is quite literally ‘seeing things as they are’. Knowledge of things in the Nikāyas consists in knowing ‘what exists as “existing” and what does not exist as “not existing”’ (santaṃ vā atthī ti ñassati asantaṃ vā natthī ti ñassati, A V 36). This knowledge, he argues, is an objective description of reality. The object of knowledge accords with fact. Jayatilleke holds that this is the definition of truth in the Nikāyas, and that the Nikāyas implicitly accept a correspondence theory of truth. Jayatilleke’s interpretation of passages that are usually taken to show the utilitarian and pragmatic nature of the suttas (such as the similes of the raft, M I 134, and the arrow, M I 429) is that the truth value of the dhamma is not dependent upon its utility. He states that the dhamma ‘ceases to have value, though it does not cease to be true, when one has achieved one’s purpose with its help by attaining salvation’. It is to this extent that he admits that Buddhism is pragmatic in character, but he holds that Buddhist doctrines state fact. This is the first part of Jayatilleke’s argument. However, in a discussion of the development of the notion of two types of truth, found in the commentaries, the nature of this correspondence is defined. Jayatilleke gives his understanding of the distinction between ‘conventional truth’ (sammuti-sacca), and ‘transcendent truth’ (paramattha-sacca). He cites a verse found in both the Kathāvatthu and Aṅguttara commentaries (Kv-a 34, Mp I 95), in which it is stated that a conventional truth is true because of convention (sanketavacanaṃ saccaṃ lokasammutikāraṇaṃ), whereas a transcendent statement is true because it discloses the true characteristics of things (paramatthavacanaṃ saccaṃ dhammānaṃ tathalakkhaṇaṃ).  Jayatilleke does not use this distinction to claim that sammuti-sacca applies to things that are not real, while paramattha-sacca corresponds with what is and applies to things that exist. He states that sammuti-sacca does not apply to dukkha, its arising, cessation, and the way to its cessation, therefore it is conventional truth, whereas paramattha-sacca does apply to these things, therefore it is transcendent truth. To apply to things as they are, to be true, a proposition must apply to dukkha. According to him the four truths are the ‘content’ of paramattha-sacca. The dhamma corresponds to the true state of things in the sense of corresponding to dukkha, its arising, cessation and the way to its cessation.
In this analysis Jayatilleke suggests, I think, something of importance about the Buddhist analysis of truth. The nature of ‘truth’ defined as the ‘four truths’ suggests that ‘truth’ itself must be analysed according to the actual nature of things. As what is; suffering, its arising, its cessation and the way to its cessation is true, so all things are subject to this. Including Truth. Truth, if held with attachment, becomes a potential hindrance on the Buddhist path. This suggests why, in certain contexts, the early Buddhist texts are somewhat ambivalent about truth. There is a statement that we shall meet throughout this thesis that describes attachment to truth with the phrase ‘only this is true, anything else is wrong’ (idam eva saccaṃ, moggam aññan ti). This defines truth as an ‘adherence’ (abhinivesa), a type of conviction and attachment. The texts state that ‘adherence to truth’ (saccābhiniveso) is an obstacle to the goal of detachment.  The early texts were not so much interested in positing a truth and negating a falsehood as in the realisation of a state of calm in which things are seen as they are. The ‘is’ and the ‘ought’ are not separate issues. In order to see things as they are we must act in a certain way. In order to act in a certain way, we must see things as they are. As stated, lack of knowledge and craving are the causes of dukkha. It follows that knowledge and the eradication of craving cause the cessation of dukkha. Buddhism, throughout its history, stressed the inter-relationship of calm and insight.
I have related the inter-relationship of ignorance to craving and calm to insight, and suggested how this may be considered in the context of a correspondence theory of truth. These issues are complex but at this point I wish to make the following clear: the early texts suggest the inter-relationship of calm and insight and this distinction can be related to the more general notions in Western thought between fact and value, between correspondence and pragmatism. At this point I wish to define certain aspects of the nature of ‘fact’ in early Buddhism. When the Buddhists texts speak of ‘seeing things as they are’, what do they mean by this?
Jayatilleke is suggesting that if we are to understand the Buddhism of the Nikāyas as stating what is, then we must understand ‘what is’ for the Nikāyas. This is dukkha, its arising, its cessation and the way to its cessation. This is not simply knowledge that there is dukkha. The four truths may be understood in this way, as stating that there is dukkha. If one knows that there is dukkha, its arising, cessation, and a way to its cessation, I may believe in this proposition, and defend this truth from other truth claims. I will believe that things are a certain way, that there is suffering. The point that I think Jayatilleke is making is that the early texts are stating what dukkha is. By stating what it is, they state how it comes to be, how it ceases and the path to its cessation. This is stating how things are. We do not just know that dukkha is, but what dukkha is. This has soteriological implications. It is not only a statement that something is a certain way but implies a course of action which should be undertaken in order to overcome dukkha. To argue that dukkha is this way clearly does not deny the metaphysical and ontological force of Buddhist discourse. Relating suffering to pre-enlightenment experience, Sue Hamilton suggests that:
‘dukkha indicates that the nature of the Truth which the Buddhist seeks to understand is epistemic: no other aspect of the Truth is in fact accessible to human beings. So dukkha in the Four Noble Truths is to be understood both in terms of the way desires fuel the continuity of the psychological experience of unsatisfactoriness, and also […] the way our cognitive process fuels and reveals to us a world that is ontologically not other than our experience.’
Others examining this issue have argued that the Nikāyas state how things are in a different way to this, and suggest that the implications for our understanding of Buddhist discourse are far-ranging. Roger R. Jackson suggests that all religions make hard truth claims:
‘When Christians assert that ‘Jesus was resurrected from the dead’ or ‘God exists’ they generally mean that these are so in the most literal sense […] The correspondence theory is at the heart of what religious people traditionally have meant by ‘truth,’ and — arguably — it must remain at the heart of any comprehensive world-view if that view is to be considered religious.’
Jackson then analyses Buddhism. He suggests that the assertion of the four truths entails various philosophical and cosmological assumptions, which he summarises in the following way:
‘If the four noble truths and their presuppositions are to be taken literally, then there ‘really’ are past and future lives at various points on the wheel of rebirth, which means that the mind ‘really’ must be independent of particular bodies; there ‘really’ exists a universal moral law in accordance with which every mental, vocal or physical action will engender an appropriate result; the mind ‘really’ is capable of attaining a state in which all suffering has been eliminated forever; and there ‘really’ do exist extraordinary practices, such as meditation, that can assure the attainment of such a state. In short, saṃsāra and nirvāṇa ‘really’ are two possible destinies for sentient life.’
Jackson believes that the four noble truths are ‘literally true statements corresponding to actual state-of-affairs’and that ‘statements regarding saṃsāra and nirvāṇa are true in the most literal sense’. Though both Jayatilleke and Jackson argue that the truths of Buddhism are descriptions of reality that correspond to a true state of affairs, I would argue that the facts that Buddhism asserts are inseparable from the value they have. In making this statement I am keeping in mind the premise held by the early texts that both craving and ignorance cause dukkha. It follows from this that the liberating insight which overcomes dukkha cannot simply be a knowledge that something is a certain way.
However, for the moment, I would like to keep these arguments at a relatively simple level. The early texts do make statements about the nature of the world and our experience of it. These statements can be interpreted as explaining the nature of the world and experience as it really is. The texts may posit cognitive truths that correspond to the actual state of things.
I would like to discuss the value which certain doctrines have by considering the arguments of those who think that Buddhist doctrines are pragmatic and primarily state what is of value. The Pāli term that seems closest to this understanding is ‘beneficial’ (atthasaṃhitaṃ) i.e., of value. This is the term familiar to us from discussions of the avyākata which are often described as ‘not beneficial to the living of the holy life’ (na atthasaṃhitaṃ n’ ādibrahmacarika, M I 395, 431), whereas the dhamma, the understanding of the four truths, is described in the opposite terms, as ‘beneficial to the living of the holy life’ (atthasaṃhitaṃ ādibrahmacarika). Richard Gombrich has argued that the dhamma is essentially pragmatic in character. He argues that the whole Buddhist enterprise was never concerned with ontology but with epistemology; this, according to him, was the Buddha’s focus:
‘Since he was interested in how rather than what, he was not so much saying that people are made of such and such components, and the soul is not among them, as that people function in such and such ways, and to explain their functioning there is no need to posit a soul. The approach is pragmatic, not purely theoretical.’
Although Gombrich explains that there is this pragmatic tendency he also holds that the Buddha’s ‘pragmatism will work because it is based on correct assumptions’. Does the dhamma, in the sense of the Buddha’s teaching, cease to be true once its purpose has been fulfilled? The simile of the raft states that one should ‘abandon good states, how much more so bad states’. Paul Williams, in discussing this simile, argues that it does not state that the dhamma is of practical value but that there is a ‘potential incompatibility between the truth of the teachings themselves and the way they are held if they are clung onto with craving and attachment’. For both Gombrich and Williams the dhamma is both true, cognitively, as fact, and is of pragmatic value.
On the other hand José Cabezón holds that Buddhism can be defined in wholly pragmatic terms, and that the cognitive, descriptive, value of the dhamma is not of primary importance.  A simple definition of pragmatism is that a proposition is true if it is useful. Cabezón’s basic idea is that Buddhism can be classified as pragmatic, but only if this pragmatism is decontextualized from Western notions of pragmatism. It is a Buddhist pragmatism, contextualized within the Buddhist tradition. The tradition is not concerned with matters of truth, but with soteriology, with freedom from dukkha. Firstly, he considers such notions as ‘skilful means’ (upāya) which may lead us to think that the validity of Buddhist doctrines is relative to the context of the individual’s ‘mental predispositions’. If the doctrine works, it is a true doctrine. This may lead to the conclusion that the truth of a doctrine is not universally valid. Cabezón argues that such notions as upāya apply to soteriology, and not to matters of truth. He observes that a doctrine can be true at one stage of the path, though this is a provisional utility that gives the doctrine validity, and the doctrine will have to be surmounted at a later stage of the path. This aspect of the validity of doctrines also gives, I think, an example of a coherence theory of truth. One aspect of a coherence theory is that the doctrine has epistemic validity for those holding the doctrine. If the epistemic validity is in the context of the different stages of the Buddhist path, then a doctrine can have coherence in the context of that stage of the path. It is coherent with other practices in achieving the goal of the path. It is coherent within the framework of the tradition. Cabezón suggests that those who claim that the Buddhist tradition espouses a correspondence theory of truth are ‘conflating literalness and correspondence’. Cabezón seems to argue that Buddhist doctrine should be taken as true, that it should be lived, and that one should see according to the doctrine in a literal sense. The question of its ultimate correspondence with fact should then be, in the context of soteriology, irrelevant. An ideal textual source for those holding that Buddhist thought is pragmatic in character would state something like the following: ‘Even if there is a self, one should live as if there were no-self, in order to contrive a means towards salvation’. I know of no such statement. However, for scholars such as Cabezón, pragmatism is the correct way to understand Buddhist thought:
‘A pragmatist […] might argue that Buddhist metaphysics and cosmology should be taken literally — for example, a Buddhist should live as though there existed past and future lives — while bracketing the question of whether or not rebirth corresponds to some external state of affairs in the world. Treating Buddhist doctrines as useful devices whose sole purpose is the mental purification of individuals, in the process refraining from judgements as to whether they correspond to objective states of affairs, in no way entails a non-literal view of doctrine, for […] Buddhists have nowhere actually claimed such correspondence. Put another way, correspondence, pragmatism and coherence are options regarding the truth of doctrines, while literalness and non-literalness are options regarding their meaning.’
As he goes on to suggest, classical Buddhist epistemology, in the guise of Dharmakīrti, treats the notion of truth in epistemological terms, and is not overtly concerned with ontological terms. For Dharmakīrti what makes a proposition true is that it is ‘validly cognizable’. Such knowledge does not need to correspond to what is but it has the capacity ‘to fulfil a human purpose’ (puruṣārthasiddhi). Certain propositions, doctrinal ones, are best understood as pragmatic truths: a proposition is true if it is useful. Cabezón refines this in the Buddhist context by stating that the doctrinal proposition is true if it is useful to believe in the proposition, and to act in accordance with it or to utilise it in meditative practice. Action for the Buddhist tradition is germane to the question of truth. Doctrine, for the Buddhist, admits of degrees of truth, degrees of validity. The truth of a doctrine at a lower stage of the path does not rest upon it reflecting how things are, in the sense of ontology, but its utility in reaching a certain goal.
This fact/value distinction is the second issue that I wish to explore. There appears to be no consensus among scholars as to whether Buddhism is primarily concerned with fact or value. Early Buddhism need not, of course, fall into either category and there are other factors involved in answering this question.
Truth and Action
In an important article Karl H. Potter claims that a number of terms in classical Indian thought should be understood and translated in a different way from what has become the norm. The term jñāna is usually translated as ‘knowledge’ and the term pramāṇya as ‘true’. Potter suggests that the relationship between ‘knowledge’ (jñāna) and its content (viṣaya), in deciding whether it is ‘true’ (pramāṇya) is one of purpose. He argues that what makes an act of cognition true is not that it sees things as they are, but that its purpose, its workability, is fulfilled. For knowledge to be defined as true, the purpose must be fulfilled. Cognising things in a certain way gives them their value: yathārtha should not be understood in the sense of ‘as the object is’ but ‘as the purpose is’. Thus yathārtha implies purpose not correspondence. He offers the example of Dharmottara for whom the relationship between right awareness (samyagjñāna) and its content, ‘requires that the awareness apprehends the content as an objective suitable for successful purposive activity’. An act of cognition which initiates a successful course of action is right knowledge. Truth is ‘workability’ (pramāṇya):
‘pramāṇya does not translate as ‘truth’ (i.e., correspondence with reality), despite standard translation practice, but rather connotes a more pragmatic criterion of being capable of producing or helping to produce satisfaction in action.’
Hence jñāna is not ‘knowledge’ but ‘awareness’ and pramāṇya not ‘truth’ but ‘workability’. Knowledge is awareness of workability, an awareness of value. Indian thought ‘adopts a utility reading of “truth”’ and thus what ‘knowledge consists in features this very relativity to purpose’. Truth is what serves a purpose, ‘to know’ is to know what one is doing, to be aware of the purpose. It is to know or be aware of the nature and course of one’s actions.
Related to the ideas of Potter are those of Mohanty. Mohanty has suggested that in Western thought truth is the predicate of a proposition, whereas in Indian thought truth is the predicate of a jñāna or act of cognition. He argues that something like a proposition is not admitted in Indian philosophy. The predicate ‘true’ has an opposite ‘false’, for a belief, or, presumably, a proposition. However, for an act of knowledge, we have ‘true knowledge’, but its opposite, ‘false knowledge’ implies something different to false proposition. He argues that for the Indian tradition the truth is connected with successful action:
For theory of action not only forms the corner-stone for practical philosophy, but also is the connecting and transitional link between theoretical philosophy and practical philosophy. This is especially so in Indian philosophy where the ultimate criterion of truth was practical success. All cognition leads to appropriate action, which, if the cognition is true, leads to successful action and, if it is false, to failure and frustration. He also argues that:
‘For all Indian thinkers, cognition issues in practical, actional response (pravṛtti), and the ultimate guarantee, for most of them, of the truth of cognition is practical success.’
Paul Griffiths has also commented on the translation of the term jñāna. He thinks that the use of the term ‘knowledge’ to translate jñāna, may express connotations unintended in the Indian context. He suggests that knowledge often implies ‘a propositional attitude, had by some subject toward some proposition’. He suggests that knowledge is often taken as a form of belief. Clearly, Western thought makes the distinction between belief and knowledge, but Griffiths’ point is that knowledge is still a type of belief, or related to it. The situation with jñāna is somewhat different to this. He argues that in the Indian context jñāna are seen as ‘episodic mental events with cognitive significance, rather than as (dispositional or occurent) propositional attitudes’. He adds that the opposite of jñāna is not falsehood, but doubt or uncertainty.
I have pointed out that some scholars hold that in Buddhism we find something comparable to a correspondence theory of truth. Alternatively, Buddhism may true because of its pragmatic value. I have suggested that there is no consensus among scholars as to whether Buddhist discourse is true because it states what is, or because it has value. I have related this dichotomy to the central Buddhist themes of there being two related causes of dukkha, craving and ignorance, which are to be overcome by calm and insight. In Buddhist soteriology fact and value, what is and how one should act, are reciprocal. It is in this context, in which knowledge and action, fact and value are essential in the realisation and overcoming of dukkha, that the notions of ‘wrong-view’ (micchā-diṭṭhi) and ‘right-view’ (sammā-diṭṭhi) can be considered.
 Erich Frauwallner, History of Indian Philosophy, Volume I, Translated by V. M. Bedekar (Delhi, 1973), pp. 127-171. First published in German (Salzburg, 1953).
 Donald S. Lopez, Jr. ‘Belief’ in Critical Terms for Religious Studies, ed. by Mark C. Taylor (Chicago, 1998), pp. 21-3 (p. 21). Lopez’s emphasis.
 Ibid., p. 28.
 W. C. Smith, The Meaning and End of Religion: A New Approach to the Religious Traditions of Mankind (New York, 1963), p. 180.
 Paths to Liberation: The Mārga and its Transformations in Buddhist Thought, ed. by Robert E. Buswell, Jr. and Robert M. Gimello (Honolulu, 1992), p. 4.
 Paths to Liberation, p. 3.
 Ibid., p. 4.
 Ibid., p. 3.
 Ibid., p. 6.
 Alasdair MacIntyre, After Virtue (South Bend, Ind, 1981), pp. 54-57, 80-81; see W.D. Hudson, ed., The Is — Ought Question: A Collection of Papers on the Central Problem in Moral Philosophy (London, 1969).
 Paul William, Buddhist Thought: A Complete Introduction to the Indian Tradition (London, 2000), p. 40.
 See A.R. White, Truth (London, 1971), pp. 102-127; D.J. O’ Conner, The Correspondence Theory of Truth (London, 1975); L.E. Johnson, Focussing on Truth (London, 1992).
 William A. Christian, Oppositions of Religious Doctrines (London, 1972), pp. 87-109.
 Ibid. pp. 43-59.
 Ibid. See especially pp. 24, 31, 85. He adds a third category: that some religious doctrines propose ‘valuations’, they give value to other doctrines (ibid., pp. 60-86 and p. 89).
 He cites the example of such a doctrine which R. B. Braithwaite uses: ‘God is love’ (William A. Christian, Oppositions of Religious Doctrines, p. 21); see R.B. Braithwaite, An Empiricist’s View of Religious Belief (Cambridge, 1955), pp. 32-33; see also William A. Christian, Meaning and Truth in Religion (Princeton, 1964).
 Christian, like other scholars in this field (i.e., D.Z. Philips) has perhaps, as he himself acknowledges (p. 20), been inspired by Wittgensteins’ 1938 lectures on religious belief; see Ludwig Wittgenstein, Lectures and Conversations on Aesthetics, Psychology and Religious Belief, ed. by Cyril Barrett (Berkeley and Los Angeles, 1966). In a classic study of religious doctrines George Lindbeck has suggested three categories for their understanding: they may be understood as propositions, reflecting the true nature of things; they may be ‘non-informative and non-discursive’ symbols’ (which would fit into the category of pragmatism), or they may be cultural linguistic categories about the sacred (something similar to a coherence theory of truth, i.e., a doctrine is true if it is coherent with the other doctrines of a given religion). George A. Lindbeck, The Nature of Doctrine: Religion and Theology in a Post Liberal Age (London, 1984), pp. 16-18; see also Carol Anderson, Pain and its Ending: The Four Noble Truths in the Theravāda Buddhist Canon (Richmond, 1999), p. 31.
 Many of these issues have been discussed by Jayatilleke, including the Abhayarājakumāra-sutta; see K.N. Jayatilleke, Early Buddhist Theory of Knowledge (London, 1963), pp 351-68.
 M III 289, S II 16-17; yathābhūtaṃ pajānāti, ‘one knows in accordance with how things are’, D I 83-84; see Jayatilleke, Early Buddhist Theory of Knowledge, p. 352.
 Mp, I 27, 355, V 66; see Peter Masefield, Divine Revelation in Pāli Buddhism (Colombo/London, 1986), p. 44.
 Jayatilleke, Early Buddhist Theory of Knowledge, p. 352.
 Ibid., p. 359.
 Ibid., p. 428.
 Ibid. Jayatilleke’s use of the Apaṇṇaka-sutta to make this point is discussed in chapter three.
 Ibid., p. 358.
 dhammānaṃ bhūtalakkhaṇaṃ, Mp I 95. Jayatilleke, Early Buddhist Theory of Knowledge, pp. 363-4.
 Ibid., pp.367-68.
 Ibid., p. 368; see also Ñāṇananda, Concept and Reality (Ceylon, 1971), pp. 44-46.
 Or idaṃsaccābhinivesa, one of the four ‘bodily knots’ (kāyagantho), at S V 59. I will list these in chapter six.
 See, for example, Carol Anderson, Pain and Its Ending (Richmond, 1999).
 I am using the word proposition in a way comparable to Wittgenstein who argued that ‘it is impossible for there to be propositions of ethics. Propositions can express nothing that is higher’. Ludwig Wittgenstein, Tractatus logico-philosophicus, trans. by D.F. Pears and B.F. McGuiness (London, 1961), 6.42.
 See Vibh-a 116-17.
 Sue Hamilton, Early Buddhism: A New Approach. The I of the Beholder (Richmond, 2000), p. 111. Hamilton’s emphasis.
 Roger R. Jackson, Is Enlightenment Possible? Dharmakīrti and rGyal tshab rje on Knowledge, Rebirth, No-Self and Liberation (New York, 1993), p. 42.
 Ibid., pp. 22-23.
 Ibid., p. 43.
 Ibid., p. 51.
 Ibid., p. 52.
 There are a number of scholars who hold that Buddhist truth is pragmatic in character such as Caroline Rhys Davids and Poussin; see Jayatilleke, Early Buddhist Theory of Knowledge, pp. 356-57 for references.
 The so-called ‘unanswered questions’. These are a group of ten questions (usually), which the Buddha refused to answer. The best analysis of them is perhaps found in Steve Collins, Selfless Persons: Imagery and Thought in Theravāda Buddhism (Cambridge, 1982), pp. 131-138. I shall consider them twice in this thesis: in chapters two and six.
 For example the Buddha states that he has declared dukkha, its arising, cessation and the way to its cessation, because it is beneficial and belongs to the fundamentals of the holy life, etc. (kiñca māluṅkyaputta mayā byākataṃ: idaṃ dukkhan ti māluṅkyaputta mayā byākataṃ, ayaṃ dukkhasamudayo ti mayā byākataṃ, ayaṃ dukkhanirodho ti mayā byākataṃ, ayaṃ dukkhanirodhagāminīpaṭipadā ti mayā byākataṃ. kasmā cetaṃ māluṅkyaputta mayā byākataṃ: etañhi māluṅkyaputta atthasaṃhitaṃ, etaṃ ādibrahmacariyakaṃ etaṃ nibbidāya virāgāya nirodhāya upasamāya abhiññāya sambodhāya nibbānāya saṃvattati, M I 431); see Jayatilleke Early Buddhist Theory of Knowledge, pp. 351-68.
 Richard Gombrich, How Buddhism Began: The Conditioned Genesis of the Early Teachings (London, 1996), p. 16; see also Ñāṇananda, Concept and Reality, pp. 42-43, who offers an argument for the pragmatic value of the dhamma.
 Gombrich, How Buddhism Began, p. 16.
 dhammā pi vo pahātabbā, pageva adhammā, M I 135. I will discuss this simile again later in this chapter and in chapters six and seven.
 Paul Williams, Buddhist Thought: A Complete Introduction to the Indian Tradition (London, 2000), p. 39.
Gombrich, How Buddhism Began, p. 16; Williams, Buddhist Thought, p. 39.
 José Cabezón ‘Truth in Buddhist Theology’, in Buddhist Theology: Critical Reflections by Contemporary Buddhist Scholars, ed. by Roger Jackson and John Makransky (Richmond, 2000), pp. 136-54.
 Cabezón ‘Truth in Buddhist Theology’, p.144.
 Ibid., p. 146.
 Ibid., p. 147.
 Ibid., p. 140. This is one of the possible objections to the ‘workability thesis’ of Karl Potter discussed later in this chapter.
 This problem goes beyond the scope of the present study. Donald Lopez has characterised this as choosing between upāya and doctrine; how one is to choose between method and truth: ‘[T]he Buddha taught many things to many people, in accordance with their aspirations, capacities and needs. How is one to choose among these myriad teachings, each “true” for its listener, to determine the final view of the teacher?’, Buddhist Hermeneutics, ed. by Donald Lopez (Honolulu, 1988), p. 50.
 Cabezón, ‘Truth in Buddhist Theology’, pp. 140-41. There is a danger of understanding Buddhism as a type of relativism in which all doctrines are somehow valid as they are all, in their way, an approximation of the truth. I do not think that this is the position of the early Buddhist texts. Early Buddhist thought is arguing that all doctrines and views are wrong if they are held with attachment. It is not arguing that all doctrines and views are relatively true from a certain perspective. All doctrines and views are not, from a certain perspective, true doctrines and views, as in the Jain anekantavāda. Another aspect of this debate, which could perhaps be utilised by both the ‘engaged Buddhist’ and the Buddhist theologian (which is part of the dialogue in which Cabezón is involved), is that relativising Buddhist doctrine, if this is what pragmatism does, could be used as a form of Buddhist apologetic. Although Cabezón does not do this, he touches upon possible aspects of its use. For example, discussing such primary Buddhist doctrines as kamma, rebirth and the theory of enlightenment, Cabezón states that the use of pragmatism is a far more useful tool for the understanding of these doctrines, for those who do not take them literally, than the correspondence theory of truth. These doctrines need not apply to a real state of affairs, as Cabezón puts it: ‘I find myself in the position of being metaphysically alienated, unconvinced of the metaphysical (lege correspondence version of the) truth of a good deal of Buddhist doctrine, while still profoundly convinced of the validity of the Buddhist doctrine as a whole. For alienated Buddhist theologians like myself — skeptics who find problematic the espousal of the metaphysical truth of Buddhist doctrines like karma and rebirth — pragmatism offers a method of finding truth in the tradition, even in those portions of the tradition which would otherwise be unacceptable.’ (Cabezón, ‘Truth in Buddhist Theology’, p. 149). Cabezón contrasts this method of approach to Buddhist doctrine to that taken by others (he cites Stephen Batchelor), whose response to such issues is one of agnosticism. William A. Christian has commented upon such a situation: ‘If some religious community, in the course of developing its doctrines, should come to a point of renouncing beliefs and constituents of its way of life, the situation would be problematical in the following way. That community would have to interpret its doctrines as purely directive or purely expressive or as combining these two functions without any admixture of assertions. And, in explanation of this policy it would have to show that those doctrines which recommend courses of action have no existential suppositions, and that those doctrines which express valuations have no valuations. This might not be easy.’ (Christian, Opposition of Religious Doctrines, pp. 91-92). Of course a simple answer may be, as Christian observes, that a doctrine such as rebirth may affect how one acts in the present life (Christian, ibid., pp. 107-8). It is a doctrine that proposes a valuation of life.
 Cabezón would disagree with this: ‘There are […] problems in maintaining that the Indo-Tibetan tradition subscribes to a coherence theory of truth. According to the coherence theory, a proposition (or doctrine) is true if it coheres with a designated set of beliefs. Given the incommensurate nature of the belief systems of the Buddhist tradition […] coherence is simply not sufficient to guarantee the truth of a doctrine. Buddhist doctrine […] is extremely diverse, and at times even contradictory. This being so, mere internal consistency with some portion of the corpus of Buddhist doctrine will not guarantee a doctrine’s truth, since it is quite possible that a doctrine that is supported in one portion of the canon will be repudiated in another. Therefore, independent criteria, over and above coherence, are considered necessary in determining which doctrines are unconditionally true.’ (Cabezón, ‘Truth in Buddhist Theology’, pp. 143-44). However, I would argue that the very diversity of a tradition, and the fact that it may not find different doctrines contradictory, may tell us much about a tradition and how it understands its own doctrines. For example, the doctrines of not-self and rebirth are, in a sense, contradictory, and this is a classic example of a dilemma which the tradition has addressed. The tradition maintains both doctrines. There must be reasons for this. We may learn much about these doctrines, and the tradition which expresses them if we attempt to understand how these doctrines are coherent with other doctrines.
 Cabezón ‘Truth in Buddhist Theology’, p. 141.
 Ibid., p. 141.
 Ibid., p. 142; though note Cabezón’s own qualifications and reservations on Dharmakīrti being understood as a pragmatist, ibid., p. 146.
 Ibid., p. 142.
 Ibid., p. 144-45.
 Ibid., p. 145.
 Ibid., p. 148.
 K. H. Potter, ‘Does Indian Epistemology Concern Justified True Belief’, Journal of Indian Philosophy, 12 (1984), 307-27.
 Ibid., pp. 311-2.
 See Chakrabarti’s comments that the latter sense was also prominent, K.K. Chakrabarti, ‘Some Remarks on Indian Theories of Truth’, Journal of Indian Philosophy 12 (1984), 339-355 (p. 340).
 Potter, ‘Does Indian Epistemology Concern Justified True Belief’, p. 314.
 See Gethin’s discussion, The Buddhist Path to Awakening: A Study of the Thirty-Seven Bodhipakkhiyā Dhammā in the Nikāyas and Abhidhamma (Leiden, 1992), pp. 108-9.
 Potter, ‘Does Indian Epistemology Concern Justified True Belief’, p. 316.
 Ibid., pp. 317-18.
 Ibid., p. 322.
 Ibid., pp. 323-324. Paul Williams would disagree with this argument. He has argued that: ‘The teachings of the Buddha are held by the Buddhist tradition to work because they are factually true (not because they work)’, Williams, Buddhist Thought, p. 40.
 J. N. Mohanty, Reason and Tradition in Indian Thought (Oxford, 1992), pp. 134-5.
 Ibid., p. 136.
 J. N. Mohanty, ‘Prāmaṇya and Workability’, Journal of Indian Philosophy 12 (1984), 329-338 (p. 332).
 Mohanty, Reason and Tradition in Indian Thought, p. 264. Hence, a non-confused, non-deluded seeing of reality will, by definition, lead to a certain wholesome course of action.
 Ibid., p. 232.
 Paul J. Griffiths, On Being Buddha: The Classical Doctrine of Buddhahood (Albany, NY, 1994), pp. 153-54.
 Ibid., p. 152.
 Ibid., p. 153.