Revista de Estudos da Religião - REVER
ISSN 1677-1222

Pós-Graduação em Ciências da Religião - PUC-São Paulo

Philosophical Analysis and the Notion of Knowledge in the Bhagavad-Gītā^[*]

Ricardo Sousa Silvestre^[**] []

1. Introduction

Those familiar with the history of ideas are aware that philosophers of all times have, in one way or another, engaged in conceptual analysis. It is therefore not surprising that a new way of philosophizing which has the practice of conceptual analysis as one of its basic foundations had arisen. What is surprising perhaps is that this new philosophy – we are of course talking about analytic philosophy – has quickly expanded its influence beyond the English-speaking world, where it was born in early twentieth century, as well as beyond the disciplines with which it is traditionally related, such as logic and philosophy of language. We find nowadays the method of philosophical analysis being applied to disciplines like ethics, political philosophy, epistemology, aesthetics, metaphysics, philosophy of mind, and also philosophy of religion.

The contributions of analytic philosophy, and in fact an understanding of what philosophical analysis effectively is, can perhaps be measured only when we look close at particular instances of philosophical investigation. A good example is the field of epistemology. Since mid-twentieth century what we call epistemology or the theory of knowledge is really two different enterprises: one which aims at analyzing the concept of knowledge and a much more traditional one centered at finding out how we get to know things. While the second approach, which is basically the philosophizing inaugurated by 17^th and 18^th century thinkers such as Descartes, Locke and Berkeley, focuses on the origins, limits and scope of human knowledge^[1], the first aims at something apparently much less ambitious: explaining or analyzing the concept of knowledge, or in a nutshell, saying what the word knowledge means. Curiously enough, even though understanding the meaning of the word knowledge is something which in principle must precede speculations about the origin and scope of knowledge, it was not until the 1960s that philosophers recognized the importance of the analysis of the concept of knowledge for epistemology. Since then, the analysis of knowledge has taken on a life of its own, mainly on articles in professional journals, so to become one of the most preeminent areas of contemporary epistemology.

This emphasis on language and on what words mean, which is certainly one of the distinguishing features of analytic philosophy, has also influenced the philosophical reflection on religion. It is well known, for instance, the controversy that took place in the 1950s and 1960s over whether theistic language was cognitively meaningful or not (Soskice, 1997), or the use made by Alvin Platinga in the 1970s of the so-called free will defense to the (logical) problem of evil (Platinga, 1974), which was accompanied by a very sophisticated analysis of the notion of necessity, or still the analysis made by Robert Adams on the relations between faith and morality (Adams, 1987). More relevant for us, however, is the contribution that conceptual analysis can give to philosophy of religion by clarifying some key notions of theistic systems. Examples of such endeavor applied to eastern religions are (Potter, 1984), (Fouts, 2004), (Stoltz, 2007) and (Ramamurty, 2008).

The purpose of this paper is to apply this so-called method of philosophical analysis to one of the most influential pieces of work in the history of religion: the Indian sacred text Bhagavad-gītā. Since epistemology has been perhaps one of the aspects of Indian philosophy most susceptible of being approached by analytic methods (Potter, 1984) (Stoltz, 2007), we are going to focus on the epistemology of Bhagavad-gītā. More specifically, we shall use the methodological and conceptual framework of contemporary analytical epistemology to analyze the notion of knowledge in the Bhagavad-gītā. By doing that, we shall inevitably try to fit the Bhagavad-gītā notion of knowledge into some key epistemological categories developed inside contemporary analytic philosophy. Therefore, our work has to do with what is nowadays called comparative philosophy. However, because the Bhagavad-gītā is primarily of course a religious text, but also because, as we shall see in the course of the paper, the concept of knowledge in the Bhagavad-gītā is intrinsically related to its theology, our work has also to do with the philosophy of religion.

The structure of the paper is as follows. In the next section we survey the relevant aspects of contemporary analytic epistemology we shall use in our conceptual analysis. We apologize for the brevity, or, someone might say, superficiality, with which we deal with some important aspects of contemporary epistemology; but unless we did that we would not be able to lay down in a paper length text our theses concerning the relations between analytic epistemology and the notion of knowledge in the Bhagavad-gītā. In Section 3 we make some methodological remarks about knowledge-related Sanskrit terms, conceptual analysis and Bhagavad-gītā translations. Sections 4 to 7, which are the core of the paper, form an interdependent sequence of argumentative attempts to clarify, according to the framework set in Section 2, the notion of knowledge in the Bhagavad-gītā. They go from a relatively simple behaviorist account of knowledge to a more sophisticated one, which includes a dual conceptual account of knowledge (Section 5), a plural foundationalist-justification based one (Section 6) and an account which includes what we might call a theist conception of knowledge (Section 7). Finally in Section 8 we make some concluding remarks.

2. Contemporary Analytic Epistemology

The first important thing to be said about analytic epistemology is the distinction acknowledged by practically all contemporary philosophers between propositional knowledge, skill knowledge and objectual knowledge. While propositional knowledge is knowledge that something is the case, as in Peter knows that Carla hates him, skill knowledge is knowledge how to do something, as in Peter knows how to ride a bicycle and objectual knowledge is one’s knowledge of persons, places and things, as in Peter knows Carla. If we talk about the content of knowledge, or what is known, we shall have that each one of these three kinds of knowledge has different kinds of contents. While the content of propositional knowledge is a proposition (in our example the proposition that Carla hates Peter), the content of skill knowledge is a skill (in the example the ability of riding a bicycle) and the content of objectual knowledge is an object (Carla in our example)^[2].

Some philosophers have related this conceptual distinction between propositional and objectual knowledge with the distinction laid down by Bertand Russell between knowledge by description and knowledge by acquaintance (Russell, 1910). Suppose you have never been in Paris before, but nevertheless have read a lot about it. Therefore, even though you are not acquainted with Paris, you can claim to have knowledge about it, namely propositional or descriptive knowledge of Paris. Supposing now that you finally get to go to Paris, you will be able not only to confirm (or disconfirm) all those descriptions you have read about it, but also to experience new aspects of Paris which might even not be susceptible of being fully expressed through words: unless you go to Paris, you might say, you will never be able to know exactly what it is. You therefore have got knowledge by acquaintance of the city of Paris, which is obviously an object, in opposition to mere propositional knowledge.

Despite acknowledging this diversity of knowledge kinds, epistemologists’ main concern, in the analytic as well as in the traditional approach, has been with propositional knowledge. Even though in Section 7 we shall make an important use of the idea of objectual knowledge, for most of the paper our main concern shall also be with propositional knowledge. Therefore in the rest of this section we shall focus exclusively on knowledge which has as content a proposition (or, as we shall see, a set of propositions.)

Besides the content of knowledge, there is also obviously the one who knows, or in other words the knowing subject: for every knowledge content P, there is possibly one knowing subject S such that S knows P. This apparently minor detail is in fact important because depending on which side of the knowledge claim S knows P we focus, we shall have two different ways of studying the notion of knowledge. If we decide to focus on the content of knowledge, we will answer the question of what knowledge is based on the features of the thing which is known. The field of philosophy of science is perhaps the best example of this approach. We can say that contemporary philosophy of science tries to say what scientific knowledge is (which is a specific kind of propositional knowledge) by laying down the structural and logical features of scientific theories, the way theories are confirmed and disconfirmed, the formal properties of sentences expressing natural laws, the nature of theoretical terms, etc. In this approach to the analysis of knowledge, which we shall call the content approach, knowledge is at least from a methodological point of view identified with the content of knowledge. What is scientific knowledge? It is the final result of scientific practice, namely the theories and the scientific claims derived from them, which are exactly the content of what scientists know^[3].

The second way to study the concept of knowledge, which we shall call the knowing subject approach, is to focus on the left-side of the claim S knows P and bring into scene the knowing subject how knows P^[4]. If we adopt this approach, the question What is knowledge? shall be replaced by the question What does it mean to say that S knows P?, for now the analysis of the notion of knowledge shall be necessarily made by making reference to the knowing subject. The most natural way to answer this is to set the necessary and sufficient conditions for S to know P, that is to say, a set of conditions which together imply that S knows P, and such that S knows P implies each one of them.

It turns out that the traditional answer to this question is to say that S knows P means that (i) S believes P, (ii) S is justified in believing P, and (iii) P is true; or in other words that (i), (ii) and (iii) are necessary and sufficient for S to know P: if S knows P then (i) S believes P, (ii) S is justified in believing P, and (iii) P is true, and if (i) S believes P, (ii) S is justified in believing P and (iii) P is true, then S knows P. For example, we can say that John knows that Carla hates him when (i) John believes that Carla hates him, (ii) he is justified or has good reasons for believing so, and (iii) Carla in fact hates him. If at least one of these conditions is not satisfied, then we cannot say that John knows that Carla hates him.

This is of course the famous justified true belief (JTB) account of knowledge traditionally attributed to Plato, the tenability of which has been seriously attacked by the American philosopher Edmund Gettier in the 1960’s (Gettier, 1962)^[5]. Without getting into the details of Gettier’s counterexamples and the various refinements of the JTB account which it gave rise (Sorell, 1988), we can say that if we want to follow this approach in the explanation of the notion of knowledge we will have to somehow account for the notions of belief, justification and truth. Due to space reasons, but also to the purpose of this survey section, we shall restrict ourselves exclusively to the notions of belief and justification.

The notion of belief has been much more studied in the analytic tradition by philosophers of mind than by epistemologists. This has been done mainly through a behaviorist-like approach called functionalism where the meaning of belief is given through the behavior of the believing subject along with the functional relations that supposedly exist between belief and other mental concepts. For example, one can say that John believes that it is going to rain if, in a situation where (i) he is about to leave home for a walk, (ii) he does not want to get wet and (ii) there is an umbrella available for him, (iv) he takes the umbrella with him^[6]. In this way, belief is seen as a disposition to act in certain circumstances. This behaviorist account sharply contrasts with the purely mentalist account given, for example, by the seventeenth century French philosopher René Descartes, to whom believing was simply an internal mental state of the soul (Descartes, 1985).

Regarding justification, the traditional account is based on what is commonly called foundationalism. According to foundationalism,

[…] our justified beliefs are structured like a building: they are divided into a foundation and a superstructure, the latter resting upon the former. Beliefs belonging to the foundation are basic. Beliefs belonging to the superstructure are nonbasic and receive justification from the justified beliefs in the foundation. (Steup, 2005: 5)

Say you ask me to justify my belief that (i) in a few minutes I shall have a strong headache. To that I reply that (ii) I feel a slight throbbing pain in the right side of my head and (iii) every time that I fell that a few minutes later a strong headache comes. So I justify the belief (i) by giving two other beliefs: (ii) and (iii). But why (ii) and (iii), which we take here as basic, serve as justification for the nonbasic belief (i)? Because there is a logical relation of some sort between these beliefs, so that the truth of (ii) and (iii) entails the truth of (i). This is how basic beliefs justify nonbasic beliefs: through logical inference.

But by virtue of what are these basic beliefs justified? Here we could say that they are justified by certain kinds of perceptual experiences, in the case of (ii) the experience of feeling the pain, and in the case of (iii) the experience of remembering the situations where I felt that pain. (See that we are not talking about the belief that I feel the pain or the belief about my past experiences, but about the experience of pain and the experience of remembrance themselves.) And how do we justify these experiences? We do not. How would you respond to someone who asks how do you know that you are felling a slight throbbing pain in your head? Probably you would say that you are felling it, and you know you are felling it, and that is it! In fact, when we make reference to perceptual experiences we are already outside our foundational building, which, it should be reminded, is made exclusively of beliefs. That is why the beliefs which perceptual experiences justify are called basic: their justification is not supported by other beliefs. As one might have noticed, this justification of basic beliefs makes direct reference to the source of belief (and consequently of knowledge), representing then the connection between the analysis of knowledge and the classical approach to epistemology. Besides the ones mentioned above, other sorts of knowledge sources and (consequently of knowledge justification) studied in the analytic tradition are testimony and introspection.

In order to finish this section, two last points should be mentioned about this JTB account of knowledge. First, there is a clear asymmetry between the three components of our definition of knowledge: while belief and justification can be called, for the purposes of this essay, internal components or conditions of knowledge, for they might be characterized exclusively by making reference to the knowing subject, truth is taken as the external component of knowledge, for the truth of a proposition, at least according to a realist standpoint, does not depend on the knower. Second, concerning propositional knowledge, this JTB account can be said to be a general one: it is not its purpose to account only for a specific type of propositional knowledge (such as scientific knowledge, religious knowledge or commonsensical knowledge), but to account for propositional knowledge in general, of any kind.

3. A Note on Translation

The most common Sanskrit term associated with the notion of knowledge is the word jñāna. There is however disagreement among scholars about the exact meaning of this word. While Alex Wayman (Wayman, 1955) for example takes jñāna as meaning the same as the English word knowledge, Karl Potter (Potter, 1984) translates it as an act of cognition or awareness, defending that the closest Sanskrit term to knowledge is actually pramā, which is an act of cognition with prāmāya or truth as essential property. That is the same view of Purusottama Bilimoria (Bilimoria, 1985), who develops a theory of knowledge which accounts for the rise of pramā from jñāna, as well as of Jitendranath Mohanty (Mohanty, 1994), who nonetheless in an earlier writing takes both jñāna and pramā as meaning knowledge (Mohanty, 1966). Even though the works of those authors are done inside delimited philosophical traditions of Indian thought, their meaning claims about jñāna are intended to have a universal scope. Besides jñāna, the terms prajñā, vijñāna and vidyā are also found in modern Bhagavad-gītā translations as Sanskrit equivalents to knowledge, which by the way also conflicts with the view of some scholars. Wayman (Wayman, 1955) for example sharply distinguishes these terms from the English word knowledge, translating prajñā as insight and vidyā as wisdom.

In this paper we shall not get into the debate over the proper linguistic meaning of knowledge-related Sanskrit terms. Even though this decision has to do with space reasons, there is a more fundamental reason which is of great importance to the approach we shall follow here. Even though syntactic and semantic considerations of a specific natural language do play an important role in explanatory endeavors such as the one we intend to do here, conceptual analysis is not exactly the same as linguistic analysis. It is at least a work hypothesis in analytic philosophy that what we call concepts lay beyond linguistic terms and words, so that the same concept may appear in different languages or even in the same language in non-syntactically related ways; or the same word might express two different concepts. So we shall adopt as a methodological strategy the acceptance of the standard English meaning given to the above mentioned Sanskrit terms by contemporary translators of Bhagavad-gītā, and use those English words and expressions as starting-points for our conceptual analysis.

In the quotations of and meaning references to Bhagavad-gītā’s verses, we shall use three different translations: Graham Schweig’s (Schweig, 2007), Winthrop Sargeant’s (Sargeant, 1994) and Bhaktivendhanta Swami Prabhupada’s one (Prabhupada, 1986). Sargeant’s work will be used solely in the translation of isolated terms. When we quote full verses, we shall lay down the translations of Prabhupada and Schweig. When we describe part of the content of a verse in a non-quotation form and without reference, we shall be using (more or less freely) the three translations. References to the verses are done as usual: 2.1 for instance means verse 1 of chapter 2.

4. A Behaviorist Account of Knowledge

In this section we actually begin our attempt to analyze, explain or, being more cautions, clarify the concept of knowledge contained in the Bhagavad-gītā. We proceed by what is perhaps the more natural way to look for a definition of knowledge in a poem: examining the verses which, from a formal point of view, say what knowledge is. We start by verses 8 to 12 of chapter 13^[7]:

Humility; pridelessness; nonviolence; tolerance; simplicity; approaching a bona fide spiritual master; cleanliness; steadiness; self-control; renunciation of the objects of sense gratification; absence of false ego; the perception of the evil of birth, death, old age and disease; detachment; freedom from entanglement with children, wife, home and the rest; even-mindedness amid pleasant and unpleasant events; constant and unalloyed devotion to Me; aspiring to live in a solitary place; detachment from the general mass of people; accepting the importance of self-realization; and philosophical search for the Absolute Truth—all these I declare to be knowledge, and besides this whatever there may be is ignorance. (13.8-12) (Prabhupada, 1986: 648)

Absence of pride, absence of deceit, nonviolence, patience, honesty; service to the guru, purity, stability, control of the self; dispassion for the objects of the senses, and also, absence of the notion of ‘I am acting’; foreseeing the perils of birth, death, old age, disease, and suffering; absence of attachment and excessive affection for children, spouse, home, and so on; also, constant same-mindedness in desirable and undesirable circumstances; and absorbed in me with no yoga other than the unwavering offering of love; dwelling in a secluded place having disregard for crowds of people; continuity in knowledge of the ‘principle of the self’, with a vision of the object of that knowledge of the truth – This is declared to be knowledge; the absence of knowledge is whatever is contrary to this. (13.7-11) (Schweig, 2007: 177-178)

Here it is said: etaj jñānam iti proktam, literally meaning this knowledge thus, declared to be (Sargeant, 1994: 539), or, as translated above, (all) this is declared to be knowledge. It is clear enough that here knowledge is being identified with things like absence of pride, nonviolence, patience, absence of excessive affection for children, spouse and home, service to the guru (spiritual teacher), control of the self, etc. But what could this mean? What could it mean, for example, to say that knowledge is service to the guru?

Well, it must first be noted that all these terms to which knowledge is identified are strongly related to specific behavioral features which people may or may not possess. Absence of pride, nonviolence, patience, absence of excessive affection for children, spouse and home, service to the guru, etc. are all properties which, if really possessed by someone, impel, so to speak, the person in question to behave in a specific and more or less predictable way. Second, even though what is being defined here is jñānam or knowledge, since all terms used in the definition are properties of human beings, we can always say that what these verses are really concerned about is the person who possess knowledge (or is in knowledge), in such a way that the definition of knowledge they give is an indirect one done through the behavior of the one who is in knowledge. Therefore, even though one can take these verses as simply describing the behaviorist result of having knowledge, or as describing the conditions one must satisfy to successfully engage in the process of attaining knowledge, what we have here in fact is a behaviorist account of knowledge (akin to the behaviorist account of belief we exemplified in Section 2) in which the question What is knowledge? is answered by pointing out how the one who is knowledge behaves.

Two other verses which might be taken as related to this behaviorist account of knowledge, perhaps complementing the list given in verses 13.8-12, and which make direct reference to the knowing subject as someone having knowledge are 2.57 and 7.19:

In the material world, one who is unaffected by whatever good or evil he may obtain, neither praising it nor despising it, is firmly fixed in perfect knowledge. (2.57) (Prabhupada, 1986: 146)

One who, everywhere, is without sentimentality upon encountering this or that, things pleasant or unpleasant, who neither rejoices nor despises – the profound knowledge of such a person is firmly established. (2.57) (Schweig, 2007: 51)

After many births and deaths, he who is actually in knowledge surrenders unto Me, knowing Me to be the cause of all causes and all that is. Such a great soul is very rare. (7.19) (Prabhupada, 1986: 393)

At the end of many births, one who has knowledge offers oneself to me, realizing, “Vāsudeva is everything!” – such an exalted self is very rarely found. (7.19) (Schweig, 2007: 112)

But which sort of behaviorist account do these verses give? Do they provide a set of sufficient and necessary conditions for knowledge or, to say better, for someone to be in knowledge? Does it seem reasonable to say that if all the conditions described in these verses are satisfied it is guaranteed that one is in knowledge (sufficient side), and if only one of them is missing then we cannot say he is in knowledge (necessary side)? While we might agree that possessing all these features is a guarantee for someone to be in knowledge, it seems too heavy to require that unless all of them are present the person in question in not in knowledge. What about if someone possesses all the mentioned features, but does not aspire to live in a solitary place and instead appreciates the association of other people? Are we ready to say that according to Bhagavad-gītā’s own standard and textual references such a person is not in knowledge?

This objection can be sorted out by taking these verses as defining not an ordinary concept, but what is known in the literature as a cluster-concept (Achinstein, 1968). In considering the concept of man, for example, we could try to define it by listing all the properties P₁, P₂, …, P_n which are attributed to a man. However, for almost all the properties P_i, we can answer the question Could there be a man without P_i? in the affirmative. We do not for example want to say that someone who has a mental illness and has lost his rationality is no longer entitled to be called a man. On the other hand, it seems absurd that the concept of man has no well-defined set of properties through which we can identify its extension. In order to resolve this sort of difficulty, philosophers have introduced the notion of cluster concept. Roughly speaking, cluster concepts are concepts that are defined not by a single property set, but by a cluster of properties. If we remove one element from a set, the set changes, but we can remove one element from a cluster without changing the identity of the cluster. Thus, while to abandon a large number of P_i’s would be felt as an arbitrary change in the meaning of the word man, if most of the properties of the cluster are present, we would still be inclined to say that we are dealing with an extension of the concept of man. In our case, if we agree to take the notion of knowledge, as behaviouristically defined in these verses, as a cluster concept, then if in a particular case just a few of the properties found in the definition are missing in someone we would still be inclined to classify such a person as someone possessing knowledge.

However and despite of this, there are two points which make this account of knowledge quite unusual. First, even though there are behaviorist accounts of belief in the philosophical literature, one hardly would find a behaviorist account of knowledge. The reason for that is very simple: knowledge is supposed to have an external component, namely truth, which, adopting a realist account (Kirkhan, 1995), does not depend on the behavior of the knowing subject, so that a purely behaviorist account to knowledge is prone to be incomplete. Note that we are not here dealing with a content approach, to which the distinction between external and internal component does not apply, but with a knowing subject approach to the analysis of knowledge. As such, it seems necessary to minimally clarify how the truth aspect of knowledge contributes to the whole idea.

Second, there is nothing in these verses that could be related to the content of knowledge, like there is in the brief example of a behaviorist account of belief we gave in Section 2. Since what we have here is merely a specification of which behavior characterizes one who is in knowledge, there always remains the question: What does the one who is in knowledge know? See that this question is in fact preceded by one which concerns how general our behaviorist account is. In other words, does it aim at accounting for all sorts of knowledge, or for just a special kind of knowledge?^[8] Neither is there any information in these verses about how this knowledge could be justified. This is of course important because the behaviorist account we developed from these verses belongs to what we have called knowing subject approach, to which the reasons why the subject knows what he knows play a crucial role.

In order to respond to these objections, we shall in the first place examine the question mentioned above of whether the Bhagavad-gītā aims at accounting for all sorts of knowledge or for just a special kind of knowledge. First of all, what if the Bhagavad-gītā deals not with only one kind of knowledge, but with a plurality of knowledge concepts? In other words, what if the word jñāna and the other knowledge-related terms used in the Bhagavad-gītā express not one, but several different notions of knowledge? As it shall become evident in the following sections, answering this question shall play a very important role in our conceptual analysis.

Second, even though these verses do not really allow us to precisely figure out what kind of knowledge the knowledgeable person knows, by describing a very singular and indeed quite rare kind of behavior they point to how extraordinary and unusual this knowledge is. In fact, there are several verses in the Bhagavad-gītā which talk about knowledge in connection with quite unusual characteristics. We have for instance that it relieves one from the miseries of all evil and inauspiciousness (9.1), that it is the king of education, the most secret of all secrets and the supreme purifier (9.2), that it enables one to see the real nature of things (4.35) and taste the eternal (13.13) and that through it one attains a transcendental platform (14.2). We have also that it includes knowledge of all other things: it reveals everything, as the sun lights up everything in daytime (5.16), so that it being known, nothing further shall remain to be known (7.2).

From this we can conclude that what we have at hand is a sort of metaphysical knowledge: a knowledge which allows one to reach a realm which transcends the world with which we daily relate through our senses and which we make sense of through common sense and science. Indeed, as we hope to be able to show in the course of the paper, one of the distinguished features of Bhagavad-gītā’s epistemology is the crucial role that the possession of knowledge plays in attaining what might be said to be the general goal of the Bhagavad-gītā.

5. A Plurality of Knowledge Concepts: jñāna vs. Vijñāna

It seems clear that the Bhagavad-gītā acknowledges the existence of more than one kind of knowledge. Verse 32 of chapter 3, for example, refers to those who are bewildered by all sorts of knowledge (sarva-jñāna-vimūhās). Also, in 18.19, it is said that according to the three different modes of material nature (tridhaiva gua-bhedata), there are three kinds of knowledge (jñānam). Verses 18.20 to 22 go on to explain what these three kinds of knowledge are. It is interesting to note that we have here three different knowledge concepts being described through the same word. Therefore, it seems reasonable to say that in the Bhagavad-gītā there is a plurality of knowledge concepts, not necessarily expressed by three different words.

Of particular interest to us is a specific conceptual distinction usually associated with the terms jñāna and vijñāna. Even though these terms have quite a great variety of meanings, being even used interchangeably, we are interested here in a specific use of them present for instance in verse 8 of chapter 6:

A person is said to be established in self-realization and is called a yogī when he is fully satisfied by virtue of acquired knowledge [jñāna] and realization [vijñāna]. Such a person is situated in transcendence and is self-controlled. He sees everything – whether it be pebbles, stones or gold – as the same. (6.8) (Prabhupada, 1986: 315)

One whose self is content in knowledge [jñāna] and in realized knowledge [vijñāna], who is focused on the highest with senses conquered – That one, “absorbed in yoga,” is said to the a yogi, for whom earth, stones, and gold are the same. (6.8) (Schweig, 2007: 93)

The same distinction seems to be present in verse 42 of chapter 18, where Ka sharply distinguishes jñāna from vijñāna while describing the natural qualities of a brāhmaa: while jñāna is translated simply as knowledge (Schweig, 2007: 231) (Sargeant, 1994: 703) (Prabhupada, 1986: 827), vijñāna is translated as realized knowledge (Schweig, 2007: 231), discrimination (Sargeant, 1994: 703) or wisdom (Prabhupada, 1986: 827).

Now, what is the difference between the concepts expressed by these two words? The official answer is to say that while jñāna means theoretical knowledge, vijñāna means wisdom or practical or realized knowledge (Wayman, 1955). In other words, while jñāna or theoretical knowledge is that knowledge which can in principle be acquired by everyone, perhaps mainly through the holy scriptures, and which does not necessarily involve change of behavior, vijñāna or realized knowledge is full understanding, grasping or realization of that theoretical knowledge, so that the one who possess it does effectively apply it in life and acts accordingly.

Considering the conceptual framework laid down in Section 2, there are at least two important points we can make about this twofold knowledge distinction. First, while jñāna is susceptible of being approached either by a content (which in fact seems to be the way it is mostly done in the Bhagavad-gītā) or by a knowing subject point of view, it seems that vijñāna has necessarily to be approached by a knowing subject standpoint (even though most verses make it to appear that it is a content approach what is at stake) in the sense that even if we adopt a content approach, the knowing subject point of view has to be there. This is so because, as something whose main raison d’être (at least so far) is to be grasped by a knowing subject so as to allow him to behave in a specific way, realized knowledge cannot dispense in its analysis reference to this knowing subject.

The second point, which is in fact very close to the first one, concerns the extent to which the analysis of these two concepts is dependent on a behaviorist account. By our tentative definition, the answer to that is quite clear: while jñāna can in principle be approached non-behavioristically – for it is at least in principle possible for someone to have jñāna at the same time that his ordinary behavior remains the same – a full analysis of vijñāna must take into account the behavior of the knowing subject.

At this point it should be recalled the reasons we have given in the previous section to conclude that the knowledge concept of Bhagavad-gītā (which perhaps we should refer at this point as the main one among the several knowledge concepts of Bhagavad-gītā) is essentially metaphysical. Given this, one might fairly wonder how this metaphysical knowledge (or how the metaphysical aspect of the concept of knowledge of Bhagavad-gītā) relates to the concepts of jñāna and vijñāna.

Given the two points laid down above about the connection between behavior, jñāna and vijñāna, one might be inclined to say that since this metaphysical knowledge seem to necessarily involve behavior in its description – for it seems reasonable to suppose that one who has reached the ultimate reality or something alike should behave somehow differently – it cannot be taken as the same as jñāna. Consequently, by exclusion, it must be what we have called realized knowledge or vijñāna. This in fact seems to be corroborated by verse 19 of chapter 13:

Thus the field of activities, knowledge [jñāna] and the knowable have been summarily described by Me. Only My devotees can understand [vijñāya] this thoroughly and thus attain to My nature. (13.19) (Prabhupada, 1986: 662)

Thus the field, as well as knowledge [jñāna] and the object of knowledge, have been briefly described. One whose love is offered to me, who realizes [vijñāya] this knowledge, comes forth to my state of being. (13.18) (Schweig, 2007: 180)

Here, after reminding Arjuna what has been described by Him – the field of activities, which can be taken as being the body, knowledge (jñāna, which has been behavioristically described in verses 8-12) and the object of knowledge (jñeya) – Ka uses the gerund form of the verb form of vijñāna (vijñāya) to say that by understanding or comprehending (vijñāya) these three things, His devotee approaches His state of being (mad-bhāvāyopapadyate) (Sargeant, 1994: 546). Needless to say, approaching the state of being of the Supreme Being, which, as those familiar with the Bhagavad-gītā know and as we shall see later, is Ka, is something which, by definition, only a metaphysical knowledge can entail.

However reasonable it may be, this account conflicts with what seems to be said in the first verse of chapter 9:

The Supreme Personality of Godhead said: My dear Arjuna, because you are never envious of Me, I shall impart to you this most confidential knowledge and realization, knowing which you shall be relieved of the miseries of material existence. (9.1) (Prabhupada, 1986: 447)

The Beloved Lord said: Now I shall reveal to you this greatest secret, for you are without envy. It is knowledge together with realized knowledge, knowing which you shall be free from inauspiciousness. (9.1) (Schweig, 2007: 127)

In this verse, Ka says that He shall declare to Arjuna the most secret thing (guhyatamam): knowledge and realization combined (jñāna vijñāna-sahita) – after which He adds that having learned this combination of knowledge and realization Arjuna shall be released from evil (Sargeant, 1994: 377). The important thing for us here is that what shall release Arjuna from evil is not vijñāna only, nor jñāna, but a combination of both. We may then conclude that the metaphysical knowledge we were talking about is not only vijñāna, but a dual concept composed by jñāna and vijñāna. Verse 6.8, which we have quoted at the beginning of this section, also seems to corroborate this thesis: it is the self who is satisfied with both knowledge and realization (jñāna-vijñāna-tptātmā) who, in Sargeant’s translation, attains the mystical stage of samādhi (Sargeant, 1994: 279), not the one who has only vijñāna (or only jñāna).

It is therefore this interpretation we shall adopt: the (main notion of) knowledge the Bhagavad-gītā is concerned about is a composite concept with a theoretical aspect, corresponding to the notion of jñāna, and a practical or realized one, corresponding to the notion of vijñāna. Regarding verses like 13.8-12 which are totally behaviorist, we would just say that they overemphasize the specific practical aspect of this knowledge concept, which does not in any way denies that it might have other aspects.

6. A Foundationalist Account of Knowledge

An important issue we have not addressed yet is the question of how these two concepts of jñāna and vijñāna relate to each other. The answer we shall give to that shows what we think to be an interesting parallel between the jñāna / vijñāna distinction and the foundationalist account of justification we showed in Section 2. It is through this parallel that we will be able to answer the question raised in section 4 about the justification of the Bhagavad-gīta concept of knowledge.

Central to most Indian schools (darśanas) is the notion of pramāa, a term which literally means evidence or proof (Mohanty, 1994). In Nyāya epistemology, for example, there are four pramāas: pratyaka or perception, anumāna or logical inference, upamāna or comparison and śabda or verbal testimony (Matilal, 1992). In Vedānta epistemology, on the other hand, there are traditionally just three pramāas: pratyaka, anumāna and śabda (Swami 1998). In what follows, we shall take the Vedānta school along with its emphasis on the three mentioned pramāas as a paradigmatic instance of orthodox Indian philosophy upon which we shall base some aspects of our analysis.

In the context of Indian epistemological debate, the term pramāa has had a double meaning, referring both to the causally effective instruments for acquiring knowledge as to the justificatory grounds on which to make knowledge claims. Thus pratyaka (perception) can be taken as a way to acquire knowledge (the way for instance I get knowledge about the color of the laptop I am in front of) as well as the justification of a special sort of knowledge claim (How do I justify the claim that I know that the color of my notebook is black? By referring to a particular visual perception of mine). The same holds for anumāna and śabda. Suppose I am seated here inside my office and get a call from the department secretary saying that Mr. such and such is coming up. Knowing her to be a reliable person who does her job very efficiently I can say I have acquired from her verbal testimony (śabda) the new piece of knowledge that Mr. such and such is coming up. But how do I justify this knowledge claim? Simply by saying that I got a call from the secretary saying that Mr. such and such is coming up and so on and so forth. Suppose now that a few minutes later I hear a noise of the door being knocked. Together with the knowledge I have acquired from the secretary’s phone call, I can use very simple logical reasoning (anumāna) to conclude that Mr. such and such is outside wishing to come in. But how do I justify the claim that I know that Mr. such and such is outside wishing to come in? By explaining the very same reasoning chain through which I got to know that in the first place.

This aspect of pramāas as justificatory grounds for knowledge claims is significant for two reasons. First, there will be, for each one of the three pramāas, a specific kind of knowledge-claim: pratyaka-claims (that is, knowledge claims which are justified through pratyaka), anumāna-claims (knowledge claims which are justified through anumāna) and śabda-claims (knowledge claims which are justified through śabda)^[9]. Second and more importantly, these knowledge-claim kinds can be linked in a very natural way to the foundationalist account of justification we discussed in Section 2. As one should remember, according to foundationalism, knowledge claims^[10] are divided into basic and nonbasic claims, the latter receiving justification from the former through logical inference. Basic claims in their turn do not get justification from any claims, but from something else. Given this, it is easy to see that while pratyaka-claims and śabda-claims are basic, owing their justification to pratyaka and śabda, respectively, anumāna-claims are nonbasic, owing their justification to anumāna (logic) applied to basic claims, nonbasic claims or to a mixture of basic and nonbasic claims (these nonbasic claims being in their turn justified in the same way, until we get only basic claims.)

Now, which of these pramāas shall be taken as the justification of our two knowledge concepts, and how shall they be related to each other in our foundationalist framework? First of all, although acknowledging the existence of the three above mentioned pramāas, Vedānta philosophy puts much greater emphasizes on śabda, defending that out of the three methods of acquiring knowledge, that of receiving knowledge from higher authorities (śabda) is the most perfect (Swami, 1998). While śabda is taken as imperative, pratyaka and anumāna are taken as supportive. Furthermore, the śabda on which Vedānta philosophy relies is not ordinary verbal testimony, but the special kind of testimony which comes from śāstra (scriptures), guru (spiritual master) and sādhu (other practitioners respected for their realization of the teachings of guru and śāstra).

Recall that we have defined jñāna as that knowledge which can in principle be acquired by everyone, mainly through the holy Scriptures, and which does not necessarily involve change of behavior. This means that, in some sense at least, the notion of jñāna, as we are using it here, is connected with knowledge given by śāstra, which is one of the sources of śabda. Now, this śāstra-śabda knowledge is clearly an important part of Bhagavad-gīta’s epistemology:

One should therefore understand what is duty and what is not duty by the regulations of the scriptures. Knowing [jñātvā] such rules and regulations, one should act so that he may gradually be elevated. (16.24) (Prabhupada, 1986: 766)

Therefore, let scripture be your authority for understanding what action should be performed and what action should not be performed. Knowing [jñātvā] the prescribed scriptural injunctions, you are obliged to enact such action in this world. (16.24) (Schweig, 2007: 208)

The other source of śabda (guru and sādhu) is also clearly mentioned in the Bhagavad-gīta:

Just try to learn the truth by approaching a spiritual master. Inquire from him submissively and render service unto him. The self-realized souls can impart knowledge [jñānam] unto you because they have seen the truth. (4.34) (Prabhupada, 1986: 262)

Learn this by humble submission, by thorough inquiry, and by serving. They will impart this knowledge [jñānam] to you, for they are knowers and seers of the truth. (4.34) (Schweig, 2007: 78)

Given this, it seems reasonable to say that what we are calling theoretical knowledge or jñāna is simply śabda-justified knowledge, that is, knowledge whose claims are justified (and acquired) through śāstra, guru and sādhu^[11].

How about vijñāna? Can we say its knowledge claims are justified in the same way, thought śabda or words? In order to answer this question we must take a look at a specific verse of Bhagavad-gītā which contains a very important hint as to the nature of this realized knowledge:

This knowledge is the king of education, the most secret of all secrets. It is the purest knowledge, and because it gives direct perception of the self by realization, it is the perfection of religion. It is everlasting, and it is joyfully performed. (9.2) (Prabhupada, 1986: 449-450)

This is the king of knowledge, the king of the secrets, the ultimate means of purification. Understood by direct perception, in harmony with dharma, it is joyful to perform and everlasting. (9.2) (Schweig, 2007: 127)

In this verse Ka continues to glorify the combined knowledge He has introduced in the previous verse (which we have quoted above.) Among other things, He says that this knowledge is pratyakāvagama, which might be translated as that whose understanding is before the eyes (Sargeant, 1994: 378) or that which can be as directly perceived (as Prabhupada and Schweig translate it) or grasped as something we perceive with our eyes. See that this expression is composed by the word pratyaka, which is the very same word used to refer to the way of acquiring (or justifying) knowledge through sensory perception. (The other word which compose the expression – avagamam – means understood.)

Does this mean that the same knowledge which Ka classifies as the king of knowledge, the most secret of all secrets and the ultimate means of purification is merely ordinary sensory-perception acquired knowledge? Clearly it does not. The word pratyaka is here to indicate that this knowledge is not mediated by anything. It is perception, in the sense of something directly experienced, but a different, supposedly higher kind of perception or phenomenal experience, which has been called by the sixteenth century Indian philosopher Jiva Goswami (Elkman and Gosvami, 1986) vaidua-pratyaka. Since we have already found enough evidence to take jñāna or theoretical knowledge as justified through śabda, we take this particular part of verse 9.2 as referring exclusively to the vijñāna part of our composed knowledge concept. So then, the way of acquiring (and justifying) vijñāna knowledge claims is not through śabda, but through a special kind of pratyaka-pramāa or perceptual evidence.

Now, how shall these things be put together in order to form a foundationalist epistemic structure? First, since jñāna and vijñāna knowledge claims are justified through śabda-pramāa and through a special kind of pratyaka-pramāa, respectively, they are of course basic knowledge claims. Nonbasic claims shall be obtained in the traditional way described above, through anumāna (logic) applied to basic claims, nonbasic claims or to a mixture of basic and nonbasic claims (these nonbasic claims being in their turn justified in the same way, until we get only basic claims.) Besides jñāna and vijñāna knowledge claims, (ordinary) pratyaka justified claims might also act as basic claims, secondarily supporting the derivation of nonbasic claims.

Now we can have a whole picture of our logical reconstruction of the concept of knowledge of Bhagavad-gīta. First, the main knowledge concept dealt with in the Bhagavad-gīta is a dual notion composed by a theoretical side, called jñāna, and by a practical, behaviorist and realized one, called vijñāna. While jñāna is acquired and justified through śabda, which in the Vedānta tradition would mean the holy scriptures (śāstra) and the words of self-realized souls (guru and sādhu), vijñāna is acquired and justified through a sort of mystical perceptual experience called vaidua-pratyaka. But when we consider that there might be further, logically derived knowledge-claims obtained both from jñāna and vijñāna, and that ordinary sensory perception (pratyaka) has always played an important role in orthodox Indian epistemological theories, such as the one of Vedānta, we have the following, considerably more complex foundationalist-structured view the notion of knowledge of the Bhagavad-gīta.

First, there are three kinds of basic knowledge claims: the claims justified through pratyaka-pramāa, the claims justified through śabda-pramāa and the ones justified through vaidua-pratyaka-pramāa. While the two latter correspond to (basic) theoretical and (basic) realized knowledge, respectively, the former is ordinary perceptual knowledge which shall always play a secondary role in the construction of nonbasic knowledge claims, being subordinate, we may say, to jñāna and vijñāna. Nonbasic knowledge claims are obtained through anumāna, applied to pratyaka-pramāa justified claims or to vaidua-pratyaka-pramāa justified claims (possibly along with pratyaka-pramāa justified claims), to nonbasic claims, or to a mixture of basic and nonbasic claims. In the case of nonbasic claims obtained from jñāna and vijñāna knowledge claims, we have an example of an effective mixing of theoretical and realized knowledge.

7. Propositional vs. Objectual Knowledge

Right before laying down the behaviorist account of knowledge in Section 4, we have mentioned that the most natural way to start out our investigation would be to look for verses in the Bhagavad-gīta saying that knowledge is so and so. Besides the ones we used in that section, there is one more verse, also from chapter 13, which deserves attention^[12]:

He is the source of light in all luminous objects. He is beyond the darkness of matter and is unmanifested. He is knowledge [jñāna], He is the object of knowledge [jñeya], and He is the goal of knowledge [jñāna-gamyam]. He is situated in everyone's heart. (13.18) (Prabhupada, 1986: 661)

Also, of luminaries it is said to be the luminary beyond darkness. It is knowledge [jñāna], what is to be known [jñeya], and the purpose of knowledge [jñāna-gamyam] – it is seated in the hearth of everyone. (13.17) (Schweig, 2007: 180)

In this verse Ka is continuing a description of the qualities of Brahman which he has started in verse 13:

I shall now explain the knowable [jñeya], knowing which you will taste the eternal. Brahman, the spirit, beginningless and subordinate to Me, lies beyond the cause and effect of this material world. (13.13) (Prabhupada, 1986: 655)

I shall describe what is to be known [jñeya], knowing which one attains immortality: The beginningless supreme Brahman is said to be neither existent nor nonexistent. (13.12) (Schweig, 2007: 178)

The important thing for us here is of course the use of the words jñāna (knowledge), jñeya (object of knowledge), and jñāna-gamyam (goal of knowledge). While in verse 18 it is said that Brahman is knowledge and the goal of knowledge, in both verses 13 and 18 it is said that this Brahman is the object of knowledge. So, apparently much of Bhagavad-gīta’s epistemology lies on the notion of Brahman: so much that it is identified both with knowledge, the object of knowledge and the goal of knowledge. But what is Brahman?

We should first say that this is not a trivial question. In fact, the concept of Brahman is among one of the most multifarious and important concepts of Indian philosophy. According to the Monier-Williams Sanskrit Dictionary, Brahman means

[…] the one-self existent impersonal Spirit, the one universal soul (or the divine essence and source from which all created things emanate or which they are identified and to which they return), the Self-existent, the Absolute, the Eternal. (Monier-Williams, 1974: 738)

This might make us fairly conclude that the concept of Brahman is somehow related to the western concept (or family of concepts) of God. While it is not our purpose here to do a conceptual analysis of the notion of Brahman (which would surely require another essay), we shall observe that as far as the Bhagavad-gīta and its epistemology are concerned, there is an intriguing relation between the notion of Brahman and the main speaker of the text, Ka:

And I am the basis of the impersonal Brahman, which is immortal, imperishable and eternal and is the constitutional position of ultimate happiness. (14.27) (Prabhupada, 1986: 707)

Truly, of Brahman, I am the foundation – and of the immortal that is everpresent, also, of dharma, that is everlasting, and of happiness that is extraordinary. (14.27) (Schweig, 2007: 192)

Here Ka says He is the pratihā, foundation, support or basis (Sargeant, 1994: 589) of Brahman. Elsewhere it is stated what seems to be a much more intimate connection between Ka, Brahman and, we might say, the concept of God:

Arjuna said: You are the Supreme Personality of Godhead [para brahma], the ultimate abode, the purest, the Absolute Truth. You are the eternal, transcendental, original person, the unborn, the greatest. (10.12) (Prabhupada, 1986: 523)

Arjuna said: You are the supreme [para] Brahman, the supreme dwelling place, the supreme means of purification; the eternal divine Person [purua śāśvata divyam], the original Divinity [ādi-devam], the unborn, all pervading one. (10.12) (Schweig, 2007: 140)

This is one of most fascinating aspects of the Bhagavad-gīta: the person who at the beginning of the Bhagavad-gīta seems to be playing the mere role of a charioteer and afterwards becomes Arjuna’s counselor, is described as being in reality the Supreme Spirit or Brahman, the divyam, Godly or divine purua (being or person) and the ādi-devam or primal God (Sargeant, 1994: 422). In fact, Ka is said elsewhere (see the quotations below) to be the father, the mother, the establisher and the grand-father of the universe (pitāham asya jagato mātā dhātā pitāmaha), the ancient Spirit (purua purāas), the supreme dwelling place (para dhāma) and that by which all the universe is pervaded (tvayā tata viśvam) (Sargeant, 1994: 393, 490). As far as our epistemological analysis is concerned, He is said also to be vedyam, the object of His own teachings in particular and of all knowledge in general:

I am the father of this universe, the mother, the support and the grandsire. I am the object of knowledge [vedyam], the purifier and the syllable o. I am also the g, the Sāma and the Yajur Vedas. (9.17) (Prabhupada, 1986: 477)

I am the father of this universe, the mother, the creator, the grandfather; That which is to be known [vedyam], the means of purification, the sacred sound OM, and indeed the ig, the Sāma and the Yajur. (9.17) (Schweig, 2007: 131)

You are the original Personality of Godhead, the oldest, the ultimate sanctuary of this manifested cosmic world. You are the knower of everything [vettā], and You are all that is knowable [vedyam]. You are the supreme refuge, above the material modes. O limitless form! This whole cosmic manifestation is pervaded by You! (11.38) (Prabhupada, 1986: 583)

You are the original Divinity, the ancient cosmic Person, you are the highest resting place of this universe. You are the knower [vettā], what is to be known [vedyam], and the highest domain. By you this universe is pervaded, O One of Unlimited Form. (Schweig, 2007: 162)

Given these textual evidences, both the ones concerning the divinity of Ka and His relation to Brahman as well as the ones which establish Him as the object of knowledge, and overlooking the conceptual distinctions that surely exist between terms such as Brahman and Purua, we shall take Ka as being that divine entity which verse 18 of chapter 13 refers to. Therefore we say that, according to the Bhagavad-gīta, Ka is both knowledge (jñānam), the object of knowledge (jñeyam) and the goal of knowledge (jñāna-gamyam).

But, what does this mean? What does it mean to say that Ka is knowledge? Needless to say, the notion of jñeyam or object of knowledge is the same as the one we have called in Section 2 content of knowledge. But notice that Ka is said here to be knowledge and the object or content of knowledge, which means that the notions of knowledge and content of knowledge are being taken as identical. It seems then we are facing something alike to a content approach to the analysis of knowledge. As one must remember, according to this approach, knowledge is, at least for the purpose of philosophical analysis, taken as being the same as the object or content of knowledge. In this way, saying that Ka is knowledge might be taken as just a (content approach) shortcut of saying that He is the content of the Bhagavad-gītā concept of knowledge. Recall however that we have been working so far under the assumption that the content of knowledge is a proposition. But here we have a verse saying that the content of knowledge is Ka, who is definitely not a proposition. How then are we to accommodate this into a content approach to the analysis of knowledge?

A simple way to sort this out is to take this verse as saying not that Ka is the content of knowledge, but that propositions about Ka are so. In fact, many verses in the Bhagavad-gītā stress the importance of knowing things about Ka (as well as the result of having such knowledge):

One who knows [vetti] the transcendental nature of My appearance and activities does not, upon leaving the body, take his birth again in this material world, but attains My eternal abode, O Arjuna. (4.9) (Prabhupada, 1986: 230)

One who truly knows [vetti] the birth and acts of my divine being, upon relinquishing the body, does not come to another birth – such a person comes to me, O Arjuna. (4.9) (Schweig, 2007: 71)

A person in full consciousness of Me, knowing Me [jñātvā mā] to be the ultimate beneficiary of all sacrifices and austerities, the Supreme Lord of all planets and demigods, and the benefactor and well-wisher of all living entities, attains peace from the pangs of material miseries. (5.29) (Prabhupada, 1986: 305)

As the beloved recipient of sacrifices and austerities, as the exalted supreme Lord of all the worlds, as the innermost heart of all beings – thus knowing me [jñātvā mā] one attains peace (5.29) (Schweig, 2007: 90)

In this view therefore, saying that Ka is the content of knowledge must be seen as a shortcut of saying that propositions about Him are the content of knowledge.

At this point one might be wondering how this new and indeed quite important piece of conclusion fits into the analytic conceptual framework we have developed so far. To start with, there clearly is a close relation between the idea that propositions about Ka are the content of Bhagavad-gīta’s (main) notion of knowledge and what we have been calling jñāna or theoretical knowledge. As one should recall, we have defined jñāna as that knowledge which can in principle be acquired by everyone through the scriptures (śāstra) and the words of realized souls (guru and sādhu) and which does not necessarily involve change of behavior. Regarding śāstra, a very good part of Bhagavad-gīta itself as well as of other Indian holy scriptures such as the Bhāgavata Purāa, Mahābhārata and Rāmāyaa are dedicated to describing the form (rūpa), qualities (gua) and activities (līlā) of Ka and His avatāras. As far as guru and sādhu are concerned, there have been powerful oral traditions inside the Vedānta school such as Gaudya Vaiavism (Elkman and Gosvami, 1986) which have put great emphasis on speaking and listening about Ka. In fact, this is something which verse 4.9 above seems to strongly encourage.

But while this account fits well jñāna or theoretical knowledge, it does not suffice for vijñāna or realized knowledge. First, knowing a specific kind of proposition, in the sense of understanding its meaning and accepting its truth, does not necessarily involve change of behavior. Second, a linguistic entity such as a proposition does not seem to be the sort of thing which would require a mystical perception as its source.

Notice that following the philosophical tradition, we have throughout most of the paper acted as if propositional knowledge – knowledge whose content is a proposition – were the only knowledge that matters. This we have done both in our exposition of contemporary epistemology, in particular in our explanation of the content approach, as in our (so far) analysis of the concept of knowledge in the Bhagavad-gītā. Recall however that at the beginning of Section 2 we have talked about another kind of knowledge, one which has an object as content and might therefore be called objectual knowledge. Knowledge of persons, places and things would be classified as objectual knowledge in this sense.

Now, despite any wondering we might have about the meaning of terms like para brahman, divyam purua or ādi-devam, we can quite fairly say that all these words, as well as the word Ka, involve in their meaning reference to an object. Thus, an alternative way to understand the claim that Ka is the content of knowledge is simply to take it literally, meaning that this object or Supreme Cosmic Person referred to by the word Ka (and not propositions about Him) is the content of Bhagavad-gītā’s knowledge. Instead of being an indirect awareness of the Lord’s features obtained through some specific kind of proposition, knowledge here will be a direct acquaintance with the Lord Himself.

Needless to say, it is this interpretation which we shall use to account for the vijñāna aspect of our knowledge concept: while jñāna has as content propositions, vijñāna has as content Ka Himself. This of course is the same as saying that while jñāna is propositional knowledge, vijñāna belongs to the class we have called objectual knowledge. Therefore, we are using in our analysis of the Bhagavad-gītā concept of knowledge both propositional and content knowledge: while one part of our composite concept of knowledge, which corresponds to one of the basic knowledge claims of our foundationalist concept, is propositional knowledge, the other part, which corresponds to the other sort of basic knowledge claims, is objectual knowledge. Recall however that in both cases we are inside a content approach to the analysis of knowledge, for we are equating knowledge with its content. And it is interesting to note that we have arrived at this account from the same line of verse 13.18 (jñāna jñeya jñāna-gamya.) The particularity was that while interpreting it non-literally we arrived at a propositional-content approach to knowledge (which we have identified with jñāna), interpreting it literally led us to an objectual-content approach (which we have identified with vijñāna).

One very important consequence of taking jñāna as propositional knowledge and vijñāna as objectual knowledge relates to the connection we made at the beginning of Section 2 between these two kinds of knowledge and what we called knowledge by description and knowledge by acquaintance. The point is that being jñāna a kind of propositional knowledge, it can be said to be an indirect awareness of the Lord’s features obtained descriptively through some sort of proposition. On the other hand, being vijñāna objectual knowledge, it can be seen as a direct, not mediated acquaintance with the Lord Himself. This explains why while jñāna can be obtained (and justified) through śabda, vijñāna requires some special kind of pratyaka or perceptual experience, in this case a higher, mystical one which allows one to see or directly experience the Supreme Being.

In fact, many passages in the Bhagavad-gītā seem to make reference to this vaidua-pratyaka-pramāa as a direct mystical experience of the divine Supreme Being. Surely the most famous ones are related to Arjuna’s experiencing Ka’s universal form or viśva-rūpa in Chapter 11, to which it is required that Arjuna possess a special, divine or heavenly (divyam) kind of vision (cakus) (Sargeant, 1994: 460):

But you cannot see Me with your present eyes. Therefore I give you divine eyes [divyam caku]. Behold My mystic opulence! (11.8) (Prabhupada, 1986: 559)

But you are unable to see me with only this, your own yes. I therefore give divine eyes [divyam caku] to you – behold my superbly powerful yoga! (11.8) (Schweig, 2007: 153)

Elsewhere, the fact that the capacity to see or experience Ka is given by Him is stated somehow more emphatically:

To those who are constantly devoted to serving Me with love [bhajatām prīti-pūrvakam], I give the understanding [buddhi-yogam] by which they can come to Me. (10.10) (Prabhupada, 1986: 520)

For them, who are constantly absorbed in yoga, who offer loving service with natural affection [bhajatām prīti-pūrvakam], I offer that yoga of discernment [buddhi-yogam] by which they come close to me. (10.10) (Schweig, 2007: 139)

This is sufficient to say that vijñāna, understood as knowledge of or acquaintance with the divine Supreme Being is knowledge by revelation. And as far as the Bhagavad-gītā in general and the above quoted verse in particular are concerned, the revelation which allows one to see the divyam purua is not gratuitous: it is given to those who offer loving service with affection to Him (bhajatām prīti-pūrvakam). This strong theological point establishing bhakti or devotional love as a means to know the Supreme Being, and as this knowledge as tantamount to going to Him, is clearly stated in one of the last verses of the Bhagavad-gītā:

One can understand Me as I am, as the Supreme Personality of Godhead, only by devotional service [bhaktyā]. And when one is in full consciousness [jñātvā] of Me by such devotion, he can enter into the kingdom of God. (18.55) (Prabhupada, 1986: 839)

By offering love [bhaktyā] one recognizes me fully and who I am in truth. Once knowing [jñātvā] me in truth, one comes to me immediately. (18.55) (Schweig, 2007: 235)

8. Conclusion

We have in this paper tried to achieve a better understanding of the notion of knowledge in the Bhagavad-gītā through a method alike to that of contemporary philosophers working in the analytic tradition. What we have done was basically to reconstruct the notion of knowledge described in some key verses of the Bhagavad-gītā from the standpoint of the standard conceptual framework of contemporary analytic epistemology. Even though, as in any other process of logical reconstruction, we went beyond what it is literally said in these verses, we do think that the ideas presented here show how it might be fruitful to use contemporary analytic methods to acquire a better understanding of some key aspects of such an important religious text as the Bhagavad-gītā^[13].

Recebido: 01/11/2008
Aceite final: 15/12/2008

página pertencente ao site http://www.pucsp.br/rever
Revista Eletrônica Rever - ISSN 1677-1222