D.M. Datta: The Six Ways of Knowing

A Critical Study of the Advaita Theory of Knowledge

Calcutta, University of Calcutta, 1972


Introduction: Prama and Pramana

pp. 20-21: "Broadly speaking there are at least four different views about
truth. According to one view the truth of knowledge consists in its practical
value and reveals an object that serves some purpose (artha or prayojana).

"Another view, that we find chiefly in the Nyaya works, regards truth as the
faithfulness with which knowledge reveals its object.

"A third view regards truth as a harmony of experience (samvada or
samvaditva). A true knowledge is in harmony with other experiences.

"The Advaita school of Vedanta favours a fourth view according to which the
truth of knowledge consists in its non-contradictedness (abadhitartha).

The [second] view cannot directly prove itself. The only way to prove
correspondence [of knowledge and real object] is to fall back on the foreign
method of consilience or coherence (samvada) - that is to infer the existence
of a real correspondence between knowledge and reality from the facts of the
harmony of experience. But all we can legitimately infer from the harmony of
knowledge with the rest of our experience up to that time is not that
knowledge is absolutely free from error, but it is not yet contradicted. As
regards the [first view's] test of pragmatic efficiency (artha-kriya-karitva),
the Advaitins argue that even a false cognition may lead to the fulfillment of
a purpose. The hypothesis that the earth is stationary and the sun is moving
has been working quite satisfactorily for ages; on the basis of this cognition
many of our actions are performed and purposes attained. It is only its
conflict with astronomical phenomena that enables us to detect its falsity."

pp. 22-23: "According to the Advaitins prama or knowledge must have as one its
characteristics truth; and the truth of prama consists in its content being
uncontradicted (abadhitartha-visayakatva).

"The second characteristic of prama or knowledge is novelty (anadhigata). The
question whether memory should be admitted to have the status of knowledge:
But though memory is not a distinct source of knowledge, it is still a
distinct experience that has to be distinguished from knowledge and given a
separate name. The experience which reveals the new (i.e. knowledge proper) is
called anubhuti, whereas reproduced knowledge is called smrti."

pp. 27-28: "A prama or knowledge can be regarded as a cognition, the object of
which is neither contradicted nor already known as an object
(anadhigata-badhita-rtha-visayam jnanam).

"The special source of a particular prama or knowledge is called pramana.
Pramana is defined as the karana of a prama. A karana is conceived as the
unique or special cause through the action of which a particular effect is
produced (vyaparavad asadharanam karanam karanam). The mind is a cause of the
existence of which is common to all sorts of knowledge.

"A cause, to be called a karana, must not be merely unique (asadharana); it
must also possess sine? active function (vyapara).

"A pramana is, then, such an active and unique cause (karana) of a prama or
knowledge."

Perception (pratyaksa)

pp. 31-33: In Western philosophy since the days of Kant a decided right of
precedence has been accorded to the problems and theories of knowledge, and
metaphysics has come to be built on the results of epistemological analysis.
In Indian philosophy metaphysics has generally preceded epistemology. In most
of the schools the nature of reality and the possibility of knowledge have
been ascertained on the grounds of the revealed texts, which have come to be
regarded as the embodiment of spiritual experiences. Consequently epistemology
has been confined to the investigation of the different sources of knowledge
(pramanas) and the problems of truth and error. The declared object of such
epistemological investigation was the ascertaining of the true method of
knowledge of reality, which was almost always conceived to be the way to
liberation.

The Buddhistic thinkers, however, challenged this method. They owed no
allegiance to the revealed texts and attempted a revision of metaphysics
through the criticism of knowledge. The enquiry, like that of Kant, ended
mostly in agnosticism, in the conception of the impossibility and futility of
metaphysics.

"The ultimate postulates of thought require to be justified by a system of
metaphysics which alone can compel its final acceptance. As regards the
problems of immediate knowledge, we can proceed to a certain point
independently of metaphysics. But when we come to the question of the
possibility of such knowledge or to the criticism of its value, no
satisfactory conclusion can be drawn without metaphysical considerations.

"If we closely examine the modern epistemological theories of perception, it
will not be difficult to find that in spite of their loud protests against
metaphysics, epistemologists have tacitly assumed without criticism certain
theories of reality, on truth of which alone their epistemological conclusions
can stand."

p. 37: "Perception may be safely defined as immediate knowledge. Or it may be
defined negatively as knowledge that is not derived through the active agency
of another knowledge. Inference is not immediate knowledge, it is derived from
previous knowledge; so also is memory."

pp. 39-40: "An indriya is not to be identified either with a sense-organ or
with the capacity of the sense-organ. It is a different material substance,
though very subtle. It is composed of the same substance, the quality of which
is sensed by it. For instance the eye, which receives impressions of form or
colour, which is a quality of light, is itself composed of the same substance
as light."

p. 41: [According to Madhvacarya] "it is necessary for every one of the senses
to reach the object and have contact with it in order that the object can be
perceived at all. This theory is stated by saying that every indriya is
prapyakari (capable of working only on reaching the object). [Cf. Vivarana-
prameya-samgraha.]

"If a sense must come into contact with its object, how can we explain the
vision of a distant object or the hearing of a distant sound?

"The real senses are invisible subtle material substances that can expand and
contract with great rapidity.

"As regards the contact of the auditory sense with sound, there is a
difference of opinion. The Nyayikas think that sound itself travels in waves
and thereby reaches the ears. But the Vedantists hold that even in this case
it is the sense that goes out to meet the sound-producing object."

p. 44: "According to the Western theory, when an object is seen, stimuli in
the form of light-waves coming from the object affect the eyes, and in
consequence of the action of the physiological apparatus in the eyes an image
is formed on the retina; thereon some brain centers are stimulated and we have
the vision of the object. The object sends its influences, the sum-total of
the action of which is represented in the formation of the retinal image. In
the place of the object, therefore, we have its representative - the small
inverted picture of the object that is directly given to the organ of sight.
But if this be the whole account, the mystery of how a small inverted image of
the object on the retina makes us see the physical object, great or small,
distant or near in the external world is too great to be ignored by any
enquiring unsophisticated mind."

p. 46: [Manas] "states for mental states and functions. Even when it is
conceived as a positive substance, as by the Nyayikas, it is primarily
conceived as an instrument - an internal sense through which we attend to
inner and outer objects. The rational activities of the mind such as memory 
[characteristic not of mind but of intelligence: SB 3.26.30], comparison, 
etc. are ascribed to the self, atman."

p. 47: "The Prabhakaras, the Bhattas, the Samkhyas, the Nyayikas, and the
Vaisesikas all agree in regarding manas as the internal organ of perception
(antarindriya). They also agree in considering manas as unconscious. The
Indian thinkers regard consciousness as an attribute of, or identical with,
the atman alone. Like the eyes, the ears, the nose, etc. manas also is an
instrument, unconscious but very subtle; and through its instrumentality the
atman obtains the knowledge of objects. Though the activity of manas is
chiefly felt in internal perception yet it is an indispensable factor in
external perception as well.

"The non-spiritual character of manas or mind receives clear emphasis in the
Samkhya school, which considers it as an evolute of ahamkara, which is
ultimately derived from prakrti, from which the entire physical world itself
evolves. As such, manas belongs to the world of unconscious objects, and is
sharply contrasted with atman or purusa (i.e. self) whose essential attribute
is consciousness."

p. 48: "According to some Vedantins antahkarana has four different aspects or
functions which are named: (1) manas; (2) buddhi; (3) ahamkara; and (4) citta.
Manas represents the indecisive state of antahkarana or mind. Buddhi stands
for antahkarana in its state of decision. Ahamkara is the state of antahkarana
having some reference to the self. Citta is antahkarana in its state of
remembering, i.e. referring to past events."

p. 49: "According to the Nyayikas, manas is a distinct dravya or substance,
that is, as such, co-ordinate to other realities like the material elements,
the self, etc. According to the Samkhyas also manas is a non-spiritual entity,
that is co-ordinate to and independent of the self. According to the
Advaitins, however, the self or atman is the only independent reality; the
material, the physical or the objective are but creations of ajnana, i.e.
ignorance or nescience. Antahkarana is but a product of this primal,
beginningless nescience, through a confused identification with which the self
gets individuated. The self is caitanya (consciousness) itself, and
consciousness is not a contingent product of any interaction between the
subject and object - it shines in its own light; it is called svayamprabha or
svayamprakasa or self-shining."

p. 51: "The atman does not appropriate the ajnana, but simply lights it up
just as a lamp."

p. 53: "In the Brahma-sutras we come across an attempt at proving the
existence of antahkarana. It is said [there] that if the self, the senses, and
the object were the only three factors in knowledge, knowledge would be
present always or never, so long as all these three are constantly present to
co-operate together. As this is not the case - as knowledge sometimes takes
place and sometime does not, in spite of the presence of these factors - we
are forced to admit the existence of a fourth variable factor, which is mind.

"We have seen that according to almost all non-advaitic thinkers, with the
exception of some Buddhists, mind is an internal sense-organ - an indriya. An
indriya is conceived as the karana or the instrumental cause of perception.
The Nyayikas and some others, who admit the existence of an abiding self as
the knower of all knowledge, consider mind to be the instrument which is
employed by the self, the agent, in internal perception."

p. 54: "Though the Vedantins admit the existence of a permanent self, which in
self-consciousness behaves as the knower or the subject, yet even among them
there are some who do not regard manas or antahkarana as an indriya or an
instrument of this subject. In the Upanisads the manas is mentioned in many
places as distinct from the indriyas."

p. 58: "As soon as the we realize the special character of internal
perception, and admit that here the mind can independently reach its object,
as the Nyayikas also must ultimately admit, the necessity of calling mind an
indriya disappears."

p. 61: "According to the Vedantins, in the perception of an external object
the mind (manas or antahkarana) goes out to the object through the indriyas,
the senses. The senses are thought of as themselves going out to reach
the object. The blind senses move, however, under the impulse and guidance of
antahkarana - rather the atman identified with antahkarana. The senses,
therefore, act as the vehicle of the movement of antahkarana towards the
object. On reaching the object the antahkarana is identified with it, just as
the water going out of a tank assumes the form of the tub on reaching it. But
the mere fact of antahkarana assuming the form of the object does not explain
the knowledge of the object. The vrtti or the mode of antahkarana is
illuminated by the atman, the cit, which is there as the observer (saksin) of
all changes. In all empirical experiences the antahkarana and the atman remain
identified or fused together, just as in a red-hot iron ball fire and iron
remain indistinguishably blended together

"Three points require critical consideration. (1) the going out of the
antahkarana to the object; (2) antahkarana's assuming the form of the object;
(3) the relation of the vrtti and its object with the self (atman)."

pp. 62-63: "Antahkarana or manas is regarded by almost all Indian thinkers as
a natural object, only of a subtler type. The Vedantins consider it to be also
finite in dimension; the movement of the mind is thus not to be regarded as
metaphysical, but actual. 

[Advaitins:] "even in the collection of raw material mind co-operates with 
senses."

p. 64: "Senses are senses in so far as they are the channels through which the
mind seeks or tends to know the external world in the process of perception.
The mental is not the terminus of a preceding physical process. The activity
of the mind is present from the very initial stages, though it may not always
be conscious of its activity."

p. 65: "If in perception mind acts prior to sense by pre-adjusting attention
to the object, it may yet be asked whether it is not sufficient for the mind
to attend to its object from within the organism without going out to the
object. When we perceive an external object our attention is directed not to
the physiological changes caused by the object within the organism, but to the
object itself, outside in space. And in fixing attention on an object mind
comes into direct contact with it. In consistency with [the general Western
view] our knowledge of the external world can at best be an inference."

p. 66: "The second point to be noted is the Vedantic statement that the mind
takes the form of the object on meeting it."

p. 68: "According to the Vedantins the antahkarana, which may be regarded as
an instrumental and an intermediate principle standing midway between the
object and the self, receives the perceived object as a whole by assuming its
form."

p. 69: "Images are commonly regarded as purely mental. But their objective
character and especially the physical characteristics of extendedness, etc.,
which some images possess, make it difficult to call them mental in the sense
in which thinking, feeling, willing, or even perceiving can be called mental."

p. 70: "Images can be classed neither with the purely psychical - the self or
the mental act - nor with the purely physical. We must place them midway
between the two. This intermediate region extends across the gulf that
otherwise seems to separate the mental from the physical - the self from the
not-self - and renders the whole of existence from the psychical to the
physical continuous. The antahkarana with all its modifications might be
considered as belonging to this region."

p. 79: "Three stages or grades of awareness of an object: 1) trying to imagine
a face without success, 2) successfully imagining it, and 3) perceiving it.

"In the second stage there is accommodation on the part of the subject to
receive the object, but still the object does not reveal itself. Other
conditions for the revelation of the object must also be fulfilled. These
conditions are the going out of the senses, and through them of the
antahkarana, to the object."

p. 82: "The predominant theories minimizing the facts of consciousness are
almost all inspired by theories of biology and evolution."

p. 90: "According to the Vedantins knowledge as knowledge or as mental act is
always immediate. Even an inference or a mere idea must be regarded as
immediate (pratyaksa) or self-evident as an act or mental process. Knowledge
thus considered would be immediate when its object is immediately known
through the going out of the antahkarana to the [object] through the senses
and its taking the form of the object establishing an identity between the
subject and the object."

pp. 106-107: "Time (kala) as perceived is not thought of as an empty form from
which all events have been abstracted, nor as the series of events present,
past, and future stretching from infinity to infinity. It is absurd to think
that time conceived in any one of these ways should be directly perceived
through the senses. The time that is said to be perceived is the present time
which subsists in the object as an element qualifying it."

pp. 107-108: "In the perception of sound the relation between the sense and
the object is said to be one of simple inherence (samavaya). Sound is a
quality inherent in the akasa and an ear is conceived as akasa in a limited
form. So that the relation between sound and the ear is the same as that
between sound and akasa, which is a relation of inherence. The universal
soundness (sabdatva) inhering in sound is perceived through an indirect
relation between the sense and the object, and this relation may be called
'inherence in the one inhering' (samaveta-samavaya)."

p. 109: "According to the Nyayikas the categories of realities are seven -
substance (dravya), attribute (guna), action (karma), universal (samanya),
particularity (visesa), inherence (samavaya), and non-existence (abhava). All
these are considered to be distinct realities. Attributes and actions have
their locus in particulars to which they bear the same relation of inherence.
There are two principal relations according to the Nyayikas: samyoga and
samavaya. Samyoga is the relation of the collocation or conjunction of two
terms that were previously unrelated or unconnected, e.g., the relation
between the table and the floor. Samavaya is the constitutive or inherent
relation that exists, for example, between the whole and its parts, between
attributes or actions and their substance, between a universal and its
particulars. Of these two relations, samyoga is considered by the Nyayikas to
be an attribute (guna) of the terms related; samavaya is not taken as an
attribute, but as an independent category by itself."

p. 111: "If the attributes and substance, universals and particulars, are
found to be inseparable in existence, there is no necessity whatever for the
conception of the foreign entity of a samavaya relation for holding them
together. On the contrary, it is reasonable to think that attributes,
universals, etc. are identical with the substance in which they are
experienced to reside. If a relation is as independent an entity as the terms
related, there arises the necessity of a second relation to connect the first
relation with each of the terms, and the second relation may similarly be
shown to require a third, a that a fourth, and so on ad infinitum. Thus the
logical analysis of the conception of a relation as a co-ordinate entity leads
us to infinite regress (anavastha)."

pp. 136-137: "The Vedantins hold that it is the nature of knowledge to be
self-manifest. It is unnecessary, therefore, either to infer its existence or
to perceive it in subsequent knowledge. The self-manifesting nature of
knowledge is regarded by the Vedantins to be a unique subjective
characteristic, which therefore cannot be classed under immediacy
(pratyaksatva) as pertaining to objects. As Madhvacarya puts it, 'Things like
pots are perceived (pratyaksa) as objects. But knowledge (pramiti), being
self-manifest (svaprakasa), cannot be regarded as an object of perception
(pratyaksa)."

p. 141: "Knowledge neither needs to be nor can be an object either of itself
or of any other knowledge. Self-manifestedness of knowledge means that
knowledge can behave as being immediate without being an object of knowledge.
This would be an exact rendering of the term svaprakasatva as given by the
Advaitins."

Comparison (upamana)

p. 158: "Admitting that dissimilarity is known through upamana, Ananta Krsna
Sastri raises a pertinent question, what is the use of this source of
knowledge for the Advaitin? He points out that it enables the Advaitins to
know the exact phenomenal nature of the world through the knowledge that the
world is dissimilar to the Absolute Reality, on the one hand, and to the
utterly unreal, on the other. For example, the Brahman, unknown to the novice,
can be understood analogically as being like the material cause of a perceived
object, or like the substratum of an illusory object. The Upanisads are full
of such uses of upamana - the attempts of the teachers to convey to the
disciples the idea of the supersensible in terms of its similarity to the
sensible."

Non-cognition (anupalabdhi)

p. 175: "There are, according to the author of the Vedanta-paribhasa, four
kind of non-existence that can be known through non-cognition (anupalabdhi).
They are called: pragabhava, pradhvamsabhava, atyantabhava, anyonyabhava.

"Pragabhava literally means previous non-existence. It is defined as the
non-existence of the effect in its material cause previous to its coming into
existence The previous non-existence of a particular object cannot be assigned
any beginning; but it is put an end to as soon as the object comes into
existence."

p. 176: "Pradhvamsabhava, or more simply dhvamsabhava, literally means
non-existence as represented by annihilation or destruction.

"Non-existence as annihilation has doubtlessly a beginning. Whether it has an
end is a disputed point."

pp. 180-181: "The third kind of non-existence, atyantabhava, literally means
total or absolute non-existence. If a particular thing does not exist in a
particular locus at any time, then it is said that there is atyantabhava or
absolute non-existence of that thing in that locus. For instance, there is
atyantabhava or total absence of color in air.

"Anyonya-bhava, the fourth kind of non-existence, is, according to the
Advaitins, nothing but difference or separateness (bheda or prthaktva) owing
to which we judge 'this is not that.' It also has beginning in time or not,
according as its locus has beginning or not. The difference of an individual
soul (jiva) from Brahman or the difference of the latter from the former is
beginningless because jiva and Brahman are beginningless."

p. 183: "According to [the Advaitins] the object of a non-perception
(anupalabdhi) can be one of the four kinds of non-existence. Ajnana being the
non-existence of knowledge should be known, like any other kinds of
non-existence, through non-cognition. But the Vivarana-prameya-samgraha does
not accept this apparent conclusion. According to it ajnana can be divided
into two classes, namely (1) absence of the knowledge of a particular object
(jnana-visesabhava) and (2) the general absence of knowledge (jnana- 
matrabhava). Of these two, the former alone can be known through non-
perception. But the latter is directly revealed to the witnessing self
(saksi-bhasya). The knowledge of the absence of the knowledge of a particular
object involves the knowledge (memory or some other kind) of the object at
that time, but the general absence of knowledge does not do so."

p. 185: "Does anupalabdhi mean only non-perception or does it mean non-
cognition in general?"

p. 187: "Though non-cognition is wider in meaning than non-perception and
though theoretically there can be as many kinds of non-cognition as of
cognition, the only kind of non-cognition that can directly yield any
knowledge about non-existence is non-perception. As a mean of knowledge,
therefore, non-cognition becomes practically equivalent to non-perception."

p. 198: "Like perception (or inference), non-perception yields a valid
judgment only under specific conditions. Like perception and inference,
therefore, it must have a canon by which its validity can be ascertained. This
canon is supplied by the tarka or hypothetical argument. Again as with
perception and inference, here the canon has not to be explicitly applied in
every case. It has to be used only in cases where there are positive reasons
for doubt or disbelief and where, therefore, necessity for ascertainment of
the truth is positively felt."

p. 199: "If the knowledge [of non-existence] is in any way doubted it is
reinstated with the help of the hypothetical argument or tarka. And even if
this validating argument be called an inference, non-perception does not cease
to be the source of the knowledge of non-existence. The content of the
knowledge (non-existence) is even then derived through non-perception, though
the validity of this knowledge has to be ascertained through a second source,
an inference."

Inference (anumana) - induction

p. 205: "The Vedanta-paribhasa defines vyapti as the coexistence of the major
term with the middle term, in all the loci in which the middle term may
exist."

p. 206: "The Nyayikas [say] that it is neither easy nor necessary for the
formation of a universal proposition to ascertain any relation of causality or
identity between two phenomena. An universal proposition can be based on the
determination of any invariable unconditional universal relation. The
Advaitins also hold the same view."

p. 207: "But the question may be asked: How often is the concomitance between
two phenomena to be observed before we can conclude that there is a relation
of vyapti between them?"

p. 208: "It should be noticed that the canon used by the Advaitins for
ascertaining the soundness of a vyapti or generalization is the method of
agreement in presence (anvaya) together with the non-observation of any
exception."

pp. 209-210: "Difference between the Nyayikas and the Advaitins as regards
the perceptual knowledge of a vyapti. The question is: How, on perceiving
a limited number of cases of smoke accompanied by fire, can we conclude - 'All
cases of smoke are cases of fire'?

"The views of the Nyayikas and the Advaitins on this controversy may be taken
as throwing some light on their conception of the universal proposition. The
Advaitins, it would appear, take a connotative view of the universal
proposition, whereas the Nyayikas seem to take a denotative or enumerative
view of it: 'all cases of smoke = all individuals of the class of smoke.' The
Advaitins say that for ascertaining a universal concomitance it is not
essential that observation should be repeated several times; even a single
observation can yield the knowledge of such concomitance, only if no exception
to it is known. They mean that under favorable circumstances a connection
between the two universals (smokeness and fireness) can be ascertained even by
a single observation, and this suffices for inference. [The Advaitins] say
that the universal proposition ['all cases of smoke are cases of fire'] is
really inferred deductively from the universal concomitance between smokeness
and fireness."

pp. 210-211: "The early Nyayikas as well as the Advaitins hold that the
knowledge of the vyapti is instrumental to inference. But they differ as to
the exact function (vyapara) that the knowledge of vyapti as an instrument
discharges.

"But the Advaitins differ from the Nyayikas as to the function (vyapara) of
the knowledge of vyapti. According to them, if one knows the vyapti, 'smoke is
accompanied by fire,' as soon as he afterwards sees smoke on the mountain the
impression of this previously acquired knowledge of vyapti is revived, which
at once yields the knowledge 'the mountain is fiery.' The function of the
knowledge of vyapti is therefore the creation of this impression (samskara),
through the revival of which inference takes place."

p. 216: "The Vedanta-paribhasa says that though the revival of past
impressions is a necessary condition of both memory and inference, there is
still a marked distinction between the two. Memory is caused by the revival of
past impressions only, but inference is caused not by that alone; it has other
causes [such as] the knowledge of the relation of the middle term with the
minor, etc.

"If the major premise expresses a relation between individuals (e.g.
individual smoke and individual fire), then the relation is not universal and,
therefore the conclusion does not follow; if it expresses a universal relation
(e.g. between all smokes and all fires), then what is already known is sought
to be proved."

p. 217: "Indian logicians generally make a distinction between inference for
one's own self (svarthanumana) and inference for others (pararthanumana),
inference used for demonstrating a truth to other persons. As to the second
kind, the Nyayikas are of opinion that it consists of the following five
steps: 1. Pratijna - statement of the proposition to be proved, the thesis to
be proved by an inference. 2. Hetu - statement of the reason. 3. Udaharana -
statement of the universal proposition along with an instance. [Hetu in
Potter's book] 4. Upanaya - statement of the presence of the mark in the case
in question. [Paksa] 5. Nigamana - conclusion proved. [Sadhya]

"But the Advaitins say that even for the purpose of demonstration the five
steps are unnecessary. It is quite sufficient either to state the first three
or the last three."

pp. 218-219: "In inference it is not always the case that the premises are
given and the conclusion is to be found. It is often the case that the
conclusion is presented first to the mind, and we are required to find the
premises that justify it. Whether we take the first three steps or the last
three, the udaharana that is common to both is preserved. And the udaharana is
the characteristic keystone of Indian syllogism; it consists of the universal
proposition supported by concrete examples. This marks off the Indian
syllogism from the mere formal and deductive of Aristotle on the one hand, and
the mere material and inductive syllogism of Mill on the other - a strong
point against the argument that the Indian syllogism had a Greek origin.

"For the Indian logician no syllogism was of any value unless based on an
universal major established through induction; consequently the processes of
induction and deduction blended together to constitute a syllogism."

Postulation (arthapatti) - deduction

p. 237: "When a known fact cannot be accounted for without another fact, we
have to assume or postulate the existence of that other fact. This process, in
which knowledge of the fact to be explained leads to the knowledge of the fact
that explains it, is called arthapatti - assumption, supposition, or
postulation of a fact.

"The instances of arthapatti have been divided into two classes:
drstarthapatti and srutarthapatti. The first literally means supposition of a
fact in order to explain perceived facts."

p. 238: "The second, srutarthapatti, literally means the assumption of a fact
in order to explain a fact known through testimony. It admits of two forms,
according as it concerns the supposition of a verbal expression (abhidhana) or
of a thing meant (abhihita)."

p. 245: "Inference and postulation - anumana and arthapatti - are two opposite
and yet complementary processes. The first tries to find out what follows from
given premises, whereas the second tries to find out what is presupposed by
the given and explains the given by showing its grounds."

Testimony (sabda)

p. 250: "Authority only changes forms from age to age. The modern man is in
the grip of authority in the new forms of the endless special sciences and
other exponents and specialists whose opinions are implicitly accepted,
because no inexpert can possibly understand or verify for himself the ever- 
increasing technicalities. Even some of the most critical and otherwise
skeptical philosophers, the logical positivists, behave toward the sciences
today just as the medieval ones used to do towards the scriptures. They regard
it as their chief, if not sole, duty to analyze, interpret and interrelate the
words of their authority, science."

pp. 252-253: "While an ordinary sound stands for itself, a symbolic sound
stands for some element of thought, and as such it bears on it a definite
impress of thought, which serves also to distinguish it from its fellows. In
fact our words are scarcely regarded as physical sounds; they pass for so many
phases of thought.

"The ancient Indians distinguished two kinds of sound, namely, the
inarticulate, indefinite sounds and the definite, articulate sounds of human
vocal organs. The former were called dhvani and the latter varna.

"Whether communication is held through spoken words or through writing, the
sensations received, whether auditory or visual, set up in the mind of the
hearer if the reader goes through a parallel process of repetition, in varying
degrees, under varying circumstances."

p. 256: "According to [the theory of sphota] the syllables of a word do not
directly present the meaning of the word, either separately or jointly, as
this process has been found to be inexplicable. In reality, however,
corresponding to every word perceived there is an unperceived, partless
(niravayava) symbol, which directly presents the meaning. This symbol is
called sphota or sabda - the word. The different syllables of a word only
serve to reveal this symbol to the mind."

pp. 257-259: sphota ð word-sound; meaning ð sphota

"When we hear the word, it cannot directly remind us of the meaning, because
neither through memory nor through perception can we grasp the whole word at
one moment, as the syllables come into consciousness one after another and not
simultaneously. But corresponding to the word we must have some unitary symbol
[sphota], in which there are no parts and consequently no question of 'before'
and 'after.' This symbol can be grasped as a single act of consciousness, as a
whole, which directly presents the meaning. The word only serves to arouse the
symbol into consciousness.

"According to the upholders of this theory sphota is ultimately one; it is in
fact the only reality and identical with Brahman. But like the Advaitins these
thinkers also admit that though reality is one from the transcendental point
of view, it is many from the empirical standpoint. They admit that though
sphota is ultimately one, it is not revealed in exactly the same form by every
word, but that different forms or aspects of the sphota are revealed by
different words.

"The sound that is heard and dies away is only a particular passing sensation
that rouses the symbol unheard, which is called the real sabda or sphota.
Thus, like an idea of Plato, sabda as sphota is both universal and eternal. It
is this that has a direct and eternal relation to meaning."

pp. 259-260: "This theory has been rejected by the Vedantins.

"Samkara, as well as his followers, holds that though a series cannot be
apprehended as a whole through perception we have, after all the members of
the series have been perceived one by one, a memory of all the members
combined together. We have in experience, in all cases of temporal and spatial
series, the knowledge of whole which preserve the internal relations obtaining
among their component members. When we already have such an experience, it is
idle to ask how it is possible. We need only admit, on the basis of such
experience, that the intellect possesses the power of synthesizing elements
which were originally apprehended at different moments of time. This function
of the intellect is called by Samkara samasta-pratyavamarsini buddhih
(intellect looking back on past experiences as a whole). The knowledge of the
unity and internal order of other series [not verbal] is possible without
sphota, in which case there is no reason why a verbal series also should not
be known in exactly the same way."

p. 261: "Samkara says that the word as a whole with its peculiar internal
order can be grasped in memory through the synthetic activity of the intellect
(samasta-pratyavamarsini buddhi), but he does not think it necessary to
explain how simultaneity and succession are compatible in the same act of
mind. The question how even in memory we can avoid remembering the syllables
in their order and therefore avoid failing also to grasp them as a whole, is
not further dealt with.

"Samkara seems to consider the testimony of consciousness regarding the
knowledge of a temporal series to be too strong to be shaken by any doubt, and
believes in the existence of a peculiar faculty of the understanding as being
the only explanation for the knowledge of such a series."

p. 264: "[Indian philosophers] distinguished between a symbol which is bodhaka
or merely indicative and a symbol which is vacaka or expressive or
connotative."

p. 265: "One of the important problems discussed almost by every school of
philosophy is: does a word primarily mean a particular (vyakti) or an
universal (jati)?"

pp. 267-268: "[Gautama and the Vedantins say] that the universal that is the
import of a word, and on the strength of which we apply a term, is not the
mere universal form or shape but the universal class character. It is not the
akrti [= generic form], but the jati which is the primary meaning of a term.

"The Nyayikas of the old school [say that] a word means all three - vyakti,
jati, and akrti. Only in some particular context some meaning is predominant
and the rest subordinate. [Footnote:] Vyakti = individual, particular; jati =
class characteristics; akrti = abiding component structure."

p. 270: "The Advaitins hold that the primary and explicit meaning symbolized
by a word is the universal alone and not the universalized particular. The
individual is the substantive (visista), which is qualified by the universal
as the adjective, and it is a well-known dictum that the knowledge of the
qualified (visista) presupposes the knowledge of the adjective (visesana) that
qualifies it. If so, then in it in virtue of the knowledge of the universal
connoted by the word that we can apply the word to the particular individual.
It follows therefore that the meaning that is logically primary is nothing but
the universal."

pp. 282-283: "The first use of the word 'John' is arbitrary, but not so the
first use of the word 'judge.' This rests on the philological assumption that
all connotative words are derived from roots, meanings of which are already
known, and the words have some pre-established connection with those meanings.
But this assumption is not at all easy to prove. When pushed to its logical
conclusion, the theory that all connotative words are radically significant
and not arbitrary must either fall back on the theory of the eternal and
internal relation between words and meanings, or be rejected. For if we can
show that at any time there was a single connotative term, which was used to
signify something, though there was no previous relation of the word with the
meaning signified, then the distinction between a connotative name and 
a proper name, on the basis of their first use, vanishes altogether.

"To this it will be replied that the logical connotation of a word is not
subjective, not even objective, but conventional."

p. 288: Proper names and abstract terms are also connotative. Therefore, all
words are connotative (i.e. all vacakas, as opposed to bodhakas).

p. 289: "In Indian logic the primary meaning of a word is called vacyartha or
sakhyartha or mukhyartha or even abhidheyartha. The secondary meaning of a
word is called laksyartha. The Indian rhetoricians however hold that in
addition to the primary and secondary meanings of a word there can be a
tertiary meaning as well, which they call vagyartha, and which can be roughly
rendered into 'suggested meaning.'

"The primary meaning of a word is known by adults mostly through memory. The
primary meaning, constituted as it is of all the common essential attributes
found in all the particulars, is not a simple thing. The secondary meaning is
generally obtained by selecting a part of the original meaning and rejecting
the rest."

pp. 290-291: "So long as the primary meaning of a word serves a particular
context, no recourse to a secondary meaning is at all necessary. It is only
when the primary meaning is found in any way incompatible that the necessity
arises for seeking a new meaning. The primary meaning of a word is known at
first either through some authority or through the inductive method, i.e.
inferring the meaning from the many contexts where the word occurs. But the
surest way of knowing the exact meaning of a word as fixed by usage is to
learn it from authority. When the meaning is once learned, in all subsequent
used of the word [it] is remembered. But in cases where doubt arises as to the
meaning of the word in a particular context memory fails as the only guide to
the knowledge of the meaning. The motive of the speaker or the writer, as
judged from that particular context or universe of discourse (prakarana), is a
second help to the ascertainment of meaning. The secondary meaning of a word,
therefore, is known through arthapatti, the data of which are the remembered
primary meaning and the inferred motive of the writer.

p. 291: "Indian rhetoricians pleaded for the recognition of a tertiary meaning
of words. Logicians, however, systematically rejected their contention, for
the simple reason that though suggested meaning possessed great literary
beauty and merit it was too vague, fleeting, and subjective to have any place
among logical meanings, the recognition of which is based purely on their
objective certainty and accuracy."

p. 292: "To logic, whose only appeal is to reason, accuracy and precision are
indispensable. So primary, direct and fixed meanings alone can be recognized
by logic. It is only a concession to the imperfection of current languages
that even secondary meanings are recognized. But a secondary meaning is
recognized because it can be fixed with tolerable certainty from the context."

p. 293: "[Summary:] A word is the symbol of thought. As a symbol, it negates
its private existence and content for the meaning it symbolizes. The meaning
is neither a particular psychical state nor the imagery called forth by the
word; it is universal. A word means a particular not primarily but
secondarily. In addition to its primary meaning, a word can have a secondary
meaning also."

p. 308: "Two ideas can combine when there is mutual affinity between them. But
this mutual affinity consists of two factors. First, there must be a want, a
feeling of incompleteness - a fellow-seeking - on the part of each constituent
member, and secondly there must also be a potency and compatibility on the
part of its fellow to satisfy its want. The first has been called akamksa
(hankering) and the second, yogyata (fitness)."

p. 310: "Akamksa is only one of the conditions of combination of meanings.
Another condition is yogyata or compatibility. Every word can raise an
expectancy and a desire for the completion of thought, so that the meaning of
the word requires something else, in combination with which it may become more
complete to thought. This fellow-seeking of a meaning is not and cannot be
satisfied by any and every other fellow."

p. 312: "The author of the Vedanta-paribhasa defines compatibility as
non-contradiction of the relation desired to be set up in a combination of
ideas (yogyata ca tatparyya-visaya-samsargabadhah). We have considered so far
only relational combinations and not non-relational ones. But there are cases,
though very rare, where the motive for placing two ideas together is not to
establish a relation at all, but is on the contrary to show the want of
relation or a mere identity. The author of Advaita-siddhi defines yogyata,
therefore, as merely the non-contradiction of the desired object of
combination (yogyatapi tatparyya-visayabadha eva) and omits the word
'relation' (samsarga).

pp. 312-313: "We have discussed hitherto only two condition, akamksa and
yogyata, under which sabda-bodha, or the knowledge of the meaning of a
combination of words, can arise. But in addition to these two material
conditions there is also a third, a formal condition, which must be fulfilled.
It is asatti or contiguity or proximity (which will be spatial in the case of
written words and temporal in the case of spoken words) between the two words
presented for combination. It is not sufficient that the meaning of two words
should be individually incomplete and should possess also the potency for
satisfying their mutual wants, but they should also be presented in close
proximity, just to show that it is these two meanings that are also meant to
be combined or construed together. Non-fulfillment of this condition may give
rise to doubt as to which words are meant to be construed together."

p. 313: "When these three conditions are satisfied, the meaning of a sentence
can be known. But even this is not all. So long as a sentence is considered by
itself, these three conditions are indeed sufficient. But a sentence is not
really an isolated, abstract entity. It is organically related to a speaker, a
context or a universe of discourse which determine, out of the many possible
meanings of a sentence, the particular meaning relevant to a particular case.
It is also necessary therefore to know the intention of the speaker or the
drift of the context (where the speaker is not known). So a fourth condition,
namely tatparyya-jnana, or the knowledge of what is intended or relevant, must
be taken into consideration. The actual meaning can be gathered from the
universe of discourse, the introduction, the conclusion, etc."

p. 314: "We see, therefore, that the knowledge of the meaning of a sentence or
a combination of words arises under the four conditions: akamksa, yogyata,
asatti, tatparyya-jnana. Sabda-bodha is therefore a peculiar kind of 
knowledge. It can, of course, be brought under the general category of
construction under which are included memory-synthesis on the one hand and
inference on the other. But the special combinations of sabda-bodha mentioned
above distinguish it even as construction from both memory-synthesis and
inference, in which these conditions are absent."

p. 330: "[Summary:] A vakya represents a significant combination of isolated
concepts as symbolized by individual words. The meaning of a sentence is
derived from a constructive combination of ideas - a process that takes place
under peculiar and specific conditions, which distinguish the knowledge of the
meaning of a sentence from ordinary types of knowledge, e.g. perception,
memory, memory-synthesis, and inference. This peculiar kind of knowledge is
called sabda-bodha. Vakyas are of two kinds, samsargavagahi and akhandarthaka,
according as they express some relation between two contents or express a
unitary unrelated content."

pp. 331-332: "We said that a vakya or a sentence, on being spoken, generates
in the mind of the hearer a constructive knowledge of the combination of ideas
or meanings, which is called sabda-bodha. But a vakya is not conceived as
sending its function here. Unless positively obstructed by some deterrent
factor, a vakya asserting a fact produces belief in the fact which forms its
objective intention (tatparya). A vakya comes, therefore, to be a source of
knowledge about facts. Consequently, sabda, as vakya, is regarded as a pramana
or method of knowledge.

p. 332: "The relation between sabda-bodha and sabda-prama, i.e. between the
knowledge of the meaning of a sentence and the knowledge of facts as derived
from a sentence, has to be clearly understood. Knowledge of meaning is an
intermediate process that is indispensable to the knowledge of facts. But the
former is organic to the latter and cannot be said to have any independent
existence."

p. 333: "We discussed the four conditions under which the knowledge of the
meaning of a proposition (i.e. sabda-bodha) takes place. Of these four
conditions, tatparya-jnana (knowledge of the speaker's object in speaking) and
yogyatva (compatibility) have also their objective aspects, which determine
the mental attitude of the hearer to the proposition."

p. 334: "Compatibility is understood in its objective aspect to be
compatibility with facts.

"We always implicitly believe in the truth of a statement made by someone, if
there be no positive ground for doubting or disbelieving him. We come thus to
the Vedantic theory of sabda-pramana, namely, that a vakya or sentence whose
import (subjective or objective) is not contradicted in any other way is a
valid source of knowledge. Thus belief in the truth of one's words is primary
and direct. Doubt or disbelief is secondary and indirect; it is otherwise
caused by some positive hindrance to belief."

p. 335: "According to the Mimamsakas the Vedas teach ritual duties; they are a
body of commandments [they] attach primary importance to the imperative of
mandatory sentences (vidhi-vakyas) in the Vedas. All other Vedic sentences
which state facts and describe the nature of realities (siddharthas) are
regarded as being only of instrumental value, their sole purpose being to help
and encourage the performance of rituals. For this school the question of
objective reference has little value. For the knowledge of a commandment it is
enough to understand the meaning of the sentence containing it.

"The Vedantins (and Advaitins in particular) attach greater importance to the
Vedas as containing the sentences that assert or declare the nature of the
world, self, God, etc. The problem of objective reference has great
importance, therefore, for these thinkers. They are interested in the Vedas
primarily as the source of the knowledge of truths. Though they also recognize
the subsidiary utility of the mandatory texts in so far as they lay down the
path of action that purifies the heart and paves thereby the way to the
knowledge of truths."

pp. 337-338: "The following three problems are involved in the consideration
of a method of knowledge: 1. Whether the alleged method is a source for the
attainment of knowledge of facts. 2. Whether the validity of the knowledge
attained by the alleged method is constituted by the very conditions which
make the knowledge itself possible, or whether it is constituted by any
external condition. 3. Whether the validity of the knowledge is also known or
ascertained by the conditions that constitute that knowledge, or by any
external condition.

"In Indian epistemology these three problems were dealt with separately as the
problem of pramana (the source or cause of true knowledge), that of
pramanyasya utpattih (the genesis or the objective constitution of validity)
and that of pramanya-grahah or pramanyasya jnaptih (the knowledge or
ascertainment of validity).

"Those who think that verbal testimony cannot be considered a method of
knowledge because the validity of such knowledge has to be ascertained through
inference and, therefore, this method becomes reduced to inference itself,
confuse the problems (1) and (3).

"Any source which gives us information about facts can be and should be
considered a method of knowledge, irrespective of the question whether its
validity is ascertained intrinsically or by some external method."

p. 340: "According to [the Mimamsakas and the Advaitins] even the validity of
verbal knowledge is constituted by and also known or ascertained through the
intrinsic conditions of verbal knowledge itself."

p. 341: "According to the Nyayikas the fact that knowledge is sometimes false
and the existence of doubt prove that every source of knowledge is considered
valid and is known as such only on the basis of an external condition, other
than that very same source of knowledge. In other words, the validity of a
certain source of knowledge and the ascertainment of that validity must be
done by means of a different source of knowledge.

"Pramanya-paratastva-vada holds that knowledge is both made true and known to
be true by special conditions, which are external to those that condition
knowledge itself. As against this line of thought the Mimamsakas and the
Advaitins hold a theory called pramanya-svatastva-vada, which means that
validity is conditioned by conditions intrinsic to knowledge itself and that
validity is known also from the conditions of knowledge itself.

"Knowledge, whether perceptual, inferential, or verbal, is essentially a
process directed to the attainment of truth. Thus truth is an intrinsic
characteristic of knowledge. If knowledge sometimes fails to attain truth, it
is because some special hindrance stands in the way of knowledge itself."

pp. 342-343: "Falsity therefore is due to or constituted by some special
conditions that are external to (and interfere with) the conditions which make
knowledge possible, and the fulfillment of which alone would otherwise make
for truth. So it is falsity that is externally conditioned, whereas validity
is conditioned by the conditions of knowledge itself.

"The difference between the two opposing views points to a fundamental
difference in the conception of the meaning and function of knowledge.
According to the former view the differentia of knowledge would be belief; and
as belief may be both true and false, knowledge also is neutral to truth and
falsity. [But] the very aim and purpose of knowledge is truth. Belief
essentially means 'believing to be true'. The Advaitins and the Mimamsakas
emphasize the fact that truth is intrinsic to knowledge while falsity is
externally conditioned.

"[Regarding the knowledge or ascertainment of validity] as against the view
that validity is ascertained through inference, the Mimamsakas and the
Vedantins argue that this view leads to an infinite regress. For if the
validity of knowledge is ascertained through another knowledge, the validity
of the second knowledge also being not self-evident has to be ascertained
through a third knowledge and so on ad infinitum. We are forced, therefore, at
some stage or other to admit that knowledge must also certify its own
validity."

p. 343: "It may be asked, however, if that be so, how could doubt at all
arise, and why is argument at all required for certifying the validity of
knowledge? To this it is replied that doubt arises only when there is positive
knowledge of conditions or facts which seem to contradict the knowledge we
have; and arguments are necessary not for the positive work of establishing
the validity of knowledge, but for the negative work of removal of doubt or
the chances of contradiction. It is the falsity and the doubtfulness of
knowledge that are inferred (from the presence of conflicting facts) and not
its validity, which in inherent in the knowledge. Non-contradiction is the
only guarantee of validity, and this guarantee does not forsake knowledge
until it is positively contradicted or doubted. [Self-evidence of the validity
of knowledge.]"

p. 344: "According to the theory of systematic coherence as held by Joachim,
the suggestion of absolute truth and absolute certainty for our fragmentary
knowledge is altogether out of the question. The process of the ascertainment
of truth cannot stop until the whole system - the Absolute itself - is known.
Ascertainment of validity cannot be said to be logically complete at any
stage."

p. 345: "Knowledge of validity takes place simultaneously with the act of
knowledge itself, which is implicitly believed. And if no amount of external
certification is logically complete, we have to treat and accept any kind of
knowledge as true if it is not as yet doubted or falsified. Thus it would
appear that non-contradiction has ultimately to be resorted to as the
guarantee for a judgment of validity. The positive and primary guarantee of
the truth of the knowledge of a thing is the knowledge itself."

pp. 346-347: "Sooner or later knowledge itself should vouch for its own truth,
make its validity self-evident, or there is no escape from relativity, no stop
to infinite regress. Falsification is not peculiar to self-evidence alone; it
is the common danger to which every other method is liable. Besides, if
knowledge ceases to be self-evident, it so far ceases to be knowledge itself.
And it is not the case that when self-evidence ceases to certify knowledge,
knowledge is left behind to be certified by some other method. When there is
positive knowledge of causes which make a particular belief false or doubtful,
what we are aware of is not merely that the knowledge is not self-evident, but
that we had no knowledge at all. Doubt or positive contradiction is not
leveled directly against self-evidence, but against knowledge itself. When ten
different arguments are adduced to ratify a particular knowledge, what is
achieved is not directly the ascertainment of validity (which is congenial
with knowledge itself), but the removal of ten actual or possible sources of
doubt or contradiction which stand in the way of knowledge itself."

p. 347: "The only consistent procedure for the real believer in relativism
would be complete silence. The judgement that would establish relativism would
be self-contradictory. Unless self-evidence of some sort is resorted to,
arguing, and still more so philosophizing, has to be abandoned.

"The conditions of knowledge will be different in different cases [for
different sources of knowledge]. When the conditions [for verbal knowledge]
are satisfied we have not only knowledge from some verbal statement, but also
belief in the validity of that knowledge."

p. 348: "The fact that knowledge derived by a particular method may sometimes
come into conflict with some other knowledge and even definitely turn out to
be false, is no reason why the method should be discarded forever. If this
principle were consistently followed, neither perception nor inference could
stand as a method."

p. 350: "The final object of the Advaita theory of knowledge is to show with
the help of the different methods (pramanas) that the world of multiplicity
cannot be consistently explained without admitting one underlying unity which
transcends all changes and diversities and which, therefore, is the highest
Reality. The Vedas including the Upanisads contain the great declarations of
unity - called the maha-vakyas (great sayings) - which are the reports of the
direct realization of unity by the seers (rsis). He who desires to liberate
himself from the delusive appearances of multiplicity and realize the One
Brahman behind all can utilize the teachings of the seers. These are, to start
with, mediate (paroksa) knowledge for him, but by constant meditation they can
be turned into immediate (aparoksa) knowledge.