Knowledge:

Framework and
Classification
R. Venkata Raghavan
Assistant Professor, Centre for Indian Knowledge Systems,
• There is nothing as purifying as knowledge.

Bhagavad Gita (4.38)

- Mahadevan et al. (2022)
Structure of the lecture
• Introduction to the Knowledge Triangle
• Brief introduction to ‘objects of knowledge’
• Introduction to ‘methods of knowledge’
Establishing New Knowledge in Indian Knowledge System

कानादं पाणिनीयं च सर्वशास्त्रोपकारकम् ।

Kānādam Pāninīyam ca sarvasastropakārakam.
Logic and grammar are indispensable aids for every branch of knowledge.

Nyāya (one of the Nine Schools of Indian Philosophy) focuses on how one can
systematically inquire into a problem and establish new knowledge in a
structured process.
5
The Questions
• India has produced enormous body of knowledge (both
documented and undocumented) over several centuries

HOW ?
The Context
• How is knowledge produced in an oral culture ?
• Discussion/Debate
• Vāda:
• A discussion in which there are two hypotheses
• Each is supported or refuted by suitable pramāṇa
• Not opposed to siddhānta
• And in which the reasoning is systematic
• Jalpa:
• A vāda which can be won even by rhetorical means
• Vitaṇḍa
• A debate just to refute the hypothesis
The Context
• Why do we seek knowledge ?
• Prayojana (purpose)
• But what is stopping us from fulfilling that purpose ?
• Saṃśaya (doubt)
• What sort of knowledge are we seeking ?
• Siddhānta (Theory)
• What do you do after this?
• Nirṇaya (Conclusion)
The Context: The Knowledge Triangle

Pramāṇa
(means of enquiry)

Prameya Pramātṛ
(Object of enquiry) (Subject of enquiry)

Pramā ?
 Prameya
6 existent categories (bhava padārthas)
• Dravya (substance)
• Guṇa (attributes)
• Karma (action)
Prameya • Sāmānya (universal)
(Object of enquiry)
• Viśeṣa (particularity)
• Samavāya (inherence)
1 non-existence (abhāva padārtha)
• Abhāva (absence)
 Pramā: Its Nature
Buddhi/Jñānam
(Cognition)
Anubhava Smṛti
(Occurrent) (Non-Occurrent/
Dispositional)
Yathārtha Ayathārtha Yathārtha Ayathārtha
(True) (False) (True) (False)

Pramā Apramā
Pramā: Its Nature
Anubhava
(Occurrent)
Yathārtha Ayathārtha
(True) (False)

Prama Apramā

1. Pratyakṣa 4. Śābda 1. Saṃśaya

2. Anumana 5. Arthāpatti 2. Viparyaya
3. Upamana 6. Anupalabdhi 3. Tarka
Seeker of Knowledge (Pramathru प्रमातृ)

• The knowledge seeker gets mentally involved in the process and

commits physical and other resources to the process.

• Once the knowledge seeker obtains the implicit knowledge, he/she

may make it available to others by way of tacit knowledge.

• The tacit knowledge is eventually transmitted systematically in the

form of explicit knowledge by proposing a new theory, framework or
literary work.

19
Objects of knowledge (Prameya, प्रमेय) - Objects of enquiry

• Objects or entities become the context for seeking knowledge.

• e.g. A scientist developing a new vaccine will direct his/her entire
focus in the knowledge creation process on the virus and all the
aspects related to it.
• The term object includes all that towards which effort is directed to
obtain the knowledge; it may include physical, meta-physical, virtual,
and other entities.
Means of obtaining the knowledge (Pramāna, प्रमाण)
Saṃśaya (Doubt, Ambiguities in existing knowledge)
Ambiguity in knowledge occurs on account of five reasons as
follows:
When an observed phenomenon resembles several
commonly known properties, there is ambiguity in
understanding the phenomenon.
• e.g. A new disease caused by a virus may resemble several
known conditions in terms of symptoms.
Recognition of properties that have no relation to anything
known so far.
• e.g. A new chemical process to extract some element may
provide results which may not make sense at the outset.
Saṃśaya Contd.
Conflicting findings from a study.
Another variation of conflicting results is considerable
variations in the results.
• e.g. inadequate sample to identify underlying
patterns.
Total absence of our understanding of the problem
itself.
• e.g. initial stages of exploratory research in new
domains.
Causes of Ambiguities in Knowledge
 Pramā: Its Means (Pramāṇa)
• Pratyakṣa
• Contact of senses with their corresponding object
• Most common means of cognising
• Could be ordinary or extra-ordinary
• Single step/Multi step
• Phenomenological theory rather than physical
• Key feature: No mediation of any other pramāṇa
Pramā: Its Means (Pramāṇa)
• Śabda
• Testimony of reliable person or source (e.g. Veda)
• Peculiar to Indian traditions
Pramā: Its Means (Pramāṇa)
• Upamāna
• Comparison or Analogy
• Based on similarity (sādṛśya upamā)
• Based on dissimilarity (vaidharmya upamā)
• Based on unique features (asādhāraṇa dharma
upamā)
Upamāna

• “upamāna” is derived from “upa” meaning “similarity” and “māna”

meaning “cognition”;
• Upamāna is “knowledge by similarity.”
• Upamāna gives us knowledge of something based on similarity with
an object which was already known. So, upamāna is the instrument of
resulting knowledge (upamiti).
 The process of the knowledge by similarity
• The process of the knowledge by similarity consists of the following steps:
• Step1: I have not seen the animal known as gavaya before;
• Step 2: I have heard that there is such an animal and that animal looks like
a cow; and
• Step 3: I have already seen a cow and known what a cow looks like.
• When these conditions are met, and when I see an animal that I have
never seen before, I notice the similarity of this animal with a cow; I then
recollect that a gavaya resembles a cow and arrive at the judgment “this is
a gavaya.”
• Knowledge by comparison always assumes the form “as…so,” which is
used to express similarity, e.g., “as the cow, so the gavaya.
• Inference (Anumānā) never assumes a similar form, e.g., “as the
smoke, so the fire”; it is not inferential knowledge.
• Therefore, upamāna is accorded the status of an independent
pramāṇa.
Pramā: Its Means (Pramāṇa)
• Arthāpatti
• When two cognitive episodes conflict, a third must be
posited to remove the conflict
• Inference to best explanation

• Anupalabdhi
• Unique pramāṇa (accepted only by Bhāṭṭa) to cognise
negation
 Arthāpatti (Arthāpatti अर्थापत्ति means postulation, derivation from circumstances.)

• Our Sastras give a clear idea of arthapatti through an illustration. "Pino

Devadatto diva na bhunkte". What does the statement mean?
• "The fat Devadatta does not eat during daytime".
• Though Devadatta does not eat during daytime, he still remains a fat fellow.
How?
• We guess that he must be eating at night.
• There is something contradictory about an individual not eating and still not
being thin.
• Here, Arthapatti helps us to discover the cause of Devadatta being fat.
• Our guess that he eats at night does not belong to the category of Anumana.
• To make an inference, there must be a hint or clue in the original statement
itself. There must be a "linga" like smoke from fire, thunder from clouds. Here,
there is no such linga.
Anupalabdhi (Anupalabdhi अनुपलब्धि means non-perception,
negative/cognitive proof)

• Anupalabdhi is the means by which we come to know a non-existent

object.
• Anupalabdhi is the means by which we know abhava.
• Suppose someone tells us, "Go and see if the elephant is in the
stable".
• We go to the stable to see for ourselves whether or not the elephant
is there. We find that there is no elephant in the stable: to recognise
such absence (non-existence) is anupalabdhi.
Basically, arthāpatti is the kind of reasoning that allows us to reason
from
(1) “Fat Devadatta does not eat during the day.”
to
Fat Devadatta eats at night.
(2) Caitra is alive; I see that Caitra is not at home.
to
Caitra is living somewhere outside of his home.
• The central idea is that there is some kind of tension which arises
from a new fact being cognized, against the background of another
fact, which cannot be resolved without the postulation of third fact.
• In example (1) above, the tension is between Devadatta being fat and
his not eating during the day.
• In (2), it is between Caitra’s being alive and Caitra’s not being at
home.
• The tension is not strict logical inconsistency, but something else.
 Pramā: Its Means (Pramāṇa)
• Anumāna (Inductive and Deductive Reasoning)
• Cognition that follows (anu) a previous cognition (māna)
• Inferring something based on its invariable connection
with what is previously known
• Four key elements:
• Hetu (the sign/reason, what is known)
• Sādhya (the conclusion, what is inferred)
• Vyāpti (the invariable relation between the two)
• Pakṣa (the locus of hetu and sādhya)
 Pramā: Its Means (Pramāṇa)
• Anumāna
Forced has been imparted to this ball, because it is moving.
Pakṣa: Ball
Hetu: Motion
Sādhya: Force
Vyāpti: Wherever there is motion, there is force
(Or, wherever there is no force, there is no motion)
Example: Like in a cart
 Pramā: Its Means (Pramāṇa)
• Anumāna
Importance of Example ?

Greek Logic Anumāna

All men are pink. All men are pink, like ?
Raghavan is a man. Raghavan is a man.
Therefore, Raghavan is pink. Therefore…
Pramā: Its Means (Pramāṇa)
• Anumāna
Standard Form
Pratijñā X has Y Ball has force imparted Hypothesis (what we want to
to it. prove)
Hetu Because of Z Because of possessing Main reason/premises
motion.
Udāharaṇa Wherever Whatever possesses Establishing sufficient
there is Z there motion has force relationship between my
is Y, like in K imparted to it, like a reason/premises and my
cart predicate + Prior studies

Upanaya This X is like This ball is like that cart Establishing that the current case
that K comparable to prior studies
Nigamana Hence, X has Y Hence, it has force Conclusion
imparted to it
Anumāna

• It is the knowledge that arises after (anu) another knowledge.

• Anumāna is measurement after something; it is that cognition which
presupposes some other cognition. It is mediate and indirect; it arises
through the knowledge of the mark or liṅga.
• Consider the case of seeing smoke on a distant hill. Upon seeing the
smoke on a hill, one infers that there is fire on the hill. In this case, the
smoke serves as the mark of the fire.
• Five-membered structure
• 1 there is fire on the hill (the proposition to be proved or pratijñā)
• 2 because there is smoke (states the reason or hetu)
• 3 wherever there is smoke, there is fire (vyāpti)
• 4 as in the case of the kitchen (example or drṣṭānta)
• 5 there is fire on the hill (conclusion or nigamana)
 Pramā: Its Means (Pramāṇa)
• Anumāna
• Types
• Pūrvavat: Inferring the effect based on the cause OR inferring
something based on previous experience of their co-occurrence
 one sees dark clouds and infers that rain will follow.

• Seśavat: Inferring based on a sample OR Inferring cause from effect

OR Inferring based on elimination
 one tastes a little water in the sea and infers that the entire sea
water is salty.
Sāmānyato-dṛsṭam
• Sāmānyato-dṛsṭam: Inferring particular from general OR any form of
non-causal reasoning
 Seeing the sun in the eastern horizon in the morning and the sun
in the western horizon in the evening, one infers the movement
of the sun from the east to the west.
 On seeing some mango trees blooming, one infers all mango
trees to be blooming.
Hetvābhāsa

• The existence of a hetvābhāsa prevents an inference from taking

place.
• A faulty hetu directly obstructs inferential knowledge. (Examples 1
and 2)
• A hetvâbhàsa indirectly obstructs inferential cognition (Examples 3, 4,
and 5)
• Example 1
• "Pain is delightful, because it is a state of mind.
• This is an example of inference which involves contradictory hetu and
this variety of faulty hetu is termed as bādhita.
Example 1:"Pain is delightful because it is a state of mind"

'pain' is the paksa, 'delightfulness' is the sâdhya, and 'being a state

of mind' the hetu .
If the opposite of sâdhya 'delightfulness' is found as a matter of
fact to exist in the paksa, i.e., 'pain‘, it is not possible to assert of
'pain' from the sâdhya 'delightfulness'.
The conclusion is invalidated by internal perception.
Hetu here fails to function and this particular failure arises
because some thing opposed to sâdhya is known to be present in
the paksa.
This particular kind of defect directly impedes the inferential
knowledge, namely, the judgment that the paksa (pain) possesses
the sâdhya.
Example 2: "Objects are of the nature of pleasure, because they produce
pleasure," and "Objects are of the nature of pain, because they produce
pain."
• The inference here is fallacious, because there is another hetu, namely,
'the quality of producing pain' which proves the opposite sâdhya, i.e.,
'of the nature of pain' in another inference of the form "Objects are of
the nature of pain because they produce pain." As neither of the two
rival hetus is better than the other, the first inferential cognition (i.e.,
"Objects are of the nature of pleasure") would be as much undesirable
as the second ("Objects are of the nature of pain because they produce
pain").
• Neither of the two hetus perform its function and this failure occurs
because they lead to contradictory conclusions; both inferences are
invalid.
• Here, the process of inference by hetu is impeded by the knowledge of
the presence of the counter hetu.
Example 3(a): "Fire is cold because it is a kind of tejas."

• This inference is fallacious because the hetu instead of establishing

the sâdhya, establishes its opposite.
• The vyâptijnâna needed for the validation of inference is "Wherever
there is a kind of tejas there is cold." The hetu of this sort has been
termed viruddha or hostile.
• This sort of hetvâbhàsa interferes with the emergence of the
knowledge of vyâpti (concomitance) between the sâdhya and the
hetu.
• Example 3(b): Air is heavy because it is empty.
• The hetu is viruddha or hostile.
Example 4 : "The hare's horn is hard because it is a horn.”

• This type of asiddha hetu is known as Sādhyasama.

• Here, since 'hare's horn' does not exist, there is no real subject.
• In this context, 'hare's horn' means 'the horn of the hare' and its
differential characteristic will be 'attached to the hare.'
• Since the characteristic is not real, the paksa which is said to be
characterized by it is equally unreal.
• When I say that my oval table has a candle on it, the subject of my
statement must be understood to possess not only tableness but also
oval shape.
• A quality such as oval shape, which specifies the subject of an
inferential judgment, is called 'the determining character of
subjecthood' .
• If the determining character of subjecthood is absent from the
subject under consideration, this would amount to the non-existence
of the subject as subject.
Example 5: "The hill has smoke because it has fire."

• The inference is fallacious because the relationship between the hetu

and the sâdhya is not invariable.
• That is, though the hetu fire is present wherever smoke (sâdhya) is
present, it is also present where smoke is absent.
• In other words, the hetu fire is present both in a similar and in a
contrary instance;
• There is fire in the hill which possesses smoke (a similar instance); fire
is also present in a red-hot iron, since it lacks smoke (a contrary
instance).
• Hence, this fallacy is designated ‘common (sâdhârana) hetu’ in the
sense that it is common to both a similar and a contrary instance.
 Pramā: Its Means (Pramāṇa)
• Anumāna
• Potential Fallacies
• Savyabhicāra: Erratic conclusions (Anakantika).
 When the hetu co-exists both with sādhya as well as
its absence OR it occurs only in one pakṣa
• Viruddha: Contradictory conclusions.
 When the hetu contradicts what is seeks to prove
• Sādhyasama: Unproven proposition. (Asiddha hetu)
 When the hetu or pakṣa or vyāti itself is as much in
need of proof as the sādhya (Presumption)
• Prakaraṇasama: Circular reasoning (Satpartipaksha)
 Using sādhya itself in the proof.
• Kālātįta: Mis-timed or Contextually irrelevant findings.
 The findings are not ‘in sync’ with the time or with
the context.
Prakaraṇasama: Circular reasoning (Satpartipaksha)

• Description: A type of reasoning in which the proposition is

supported by the premises, which is supported by the proposition,
creating a circle in reasoning where no useful information is being
shared. This fallacy is often quite humorous.
• Logical Form:
• X is true because of Y.
• Y is true because of X.
Five-types of Pseudo-reasoning

• Since there can be fire without smoke (as in a red-hot iron ring), if somebody wants to infer
presence of smoke in the kitchen on the basis of the presence of fire there, his evidence
would be pseudo-evidence called the "deviating."
• Where the evidence (say a pool of water) is usually the sign for the absence of fire, rather
than its presence, it is called the contradictory.
• An evidence-reason must itself be established or proven to exist, if it has to establish
something else. Hence, an "unestablished" evidence-reason is a pseudo-evidence or a
pseudo-sign.
• A purported evidence-reason may be countered by a purported counter-evidence showing
the opposite possibility. This will be a case of the "counter-balanced.“
• An "untimely" is one where the thesis itself precludes the possibility of adducing some sign
as being the evidence-reason by virtue of its incompatibility with the thesis in question. The
"untimely" is so-called because as soon as the thesis is stated, the evidence will no longer be
an evidence.
 Pramā: Its Means (Pramāṇa)
• Rhetorical Tricks
• Jāti: Refuting the argument based on superficial analogy
A: “Soul is a substance (dravya), because it has
properties, just like a pot”
B: “Ok. Then like a pot is it also made of clay ?”
 Pramā: Its Means (Pramāṇa)
• Nigrahasthāna
• ‘Defeat situation’ or ‘Clincher’
• 22 listed by Nyāya sutra
• Some are logical: Like contradicting one’s thesis,
fallacious reasoning
• Others are procedural: Not using all the 5 steps, Using
more steps, Repeating one self, Being silent, Not calling
our rhetorical tricks, engaging in whataboutary.
 Knowledge via
Science
Induction Deduction

(Cooper and Schindler, 2009)

• Induction occurs when we observe a fact and ask
“Why is this?”
• In answer to this question, we make a tentative
explanation (i.e., formulate a hypothesis or develop
a model).
• Deduction is the process by which we test whether
the hypothesis/model is capable of explaining the
fact.
57
 Deduction
• Deduction involves techniques that depend upon the
nature of the hypothesis/model.
• It may involve
hypothesis testing
solving equations
running a computer program
expressing a sequence of logical statements
whatever necessary to solve the problem of
interest relative to the model.

58
The Scientific Method (Ravindran et al. 2007)

59
 Example
(Cooper and Schindler, 2009)

• Suppose a company spends 1 crore on a regional

promotional campaign and sales do not increase.
• This is a fact – sales did not increase during or after the
promotional campaign.
• The question is “Why did not sales increase?”
• One answer to this question is a conclusion that the
promotional campaign was poorly executed.
• The conclusion is only a hypothesis, i.e., one explanation.

60
• The following hypotheses may also explain why sales did
not increase:
• Regional retailers did not have sufficient stock to fill
customer requests during the promotional period.
• A strike by the employees of the trucking firm
prevented stock from arriving in time for the
promotion to be effective.
• A cyclonic rain closed all the retail locations in the
region for 10 days during the promotion.

61
61
• We promote a product but, sales do not increase. (Fact
1)
• We ask the question “Why did not sales increase?”
(Induction)
• We infer a conclusion (hypothesis) to answer the
question: The promotion was poorly executed.
(Hypothesis)
• We use this hypothesis to conclude (deduce) that sales
will not increase during a poorly executed promotion.
We know from experience that ineffective promotion
will not increase sales. (Deduction 1)

62
62
• To test a hypothesis, we must be able to deduce from it
other facts that can then be investigated.
• This is what research is all about. We must deduce
other specific facts or events from the hypothesis and
then gather information to see if the deductions are
true.
In this example,
• We deduce that a well-executed promotion will result
in increased sales. (Deduction 2)
• We run an effective promotion, and sales increases.
(Fact 2)

63
63
 Why Did Not Sales Increase?

Induction
Fact 1:
We promote a product Why?
Ded
but sales do not increase.
pro uction
m :
sale otion Ineffe
s. wil c
l no t i ve
t in
c re a
se
Hypothesis:
The promotion was
poorly executed.
Fact 2: t i o n: E ffective
Dedu c
i n c r e a se sales
We run an effective will
promotion
promotion and sales
increases.

64
Steps in research problem formulation

Broad areas

Dissect into subareas

Select the most interesting topic

Raise research questions

Formulate objectives
આભાર‫شکریہ‬న్యవాదాలు
ধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ நன்றிధన్ దా లు ಧನ್ಯವಾದനന്ദി धन्यवाद
આભાર‫ شکریہ‬ধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ நன்றி ధన్ దా లు న్యవాదాలు ಧನ್ಯವಾದ
നന്ദിઆભાર‫ شکریہ‬ধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ நன்றிధన్ దా లు న్యవాదాలు धन्यवाद
ಧನ್ಯವಾದനന്ദിઆભાર‫ شکریہ‬ধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ நன்றிధన్ దా లు న్యవాదాలు
धन्यवाद
ಧನ್ಯವಾದനന്ദി આભાર‫ شکریہ‬धन्यवाद ধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ धन्यवाद

ಧನ್ಯವಾದ
Thank you
நன்றிధన్ దా లు న్యవాదాలు
நன்றிధన్ దా లు న్యవాదాలు
धन्यवाद ಧನ್ಯವಾದനന്ദി આભાર‫ شکریہ‬ধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ
നന്ദി धन्यवाद ಧನ್ಯವಾದനന്ദി આભાર‫شکریہ‬
ನ್ಯವಾ
న్యవాదాలు
ধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ நன்றிధన్ దా లు ಧನ್ നന്ദി ধ ದઆભાર ন্যবাদ ਤੁਹਾਡਾ
ਧੰਨਵਾਦ நன்றிధన్ దా లు న్యవాదాలుধন্যবাদ ਤੁਹਾਡਾ ਧੰਨਵਾਦ நன்றிధన్ దా లు న్యవాదాలు