UNIT 3 INFERENCE

Contents
3.0 Objectives
3.1 Introduction
3.2 Definition of Inference
3.3 Kinds of Inference
3.4 Indian Theory of Inference
3.5 Critique of Inference
3.6 Let Us Sum Up
3.7 Key Words
3.8 Further Readings and References
3.9 Answers to Check Your Progress

3.0 OBJECTIVES
The goal of this Unit is to give an insight about one of the important sources of
knowledge through which we come to cognise some new knowledge. Although
the details of the logical inference are not considered in this unit, it is a very
important Unit as it explains one of the sources of knowledge which is important
for the course on theory of knowledge, which provides an important discussion
on truth and validity of our knowledge. In this unit, we shall attempt to give a
definition of inference and how they are classified and the importance of its role
in the acquisition of new knowledge. We shall advance this concept both from the
western as well as the Indian approaches and how they are both differently and
similarly conceived by them. We shall also point out some important objections
against inference and discuss whether inference can offer us new knowledge.
Thus by the end of this Unit you should be able:
 to have a basic understanding of Inference;
 to understand the different kinds of inference;
 to have an overview of Indian inference;
 to have an idea about the objections against inference
 to realize the need of inference in obtaining new knowledge

3.1 INTRODUCTION
‘Inference’ is, in general, a subject matter of epistemology and Logic. To have a
better understanding of the concept of inference, it is essential that we comprehend
the association between the two major subjects of philosophy. Epistemology is
“the science of sure knowledge.” It deals with the nature and validity of knowledge:
that is about the truthfulness of our knowledge. On the contrary, Logic is interested 29
Sources of Belief in the correct form of the argument. “Logic teaches us how to use one’s mind;
how to draw a conclusion from the given premises;” but it does not teach us
whether what we thought or arrived at is true or false. It is outside the scope of
logic to guarantee us whether the conclusion arrived at is true or false. Truth and
falsity belong to the field of epistemology. Although both are concerned about
knowledge, their scope is different. In this unit, we shall try to understand inference
from the epistemological point of view rather than that of logic; however, it is
inevitable to avoid certain logical arguments, mostly propositional logic, to have
a better understanding of inference.
The Place of Inference
One of the essential discussions of epistemology is on the valid sources of
knowledge: how or through which, one comes to the process of cognition. The
principal sources of knowledge are classified into two: sensible and intellectual.
The first does not belong to this unit and therefore not the scope of this unit. The
second principal source is ‘intellectual’.
For most part of it, we acquire new knowledge of the reality through the intellect.
The Intellect gives us two types of knowledge: immediate and mediate. By
Immediate, we mean that knowledge that we gain intuitively, by looking at an
object. We identify a person that he/ she is somebody and he / she is not
somebody else. For example, you identify that somebody is Praveen and Praveen
is the son of Prakash – this is known as the principle of identity. So also there
are other principles which are given to us by the intellect immediately. Mediate or
reflective knowledge is acquired with “different operations of our intellect and
through the secondary sources.” By secondary source, we mean that the knowledge
already obtained through perception, or other previous knowledge. This intellectual
process is called reasoning or inference. Therefore, inference will have its place
right here in the mediate / reflective knowledge in which we move from the
perceptual knowledge to the new knowledge.

3.2 DEFINITION OF INFERENCE

Aristotle (384 – 322 B.C.), one of the prominent philosophers of the ancient
Greece, introduced the process of inference in the western philosophical world
through syllogism – a three statement propositions in which the conclusion is
drawn from the previous two propositions.
Dictionary of philosophy defines inference as the process that refers to “the
drawing of a conclusion.” It is also called reasoning. Thomas Aquinas says, “to
reason is to advance from one thing understood to another.” That is, we pass
from what is known to the unknown. We shall give an example: We are sitting
in a room and the door is closed. All of us hear a ‘triple knock’ on the door. All
that your senses tell you is that there was a sound produced on the door. It is
the intellect, which does not see the person(s) knocking at the door conceives of
different possibilities. It may be a single person who knocked at the door thrice:
or it may be two persons, one knocked twice and the other once: or still, there
were three persons who knocked at the door at regular intervals. Although it is
insignificant to know how many knocked at the door, it is important to know that
from what is given to us through the senses, the intellect goes through a process
30 to have knowledge of what happened. This process of drawing conclusion is
 Inference
known as inference. This process can be related to the past, present or future
events and occurrences.
We apprehend an event through perception and the intellect makes a judgment
and “on the basis of judgment previously made” we infer or draw a conclusion.
This process of coming to a new knowledge is called inference. We shall attempt
to explain this from propositional logic. In logic, “the propositions which lead up
to the new truth are called the antecedent. They give the reasons why we can
assert the new truth. The proposition which expresses the new truth is called the
consequent. The consequent flows from the antecedent as necessarily caused by
it.” Because of the antecedents we are able to know the consequent for certain.
The certainty of knowledge can be attained through this form inference.
While attempting to understand inference, it is necessary that we introduce another
notion, which is apparently similar to it, but very much different from it. That
notion is ‘argument.’ We should not misunderstand that inference is an argument.
There are lot of differences between an argument and inference. “An inference
can be defined as the psychological process of moving from one thought to
another. An argument can be constructed that corresponds to an inference. But
an inference is not equivalent to an argument. Furthermore, the premises of a
good argument imply, they do not infer, its conclusion, since only persons can
make inferences.” For inference itself is not an argument. Arguments are constructed
to correspond to the inference. Therefore “it is only correct to say that the
persons make inferences and the premises of a good argument imply the
conclusion.” So we could say that reasoning or inference is the process of the
intellect which infers a new cognition from an already known cognition.

3.3 KINDS OF INFERENCE

While there are disputes among scholars with regard to the classification of inference,
we shall classify it in the most known way. Inference or reasoning is of two kinds:
one is deductive and the other is inductive: the former is subdivided further into
two as immediate and mediate, while the latter is divided into many kinds of which
we shall discuss the important kinds in a while in this unit.
Deduction
Aristotle had held high deductive inference over induction, so much so Immanuel
Kant (1724 – 1804) wrote, and rightly so, in the Critique of Pure Reason that
Aristotle’s theory of logic completely accounted for the core of deductive inference.
For him certainty or being closer to the truth is attained only through deductive
inference. We shall first define deduction. A deduction is defined as “a valid
inference from necessary premises.” By necessary, we mean premises that are
self-evident truths or well established truths. The statement with which we begin
a proposition is called major premise. In deductive inference, the major premise
is true. For example: “all humans are mortal.” This is a necessary premise. From
this truth, you may arrive at new knowledge. You find then that Raja is a human
being. Therefore you infer that Raja is mortal. The knowledge you have about the
particular is true. Therefore according to him this form of inference is the best way
to have new and true knowledge. Since we move from a general truth to the
particular, deduction is understood as “a valid inference from more general
premises to a less general, i.e. more specific, conclusion” (Dictionary of 31
Sources of Belief Philosophy). In deduction it is very vital to note the very essential and basic
point that if the premise is true then the conclusion must be true, which is not the
case in induction.
Deductive inference could be divided into two: Namely, immediate inference and
mediate inference, of which we shall discuss in the next subdivision.
Immediate Inference
Strictly, immediate inference belongs to the field of logic; nevertheless we include
it here that we have an overall understanding of inference. In the above said
example, we notice that there were three statements, of which the first two are
called as antecedent and the last as consequent. Immediate inference is a different
kind of deductive inference which does not need two premises (antecedent) to
arrive at a conclusion but a single premise is sufficient. “Immediate inference is a
kind of deductive inference, in which, the conclusion follows from one premise.”
Because we have classified it under deduction, it is important to note that the
conclusion cannot be more general than the premise. To put it differently, it is a
process in which you infer one proposition from the given proposition. Immediate
inferences are of many kinds of which we shall see only the four.
Conversion
The first type of immediate inference is known as ‘conversion.’ “Conversion is a
kind of immediate inference, in which there is a legitimate transposition of the
subject and the predicate of a proposition.” for example, from the given example
of “No dogs are felines” you infer that “No felines are dogs.” We shall give
another example: the converse of “Some snakes are poisonous animals” is “Some
poisonous animals are snakes.”
Obversion
“obversion is a kind of immediate inference in which there is a change in the
quality of the given proposition, while its meaning remains unchanged.” For example,
the obverse of “All ants are insects” is “No ants are non-insects”; the obverse of
“Some musicians are males” is “Some musicians are not non-males.”
Contraposition
“Contraposition is a kind of immediate inference in which from a given proposition
we infer another proposition, having its subject the contrary of the given predicate.”
For example, the contrapositive of “All crows are birds” is “All non-birds are
non-crows.”
Inversion
“Inversion is a kind of immediate inference in which from a given proposition we
infer another proposition, having its subject the contradictory of the given subject.”
For example the inversion of “all men are mortal” is some not-men are not-
mortal.”
These four of them are called ‘eductions.’ Eductions may be defined as those
forms of immediate inference in which from a given proposition, accepted as true,
we derive others implied in it, though differing from it in subject or predicate or
32
both.” Apart from the above mentioned immediate inferences, there are also other
 Inference
kinds, like oppositions, modal consequence, change of relation, inference by
Added Determinants and inference by complex conception.
What occurs in the immediate inference is that we infer another proposition which
is already implied in it. Therefore there is a discussion among scholars about
‘immediate inference’ whether it is an inference at all, because there is no new
knowledge is arrived at through this inference; what happens is only an explication
of what is implicit. We are not offering any justification to state that immediate
inference is a genuine form of inference but it is good to know that when we
discover what is hidden in a proposition or an object it offers us a new of
knowledge.
Mediate Deductive Inference
Contrast to immediate inference, the consequent or the new knowledge is deduced
from more than one proposition: that is “the conclusion follows from more than
one proposition. Where there are only two premises, and the conclusion follows
from them taken jointly.” This form of mediate inference is called “syllogism.”
“The nature of syllogistic reasoning was first disengaged, as said already, by
Aristotle in the Prior Analytics in which he undertook to lay bare the essential
structure of scientific knowing.” He defines syllogism as “an argument in which,
certain truths having been assumed, something other than these follows of necessity
from their truths, without needing any term outside” (I .1 24b 18). We could say
that syllogism is “the outward expression of deductive sequence.” We are probably
familiar with the famous syllogism.
i) All men are mortal
ii) Socrates is a man
iii) Therefore, Socrates is mortal.
The conclusion of the syllogism rests upon the preceding two propositions (i and
ii). We arrive at a specific and a particular knowledge from a more general
knowledge. In this way of reasoning, one can be sure that what he knows is true.
But in our ordinary life we won’t be using formal syllogism to deduce certain
knowledge. Aristotle is aware of this and suggests the use of what he calls as
Enthymeme. It is a kind of syllogism in which the any of the premise will be
missing. For example to try to prove the certainty of truth we say: Socrates was
mortal for he was only a man. What is missing here is the Major premise that all
men are mortal: but it is so obvious that it is not needed in persuasion. Mostly
it is used in debates, in court rooms for the benefit of persuasion. We know that
we cognise through deduction, complete or incomplete, to know and to prove
that something is true.
Check Your Progress I
Note: a) Use the space provided for your answer.
b) Check your answers with those provided at the end of the unit.
1) What is inference?
.................................................................................................................
................................................................................................................. 33
Sources of Belief
.................................................................................................................
.................................................................................................................
2) What is deduction?
.................................................................................................................
.................................................................................................................
.................................................................................................................
.................................................................................................................
Induction
In the history of western philosophy from the period of Aristotle majority of them
accepted deductive inference to be a valid form of inference. But From the
modern period philosophers, particularly Francis Bacon (1561 – 1626) “lamented
the powerlessness of deduction” because they insisted it is not a useful form of
inference to advance to new knowledge. Because what happens in deduction is
simply the explication of what is implied in the major premise and the conclusion
cannot be more general than the major premise. Therefore, they propose inductive
inference to be the sole inference which can offer us to have new knowledge.
Compared to deduction, induction moves from specific instances to have a general
conclusion. The traditional definition of Induction is that it is an “inference in which
the intellect moves to cognise from a finite number of particular cases to a further
case or to a general conclusion.” We shall give a familiar example: we observe
that swan A is white, then swan B is white, and therefore we conclude that all
swans are white. We make a movement from the particular to make a larger
conclusion which is not implied in the antecedent. Bacon and J. S. Mill (1806 –
1873) did not see any problem with this conception of inductive inference. But
others pose different problems with this kind of understanding of inference: namely
“to categorically accept another statement on the basis of premises that are
categorically accepted.” The problems arise out of the uncertainty involved in this
inference. That has given rise to understanding inductive inference differently by
different people. We shall therefore first explain how mill and others understood
induction and the other kinds of induction in the following section.
Inductive Generalization / Enumerative Induction
It is a kind of an inference, where lot of particulars are observed from which we
try to generalise the conclusion. In fact the definition we gave in the last passage
directly applies to the enumerative induction. The standard example of this kind
of inference is the following: from all the observed ravens being black, we infer
that all ravens are black. Mill argued that “inductive generalization is the only
legitimate kind of induction.” That is why this is called ampliative. But the problem
with this kind of induction is, (with any kind of induction for that matter) how
could we move from the observed objects to the unobserved objects. For example,
all the swans observed in the 18th century Europe were white. And therefore we
make a conclusion that all swans are white. But that is not true. Swans in Australia
were observed to be black. Therefore the conclusion is false. This uncertainty of
the conclusion in induction is a hindrance to have a true knowledge.
34
 Inference
Statistical Inductive Generalisation
Some scholars propose therefore that induction has to be conceived differently.
In their view, the conclusion has to be inferred based on the percentage of the
observed particulars. Suppose we know by induction that 90 percentages of
women in Japan are less than 5 feet; then we could infer the conclusion that the
next woman from Japan will have a 90percentage chance of being less than 5 feet
and make a conclusion that 90 percentage of women in Japan are less than five
feet.
Probability Theory
Rudolf Carnap (1891 – 1970), Richard C. Jeffrey (1926 – 2002) and others
hold that “induction should be conceived not as a process by which we pass from
some accepted statements to others, but rather as a process by which we assign
probabilities to various hypothesis in the light of our evidence.” This type of
induction involves a two-step process: the first is in identifying a broad class of
possible confirmation functions and the second in identifying either a unique function
in that class or a parametric family of specific confirmation functions. In Fact, the
modern probability theory is influenced by Thomas Bayes (c. 1702 – 1761).
There are also some problems in this view. We shall explain it with a thought
experiment ‘the Lottery Paradox’ proposed by Henry E. Kyburg (1928 – 2007)
in which we will be forced rationally to accept the contradictory propositions that
one ticket wins and no ticket wins: because it is probable that any ticket can win
but at the same time every ticket has more probability of losing than winning.
Predictive Inference
It is a form of inference that emphasises the prediction of future occurrences
based on the past observation. It could be based on cause and effect relationship
or analogy. That there is fire because there is smoke or that it will rain because
there are dark clouds. The Indian scholars in fact have different names for this
kind of inference.
The principle of Induction
From what we have seen so far, we could realize that everyone proposes different
understanding of inductive inference to eliminate the possibility of having false
knowledge. We could also understand that there is a principle that is operative
in this form of inductive inference which is known as the principle of induction.
This principle is formulated as “the assertion that events in the future will resemble
events in the past, or that unobserved cases will resemble observed cases.” Some
even argue that this principle may be used “to reduce all inductive arguments to
deductive arguments.” But the question how one could really justify the principle
of induction has given rise to what is known as the problem of induction, which
will be discussed in final section.
Check Your Progress II
Note: a) Use the space provided for your answer.
b) Check your answers with those provided at the end of the unit.
1) What is the classical definition of inductive inference?
................................................................................................................. 35
Sources of Belief
.................................................................................................................
.................................................................................................................
.................................................................................................................
2) What is the principle of induction?
.................................................................................................................
.................................................................................................................
.................................................................................................................
.................................................................................................................

3.4 INDIAN THEORY OF INFERENCE

Introduction
Indian philosophy which is older than the western tradition had a deep philosophical
outlook on the theory of knowing. They discuss in detail the sources of knowledge,
particularly inference. In Sanskrit the term Pramana conveys the meaning of the
source of knowledge. The chief Pramanas which are discussed by majority of
them are two: namely perception (Pratyaksa) and inference which is indicated by
the term Anumana. Of course there are also other valid sources which are
accepted like Verbal testimony (sabda), comparison (Upamana) etc. In Indian
philosophy, logic and epistemology were joined together and were not separated
as in the western thought. “What was meant by syllogistic reasoning corresponds
in India to what is known as anumana – inference. Inference in Indian understanding
includes both ‘deduction’ and ‘induction.’
Meaning of the Term Anumana:
The Sanskrit word Anumana is the combination of two words. Anu means after
and mana means measurement. “The whole word literally means the measuring
after something.” According to them it is a knowledge that is obtained after proof.
We know by now that knowledge derived through anumana is not direct “since
it makes use of previous knowledge obtained” from other sources of knowledge
like perception, testimony etc., and “enables one to explore further knowledge.”
Not all the major Indian philosophical systems accept all the pramanas. For
example, the Carvakas – the Indian materialists who hold the theory that matter
is the only reality – do not accept anumana as a valid source of knowledge.
Structure of Anumana
Although all the major schools accept anumana as a valid source of knowledge,
the understanding and the explanation of each school will have certain variation
according to their understanding of knowledge. In Indian philosophy inference is
used for oneself and inference for others. When inference is used for oneself the
propositions are not well structured since its primary aim is the acquisition of
personal knowledge without error, whereas inference for others has to be well
structured because it is used to convince the other of the truth. We shall in this
unit concentrate mainly the understanding on Nyaya because it is well known for
36 its logic.
 Inference
Inference is defined by them as “a process of reasoning in which we pass from
the apprehension of some mark (linga) to that of something else by virtue of an
invariable relation (vyapti) that exists between them.” Vyapti is essential in Indian
philosophy for making a valid inference: however, it is good to know that different
schools had different names for vyapti; For example, Vaisesikas called it Prasiddhi
and Samkhya called it pratibandha.
Nyaya proposes a longer syllogism; it has five propositions. An argument according
to them has five parts: Namely, Paksa or Pratinjna, hetu, drastanta, upanaya
and nigamana. We shall give a standard example to understand this.
1. Paksa – The Thesis / Pratijna – Proposition = The hill has fire
2. Hetu – Reason or the ground = Because it has smoke
3. Drstanta – The corroboration = wherever there is smoke
there is fire, as in the
kitchen
4. Upanaya – The application = the hill is so
5. Nigamana – the conclusion = Therefore the hill has
fire.
In this process, we begin asserting something, then we provide the reason / the
ground for the assertion and make a universal proposition which shows the
concomitant relationship between the two with an example then we apply the
universal proposition to the present case and make a conclusion from the preceding
propositions. This type of syllogism is said to have anvaya vyapti – since it
denotes a positive concomitance – if there is smoke then there is fire. We shall
give a specimen from the western example: 1) Ram is mortal 2) Because he is
a man 3) All men are mortal like my grandfather 4) Ram is also a man 5)
Therefore Ram is mortal. The purpose of giving this example is also to show how
Indian philosophy combined both induction and deduction together in the same
syllogism. The first 3 propositions (1 – 3) form inductive syllogism, while the last
three (3 – 5) form as a deduction. The proposition 3 is the conclusion for the
induction and the major premise for the deduction.
When it denotes negative concomitance it is said to have vyatireka Vyapti. An
example of this is the opposite of what we have stated above. The hill has no
smoke; because there is no fire; wherever there is no fire there is no smoke as
in the lake (because water and fire are opposed substances); there is no fire in
the hill; therefore the hill has no smoke.
Classification of Inference
Inference here is classified based on the nature of vyapti between hetu (smoke)
sadhya (fire). Vyapti denotes a correlation between two facts of which one is
pervaded and the other which pervades. E.g. Smoke is pervaded by fire and fire
pervades smoke. Vyapti is established based on its presence of both in all such
events (wherever there is smoke there is fire) and the absence of both (wherever
there is no fire there is no smoke). The classification is based on the relationship
(causal uniformity or non-causal uniformity) between the reason and what is
inferred. There are three types of inference. 37
Sources of Belief 1. Purvavat inference
“It is that in which we infer the unperceived effect from a perceived cause.” E.g.
we infer of future rain from the appearance of dark heavy clouds.
2. Sesavat inference
“It is that in which we infer the unperceived cause from a perceived effect.” E.g.
we infer of the past rain from swift muddy current of water in the river.
3. Samanyatodrasta inference
“It is that which we infer not based on causal relation but on experience of
uniformity.” E.g. on seeing the different positions of the moon at long intervals, we
infer that it moves although the motion might not have been perceived by us.

3.5 CRITIQUE OF INFERENCE

Having seen both Indian and western understanding of inference, it is good to
evaluate them. There are quite a few objections raised by some philosophers who
are suspicious about the inferential knowledge we acquire. Since this needs an
elaborate study by itself we shall restrict ourselves to some important objections
alone.
The first one is with regard to deduction. If one has to obtain a true cognition
through inference, it is essential and necessary that the major premise as well as
the minor premise be true. We can easily recognize the truthfulness of the minor
premise “since it is an object of direct perception. But the problem is, “how can
the major premise be recognized to be true since it is not such an object of direct
perception?” we shall enumerate the following example to make it clear. Suppose
we say that ‘all that begin to exist must someday cease to exist’ (major): you have
begun to exist. Therefore you must someday cease to exist. The question regarding
the major premise is this. How does one know that ‘all that begin to exist must
someday cease to exist’ is unconditionally or necessarily true? Has anyone sense
experienced all things – not only the present but also of the past and future.
Therefore such a statement is a generalization which cannot be proved to
correspond to facts. Scholars reply that the truth of deduction depends on self
evident principles or self evident truth. Take for example the principle of non-
contradiction which is a self evident principle – A being cannot be and not be at
the same time under the same respect.” It does not need any proof.
Another objection with regard to deduction is that it does not offer any new
knowledge. As to whether inference can yield new knowledge even granting that
the conclusion is implied in the major premise – the answer seems to be that to
come to an explicit awareness of what is only logically implied in a already
known truth is surely a process of every act of understanding. Whether is not
one prepared to call this newly explicated awareness ‘knowledge’ depends on
one’s understanding of the term.
The next objection is with regard to induction. We have explained about the
principle of induction. But the problem is how to justify that principle? The problem
was first raised by David Hume. Hume does not use the word induction; nevertheless
it has come to be known as the problem of induction, wherein he is stating that
38 this form of inference cannot be justified. There are two things involved in induction.
 Inference
First, What we do in inference is to generalise the properties of a class from a
certain number of observed instances of that class: for example that swan A, B,
C, … X are white therefore all swans are white. We apply whiteness to all swans.
The next we do is to infer certain laws based on our observation in the past. How
to justify these steps is the philosophical question known as the problem of
induction. Some scholars try to answer that “in the past, similar things behaved
similarly under similar circumstances. Therefore, all similar things behave similarly
under similar circumstances.” What has happened here is that an inductive argument
has been used to justify the principle of induction and therefore ending up in
circularity; but some claim that this circularity is legitimate which is not universally
accepted. Hume raises the question that the past cannot be foundation for the
future events which we infer based on their causal connections. Hume says:
For all inferences from experience suppose, as their foundation, that the future will
resemble the past, and that similar powers will be conjoined with similar sensible
qualities. If there be any suspicion that the course of nature may change, and that
the past may be no rule for the future, all experience becomes useless, and can
give rise to no inference or conclusion.” (Section IV 32 in the Enquiry)
Let us explain the problem from the same example. I can perceive some things
which begin to exist and then cease to exist. For example I plant a tree and I see
it dying. I have seen my friend being born and dying. But the question is: how
could one say that they cease to exist because they began to exist? In other
words, even if decomposition was an essential characteristic of things which I
sense perceived to have ceased to exist, how can I say that all things are, in this
respect, similar to them? How can I form a ‘class’ of things characterized by the
same characteristics? What I sense perceive is in this particular thing and in that
particular thing, but not any ‘universal.’ How could I make a jump from the
particular to the class of things? It cannot be merely based on the regularity of
nature because there is no certainty that nature will behave the same way tomorrow.
It leaves lots of uncertainty about the fact that we assert. So, many claim that
inductive inference cannot give us a sure knowledge. Karl Popper (1902 – 1994)
maintained that “what is called induction is a myth in as much as what passes
under the title ‘is always invalid and therefore clearly not justifiable.” He says that
only deduction has the power to prove a scientific theory. Therefore Popper
proposes what is called Hypothetico-deductive method.
Hypothetico-deductive Method
Since there is a problem about making a generalisation from the observed to the
unobserved, this method is suggested. This has three steps: “1) the formulation of
a “hypothetical” generalisation; 2) the deduction of particular observation statements
from this generalisation; and 3) the testing of the observation statements whether
they confirm or falsify the generalisation.” We shall explain it with the example
given by Robert Baum. 1) All sea otters use rocks to crack open the sea shells.
It is simply a working hypothesis. 2) Next sea otter will use rock to crack open
the sea shells. We deduce the observation statements. We may deduce any
number of propositions like the 23rd sea otter will crack open the sea shells. 3)
The third step involves in justifying the proposition which involves the process of
falsification. The more we observe the greater certainty we can have about the
generalisation.
39
Sources of Belief It is good to note at this juncture that there was a different response to the
problem from India. While Popper suggests hypothetico method, the Indian schools
propose hypothesis itself as an independent source of knowledge apart from
inference. For examples, Mimamsa accepts arthapathi (postulation or hypothesis)
as an independent response. From the western point of view, it may said to be
a kind of inference. But Mimamsikas explain that it is not; because there are
cases in which there may not be any (vyapti) invariable concomitance between
hetu and sadhya. For example; a man is seen fasting during the day. Yet he is
growing fat. Therefore we say that he should be eating at night – this is a
hypothesis / postulation. Here there is no invariable concomitance between fatness
and eating at night. Therefore arthapathi is an independent source of knowledge.
Hypothesis as an independent source was an answer to the jump we make from
the observed to the unobserved facts.
To conclude let us say that unless and until one is prepared to posit some kind
of insight into the whole process – obtaining any new knowledge – including
perceptually knowable – one cannot start even meaningfully about anything at all.
For the critique against inference themselves fall into using categorical statements
and concepts which are cognized through inference. Hence it is difficult to do
away with inference. We shall conclude that as Kant in his Critique of Pure
Reason says that “Our knowledge springs from two fundamental sources of the
mind; the first is the capacity of receiving representations (receptivity for
impressions), the second is the power of knowing an object through these
representations (spontaneity [in the production] of concepts).” Inference then is
necessary, although there are lot of problems connected with this, to obtain true
and valid knowledge.
Check Your Progress III
Note: a) Use the space provided for your answer.
b) Check your answers with those provided at the end of the unit.
1) What is anumana?
.................................................................................................................
.................................................................................................................
.................................................................................................................
2) What is the problem of induction?
.................................................................................................................
.................................................................................................................
.................................................................................................................

3.6 LET US SUM UP

We began this unit in situating inference as one of the important sources of
knowledge. We gave a definition of inference which is a psychological process in
which we move from the previous knowledge to have a new knowledge. We
differentiated it from argument saying that only persons can infer while an argument
40 could only imply. We classified inference into two: deduction and induction.
 Inference
Deduction is a kind of reasoning in which we arrive at a new knowledge from the
given propositions or self evident principles. Deduction is divided into two as
immediate inference and mediate inference. The difference between them lie in the
number of propositions; while the former infers the new proposition from one
proposition the latter needs at least two or more premises.
We discussed then about inductive inference which was important particularly in
the field of science to arrive at new knowledge. We examined the different
understanding of induction because of the problem involved in the principle of
induction which is at the bottom of making any inductive inference. We then gave
an overview of Indian theory of inference which unites both deduction and induction
in the same syllogism. We also brought to our notice the different kinds of inference
which is classified based on vyapti. We then discussed the problems related to
inference and how we could respond to those problems. Hence at the end of the
unit we know by now that inference is part and parcel of the process of cognition
because it is only reasoning that helps to acquire new knowledge without which
our knowledge would be always stand still. That is why with Kant we could say
that “though all our knowledge begins with experience, it does not follow that it
all arises out of experience.”

3.7 KEY WORDS

Syllogism : “it is a form of mediate deductive inference,
in which the conclusion is drawn from two
premises, taken jointly.” Syllogism consists
of three propositions.
Enthymeme : “an enthymeme is a syllogism with some
of its constituent propositions suppressed.”
Suppose I say that Socrates is mortal for he
is a man, what is suppressed is the major
premise “all men are mortal.” It can then
said to be an incomplete syllogism.
Vyapti : It denotes a correlation between two facts
of which one is pervaded and the other which
pervades. E.g. Smoke is pervaded by fire
and fire pervades smoke. Vyapti is
established based on its presence of both in
all such events (wherever there is smoke
there is fire) and the absence of both
(wherever there is no fire there is no smoke).

3.8 FURTHER READINGS AND REFERENCES

Baum, Robert. Logic. U.S.A: Holt, Rinehart and Winston, Inc., 1975.
Bhattacharya, Birendra Kumar. Inference in Indian and Western Logic. Calcutta:
Sanskrit Pustak Bhandar, 1976.
Bhattacharyya, Hari Mohan. Jaina Logic and Epistemology. Calcutta: K.P.
Baachi & company, 1994.
41
Sources of Belief Blandino S.I., Giovanni. An Outline of the Philosophy of Knowledge and
Science. Roma: Pont. Universitas Lateranensis, 1989.
Edwards, Paul, ed. The Encyclopedia of Philosophy. Vol. 3 & 4. New York:
Macmillan Publishing co., Inc & the Free Press, 1967. Reprint. 1972.
Francis, B. Joseph. Epistemology in Theory & Practice. Bangalore: St. Peter’s
Pontifical Institute, 2005.
Kandaswamy, S.N. Indian Epistemology: As Expounded in the Tamil Classics.
Chennai: International Institute of Tamil Studies, 2000.
Mercier, Jean L. Epistemology and the Problem of Truth. Bangalore: Asian
Trading Corporation, 2000.
Rodrigues, Antonio F. X. Epistemology or Philosophy of Knowledge. Bangalore:
Redemptorist Publications India.
Roy, Bhola Nath. Text Book of Deductive Logic with Copious Examples and
Questions. Calcutta: S.C. Sarkar & Sons Ltd, 1953 (7th edition).
Subba Rao, Veluri. Theories of Knowledge: its validity and its sources. Delhi:
Sri Satguru Publications, 1998.
Sullivan, Daniel J. Fundamentals of Logic. New York: McGrew – Hill Book
Company, Inc., 1963.

3.9 ANSWERS TO CHECK YOUR PROGRESS

Answers to Check Your Progress I
1. Inference is the process that refers to “the drawing of a conclusion.” It could
be also called reasoning. Reasoning or inference is the process of the
intellect which moves to a new cognition from an already known cognition.
2. A deduction is defined as a valid inference from necessary premises. Deduction
is in general understood as a valid inference from more general premises to
a less general, i.e. more specific, conclusion.
Answers to Check Your Progress II
1. The classical definition of Induction defines induction as an inference in which
the intellect moves to cognise from a finite number of particular cases to a
further case or to a general conclusion.
2. The principle of induction is formulated as “the assertion that events in the
future will resemble events in the past, or that unobserved cases will resemble
observed cases.”
Answers to Check Your Progress III
1. The Sanskrit word Anumana is the combination of two words. Anu means
after and mana, which is the same as in the end of pramana, literally means
measurement. “The whole word literally means the measuring after something.”
2. The problem of induction is the philosophical question of inductive inference
whether it leads to truth. It raises the need for justification for the two
important functions that are part of the inductive inference: generalization and
42 the belief in the causal link of events which rests on the regularity of nature.