|
Epistemology
is one of the main branches of philosophy; its subject matter concerns the
nature, origin, scope, and limits of human knowledge. The name is derived from the Greek terms episteme
(knowledge) and logos (theory), and
accordingly this branch of philosophy is also referred to as the theory of
knowledge.
Why should there be such a subject as
epistemology? Aristotle provided the answer when he said that philosophy begins
in wonder, in a kind of puzzlement about things. Nearly all human beings wish to
comprehend the world they live in, a world that includes the individual as well
as other persons, and most people construct hypotheses of varying degrees of
sophistication to help them make sense of that world. No conjectures would be
necessary if the world were simple; but its features and events defy easy
explanation. The ordinary person is likely to give up somewhere in the process
of trying to develop a coherent account of things and to rest content with
whatever degree of understanding he has managed to achieve.
Philosophers, in contrast, are struck
by, even obsessed by, matters that are not immediately comprehensible.
Philosophers are, of course, ordinary persons in all respects except perhaps
one. They aim to construct theories about the world and its inhabitants that are
consistent, synoptic, true to the facts and that possess explanatory power. They
thus carry the process of inquiry further than people generally tend to do, and
this is what is meant by saying that they have developed a philosophy about
these matters. Epistemologists, in particular, are philosophers whose theories
deal with puzzles about the nature, scope, and limits of human knowledge.
Like ordinary persons, epistemologists
usually start from the assumption that they have plenty of knowledge about the
world and its multifarious features. Yet, as they reflect upon what is
presumably known, epistemologists begin to discover that commonly accepted
convictions are less secure than originally assumed and that many of man's
firmest beliefs are dubious or possibly even chimerical. Such doubts and
hesitations are caused by anomalous features of the world that most people
notice but tend to minimize or ignore. Epistemologists notice these things too,
but, in wondering about them, they come to realize that they provide profound
challenges to the knowledge claims that most individuals blithely and
unreflectingly accept as true.
What then are these puzzling issues?
While there is a vast array of such anomalies and perplexities, which will be
discussed below in the section on the history of epistemology, two of these
issues will be briefly described in order to illustrate why such difficulties
call into question common claims to have knowledge about the world.
Most people have noticed that vision
can play tricks on them. A straight stick put in water looks bent to them, but
they know it is not; railroad tracks are seen to be converging in the distance,
yet one knows that they are not; the wheels of wagons on a movie screen appear
to be going backward, but one knows that they are not; and the pages of
English-language books reflected in mirrors cannot be read from left to right,
yet one knows that they were printed to be read that way. Each of these
phenomena is thus misleading in some way. If human beings were to accept the
world as being exactly as it looks, they would be mistaken about how things
really are. They would think the stick in water really to be bent, the railway
tracks really to be convergent, and the writing on pages really to be reversed.
(see also illusion, appearance)
These are visual anomalies, and they
produce the sorts of epistemological disquietudes referred to above. Though they
may seem to the ordinary person to be simple problems, not worth serious notice,
for those who ponder them they pose difficult questions. For instance, human
beings claim to know that the stick is not really bent and the tracks not really
convergent. But how do they know that these things are so? (see also optical
illusion, refraction)
Suppose one says that this is known
because, when the stick is removed from the water, one can see that it is not
bent. But does seeing a straight stick out of water provide a good reason for
thinking that it is not bent when seen in water? How does one know that, when
the stick is put into the water, it does not bend? Suppose one says that the
tracks do not really converge because the train passes over them at that point.
How does one know that the wheels on the train do not happen to converge at that
point? What justifies opposing some beliefs to others, especially when all of
them are based upon what is seen? One sees that the stick in water is bent and
also that the stick out of the water is not bent. Why is the stick declared
really to be straight; why in effect is priority given to one perception
over another? (see also sense)
One possible response to these queries
is that vision is not sufficient to give knowledge of how things are. One needs
to correct vision in some other way in order to arrive at the judgment that the
stick is really straight and not bent. Suppose a person asserts that his reason
for believing the stick in water is not bent is that he can feel it with his
hands to be straight when it is in the water. Feeling or touching is a mode of
sense perception, although different from vision. What, however, justifies
accepting one mode of perception as more accurate than another? After all, there
are good reasons for believing that the tactile sense gives rise to
misperception in just the way that vision does. If a person chills one hand and
warms the other, for example, and inserts both into a tub of water having a
uniform medium temperature, the same water will feel warm to the cold hand and
cold to the warm hand. Thus, the tactile sense cannot be trusted either and
surely cannot by itself be counted on to resolve these difficulties.
Another possible response is that no
mode of perception is sufficient to guarantee that one can discover how things
are. Thus, it might be affirmed that one needs to correct all modes of
perception by some other form of awareness in order to arrive at the judgment,
say, that the stick is really straight. Perhaps that other way is the use of reason.
But why should reason be accepted as infallible? It also suffers from various
liabilities, such as forgetting, misestimating, or jumping to conclusions. And
why should one trust reason if its conclusions run counter to those gained
through perception, since it is obvious that much of what is known about the
world derives from perception?
Clearly there is a network of
difficulties here, and one will have to think hard in order to arrive at a clear
and defensible explanation of the apparently simple claim that the stick is
really straight. A person who accepts the challenge will, in effect, be
developing a theory for grappling with the famous problem called "our
knowledge of the external world." That problem turns on two issues, namely,
whether there is a reality that exists
independently of the individual's perception of it--in other words, if the
evidence one has for the existence of anything is what one perceives, how can
one know that anything exists unperceived?--and, second, how one can know what
anything is really like, if the perceptual evidence one has is conflicting.
The second problem also involves seeing
but in a somewhat unusual way. It deals with that which one cannot see, namely
the mind of another. Suppose a woman is
scheduled to have an operation on her right knee and her surgeon tells her that
when she wakes up she will feel a sharp pain in her knee. When she wakes up, she
does feel the pain the surgeon alluded to. He can hear her groaning and see
certain contortions on her face. But he cannot feel what she is feeling. There
is thus a sense in which he cannot know what she knows. What he claims to know,
he knows because of what others who have undergone operations tell him they have
experienced. But, unless he has had a similar operation, he cannot know what it
is that she feels. (see also sensation)
Indeed, the situation is still more
complicated; for, even if the doctor has had such a surgical intervention, he
cannot know that what he is feeling after his operation is exactly the same
sensation that the woman is feeling. Because each person's sensation is private,
the surgeon cannot really know that what the woman is describing as a pain and
what he is describing as a pain are really the same thing. For all he knows, she
could be referring to a sensation that is wholly different from the one to which
he is alluding.
In short, though another person can
perceive the physical manifestations the woman exhibits, such as facial grimaces
and various sorts of behaviour, it seems that only she can have knowledge of the
contents of her mind. If this assessment of the situation is correct, it follows
that it is impossible for one person to know what is going on in another
person's mind. One can conjecture that a person is experiencing a certain
sensation, but one cannot, in a strict sense of the term, know it to be the
case.
If this analysis is correct, one can
conclude that each human being is inevitably and even in principle cut off from
having knowledge of the mind of another. Most people, conditioned by the great
advances of modern technology, believe that in principle there is nothing in the
world of fact about which science cannot obtain knowledge. But the
"other-minds problem" suggests the contrary--namely, that there is a
whole domain of private human experience that is resistant to any sort of
external inquiry. Thus, one is faced with a profound puzzle, one of whose
implications is that there can never be a science of the human mind.
These two problems resemble each other
in certain ways and differ in others, but both have important implications for
epistemology.
First, as the divergent perceptions
about the stick indicate, things cannot just be as they appear to be. People
believe that the stick which looks bent when it is in the water is really
straight, and they also believe that the stick which looks straight when it is
out of the water is really straight. But, if the belief that the stick in water
is really straight is correct, then it follows that the perception human beings
have when they see the stick in water cannot be correct. That particular
perception is misleading with respect to the real shape of the stick. Hence, one
has to conclude that things are not always as they appear to be.
It is possible to derive a similar
conclusion with respect to the mind of another. A person can exhibit all the
signs of being in pain, but he may not be. He may be pretending. On the basis of
what can be observed, it cannot be known with certitude that he is or that he is
not in pain. The way he appears to be may be misleading with respect to the way
he actually is. Once again vision can be misleading.
Both problems thus force one to
distinguish between the way things appear and the way they really are. This is
the famous philosophical distinction between appearance and reality. But, once
that distinction is drawn, profound difficulties arise about how to distinguish
reality from mere appearance. As will be shown, innumerable theories have been
presented by philosophers attempting to answer this question since time
immemorial.
Second, there is the question of what is
meant by "knowledge." People claim to know that the stick is really
straight even when it is half-submerged in water. But, as indicated earlier, if
this claim is correct, then knowledge cannot simply be identical with
perception. For whatever theory about the nature of knowledge one develops, the
theory cannot have as a consequence that knowing something to be the case can
sometimes be mistaken or misleading.
Third, even if knowledge is not simply
to be identified with perception, there nevertheless must be some important
relationship between knowledge and perception. After all, how could one know
that the stick is really straight unless under some conditions it looked
straight? And sometimes a person who is in pain exhibits that pain by his
behaviour; thus there are conditions that genuinely involve the behaviour of
pain. But what are those conditions? It seems evident that the knowledge that a
stick is straight or that one is in great pain must come from what is seen in
certain circumstances: perception must somehow be a fundamental element in the
knowledge human beings have. It is evident that one needs a theory to explain
what the relationship is--and a theory of this sort, as the history of the
subject all too well indicates, is extraordinarily difficult to develop.
The two problems also differ in certain
respects. The problem of man's knowledge of the external world raises a unique
difficulty that some of the best philosophical minds of the 20th century (among
them, Bertrand Russell, H.H. Price, C.D. Broad, and G.E. Moore) spent their
careers trying to solve. The perplexity arises with respect to the status of the
entity one sees when one sees a bent stick in water. In such a case, there
exists an entity--a bent stick in water--that one perceives and that appears to
be exactly where the genuinely straight stick is. But clearly it cannot be; for
the entity that exists exactly where the straight stick is is the stick itself,
an entity that is not bent. Thus, the question arises as to what kind of a thing
this bent-stick-in-water is and where it exists.
The responses to these questions have
been innumerable, and nearly all of them raise further difficulties. Some
theorists have denied that what one sees in such a case is an existent entity at
all but have found it difficult to explain why one seems to see such an entity.
Still others have suggested that the image seen in such a case is in one's mind
and not really in space. But then what is it for something to be in one's mind,
where in the mind is it, and why, if it is in the mind, does it appear to be
"out there," in space where the stick is? And above all, how does one
decide these questions? The various questions posed above only suggest the vast
network of difficulties, and in order to straighten out its tangles it becomes
indispensable to develop theories.
Philosophy viewed in the broadest
possible terms divides into many branches: metaphysics, ethics, aesthetics,
logic, philosophy of language, philosophy of mind, philosophy of science, and a
gamut of others. Each of these disciplines has its special subject matter: for
metaphysics it is the ultimate nature of the world; for ethics, the nature of
the good life and how people ideally ought to comport themselves in their
relations with others; and for philosophy of science, the methodology and
results of scientific activity. Each of these disciplines attempts to arrive at
a systematic understanding of the issues that arise in its particular domain.
The word systematic is important in this connection, referring, as explained
earlier, to the construction of sets of principles or theories that are
broad-ranging, consistent, and rationally defensible. In effect, such theories
can be regarded as sets of complex claims about the various matters that are
under consideration.
Epistemology stands in a close and
special relationship to each of these disciplines. Though the various divisions
of philosophy differ in their subject matter and often in the approaches taken
by philosophers to their characteristic questions, they have one feature in
common: the desire to arrive at the truth about
that with which they are concerned--say, about the fundamental ingredients of
the world or about the nature of the good life for man. If no such claims were
asserted, there would be no need for epistemology. But, once theses have been
advanced, positions staked out, and theories proposed, the characteristic
questions of epistemology inexorably follow. How can one know that any such
claim is true? What is the evidence in favour of (or against) it? Can the claim
be proven? Virtually all of the branches of philosophy thus give rise to
epistemological ponderings.
These ponderings may be described as
first-order queries. They in turn inevitably generate others that are, as it
were, second-order queries, and which are equally or more troubling. What is it
to know something? What counts as evidence for or against a particular theory?
What is meant by a proof? Or even, as the Greek Skeptics asked, is human
knowledge possible at all, or is human access to the world such that no
knowledge and no certitude about it is possible? The answers to these
second-order questions also require the construction of theories, and in this
respect epistemology is no different from the other branches of philosophy. One
can thus define or characterize epistemology as that branch of philosophy which
is dedicated to the resolution of such first- and second-order queries.
As indicated above, one of the basic
questions of epistemology concerns the nature of knowledge. Philosophers
normally interpret this query as a conceptual question, i.e., as an issue about a certain conception or idea or notion
called knowledge. The question raises a perplexing methodological issue, namely,
how does one go about investigating such conceptual questions? It is frequently
assumed, though the matter is controversial, that one can determine what
knowledge is if one can understand what the word "knowledge" means,
that is, what notion or concept the word "knowledge" expresses or
embodies.
Philosophers who proceed in this way
draw a distinction between a word and its meaning, and a meaning is generally
considered to be the concept which that particular word has or expresses. It is
usually further assumed that though concepts are not identical with words, that
is, with linguistic expressions, language is the
medium in which the meaning of such concepts is displayed or expressed.
The investigation into the nature of
knowledge often begins in a similar fashion with the study of the use of the
word "knowledge" and of certain cognate expressions and phrases found
in everyday language. A survey of such locutions reveals important differences
in their uses: one finds such expressions as "know him," "know
that," "know how," "know where," "know why,"
or "know whether." These differences have been explored in detail,
especially in the 20th century. The expression "know x," where "x"
can be replaced by a proper name, as in "I know Jones" or "He
knows Rome," has been taken by some philosophers, notably Bertrand
Russell (1872-1970), to be a case of knowledge
by acquaintance. Russell thought its characteristic use was to express
the kind of knowledge one has when one has first-hand familiarity with a certain
object, person, or place. Thus, one could not properly say in the 20th century,
"I know Julius Caesar," since this would imply that one had met or was
directly acquainted with a person who had died some 2,000 years ago. This sense
or use of "know" becomes important in the theory of perception and in
sense-data theory, since some philosophers, such as Russell and G.E. Moore
(1873-1958), have held that one's awareness of a sense-datum (a notion to be
discussed later) is a case of direct acquaintance, whereas one's acquaintance
with a physical object, such as a human hand, is not.
The phrases "know that" and
"know how" have also played fundamental roles in the theory of
knowledge. The British philosopher Gilbert Ryle
(1900-76), for instance, argued that "know how" is normally used to
refer to a kind of skill that a person has, such as knowing how to swim. One
could have such knowledge without being able to explain to another what it is
that one knows in such a case, that is, without being able to convey to another
the knowledge required for that person to develop the same skill. "Know
that," in contrast, does not seem to denote the possession of a skill or
aptitude but rather the possession of specific pieces of information, and the
person who has knowledge of this sort can generally convey it to others. To know
that the Concordat of Worms was signed in the year 1122 would be an example of
this sort of knowledge. Ryle has argued that, given these differences, some
cases of knowing how cannot be reduced to cases of knowing
that and, accordingly, that the kinds of knowledge expressed by these
phrases are independent of one another. (see also operational
knowledge)
In general, the philosophical tradition
from the Greeks to the present has focused on the kind of knowledge expressed
when it is said that someone knows that such and such is the case, e.g.,
that A knows that snow is white. This sort of knowledge, called propositional
knowledge, raises the classical epistemological questions about the truth or
falsity of the asserted claim, the evidence for it, and a host of other
problems. Among them is the much debated issue of what kind of thing is known
when one knows that p, i.e., what counts as a
substitution instance of p. The list
of such candidates includes beliefs, propositions, statements, sentences, and
utterances of sentences. Each has or has had its proponents, and the arguments
pro and con are too subtle to be explored here. Two things should, however, be
noted in this connection: first, that the issue is closely related to the
problem of universals (i.e., whether
what is known to be true is an abstract entity, such as a proposition, or
whether it is a linguistic expression, such as a sentence or a sentence-token)
and, second, that it is agreed by all sides that one cannot have knowledge, in
this sense of "knowledge," of that which is not true. One of the
necessary conditions for saying that A knows that p is that p must be true,
and this condition can therefore be regarded as one of the main elements in any
accurate characterization of knowledge.
Mental versus nonmental conceptions of
knowledge. Philosophers have asked whether knowledge is a state of mind, i.e.,
a special kind of awareness of things. That it is has been argued by
philosophers since at least the 5th century BC. In The
Republic Plato provided the first
extensive account of such a view. He regarded knowing as a mental faculty, akin
to but different from believing or opining. Contemporary versions of this sort
of theory regard knowing as one member of a sequence of mental states that
involve increasing certitude. This spectrum would begin with guessing or
conjecturing at the lowest end of certitude, would include thinking, believing,
and feeling sure as expressing stronger attitudes of conviction, and would end
with knowledge as the highest of all these states of mind. Knowledge, in all
views of this type, is a form of consciousness,
the strongest degree of awareness humans possess, and accordingly it is common
for proponents of such views to hold that, if A knows that p, A must be conscious of what he knows. This view is normally
expressed by saying that, if A knows that p,
A knows that he knows that p.
Many 20th-century philosophers have
rejected the notion that knowledge is a mental state. In On Certainty (1969) Ludwig Wittgenstein says: " 'Knowledge' and
certainty belong to different categories.
They are not two mental states like, say surmising and being sure." But, if
knowing is not a mental state, then what is it? These philosophers have accepted
the challenge of trying to give a different characterization of what it means to
say that a person knows something. They typically begin by pointing out that a
person can know that p without knowing
that he knows it (a good example is in fact to be found in Plato's Meno,
where Socrates gradually elicits from a slave boy geometrical knowledge that the
boy was not aware he had). They then proceed to argue that it is a mistake to
assimilate cases of knowing to cases of doubting, feeling a pain, or having a
certain opinion about something. All of these latter are mental states, and they
are such that a person who has such a state is aware that he does.
These philosophers, moreover, typically
deny that knowing can be described as being a single thing, such as a state of
consciousness. Instead, they claim that one can ascribe knowledge to someone, or
to oneself, when certain complex conditions are satisfied, among them certain
behavioral conditions. For example, if a person can always give the right
answers to questions under test conditions, one would be entitled to say that
the person has knowledge of the issues under consideration. Knowing on this
account seems tied to the capacity to perform in certain ways under certain
standard conditions. Accordingly, though such performances may involve the
exercise of intelligence or other mental factors, the attribution of knowledge
to someone is not merely the attribution of a certain mental state or state of
awareness to that person (as seen in the case of the slave boy in the Meno).
A well-known variant of such a view was
advanced by J.L. Austin in his 1946 paper
"Other Minds." Austin claimed that, when one says "I know,"
one is not describing anything, let alone one's psychology or a mental state.
Instead, one is engaging in a social act, i.e.,
one is indicating that one is in the position (has the credentials and the
reasons) to assert p in circumstances
where it is necessary to resolve a doubt. When these conditions are satisfied,
one can correctly be said to know.
A distinction closely related to the
previous one is that between occurrent and dispositional conceptions of
knowledge. The difference between occurrences and dispositions can be
illustrated with respect to sugar. A sugar cube will dissolve if put into water.
One can thus say that, even if the cube is not now dissolving as it sits on the
table, it will do so under certain conditions. This propensity to dissolve is
what is meant by a disposition, and it is a feature sugar has at all times and
in all conditions. It can be contrasted with sugar's actually dissolving when
immersed in liquid, which is an occurrence, that is, an event happening at a
specific place and at a specific time.
These terms also apply to mental events.
One can say of Smith, who is working on a problem, that he has just seen the
solution. According to this way of speaking, there is a certain answer that
Smith is presently aware of and to which he is attending. In such a case Smith's
knowledge is occurrent. But one can also ascribe a different sort of knowledge
to Smith. Though Smith is perhaps not now thinking of his home address, he
certainly knows it in the sense that, if he were asked, he could produce the
correct answer. One can thus have knowledge that one is not aware of at a given
moment. One can thus say, as with sugar, that knowledge may be either occurrent
or dispositional in character, i.e., that one may or may not be in an immediate state of
self-awareness with respect to p, but
that in either case it can be said that the person knows that p.
It should be noted that the distinction
between dispositional and occurrent knowledge thus applies to cases of
"knowing that" as well as to cases of "knowing how" and thus
is a powerful conceptual tool for analyzing different sorts of epistemic
notions. The concept of a disposition has itself been further analyzed, for
example by Roderick M. Chisholm (b. 1916), in counterfactual terms, and it has
been proposed by many philosophers that the knowledge expressed by causal laws
(laws of nature) is counterfactual and thus dispositional in character.
A sharp distinction has been drawn since
at least the 17th century between two types of knowledge: a
priori knowledge and a posteriori knowledge.
The distinction plays an especially important role in the philosophies of David
Hume (1711-76) and Immanuel Kant (1724-1804). It is also found in many
contemporary, empirically oriented theories of knowledge, which typically hold
that all knowledge about matters of fact derives from experience and is
therefore a posteriori and that in consequence such knowledge is never certain
but at most only probable.
The difference between these types of
knowledge is easy to illustrate by means of examples. Consider the sentences
"All husbands are married" and "All Model-T Fords are black"
and assume that both statements are true. But how does one come to know that
they are true? In the case of the first, the answer is that, if one thinks about
the meaning of the various words in the sentence, one can see that the sentence
is true. One can see that this is so because what is meant by
"husband" is the same as what is meant by "married male."
Thus, by definition, every husband is a married male, and, accordingly, every
husband is married. In calling such knowledge a priori, philosophers are
pointing out that one does not have to engage in a factual or empirical inquiry
in order to determine whether the sentence is true or not. One can know this
merely on the basis of reflection and thus prior to or before any investigation
of the facts.
In contrast, the second statement can be
determined to be true only after such an investigation. One may well know that
the Model-T Ford was an automobile built prior to World War II and accordingly
would understand what all the words in the sentence mean. Nonetheless,
understanding alone would not be sufficient to allow one to determine whether
the sentence is true or not. Instead, some kind of empirical investigation is
required in order to arrive at such a judgment; the knowledge thus acquired is a
posteriori, or knowledge after the fact.
There are sets of distinctions related
to the one just developed and in terms of which the two propositions
can also be differentiated. They are necessary versus contingent, analytic
versus synthetic, tautological versus significant, and logical versus factual.
A proposition is said to be necessary
if it holds (is true) under all possible circumstances or conditions. "All
husbands are married" is such a proposition. There are no possible or
conceivable conditions under which this statement would not be true (on the
assumption, of course, that the words "husband" and
"married" are taken to mean what they ordinarily mean). In contrast,
"All Model-T Fords are black" holds in some circumstances (those
actually obtaining, and that is why the proposition is true), but it is easy to
imagine circumstances in which it would not be true--for instance, if somebody
painted one of those cars a different colour. To say, therefore, that a
proposition is contingent is to say that it
holds in some but not in all possible circumstances. Some necessary
propositions, such as "All husbands are married" are a priori (though
not all are) and most contingent propositions are a posteriori.
A proposition is often said to be
analytic if the meaning of the predicate term is contained in the meaning of the
subject term. Thus, "All husbands are married" is analytic because the
term "husband" includes as part of its meaning "being
married." A term is said to be synthetic if this is not so. Therefore,
"All Model-T Fords are black" is synthetic since the term
"black" is not included in the meaning of "Model-T Ford."
Some analytic propositions are a priori, and most synthetic
propositions are a posteriori. These distinctions were used by Kant to
ask one of the most important questions in the history of epistemology, namely,
whether a priori synthetic judgments are possible (see below for a discussion of
this question). (see also analytic
proposition )
A proposition is said to be tautological
if its constituent terms repeat themselves or if they can be reduced to terms
that do, so that the proposition is of the form "a = a." In such a
case the proposition is said to be trivial and empty of cognitive import. A
proposition is said to be significant if its constituent terms are such that the
proposition does provide new information about the world. It is generally agreed
that no significant propositions can be derived from tautologies. One of the
objections to the ontological argument is that
no existential (significant) proposition can be derived from the tautological
definition of "God" with which the argument begins. Tautologies are
generally known to be true a priori, are necessary, and are analytic; and
significant statements are generally a posteriori, contingent, and synthetic.
(see also meaning)
In the ontological argument, for
example, God is defined (roughly speaking) as the only perfect being. It is then
argued that no being can be perfect unless it exists; therefore, God exists.
But, as Hume and Kant pointed out, it is fallacious to derive a factual
statement about the existence of God from the definition of God as a perfect
being (see the discussion of St. Anselm below).
The term "logical" in this
connection is used in a wide sense to include a proposition such as "All
husbands are married." By analyzing the meaning of its constituent terms
one can reduce the proposition to a logical truth,
e.g., to "A and B implies
A." In contrast, factual propositions, such as "All Model-T Fords are
black," have syntactical and semantic structures that differentiate them
from any propositions belonging to logic, even in the broad sense mentioned
above. The theorems of logic are often a priori (though not always), are always
necessary, and are typically analytic. Factual propositions are generally a
posteriori, contingent, and synthetic.
These various distinctions are widely
appealed to in present-day philosophy. For instance, Saul
Kripke (b. 1941) in "Naming and Necessity" (1972) has used
these notions in an effort to solve a long-standing problem, namely, how true
identity statements can be nontrivial. The problem, first articulated by Gottlob
Frege in "On Sense and Reference" (1893) and later
independently addressed by Russell, begins with the assumption that the
sentences "Scott is Scott" and "Scott is the author of Waverley"
are both identity sentences and are true and that the former is trivial while
the latter is not. The puzzle arises from the further assumption that any true
identity sentence simply says of some object that it is identical with itself.
Hence, all such sentences should be trivial. Clearly, however, "Scott is
the author of Waverley" is not
trivial. But, if it is not, how is this possible?
Kripke argues that all true identity
sentences are necessary (i.e., that
they hold in all possible worlds) and that some of these, such as "Scott is
Scott," are known a priori and accordingly are trivial; but, he argues,
some true identity sentences are not known a priori but only a posteriori and
are not trivial. In cases of the latter sort, their nontriviality is a function
of their being known to be true only after some sort of inquiry or
investigation. It is the investigation that provides new information.
A good example would be the following.
At one time in human history, ancient peoples did not know that what they called
"the evening star" was the same planet called "the morning
star." But eventually the Babylonians discovered through astronomical
observation that the morning star is the planet Venus as it appears in the
morning sky and that the evening star is the planet Venus as it appears in the
evening sky. The discovery that these two appearances are appearances of the
same object amounted to discovering more than that Venus is Venus. It provided
new information, and that is why "the morning star is identical with the
evening star" is significant in a way in which "Venus is Venus"
is not, even though all of the descriptive terms in both sentences refer to
exactly the same object. In similar fashion, the a posteriori finding that it
was Scott who wrote Waverley explains
the nontriviality of "Scott is the author of Waverley." But no such investigation was needed to determine
that Scott is Scott.
The distinction between knowledge by
acquaintance and knowledge by description was introduced by Bertrand Russell in
connection with his celebrated theory of descriptions. Here only the
epistemological (as distinct from the logical) version of his theory will be
considered. It was invented by Russell to lend support to the basic thesis of
empiricism that all knowledge of matters of fact (i.e., all a posteriori knowledge) derives from experience.
Russell's program is both reductive and foundationalist. It tries to show that
man's system of knowledge is stratified: that some types of knowledge depend on
others but that some do not and that the latter form the foundational units
which give support to the whole epistemic system. He argued that, because these
basic units rest upon direct experience, ultimately all factual knowledge is
derivable from experience.
Russell's argument begins with a
distinction between two different types of knowledge, that which is and that
which is not based on direct experience. Nearly all of man's knowledge is of the
latter type. For example, it is known that some 2,000 years ago there lived a
Roman statesman named Augustus, that he was the successor to Julius Caesar, who
had been assassinated, and that he was a friend of the historian Livy. But,
since none of these pieces of information is presently known on the basis of
personal experience, what justification is there for calling them instances of
knowledge?
Russell argued that information based on
direct experience is basic and needs no justification; he called it
"knowledge by acquaintance." Information not based on direct
experience he called "knowledge by description." One is justified in
calling such information knowledge, if one can show that it can be traced back
to and thus ultimately rests upon knowledge by acquaintance. To show how this is
so in a particular case is to legitimate that particular piece of information as
a specimen of knowing. Here is how this reductive process would work in the case
of what is known about Augustus.
Whatever information people in the 20th
century have about Augustus probably comes to them from literary works, such as
Livy's history of Rome. Such information thus comes secondhand, via descriptions
in books about the life and activities of Augustus. But why call such
descriptions knowledge? The answer is that through a historical process one can
trace such information back to an original source like Livy, who was a
contemporary of Augustus. One learns, via this process, that Livy in his history
of Rome is reporting events that he had witnessed himself or that he had learned
from other eyewitnesses. One can call what he tells about Augustus knowledge,
because it is testimony that is based upon his or someone else's direct
experience. Thus, knowledge by description is a legitimate form of knowledge,
even though it is ultimately dependent upon knowledge by acquaintance.
Russell's reductive thesis then was that
all legitimate specimens of knowledge are either based upon direct experience or
can be shown to be dependent upon such direct experience via a chain of tight
historical or causal links. His theory was therefore a form of empiricism,
because it tried to show how all knowledge of matters of fact could be derived
from experience.
But there is a further feature of the
theory, stemming from the empirical tradition of John Locke (1632-1704) and David
Hume, that gives a special twist to the notion of "knowledge by
acquaintance." According to this tradition, knowledge by acquaintance is
always knowledge based upon what Hume called "impressions," or upon
what Russell called "sense-data." These for Russell were mental
entities that generally, but not always, reflected the characteristics actually
possessed by physical objects. But, unlike physical
objects, sense-data were the objects directly apprehended in an act of
perception. What Russell meant by "direct apprehension" or
"direct perception" was itself explicated in terms of the concepts of inference
and non-inference. He held that direct perception, i.e., the perceptual awareness of a sense-datum,
involves no inference and, accordingly, that knowledge by acquaintance is
identical with the perception of sense-data.
The difference between inferential and
noninferential perception can be illustrated by an example. Suppose one is
working in a room and hears a sound that emanates from an outside source.
(Russell considered hearing to be a form of perception.) In such a case the
sound is a sense-datum. One need not infer that one is hearing a sound; there is
a direct awareness of it. This would be a case of knowledge by acquaintance. On
the basis of what one hears in this direct fashion, one might then infer (guess,
conjecture, hypothesize) that what is causing the sound is a motorcycle located
outside of the room, something that one who is in the room cannot see directly.
If one is correct in this supposition, the information obtained in this way
would be a case of indirect knowledge. In such a case, one's knowledge that
there is a motorcycle in the street is dependent on (and in Russell's sense,
reducible to) one's direct awareness of a sound. The example illustrates how
indirect knowledge, such as knowledge by description, is derived from direct
knowledge, such as knowledge by acquaintance, and in turn how this latter
depends upon the direct awareness of sense-data.
It should be mentioned that the
distinction between knowledge by acquaintance and knowledge by description can
be defended as legitimate and useful independently of a commitment to sense-data
theory. In Russell's work the objects of direct awareness are sense-data, but
sense-data theory today has few proponents. A philosopher thus might hold that
one at least sometimes directly perceives physical objects (which are not
sense-data) while accepting that one's knowledge of past events and persons is
indirect and is thus knowledge by description.
Epistemology during its long history has
engaged in two different sorts of tasks. One of these is descriptive in
character. It aims to depict accurately certain features of the world, including
the contents of the human mind, and to determine whether these should count as
specimens of knowledge. A philosophical system with this orientation is, for
example, the phenomenology of Edmund Husserl
(1859-1938). Husserl's aim was to give an exact description of the notion of
intentionality, which he characterized as consisting of a certain kind of
"directedness" toward an object. Suppose the object is an ambiguous
drawing, such as the duck/rabbit sketch found in the Philosophical Investigations of Ludwig Wittgenstein. A person
looking at the sketch is not sure whether it is a drawing of a duck or of a
rabbit. Husserl claimed that the light rays reaching the eye from such an
ambiguous drawing are identical whether one sees the image of a duck or of a
rabbit and that the difference in perception is due to the viewer's structuring
of what he sees in the two cases. The theory tries to describe how such
structuring takes place, and it ultimately becomes very complex in the account
it gives.
In a famous passage in Philosophical Investigations (1953), Wittgenstein states that
"explanation must be replaced by description," and much of his work
was devoted to carrying out that task, as, for example, in his account of what
it is to follow a rule. Another example of descriptive epistemology is found in
the writings of such sense-data theorists as Moore, Price, and Russell. They
begin with the question of whether there are basic apprehensions of the world,
free from any form of inference, and in those cases where they have argued that
the answer is yes, they have tried to describe what these are and why they
should count as instances of knowledge. Russell's thesis that the whole edifice
of knowledge is built up from a foundation composed of ingredients with which
human beings are directly acquainted illustrates the close connection between
the attempt to characterize various types of knowledge and this descriptive
endeavour. The search by some logical positivists, such as Moritz Schlick
(1882-1936), Otto Neurath (1882-1945), and A.J. Ayer (b. 1910) for protocol
sentences, sentences that describe what is given in experience without
inference, is a closely related example of this kind of descriptive practice.
Epistemology has a second function,
which, in contrast to the descriptive one, is justificatory or normative.
Philosophers concerned with this function start from the fact that all human
beings have beliefs about the world, some of
which are erroneous and some of which are not. The question to them is how one
can justify (defend, support, or provide evidence for) certain sets of beliefs.
The question has a normative import since it asks, in effect, what one ideally
ought to believe. (In this respect epistemology has close parallels to ethics,
where normative questions about how one ought ideally to act are asked.) This
approach quickly takes one into the central domains of epistemology. It raises
such questions as: Is knowledge identical with justified true belief? Is the
relationship between evidence for a belief and the belief itself a probability
function? If not, what is it? What indeed is meant by "justification"
and what sorts of conditions have to be satisfied before one is entitled to say
that a belief or set of beliefs is justified? These two differing aspects of
epistemology are not inconsistent and indeed are often found intertwined in the
writings of contemporary philosophers.
The relationship between knowledge and certainty
is complex, and there is considerable disagreement about the matter. Are these
concepts the same? If not, how do they differ? Is it possible for someone to
know that p without being certain that p?
Is it possible for someone to be certain that p without knowing that p?
These are the central issues around which the debate revolves. The various
answers that have been proffered depend on how the concepts of knowledge and
certainty are analyzed. If one holds, for instance, that knowing is not a
psychological state but that certainty is, then one would deny that the concepts
are identical. But if one holds that knowing represents the highest degree of
assurance which humans can obtain with respect to the truth of p,
and that such a maximal degree of assurance is a psychological state, one will
interpret the concepts to be equivalent. There have been proponents on both
sides of this issue.
Further complicating the discussion are
subtle distinctions drawn by 20th-century philosophers. For instance, in
"Certainty" (1941) G.E. Moore claimed
that there are four main types of idioms in which the word "certain"
is commonly used: "I feel certain that," "I am certain
that," "I know for certain that," and finally "It is
certain that." He points out that "I feel certain that p"
may be true when p is not true but
that there is at least one use of "I know for certain that p"
and "It is certain that p"
which is such that neither of these sentences can be true unless p
is true. Moore argues that it would be self-contradictory to say "I knew
for certain that he would come but he didn't," whereas it would not be
self-contradictory to say "I felt certain he would come but he
didn't." In the former case, the fact that he did not come proves that one
did not know that he would come, but, in the latter, the fact that he did not
come does not prove that one did not feel certain he would. "I am certain
that" differs from "I know for certain that" in allowing the
substitution of the word "sure" for the word "certain." One
can say "I feel sure (rather than certain)" without a change of
meaning, whereas in "I know for certain" or "It is certain
that" this substitution is not possible. On the basis of these sorts of
considerations Moore contends that "a thing can't be certain unless it is known."
He states that this is what distinguishes the word "certain" from the
word "true." A thing that nobody knows may well be true, but it cannot
possibly be certain. He thus infers that a necessary condition for the truth of
"It is certain that p" is
that somebody should know that p is
true. Moore is therefore one of the philosophers who answers in the negative the
question of whether it is possible for p
to be certain without being known.
Moore also argues that to say
"Someone knows that p is
true" cannot be a sufficient condition for "It is certain that p." If it were, it would follow that, in any case in which at
least someone did know that p was
true, it would always be false for anyone to say "It is not certain that p";
but clearly this is not so. If one person says that it is not certain that Smith
is still alive, he is not thereby committing himself to the statement that
nobody knows that Smith is still alive: the speaker's statement is consistent
with Smith's still being alive, and both he himself and other persons know this.
Moore is thus among those philosophers who would answer in the negative the
question of whether the concepts of knowledge and certainty are the same. Though
it is widely accepted that to affirm that somebody knows that p
implies that somebody is certain that p,
the case of the slave boy in Plato's Meno
seems, at least at first glance, to be a counterinstance. Meno may know, in a
dispositional sense, certain theorems of geometry without knowing that he knows,
and, if he does not know that he knows, then it would seem that he cannot be
certain that he does know. But it has also been argued that, once his
disposition to know has been actualized and his knowledge has become occurrent,
then, insofar as he does know in this occurrent sense, he is certain of what he
knows.
The most radical position on these
matters is to be found in Wittgenstein's On
Certainty, published posthumously in 1969. Wittgenstein holds that knowledge
is radically different from certitude and that neither concept entails the
other. It is thus possible to be in a state of knowledge without being certain
and to be certain without having knowledge. As he writes: "Instead of 'I
know' . . . couldn't Moore have said: 'It stands fast for me that . . .' ? and
further: 'It stands fast for me and many others. . . .' " "Standing
fast" is one of the terms Wittgenstein uses for certitude and is to be
distinguished from knowing. For him certainty is to be identified with acting,
not with seeing propositions to be true, the kind of seeing that issues in
knowledge. As he says: "Giving grounds, justifying the evidence comes to an
end--but the end is not certain propositions striking us immediately as true--i.e., it is not a kind of seeing
on our part; it is our acting which
lies at the bottom of the language game."
Philosophers not only wish to know what
knowledge is but also how it originates. This motivation is based, at least in
part, on the supposition that an investigation into the provenance of knowledge
can help cast light on its nature. From the time of the Greeks to the present,
therefore, one of the major themes of epistemology has been a quest into the
sources of knowledge.
Plato's The
Republic contains one of the earliest systematic arguments to the effect
that sense experience cannot be a source of knowledge. The argument begins with
the assertion that ordinary persons have a clear grasp of certain concepts, that
of equality, for instance. In other words, people know what it means to say that
A and B are equal, no matter what A and B are. But where does such knowledge
come from? One may wonder, for instance, whether it is provided by vision and
consider the claim that two pieces of wood are of equal length. A close
inspection of these pieces of wood, however, shows them to differ slightly, and
the more detailed the inspection, via various degrees of magnification, the more
disparity one notices. It follows that visual experience cannot be the fount of
the concept of equality. Plato applies this result to the operations of all the
five senses and concludes that sense experience in general cannot be the origin
of such knowledge. It must therefore have another source, which he regards as
prenatal (one such account is found in the myth of Er in Book X).
The mathematical
example Plato selects to illustrate that the origin of knowledge is not in sense
experience is highly significant; indeed it is one of the signs of his
perspicacity that he should pick such an example. For, as the subsequent history
of philosophy reveals, the strongest case for the notion that at least some
knowledge does not derive from sense experience lies in mathematics.
Mathematical entities are abstractions--perfect triangles, disembodied surfaces
and edges, lines without thickness, and extensionless points--and none of these
exists in the physical world, i.e.,
the world apprehended by the senses. It might be thought that, had Plato
selected a different example, say, the colour red, his argument would have been
less convincing. But it is a further sign of his genius that he discusses
colours as well as mathematical notions and provides good reasons for holding
that seeing examples or specimens of red (or any other colour) is not equivalent
to knowing what that colour is. Such knowledge must therefore have a different
genesis than sense experience.
The puzzle about origins of knowledge
has led historically to two different kinds of issues. One of these is the
question of whether knowledge (or at least certain kinds of knowledge) is
innate, meaning that it is not acquired or learned through experience but in
some important sense is present in the human psyche at birth. The matter is
still a live issue today, not only in philosophy but also in linguistics and
psychology. The linguist Noam Chomsky (b. 1928),
for example, has asserted that the "projection phenomenon"--the
ability of children to construct sentences that they have never heard before and
that are grammatical--is proof of inherent conceptual structures, whereas the
experimental psychologist B.F. Skinner (b. 1904)
has tried to show that all knowledge is the product of learning
through environmental conditioning by means of the processes of reinforcement
and reward. (see also innate
idea)
In the extensive historical literature
on this topic both the notion of "innateness" and that of
"learning" have been given various interpretations. Sometimes, for
instance, innateness carries only the sense of a disposition or propensity, but
in stronger versions of the thesis, such as Plato's, it is affirmed that humans
possess actual pieces of prenatal knowledge. "Learning" also is given
a variety of meanings, ranging from trial-and-error methods to inexplicit types
of "absorption" of information. There are also a range of
"compromise" theories. These typically claim that humans have some
knowledge that is innate--the awareness of God, the principles of moral
rightness and wrongness, and certain mathematical theorems being favoured
examples--whereas other kinds of knowledge--such as knowledge by
acquaintance--are gained through experience.
The second issue that emerges from
considerations of the origins of knowledge focuses on the distinction between
rationalism and empiricism. Though closely related to the issue of innateness
versus learning, the question in this case concerns the nature of the source
from which knowledge arises. The history of discussion of the issue indicates
that two main sources have been identified and argued for: reason and
experience.
Rationalism is the thesis that the
ultimate source of knowledge is to be found in human reason. What reason is, in
turn, is a difficult question. But, generally speaking, it is assumed that
reason is a feature of the human mind that differs not just in degree but in
kind from bodily sensations, feelings, and certain psychological attitudes, such
as disgust or enthusiasm. For some writers, such as Plato, reason is a faculty,
a special facility or structure of the mind. Many later philosophers reject any
sort of faculty psychology, and some of them tend to interpret reason in
dispositional or behavioral ways. But, whatever the interpretation, a
rationalist must hold that reason has a special power for grasping reality. It
is the exercise of reason that allows human beings to understand the world they
live in. Such a thesis is double-sided: it holds, on the one hand, that reality
is in principle knowable and, on the other hand, that there are human,
distinctively mental, powers capable of apprehending it. One thus might define
rationalism as the theory that there is an isomorphism (a mirroring
relationship) between reason and reality which makes it possible for the former
to apprehend the latter just as it is. Rationalists affirm that, if such a
correspondence were lacking, the effort of human intelligence to understand the
world would be impossible.
Empiricism is often defined as the
doctrine that all knowledge comes from experience. Almost no philosopher,
however, has ever literally held that all knowledge comes from experience. Locke,
who is the empiricist par excellence, thought there is some knowledge human
beings have--which he calls "trifling ideas" (or trivialities), such
as a = a--that
does not derive from experience; but he regarded such knowledge as empty of
content. Hume held similar views.
Empiricism thus generally allows for a
priori knowledge while denigrating its significance, and accordingly it is more
accurate to define it as the theory that all knowledge about matters of fact
derives from experience. When defined in this way empiricism does represent a
significant contrast to rationalism. Rationalists hold that human beings have
knowledge about matters of fact which is anterior to experience and yet which
does tell them something significant about the world and its various features.
Empiricists would deny that this is possible.
The meaning of the term experience is
generally limited to the impressions and sensations received by the senses.
Thus, knowledge is the information apprehended by the five sense
modalities--hearing, seeing, touching, tasting, and smelling. Such knowledge is
always about matters of fact, about what one can see, touch, hear, taste, or
smell. For strict empiricists this definition has the implication that the human
mind is passive -- a tabula rasa, in Locke's
idiom; it is an organ that receives impressions and more or less records them as
they are. This conception of the mind has seemed counterintuitive to many
philosophers, especially those in the Kantian tradition. But it also poses
serious challenges for empiricists. For example, it raises the question of how
one can have knowledge of items, such as a dragon, that cannot be found in
experience.
In response, the classical empiricists
such as Locke and Hume have tried to show how the complex concept of a dragon
can be reduced to simple concepts (such as wings, the body of a snake, the head
of a horse), all of which derive from direct impressions of such items. On such
a view the mind is still considered to be primarily passive, but it is conceded
that it has some active functions, such as being able to combine simple
impressions and ideas into complex ideas. (see
also simple
idea)
There are further difficulties: the
empiricist must explain how abstract ideas, such as the concept of a perfect
triangle, can be reduced to elements apprehended by the senses when no perfect
triangles are actually found in nature, and he must also give an account of how
general notions are possible. It is obvious that one does not experience
"mankind," but only particular individuals, through the senses; yet
such general notions are meaningful, and propositions containing these concepts
are known to be true. The same difficulty applies to colour concepts. Some
empiricists have argued that one arrives at the concept of red, for example, by abstracting
from individual items that are red. But the difficulty with this suggestion is
that one would not know what to count as an instance of red unless one already
had such a concept in mind; and, if that is so, it would seem that experience
cannot be the source of the concept. It is generally
felt that, despite ingenious attempts by empiricists to deal with such issues,
their solutions have not been wholly successful. Indeed, the history of
epistemology has to a large extent been a dialectic between rationalism and
empiricism in an effort to meet skeptical challenges that are designed to
undermine both positions.
Many philosophers past and present and
many nonphilosophers who are studying philosophy for the first time have been
struck by the seemingly indecisive nature of philosophical argumentation. For
every argument, there seems to be a
counterargument; and for every position, a counterposition. To a considerable
extent skepticism is born of such reflection. Some of the ancient skeptics
contended, for example, that all arguments are equally bad and, accordingly,
that nothing can be proved. The American philosopher Benson Mates claims to be a
modern representative of this tradition, except that he believes all
philosophical arguments to be equally good. But he insists that, because they
are, they invariably issue in conceptual deadlocks and resolve nothing.
Ironically, skepticism is itself a type
of philosophy, and the question has been raised whether it manages to escape its
own demurrers. Does it offer arguments, and, if so, are they decisive? The
answers to these questions depend on what is meant by skepticism. Historically,
the term refers to a complex set of practices taking many different forms--from
stating explicit theories to assuming negative attitudes without much
propositional content. Thus, it is difficult to define. But, however it is
understood, skepticism represents a set of challenges to the claim that human
beings do possess or can acquire knowledge.
In giving even this minimal
characterization, it is important to emphasize that both dogmatists and skeptics
accept a definition of knowledge that implies two things: that, if a person, A,
knows that p, then p
is true and that, if a person, A, knows that p,
then A cannot be mistaken, meaning that it is logically impossible that A could
be wrong. If a person says that he knows Smith will arrive at 9:00 AM, and Smith
is not there at 9:00 AM, then that person would have to withdraw the claim to
know. He might say instead that he thought he knew or that he felt sure. But he
could not continue rationally to insist that he knew if what he claimed to know
turned out to be false. (see also propositional
knowledge)
It should also be stressed that, given
this definition of knowledge, the skeptic does not have to show that A is
actually mistaken in claiming to know that p.
All he has to show is that it is possible that A might be mistaken. Hence arises
the skeptic's practice of searching for a possible counterexample to a claim. If
A states that he has had a certain experience, for instance, that of having
personally spoken with Smith, who assured him he would keep his appointment at
9:00 AM, then the skeptic can point out that, although one could have such an
experience, it is still possible that Smith might not show up; and, if so, A's
claim to know is untenable. In effect, by emphasizing the notion of possibility,
the skeptic is pointing out that there is a logical gap between the criteria
that support the claim and the claim itself. The criteria might be satisfied,
and yet the claim might be false; but, if such a possibility exists, the
original assertion cannot be a specimen of knowing.
More generally, radical skepticism has
tried to show that one might (i.e.,
could possibly) have all the experiences associated with normal perception or
behaviour and yet be wholly mistaken in thinking that these experiences
correlate with anything in the external world. For example, a brain in a vat
might be programmed by scientists to have the sensation of seeing a tree, even
though it is not in fact seeing a tree. Thus, there is a gap between the
experience the brain is having and external reality; accordingly, its claim to
know on the basis of such a visual experience is mistaken. The skeptic's point
is that the disparity between external reality and felt experience is always
possible and, accordingly, that knowledge claims based upon such experience
cannot be defended.
The ability to find counterexamples
explains why skeptics do not challenge but indeed accept the dogmatist's
definition of knowledge. That they do so is important because it means that they
are not arguing at cross-purposes with their opponents. What they challenge is
not the meaning of knowledge but the contention that anybody actually has
knowledge in that sense.
Nearly all of the major epistemological
theories of philosophy have given rise to skeptical reactions. Many of the
greatest thinkers in the Western tradition have assumed that by means of reason
or sense experience one can come to have knowledge of reality. But skepticism
has challenged the validity of both of these appeals. Skeptics have developed
wholesale arguments to undermine the efforts to show that reason and sense
experience, which seem to be the only possible candidates, are reliable sources
of knowledge. Descartes, for example, considered
the hypothesis that an evil genius may delude people into thinking that they are
experiencing the real world when they are not. With regard to major
epistemological problems, such as the "other-minds problem," the
problem of memory, the problem of induction, and the problem of self-knowledge,
skeptical doubts have challenged the validity of reason and of sense experience
and thus of claims to have knowledge of various aspects of reality. How some of
these moves and countermoves actually take place are addressed below. (Av.S.)
The central focus of ancient Greek
philosophy was its attempt to solve the problem of motion. Many
pre-Socratic philosophers thought that no logically coherent account of motion
and change could be given. This problem was a
concern of metaphysics, not epistemology, however, and in the present context it
suffices merely to allude to the arguments of Parmenides
and Zeno of Elea against the possibility that anything moves or changes. The
consequence of this position for epistemology was that all major Greek
philosophers held that knowledge must not itself change or be changeable in any
respect. This requirement motivated Parmenides, for example, to hold that
thinking is identical with being (what exists or is unchanging) and that it is
impossible to think of "nonbeing" or "becoming" (what
changes) in any way.
Plato (c.
427-347 BC) accepted the Parmenidean constraint on any theory of knowledge that
both knowledge and its objects must be unchanging. One consequence of this, as
Plato pointed out in Theaetetus, is
that knowledge cannot have physical reality as its object. In particular, since
sensation and perception have various kinds of motions
as their objects, knowledge cannot be the same as sensation or perception. The
negative thesis of Plato's epistemology consists, then, in the denial that sense
experience can be a source of knowledge on the ground that the objects
apprehended through the senses are subject to change. To the extent that humans
have knowledge, they attain it by transcending the information provided by the
senses in order to discover unchanging objects. But this can be done only by the
exercise of reason, and in particular by the application of the dialectical
method of inquiry inherited from Socrates.
The Platonic theory of knowledge is thus
divided into two parts: a quest first to discover whether there are any
unchanging objects and to identify and describe them and second to illustrate
how they could be known by the use of reason, that is, via the dialectical
method. Plato used various literary devices for illustrating his theory; the
most famous of these is the allegory of the cave
in Book VII of The Republic. The
allegory depicts ordinary people as living locked in a cave, which represents
the world of sense-experience; in the cave people see only unreal objects,
shadows, or images. But through a painful process, which involves the rejection
and overcoming of the familiar sensible world, they begin an ascent out of the
cave into reality; this process is the analogue of the application of the
dialectical method, which allows one to apprehend unchanging objects and thus
acquire knowledge. In the allegory, this upward process, which not everyone is
competent to engage in, culminates in the direct vision of the sun, which
represents the source of knowledge.
In searching for unchanging objects,
Plato begins his quest by pointing out that every faculty in the human mind
apprehends a set of unique objects: hearing apprehends sounds but not odours;
the sense of smell apprehends odours but not visual images; and so forth.
Knowing is also a mental faculty, and therefore there must be objects that it
apprehends. These have to be unchanging, whatever they are. Plato's discovery is
that there are such entities. Roughly, they are the items denoted by predicate
terms in language: such words as "good," "white," or
"triangle." To say "This is a triangle" is to attribute a
certain property, that of being a triangle, to a certain spatiotemporal object,
such as a particular figure drawn on a blackboard. Plato is here distinguishing
between specific triangles that can be drawn, sketched, or painted and the
common property they share, that of being triangular. Objects of the former kind
he calls particulars. They are always located
somewhere in the space-time order, that is, in the world of appearance. But such
particular things are different from the common property they share. That is, if
x is a triangle, and y
is a triangle, and z is a triangle, x, y, and z are
particulars that share a common property, triangularity. That common property is
what Plato calls a "form" or "idea"
(not using this latter term in any psychological sense). Unlike particulars,
forms do not exist in the space-time order. Moreover, they do not change. They
are thus the objects that one must apprehend in order to acquire knowledge.
Similar remarks apply, for example, to
goodness, whiteness, or being to the right of. Particular things change; they
come into and go out of existence. But whiteness never changes, and neither does
triangularity; and, if they do not change, they are not subject to the ravages
of time. In that sense, they are eternal.
The use of reason for discovering
unchanging forms is exercised in the dialectical method. The method is one of
question and answer, designed to elicit a real
definition. By a "real definition" is meant a set of necessary
and sufficient conditions that exactly delimit a concept.
One may, for example, consider the concept of being the brother of Y. This can
be explained in terms of the concepts of being male and of being a sibling of Y.
These concepts together lay down necessary and sufficient conditions for
anything's being a brother. One who grasps these conditions understands
precisely what it is to be a brother.
The
Republic begins with the use of the dialectical
method to discover what justice is. Cephalus proposes the thesis that
"justice" means the same as "honesty in word and deed."
Socrates searches for and finds a counterexample to this proposal. It is just,
he points out, under some conditions, not to tell the truth or to repay debts.
If one had borrowed a weapon from an insane person, who then demanded it back in
order to kill an innocent person, it would be just to lie to him, stating that
one no longer had the weapon. Therefore, "justice" cannot mean the
same as "honesty in word" (i.e., telling the truth). By this technique of proposing one
definition after another and subjecting each to possible counterexamples,
Socrates attempts to find a definition that would be immune to counterexamples.
To find such a definition would be to define the concept of justice, and in this
way to discover the true nature of justice. In such a case one would be
apprehending a form, the common feature that all just things share.
Plato's search for definitions and
thereby the nature of forms is a search for knowledge. But how should knowledge
in general be defined? In Theaetetus
Plato argues that it involves true belief. No one can know what is false. A
person may mistakenly believe that he knows something, which is in fact false,
but this is only thinking that one knows, not knowing. Thus, a person may
confidently assert, "I know that Columbus was the first European to land in
North America" and be unaware that other Europeans, including Erik the Red,
preceded Columbus. So knowledge is at least true belief, but it must also be
something more. Suppose that someone believes there will be an earthquake in
September because of a dream he had in April and that there in fact is an
earthquake in September, although there is no connection between the dream and
the earthquake. That person has a true belief about the earthquake but not
knowledge. What the person lacks is a good reason supporting his true belief. In
a word, the person lacks justification for it.
Thus, in Theaetetus, Plato concludes that knowledge is justified true belief.
Although it is difficult to explain what
justification is, most philosophers accepted the Platonic
analysis of knowledge as fundamentally correct until 1963, when the American
philosopher Edmund L. Gettier produced a counterexample that shook the
foundations of epistemology: suppose that Kathy knows Oscar very well and that
Oscar is behind her, out of sight, walking across the mall. Further, suppose
that in front of her she sees walking toward her someone who looks exactly like
Oscar; unbeknownst to her, it is Oscar's twin brother. Kathy forms the belief
that Oscar is walking across the mall. Her belief is true, because he is walking
across the mall (though she does not see him doing it). And her true belief
seems to be justified, because she formed it on the same basis she would have if
she had actually seen Oscar walking across the mall. Nonetheless, Kathy does not
know that Oscar is walking across the mall, because the justification for her
true belief is not the right kind. What her true belief lacks is an appropriate
causal connection to its object.
In Posterior
Analytics, Aristotle (384-22 BC)
analyzes scientific knowledge in terms of necessary propositions that express
causal relations. Such knowledge takes the form of categorical
syllogisms, in which the middle term causally and necessarily connects
the major and minor terms. For example, because all stars are distant and all
distant objects twinkle, it follows that all stars twinkle. That is, the middle
term, "distant objects," connects the minor term, "stars,"
to the major term, "twinkle," in order to yield the conclusion that
all stars twinkle. Aristotle, however, recognizes that not all knowledge is
provable. Thus, the premises of the most basic syllogisms are known but not
provable. In contrast with scientific knowledge, there is opinion, which is not
provable and is about what happens to be true but need not be. (see also science)
Since the knowledge formulated in
syllogisms resides in the mind, which is part of or one faculty of the soul,
much of what Aristotle says about knowledge is part of his doctrine about the
nature of soul and, in particular, human soul.
As he uses the term, every living thing, including plant life, has a soul (psyche),
a soul being what makes a thing alive. Thus it is important not to equate soul
with mind or intellect. The intellect (nous)
might variously be described as a power, faculty, part, or aspect of the human
soul. It should be stressed that for Aristotle the terms soul (psyche)
and intellect (nous) and its
constituents were understood to be scientific terms.
Knowledge is something that a person
has. Thus it must be in him somewhere, and the location must be his mind or
intellect. Yet there can be no knowledge if the knower and the thing known are
wholly separate. What then is the relation between the knowledge in the person
or his mind and the object of his knowledge? Aristotle's answer is one of his
most enigmatic claims. He says, "Actual knowledge is identical with its
object."
Here is one suggestion about what
Aristotle means. When a person learns something, he acquires something. What he
acquires must either be something different from the thing he knows or identical
with it. If it is something different, then there is a discrepancy between what
he has in mind and the intended object of his knowledge. But such a discrepancy
seems to be incompatible with the existence of knowledge. For knowledge, which
must be true and accurate, cannot deviate from its object in any way. One cannot
know that blue is a colour if the object of that knowledge is something other
than that blue is a colour. This idea that knowledge is identical with its
object is dimly reflected in the repetition of the variable p
in the standard formula about knowledge: S knows that p
just in case it is true that p.
Although the line of thinking being attributed to Aristotle is defective in
several ways, something like it seems to have motivated Aristotle and many other
thinkers over the centuries.
To assert that knowledge and its object
must be identical raises a question: In what way is knowledge in a person?
Suppose that Smith knows Fido. Then Fido is in Smith. Obviously, Fido is not
there as he exists in the nonmental world of space and time. In what sense can
it be true that a person who knows what a dog is has that object in his mind?
Aristotle derives his answer from his general theory of reality. According to
him, all (terrestrial) substances are composed of two principles: matter
and form. If there are four dogs--Bowser, Fido,
Spot, and Spuds--they are the same in some respect and different in some
respect. They are the same in that each belongs to the same kind and each
functions similarly. Thus, Aristotle reasons, just as Plato had, that there must
be something in virtue of which they are the same, and this he calls
"form." That is, Bowser, Fido, Spot, and Spuds each have the very same
form of being a dog. They are different in that they are made out of different
matter, different parcels of stuff. The form that a thing has is more important
than its matter because it is the form that makes the thing what it is. If Fido
were to lose the form of being a dog and acquire another, he would no longer be
the same thing. The stuff out of which Fido is made is not similarly important,
and in fact that stuff changes periodically, as body cells change through
metabolic processes, without Fido ceasing to be Fido.
To return to the explanation of
knowledge, what is in the knower when he knows what dogs are is the form of
being a dog minus the matter. According to Aristotle, matter is literally
unintelligible and not essential to what Fido or any other dog is; thus its
absence is inconsequential for knowledge, though not for Fido.
In his sketchy account of the process of
thinking in De
anima (On the Soul),
Aristotle says that the intellect, like everything else, must have two parts:
something analogous to matter and something analogous to form. The first of
these is the passive intellect; the second is active
intellect, of which Aristotle speaks tersely. "Intellect in this
sense is separable, impassible, unmixed, since it is in its essential nature
activity. . . . When intellect is set free from its present conditions it
appears as just what it is and nothing more: it alone is immortal and eternal .
. . and without it nothing thinks."
This part of Aristotle's views about
knowledge is an extension of what he says about sensation. According to
Aristotle, sensation occurs when the sense organ is stimulated by the sense
object, typically through some medium, such as light for vision and air for
hearing. This stimulation causes a "sensible species" to be generated
in the sense organ itself. This "species" is some sort of
representation of the object sensed. As Aristotle describes the process, the
sense receives "the form of sensible objects without the matter, just as
the wax receives the impression of the signet-ring without the iron or the
gold." But, since there are different species for each of the five external
senses that Aristotle recognized--sight, hearing, touch, taste, and
smell--"species" does not mean "image."
After the development of Aristotle's
psychology the next significant event for the theory of knowledge was the rise
of Skepticism, of which there were at least two kinds. The first, Academic
Skepticism, arose in the Academy after Plato's death and was propounded
by the Greek philosopher Arcesilaus (c.
315-c. 240 BC), about whom the
philosophers Cicero, Sextus Empiricus, and Diogenes Laërtius provide
information. Academic Skepticism is also called "dogmatic Skepticism"
when it is interpreted as arguing for the thesis that nothing is known. The
thesis was inspired by Socrates' avowal that the
only thing he knew was that he knew nothing. Thus, it asserts that knowledge is
impossible. This form of Skepticism seems to be susceptible to an objection
raised by the Stoic Antipater (fl. c. 135 BC) and others that the view is self-contradictory. To know
that knowledge is impossible is to know something; hence, dogmatic Skepticism is
false.
Carneades
(c. 213-129 BC), a member of the
Academy, gave a subtle reply. Academic Skepticism, he claimed, should not be
interpreted as a claim about how the world is in itself or about a
correspondence between thought (or language) and the world, but as a judicial
decision. Just as a defendant in a trial does not prove his innocence but relies
upon its presumption and defends it against attack, so the Skeptic does not try
to prove that he knows nothing but presumes it and defends this presumption
against attacks.
Carneades' construal of Academic
Skepticism brings it close to the other kind, Pyrrhonism,
named after Pyrrho of Elis (c. 365-275
BC). None of his works survive, and scholars rely principally on the early
3rd-century-AD writings of Sextus Empiricus to understand Pyrrhonism.
Pyrrhonists assert or deny nothing but lead people to give up making any claims
to knowledge. The Pyrrhonist's strategy is to show that, for each proposition
with some evidence for it, an opposed proposition has equally good evidence
supporting it. These arguments for refuting each side of an issue are called "tropes."
For example, the judgment that a tower is round when seen at a distance is
contradicted by the judgment that the tower is square when seen up close. The
judgment that Providence cares for all things, based upon the orderliness of the
heavenly bodies, is opposed by the judgment that many good people suffer misery
and many bad people enjoy happiness. The judgment that apples have many
properties--shape, colour, taste, and aroma--each of which affects a sense
organ, is opposed by the equally good possibility that apples have only one
property that affects each sense organ differently.
Pyrrhonists diagnose dogmatism as the
unjustifiable preference for one mode of existence over another. Dogmatists
prefer wakefulness and sanity over sleep and insanity. But why should sleep and
insanity not be the norm? If the dogmatist answers that it is because sleep and
insanity involve some deficiency or abnormal physical states, the Skeptic
replies, "By what nonquestion-begging criterion are these things said to be
deficient or abnormal? Why should insanity not be taken as the primary notion
and sanity be defined as the lack of insanity? If it were, then it would not be
difficult to see sanity as a deficiency or abnormality, just as insanity
currently is. Or why should wakefulness not be seen as the deficient condition
in which people do not dream?" The Skeptic does not advocate insanity or
sleep but merely argues that a preference for them is no less justified than a
preference for sanity and wakefulness.
What is at stake in the preceding
Skeptical arguments is "the problem of the criterion," that is, the
problem of deciding how one can determine a justifiable standard against which
to measure judgments. Truth seems to need a
criterion. But every criterion is either groundless or inconclusive. Suppose
that something is proffered as a criterion. The Skeptic will ask what proof
there is for it. If no proof is offered, the criterion is groundless. If, on the
other hand, a proof is produced, a vicious circle begins to close around the
dogmatist: What judgment justifies belief in the proof? If there is no judgment,
the proof is unsupported; and if there is a judgment, it requires a criterion,
which is just what the dogmatist was supposed to have provided in the first
place.
If the Skeptic needed to make judgments
in order to survive, he would be in trouble. In fact there is another method of
survival that bypasses judgment. The Skeptic can live quite nicely, according to
Sextus, by following custom and the way things appear to him. In doing this, the
Skeptic does not judge the correctness of anything but merely accepts
appearances for what they are.
Ancient Pyrrhonism is not strictly an
epistemology since it has no theory of knowledge and is content to undermine the
dogmatic epistemologies of others, especially of the Stoics and Epicureans.
Pyrrho himself was said to have had moral and ethical motives for attacking
dogmatists. Being reconciled to not knowing anything, Pyrrho thought, induced
serenity (ataraxia).
St. Augustine of Hippo (354-430) claimed
that human knowledge would be impossible if God did not illumine the human mind
and thereby allow it to see, grasp, or understand ideas.
There are two components to his theory: ideas and illumination.
Ideas as Augustine construed them are the same as Plato's; they are timeless,
immutable, and accessible only to the mind, not to the senses. They are indeed
in some mysterious way part of God and seen in God. Illumination, the other
element of the theory, was for Augustine and his many followers, at least
through the 14th century, a technical term, built upon a metaphor. Since the
mind is immaterial, it cannot be literally lighted. Yet the entire theory of
illumination rested upon the extended visual metaphor, inherited from Plotinus
(205-270) and other Neoplatonic sources, of the human mind as an eye that can
see when and only when God, the source of light, illumines it. Still, it is a
powerful metaphor relied upon even in the 17th century by René Descartes
(Discourse on Method; 1637). Varying
his metaphor, Augustine sometimes says that the human mind participates in God
and even, as in On the Teacher, that Christ illumines the mind by dwelling in it. It
is important to emphasize that Augustine's theory of illumination concerns all
knowledge, and not specifically mystical or spiritual knowledge. In addition to
its historical significance, his theory is interesting for showing how diverse
epistemological theories have been. (see also Middle
Ages, Christianity)
Before he articulated this theory in his
mature years and soon after his conversion to Christianity, Augustine was
concerned to refute the Skepticism of the Academy. In Against the Academicians Augustine
claims that, if nothing else, humans know such disjunctive tautologies as that
either there is one world or there is not one world and that either the world is
finite or it is infinite. Humans also know many propositions that begin with the
phrase "It appears to me that," such as "It appears to me that
what I perceive is made up of earth and sky, or what appears to be earth and
sky." And they know logical (or what he calls "dialectical")
propositions, for example, "If there are four elements in the world, there
are not five; if there is one sun, there are not two; one and the same soul
cannot die and still be immortal; and man cannot at the same time be happy and
unhappy."
Many other refutations of Skepticism
occur in later works, notably, in On the
Free Choice of the Will, On the Trinity, and The
City of God. In the latter work Augustine proposes other examples of
things about which people are absolutely certain. Again in explicit refutation
of the Skeptics of the Academy, Augustine argues that if a person is deceived,
then it is certain that he exists. Like Descartes, Augustine puts the point in
the first person, "If I am deceived, then I exist" (Si
fallor, sum). A variation on this line of reasoning occurs in On the Trinity, when he says that if he is deceived, he is at least
certain that he is alive.
Augustine also points out that, since he
knows, he knows that he knows; and he notes that this can be reiterated an
infinite number of times: If I know that I know that I am alive, then I know
that I know that I know that I am alive. This point was codified in 20th-century
epistemic logic as the axiom "If X knows that p,
then X knows that X knows that p."
In The City of God Augustine claims that he knows that he loves:
"For neither am I deceived in this, that I love, since in those things
which I love I am not deceived." With Skepticism thus refuted, Augustine
simply denies that he has ever been able to doubt what he had learned through
his sensations or even the testimony of most people.
Skepticism did not recover from
Augustine's criticisms for a thousand years; but then it arose again like the
phoenix in Egyptian mythology. Augustine's Platonic epistemology dominated the
Middle Ages until the mid-13th century, when St. Albertus Magnus (1200-80) and
then his student St. Thomas Aquinas developed an alternative to Augustinian
illuminationism.
The phrase St.
Anselm of Canterbury (c.
1033-1109) used to describe his own project, namely, "faith seeking
reason" (fides quaerens intellectum),
well characterizes medieval philosophy as a whole. All the great medieval
philosophers, Christian, Jewish, and Islamic alike, were also
theologians. Virtually every object of interest was related to their belief in
God, and virtually every solution to every problem, including the problem of
knowledge, contained God as an essential part. Anselm himself said that, while
true propositions are those that signify what is, ultimately truth is God. This
presented Anselm with a problem, which he discusses at the beginning of Proslogium
as a prelude to his famous ontological argument for the existence of God. There
is a tension between the view that God is truth and intelligibility and the fact
that humans have no perception of God. How can there be knowledge of God, he
asks, when all knowledge comes through the senses and God, being immaterial,
cannot be sensed? His solution is to distinguish between knowing something by
being acquainted with it in sensation and knowing something by describing it. Knowledge
by description is possible because of the concepts that one forms from
sensation. All knowledge about God depends upon the description that he is
"the thing than which a greater cannot be conceived." From this
premise Anselm argues that humans can know, for example, that God exists, is
all-powerful, all-knowing, all-just, all-merciful, and immaterial. Eight hundred
years later Bertrand Russell would use the same distinction between knowledge by
acquaintance and knowledge by description to develop his influential philosophy,
although he would have vigorously denied that the distinction could be employed
as Anselm had, namely, to prove that God exists. (see also religious
belief)
While a Platonic and Augustinian
epistemology dominated the early Middle Ages, the translation of Aristotle's On
the Soul in the early 13th century had a dramatic effect. Following
Aristotle, Thomas Aquinas (1225-74) recognized that there are different kinds of
knowledge. Sense knowledge is what results from sensing individual things: thus,
one sees a tree, hears the song of an oriole, and tastes or smells a peach.
Thomas considered sense knowledge to be low-grade because it has individual
things as its object and is also shared with brute animals. Sensation itself
does not involve the intellect and is not properly speaking knowledge (scientia).
(see also Aristotelianism)
It is characteristic of scientific
knowledge to be universal; the more general in scope a piece of knowledge is,
the better. This is not to diminish the importance of specificity. Scientific
knowledge should also be rich in detail, and God's knowledge is the most
detailed. The detail, however, must be essential to the thing being studied and
not peculiar to just some instances of that kind. Although Thomas thought that
the highest knowledge humans can possess is knowledge of God, knowledge of
physical objects is more attuned to human capabilities, and only that kind of
knowledge will be discussed here.
In his discussion of knowledge in Summa
theologiae, Thomas Aquinas argues that human beings do not know
material objects directly, nor are such things the principal object of
knowledge. Knowledge aims at what is universal, while material things are
individual and can be known only indirectly. Elaborating on the thought of Aristotle,
Thomas claims that the process of thinking that acco |
