Abstract of a thesis entitled

'Suspension of Judgement: Agrippa and

submitted by
Yung Yeuk Yu
for the degree of Master of Philosophy
at the University of Hong Kong
in February 2001
It is believed that Agrippa (c.100 B.C.
100 A.D.) formulates or
assembles five general patterns of argument forms, known as the Five Modes,
to induce epochs (i.e., suspension of judgement), in which people neither posit
nor reject anything. Since then the Five Modes and the notion of epoche have
remained the core legacy of Pyrrhonian scepticism (or Pyrrhonism for short).
By pursuing the Five Modes and epoche, I attempt to set Pyrrhonism against its
background and indicate its importance and relevance to philosophical
enterprise and approach.
I incorporate in Chapter Two a textual study of the Five Modes from
two ancient sources, namely Sextus Empiricus and Diogenes Laertius. I also
argue that the Pyrrhonist can induce epoche on the strength of the Five Modes
without the necessity of committing to the notion of akatalepsia (i.e., inability
to comprehend).
In Chapter Three I discuss the Agrippan problem posed by the Five
Modes. The Agrippan problem is a lively issue in philosophy. The Pyrrhonian
attack on the Dogmatic account of criterion found in Sextus' Outlines of
Pyrrhonism (PH) demonstrates how the Five Modes constrain theories of
epistemic justification and block every way out of the epistemic predicament of
belief justification. I go through Barnes's modern version of the Agrippan
problem and address various contemporary responses to the problem, including

Fogelin's remark that 'If the Agrippa problem cannot be resolved, there is no
reason to suppose that knowledge of the kind sought by justificationalist
philosophers exists', and Chisholm's contention that the problem could be
resolved only by begging the question.
In Chapter Four I review the philosophical debate between Frede,
Burnyeat and Barnes over the nature and scope of Pyrrhonian scepticism. The
disagreement centres on the scope of epochs and the interpretation of PH113,
the passage in which Sextus mentions two senses of dogma and explains in
what sense the sceptic does not dogmatize. The debate proceeds as if PH 113
specifies the scope of epochs and then in turn defines the scope of Pyrrhonian
scepticism until Barnes points out that PH I 13 is silent on the status of
ordinary beliefs and so 'epoche may be broad or narrow'. After assessing each
interpretation and examining the arguments put forward, I conclude that
Barnes's interpretation is more plausible.
Finally, I offer some concluding remarks in Chapter Five by discussing
the insight and assets bequeathed to us by the Pyrrhonist. The discussion
focuses on the nature of philosophy and the philosophical approach. I draw on
the different views of Sextus, Wittgenstein, Strawson and Husserl on the nature
of philosophy and the philosophical approach.

Suspension of Judgement: Agrippa and

by
Yung Yeuk Yu

B.A. H.K.U.

A thesis submitted in partial fulfilment of the requirements for

the Degree of Master of Philosophy
at the University of Hong Kong
February 2001

Declaration
I declare that this thesis represents my own work, except where due
acknowledgement is made, and that it has not been previously included in a
thesis, dissertation or report submitted to this University or to any other
institution for a degree, diploma or other qualification.

Signed
Yung Yeuk Yu

Acknowledgements
This thesis is an attempt to explore deep into the heart of Pyrrhonian
scepticism. The process of researching and writing is painstaking. But it is
worthwhile. Intellectual achievement gives satisfaction. I am indebted to
increasing amount of scholarly works on the subject. My greatest gratitude
goes to Prof. F. C. T. Moore, my supervisor, without his inspiration and
support after his retirement from the department nothing would have been
possible.

ii

Contents
Declaration
Acknowledgements

ii

Table of Contents

iii

Foreword

iv

Chapter One

Overview

Chapter Two

Agrippa and the Five Modes

Chapter Three

The Agrippan Problem: A Pyrrhonian Challenge to Epistemic Justification... 37

Chapter Four

61

Chapter Five

Concluding Remarks on Pyrrhonian Scepticism and Philosophy

118

Glossary

133

Bibliography

135

iii

Foreword
There is a controversy over the interpretation of Pyrrhonian scepticism,
the protagonists being Frede, Bumyeat and Barnes. This controversy is
exemplary for its high level of scholarship, but also because of the way in
which the participants, while debating detailed points about the texts of Sextus
Empiricus (and others), explicitly look to the broadest questions about
philosophy, its nature and its history. At the heart of this thesis is a critical
review of the debate (which comes down on the side of Barnes). I attempt to
situate the Pyrrhonist in the conclusion, showing how he may lead us to rethink
scepticism and philosophy as well.
In this thesis, Sextus Empiricus's Outlines of Pyrrhonism, a work
frequently referred to, is abbreviated as PH (according to the Latin title), and
similarly his Against the Mathematicians is abbreviated as M. Though there are
several translations of the PH into English, the main one used is Annas and
Barnes [1994]. Mates [1996] and Sextus Empiricus [1933-49] are also
consulted. (For full bibliographic details please see the bibliography.)
References to Greek words are normally given, where necessary, in the
conventional transcription system into the Roman alphabet, including macrons
to mark long vowels, to avoid ambiguity. For the reader's convenience, a short
glossary of these words is given at the end of the thesis, with the relevant
definitions as given in a standard Greek-English dictionary. In some cases, such
as the word dogma, there is a scholarly controversy about the appropriate
definitions. These are indicated and sometimes discussed and documented in
the body of the text.
I have adopted the spelling 'Pyrrhonian' in this thesis. Some writers
prefer 'Pyrrhonean', a spelling that is retained in this thesis only within
quotations. 'Pyrrhonian scepticism' and 'Pyrrhonism' are used interchangeably;
'Pyrrhonian sceptic' and 'Pyrrhonist' are also used in the same manner.
iv

CHAPTER ONE

Overview

YRRHONIAN SCEPTICISM (or Pyrrhonism for short) is an ancient

form of scepticism which originated in Hellenistic philosophy in the
4th century B.C.. Pyrrho of Elis (c.365-270 B.C.) is thought to be the
founder. This thesis focuses on the Five Modes of Agrippa, a set of arguments
which has been described as the soul of Pyrrhonian scepticism.l The Five
Modes and the notion of epoche (suspension of judgement) are the point of
departure.
The Five Modes are patterns of inducing epoche. There is a review of the
controversy between Frede, Burnyeat and Barnes on how we should give
content to the Pyrrhonian epoche in Chapter Four. The review is philosophical.
It evaluates conflicting interpretations of epoche and arguments are examined in
detail.
This thesis aims at a critical evaluation of Pyrrhonian scepticism and its
relevance to philosophy. I do touch upon some historical matters related to the
Five Modes and Pyrrhonian scepticism in Chapter Two. The discussion of
historical matters is preparatory and it helps give a comprehensive treatment of
the subject. It should be made clear in the discussion that this is not a historical
1

Barnes [1990] p.ix

1

reconstruction of Pyrrhonian scepticism and I offer no novel interpretation of

it.2
Chapter Two provides a broad basis for discussion. It opens with a
preliminary treatment of some background issues about the Five Modes, for
instance, Agrippa's authorship and the role of his Five Modes in Pyrrhonian
scepticism. Four points are considered:
1. There are two accounts of the Five Modes separately compiled by Sextus
and Diogenes. Sextus produces our principal texts of Pyrrhonian scepticism
but he is rather more obscure than Diogenes on the authorship of the Five
Modes. As far as we can tell, only Diogenes ascribes the Five Modes to
Agrippa.
2. By comparing the two accounts of the Five Modes, I find that there are some
deviations between them worth mentioning though they are more or less the
same in many places.
3. I maintain that even though the Five Modes of Agrippa are superior to the
Ten Modes of Aenesidemus, they do not necessarily replace the Ten Modes
in the Pyrrhonian framework.
4. I argue that the Pyrrhonist can induce epoche on the strength of the Five
Modes without the necessity of committing to the notion of akatalepsia
(inability to comprehend).
Chapter Three addresses the Agrippan problem posed by the Five Modes.
The Agrippan problem brings the foundation of belief under severe scrutiny to
the effect that belief is rendered groundless. In this regard no belief is preferable
to its contrary; people could neither posit nor reject anything. The Agrippan

For historical reconstruction of Pyrrhonism, see Annas and Bames [1985], Brunschwig [1994],

Mates [1996] and Algra, Bames, Mansfeld, & Schofield [1999].

2

problem is a problem of epistemic justification. Barnes [1990] argues that the

Five Modes have raised the fundamental issue concerning the status of
epistemic extemalism. We shall see how the Five Modes are capable of
constraining theories of justification rigorously and the issue of epistemic
extemalism in Chapter Three.
Presumably, the Five Modes are capable of undermining theories of
epistemic justification, resulting in a full-scale epochs. It is one thing for the
Five Modes to be capable of yielding this full-scale epoche. It is another thing
for Sextus or the Pyrrhonist of the PH to really eschew all sorts of beliefs. The
Five Modes can stand on their own. But the Pyrrhonist is not bound to align with
them. To find out if Sextus' Outlines of Pyrrhonism (PH) has spelled out the
scope of epoche, the review in Chapter Four examines how various conflicting
accounts by Frede, Burnyeat and Barnes interpret the meaning of PH 113 and its
exact bearing on the scope of epoche.
Finally, I offer some concluding remarks in Chapter Five by discussing the
insight and assets bequeathed to us by the Pyrrhonist. The discussion focuses on
the nature of philosophy and philosophical approach. I draw on the different
views of Sextus, Wittgenstein, Strawson and Husserl on the nature of
philosophy and the philosophical approach.

CHAPTER TWO

Agrippa and the Five Modes

Historical background

CEPTICISM ABOUT KNOWLEDGE seems peculiar in the

philosophical scenery. On the one hand, sceptical arguments seem to

undermine all belief and knowledge claims. On the other hand they are

feeble and idle in the sense that they leave everything as it is. People may feel

intellectually uneasy about scepticism. But this is all it can do. Philosophical
scepticism about knowledge rarely carries any practical implication in real life.
It dwells in the reflective plane. There are several attempts to deal with this
scepticism. Some people, like Barry Stroud in The Significance of Scepticism,
suggest that we should defuse it, making it philosophically impotent while
others, like G. E. Moore with his Proof of an External World and A Defence of
Common Sense, want to refute it, eliminating the habitat of scepticism.
Regardless of how many arguments people put forward to establish that we do
know the things that the sceptic denies we know for certain, scepticism is a
persistent item on the philosophical agenda. The 'struggle' between the sceptic
and his rival is a never-ending story. Its origin can be traced to ancient Greece.
Pyrrhonian scepticism was named after Pyrrho, who left no writings. But
the works of his student Timon of Philius (c.320-230 B.C.) provide us with
scattered pieces of information about his life and teachings. Other ancient

sources, for instance Lives of the Philosophers by Diogenes Laertius and

Preparation for the Gospel by Aristocles of Messene, also give us bits of
information about Pyrrho. Despite that, Pyrrho remains a shadowy figure for
us. It is generally believed that Aenesidemus of Cnossus, a Hellenistic
philosopher unhappy with the Middle or New Academy in the first century B.C.,
revived Pyrrhonism with his formulation of the Ten Modes and some other
modes, through which epoche would be induced. The Academy was another
sceptical force in Greek philosophy. Under the headship of Arcesilaus of Pitane
(315-240 B.C.) the Academy became sceptical and remained so for more than
two hundred years. Sextus reports that the sceptical Academics argued that we
could have no knowledge of any sort while the Pyrrhonian sceptic suspended
judgement.4
The key figure for our acquaintance with Pyrrhonism is Sextus, whose PH
and M, written in the second century A.D., were rediscovered and published in
Latin in Medieval Europe in the sixteen century A.D., on the eve of the birth of
early modern philosophy.
The PH consists of three books. Book I is a general exposition of
Pyrrhonism while Book II and III are collections of arguments against dogmatic
enterprises in logic, physics and ethics, which make up philosophy for most
Hellenistic philosophers. M, in eleven books, is Sextus' other preserved work.
In it we find sceptical arguments against linguistics, rhetoric, geometry,
arithmetic, astrology, musical theory, and arguments against dogmatic
enterprises in logic, physics and ethics.
Hellenistic philosophers experienced different philosophical approaches.
The Socratic Method or the elenchus proceeds from tentative definition given
3
4

For Pyrrho and his legacy, see Bert [2000].

For a brief but useful introduction to philosophy in the Hellenistic age, see Sedley [1980].
5

by the interlocutor to the rejection of the definition. Euthyphro, for instance,

offered an account of piety. Socrates then proceeded from Euthyphro's account
of piety and his other beliefs related to piety to a conclusion that contradicted
the tentative account of piety offered in the first place. Euthyphro was required
to make substantial modification to his account of piety or to confess that he did
not know piety at all. The Socratic Method or the elenchus is ad hominem; it
undermines Euthyphro's tentative account of piety by exposing its inherent
contradictions.
While Socrates is renowned for his Socratic method of elenctic testing,
Agrippa is credited with reviving interest in his Five Modes. As we shall see in
Chapter Three the Agrippan modes of argumentation work in a different manner;
they render a belief claim groundless by constraining the justification process.
Agrippa is a shadowy figure, to whom the Five Modes are attributed.
Besides the Five Modes (PHI 164-177), the Pyrrhonian sceptic of the PH was
equipped with the Two Modes (PHI 178-9)5, the Eight Modes against dogmatic
causal explanations (PH I 180-6)6 and the Ten Modes (PH I 35-163). Among
these sets of modes, the Five Modes by Agrippa, with which we are mainly
concerned, are the most powerful and tactical way of inducing epochs. Sextus
was so confident in the Five Modes that after introducing the Five Modes he
claimed 'every object of investigation can be referred to these modes'. (PH I
169) The Pyrrhonian modes (tropoi) are sometimes called the modes of
5

Sextus seems to suggest at the passage that the Two Modes are also offered by 'the more recent

sceptics' who 'handed down' the Five Modes. In other words, the Two Modes were another set
of modes by Agrippa. But neither Sextus nor anybody else explicitly acknowledged the
relationship between Agrippa and the Two Modes.
6

At M VII 345 Sextus reports that Aenesidemus produced both the Eight Modes and the Ten

Modes.
6

suspension of judgement, as they are formulated to induce epoche, which was

characterised as 'a standstill of the intellect, owing to which we neither reject
nor posit anything'. (PH110) Sextus suggested that this epoche would give rise
to the ultimate aim of Pyrrhonian scepticism i.e., 'tranquillity in matters of
opinion (ataraxia) and moderation of feeling in matters forced upon us.' (PHI
25)
Agrippa and his Five Modes pose important and profound problems in the
history of philosophy and the development of Pyrrhonian scepticism. Though
Diogenes ascribed the Five Modes to Agrippa, we are still unable to decide the
extent to which Agrippa should be credited with novelty in putting forward the
Five Modes. Barnes, for instance, is impressed by 'the close thematic similarity',
and 'linguistic parallels between Aristotle's text (Posterior Analytics) and
Sextus' exposition of Agrippa's modes.'7 He conjectures that the philosophical
core of Agrippa's Five Modes might be derived historically from some ideas
discussed in the Posterior Analytics. In the light of insufficient evidence Barnes
is reserved about this 'historical tale'. There is, indeed, philosophical continuity
between Pyrrhonian scepticism and the prior development of ancient Greek
Philosophy. For instance, the revival of Pyrrhonian scepticism by Aenesidemus
was actually the result of his reaction to the Sceptical Academy. I shall not here
offer an account of the development of Pyrrhonian scepticism from the
perspective of philosophical continuity.
The objective of the following discussions in this chapter is to explore
different aspects of the Five Modes. Textually, I shall compare and contrast two
accounts of the Five Modes from Sextus and Diogenes. Then I shall discuss the
uncertain relationship between the Five Modes and the Ten Modes within the

Barnes [1990] p.121

7

framework of PH. Finally, with regard to the philosophical merit of the Five
Modes, I shall argue that the Pyrrhonist can induce epoche on the strength of the
Five Modes without invoking akatalepsia or closing off possible paths of further
enquiry. My discussions take off with the problem of authorship of the Five
Modes and Agrippa.

Agrippa

HOUGH AGRIPPA is credited with increasing concern for his

Five Modes, there is very little else that can be said about him.
Even his date is open to conjecture. He probably flourished during

the period between Aenesidemus and Sextus. Roughly speaking, his time would

range from the second half of 1st century B.C. to 1st century A.D. We possess
none of his written works. In fact, we do not even know if he ever produced any
written works. The only textual evidence in favour of his authorship of the Five
Modes is found in the Lives of Eminent Philosophers written by Diogenes in the
3 rd century A.D.8 Otherwise Agrippa is rarely mentioned by any ancient authors.
The time gap between Agrippa and Diogenes could be as large as three and a
half centuries. In Sextus' works, we find an earlier account of the Five Modes.
The two accounts by Sextus and Diogenes are the only source of our
acquaintance with the Five Modes. Unlike Diogenes, Sextus does not ascribe

Book IX 88 'But Agrippa and his school add to them [the Ten Modes by Aenesidemus] five

more other modes, resulting respectively from disagreement, extension ad infinitum, relativity,
hypothesis and reciprocal inference.'

the Five Modes to Agrippa. When he introduces the Five Modes, he just vaguely
notes that 'The more recent sceptics offer the following five modes of
suspension of judgement'. (PH I 164) Prior to the Five Modes, when Sextus
introduces the Ten Modes, he writes,
The older sceptics normally offer ten modes in number through which we
are thought to conclude to suspension of judgement. (PHI 36)

The Ten Modes by Aenesidemus are earlier than the Five Modes. In
Sextus' line, 'the older sceptics' who passed down the Ten Modes are
contrasted with 'the more recent sceptics' who added the Five Modes. In book
VII of M, Sextus identifies Aenesidemus as one of 'the older sceptics' and
ascribes the Ten Modes to him.9 But interestingly, the identity of 'the more
recent sceptics' remains an unresolved riddle in Sextus' pages. Curiously, for no
reasons that we can discern, Sextus does not mention the name of Agrippa at all
in his exposition of the Five Modes and Pyrrhonian scepticism. Hence,
Diogenes' line about Agrippa in the Lives of Eminent Philosophers is the only
testimony on the authorship of the Five Modes made in an extant work.
There are some obscurities about the authorship of the Five Modes. It is
suggested that 'Agrippa and his school' added the Five Modes to Pyrrhonian
scepticism. However, who was 'his school'? Does the phrase imply that there
were a number of co-authors? The name of Agrippa is given but 'his school' is
curiously nameless. Are the Five Modes a piece of individual work solely by
Agrippa? Are they the result of collective effort by Agrippa and some unknown
co-authors? We do not know. It could also be the case that Agrippa in fact
leaves no writings and it is his apostles, unknown to us, who compile the Five

See M VII 345

9

Modes, in a way similar to what happens with Pyrrho and Timon. Unless new
materials are found we can never know how things happened and the obscurities
of authorship would remain. At any rate, it is still appropriate to accept
Agrippa's authorship over the Five Modes until conflicting evidence comes up.
This is all I want to say about Agrippa. With that, I proceed to a comparison of
the two accounts of the Five Modes.

10

Two Accounts of the Five Modes

S I HAVE noted, Sextus' account of the Five Modes is closer to the

date of Agrippa. Nevertheless, it is Diogenes, rather than Sextus,
who ascribes the Five Modes to Agrippa. This suggests that

Diogenes does not rely solely on Sextus' works when he compiles the Five

Modes. However, as to what the other materials are and how Diogenes comes to
recognize Agrippa as the author of the Five Modes, we do not know. Obviously,
it is rather futile to speculate on these historical obscurities when we possess so
few textual resources about the issues. The two accounts appear more or less the
same in many places. However, there are some deviations between them worth
mentioning. I will discuss the similarities as well as the differences between
them. For comparison, I shall first lay out Diogenes' account:

The mode arising from disagreement proves, with regard to any inquiry
whether in philosophy or in everyday life, that it is full of the utmost
contentiousness and confusion.
The mode which involves extension ad infinitum refuses to admit that
what is sought to be proved is firmly established, because one thing
furnishes the ground for belief in another, and so on ad infinitum.
The mode derived from relativity declares that a thing can never be
apprehended in and by itself, but only in connexion with something else.
Hence all things are unknowable.
The mode resulting from hypothesis arises when people suppose that you
must take the most elementary of things as of themselves entitled to
credence, instead of postulating them: which is useless, because some one
else will adopt the contrary hypothesis.

11

The mode arising from reciprocal inference is found whenever that which should be
confirmatory of the thing requiring to be proved itself has to borrow credit from the
latter, as for example, if anyone seeking the existence of pores on the ground that
emanations take place should take this (the existence of pores) as proof that there
are emanation. (Lives of Eminent Philosophers book IX 88-90)

The following is Sextus' account.

According to the mode deriving from dispute, we find that undecidable

dissension about the matter proposed has come about both in ordinary life
and among philosophers. Because of this we are not able either to choose
or to rule out anything, and we end up with suspension of judgement.
In the mode deriving from infinite regress, we say that what is brought
forward as a source of conviction for the matter proposed itself needs
another such source, which itself needs another, and so ad infinitum, so
that we have no point from which begin to establish anything, and
suspension of judgement follows.
In the mode deriving from relativity, as we said above, the existing object
appears to be such-and-such relative to the subject judging and to the
things observed together with it, but we suspend judgement on what it is
like in its nature.
We have the mode from hypothesis when the Dogmatists, being thrown
back ad infinitum, begin from something which they do not establish but
claim to assume simply and without proof in virtue of a concession.
The reciprocal mode occurs when what ought to be confirmatory of the
object under investigation needs to be made convincing by the object
under investigation; then, being unable to take either in order to establish
the other, we suspend judgement about both. (PHI 165-169)

12

Some general points first. Sextus and Diogenes put the modes under the
same headings and in the same order: Disagreement, Infinite Regression,
Relativity, Hypothetical and Reciprocal. My discussion shall also follow this
order. Sextus mentioned the Dogmatists being confronted with the mode of
Infinite Regression, and getting into trouble with the Hypothetical mode.
Diogenes made no reference to the Dogmatists. To illustrate the Reciprocal
mode, Diogenes gave an example of circular argument arguing for the existence
of pores. Sextus gave no example at all in his account. After introducing the
Five Modes, Sextus went on to demonstrate briefly how the Modes, working in
collaboration, are capable of inducing epoche in every object of investigation.
Diogenes, on his part, rehearsed sceptical arguments against demonstration,
criterion, sign, cause, motion, the process of learning, and coming into being. In
his rehearsal, he made explicit reference to the Five Modes at some points.
Sextus concluded with epoche four times. The only mode that he did not
finish with a conclusion is Infinite Regression. Diogenes drew one conclusion
only and so his presentation is a little bit shorter than that of Sextus. It is worth
notice that the only conclusion that Diogenes drew is not epoche. He finished
the Relativity mode with 'Hence all things are unknowable'. I shall go into more
detail when I discuss the Relativity mode. Now I move on to the mode of
Disagreement.

10

For examples, see Book IX 90-92, 92-94,94-96.

13

The Mode of Disagreement

OTH SEXTUS AND DIOGENES mentioned disagreement arising

in philosophy and ordinary life. In Diogenes' account, the mode of
Disagreement was said to establish the presence of 'the utmost

contentiousness and confusion' in philosophy or everyday life. Diogenes had

nothing more to say on the purpose or function of disagreement with regard to

epoche. Sextus explained how the presence of disagreement results in epoche. In
the light of disagreement, Sextus concluded for us that 'we end up with epoche.
For 'we are not able either to choose or to rule out anything'. In general, epoche
is brought by 'the equipollence in the opposed objects and accounts' (PHI 8).
For example, after examing both sides of the disagreement on whether time has
a beginning, one cannot decide what to accept and what to reject and so one ends
up with epoche. Put it in schematic terms, we suspend judgement about the
proposition that p because after taking the disagreement about p into account we
cannot decide whether to believe that p or disbelieve that p. In that case, the
equipollence is the result of unresolved disagreement. One point to note is that
Sextus described the 'dissension about the matter proposed' as anepikritos (not
decided, indeterminate) and this anepikritos dissension forced us to suspend
judgement when we fail to choose or rule out anything. The notion of
anepikritos invites the connotation of akatalepsia (inapprehensibility or
unknowability). It could be argued that the unresolved disagreement about p is
not decided because p is in fact inapprehensible or unknowable. Therefore the
sceptic who is either aware of this fact or supposes that it is so recommends
epoche. I think this is a mistaken interpretation. I shall postpone my discussion
of this issue to the end of this chapter. For the time being, I simply want to
14

suggest that the sceptic would not induce epoche by appealing to the notion of
akatalepsia.

The Mode of Infinite Regression

HE MODE OF INFINITE REGRESSION is concerned with the

structural regression of reasoning. Sextus argued, 'what is brought

forward as a source of conviction for the matter proposed itself needs

another such source'. To prove one thing you need a proof. However, to support

this proof you need another proof. Another proof needs a further proof, and so
on ad infinitum. The result is that 'we have no point from which to begin to
establish anything'. Of course, what follows is epoche. This argument assumes
that a proof needs another proof to support it, if it is to be credited as trustworthy.
Otherwise, it is subject to epoche. The mode of Infinite Regression dismisses
any reasoning conducted in infinite regressive manner.
Diogenes, on his part, argued 'what is sought to be proved'. Obviously,
'what is sought to be proved' is not a proof. In fact, 'what is sought to be
proved' is in need of a proof or 'something brought forward as a source of
conviction'. Therefore, while Sextus mentioned the proof, Diogenes talked of
things sought to be proved. The objects of their presentations are different. But
this does not alter the structure of the mode of Infinite Regression in Diogenes'
account.
In Diogenes' account, the mode of Infinite Regression 'refuses to admit
that what is sought to be proved is firmly established'. For 'one thing furnishes
the ground for belief in another, and so on ad infinitum1. Diogenes implicitly
suggested that the target of the mode of Infinite Regression is 'belief. Suppose
15

a whole set of beliefs is structured in such a way that every single belief is
dependent on a prior belief while this prior belief is founded on a further prior
belief. The mode of Infinite Regression exerts pressure on the ground of any
belief claim if that belief claim is justified by a prior belief.
It is worth notice that Diogenes was more reserved about the power of the
mode of Infinite Regression. While Sextus argued, 'we have no point from
which to begin to establish anything', Diogenes limited the mode of Infinite
Regression to the denial that what is sought to be proved in regressive manner is
firmly established. He did not argue that it should be rejected. There would be
room for people who operated with less demanding epistemological criteria to
accept belief grounded on structural regressive reasoning.

16

Relativity Mode

HE TWO ACCOUNTS differ from each other about the conclusion

of the Relativity mode. After making the claim that things are
apprehensible only in connection with other things but not in and by

themselves, Diogenes finished the Relativity mode with the conclusion: 'Hence
all things are unknowable'. Curiously, this was the first time but also the last
time that Diogenes drew a conclusion in his articulation of the Five Modes. This
conclusion was very different from the Pyrrhonian conclusion, i.e. epoche,
which Sextus drew in his presentation of the Relativity mode and other modes

as well. Diogenes' conclusion incorporated akatalepsia. I shall argue that the

intrusion of akatalepsia was a substantial deviation from the usual Pyrrhonian
epoche. We do not know whether Diogenes himself added akatalepsia to the
Relativity mode or he followed his source. At any rate, this addition gave the
impression that the Relativity mode here made a categorical assertion about
unknowability of things. But as I shall argue, the Pyrrhonist does not
incorporate akatalepsia.
In the meantime, this impression could be explained away with reference
to Sextus' reminder that the Pyrrhonist uses 'is' and 'are' not in the categorical
senses of 'really is' and 'really are' but in the non-categorical and loose senses
of 'appear(s) to be'. 11 So even if a Pyrrhonist happens to say something like 'all
things are unknowable', what he means would be simply that 'all things appear
to him to be unknowable'. He is just reporting his state of mind. It seems to him
that all things are unknowable. But as to whether or not all things are really
unknowable, the Pyrrhonist would suspend judgement. It seems very unlikely

11

See PH1135 and M XI18-9

17

that the Pyrrhonist would have more than this in his mind when he makes the
seemingly categorical statement that all things are unknowable. Likewise,
'everything is relative' would be taken as 'everything appears relative'.
Sextus' presentation of the Relativity mode does not give the impression
that the mode admitted of akatalepsia. He simply dwelled on the fact that an
object of perception or of thought, 'appears to be such-and-such relative to the
subject judging and to the things observed together with it'. Therefore, the
sceptic recommended epoche on the real nature of things. This argument looks
similar to the discussion of colour in modern epistemology in which people
argue that the sky is not blue in itself; it just appears to be blue to us. In general
things are colourless, but when they reflect light in different wavelengths, they
appear to have various colours. But what appears blue to a human being does
not appear to be so to a dog. Hence, colour is relative to the perceiving subjects.
Despite different formulations of the mode, the content is the same. For the
notion of 'an existing object' in Sextus' account can be taken in its fullest sense
to cover objects of perception and objects of thought. Hence, to say that things
are unknowable in and by themselves and we can only apprehend them in
connection with something else is more or less the same as saying that objects
are (or appear to be) relative to the subject judging and to the things observed
together with them. Epoche follows.

18

The Hypothetical Mode

N SEXTUS' account of the Hypothetical mode, the 'Dogmatists, being

thrown back ad infinitum, begin from something which they do not
establish but claim to assume simply and without proof. There are two

points to note. First, he introduced it with the mode of Infinite Regression. This
is the first example of how the Five Modes can work in collaboration with each
other to induce epoche. The second point is about the dogmatists who were
challenged by the mode of Infinite Regression and the Hypothetical mode. It
implicitly suggested that the dogmatists were the targets of Pyrrhonian sceptic.
The 'Dogmatists' were those who claimed that they had found the truth. In PHI
3 Sextus gave a short list of the Dogmatists. They were 'the schools of Aristotle
and Epicurus and the Stoics, and some others.' Their philosophy was called
'Dogmatic philosophy'. There was also 'Academic philosophy'. Dogmatic and

Academic philosophies are contrasted with Sceptical philosophy in Sextus'

exposition of Pyrrhonian Scepticism.
The Hypothetical mode is opposed to the idea of an axiomatized model in
epistemic justification. According to this model, there are two types of beliefs:
derived beliefs and beliefs that are the most elementary and entitled to credence
by themselves. Like the relationship between theorems and axioms, derived
beliefs are logically generated from a set of elementary beliefs. The idea of an
axiomatised system attracted various Greek philosophers like Plato and
Aristotle. But the Hypothetical mode rejects the presupposition that we must
take the most elementary of things as of themselves entitled to credence when
they are not postulated.

19

The quality of being unsupported is a defining characteristic of hypothesis.

To hypothesize that p we assume p is the case without any proof. Otherwise, it
could be anything else but not a hypothesis. According to Diogenes, hypothesis
is 'useless, because some one else will adopt the contrary hypothesis'. If it were
acceptable to hypothesize that/?, then there would be no reason that the opposite
of p could not be hypothesized with the same degree of credibility. For
obviously, there was nothing put forward to support/? when it was hypothesized.
What could restrain the opposite of p from being hypothesized, provided that
the opposite of p did not appear to be unreasonable or unintelligible at all? For
example, a creationist may assume the existence of God to account for the orgin
of human beings. But an evolutionist may equally assume the non-existence of
God and explain the emergence of human beings in terms of evolution. An
unsupported assumption by itself can never rule out its opposite. Unless the
assumption appeals to some other propositions to rule out its opposite, the
assumption is vulnerable to the Hypothetical mode (so is the presupposition of
the axiomatized model). But if the assumption draws on other propositions, the
sceptic can resort to the rest of the Five Modes. In a word, the Hypothetical
mode represents the Pyrrhonian sceptic's uncompromising and demanding
attitude to the fundamental presupposition of hypothesizing. I shall touch upon
the Hypothetical mode again in Chapter Four.

20

The Reciprocal Mode

EXTUS AND DIOGENES were complementary to each other in their

articulations of the Reciprocal mode, the last of the Five Modes.
Suppose someone puts forward the proposition that p as a source of

conviction and purports to support it with q, which in turn draws on p for

credibility. This is an example of circular inference with which the Reciprocal

mode is concerned. Diogenes offered an example of circular inference in which
emanations were put forward as a proof of the existence of pores. Sextus drew
the conclusion for us. Given the fallacious proof ofp with reference to q, we are
'unable to take either in order to establish the other, we suspend judgement
about both'.
One argues reciprocally when the conclusion of his first argument is a
premise in his second argument which itself establishes a premise of his first
argument. Reciprocal arguments are circular arguments with just two pairs of
arguments. But not all circular arguments are reciprocal arguments. Circular
arguments can have more than two components.

21

Relationship between the Five Modes and the Ten Modes

EXTUS FINISHED his articulation of the Five Modes with the

following remark. 'Such are the Five Modes which have been handed
down by the more recent Sceptics. They put them forward not as

rejecting the Ten Modes but in order to refute the rashness of the Dogmatists in
a more varied way by using both sets together.' (PH I 177) This is the only
statement that Sextus made about the relationship between the two sets of
modes. He stressed that the Five Modes are 'handed down' by the more recent
sceptics to work with the earlier Ten Modes of Aenesidemus 'in order to refute
the rashness of the Dogmatists in a more varied way'. Why did Sextus stress that
the Five Modes did not reject the Ten Modes? Was it because some people or
even some sceptics at Sextus' time were confused about the relationship

between the two sets of modes and some of them even attempted to replace the
Ten Modes with the Five Modes? These are speculations. The relationship
between the two sets of modes is confusing.
The Ten Modes of Aenesidemus induce epoche by appealing to
dissimilarity mphantasia (appearance).12 The underlying argument of the Ten
Modes is that p appears x but also y, where x and v are incompatible. We could
not decide ifp is really x or v. Hence, epoche follows. To produce dissimilarity
in appearance, the Ten Modes compare humans and other animals, human
variations, the senses, circumstances, places and positions, mixtures, quantities,
relativity, the common and the rare, and finally customs and persuasions.

12

For a comprehensive discussion of the Ten Modes, see Annas and Barnes [1985]. My later

discussion on Aenesidemus' Relativity mode and Agrippa's Relativity mode is based on their
hypothesis.

22

Although Sextus explicitly stated that the Ten Modes and the Five Modes
were complementary, there is textual evidence leading people to think
otherwise. Sextus' formulation of Agrippa's Relativity mode (PH I 167)
contains an extra phrase, which is 'as we said above'. This seemingly
unimportant phrase is worth looking into. Annas and Barnes [1985] have
investigated the phrase and they found that it refers to Sextus' earlier
presentation of Aenesidemus' Relativity mode (PH I 135-40), which came
eighth in the Ten Modes. In Agrippa's Relativity mode, an existing object is
'relative to the subject judging and to the things observed together with it'. In
Aenesidemus' Relativity mode, relativity is broken down into two senses:
'relative to the subject judging' and 'relative to the things observed together
with the object'. Agrippa's Relativity mode appears almost completely identical
to Aenesidemus' Relativity mode. The same conception of relativity appeared
in two supposedly separate modes of relativity in Sextus' exposition. The phase
'as we said above' is intended to remind the readers about the earlier
presentation of Aenesidemus' Relativity mode does it imply that Agrippa's
Relativity mode mainly draws on Aenesidemus' Relativity mode for content
and substance?
On introducing the Ten Modes, Sextus arranged them at PH I 38-9 under
the headings of three 'superordinate' modes, namely 'that deriving from the
subject judging', 'that deriving from the object judged' and 'that combined
from both'. While the superordinate modes were said to be more 'specific', the
Ten Modes were labelled as 'subordinate'. The ten subordinate modes were put
under the three specific superordinate modes according to the context to which
they appeal. The first four of the Ten Modes shared the same context as 'what
judges is either an animal or a human or a sense, and is in some circumstance'
and thus they fell under the first superordinate mode deriving from the subject
23

judging. The seventh and tenth were put under the second superordinate mode
deriving from the object judged. The fifth, sixth, eighth and ninth belonged to
the third superordinate mode combined from both.
The superordinate modes were in turn subsumed under the Relativity mode.
The Relativity Mode was thus made a more outstanding mode than the other
nine modes. It was one of the Ten Modes but at the same time, it was made
superior to the other nine modes. The Relativity Mode argues that everything
appears relative to the subject judging and/or to the things judged. We find no
better reason to prefer this than that or vice versa, so we have to suspend
judgement. The Relativity mode is so broad that it virtually covers the other
nine modes. It is the 'most generic' or the most comprehensive, summing up the
structure of the other nine modes.
It seems problematic for the same Relativity mode to be one of the ten
subordinate modes on the one hand and the most generic on the other hand at the
same time. Moreover, the taxonomy is inconsistent. When Aenesidemus'
Relativity mode was made the most generic at PHI 38-9 it consisted of three
components. But when it was put together with the other nine modes at PH I
135-40, the third component 'that deriving from both' was omitted; only the
subject judging and the things observed together with the object were preserved.
In Agrippa's Relativity mode, the concept of relativity also involves the subject
judging and the things observed together with the object. The omission of the
third component is puzzling if we try to match the first taxonomy with the
second taxonomy.
The Ten Modes are found in three ancient sources. The earliest record is
found in Philo's work, On Drunkenness (c.20 B.C. to 45 A.D.). Sextus' account is
the second source. Diogenes' Lives of Eminent Philosophers contains the third
account. Annas and Barnes have conducted a detailed survey of the Ten Modes
24

exhausting the three accounts. According to them, Sextus' articulation of

Aenesidemus' Relativity mode is corrupt; he possibly not only 'ejects'
Aenesidemus' original Relativity mode but also replaces it by Agrippa's
Relativity mode. Therefore, the distinction between Aenesidemus' Relativity
mode and that of Agrippa is blurred or even disappears in Sextus' presentation.
If Annas and Barnes are right, then Agrippa's Relativity mode does not draw on
that of Aenesidemus. On the contrary, Agrippa's Relativity mode is the source
of Sextus' articulation of Aenesidemus' Relativity mode. As a result, what we
have in the PH is a distorted account of the Ten Modes in which Agrippa's
Relativity mode is made the most generic, embodying Aenesidemus' other
modes. It encourages the speculation that Agrippa intends to replace the Ten
Modes by his Five Modes.13
On the one hand, Sextus remarked that the Five Modes and the Ten Modes
were used together to refute the rashness of the Dogmatists. However, on the
other hand, as Annas and Barnes have suggested, he ejected Aenesidemus'
Relativity mode and replaced it with Agrippa's Relativity mode. To make
things worse, he went further to subsume Aenesidemus' other nine modes under
Agrippa's Relativity mode. Given the apparent superiority of Agrippa's
Relativity Mode (and the Five Modes in general), what else can the Ten Modes
do for the sceptic?
To induce epoche, the Ten Modes appeal to the appearance of dissimilarity
in different contexts. Yet, they all conform to the same structure as laid down by
Agrippa's Relativity mode. Annas and Barnes have argued that presumably
Sextus' identification of the Agrippa's Relativity mode as the most generic

13

See Hicks' note on Diogenes [1950] p.500. The phase he quoted does not support his claim

that Agrippa intends to 'replace the ten modes by his five'.

25

suggests that he was aware of the common structure of the Ten Modes.
However, as his version of Aenesidemus' Relativity mode draws on Agrippa's
Relativity mode, it follows that Agrippa must have recognized the common
structure of the Ten Modes when he introduces his Relativity mode. It could be
the case that after studying the Ten Modes thoroughly, Agrippa came to realize
that all the modes conform to the same structure. By capturing the most
essential elements of the common structure, he thus came up with his version of
Relativity mode. If this speculation were right, then Agrippa's Relativity mode,
being the refinement of the Ten Modes, would apparently be superior to the Ten
Modes.14
The development from Aenesidemus' Ten Modes to Agrippa's Five
Modes can be put under the perspective of evolution of the Pyrrhonian modes.
To begin with, Pyrrho 'attached himself to Scepticism more systematically and
more conspicuously than anyone before him' (PH I 7). About some two
hundred years later, Aenesidemus, as a reaction to the Sceptical Academy,
produced the Ten Modes to revive Pyrrhonian scepticism. Then with his Five
Modes, Agrippa skilfully refined the way the Pyrrhonist induces epoche. The
Ten Modes are rather narrow in scope as they are confined to oppositions of
appearances. The Five Modes are more wide-ranging. They are virtually
capable of rendering any belief claim rationally groundless.
Philosophically, the Five Modes exhibit a much deeper and a much more
theoretical understanding of the nature of justification. All sorts of oppositions
fall under the mode of Disagreement. The Relativity mode captures the most
essential technique of how the Ten Modes bring about epoche. The mode of

14

What about Aenesidemus himself? Did he notice the common structure of his modes? This is

a crucial question about which I do not want to speculate.

26

Infinite Regression, the Hypothetical mode and the Reciprocal mode focus on
the structure ofjustification by which a belief claim is established as a source of
conviction. The Five Modes could be regarded as masterly regimented tropoi
combating the rashness of the Dogmatists in highly skilful and varied ways. The
'regiments' can fight against the Dogmatists individually or in a group. When
they work in a group, they are more powerful. For instance, as we shall see in
the next chapter, the mode of Infinite Regression, the Hypothetical mode and
the Reciprocal mode can work together to make seemingly unbearable demands
for rational justification. If a Dogmatist does not want to be caught up in circular
reasoning, then he may be forced to find himself becoming entangled with the
endless urge to give infinite reasons to justify his belief claim. He may
otherwise stop giving proof. But then the belief claim is left unsupported. The
Five Modes do not replace the Ten Modes. In an analogy, the Five Modes are
highly competent all-round regiments while the Ten Modes are specialized in
combating the Dogmatists with oppositions of appearances. They may overlap
in one or two aspects but the Five Modes need not replace the Ten Modes in the
sceptical force. There could be division of labour between the Five Modes and
the Ten Modes.

27

Akatalepsia

OW I WANT to finish this chapter with an attempt to challenge

what I take to be a misconception about Pyrrhonian scepticism.

Akatalepsia eliminates all hope of bringing an enquiry to a fruitful

end. Continuing the enquiry would not bring people any closer to the answer.
There is actually no answer to be discovered. So, to make his conclusion or

recommendation of epoche compelling if not irresistible, it seems strategically

desirable for the sceptic to appeal to akatalepsia. 'Things are unknowable! So
why bother to insist on pursuing the enquiry? Better suspend judgement if you
want peace of mind.' Obviously the sceptic could not have asserted that things
are unknowable without being dogmatic about akatalepsia. But this line of
thought captures the general appeal of Paula Gottlieb's arguments. Her
arguments challenged Barnes [1990] on disagreement which is anepicritos
(mentioned in the mode of Disagreement). She criticized Barnes for
underestimating 'the sceptic's ingenuity' when he argued that

anepikritos

disagreement is undecided disagreement instead of undecidable disagreement.

She sketched his position as follows:

Barnes argues that the sceptic thought that suspension of belief (epoche)
follows from undecided {anepikritos) disagreement. Barnes takes issue
with those who would translate anepikritos as 'undecidable'. He argues
that Pyrrhonian sceptics such as Sextus, unlike the Academics, did not
take epoche to be the same as akatalepsia (unknowability). The
Pyrrhonian sceptics, he argues, were in principle tolerant of future
progress, although they did not undertake research themselves. Barnes
admits that often Sextus says, or at least appears to say, that Pyrrhonism

28

embraces unknowability. He says, 'These passages are admittedly

puzzling. But most of them can, I think, be explained away (Sextus does
not actually mean what he appears to say); and the rest may be put down
to carelessness.' (p. 10)

Gottlieb also stated that 'Agrippa's most important and exciting

innovation in the sceptical tradition [via the Five Modes] was his closing off of
any possible avenues of further inquiry'. 1 Therefore she complained that
Barnes had made the sceptic's conclusion of epochs too weak to be accepted by
other people, as it was supported by undecided disagreement only.

The Pyrrhonian sceptic's goal was ataraxia, a peaceful state of mind, a

state undisturbed by needless worries. It is not clear why the sceptic
should expect to induce this state in his hearers if his conclusion was as
weak as Barnes suggests. To be sure, the sceptic could not say that any
question is unknowable without entering the dogmatists' camp.
Nevertheless, the sceptic would surely want his listener to assume that
there is no good reason to continue the search. It is not clear how the
sceptic's weaker conclusion will prompt the listener to stop worrying
about finding an answer.

Gottlieb appeared to argue that people would stop worrying about finding
the answer only when they believe in akatalepsia. And it was the sceptic's job to
make people see that things are unknowable and there is no good reason to
continue the search. Otherwise, it would be difficult for the sceptic to induce

15

Gottlieb [1992] reported that this had been an argument made by Victor Brochard in Les

Sceptiques Grecs, Paris: Imprimerie Nationale, 1887, book iii ch. 6, especially page 304.

29

ataraxia in his hearers if they still believed that things were knowable and thus
that it was worthwhile to continue the search.
I shall argue that this line of argument is wrong-headed. It is founded on a
few mistaken assumptions about epoche: (1) epoche is the result of akatalepsia,
(2) epoche entails cessation of enquiry and (3) epoche is compatible with the
conviction of akatalepsia. The argument also fails to understand ataraxia. It
could be granted that akatalepsia would better prompt people to stop worrying
about finding an answer. But I shall argue that it is only in a very restricted sense
that the Pyrrhonist would tolerate this intrusion of akatalepsia to bridge the gap
between epoche and ataraxia.
According to Sextus, ' [Pyrrhonian] Scepticism is an ability (dunamis
antithetikos) to set out oppositions among things which appear and are thought
of in any way at all, an ability by which, because of the equipollence in the
opposed objects and accounts, we come first to epoche and afterwards to
ataraxia.' (PH I 8) When confronted with the equipollence in the opposed
accounts, people will naturally find 'none of the conflicting accounts takes
precedence over any other as being more convincing'. Epoche follows; the
intellect comes to a standstill, neither rejecting nor positing anything. It is not
clear why 'the sceptic', as Gottlieb argued, 'would surely want his listener to
assume that there is no good reason to continue the search'. Actually, we cannot
find Sextus urging people to be pessimistic about their enquiry so as to induce
epoche. In Sextus' account, epoche is the natural result of equipollence in the
opposed objects and accounts; one does not need to call upon akatalepsia in
order to induce epoche. The sceptic's job is not to make people see that things
are unknowable. His job is to set out oppositions among things to bring about
equipollence, so that people will then come to epoche.

30

After all epoche is only a standstill of the intellect in which one neither
rejects nor posits anything. In other words, one just does not know which
propositions deserve assent and which do not. There is no reason why epoche
should bar one from continuing the search. On the contrary, it is precisely the
fact that one does not know which propositions deserve assent and which do not
that motivates one to set off and continue the search. In addition, Mates [1996]
also argued 'certainly some searching is required to bring the Skeptic into a state
of aporia and from there to epoche, but there seems to be no reason why, just
because he is withholding assent, he must close his mind to all further
consideration of the matter in question.'(p.226)
In reality, it is very common for people to suspend judgement while
pursuing an enquiry. For instance, when confronted with extremely puzzling
and difficult issues such as the origin of human beings or the coming into being
of the universe, or the nature of consciousness some people may find it difficult
to make up their mind and so they should suspend judgement. Some people are
optimistic and they do not think that the issues are unknowable. Among them,
some may take sides with a certain theory. The rest may find none of the
existing theories compelling and so they suspend judgement and hope for
discovery in the future. They may engage in the search for discovery or they
may leave it to other people. Some other people may be pessimistic about
finding the answers. They may think that these issues are undecidable in
principle. But then they are not suspendingjudgement. Their intellect is not in a
standstill. When they think that some issues are undecidable in principle, they
are actually positing something. An epoche founded on the conviction of
akatalepsia is unorthodox and the very conviction of akatalepsia is actually
incompatible with epoche. Gottlieb's argument is seriously mistaken.

31

Gottlieb advanced

a further

argument in which the

anepikritos

disagreement mentioned in the mode of Disagreement could be made

undecidable if other modes are involved:

although only suspension of judgment is warranted by the fact of

undecided disagreement alone, unknowability would be warranted if the
other sceptical arguments, that attack any possible standards of
justification for accepting one side of a dispute over another, were
successful. Ataraxia would therefore be the outcome of all the
arguments together, rather than just the argument from disagreement.
Whether Sextus is claiming (or can justifiably claim) that a dispute is
merely undecided or whether it is undecidable may depend on how
much of the sceptic's extra machinery he is wheeling into the argument
from disagreement.

Gottlieb's argument jumps to the conclusion of akatalepsia.

Strictly

speaking, even if the sceptic happens to attack possible standards of justification

for accepting one side of a dispute over another with all the Five Modes, what
would be yielded is not akatalepsia. The attack, if successful, should only force
one to concede that one does not know which propositions deserve assent and
which do not and epoche follows. One may suspend judgement and continue the
search. But if one is convinced by the attack that the disagreement is
undecidable in principle, one has already made his judgement. Epoche, as I have
argued, is incompatible with the conviction of akatalepsia.
Now I shall expose a fundamental problem with Gottlieb's line of
argument. The intrusion oi akatalepsia suggests a Pyrrhonian scepticism which
is foreign to Sextus's accout. When Sextus introduced Pyrrhonian scepticism in

32

the first place, he explicitly stated, 'the sceptics are still investigating'. (PHI 3)
I do not see why we should ignore this written 'declaration'. In fact, it would be
very difficult to set aside this statement. After all, The Greek adjective skeptikos
originally derives from a verb meaning 'to inquire' or 'to consider'.16 One may
argue that the word need not be taken so strongly as to reject the interpretation
in which epoche entails cessation of enquiry. It might also be the case that the
sceptic pays only lip service to the cause of enquiry and they do not undertake
research themselves. Nevertheless due weight should be given to the recurring
emphasis that the sceptic wants to be associated with the pursuit of enquiry. In
addition, Sextus differentiated Pyrrhonian scepticism from the New Academy
precisely on the basis that the Pyrrhonists 'are still investigating' hoping that it
is 'possible for some things actually to be apprehended' while 'the school of
Clitomachus and Cameades, and other Academics, have asserted that things
cannot be apprehended.'17 Even so, Sextus' account of the New Academy may
not be historically accurate. At any rate, the repeated contrast between the
sceptics who were described as still investigating and the New Academy that
was said to be committed to akatalepsia demonstrated the fact that Sextus was
keen on keeping the notion of akatalepsia away from Pyrrhonian scepticism.
There is no reason why we should overlook his effort. Therefore, I suppose
Pyrrhonian scepticism should not be associated with akatalepsia.
Moreover as I have argued in the earlier discussion of the Relativity mode,
if a sceptic happens to say something like 'all things are unknowable', what he
means should be simply that 'all things appear to him to be unknowable'. He
uses 'are' in a non-epistemic sense. He does not assert that things are

16

Annas and Barnes [1985] p.l.

17

PH11-3and 226
33

unknowable. He is just reporting his state of mind. It seems to him that all things
are unknowable. It is only in this restricted sense that the sceptic would entertain
the impression of akatalepsia. Actually things are unknowable for certain
dogmatists, but not for the sceptic. The sceptic takes the positions of the
Dogmatist to conclude that things are unknowable for the Dogmatist. But as to
whether or not all things are really unknowable, the sceptic should suspend
judgement.
Even if a disagreement strikes the sceptic as undecidable at the moment, he
might assent to his impression but this does not admit of akatalepsia. The
sceptic would only suspend his judgement. When the sceptic cannot refute a
present argument propounded to him, he will reply by appealing to the future.

Before the founder of the school to which you adhere was born, the
argument of the school, which is no doubt sound, was not yet apparent,
although it was really there in nature. In the same way, it is possible that
the argument opposing the one you have just propounded is really there in
nature but is not yet apparent to us; so we should not yet assent to what is
now thought to be a powerful argument. (PHI 34)

The Pyrrhonists 'are still investigating' hoping that it is 'possible for some
things actually to be apprehended'. So they will leave it open that the seemingly
undecidable disagreement might become decidable in the future.
The motivation behind Gottlieb's line of argument is to make the
Pyrrhonian conclusion or recommendation of epoche compelling. It is assumed
that a compelling epoche founded on the conviction of akatalepsia will prompt
people to stop worrying about finding an answer and thus achieve ataraxia
more effectively. There is substance in this line of thought. In the first place, the

34

Five Modes, as we shall see in Chapter Three, do give one all the resources one
needs to undermine all claims to justified belief. And akatalepsia seems follow
naturally. In addition, if, as we have seen in the passage, it is always possible to
come up with counter-arguments in the future, it seems things are undecided in
principle.
The problem I have been discussing concerns epoche, ataraxia and
akatalepsia.

I have argued that an epoche founded on the conviction of

akatalepsia is unorthodox and actually the very conviction of akatalepsia is

incompatible with epoche, not to see this is not to understand what Pyrrhonian
scepticism is all about. It is assumed that akatalepsia would better prompt
people to stop worrying about finding an answer. But consider this:

For Sceptics began to do philosophy in order to decide among appearances

and to apprehend which are true and which false, so as to become tranquil;
but they came upon equipollent dispute, and being unable to decide this
they suspend judgement. And when they suspended judgement, ataraxia
in matters of opinion followed fortuitously. (PH I26)
Now the Sceptics were hoping to acquire tranquillity by deciding the
anomalies in what appears and is thought of, and being unable to do this
they suspended judgement. But when they suspended judgement,
tranquillity followed as a shadow follows a body. (PH I29)

In the first passage, the sceptic comes to epoche and then ataraxia follows
fortuitously. In the second passage, the sceptic comes to epoche and then
ataraxia follows as a shadow follows a body. There is inconsistency between
the image of a shadow following a body which suggests a natural and
inseparable connection between epoche and ataraxia and the passage, which

35

states that the coming of ataraxia in these circumstances is fortuitous. The

intrusion of akatalepsia is intended to bridge the gap between epoche and
ataraxia. But as I have argued, the intrusion risks not only minsunderstanding
the notion of epoche but also suggesting a Pyrrhonism which is foreign to
Sextus' account. The Pyrrhonian sceptic may not grapple for a solution with
determined effort. He may just pay lip service to the pursuit of enquiry. But one
thing is clear: the Pyrrhonist induces epoche on the strength of the Five Modes
and this should be done without invoking akatalepsia or closing off possible
paths of further enquiry.
The Five Modes appear capable of compelling people to concede that they
do not have the legitimate justification for accepting one side of a disagreement
over another. Epoche would prevail. But two questions need to be asked. How
can the Five Modes do this? Was the sceptic of the PH obligated by his
Pyrrhonian scepticism to destroy possible standards of justification in all fields
of study? I shall pursue these two questions in Chapter Three and Four
respectively. In addition, the sceptic is accustomed to the accusation that his
Pyrrhonian scepticism would make him inactive. However, the sceptic replies
that he lives by following the everyday observances that consist of 'guidance by
nature, necessitation by feelings, handling down of laws and customs, and
teachings of kinds of expertise'. (PH I 23) I shall also explore this issue with
reference to epoche in Chapter Four.

36

CHAPTER THREE

The Agrippan Problem:

A Pyrrhonian Challenge to Epistemic Justification

OST PHILIOSPHERS AGREE that Agrippa's Five Modes pose

profound problems for the theory of justification. According to

Barnes, Agrippan argumentation, with its forms and structures,

'were among the most important aspects of Pyrrhonism, so that to study them is

to study the soul of ancient scepticism.'1 Historically, they 'had a unique

influence on the subsequent history of sceptical enquiry, and hence, more
generally, on the history of epistemology or the enquiry into the nature and
scope of human knowledge: the Agrippan forms lie at the heart of the western
philosophical tradition.' Philosophically, they 'remain today among the central
issues in the theory of knowledge; that every modem epistemologist must take
notice of them; and that they still provide the subject of epistemology with some
of its most cunning puzzles and most obdurate problems.' Fogelin shared a
similar opinion with regard to the Five Modes. He observed, 'there is an
uncanny resemblance between problems posed by Agrippa's Five Modes and
those that contemporary epistemologists address under the heading of the
theory of justification..'19 He was so impressed that he is prepared to 'define the
philosophical problem of justification as the attempt to take seriously and then

18

Barnes [1990] p.ix.

19

Fogelin [1994] p. 11
37

avoid the consequences of Agrippa's Five Modes' despite the fact that most
writers on the subject seem never to have heard of Agrippa and his Five Modes.
This chapter addresses the Five Modes inasmuch as they constrain theories
of justification. The attack on the criterion of truth (PH II 18-20) serves as an
excellent example to demonstrate how the Five Modes block every way out of
the sceptical predicament of justification. I mention different contemporary
formulations of the problem posed by the Five Modes. Much weight is given to
Barnes's system of three modes and his unparallelled appreciation of the
Hypothetical mode. However, I also give sufficient attention to Fogelin who
argued that the Agrippan problem poses epistemic challenges in an even-handed
way. Finally, I discuss how the Five Modes pose a seemingly unanswerable
problem for the justification of belief.

38

The Problem and The Attack on the Criterion

GRIPPA'S FIVE MODES raise serious questions about the

foundation of belief. The Modes can work individually or in varied
groups of two to five, as Sextus has demonstrated elsewhere in PH.

When the modes work in a group, they are more powerful. In such cases, they
make successive demands for justification. Epistemic justification runs three

risks: i.e., infinite regression, circularity, and unwarranted assumption. These

dangers seem insuperable. They may lead us to think that no belief claim is
justified. Suppose there is disagreement over the trustworthiness of a belief
claim. The belief claim is trustworthy only if it is justified. Intuitively, a
justified belief is more likely to be true than an unjustified belief. But if the
belief claim is to be justified, its justification needs to be grounded on a further
justification. For a justified justification would better establish the truth of the
belief claim. The further justification would also need to be justified. The
process goes on to form a need for a chain ofjustification. If the end of the chain
turns back to an earlier section of the chain, circular justification results. If the
chain of justification goes on and on, an infinite regress threatens. If it stops
somewhere, unwarranted assumption instead of justification results.
The attack on the criterion of truth (PH II 18-20) serves as an excellent
example of how the Five Modes constrain attempts to justify a belief claim.

Of those who have considered the matter, some, for example, the Stoics
and others, have asserted that there is a criterion; others, including the
Corinthian Xeniades and Xenophanes of Colophon,... have asserted that
there is not; while we have suspended judgment as to whether there is or
not.

39

This dispute, then, they will either declare to be decidable or to be

undecidable; if undecidable, they will be granting at once that judgement
should be suspended; but if decidable, let them say with what it is to be
decided, seeing that we do not have nay agreed-upon criterion and do not
know indeed, are inquiring
whether one exists.
And anyhow, in order to decide the dispute that has arisen about the
criterion, we have need of an agreed-upon criterion by means of which
we shall decide it; and in order to have an agreed-upon criterion it is
necessary first to have decided the dispute about the criterion. Thus, with
the reasoning falling into the circularity mode, finding a criterion
becomes aporetic; for we do not allow them to adopt a criterion
hypothetically, and if they wish to decide about the criterion by means of
a criterion we force them tinto an infinite regress.
Further, since proof requires a criterion that has been proved, while the
criterion has need of what has been determined to be a proof, they land in
circularity. (PH II 18-20)

This example

involves the modes

of Disagreement,

Reciprocal,

Hypothetical, and Infinite Regression. The task of the Disagreement mode is

twofold. It first induces epoche by showing that there are incompatible claims
concerning the criterion of truth and so epoche follows. It also calls for
justification from those who insist on a certain position instead of suspending
judgement in the light of the disagreement. The Relativity mode could serve the
same purpose. But it is not used in the example. The other three modes proceed
systematically to render different belief claims about the criterion of truth
unjustified by denying unwarranted assumption and justification that is circular
or regressive infinitely.

40

In schematic terms, suppose there is disagreement about the belief claim

that p. Either epoche follows or reasons will be given to support one side of the
disagreement. If p is asserted without any proof, the Hypothetical mode applies.
If p is supported by q, then q itself needs support too or else the Hypothetical
mode applies again. The mode of Infinite Regression rejects infinite sets of
reasons. The reciprocal mode does not allow circular reasoning. Therefore, if
we are to accept the belief claim that p, p must not be merely asserted nor, at the
same time, be supported by a set of circular or infinite reasons. Otherwise, p
could not be accepted. Thus, the key to the Agrippan problem, as it appears, is to
find a way to justify the belief claim that p in a non-hypothetical manner and
without invoking reasoning that is infinitely regressive or circular. But how
could this be done?

The System of Three Modes and The Hypothetical Mode

HERE ARE different formulations of the Agrippan problem.

Chisholm discussed the formulation that Descartes had put forward

in his reply to the VIIth set of objections and that which Coffey [1917]

took up in his discussion of the problem of the criterion. This version of the

problem is concerned with 'the proper method for deciding which are the good
beliefs and which are the bad ones which beliefs are genuine cases of
knowledge and which beliefs are not'. 20 Amico, following Chisholm, has
discussed the problem of the criterion; he described it as 'a metaepistemological

20

Chisholm [1973] p.10

41

problem concerning the justification of first order knowledge claims among

disagreeing disputants'.21 Fogelin [1994] mentioned two 'modern restatements
of the Agrippa problem' made by Laurence Bonjour [1985] and Paul K. Moser
[1985]. Combining three of Agrippa's Five Modes, Barnes advances a tactical
Pyrrhonian mode of argumentation. He believes that his System of Three
Modes 'conveys what is epistemologically most important and most
challenging about this [Agrippan] aspect of ancient Pyrrhonism'.23 He claims
that the System of Three Modes is his own invention in a sense, although Sextus
exhibits a different system of three modes. Sextus' system consists of the modes
of Disagreement, Infinite Regression and Reciprocal whereas Barnes assembles
the modes of Hypothetical, Infinite Regression and Reciprocal to constrain
theories of epistemic justification. The mechanism of Barnes's system is given
as follows.

Suppose you are considering the claim that P. Then either (1) the claim is
merely asserted, or else (2) it is supported. If (1), then the hypothetical
mode applies.
If (2), then P rests on some reason or set of reasons, Rl. Either (2a) Rl is
an "old" item, i.e. (in this case) it is the same as P, or else (2b) it is a new
item. If (2a), then the reciprocal mode applies.
If (2b), then either the (2bi) Rl is merely asserted or (2bii) Rl is supported.
If (2bi), then the hypothetical mode applies. And so on ... until the
regressive mode is invoked. (P.l 19)

21

Amico[1993]p.l43

22

Fogelin [1994] p.l 17

23

Barnes [1990] p.l 19

42

Apparently, there is division of labour between the three modes in the

system. The Hypothetical mode first asks for justification. The Infinite
Regressive mode forces people to embark on an infinite series of justifications
and the Reciprocal mode does not allow circular justification. The Hypothetical
mode recurrently shuts the doors leading the way out of the predicament.
Whenever one attempts to assert p without any reason, however self-evident p
may appear to be, the Hypothetical mode does not allow it; it requires further
reasons for p. In Barnes's system, the challenge comes largely from the
Hypothetical mode. According to him, the Hypothetical mode 'is a mode of the
first importance to the Pyrrhonists'. It is so important that he observes
'numerous implicit references to hypotheses' in Sextus' exposition of
Pyrrhonian scepticism.24
Sextus briefly explained how the Five Modes lead to epoche when he
introduced them. But he did not explain how the Hypothetical mode induces
epoche. Because of anepikritos disagreement, we cannot choose what to believe
and what to disbelieve and hence 'we end up with epoche. Because of infinite
regression in our reasoning, we 'have no point from which to begin to establish
anything, and epoche follows'. Since things appear relatively different to people
who perceive them, 'we suspend judgement on what it is like in its nature'.
Finally we suspend judgement 'when what ought to be confirmatory of the
object under investigation needs to be made convincing by the object under
investigation'. However, it was not explained how the Hypothetical mode
would lead to epoche. Sextus simply revealed that the sceptic would bring in the
Hypothetical mode when the dogmatists 'begin from something which they do
not establish but claim to assume simply and without proof in virtue of a

24

Ibid, p.96

43

concession'. In my earlier discussion of the Hypothetical mode in Chapter Two,

I noted that the Hypothetical mode represents the Pyrrhonian sceptic's
uncompromising and demanding attitude to the fundamental presupposition of
hypothesizing, which suggests that we must take the most elementary of things
as of themselves entitled to credence when they are not postulated. Now I want
to finish my discussion of the Hypothetical mode to see how it gives rise to
epoche and what role it plays in the sceptical framework.
Sextus' repeated criticisms of hypothesizing reveal implicitly the
problematic feature of hypothesizing in the process ofjustifying a belief claim.

(1) If it is acceptable for a Dogmatist to hypothesize that p, i.e. to lay down

p, by a bare assertion, as a first principle, then it must be equally
acceptable for a sceptic - or another Dogmatist to hypothesize that p*,
where p* is the "opposite" ofp. But if p* is no less acceptable than/?, we
cannot accept p as a first principle just because the Dogmatist
hypothesizes it. (See PH I 173; MVIE 370; M III 8)
(2) What the Dogmatists hypothesize is either true or false. If it is true,
they should not hypothesize it (for hypothesis is "a matter full of
suspicion") but rather assume it straight off. If it is false, it can do them no
good for a false starting-point cannot ground a science or a branch of
knowledge. (SeePHI 173;M VIH 371; M III 9-10)
(3) If the Dogmatists hold that the consequences of any hypothesis are
acceptable, then all enquiry is subverted. For, given any absurd
proposition, we can find some hypothesis from which it follows; hence
any proposition whatsoever will be acceptable. And this is evidently silly.
(SeeM VIII 372-3; M III 11-12)
(4) If in order to establish that p 2 you first hypothesize thatpl and then
derive p2 from pl, why not establish p2 directly, by hypothesizing it, and

44

thus save yourself the labour of looking for arguments? (See PHI 174; M
Vm374;M III 13)

Barnes collected the arguments from various places in Sextus' works.25 He

suggested that the first argument is the most important one and Sextus had
frequently alluded to it in several places. The quality of being unsupported is a
defining characteristic of hypothesis and an unsupported assumption by itself
can never rule out its opposite unless it appeals to something external e.g., its
consequences and that of its opposite. If we can infer some true propositions
from p but we cannot do so with the opposite of p, then we could judge that we
should not hypothesize the opposite of p. But then we are actually grounding p
on some other true propositions; p would no longer be a hypothesis. And of
course, the sceptic could proceed with his other modes. Moreover, it is pointless
to establish other propositions first and then establish p based on the established
propositions. For other true propositions are obtained by hypothesizing p. We
will be subject to the Reciprocal mode if we obtain some true propositions by
hypothesizing p and, in return, justify p on the ground of those true propositions.
Therefore, 'If the only thing that can be said for or against p is that some
Dogmatist has hypothesized it, and if hypothesizing that p does not establish or
warrant belief in p, then we should suspend judgement over p". (p.99) This is
how the Hypothetical mode induces epoche.
Barnes argued that 'the hypothetical mode is more closely connected to
epoche than is either the regressive mode or the reciprocal mode', (p. 108) His
reason was that while the two modes induce epoche when there are bad
arguments, either regressive or circular, the Hypothetical mode induces epoche

25

Ibid. p. 100

45

just because of the defining characteristic of hypothesizing. His account of the

Hypothetical mode is closely relevant to his argument that 'the deep and
fundamental

issue raised by Agrippan scepticism' was 'the status of

epistemological externalism'. (p.141) I shall return to this point in the next

section.
Fogelin [1994] held a different view on the role played by individual
modes in the Pyrrhonian challenge to epistemic justification. Unlike Barnes,
who regarded the Hypothetical mode as the leading mode in constraining
epistemic justification, Fogelin argued that the sceptic posed the Agrippan
problem in an even-handed way. Therefore, he complained that the Agrippa
problem is often narrowly characterized as the infinite regress problem in
contemporary literature on epistemic justification. He pointed out that

the problem that presents itself is not simply that of avoiding a bad
infinite regress; the challenge is to avoid this regress without falling
into a bad form of circularity or a bad form of unjustified acceptance...
If we think the threat of an infinite regress of reasons as the central
challenge to justified belief, then theories, despite their own
difficulties, may lay claim to our acceptance just because they seem to
deal with this single aspect of the Agrippa problem. If, however, we
begin with an antecedent horror of circularity, an appeal to an infinite
regress might recommend itself as a way out. It is important, then, not
to grant unwarranted dialectical advantages, but to insist, instead, that
a philosophical theory of justification must simultaneously avoid
involvement in a bad infinite regress, in a bad form of circularity, and
in a bad appeal to unwarranted assumption. The Agrippa problem
poses these challenges in an evenhanded way. (p. l14)

46

The Agrippan Problem and Modern Philosophy

OGELIN GAVE two reasons in terms of what he called Cliffordism

and the Cliffordian project to account for the significance of the

Agrippan problem in modern philosophy. He proposed to call 'a

strong commitment to a strong normative principle of epistemic justification'

Cliffordism.26 In the words of W. K. Clifford, [1879] 'It is wrong always,

everywhere, and for anyone, to believe anything upon insufficient evidence.
(2:186)' Essentially, Cliffordism is a commitment to high standards of
epistemic responsibility. Any epistemic act that falls short of the standards
amounts to epistemic

irresponsibility.

Fogelin

found

that

different

epistemologists embraced this Cliffordism. Both Moser and Bonjour expressed

this attitude to belief in terms of epistemic irresponsibility. For Moser [1985],
'To accept a proposition in the absence of good reason is to neglect the
cognitive goal of truth. Such acceptance, according to the present normative
conception of justification, is epistemically irresponsible, (p.4)' And for
Bonjour

[1985], 'To accept a belief in the absence of... a [good] reason,

however appealing or even mandatory such acceptance might be from some

other standpoint, is to neglect the pursuit of truth. My intention is that the idea of
avoiding such irresponsibility, of being epistemically responsible in one's
believing, is the core of the notion of epistemic justification, (p.8)'
The task of the Cliffordian project was to demonstrate how it would be
possible for us to meet the high standards of epistemic responsibility. We may
think of ourselves as knowing all sorts of things. This body of knowledge may
satisfy our ordinary understanding of knowing things. However, it remained to
26

Fogelin [1994] p.l 14

47

be seen if they also lived up to the high standards of epistemic responsibility set
by Cliffordism. This was the task that theorists of epistemic justification had
assigned to themselves. 'For the Agrippa problem to emerge as a serious
problem, the Cliffordian must further believe that knowledge does exist or at
least could exist.'(p.115) The sceptic did not set high standards of epistemic
responsibility; the sceptic simply drew the dogmatists' attention to the
constraints of infinite regress, circularity and unwarranted assumption.
Fogelin discussed how a number of theories of justification had failed to
resolve the Agrippan problem. He suggested that 'It is not possible to deal in
detail with all theories of justification that have been presented there has been
rather an explosion of them in recent years', (p.l 19) Therefore, he contended
with 'the strongest representative samples of the main types of theories of
justification', viz., Chisholm's foundationalism, Bonjour's internal coherentism
and Davidson's external coherentism. He noted that the justificationalists
tended to 'exploit the fundamental structure of the Agrippa problem' in order to
evade the problem.27 Roughly speaking, it is done by 'a complex disjunctive
syllogism eliminating all contending positions, followed by a constructive
attempt to show that the remaining Agrippan mode (properly understood and
qualified) provides a way of solving the Agrippa problem.' For instance,
Bonjour, as a coherentist, dwelled on the inadequacy of unwarranted
assumption and infinite regress but supported circularity as the only desirable
process of justifying a system of beliefs. Chisholm and Moser discussed the
problems with regard to circularity, infinite regress and unwarranted
assumption, and they concluded with foundationalism as the only feasible
option.

27

Ibid, p.l 17

48

Fogelin set three success conditions for the justification theories. The first
condition was that 'the presentation of a theory of justification should be
governed by... the principle of philosophical candor. In conformity with this
principle, the author should specify, as desiderata, just which beliefs she takes to
be justified, and which not, for without this specification the scope of the theory
will be indeterminate.'(p. 118) The second condition required the theory to show
in some detail how the supposedly justified beliefs were justified. For the third
condition, 'An answer to the Agrippa problem may not beg the question by
assuming for argumentative purposes that there must be some positive solution
to it.'(p.119) Fogelin complained that epistemic justificationalists hardly ever
insisted upon these three success conditions. He went even further to suspect
that 'There exists what might be called "the Epistemologists' Agreement" not to
hold each other to such standards, perhaps because it was tacitly understood that
no theory could meet them.'
Interestingly enough, Chisholm in The Problem of the Criterion claimed
that the problem could be resolved only by begging the question.

What few philosophers have had the courage to recognize is this: we can
deal with the problem [of the criterion] only by begging the question. It
seems to me that, if we do recognize this fact, as we should, then it is
unseemly for us to try to pretend that it isn't so. One may object:
"Doesn't this mean, then, that the sceptic is right after all?" I would
answer: "Not at all. His view is only one of the three possibilities and in
itself has no more to recommend it than the others do. And in favor of
our approach there is the fact that we do know many things, after all.
(pp.37-8)

49

It is no wonder that at the end of his examination, Fogelin concluded that

Chisholm's foundationalism, together with Bonjour's internal coherentism and
Davidson's external coherentism had all failed to meet the success conditions.
He remarked that 'no justificatory program seems to show the prospect of
solving the Agrippa problem.'(p.l93) But he did not conclude that the Agrippa
problem definitely resists solution. 'It is possible that someone will produce a
wholly new sort of theory of empirical justification that will provide a
satisfactory solution to the Agrippa problem, or perhaps someone will
accomplish this through hitting upon an utterly novel way of developing one of
the traditional theories of empirical justification. It would be an unseemly
dogmatism to rule these possibilities out in advance.'(p.l94)
Barnes discussed the Agrippan problem with epistemological externalism.
Although the Dogmatists did not develop any externalist theory deliberately, he
argued, 'it is most plausible to construe the Dogmatists as offering externalist
theories, i.e. as offering theories which in fact are externalist.'

The

externalism involved was ontological externalism. By the externalist account of

justification, a justification was external when the belief concerned was justified
by virtue of something external to the believer, like a real state of affairs, instead
of by virtue of other mental contents of the believer. Based on Sextus' reading
of Xenophanes of Colophon (M VII 52 and VIII 325), Barnes formulated an
argument against externalist epistemology.

Suppose a Dogmatist to adhere to an externalist epistemology of the sort

we may ascribe to the Stoics. Suppose him to assert that P (and to take
himself to be asserting a basic belief). Then his assertion may, of course,

28

Barnes [1990] p.137

50

quite well be true. Moreover, P may well be a member of the class [of
basic beliefs] . But - ex hypothesi he will not claim to know that P is a
member of the class

for he holds that he need not know that P is a

member of in order to be justified in claiming that P is true. The whole

point of the externalist theory is to allow the Dogmatist to claim that P,
when P is in

without also claiming that P is in

But in that case (so

Sextus might have urged), he cannot properly claim to know that P.

(pp. 139-40)

Barnes further argued 'the Dogmatist is relying on a sort of externalist

theory and that the sceptic is pressing the inadequacy of any externalist account
of knowledge or justified belief. And in this way the status of epistemological
externalism can be seen to be the deep and fundamental issue raised by
Agrippan scepticism.'(p.l41) According to him 'the Pyrrhonian [sceptic] is not
concerned with whether the Dogmatist knows that he knows that p.' Rather the
upshot was whether the Dogmatist as an externalist can claim that p. (p. 140)

The Pyrrhonian point is this. Suppose our Dogmatist continues to claim

that/? that this is an apple or that honey is sweet. The claim is advanced
as an "hypothesis". No reasons for it are given. It is offered as a putative
item of basic knowledge. The Pyrrhonist wonders if the Dogmatic claim
is justified. He runs through the arguments for the hypothetical mode,
and then asks the Dogmatists: "Well, do you think you're justified in
claiming that p?" What reply will he get? Now the Dogmatist will
actually be justified in claiming that p (on the externalist hypothesis)
provided that because p is in

[a class of "natural" beliefs] he believes

that p. Hence if he claims that he is justified in believing that p, he will in

effect be claiming that because p is in

51

he believes that p. But ex

hypothesi he will not make this further claim. For, as I said, the whole
point and purpose of his externalist invocation of basic belief is that he
may justifiably believe that p without making the further claim that
because p is in he believes in p. Thus whether or not he is justified in
claiming that p, he will not respond to the sceptic's challenge by
claiming that he is so justified, (p. 143)

Barnes concluded that 'if a Dogmatist is rationally to claim that/? (where p

is a basic belief), then he must also be entitled to claim that p is a basic belief. If
our Dogmatist is to remain a foundationalist and an externalist, then he must be
able to claim (1) that/?, and also (2) p is in

but he must not offer (2) as his

justification for (l).'(p.l44) Barnes gave a sceptical note on how the Dogmatist
can claim both (1) and (2).

Historical Development of Scepticism

S AN INTERLUDE before the conclusion of this chapter, I

want to mention the historical development of scepticism
from ancient to early modern philosophy with reference to

the question of the possibility of knowledge as depicted by Frede. Frede

suggested that the Pyrrhonian sceptics did not raise the question of the
possibility of knowledge; nor did they return the negative answer that
nothing could be known.29 Despite their differences, Frede proposed to

29

See' The Sceptic's Two Kinds of Assent and the Question of the Possibility of Knowledge' in

Bumyeat and Frede [1997] ppl27-151.

52

group Arcesilaus, Carneades and the Pyrrhonists under the heading of

'classical scepticism' as they exhibited the same feature, namely that they
did not take the position that nothing was, or could be, known for certain. He
argued people in the modern period have failed to appreciate classical
scepticism and identified scepticism in general with a limited and dogmatic
scepticism. So that nowadays, people thought of scepticism as narrowly
concerned with knowledge and certainty and the sceptic not only challenged
us with the question of the possibility of knowledge but also denied any
satisfactory answer to it. According to Frede, it was some other ancient
sceptics who, with their 'dogmatic scepticism' returned the answer that
nothing could be known. The question of the possibility of knowledge was a
dogmatic enterprise in ancient Greece.

For the dogmatic... something is at stake. It does make a great

difference to him whether his impressions really are true and whether he
has made a mistake in taking them to be true. For in actively giving
assent to them he has become responsible for them, and hence feels a
need to defend them and to prove them to be true. The dogmatic, in
taking a position, has made a deliberate choice, a hairesis, for which he
is accountable. But because so much is at stake for him, he no longer is
in a position openly to consider alternatives, to realize and accept the
weight of objections; he has become dogmatic in his attitude, (p. 138)

Dogmatists who wanted real and certain knowledge had raised the
standards for what was to count as knowledge. They called into question all the
truths we ordinarily went by and subjected the purported truth claims to some
unnecessarily strict canons. As a result, things that we ordinarily would count as

53

knowledge would no longer be qualified as knowledge. It was in this way that

the global contrast between appearance and truth or reality took shape. For
instance, we may think it was true that the sky was blue. But it turns out that on
the true theory of things this was a mere appearance. Indeed, there was a certain
configuration of atoms and particles in the atmosphere that may, or may not,
produce this appearance. Therefore, it was not true that the sky was blue in
reality, though it appeared to be so in appearance. The same thing may happen
for all other ordinary truths.
The classical sceptic, including the Pyrrhonian sceptic, on the contrary, did
not demand that anybody who claimed to know should have the right kind of
reason or justification which would not only support his belief but also allow
him to rule out all incompatible beliefs. The sceptic did not maintain that
knowledge had to be firm or certain. However, it was the dogmatic sceptic who
came up with the position that nothing could be known for certain.
Unfortunately, as time went by, people came to think of scepticism as
dogmatic and even associated the classical sceptics with the positions of the
dogmatic sceptics. As Frede depicted it, the process of historical misconception
began in late antiquity, passed through the medieval and lingered in early
modern philosophy.

It was largely a matter of ignorance that in late antiquity scepticism came

to be identified with dogmatic scepticism. In the Latin West this was, no
doubt, in good part due to Cicero's influence, who himself was a dogmatic
sceptic and who, moreover, would be the only substantial source
concerning scepticism available to those who did not read Greek. And
Cicero's influence was magnified by St. Augustine's authority, who for
his attack on scepticism in his Contra Academicos primarily, if not

54

exclusively, relied on Cicero, but unlike Cicero, gave no indication of the

possibility of a nondogmatic scepticism and treated Carneades as taking
the kind of position espoused by Cicero. And given Augustine's standing
far into early modern times, it is not surprising that the Western view of
scepticism should have been determined by him throughout the Middle
Ages, especially since for a long time his Contra Academicos would have
been the only readily available source which discussed scepticism in any
detail... It may also be of relevance in this context that the question of
knowledge became a live issue again in the late Middle Ages owing in
part to Ockham's doctrine of intuitive cognitions, (pp. 146-4)

Looking into Ockham's doctrine of intuitive cognitions and its objections,

Frede concluded that 'the question of the possibility of knowledge came to be a
live issue again more or less exactly in those terms in which dogmatic
scepticism had formulated it.'(p. 147) In addition, for some reasons, like inertia
and continuation of tradition and heritage, people went on with the question of
the possibility of knowledge, together with the historical misconception of
classical scepticism into the medieval and then to the modern time.
Finally, I want to sketch the view (also found in Frede) that something
internal to our epistemic practice made the problem of epistemic justification
resist solution. For the sceptic, assent did not require truth and people might
well act on this assent.

Assent may be a purely passive matter. It may be the case that human
beings work in such a way that impressions are more or less evident to
us. Evidence is a purely internal feature of our impressions. Now we
also attribute different importance to different questions. We might be
constructed in such a way that if we have an impression on a matter

55

whose degree of evidence does not correspond to the degree of

importance we attach to the matter, we naturally, unless we are
prevented, e.g., by lack of time or energy or have decided to take a risk,
go on to consider the matter further till we get an impression which has a
sufficient degree of evidence. It would not even have to be the case that
at a certain point we decide that we now have a clear enough impression
and stop to consider the matter further. It may just be the case that as
soon as we have a clear enough impression we, without any further
thought, act on it. And this may be all acquiescence and assent consist in.
(p. 135)

Every impression bears a degree of evidence. 'Evidence is a purely internal

feature of our impressions.' But as Frede suggested, other factors may affect
how we perceive the evidence. An impression whose degree of evidence did not
match up with the degree of importance we attached to the matter would trigger
reflection on the matter until we reached an impression that had a corresponding
degree of evidence. Likewise, a belief claim whose degree of evidence did not
correspond to the degree of importance we attached to the matter would initiate
contemplation. When the degree of evidence falls short of the degree of
importance, we might withhold the belief and go on to consider the matter in
detail.
Other factors may also affect our response to the evidence of our
impression or a belief claim. For instance, a strong commitment to high
standards of epistemic responsibility would require us to meet very high
standards of epistemic justification. Fogelin [1994] advanced the idea of levels
of scrutiny to account for our epistemic practice. The idea was that certain
circumstances demanded stricter scrutiny. 'Although our common justificatory

56

procedures do not demand that we eliminate all potential defeaters, it is part of

these procedures to have built-in mechanisms for dealing with epistemically
risky circumstances.'30 For instance, we would operate with a higher level of
scrutiny when we were dealing with an estate agent but not with, say, our
friends. Fogelin claimed that the problem of epistemic justification was so
difficult to resolve because reflection alone would heighten the level of scrutiny.
The circumstances may look all right. However, reflection might call for a
higher level of scrutiny to eliminate remote defeating possibilities. If we cannot
reject all the defeating possibilities, it seems that we do not know for certain or
our belief claim is not well justified. In his words, 'the theory of knowledge, in
its traditional form, has been an attempt to find ways of establishing knowledge
claims from a perspective where the level of scrutiny has been heightened by
reflection alone.'

31

The dogmatic sceptics privileged this heightened

perspective but the Pyrrhonian sceptic did not.

30

Fogelin [1994] p.92

31

Ibid, p.99
57

Conclusion

HE AGRIPPAN PROBLEM seems devastating. For how can any

theory of justification evade the constraints of infinite regression,
circularity, and unwarranted assumption at the same time? Anyone

who purported to justify his belief claims would be embarrassed by the

Agrippan problem. 'If the Agrippa problem cannot be resolved,' as Fogelin
noted, 'there is no reason to suppose that knowledge of the kind sought by

justificationalist philosophers exists.'

32

Some justification theorists claimed that the Agrippan problem could be

resolved. For instance, coherentism suggested that a properly construed
circularity could provide a practical platform for a system of beliefs.
Foundationalism appealed to some basic beliefs, which were thought of as
immune from the charge of unwarranted assumption. The basic beliefs were
justified non-inferentially and so they were ideal candidates to provide
inferential justification for other ordinary beliefs. 'The crucial question' as
Fogelin suggested, 'is whether any justificatory procedure is privileged in the
sense that justification of that kind is final.'(p. 195) Chisholm [1973] contended
that 'we can deal with the problem only by begging the question.' Barnes [ 1990]
reported that Aristotle was one of those philosophers who preferred having
knowledge even at the risk of hypothesizing. Thus the Aristotelians maintained
that the basis of our belief systems was a class of propositions 'which we may
reasonably accept as hypotheses because in their case either bare assertion is

32

Ibid. p. 195

58

admissible or else hypothesizing is something other than bare assertion.'33 In

this way, we were looking for something like self-evident truth. However, did
we have any self-evident truth? Could there be any self-evident proposition,
which justifiably 'stands' on its own? Must all known propositions be based on
some other known propositions? The preference for knowing many things was
the product of pragmatic reasoning. If we wanted to have knowledge, then we
had to accept some propositions that were not grounded on other propositions. If
we were given a fulcrum, we would be able to lift up the Earth. In very much the
same way, if we were to have knowledge we should be 'given', so to speak,
some propositions that can 'stand' on their own. Nevertheless, the existence of
some propositions whose justification does not depend on other propositions
but relies on their own merit is still a matter of controversy.
There are some questions about the Five Modes that are worth looking into.
For instance, Moser who was sympathetic to conceptual pluralism asked some
important questions. 'Specifically, is it epistemically mandatory that every
epistemologist be a Pyrrhonian sceptic? If so, what epistemic considerations
make Pyrrhonian scepticism universally mandatory?'34 'Why, in other words,
should non-Pyrrhonians adopt the epistemic concepts and standards that yield
Pyrrhonian worries?'35
These questions are not only about the Five Modes. Their implications go
further down to a more fundamental issue. The Five Modes could be used to
show that no beliefs were justified and epoche followed. But what was
Pyrrhonian scepticism all about? Fogelin [1983] distinguished between

33

Barnes [1990] p. 125

34

Moser [1997] p.401

35

Ibid, p.405
59

prescriptive intent and theoretical intent. In our case with the Pyrrhonian sceptic,
we needed to make a few things clear. What was the normative intent of the
sceptic when he argued against the standard of truth? Was it prescriptive or
theoretical? Similarly, what was the normative intent when the sceptic argued
that one's belief was groundless? But the questions could not be answered in
isolation. It would require us to take the nature of Pyrrhonian scepticism into
account. Consequently, we shall consider a more fundamental question: what is
the nature of the Pyrrhonian scepticism? Was it therapeutic or compelling
reasoning? I shall pursue these questions in the next chapter.

60

CHAPTER FOUR

Introduction

HIS CHAPTER looks into the different interpretations of

Pyrrhonian scepticism offered by Frede, Burnyeat and Barnes.36 At
PH I 13 Sextus explains in what sense the Pyrrhonian sceptic does

not dogmatize with reference to a contrast between two senses of dogma. The
first sense of dogma is acquiescing in something or consenting to a thing
(eudokein tint pragmati). The second sense of dogma is assent to anything
non-evident. Sextus assures us that the sceptic does not dogmatize in the second
sense of dogma. In his own words, 'the Pyrrhonist assents to nothing that is
non-evident.' The contrast between two senses of dogma, however, gives rise to
some interpretative problems. On the one hand the sceptic does not assent to
anything non-evident, yet on the other hand he does 'assent to the pathe that are
forced upon him by a phantasia'. 37 What makes the sceptic consent to his
affections but withhold assent from anything non-evident? Does the sceptic
come to hold any belief when he assents to the pathe that are forced upon him by
a phantasial Does Sextus imply in this passage that the sceptic holds beliefs in

36

The debate, featured in five separate essays, is incorporated in chronological order in

Bumyeat and Frede [1997].

37

For the meaning of these words, see the glossary.

61

the first sense of dogma, if holding this kind of belief is compatible with his
Pyrrhonian Scepticism?
The interpretations by Frede, Burnyeat and Barnes are founded on different
readings of the contrast at PH I 13. As a result, the nature and scope of
Pyrrhonian scepticism varies from one interpretation to another. Frede in 'The
Sceptic's Beliefs'38, for instance, argues that 'there can be no doubt whatsoever
that, according to Sextus, a serious Pyrrhonean sceptic can have beliefs.'
Burnyeat however, disagrees in 'Can the Sceptic Live His Scepticism?'39 and he
argues that the sceptic has no beliefs though 'the supposed life without belief is
not, after all, a possible life for man.' In 'The Beliefs of a Pyrrhonist'40 Barnes
writes, '"What is the scope of Pyrrhonian epoche?" is of the last importance for
an understanding of ancient Scepticism'. This chapter features a dialectical
structure. Frede's interpretation is a good starting point. I give an account of his
interpretation. Then I present the objections to his interpretation by Burnyeat
and Barnes. Burnyeat, for his part, offers a different interpretation of Pyrrhonian
scepticism. But we shall see how Barnes and Frede seriously undermine it.
Finally, Barnes appears to defuse the debate by arguing that the question about
the extent of Pyrrhonian scepticism may actually be ill conceived.

38

Frede's article is the earliest among thefive.It initiates the debate in the sense that Burnyeat

and Bames make several references to it in their own discussions.

39

Bumyeat and Frede [1997] pp.25-57

40

Bumyeat and Frede [1997] pp.58-91

62

Frede

REDE DISMISSES the common interpretation of Pyrrhonian

scepticism according to which 'the sceptic not only claims to have no
deeper insight into things, he also claims not to know anything at all; not

only does he maintain no specifically sceptical doctrines, he also has no views

or beliefs about anything.' 41 This interpretation appears 'fundamentally
mistaken' to Frede. For it seems impossible to live without having any beliefs,
holding any views or making any judgement. Radical scepticism appears
inconsistent.For Frede, 'if we turn from our own conception of scepticism to the
words of Sextus Empiricus, we can see clearly that the sceptic, in many
instances, does think of himself as knowing something...it is perfectly
compatible with his scepticism.' (p.2) He leaves the issues with scepticism and
knowledge aside and restricts his concern to scepticism and belief. In his own
words, 'whatever the case may be with regard to knowledge, it seems clear to
me both that there are many things the sceptic thinks or believes are the case and
that it is perfectly compatible with his scepticism for him to have all sorts of
views and beliefs. And it is just this last point which shall be our concern
here
can the sceptic have beliefs?'42
At PHI 12 Sextus introduces the chief constitutive principle of scepticism,
which is 'the claim that to every account an equal account is opposed'. Sextus,
in addition, tells us that it is from this principle that the sceptics 'come to hold
no beliefs'. This might be taken as if Sextus assures us that the sceptic withholds

41

Ibid. p. 1

42

Ibid, p.2

63

judgement on whatever issue is under consideration. Frede, however, argues the

sceptic can have beliefs. He maintains, 'it is only true with a restricted sense that
the sceptics suspend judgement on all matters and that everything depends on
how one construes this restriction.' (p.3) The restriction that Frede has in mind
confines epoche to some kinds of beliefs but it allows the sceptic to have some
other kinds of beliefs. PH I 13 provides textual substance for his interpretation
of the restricted epoche. He offers his own translation of PH I 13 as follows:

We say that the sceptic does not dogmatize, not in the sense of "belief
(dogma) in which some say, speaking quite generally, a belief consists
in consenting to a thing (eudokein tini pragmati); for the sceptic does
assent to such affections which necessarily result when things appear to
him in certain ways; he would not, for example, when he is hot or cold,
say, "I believe I am not hot (cold)"; We rather say, he does not
dogmatize, in the sense of "belief, in which some say a belief consists
in assenting to one of the nonevident things which the sciences have as
their objects of inquiry; for the Pyrrhonean assents to nothing
nonevident. (p. 16)

By and large, the debate on the scope of epoche hinges on conflicting

interpretations of PHI 13, the passage in which Sextus mentions two senses of
dogma and explains in what sense the sceptic does not dogmatize. For Frede,
PHI 13 defines the scope of Pyrrhonian scepticism. It introduces a distinction
between a wider and a narrower sense of belief. The distinction is coupled with
the claim that the sceptic assents to his affections but not to the non-evident.
Therefore, it appears to Frede that the sceptic can have beliefs in the sense of
eudokein tini pragmati (the wider sense of dogma) though he eschews beliefs in

64

the narrower sense of dogma. The distinction defines the sense in which the
sceptic suspends judgement on all matters but it also allows the sceptic to have
some beliefs. The sceptic suspends judgement on all non-evident matters, but he
can have beliefs in the sense of eudokein. Therefore Frede contends that, 'In PH
I 13ff., Sextus explains in what sense the sceptic is not dogmatic. What is not in
question, at least if we follow Sextus, is whether the sceptic has no dogmas, no
beliefs at all but whether he has no beliefs of a certain sort. Sextus distinguishes
between a wider {koinoteron) and a narrower sense of 'belief; and only beliefs
in the narrower sense count as dogmatic. Hence, there can be no doubt
whatsoever that, according to Sextus, a serious Pyrrhonean sceptic can have
beliefs.'43
But the passage does not explicitly say the sceptic can have beliefs. It
merely suggests that the sceptic does not dogmatize, or he does not assent to
anything non-evident. At best, it simply says the sceptic assents to his affections.
But Frede is ready to take this assent to affections (in the sense of consenting to
a thing or eudokein tini pragmati) as a belief. And he is prepared to support this
interpretation with arguments. He observes that the word eudokein is common
in legal contexts and Hellenistic literature. It is used in the sense of 'be content
with', 'assent to', 'agree', 'consent to', 'recognize', 'accept', or 'suppose'.44 In
the paragraph, eudokein is contrasted with the assent to non-evident things.
Hence, Frede argues if the sceptic is willing to assent to something, it will be
something evident or something that seems to him to be the case. The meaning
of eudokein in the paragraph would then be 'what the sceptic literally accepts,
what he is content with, what he has no objection to is whatever seems to him to

43

Ibid, pp.8-9 Koinoteron could just mean 'more common'. See glossary.

44

Ibid. p. 17

65

be the case, whatever seems evident to him.' (p. 17) This inference is supported
by the contrast between the dogmatist and the sceptic. For Frede 'the dogmatist
is so concerned that things might, in reality, be quite different, that he does not
accept the verdict of phantasia; instead, he relies on reason in order to find out
how things really are (cf. PHI 12).'45 The sceptic, however, is different from
the dogmatist in this respect. The sceptic, as Frede projects, 'has learned from
experience, that reason if he tries to follow it seriously and fully, gets him no
further and, thus that, he must rest content with how things appear (cf. PH I
12).' Though the sceptic sometimes argues against what seems evident, it is for
the sake of dialectical argumentation. In general, the sceptic accepts the verdict
ofphantasia and he does not bother to join the dogmatist to find out how things
really are.
Having argued to the effect that the sceptic is content with whatever seems
to him to be the case, Frede goes on to argue that the sceptic comes to hold
beliefs when he is content with whatever seems to him to be the case. He claims
that 'to judge by the passage at hand', Sextus exhibits a conception of the origin
of beliefs. 'The dogmatists', Frede suggests, 'see assent as a voluntary act, a
judgement about the impression which presents itself to us; it is only this
judgement that leads to a belief.'(p. 18)46 Whereas for Sextus 'something which
can count as a belief, a judgement, arises in us when we do not object and
consequently consent.' In this way, the sceptic is said to have beliefs when he
consents to the affections forced upon him. The beliefs that the sceptic has
would be nothing more than consenting to a thing forced upon him. Frede
conjectures that Sextus intentionally used eudokein to denote the beliefs that the

45

Ibid. p. 18

46

On issues related to assent as voluntary, and belief as involuntary, see Cohen [1992].
66

sceptic came to hold, as opposed to other beliefs that had to do with assent to
something non-evident. He argues that eudokein is a word that is
philosophically neutral. 'It hardly appears at all in philosophical texts; as a
philosophical term, it occurs nowhere else' and 'it has no philosophical or
technical meaning, no philosophical associations and is connected with no
special philosophical claims; presumably, it is exactly this fact that leads Sextus
to choose the word', (p. 17) The notion of eudokein was deliberately adopted to
highlight the fact that the sceptic consents to his affections in an ordinary
manner or in a way that is free of philosophical colouring. The sceptic was
content with whatever seems to him to be the case and he would come to hold
beliefs in this way. But his beliefs would be nothing more than consenting to a
thing forced upon him, but on the other hand 'the sceptic may not have beliefs of
a certain kind, viz., philosophical or scientific ones which depend on reasoned
grounds', (p. 19) It is worth notice that assent to the affections is made passive
for the sceptic in Frede's account. As this passive assent matches up with the
notion of belief in the account, belief is also a passive thing.
To sum up, Frede's interpretation presumes that the sceptic's assent to
affections or the first sense of dogma in the sense of consenting to a thing
matches up with the notion of belief. It is given that the sceptic is happy to
assent to his affections; therefore, the sceptic can have beliefs. Though the
sceptic is said to hold no beliefs at PH 112. Frede thinks this is only true in a
restricted sense. He suggests construing the restriction to dogmatic beliefs, so
that the sceptic comes to hold no beliefs only in the sense that he comes to hold
no dogmatic beliefs, which are related to non-evident things. However, is it
really the case that the sceptic's assent to impression amounts to belief? Is it
appropriate to take PH I 13 as the key passage 'in which Sextus explicitly
discusses what sorts of beliefs the sceptic can have without being dogmatic'?
67

As for beliefs, Frede interprets the passage as suggesting that some beliefs
depend on reasoned grounds while some others are simply consenting to the
affections. The sceptic avoids beliefs that depend on reasoned grounds but he
holds beliefs when he consents to the affections or even when he fails to object
to the affections. It is, however, not the content of a belief that makes the belief
dogmatic or ordinary. 'Any belief, whatever its content may be, can be a
dogmatic belief; conversely, every belief can be an undogmatic one.' (p. 19)
Whether a belief is dogmatic or not depends decisively on the attitude of the
believer who holds the belief. If one puts in anything more than what would be
sufficient for consenting to a thing, one's belief would not be an ordinary belief.
However, is the notion of ordinary belief in Frede's account too general and too
broad? Obviously, Frede does not worry about this.
At the end of his essay, Frede briefly considers an objection against his
interpretation. 'It might be objected that what... Sextus is prepared to call
"dogmata" are not even beliefs. For example, we might think that the mere
feeling that something is the case is not to be regarded as a belief just because
we do not object to this feeling or impression.' (p.21) To defend his
interpretation, Frede argues that what Sextus is prepared to call dogmata are
beliefs unless we are talking about 'some specific, dogmatic definition of belief.
His argument appeals to the ordinary use of 'believe', 'think', or 'suppose' (or
the ordinary use of 'dokein'). He holds, 'it is clear that the conditions for
employing these verbs are so weak that the sceptic's beliefs will satisfy them
without any difficulty.' (p.22) Frede supposes we ask someone who has just
come in the room if it is still raining outside and the man replies that it is. The
reply, as Frede claims, expresses the man's belief that it is still raining. In
addition, to deny this would involve 'a dogmatic view about what is to count as
a belief. When we ask the sceptic the same question under the same conditions,
68

the sceptic will give the same reply. Likewise, unless we are talking with a
dogmatic definition of belief, we cannot deny that the answers offered by the
sceptic express his belief that it is now raining.
When the sceptic goes out and finds that it is raining, he has no choice but
to assent to the impression forced upon him. When we ask the sceptic about the
weather, he simply replies that it is now raining. His answer is a report of what
seems to him to be the case. When he is impressed in a certain way, he reports
accordingly. The sceptic's report is azetetos, i.e., his report is outside the scope
of enquiry (PH I 22). We are in no position to challenge the sceptic's report of
the present state of his own mind. However, is this assent to the impression
belief? Drawing on PH I 14,1 used to argue that assent to impression does not
match up with belief. Sextus reports in what sense the sceptic does not
dogmatize at PH 113. After that he further explains at PH 114 that the sceptic
does not come to hold beliefs even when the sceptic utters sceptical phases like
'I determine nothing'. Most importantly, I thought Sextus had explicitly
explained what it was to hold beliefs when he put down, 'if you hold beliefs,
then you posit as real the things you are said to hold beliefs about'.47 Surely to
posit things as real is very different from merely consenting to things.
Consenting to things is much weaker than positing things as real. To consent to

47

PHI 14 'Not even in uttering the Sceptical phrases about unclear matters for example, "In

no way more", or "I determine nothing", or one of the other phrases which we shall later
discuss do they hold beliefs. For if you hold beliefs, then you posit as real the things you are
said to hold beliefs about; but Sceptics posit these phrases not as necessarily being real. For they
suppose that, just as the phrase "Everything is false" says that it too, along with everything else,
is false (and similarly for "Nothing is true"), so also "In no way more" says that it too, along
with everything else, is no more so than not so, and hence it cancels itself along with everything
else. And we say the same of the other Sceptical phrases.'

69

things, we simply accept the verdict of phantasia and raise no objection against
it. Whereas to posit things as real, we need to, at least, give reasons to support
the things as real. If somebody challenges our belief claim, then we need to
defend it with reasons again. To posit things as real, we have to be ready to
maintain it. As a result, I was inclined to argue that the sceptic's mere assent to
impression forced upon him did not amount to belief. Assenting to a certain
impression is not belief. It requires the sceptic to go far beyond merely
consenting to things to substantially posit things as real. But that is the last thing
the sceptic would do. For he claims no insight into how things really are and he
is not prepared to posit his impressions as real. In addition, the sceptic is
notorious not for establishing any doctrine but for undermining every doctrine
that he comes across; as the chief constitutive principle of scepticism states, 'to
every account an equal account is opposed'.
Moreover, the sceptic's report of what seems the case is beyond real or
unreal. So when it is raining and the sceptic is impressed that it is so, he consents
to the impression that it is raining and he raises no objection against it. He has
no need to give any reason to support his impression that it is now raining.
When he is impressed in a certain way, he reports his impression without
making any bold judgement on how things really are. My conclusion was, while
Sextus had explicitly suggested that to have beliefs was to posit as real the
things believed, and the sceptic's assent to impressions was neither real nor
unreal, it was not sensible to maintain that the sceptic's assent amounted to
beliefs. To identify the assent to affections as beliefs would be a false departure
from what Sextus said on what it is to believe in something. Dogma and belief
were two different things for the sceptic. They should not be confused in any
interpretation of Pyrrhonian scepticism. Dogma has to do with eudokein tini
pragmati, viz., consenting to things. Whereas to hold a belief, for the sceptic,
70

was to posit the thing that you believe as real. The sceptic never sets off to posit
any thing as real. Therefore, he comes to hold no beliefs, he acts in a
non-belief-related way (adoxastos).
My arguments, as I used to believe, did not rely on any dogmatic definition
of belief. It simply considered what it was for the sceptic to hold beliefs. If it is
true that the sceptic comes to hold no beliefs, he puts in nothing more than what
would suffice for consenting to a thing when he assents to his affections; as for
anything non-evident, he withholds his assent.
What is in question with PH I 13 is whether the sceptic's assent to
impressions or the first sense of dogma matches up with belief. Strictly speaking,
the passage does not explicitly say anything on the question. However, Frede
claims 'to judge by the passage at hand', for Sextus 'something which can count
as a belief, a judgement, arises in us when we do not object and consequently
consent.' Dogma is made equivalent to belief and belief has two senses; the
sceptic is said to have beliefs in the first sense but avoid beliefs in the second
sense. However, is it really the case that what Sextus refers to as dogma is belief?
Burnyeat has a different story to tell.

71

Objections to Frede's interpretation

URNYEAT CONSIDERS Frede's interpretation as a challenge to

his own interpretation. For Bumyeat, the notion of 'belief is tied to

truth; it is a sub-category of assent that necessarily has to do with

non-evident things. Since the sceptic does not assent to non-evident things, he
eschews all beliefs. Burnyeat notes that the sceptic's assent is expressed in
prepositional form like 'the honey appears to us to be sweet', an example from
PH I 20. If some of the prepositional expressions of his pathe can be properly
interpreted as making epistemic assertions then it can be argued with effect that
the sceptic is committed to some beliefs. The sceptic, by this interpretation,
would be making non-dogmatic assertions about how things are in the world
with sentences like 'the honey appears to us to be sweet'. Frede's interpretation,
as we have seen, gives the same picture of the sceptic. Therefore, when
Burnyeat attempts to rebut this line of interpretation, he is also taking issue with
Frede's interpretation.
By the epistemic reading of the sceptic's talk of appearances, the sceptic's
assent to appearance expresses his ordinary belief about what is the case in the
world. The general idea of the epistemic reading is as below.

The sceptic is supposed to content himself with appearances in lieu of

beliefs, but it maybe objected that, whatever Sextus may say, at least some
of these appearances are beliefs in disguise. "Honey tastes sweet" may
pass muster as the record of a perceptual or bodily experience, but when it
comes to "All things appear relative" (PH 1135) or "Let it be granted that
the premisses of the proof appear" (M VIII 368) or "Some things appear
72

good, others evil" (MXI 19), we can hardly take "appear" (phainesthai)
other than in its epistemic sense. That is, when the sceptic offers a report
of the form "It appears to me now that p", at least sometimes he is
chronicling the fact that he believes or finds himself inclined to believe
that something is the case, (p.47)

Though some of the sceptic's appearance-statements may be properly

accepted as records of perceptual or bodily experiences, some others should be
taken as registering what the sceptic believes. The epistemic reading seems to
rest on the assumption that it is the content of the appearance-statement that
decides whether the statement should be taken as bearing an epistemic sense or
not. If that is the case then what should be done is to decide if the contents of the
sceptic's appearance-statements are records of his perceptual or bodily
experiences, or endorsements of what the sceptic believes. Bumyeat does not
take interest in this approach to deal with the epistemic reading of the sceptic's
appearance-statement. To refute the epistemic reading, he first lays down an
argument which looks like Frede's interpretation of the sceptic's assent to
appearance, traces its origin to PH I 13 and then re-interprets the paragraph. The
argument is as below:

the sceptic's assent to appearance, as Sextus describes it, is not the

assertion of the existence of a certain impression or experience but the
experience of a non-dogmatic belief about what is the case in the world. It
will then follow that what the sceptic eschews, when he suspends
judgement about everything, is not any and every kind of belief about
things, but belief of a more ambitious type, which we may call (pending
further elucidation) dogmatic belief, (p.48)

73

This argument resembles Frede's interpretation of Pyrrhonian scepticism.

Unlike the epistemic reading, Frede's interpretation is concerned with the
attitude of the believer who holds the belief, whether a belief is dogmatic or not
depends decisively on the attitude of the believer. Drawing on PH I 13 Frede
contends that the sceptic comes to hold ordinary beliefs about what is the case
when he consents to the phantasia forced upon him. On the other hand, the
sceptic holds no dogmatic beliefs about how things are in reality, which has to
do with assenting to non-evident things. This interpretation of Sextus is very
different from Bumyeat's interpretation. It poses a challenge to his
interpretation. It is necessary for Bumyeat to turn to PH 113 in order to dismiss
the epistemic reading or Frede's interpretation. First of all, Burnyeat admits, 'a
good number of the appearance-statements in Sextus Empiricus can be read
epistemically.'48 However, he thinks the thrust is whether it can be properly
shown that the appearance-statements should be read epistemically or that
Sextus himself approves an epistemic reading. For Burnyeat, an epistemic
reading is most likely to find its justification in PH 113. Instead of giving his
own translation of the paragraph he offers his impression of PH I 13 as below.

There Sextus says that some people define a broad sense of "dogma"
meaning to accept something or not contradict it, and with this he
contrasts a narrower sense explained by some (? the same) people as
assent to one of the non-evident things investigated by the sciences. The
point of this distinction is to clarify the sense in which the sceptic does not
dogmatize: he will have nothing to do with dogma in the second and
48

The fact that many appearance-statements in PH and M seem to demand epistemic reading

may pose an objection to Sextus as well. Bumyeat dwells on this too, see pp.53-7.
74

narrower sense, "for the Pyrrhonist does not assent to anything that is
non-evident". But he does assent to states with which he is forcibly
affected in accordance with an impression, and such assent (we are given
to understand) is or involves dogma in the broader sense to which the
Pyrrhonist has no objection. For example (an example we have met
before), He would not say, when he is warmed or chilled, "I think I am not
warmed or chilled." (p.48)

With regard to the passage, Bumyeat considers two interpretative

questions. 'First, does Sextus' tolerance of the broad sense signify approval of
an epistemic reading for appearance-statements generally?' If it does not then
the sceptic's assent to affections would not be an expression of a non-dogmatic
belief about what is the case. It would simply be assent to affections only.
'Second, does his account of the narrower sense restrict his disapproval to what
we have provisionally called dogmatic belief?' 49 If it does not then the epoche is
not restricted to dogmatic belief; it would cover non-dogmatic belief as well.
The sceptic would eschew all beliefs. If this is the case, there will be no room for
any epistemic reading or Frede's interpretation according to which the sceptic is
said to hold ordinary belief but eschew dogmatic beliefs.
Bumyeat takes the questions in order. From the example of 'be warmed' or
'be chilled', he infers that 'What the sceptic accepts or does not contradict is "I
am warmed/chilled".'(p.48) Then he argues, 'unless its content "I am
warmed/chilled" is a proposition about what is the case in the world rather than
a proposition about the sceptic's experience', the statement is not 'an epistemic

49

Bumyeat does not favour the distinction of dogmatic and non-dogmatic belief. According to

him, to believe something is to believe it as true and 'Sextus has no other notion of belief than
the accepting of something as true.' (p.53)
75

seeming'. It is not an epistemic seeming since it says nothing about what is the
case in the world. The statement is not an expression of a non-dogmatic belief
about what is the case. It is, at best, a report of the pathos of the sceptic, which
makes no truth-claim about things in the world. According to Sextus, the
sceptic's report of his impression is not subject to enquiry, azetetos (PH 122).51
We cannot properly challenge the sceptic's report of the present state of his own
mind. Therefore, if the sceptic thinks that he is warmed and he says that he is
warmed, his statement should not be taken in any epistemic sense. It is just an
expression of his pathos or a record of his bodily experience. My earlier
objection to Frede's interpretation relies on the same characterization of the
appearance-statement made by Sextus. Burnyeat pursues a different line of
argument. Now since it is most likely for epistemic reading to get approval from
Sextus at PH I 13, one could bring the paragraph under scrutiny to see if it really
gives the approval for an epistemic reading. Burnyeat adopts this approach. His
contention is that the broad sense of dogma would not support an epistemic
reading of appearance-statements if the example of being warmed/chilled at PH
113 is not a proposition about what is the case in the world. For if the example is
a proposition about what is the case in the world, then it would be an epistemic
seeming; it would be an example by which the sceptic expresses his belief about
what is the case in the world. In addition, since Sextus intends to explain in what
sense the sceptic does not dogmatize at PH I 13, it is natural to take the

50

Ibid, p.49

51

PH I 22 'Accordingly, we say that the criterion of the Sceptic Way is the appearance in

effect using that teim here for the phantasia for since this appearance lies in feeling and
involuntary pathos it is not open to question. Thus nobody, I think, disputes about whether the
external objects appears this way or that, but rather about whether it is such as it appears to be. '

76

paragraph as an overall conclusive statement on the issue. If it is found that

Sextus makes a belief claim in this particular paragraph, that could be made a
forceful if not a decisive case for the epistemic reading.
Burnyeat's discussion opens with an analysis of two verbs from PH I 13,
i.e., thermainesthai and psuchesthai. He argues that these two verbs 'do not
normally mean "I feel hot/cold", although translators (Bury, Hossenfelder) have
a tendency to render them in such terms here, just because Sextus is illustrating
an affection {pathos). They normally mean "be warmed/chilled". On the other
hand, neither does "I am warmed/chilled" necessarily refer to an objective
process of acquiring or losing heat. And my own view is that to insist that
Sextus' illustrative pathos must be either a subjective feeling or an objective
happening is to impose a Cartesian choice which is foreign to his way of
thinking.' (p.49)
Bumyeat makes three points in the paragraph. First, he takes
thermainesthai and psuchesthai as meaning be warmed or be chilled
respectively. Secondly, 'I am warmed/chilled' does not necessarily imply that
the sceptic endorses the existence of any objective happening of heat transfer.
Hence, the sentence does not necessarily commit the sceptic to any
non-dogmatic belief about what is the case in heat transfer. This argument is
barely supported or elaborated. It is a weak argument. It is on a par with the
contrary argument, namely that the sentence necessarily implies that the sceptic
endorses what is the case in heat transfer. This weak argument, however, is not a
mistake or a careless fault. For Burnyeat, in fact, believes that the
terminological ambiguity involved makes it impossible to decide conclusively
if thermainesthai and psuchesthai together should be identified with the
physical event of heat transfer, or with the way it feels about gaining or losing

77

heat.

The best he can argue is that "The reference of these funny verbs

[thermainesthai and psuchesthai] is plainly to a perceptual process rather than

to the transmission of heat, ... but we should keep the translation "be
warmed/chilled". The man is being affected perceptually (cf. "We are
sweetened perceptually", glukazometha aisthetikos, at PH I 20 and the use of
thermainein at PH 1110, II 56, MI 147, VII 368, IX69)'.53 In a word, Burnyeat
argues that thermainesthai and psuchesthai do not necessarily refer to what is
the case in the process of heat transfer but rather refer to a perceptual process.
The third point states that to insist that the sceptic's pathos must fall under
the category of either subjective feeling or objective happening is to impose an
anachronistic

dichotomy between

'subjective

feeling'

and

'objective

happening' on the sceptic's pathos. Burnyeat suggests that the dichotomy is 'a
Cartesian choice' which is alien to the sceptic's way of thinking.54 In other
words, 'we cannot "split" the affection (pathos) into separate mental (subjective)
and physical (objective) components.' (p.50) So if this is right, it would be
possible for the sceptic's pathos to be free of any epistemic association with the
process of heat transfer when he says something like, 'I am warmed/chilled'.

52

Burnyeat reports that the ambiguity encountered existed in Cyrenaic vocabulary as well, (p.49)

Frede also discusses thermainesthai. Like Bumyeat, he does not approve the translation by Bury
and Hossenfelder on thermainesthai. And he admits the same difficulty; ' A precise analysis of
the example is difficult for both linguistic and intrinsic reasons.' (p.20) But his concern is to
show that the sceptic can have beliefs that are more than his own impressions. See pp.20-1.
53

Ibid, p.49

54

Bumyeat [1982] argues that 'it was Descartes who put subjective knowledge at the centre of

epistemology and thereby made idealism a possible position for a modem philosopher to
take.' Before that the sceptic and his fellow Greeks had not disintegrated the pathos into
separate subjective and objective components.

78

Eventually the pathos, as Burnyeat has tentatively argued earlier, refers to a

perceptual process of being warmed/chilled. All possibilities of intrusion by
epistemic reading are exhausted and ruled out. For Burnyeat 'The moral to draw
is not that the Pyrrhonist allows himself some beliefs about what is the case, but
that scepticism is not yet associated with a Cartesian conception of the self.'
(p.50) And the conclusion, in his own words, is that 'PH I 13 offers no
justification for an epistemic reading of the sceptic's appearance-statements.
The broader sense of "dogma" is simply the accepting of a perceptual
experience as the experience it is... Sextus is not going out of his way to leave
room for a non-dogmatic type of belief about matters of real existence.' 55
For the second interpretative question, Burnyeat first notes, 'any thing
which is non-evident is something for the sciences to investigate, the
non-evident being by definition that which can only be known by the mediation
of inference.'(p.50) Then he reminds us 'the Pyrrhonist attack on the criterion of
truth abolishes the evidence of everything that the dogmatists consider evident
(PH II 95, M VIII 141-2).' The sceptic's attack on the criterion of truth makes
everything non-evident to him, even the seemingly evident as 'It is day'. The
sceptic, as Burnyeat thinks, 'accepts it is day only as a non-epistemic statement
of appearance, "It appears to be day [sc. but may not actually be so]". Anything

55

Before dealing with the second interpretative question, Burnyeat adds a supplementary

argument to his discussion of the first question. The sceptic uses the verb 'to be ' in the sense of
'to appear' (PH 1135,198,200) and at MXI 18 Sextus explains it in non-epistemic terms. So, he
argues, we should take 'honey is sweet' meaning 'honey appears sweet'. The fact that Sextus is
cautious about 'to be ' and his conscious avoidance of giving the false impression that he
endorses any epistemic seeming seems to support the idea that Sextus does not approve any
epistemic reading of the sceptic's appearance-statements.

79

which goes beyond (non-epistemic) appearances is subject to enquiry'. The

distinction between the evident and the non-evident is blurred. Anything
seemingly evident can rum non-evident. Anything non-evident is subject to
enquiry. It follows that anything can be the subject of investigation, and thus
that withholding assent can extend to anything. The narrower sense of dogma,
hence, does not restrict the epochs to so-called dogmatic beliefs and leave other
beliefs intact. When the sceptic suspends judgement, he eschews all beliefs with
no restriction.
Frede's interpretation of the epistemic reading appears to be greatly
undermined by Burnyeat's arguments. Contrary to Frede's interpretation,
Burnyeat argues that the sceptic's assent to affections is not an expression of
any non-dogmatic belief about what is the case. 'The sceptic's assent is simply
the acknowledging of what is happening to him". 57 In addition, when the
sceptic suspends judgement he eschews all beliefs with no reservation. As I
have noted before, Bumyeat operates with a confined notion of 'belief and he
claims, 'Sextus has no other notion of belief than the accepting of something as
true.' 58 His discussion rests heavily on this confined notion of belief. I shall
return to this when I discuss objections to Burnyeat's interpretation.
Frede's interpretation contents itself with a distinction between dogmatic
and non-dogmatic belief. But consider the following passage:

We proceed in the same way when asked whether the sceptic has
hairesis (a system). If one defines hairesis as an attachment to a number

56

Ibid.p.51

57

Ibid, p.43

58

Ibid.p.53

80

of dogmata that agree with one another and with appearances, and
defines a dogma as an assent to something non-evident, we shall say that
the sceptic does not have hairesis. But if one says that hairesis is a way
of life that, in accordance with appearances, follows a certain rationale,
where that rationale shows how it is possible to live rightly ("rightly"
being taken, not as referring only to virtue, but in a more ordinary sense)
and tends to produce the disposition to suspend judgement, then we say
that he does have hairesis. For we do follow a certain rationale that, in
accord with appearances, points us toward a life in conformity with the
customs of our country and its laws and institutions, and with our own
particular paths. (PH I 16-7)

Burnyeat argues at the end of his discussion of the epistemic reading, 'one
solitary reference to the sciences (for it is not repeated elsewhere in Sextus) in a
definition borrowed from someone else is not sufficient to credit Sextus with a
distinction between dogmatic and non-dogmatic belief.' (p.51) In the first place,
it seems more likely to Burnyeat that the distinction is borrowed from someone
else. In the distinction, Sextus makes explicit reference to some people who
define the broad sense of dogma and some other people who define the narrower
sense of dogma. (Burnyeat wonders whether they are the same people).
Burnyeat takes it as Sextus' admittance that the distinction is 'borrowed from
some previous sceptic writer'. Moreover, he notes that the distinction is
structurally parallel to PH 116-7 in which Sextus mentions a contrasting pair of
the term hairesis. He points out that the first definition of hairesis is 'couched (it
would appear) in terms of the narrower sense of "dogma"' and it 'can be found
almost verbatim in an unfortunately truncated passage of Clement (SVF II, p.37,
8-10), where it is again attributed to "some people".' (p.51) In the second place,
81

Burnyeat notes that Sextus repeatedly restricts his epochs with various forms of
qualification.

According to him, 'Just how restrictive these qualifications are

depends on what they are contrasted with, and in every case the contrast is with
how things appear, where this, as we have seen, is to be taken non-epistemically.
All we are left with, then, is a passive impression (phantasia) or experience
(pathos), expressed in a statement which makes no truth-claim about what is the
case.' Drawing on Sextus' characterization of the sceptic at PH I 15, he
maintains, the sceptic 'states what appears to himself and announces his own
experience without belief, making no assertion about external things'.
Actually, Burnyeat thinks that the distinction between dogmatic and
non-dogmatic belief cannot really be made out. In the first place, the distinction
is inconsistent with the internal logic of Pyrrhonian scepticism. To induce
epochs, the sceptic may need to undermine our hopes and fears. However, hope
and fear can be founded on any type of belief, let it be dogmatic or
non-dogmatic. So to undermine our hopes and fears, the sceptic has to deal with
any beliefs about what is good and what is bad for us. 'Belief, in the sense
Sextus is attacking, is responsible for all the things men pursue and avoid by
their own judgement (M XI 142, using doxa). The internal logic of Pyrrhonism
requires that dogma and doxa Sextus does not differentiate between these
two terms really do mean: belief.' (p.52)
Secondly, for Burnyeat it is rather uncertain how a distinction between
dogmatic and non-dogmatic beliefs can be made out. Sensible qualities do not
59

Burnyeat gives the following examples of restriction. The Pyrrhonist's epochsis restricted 'to

the question how things are "in nature" (pros ten phusin etc., PH I 59, 78, 87, et al.)\ 'how
things are "so far as concerns what the dogmatists say about them" (PH II 26, 104, III 13, 29,
135, M Vin 3)', or 'how things are "so far as this is a mater for logos (statement, definition,
reason)" (PHI 20,215)'.
82

pose any great problem to the distinction; 'but it would need to be explained
what it amounted to when applied to such examples as "It is day", "I am
conversing" (M VIII144), or "This is a man" (M VIII 316).' (p.53) According to
Burnyeat, it is not practical to define non-dogmatic belief as belief not grounded
in reasoning. For this attempt will disconnect belief with truth and reason. He
insists, 'Sextus has no other notion of belief than the accepting of something as
true.'
An epistemic reading seems most likely to find its justification at PH I 13
because the distinction involved a reference to the sciences. The narrower sense
of dogma is 'assent to one of the non-evident matters investigated by the
sciences'. The reference to sciences is not repeated elsewhere in Sextus.
Therefore, Burnyeat argues that the distinction is borrowed from someone else
and Sextus is not committed to the distinction. However, Barnes also notices
that the reference is mentioned once, but he argues that Sextus' normal way of
identifying dogmata is simply 'assent to something unclear' (PH I 97). 60 A
distinction between dogmatic and non-dogmatic belief could be preserved
without the reference to the sciences. This distinction still allows and supports
Frede's interpretation even if we grant that Sextus borrows the distinction from
someone else, as Burnyeat suggests. By this distinction, belief loses connection
with truth and reason. As we shall see in Frede's objection to Burnyeat, it could
be argued that belief is not necessarily related to truth and reason.
Burnyeat's objections to Frede are not conclusive so far. Now I move on to
see how Barnes adds force to the challenge to Frede and then I will get back to
Burnyeat's other powerful arguments.

60

Ibid, p.76
83

There are numerous utterances of

sentences in PH. For instance,

'the honey appears to us to be sweet' (PH120). Barnes contends that

in Greek may carry epistemic sense, like 'appears' or 'seems' in English. But
. in PH I does not carry this epistemic

then he agrees with Bumyeat that

sense and 'the appearing is "phenomenological"

reports the way

things look.' (p.63) It can be argued that the Pyrrhonist of PH is committed to

his utterances of phenomenological

sentences and thereby comes to

hold some beliefs, like beliefs about his own existence, space, time and external
objects. 'For the utterance of an indicative sentence functions characteristically
as a manifestation of belief in the proposition expressed by the sentence.' (p.64)
Barnes, however, suggests that not every utterance of an indicative sentence is a
statement or an affirmation. In addition, the Greek philosophers know that there
are many speech acts while stating is only one of them. Therefore, he argues that
we have to find out 'what speech act the Pyrrhonist is performing when he utters
sentences.'(p.65) First, he reports that Sextus use the term 'avowals'
(apangeliai) to refer to Pyrrhonist utterances. According to Wittgenstein,
avowals (or Ausserungen) are expressions of feeling and 'to call the avowals of
a feeling a statement is misleading' (Zettel, 549). Drawing on Wittgenstein on
avowal, Barnes argues that when the Pyrrhonist of PH is mentally affected and
utters a phenomenological

sentence, he is not stating anything nor

evincing any belief; he is rather expressing his mental state (pathos). 'Avowals
are not statements; and they bypass belief. The avowals of a Pyrrhonist may
similarly bypass belief.' Hence, 'if we are prepared to take seriously Sextus'
talk of avowals, the Pyrrhonist may support

84

volubly while

remaining an exemplary rustic.'61 In this way, the PH Pyrrhonist's epistemic

commitment to

is minimal.

If Barnes's analysis of the nature of the utterances of

sentences

in PH is right, there would be not much room for any epistemic reading or
Frede's interpretation. But his reliance on Wittgenstein could be anachronistic.
Groarke [1990] argued that the Pyrrhonist and his contemporaries understand
belief in quite a different way than we do. Epoche leaves room for belief in
contemporary sense but not belief in ancient sense. And so it is anachronistic to
conflate the two senses of belief in interpreting Pyrrhonian scepticism and the
epochs. Burnyeat, as we now turn to see, also detects the shadow of
anachronism in Frede's interpretation.
In 'The Sceptic in His Place and Time' Bumyeat advances a round-about
argument against Frede's interpretation. He presents a historical perspective
according to which for the Greeks, philosophizing would have direct bearing on
first order judgement so that philosophical puzzlement would impede the
ordinary business of life. Greek philosophy is under 'the atmosphere of
"belatedness"'. Another way to describe it is the lack of 'insulation' between
philosophizing and living. By contrast, modern philosophy is associated with
insulation, so that philosophizing would be separated from the ordinary
business of life. His main idea is that we need to put the sceptic in his place and
time, and be aware of the lack of insulation if we are to have a sound
understanding of the sceptic. Based on this historical perspective, Frede's
interpretation is rendered anachronistic.
According to Bumyeat's thesis of insulation, a modern philosopher
'insulates his ordinary first order judgements from the effects of his

61

Ibid. p.67. A Pyrrhonist is rustic in Barnes's sense if he holds no beliefs whatsoever.

85

philosophizing'. 'Insulation... is a two way business.' (p.93) On the one hand,

insulation safeguards ordinary life from philosophical scepticism so that
philosophical scepticism does not alter judgements and knowledge-claims made
in ordinary life. On the other hand, it protects philosophical scepticism from
common sense. Insulation, moreover, is the result of centuries-long
development in philosophy. 'Ancient philosophers would find it puzzling; so
would the philosophers of the Renaissance. This sense of the separateness,
sometimes even the strangeness, of philosophical issues is not a timeless thing,
intrinsic to the very nature of philosophy. It is a product of the history of
philosophy', (p.94)
Gassendi and Montaigne were actively involved in the modern revival of
Pyrrhonian scepticism. According to Bumyeat their interpretation of
Pyrrhonian scepticism already featured the sceptic practising insulation. And
'their brand of insulation is still to be met with in modern accounts of ancient
scepticism', (p.96) He regards PH I 13 as the key source for all insulating
interpretations that render Pyrrhonian scepticism transcendental and Frede's
interpretation as exemplary. The following is his translation of the paragraph.

When we say that the sceptic does not dogmatize, we are not using
"dogma" in the more general sense in which some say it is dogma
to accept anything (for the sceptic does assent to the experiences he
cannot help having in virtue of this impression or that: for example,
he would not say, when warmed or cooled, "I seem not to be
warmed or cooled"). Rather, when we say he does not dogmatize,
we mean "dogma" in the sense in which some say that dogma is
assent to any of the non-evident matters investigated by the

86

sciences. For the Pyrrhonist assents to nothing that is non-evident.

(PH113)62

The sceptic's assent to his affections is important to how different

interpretations decide on the scope of Pyrrhonian scepticism as depicted by
Sextus in the PH. This assent is also called assent to appearances. Burnyeat
sums up the debate on the scope of Pyrrhonian scepticism as a matter of how the
notion of appearance is to be defined. So 'if one gives the sceptic a generous
notion of appearance, the area of his assent expands and the scepticism contracts,
while conversely scepticism spreads and assent draws back if... one takes a
more restricted view of appearance.' (p.97) Burnyeat admits that he takes a
restricted view of appearance.
Instead of giving an analysis of the assent to appearance, Burnyeat focuses
on the other half of the contrast in PH I 13. To start with, he defines the notion of
the non-evident as 'the notion of that which we can only know about, if we can
know it at all, by reference from what is evident. If knowledge of the
non-evident is possible, as Sextus' dogmatist opponent believes, it is mediate
knowledge as contrasted with the immediate, non-inferential knowledge of what
is evident (PH II 97-9).'63 This notion of the non-evident together with the
reference to the sciences in the contrast gives rise to what Burnyeat calls 'the
country gentleman's interpretation', an insulating interpretation.

Now, if one takes "The stove is warm" as an example of something

evident, and couples it with reference to the sciences, it becomes rather

62

Ibid, p.96

63

Ibid, p.98

87

natural to suppose and this is what Montaigne and Gassendi did

suppose that dogma in the sense Sextus wishes to eschew is any
scientific pronouncement about, for example, the underlying physical
structure which makes warm things warm, any theory about the real
nature of heat, perhaps even the assertion or the belief that there is such a
thing as the real nature of heat about which a theory could in principle be
given. On this type of interpretation in honour of Montaigne I should
like to call it the country gentleman's interpretation Pyrrhonian
scepticism is scepticism about the realm of theory, which at this period
will include both what we would consider philosophical or metaphysical
theory and much that we can recognize as science, (pp.98-9)

According to the country gentleman's interpretation,

Pyrrhonian scepticism is scepticism about the realm of theory, which at

this period will include both what we would consider philosophical or
metaphysical theory and much that we can recognize as science. The
non-theoretical judgements of ordinary life are insulated from the
scepticism and the scepticism is insulated from them, not because
Sextus, like Thompson Clarke, assigns a special status to philosophical
doubt, but because he assigns it a special subject matter, different from
the subject matters with which the ordinary man is concerned in the
ordinary business of life. This is insulation by subject matter or content,
a disengagement of life from theory, (p.99)

The sceptic practises insulation by subject matter. Therefore, he can hold

beliefs about the ordinary business of life but suspend judgement over a number
of theoretical things. This interpretation appears similar to the interpretation
given by Frede. However there is a subtle difference. In Frede's account, the

88

sceptic does not practise insulation by subject matter. It is not the content that
makes the belief dogmatic. 'Any belief, whatever its content may be, can be a
dogmatic belief; conversely, every belief can be an undogmatic one.' (p. 19)
Whether a belief is dogmatic or not depends decisively on the attitude of the
person who holds the belief. Frede's interpretation differs from the country
gentleman's interpretation in this respect. As we shall see, for Burnyeat, Frede's
interpretation features insulation between two ways of understanding
statements such as 'The soup is hot'. Therefore, his objections to the country
gentleman's interpretation could be directed to Frede's interpretation.
For Bumyeat, the manner in which 'Gassendi aligns Sextus' contrast [at
PH I 13] with a contrast between the outside of things (what is accessible to
everyday observation through the senses) and their inner nature' in his
exposition of Sextus is an example of anachronism in the way Pyrrhonian
scepticism has been interpreted since the seventeenth century. He claims that
one can never 'find Sextus adding the epithet "inner" on the innumerable
occasions when he concludes, "We can say how things appear, but not what
their nature is". The inner/outer contrast bespeaks a new world, in which the
interpretation of ancient Pyrrhonism has been overlaid with the preoccupations
of seventeenth-century science.' (p. 100)
Referring to a passage by J. L. Mackie, Burnyeat spots a contemporary
example of insulation in ethics.64 According to Burnyeat, Mackie 'insulates first
order moral judgements so securely that he thinks they can survive the second

64

Mackie [1977] Ethics: Inventing Right and Wrong p.34. 'The denial of objective values can

carry with it an extreme emotional reaction, a feeling that nothing matters at all, that life has lost
its purpose. Of course this does not follow; the lack of objective values is not a good reason for
abandoning subjective concern or for ceasing to want anything.'

89

order discovery that all first order value judgements involve error, viz. an
erroneous (false) claim to objective truth.' (p.112) He then compares Mackie
with the sceptic. "The original Pyrrhonists, by contrast, thought that if
philosophical argument could cast doubt on the objectivity of value in their
terms, if it could be shown that nothing is good or bad by nature that would
precisely have the effect of making you cease to want anything, or to hope for
anything, or to fear anything. Their name for this detached view of one's own
life was ataraxia.' For the sceptic, epoche is the gateway to ataraxia. Burnyeat
argues that insulation would make it difficult for the sceptic to induce ataraxia.
His point is that it would be necessary for the sceptic not to leave ordinary belief
intact if ataraxia were to be achieved. Otherwise, epoche on second order
judgement alone would not have much bearing on our desires and orientation.

It is above all the judgments which underlie the ordinary man's hopes
and fears which must be put in doubt and withdrawn if ataraxia is to be
achieved. The target of the sceptical arguments is, first, the ordinary
man's ordinary belief that it is good and desirable to have money, say, or
fame or pleasure (M XI 120-4, 144-6; cf. PH I 27-8); and second, the
first order judgements of ordinary life about what is happening in the
world around, which bear upon our achievement of these goals (if it is
good and desirable to have money, it is important to know where the
money is). The method of attack is philosophical argument, but the
target is our innermost selves and our whole approach to life. Any
attempt to insulate our first order judgements would frustrate the
sceptic's philanthropic enterprise of bringing us by argument to ataraxia
ofs oul(cf.PH III 280).65

65

Bumyeat and Frede [1997] p.l 12-3

90

Burnyeat claims this perspective should prevail in Sextus' discussion of

space and time as well. For instance, we see a difference between doubting the
claim that Pyrrho lived in the third century B.C. on empirical grounds and
doubting it on the grounds of scepticism about time. If it is on empirical grounds,
the doubt calls for historical evidence to show it is true that Pyrrho lived in the
third century B.C. If it is on the grounds of scepticism about time, the whole past
is rendered dubious. Burnyeat thinks Sextus would not have this sense of
difference. He contends that for Sextus

anyone who says that Plato now is in the place where Socrates was when
he was alive, and intends thereby to make a truth-claim, says something
which is open to inquiry in that he can be challenged to give reasons or
evidence for his claim and to defend its legitimacy, where this may
include... defending a conception of place or the reality of time. If the
defence fails, that has much the same effect as failure to produce decent
historical evidence. It begins to look as if there is no good reason to
believe the statement. And if you can find no good reason to believe a
statement, what can you do but suspend judgment about it? All that
remains for you is the standard sceptic retreat to a statement which makes
no truth-claim, for which, consequently, reasons and legitimacy cannot be
demanded, namely, "It appears to me that Plato now is in the place where
Socrates was when he alive." That you can say without opening yourself
to the sceptical arguments, (p. 113)

This argument, as it appears to me, appeals to PH I 14 where Sextus

mentions 'if you hold beliefs, then you posit as real the things you are said to
hold beliefs about'.

91

Having elaborated the thesis of insulation, Burnyeat returns to the second

half of the contrast in PH I 13 to discuss a major problem for the country
gentleman's interpretation. At first sight, the second half of the contrast appears
to confine the scope of epoche to theoretical statements. The country gentleman
would be happy to take this part of the contrast at its face value to support his
interpretation of Pyrrhonian scepticism. However 'Sextus plainly states that the
outcome of his critique of the criterion and of truth is that one is forced to
suspend judgement about the things which the dogmatists take to be evident as
well as about the abstruse matters they describe as non-evident. (PH II 95, M
VIII 1411-2)' (p. 115) Burnyeat used to think the critique blurred the distinction
between the evident and the non-evident and so made all simple statements
about external objects non-evident and thus dogmatic. But Burnyeat changes his
mind in 'The Sceptic in His Place and Time'. Now it appears to him that,

the distinction between the evident and the non-evident is itself one of
those dogmatists' distinctions which the sceptic makes light of (cf. PH II
97). The definition of dogma as assent to any of the non-evident matters
investigated by the sciences is explicitly taken from someone else (PH I
13). Sextus will use it, but not for the purpose of insulating the ordinary
from the theoretical. About both sides of the dogmatists' distinction he
speaks with a clear voice: it is impossible not to suspend judgment,
(p. 115)

The distinction makes no difference to the scope of Pyrrhonian scepticism

because of the lack of insulation. Every statement making a truth-claim uses
concepts like motion, time, place, body, which are the subject of theoretical
speculation. Therefore, all statements fall under the scope of scientific

92

investigation. If these concepts are problematic, as Sextus argues they all are,
and philosophical doubt is not separated from empirical doubt, ordinary
statement will then be equally problematic. Therefore, if the sceptic suspends
judgement on theoretical concepts, ordinary life statements that presuppose the
concepts would be subject to epoche as well. The sceptic may assent to his
affections and dwell on ordinary life statements. But he would suspend
judgement on them. It seems it would be difficult for the country gentleman's
interpretation to reply to this argument.
Bumyeat suggests that Kant played an important role in the invention of
insulation in philosophy. 'It was Kant who persuaded philosophy that one can
be, simultaneously and without contradiction, an empirical realist and a
transcendental idealist.' (p. 121) In addition, it was after Kant that scepticism
goes to the transcendental level. Insulation was thus invented. Since then
ordinary life statements like 'The soup is hot' can be taken empirically,
invoking no philosophical colouring at the transcendental level. Statements like
these became immune from transcendental scepticism. But things were different
in ancient Greece, if the historical perspective given by Burnyeat is correct.
Frede's interpretation of Pyrrhonian scepticism also features insulation between
two ways of understanding statements: the ordinary man's plain way of
understanding and the philosophical way of understanding. If things had not
been so in ancient Greece, this line of interpretation would be seriously wrong.
Burnyeat observes that Thomas Clarke [1972] reiterated some of what
Kant had said on scepticism, but in a quite different tone of voice.

Clarke's sceptic takes up what is called the absolute point of view and
declares that the plain man's knowledge claims are all very well in the
context of ordinary life but they do not embody an absolute knowledge
93

of things as they are in themselves; they are knowledge only in a manner

of speaking the plain man's manner of speaking. Which has no
foundation outside the practices of ordinary life. So we reach the idea
that there are two ways of understanding a statement like "The stove is
warm", the plain way and the philosophical way, and it is only the
philosophical claim to an absolute knowledge that the sceptic wants to
question. What he questions is precisely that "The stove is warm" can
embody any further or deeper kind of knowledge and truth than the plain
man puts into it. Once the Kantian insulation by levels is established,
scepticism itself goes transcendental, (p. 122)

Peter linger's hypothesis of semantic relativity is an

influential

elaboration of the phenomenon of different understanding of statements. The

concept of semantic relativity assumes that 'there is no objectively right answer
as to how a certain expression should be interpreted; no unique determinate
meaning to be assigned... One set of assumptions leads to one semantic
interpretation, another set leads to another, and there is nothing to decide
objectively in favour of either set.' 66
Semantic relativity contributes to philosophical relativity, which maintains
that some traditional problems in philosophy could not be objectively answered
because 'the answer one prefers for a certain philosophical problem will depend
upon what assumptions one has adopted in relation to that problem...A certain
set of assumptions yields one answer, another set another; whatever facts pertain
to the problem fail to decide between the one set and the other.'

64

Unger[1984]p.5
94

A crucial aspect of a philosophical problem may depend on the meaning

of, or on the semantic conditions of, certain linguistic expressions in
terms of which the problem directly and standardly formulated. For
example, the problem of knowledge might thus turn upon the meaning
of "know" as it occurs in typical sentences of the form "Someone knows
that such and such is the case". In much the same way, which answer
one gives to the problem may turn on one's specification of the
truth-conditions of such sentences. Even if there are other aspects of a
given philosophical problem that are not undecidable, the existence of
only one undecidable semantic aspect may be enough to lead to
philosophical relativity in the case of that problem, (p.5)

Semantic

relativity

goes

far

beyond

the

concept

of

pluralistic

understanding. It gives a more sophisticated picture of how people understand

statements. There would be diverse, sometimes conflicting, understandings of
key notions that are crucial to the apprehension but also to the solution of certain
philosophical problems.
Though modem philosophy is associated with insulation and/or semantic
relativity, G. E. Moore is an exception. Burnyeat points out that the manner in
which Moore attempted to refute scepticism resembles the ancient sceptic's
approach to philosophizing. Insulation is absent in both cases.

Moore is notorious for insisting that a philosophical thesis such as "Time

is unreal" be taken with a certain sort of seriousness, as entailing, for
example, that it is false that I had breakfast earlier today. And he thinks
it relevant and important to argue the contrapositive: it is true that I had
breakfast earlier today, therefore it is false that time is unreal. People
always feel that these arguments and attitudes of Moore's miss the point.

95

That is not the way philosophical questions should be treated; it is a

naive and wrong sort of seriousness. But I think that Sextus would
recognize a kindred spirit. If we look a third time at the texts before us
we can see that Sextus' dogmatist argues in a manner exactly like Moore:
One thing is to the right, another to the left, therefore there are places;
Plato is where Socrates was, so at least one place exists. Compare: here
is one hand, here is another, so at least two external things exist. Sextus
complains that this is circular; he does not complain that it is the wrong
sort of argument to establish the thesis that place exists. And he
propounds a modal version of the same inference in reverse: it is
problematic whether place exists, therefore it is problematic whether
Plato is where Socrates was or whether one thing is to the right of
another.67

Finally, Burnyeat concludes that Frede's interpretation of Pyrrhonian

scepticism finds in the PH, especially at PH 13, 'something very like a
transcendental scepticism'. A transcendental sceptic would tolerate common
sense belief and ordinary way of understanding but he targets his scepticism on
the dogmatist's excessive philosophical way of understanding. But Bumyeat
thinks this is anachronistic.

67

Burnyeat and Frede [1997] p. 115-6

96

Burnyeat

OW I MOVE ON to Burnyeat's interpretation. At the outset of his

discussion, he claims 'Pyrrhonism is the only serious attempt in
Western thought to carry scepticism to its furthest limits and to live

by the result', (p.26) To live by the result of scepticism, it follows, is to make a

way of life without belief. The possibility of adoxastos bioumen {let us live in a
way free of belief) has been challenged and defended since antiquity. Burnyeat

revisits 'those old controversies from the perspective of a modern scholarly

understanding of Sextus Empiricus.' He argues that the sceptic eschews all
beliefs and that adoxastos bioumen 'was a fundamental feature of Pyrrhonism
from Aenesidemus onwards' (p.34) His arguments are shaped by the concept of
truth according to which what is true for the Greeks, including the sceptic, is
restricted to what conforms to real existence as contrasted with appearance.

68

He discusses the issue in greater detail in Burnyeat [1982]. Actually, his whole interpretation

of Pyrrhonian scepticism is characterized by the concept of truth. At PH I 8 Sextus concisely

describes Pyrrhonian scepticism as ' a disposition to oppose phenomena and noumena to one
another in any way whatever, with the result that, owing to the equipollence among the things
and statements thus opposed, we are brought first to epoche and then to ataraxia.'' Bumyeat ties
all key terms in the description to the concept of truth. So that 'the conflict of opinions is
inconsistency, the impossibility of being true together (cf. M VII392); the undecidability of the
conflict is the impossibility of deciding which opinion is true; the equal strength of conflicting
opinions means they are all equally worthy (or unworthy) of acceptance as true; epoche is a state
in which one refrains from affirming or denying that any one of them is true; even ataraxia is
among other things a matter of not worrying about truth and falsity any more.' Bumyeat and
Frede[1997]p.30.

97

According to Burnyeat,

When the sceptic doubts that anything is true (PH II 88ff., M VIE 17 ff.),
he has exclusively in view claims as to real existence. Statements which
merely record how things appear are not in question they are not called
true or false only statements which say that things are thus and so in
reality. In the controversy between the sceptic and the dogmatists over
whether any truth exists at all, the issue is whether any proposition or class
of propositions can be accepted as true of a real objective world as distinct
from mere appearance. For "true" in these discussions means "true of a
real objective world"; the true, if there is such a thing, is what conforms
with the real, an association traditional to the word alethes since the
earliest period of Greek philosophy (cf. M XI 221). (p.30)

Based on this contextual framework of Greek philosophy, Burnyeat argues

that the sceptic eschews all beliefs. An appearance-statement, which records
only how things appear, is neither true nor false. Only statements purporting to
describe real existence can be properly regarded as true or false. Truth is
essentially

matter

of

conforming

to

real

existence

and

an

appearance-statement says nothing about real existence at all. Belief is tied to

truth; it is related to real existence as contrasted with appearance. To believe
something is to accept it as true or conforming to real existence. An appearance
statement, which is neither true nor false, does not express any belief. In
contrast, a statement about real existence expresses a belief. For it can be
properly described as true or false. There is, it follows, only belief about real
existence as contrasted with appearance. There is no such thing as belief about
appearance. Now 'if epoche is suspending belief about real existence as

98

contrasted with appearance, that will amount to suspending all belief, (p.31)
Burnyeat further argues that 'the Greek word dogma originally means
simply "belief (cf. PI. Rep. 538c, Tht. 158d) [and] Sextus defines

"dogma"...

as assent to something non-evident (PH I 16).' As a result, 'Assent is the genus;

opinion, or belief, is that species of it which concerns matters of real existence
as contrasted with appearance.' Contrary to Frede on dogma and belief,
Burnyeat argues belief via dogma is a sub-category of assent that has to do with
non-evident things. They both agree that the sceptic does not have beliefs that
have to do with non-evident things. But for Frede, the sceptic can have beliefs in
the sense of eudokein tini pragmati, Burnyeat, however, denies beliefs in this
sense.Given his views on the true and dogma, Burnyeat seems to ground his
interpretation on the assumption that Sextus is committed to two doctrines: (1)
the true is what conforms to reality; and (2) belief is tied to truth; it is assent to
something not evident.

Nevertheless is it really the case that Sextus endorses

the doctrines (1) and (2)? I am not going to argue against the doctrines by
themselves. However, I shall argue in the following that Sextus may appear to
endorse truth and belief as articulated by Burnyeat. Yet the fact is that he is not
under any yoke to commit himself or the sceptic of the PH to the doctrines.

69

Ibid.p.31

70

This assumption seems reasonable to Bumyeat. He claims, 'it is a fact of central importance

that truth, in the sceptic's vocabulary, is closely tied to real existence as contrasted with
appearance', 'The internal logic of Pyrrhonism requires that dogma and doxa Sextus does not
differentiate between these two terms really do mean: belief and, 'Sextus has no other notion
of belief than the accepting of something as true.'

99

Objections to Burnyeat's interpretation

PYRRHONIAN SCEPTIC is someone who possesses dunamis

antithetikos, the power or ability to set in opposition in any way

whatever. (PH I 8-11) In practice, the sceptic takes issue with the

dogmatist on a wide range of topics not restricted to philosophy. For the sake of
argument, the sceptic is generally accustomed to making temporary concessions
to the dogmatists without committing himself to any doctrine under scrutiny in
any sense. This is exactly what happens with the doctrines that the true is what
conforms to reality, and belief is tied to truth; it is assent to the non-evident. Far
from being committed to these dogmatic doctrines, Sextus just takes them for
granted in order to induce epoche dialectically. They are only hypotheses for
him. He renders them aporetic dialectically. The purpose of his arguments is not
to assert anything about the true, the apparent and the non-evident. There is no
reason to take Sextus as endorsing the claim that the true is what conforms to
reality. It is the dogmatists who hold this doctrine. Sextus defuses the
plausibility of this dogmatic doctrine with his arguments. His concern is to bring
about equipollence and epoche. It is wrong to suppose that the sceptic is
committed to the doctrine that the true is what conforms to reality.
As for the second doctrine, I want to argue that Sextus would not be limited
to the notion of belief as the accepting of something as true. Obviously, this

doctrine counts on the first doctrine. If Sextus is not committed to the true as
what conforms to real existence and as he argues that 'the criterion of truth has
appeared to be aporetic' at PH II 14-96, it seems that he would not be inclined
to load belief with truth, if he were to hold a position on belief. Actually, it
seems more likely that Sextus operates with a notion of belief that is not

100

restricted to the accepting of something as true. Barnes surveys the use of the
word dogma in 'The Beliefs of a Pyrrhonist'. His survey deals with the sense
and the colour of the word. He not only draws on Sextus' usage but also that of
all the major prose-writers from 400 B.C. to A.D. 250. He contends, 'outside
political contexts, "belief generally conveys the sense of dogma.''72 However
when it comes to the colour of the word, he observes that not every belief is a
dogma. 'Dogmata are weighty, substantial beliefs tenets, doctrines,
principles.' For instance, Philo of Alexandria, who is the first author to make
frequent use of the word, denotes philosophical tenets or religious beliefs by
dogma. Plato, Aristotle, Epicurus, Plutarch and Alexander of Aphrodisias all
use dogma to refer to philosophical beliefs. In addition, Lampe shows that
dogma is used to denote philosophical principles and religious doctrines.
Moreover Clement, Hippolytus and Origen use dogmata to refer to
philosophical, religious, or scientific beliefs. Furthermore, Galen in 'his works
shows clearly that a man may reject all dogmata and yet retain innumerable
beliefs.' Finally, dogma is also common in the works of the imperial Stoics. For
instance, Epictetus uses dogma to refer to philosophical tenets and practical or
evaluative judgements.73 But Barnes goes back to Sextus' use of dogma and its
cognates in PH and M. He declares that 'Sextus' use of dogma is entirely
comparable to the usage of Galen or of Clement or of any other Greek of that
era.' 'It is really plain that when Sextus uses a term from the dogma family he is
designating a philosophical principle or a scientific theory.'(p.73) Contrary to
Burnyeat's claim that Sextus does not differentiate between dogma and doxa,

71

But this does not mean that he holds that particular position on belief.

72

Ibid, p.69

73

Ibid, p.72
101

the survey suggests that Sextus and his fellow Greeks do not mix up dogma and
doxa or belief. Dogmata refer to philosophical, religious, or scientific beliefs.
Other beliefs are not classified as dogmata. They are ordinary beliefs and they
will be preserved though all dogmata are given up. Hence drawing on the
survey it is not the case that 'Sextus has no other notion of belief than the
accepting of something as true.'
In 'The Sceptic's Two Kinds of Assent', Frede advances a series of
comprehensive arguments against the claim that 'to accept an impression surely
is to accept it as true.' (p. 135) He suggests several ways in which one could
have the view that p without thinking that it is the case that p or p is true. The
main idea is a distinction between 'two kinds of assent such that having a view
involves one kind of assent, whereas taking a position, or making a claim,
involves a different kind of assent, namely the kind of assent a sceptic will
withhold.' (p. 128)

To just have a view is tofindoneself being left with an impression, to find

oneself having an impression after having considered the matter, maybe
even for a long time, carefully, diligently... it does not follow that the
impression one is left with is true, nor that one thinks that it is true, let
alone that one thinks that it meets the standards which the dogmatic
philosophers claim it has to meet if one is to think of it as true.
To make a claim, on the other hand, is to subject oneself to certain canons.
It does, e.g., require that one should think that one's impression is true and
that one has the appropriate kind of reason for thinking it to be true.
To be left with the impression or thought that/?, on the other hand, does
not involve the further thought that it is true that/?, let alone the yet further
thought that one has reason to think that p, that it is reasonable that p.
(p.133)
102

Having sketched the general idea of the distinction, Frede admits that it is
difficult to formulate it in detail. The problem lies in the fact that 'there is a
whole spectrum of distinctions with a very weak notion of having a view at one
extreme and a strong notion of taking a position at the other extreme.' (p. 134)
He finds Sextus drawing a distinction between two kinds of assent at

PH113.

By this distinction, the first kind of assent is assent (eudokeiri) to the affections
forced upon the sceptic by a phantasia. This assent to affections is passive. It is
a passive acquiescence in the affections. Frede contends there are two senses in
which one might accept or approve of affection. The dogmatist holds that to
assent to affection is to take it to be true, whereas for the sceptic to assent to
affection is simply to acquiesce in the affection without taking it to be true. The
first kind of assent is not a matter of choice while the second kind of assent can
be withheld at will.
Then he suggests several ways in which one could have the view that p
without thinking that it is the case that p or p is true. This is his first argument.

It might be the case that action does not, in addition to the impression
that p, require a positive act of assent or the further thought that it is true
that p. All that may be needed is one's acquiescence in the impression,
and all this may amount to is that in the series of impressions one has
reached an impression which produces an action rather than the kind of
disquiet which would make one go on to consider the matter further till
one reached an impression which one no longer resists and which
produces an action. Indeed one may have the view that p without even
entertaining the thought that p, let alone the further thought that p is true.
Things may have left us with the impression that/?, and we may act on
that view, without being aware of it. (p. 135)

103

For instance, an expert carpenter would be able to act on his professional

beliefs even though he is not thinking of what he is doing when he is actually
acting on them. But if his apprentice asks him about his beliefs when he is acting
on them, the carpenter might have to slow down his action and start to think
about them in order to explain to the apprentice. The carpenter might well
explain the beliefs and his activity to his apprentice after finishing his work. In
addition, if the carpenter had no apprentice before, it might be the case that it
would be the first time that he articulates some of his beliefs. It is not necessary
that to act on our view we have to entertain the corresponding thought and to
assent positively to it on that occasion.
The dogmatist may not object to the first sense of assent. In addition, the
first sense of assent would already be sufficient for action though it does not
involve taking the impression to be true.

For even the Stoics assume that the wise man will often act, not on the
basis of certain knowledge, but of wise conjecture. He is not omniscient,
and his rationality and wisdom are characterized exactly by his ability to
be rational or reasonable in his assumptions and actions even when he
lacks knowledge, as he inevitably will, in the complex situations of
everyday life. Nevertheless, he will do what is fitting or appropriate
because he will be able, as the Stoics themselves say, to give a
reasonable (eulogon) account of what he has done. (p. 136)

According to Frede, Aristotle has a rather similar view. It is the fact that
after the action we can correctly account for it that makes our action voluntary.

104

Similarly, the Stoic wise man, in order to do what is fitting, does not
necessarily actually have to go through some reasoning, overtly accept
or assent to the conclusion, and act on the basis of this. It, rather, is that
his action in hindsight can be explained in terms of such reasoning.

Moreover, the outlook of the sceptic would not be different from those who
take their views to be true if we consider the behaviour alone. In addition even if
the sceptic utters appearance sentences in reply to questions about his views on
any subject, he is just 'giving an autobiographical report' on the views that
guide his activity without any attempt to establish the truth of his views.
Barnes and Frede give separate but strong cases in which it is suggested
that belief could be separated from truth or assent to the non-evident. So it
appears that Sextus would not be confined to a restricted notion of belief as
Burnyeat has maintained. To insist that belief be tied to truth for Sextus is itself
sheer dogmatism.

Barnes

HE NOMENCLATURE of Pyrrhonian Scepticism is Zetetic

(questioning), Ephectic (suspensive), Aporetic (inclined to aporiai)

and Pyrrhonian (PH I 7). According to Sextus, it is 'the pathos that

arises concerning the subject of inquiry' that makes Pyrrhonian scepticism

suspensive. The suspensive pathos is epoche, 'a state of the intellect on account

of which we neither deny nor affirm anything' (PH I 10). We should note that
epoche as a pathos not only is suspensive but also has an object. It arises

105

concerning a certain subject of enquiry; we neither deny nor affirm a certain

subject of enquiry. Epochs appears to require an object. This characterization of
epoche has escaped the attention of interpreters of Pyrrhonian scepticism.
Barnes, however, is an exception. He is aware of this feature of epoche.
Any investigation attacks some specific subject matter and poses some
particular question. The state of epoche resulting from any investigation will
therefore itself be directed towards some specific subject matter and some
particular question.

Pyrrhonism thus works piecemeal. The dunamis antithetikos is a general

capacity, but it can only be exercised on particular issues. Epoche is not
a global state a state of total intellectual paralysis; rather, it is a
particular attitude, essentially directed towards some specific issues.
Epoche on one issue does not imply epoche on any other issue. Hence if
you ascribe epoche to a man you must indicate the object of his epoche:
towards what issue is epoche directed? and if a Pyrrhonist claims that
epoche is the route to ataraxia we must equally ask him to specify the
object of that epoche: over what range of issues is his epoche extended?
(p.59)

This observation about epoche is the underlying theme of Barnes'

interpretation of the nature and scope of Pyrrhonian scepticism as depicted in
the PH. It helps him to distinguish legitimate questions about Pyrrhonian
scepticism from questions that are wrong-headed. He suggests that

It is pointless to ask a Pyrrhonist whether we ought to suspend

judgement on this or that specified topic: epoche is not something to be
adopted or rejected at will. But it is wholly appropriate to ask where

106

over what range of topics a Pyrrhonist will exercise his dunamis

antithetikos, and hence to ask what is the scope of his Scepticism, (p.60)

Moreover, the question like '"What may a Pyrrhonist believe?", or "What

is the scope of Pyrrhonian epochs",

is of the last importance for an

understanding of ancient Scepticism'. The questions are wrong-headed. They

ignore the fact that epoche requires an object; 'it is a particular attitude,
essentially directed towards some specific issues.' In practice, the sceptics may
exercise their dunamis antithetikos over different topics and so end up with
different level and intensity of epoche. Some sceptics are radical or rustic; some
are moderate or urbane. Epoche cannot properly be detached from the related
subject of enquiry. This feature of epoche makes it futile to speculate on the
range of epoche in isolation. For these reasons the controversy between Frede
and Bumyeat on the scope of epoche is misplaced. Barnes, therefore, restricts
his discussion to a more plausible question: 'What is the scope of epoche in
Sextus' Outlines of Pyrrhonism? what, if anything, may the Pyrrhonist of the
Outlines believe?'(p.61) Two answers that define rival types of scepticism are
discussed. The first type of scepticism is called rustic Pyrrhonism and the
second type is urbane Pyrrhonism. Accordingly, 'The rustic Pyrrrhonist has no
beliefs whatsoever: he directs epoche towards every issue that may arise'
whereas 'The urbane Pyrrhonist is happy to believe most of the things that
ordinary people assent to in the ordinary course of events: he directs epoche
towards a specific target
roughly speaking, towards philosophical and
scientific matters.' (pp.61-2)
Frede argues that the Pyrrhonian sceptic has ordinary beliefs about what is
the case, but at the same time, he eschews all dogmatic beliefs that have to do

107

with assenting to non-evident things investigated by the sciences. This

interpretation gives what Barnes called urbane Pyrrhonism.

Now according to the urbane interpreter of PH, dogmata are beliefs of a

special sort: they are, roughly speaking, philosophico-scientific
opinions doctrines, principles, tenets. In rejecting dogmata, then, the
Pyrrhonist rejects not beliefs but doctrines; and insofar as the Pyrrhonist
is defined as a non-dogmatist, he is apparently able to admit and to
profess all ordinary beliefs, (p.67)

In reply to this 'urbane contention', Barnes surveys the use of dogma in

Greek writings in order to determine its sense and colour. We have seen this
survey earlier. His conclusion is that Sextus and his fellow Greeks do not mix up
dogma and doxa or belief. The survey is followed by a discussion of the
distinction between two sense of dogma made at PH I 13. Barnes takes the
distinction as explicit remarks by Sextus on the sense of dogma. This is his
translation of the passage.

We say that Sceptics do not dogmatise not in the sense [i] in which some
people say, fairly broadly, that dogma is eudokein tini pragmati...;
rather, we say that they do not dogmatise in the sense [ii] in which some
people say that dogma is an assenting to some object from among the
unclear things being investigated by the sciences, (p.74)

Drawing on the same passage, Frede argues that a serious Pyrrhonian

sceptic can have beliefs in sense [i] and reject scientific or philosophical beliefs
in sense [ii]. Barnes agrees with Frede on the meaning of the notion eudokeo

108

which means 'be content with something'and a proper English translation is

'acquiesce in'. 74 However, he disagrees with Frede on the status of the sceptic's
eudokein. Dogmata, as the survey has suggested, are weighty and substantial
beliefs; the word is used to denote philosophical, religious, or scientific beliefs.
Sextus' characterisation of dogma in sense [ii] conforms to this suggestion. It is
clear from Sextus that the sceptic does not have dogmata in sense [ii]; he will
not have philosophical or scientific beliefs. But when dogmata are taken in
sense [i] viz., eudokeo, it loses its colour of being weighty and substantial.
Sextus deploys this weak sense of dogmata only at PH I 13. Barnes conjectures
that it is 'a dialectical concession by the Pyrrhonists' making their avowal of
pathos a harmless dogmatizing. He argues that Sextus uses eudokeo 'to convey
a minimal notion of contentment a Pyrrhonist acquiesces in his pathe, he does
not speak out against them or deny them', (p.75) And 'his acquiescence, as
Sextus describes it, does not involve any beliefs..' For 'his eudokeo is a matter of
refraining from belief (he will not say 'I believe...'), and not a matter of
believing anything at all.' So for Barnes the sceptic may dwell on his pathe by
using phenomenological

sentences. But the dogmata he comes to hold

in this way are not any beliefs at all. They are merely expressions of
acquiescence.
However, for Frede, the sceptic's eudokeo amounts to non-dogmatic or
ordinary belief. Barnes supposes that dogmata in the sense of eudokeo are 'a
dialectical concession by the Pyrrhonists'. But Frede thinks differently. He
notices that eudokein 'has no philosophical or technical meaning, no
philosophical associations and is connected with no special philosophical
claims'. He believes that it is precisely for this reason that Sextus consciously

74

Ibid, p.74

109

explains dogmata in sense [i] by the phase eudokein tini pragmati. He believes
that dogmata in the sense of eudokein give a proper contrast to dogmata in the
usual dogmatic sense. The contrast reinforces the distinction between dogmatic
belief and ordinary belief. Actually, Frede goes even further to claim that Sextus
exhibits a conception of the origin of ordinary beliefs at PH I 13, according to
which 'something which can count as a belief, a judgement, arises in us when
we do not object and consequently consent.' Barnes, on the contrary, believes
that the eudokeo 'is a matter of refraining from belief. If Barnes is right about
the sense and colour of dogmata, then ordinary beliefs are not dogmata and
dogmata in the sense of eudokein are not beliefs of any sort. But this objection
to Frede's interpretation is not conclusive. The fact that Sextus uses dogmata
and its cognates to denote philosophical or scientific beliefs seems to support
the urbane contention that the sceptic's repudiation of dogmata concerns
dogmatic beliefs only. Frede's speculation that Sextus shows his conception of
the origin of ordinary beliefs at PH I 13 may not be true. A dogma in the sense
of eudokeo, in fact, may not be a belief of any sort. However, it might still be
possible for the sceptic to be 'able to admit and to profess all ordinary beliefs' in
some other way. At any rate, if the sceptic does eschew ordinary beliefs, it needs
to be shown that it is so.
It can be argued that ordinary beliefs are not dogmata; they do not refer to
are 'those things which do not have a nature of the sort to
fall under our direct perception (e.g. imperceptible pores) {PH II 98).' 75 Things
that have a nature to fall under our direct perception are called xa 7ipo8r|A.a.
They are the subject matter of our ordinary beliefs. It seems that ordinary beliefs
and dogmata are separated by their different subject matters. Yet, they are

75

Ibid, p.76
110

somehow related. Barnes argues that the sceptic's repudiation of dogmata will
commit the sceptic to abandon not only philosophico-scientific beliefs but also
ordinary beliefs. This is his argument.

A Pyrrhonist will only believe that the water is tepid if he judges it to be

so; and he can only judge it to be so if he possesses a criterion of truth by
which to judge it. But the thesis that there is a criterion of truth is itself a
dogmaindeed it is a perfect specimen of those philosophico-scientific
tenets which the Greeks called dogmata. Now the Pyrrhonist of PH
rejects all dogmata. Hence he will not haveor rather, will not believe
that he hasa criterion of truth. Hence he will not be able to judge, or to
believe, that the water is tepid.
In general, the Pyrrhonist of PH will have no ordinary beliefs at all.
Ordinary beliefs are not dogmata, nor do they advert to
Nonetheless, in rejecting dogmata the Pyrrhonist must reject ordinary
beliefs; for the possession of ordinary beliefs presupposes the
possession of at least one dogma
the dogma that there is a criterion of
truth. (P.78)

Fogelin (1994) finds this argument from the criterion 'not only wrong, but
deeply wrong'.76 This is his criticism of the argument.

The passage simply misrepresents the dialectical character of the

Pyrrhonian attack on the dogmatists. The Pyrrhonist does not hold the
view that judgments may not be made in the absence of a criterion of truth.
That view is held by the dogmatists... If the argumentfromthe criterion is
correct, it will have as a consequence that the dogmatist ought to suspend
76

Fogelin [1994] p.7

111

judgment on her dogmatic philosophical beliefs, and also on her ordinary

beliefs, for, as Barnes rightly notes, the argument applies equally to both.
But this leaves the Pyrrhonist untouched, for it is no part of his position to
suppose that judgment may only be made on the basis of a criterion of
truth. Not to see this is not to see what Pyrrhonian scepticism (whether it
is right or wrong persuasive or unpersuasive) is all about.

To judge by the passage alone, the criticism made by Fogelin may seem
overwhelmingly devastating. However, actually Barnes is aware of 'the
dialectical character of the Pyrrhonian attack on the dogmatists'. For instance,
as we have seen earlier he regards the occurrence of the weak sense of dogmata
at PH I 13 as 'a dialectical concession by the Pyrrhonists'. Hence, I think he
does not fail to 'see what Pyrrhonian scepticism is all about'. Nonetheless,
Fogelin's criticism makes a point. For dialectical purposes the Pyrrhonist
temporarily takes up the dogmatic view that judgements may not be made in the
absence of a criterion of truth and then proceeds to argue that the dogmatist
ought to suspend judgement on both his dogmatic beliefs and ordinary beliefs.
The Pyrrhonist is not affected by his own arguments against the dogmatist. It
could be the case that he does not eschew ordinary beliefs. Therefore, Barnes's
argument from the criterion does not establish the conclusion that the sceptic
gives up ordinary beliefs because of his abandoning of dogmatic beliefs about
the criterion of truth. But it might be feasible in some senses for the sceptic to
give up ordinary beliefs. To see how this is feasible, we have to make a few
things clear first.
Bames dismisses questions like 'What may a Pyrrhonist believe?' or
'What is the scope of Pyrrhonian epoche?' from the beginning of his discussion.
The questions, generally posed, fail to relate epochs to any specific issue. Hence,

112

they are wrong-headed. Therefore, Bames focuses on Sextus and the scope of
epoche in PH. This concern transforms into a question: is PH rustic or urbane?
However, 'the question itself may be ill-conceived.'77 After drawing the
conclusion that PH is inconsistent, he finishes his whole discussion with an
attempt to show that 'the problem of the scope of epoche is in a certain sense
unreal.'(p.89)
At PH I 12 Sextus reports, 'the causal origin of Pyrrhonian Scepticism is
the hope of attaining ataraxia. Certain talented people, upset by anomalia in
"the facts" and at a loss as to which of these "facts" deserve assent, endeavoured
to discover what is true in them and what is false, expecting that by settling this
they would achieve ataraxta.' Before that Pyrrhonian scepticism is described as
a disposition {dunamis antithetikos) to induce epoche by the equipollence
among arguments and then ataraxia. {PH I 8) Barnes argues that

a Pyrrhonist will only reach epoche if he exercises his dunamis

antithetikos; he will only exercise his dunamis antithetikos if he finds
himself suffering from tarache; he will only suffer from tarache if he
perceives a worrying anomalia in things...
The point of Pyrrhonism is ataraxia. Pyrrhonist strategies are relevant
only where tarache exists. A man who suffers only mildlyfromtarache
may be a perfect Pyrrhonist; for he may achieve complete ataraxia by
exercising his dunamis antithetikos and reaching epoche in a very
modest way. Others, whofindthe whole of life a sea of troubles, will not
be set at rest until they have achieved universal epoche. (p.90)

77

Bumyeat and Frede [1997] p.63

113

A worrying anomalia in things gives rise to tarache. Tarache triggers off

the Pyrrhonists' dunamis antithetikos. Dunamis antithetikos brings about
epochs. People suffer from different worrying anomaliai in things and so they
exercise dunamis antithetikos over different topics. The level and intensity of
epoche brought about varies from one person to another. Some Pyrrhonists are
radical or rustic; some are moderate or urbane. Barnes supports this argument
by the 'medical simile' found at PH III 280-1.

Tarache is a disease, epoche the cure. The Pyrrhonist is a doctor a

psychiatrist who claims the ability to cure tarache in most of its
forms... Some conditions require massive doses and major surgery,
others are assuaged by an aspirin. It is absurd to imagine that doctors can
produce a single formula, applicable to all men in all conditions, or
pronounce generally that every patient needs so many pills a day.
How much epoche does a man need for ataraxia or mental health? How
far will a competent Pyrrhonist apply his tropoi and exercise his
dunamis antithetikos? Plainly it all depends on the disease. Serious
mental conditions require strong remedies, minor maladies are righted
by a simple argument or two. It is absurd to suppose that a Pyrrhonist
can produce a single formula, applicable to all men in all conditions, or
pronounce generally that every patient needs so much epoche a day and
so many tropoi a day. (pp. 90-1)

The tension caused by the apparently inconsistent conclusion that 'The

general tenor of PH is... indubitably rustic. But PH also contains important
intrusions of urbanity' is resolved.

'Epoche may be broad or narrow.

Pyrrhonism may be rustic or urbane. Everything depends on the state of the

particular patient.'(p.91) The scope of every sceptic's epoche is not stagnant. It

114

varies and it corresponds to the tarache suffered. Some Pyrrhonists are radical
or rustic; some are moderate or urbane. Some patients need a wide-ranging
epoche; some will be fine with a confined epoche. Therefore, the idea that the
sceptic eschews ordinary beliefs is feasible. Some Pyrrhonists may 'find the
whole of life a sea of troubles,' and they 'will not be set at rest until they have
achieved universal epoche.'

Conclusion

OW, IT IS TIME for the conclusion of this chapter. So far, I have

reviewed different interpretations on the nature and scope of

Pyrrhonian scepticism put forward by Frede, Bumyeat, and Barnes.

The review is concerned with how they see the true meaning of PH I 13 and its

exact bearing on belief and epoche. In this passage, the notion of dogmata in its
usual sense is contrasted with dogmata in the sense of eudokeo. The three
interpretations agree with each other on dogmata in its usual sense; none of
them suggests that the sceptic comes to hold dogmata in its usual sense. It is the
dogmata in the sense of eudokeo that gives rise to the controversy. Frede's
perception of dogmata in the sense of eudokeo sides with the urbane
interpretation. For him, the sceptic eschews dogmatic beliefs or doctrines in the
usual sense of dogmata, but his eudokeo amounts to ordinary beliefs. Bumyeat
advocates the rustic interpretation. His interpretation accommodates no
ordinary beliefs. He thinks that dogma in the sense of eudokeo is not a belief at
all; it 'is a passive phantasia (impression) or pathos (experience), expressed in a
statement which makes no truth-claim about what is the case.' (p.51) Barnes

115

does not indulge in urbane or rustic interpretations. For him Pyrrhonian

scepticism could be rustic or urbane; there is no single Pyrrhonian orthodoxy on
the scope of Pyrrhonian scepticism. The question about the scope of Pyrrhonian
scepticism is indeed an open-ended question. As far eudokein is concerned, he
argues that it 'is a matter of refraining from belief (he will not say 'I believe...'),
and not a matter of believing anything at all.' Up to a certain point, Bumyeat
and Barnes share the same view, namely that dogma in the sense of eudokeo is
not a belief. However, when it comes to the bearing of this upon belief and
epoche, I think Barnes is more persuasive. He seems to have suggested a liberal
interpretative option on Pyrrhonian scepticism that is less rigid. The Pyrrhonian
sceptic would be free from the dichotomy between rustic and urbane. He points
out that PH I 13 is silent on the status of ordinary beliefs. It involves the notion
of dogmata in its usual sense but also in the sense of eudokeo, which are not
beliefs of any sort. It says nothing about ordinary beliefs. As far as epoche is
concerned, he notices that it requires an object. This comprehension leads him
to make a profound point: 'Epoche may be broad or narrow. Pyrrhonism may be
rustic or urbane. Everything depends on the state of the particular patient.'
Finally, I want to finish this extended discussion with a few words from
Burnyeat, who calls for a relocation of scepticism in modern philosophy from
the ancient scepticism.

The other important thing about Clarke's sceptic, and about most of the
references to "the sceptic" in modem philosophical literature is that this
sceptic has no historical reality. It is a construction of the modem
philosophical imagination. The point is that when scepticism goes
transcendental, the expression "the sceptic" has to lose the historical
reference it still carries in Hume, its connection with what certain

116

historical figures actually said and thought. It becomes the name of

something internal to the philosopher's own thinking, his alter ego as it
were, with whom he wrestles in a debate which is now a philosophical
debate in the modern sense, (p. 122)

117

CHAPTER FIVE

Concluding Remarks on Pyrrhonian Scepticism and Philosophy

HILOSOPHICAL SCEPTICISM is deeply rooted in philosophy. If

we were to have a conceptual museum of scepticism, Pyrrhonian

scepticism would arguably occupy a main, if not the greatest, hall in

the museum. The reasons are not hard to tell. As my discussions have revealed,
the issues raised by the Pyrrhonists are profound and persistent in contemporary

philosophical discussions. For instance, epistemic justificationalists, as we have

seen, are still struggling with the Agrippan Problem and the status of epistemic
extemalism remains under strict scrutiny.
The main subject of this thesis is Pyrrhonian scepticism and philosophy. At
several points of my discussions, I have, sometimes implicitly, touched upon
some fundamental issues in philosophy. To match them in pairs, they are the
possibility of knowledge and the implication of unknowability, philosophical
scepticism and philosophical relativity. All of these issues deserve a serious
treatment on their own accord. I have singled out scepticism and referred to
other issues when appropriate.
Pyrrhonian scepticism is so prolific that it flooded philosophy and other
discourses with arguments. In addition, it was even described as a way of
attaining ataraxia. But time has passed since the moment championed by the
Pyrrhonists like Pyrrho, Aenesidemus and Agrippa. There was a long period in
philosophy as if the Pyrrhonists were forgotten. Recently, the Pyrrhonists have
come to people's attention again. Though so much has changed since the ancient
118

Greeks, Pyrrhonian scepticism continues to provide a stimulating perspective

from which we could reflect on the nature of philosophy and its course of
development. Indeed, the recent revival of interest in Pyrrhonian scepticism
opens up a refreshing opportunity for us to rethink not only scepticism but also
philosophy.
In the following discussion, I attempt to indicate some lines which such
rethinking might take by making some (necessarily cursory) comparisons. I
shall indicate two themes: the therapeutic conception of philosophy and
Husserlian epoche as a methodological variant of the Pyrrhonian epoche. For the
the first theme, I sketch Wittgenstein's notion of the therapeutic role of
philosophizing, to compare and contrast it with that of Sextus. Strawson's
rebuttal of the therapeutic role of philosophizing and defence of the call of
theory is outlined. Then I discuss Husserl's appropriation of the term epoche for
his phenomenological reduction, which while having some affinities with
Pyrrhonism, is used in a quite different way to provide a starting point for
elaborate and systematic philosophical theorizing.
I shall first summarize the results of my previous discussions on Pyrrhonian
scepticism. At the outset of his exposition of Pyrrhonian scepticism, Sextus
sorted out the most fundamental kinds of philosophy as: the Dogmatic, the
Academic, and the Sceptical. Sceptical philosophy (i.e. Pyrrhonian scepticism)
was characterized as 'an ability to set out oppositions among things which
appear and are thought of in any way at all'. The sceptic was someone who
possessed the ability to set out oppositions among things in order to induce
equipollence among arguments. It was also claimed that ataraxia would follow
equipollence among arguments as a shadow followed a body. Being framed in
this way, Pyrrhonian scepticism appeared to be therapeutic; it offered treatment
of mental disturbance. This therapeutic view of Pyrrhonian scepticism might be
119

inaccurate from the original Pyrrhonian scepticism. For actually, Sextus'

account of Pyrrhonian scepticism in the PH was the cumulative result of
sceptical movements led by various figures. It is highly probable that Pyrrhonian
scepticism underwent significant changes over the time span from Pyrrho via
Aenesidemus and Agrippa to Sextus, a period of about five hundred years that
witnessed its ups and downs. It could also be possible that Sextus, who was a
doctor of the Empirical school, brought in 'new elements' in his articulation of
Pyrrhonian scepticism. The result was a dyed Pyrrhonian scepticism. These are
interesting questions that go beyond my discussion here. While Sextus remains
the only extensive textual source of Pyrrhonian scepticism, my discussion has to
draw heavily on Sextus, assuming that his articulation gives a prima facie true
picture of Pyrrhonian scepticism. At worst, my discussion would be confined to
Sextus' version of Pyrrhonian scepticism, should we discover an earlier account
of Pyrrhonian scepticism that was drastically different.
Sextus gave a brief outline of the disease and the intellectual treatment
offered by the sceptic. By the sceptic's diagnosis, the source of illness is
anomalia (irregularity) in things. People wanted to resolve the anomalia that
troubled them and acquire ataraxia by deciding what they should assent to.
Unfortunately, after pursuing the issue people remained disturbed and puzzled
over which propositions deserved assent and which did not. There came the
sceptic and his recommendation. Epoche was the cure and the various tropoi
were the prescription. The treatment of mental disturbance by arguments and
epoche seemed to be supported by the experience of some sceptics who acquired
ataraxia after epoche. 'For Sceptics began to do philosophy in order to decide
among appearances and to apprehend which are true and which false, so as to
become tranquil; but they came upon equipollent dispute, and being unable to
decide this they suspend judgement. And when they suspended judgement,
120

ataraxia in matters of opinion followed fortuitously.' (PH I 26) Sextus explicitly

recognized 'ataraxia in matters of opinion and moderation of feeling in matters
forced upon' as the aim of Pyrrhonian scepticism. Hence, the purpose of
Pyrrhonian scepticism, as it follows, is to achieve ataraxia by epoche, which in
turn, is the result of equipollence among arguments.78 The Five Modes, which
were founded on a discerning understanding of the nature ofjustification, are the
most powerful and tactical patterns of argument formulated by Agrippa to assist
the sceptic to induce epoche and Sextus was so convinced of the potency of the
Five Modes that he claimed, 'every object of investigation can be referred to
these modes'.
At the end of Book III, the last book of PH, Sextus finished his exposition
with a final remark on Pyrrhonian scepticism. This final remark explained in
some detail the practice of therapeutic scepticism. It also identified the 'virus'
that the sceptic wanted to combat.

Sceptics are philanthropic and wish to cure by argument, as far as they can,
the conceit and rashness of the Dogmatists. Just as doctors for bodily
afflictions have remedies which differ in potency, and apply severe
remedies to patients who are severely afflicted and milder remedies to
those mildly afflicted, so Sceptics propound arguments which differ in
strength
they employ weighty arguments, capable of vigorously

78

1 am not going to examine the validity of the claim that epoche induces ataraxia. Nor will I

attempt to evaluate how effectively epoche induces ataraxia. I would like to pass them over and
focus on the philosophical consequences of the therapeutic conception of philosophy. Hence, I
leave it aside, the inconsistency between the image of a shadow following a body (which
suggests a natural and inseparable connection between equipollence/epoc/ie and ataraxia) and
the passage, which states that the coming of ataraxia in these circumstances is fortuitous.

121

rebutting the dogmatic affliction of conceit, against those who are

distressed by a severe rashness, and they employ milder arguments against
those who are afflicted by a conceit which is superficial and easily cured
and which can be rebutted by a milder degree of plausibility.This is why
those with a Sceptical impulse do not hesitate sometimes to propound
arguments which are sometimes weighty in their plausibility, and
sometimes apparently rather weak. They do this deliberately, since often a
weaker argument is sufficient for them to achieve their purpose. (PH HI
280-1)

Sextus expressed elsewhere the view that the sceptic was concerned with
the rashness of the Dogmatists. Many if not all of the arguments in the PH
targeted the rashness of the Dogmatists. It seemed as if the sceptic of the PH or
Sextus shouldered the responsibility of refuting the rashness of the Dogmatists.
For instance, he maintained at PH I 20 that the sceptic did not want to reject
what was apparent, but if he needed to expose the rashness of the Dogmatists by
propounding arguments against what was apparent, he would do it. In addition,
as we have seen earlier, the Five Modes were handed down to reinforce the
challenge against the Dogmatists launched by the Ten Modes. At the end of his
exposition, Sextus explicitly revealed that the philanthropic sceptic wished to
cure by argument, as far as they could, the conceit and rashness of the
Dogmatists. Therefore, to cure the rashness of the Dogmatists must be a very
important endeavour for the sceptic, besides ataraxia, the aim of scepticism.
The PH, as it appeared, was a series or a record of intellectual treatment by
arguments against the rashness of the Dogmatists.
The therapeutic conception of philosophy did not pass into history with
Sextus and the Pyrrhonian sceptics. Contemporary philosophy observed a

122

radical development of the therapeutic conception of philosophy in one of the

most distinguished philosophers of our time. Wittgenstein expressed something
which looked like a modem version of the therapeutic conception of philosophy
in his Philosophical Investigations?9 The ancient sceptic wished to cure by
arguments the rashness of the Dogmatists and to achieve ataraxia by epochs.
For Wittgenstein too, 'The philosopher's treatment of a question is like the
treatment of an illness.' (255) There is a fundamental difference in the first
place. Wittgenstein's treatment was intended to solve problems. But the sceptic
just wanted to induce epochs. So while the sceptic recommended epochs and
gave arguments to balance the force of the Dogmatists' arguments, Wittgenstein
focused on language and proposed 'grammatical investigation' that 'sheds light
on our problem by clearing misunderstandings [concerning the use of words]
away'. (90) This is his brief outline of the grammatical investigation:

It was true to say that our considerations could not be scientific ones...
And we may not advance any kind of theory. There must not be anything
hypothetical in our considerations. We must do away with all explanation,
and description alone must take its place. And this description gets its
light, that is to say its purpose, from the philosophical problems. These are,
of course, not empirical problems; they are solved, rather, by looking into
the workings of our language, and that in such a way as to make us
79

Though I shall be referring to what I called 'the Wittgensteinian account of philosophy', it

does not mean I found a theory of philosophy in Philosophical Investigations. As I understand

the book, he has no intention of giving such a theory in it. In his own words, what appeared in the
book were his 'philosophical remarks' on a wide range of subjects. The so-called
Wittgensteinian account of philosophy composed of a few of his remarks. I am not going to offer
any formal comparison of the PH and Philosophical Investigations. I just want to review some
common features as well as some divergences between the two works.
123

recognize those workings: in spite of'an urge to misunderstand them. The

problems are solved, not by giving new information, but by arranging
what we have always known. Philosophy is a battle against the
bewitchment of our intelligence by means of language. (109)

Grammatical investigation is opposed to theory and explanation. Problems

are not solved by explanations or any theory. They are rather solved by
descriptions of how language works. Wittgenstein did not favour theory or
explanation as a means to resolve problems in philosophy. In addition, he
seemed to hold a restricted view of philosophy in the following remarks.

Philosophy simply puts everything before us, and neither explains nor
deduces anything. Since everything lies open to view there is nothing to
explain. For what is hidden, for example, is of no interest to us...
The work of the philosopher consists in assembling reminders for a
particular purpose. (126-127)

Wittgenstein once urged us to 'do away with all explanation' and not to
'advance any kind of theory' in philosophy. Now he explicitly suggested the
philosopher's job was to lay out things or what we have always known clearly in
the manner of assembling reminders for a particular problem, so that the
problem would be solved. To be more accurate, the problem would actually
disappear when things lie open to view. Wittgenstein thought that once we are
reminded of the proper employment of the words and concepts concerned we
would come to realize that the problem arise because of 'the bewitchment of our
intelligence by means of language'. Hence, the problem would cease to be a
problem any more when reminders are assembled.

124

Strawson made his brief criticism of the therapeutic conception of

philosophy in Analysis and Metaphysics.

According to him, the conception

'may seem very implausible, perhaps even shocking; at least exaggerated and
one-sided'.

Actually, he saw in it 'a total abdication of philosophical

responsibility', (p.5) The philosophical responsibility arises in the absence of an

account of the general conceptual structure.

we know, in a sense, what knowing is perfectly well long before we hear

(if we ever do hear) of the Theory of Knowledge. We know, in one sense,
what it is to speak the truth without perhaps suspecting that there are such
things as Theories of Truth. We learn to handle the words "the same",
"real", "exists", and to handle them correctly, without being aware of the
philosophical problems of Identity, Reality, and Existence. In the same
way, we learn to operate with a vast and heterogeneous range of notions:
ethical notions...; temporal and spatial concepts; the ideas of causality and
explanation; ideas of emotions...; [ideas] of mental operations of various
kinds...; [ideas] of perception and sense experience; whole ranges of
classificatory concepts... Of course we leam the words which express
these concepts in a variety of ways; but we learn them largely without
benefit of anything which could properly be called general theoretical
instruction. We are not introduced to them by being told their place in a
general theory of concepts. Such instruction as we do receive is severely
practical and largely by example. We leam largely by copying and by
occasional correction; as children leam to speak grammatically before
they hear of grammars, (pp.6-7)

80

Strawson [1992] p.3

125

'Mastery of a practice does not involve an explicit mastery (though it may

sometimes be allowed to involve an implicit mastery) of the theory of that
practice.' For example, speakers of a language speak grammatically and
correctly but they might not be able to state the grammatical rules and theories as
the grammarians would have stated them. It is the philosophical responsibility of
the philosopher to work out a general theory of concepts by conceptual analysis;
'the philosopher labours to produce a systematic account of the general
conceptual structure of which our daily practice shows us to have a tacit and
unconscious mastery.' (p.7)
Strawson preferred the call of theory in philosophy to philosophy as
therapy, as did Husserl, in a quite different way, as we are going to see soon.
Wittgenstein considered that feeling that call is itself a symptom of something
requiring a cure. Both Wittgenstein and the Pyrrhonian sceptic expressed a
therapeutic conception of philosophy. How would the Pyrrhonian sceptic reply to
the call of theory? Wittgenstein radically attempted to eliminate the very
enterprise of theorizing in philosophy. By contrast, the ancient sceptic was not
opposed to theorizing. Indeed, as we have seen, the word skeptikos itself comes
from the verb meaning 'to inquire'. What the sceptic disputed was the rashness of
the Dogmatists in the process of theorizing. 'And the Sceptics are still
investigating.' Though the form of their investigation is rather unclear to us.
Nevertheless, we can expect (and we find in the writings of Sextus) that the
sceptics would be keen on giving arguments of the sort that seem indispensable
in philosophy, regardless of how one might conceive of it. In this way,
Pyrrhonian scepticism represents the critical character of philosophy.
Now I move on to discuss the methodological significance of Pyrrhonian
scepticism

and Husserl's

appropriation

of the term

epoche for his

phenomenological reduction. Husserl was no Pyrrhonist but the notion of epoche

126

is crucial to his phenomenology. This could be seen in his stress on the idea that
philosophy should not 'model itself after the exact sciences methodologically' and
his advocacy of epoche as 'a new and radically new method' for philosophy. 81
Husserl's conception of philosophy and phenomenology requires detailed
eludication, though I can only sketch this here. We shall begin with his concern
about the problem of the possibility of cognition. According to Husserl, 'the
correlation between cognition as mental process, its referent (Bedeutung) and
what objectively is... is the source of the deepest and most difficult problems.
Taken collectively, they are the problem of the possibility of cognition.' (p. 15)
The gist of the problem of the possibility of cognition, as depicted by Husserl, is as
follows.

In perception the perceived thing is believed to be directly given. Before

my perceiving eyes stands the thing. I see it, and I grasp it. Yet the
perceiving is simply a mental act of mine, of the perceiving subject.
Likewise, memory and expectation are subjective process; and so are all
thought processes built upon them and through which we come to posit
that something really is the case and to determine any truth about what is.
How do I, the cognizing subject, know if I can ever really know, that
there exist not only my own mental processes, these acts of cognizing,
but also that which they apprehend? How can I ever know that there is
anything at all which could be set over against cognition as its object?82

It is plain here that Husserl was addressing traditional problems in the

theory of knowledge. He had his expectations of the theory of knowledge. 'The
task of the theory of knowledge or the critique of theoretical reason is, first of all,
81
82

Husserl [1973] p.21

Ibid. p. 15-16
127

a critical one.' The task he gave to the theory of knowledge was rather unusual.
For Husseri:

the positive task of the theory of knowledge is to solve the problems of

the relations among cognition, its meaning and its object by inquiring
into the essence of cognition. Among these, there is the problem of
explicating the essential meaning of being a cognizable object or what
comes to the same thing, of being an object at all: of the meaning which
is prescribed (for being an object at all) by the correlation a priori (or
essential correlation) between cognition and being an object of
cognition, (pp. 17-8)

The theory of knowledge, being described as such, was also phrased as

'the critique of cognition'. 'Precisely by solving these problems the theory of
knowledge qualifies as the critique of cognition, more exactly, as the critique of
natural cognition in all the sciences of a natural sort. It puts us, in other words, in
a position to interpret in an accurate and definitive way the teachings of these
sciences about what exists.' (p. 18) It seemed here that Husseri allowed philosophy
and science to be complementary to each other. But as we shall soon come to see,
philosophy and science were fundamentally different and the call for 'a new and
radically new method" for philosophy i.e., epoche or the phenomenological
reduction arises precisely out of the fundamental difference that Husseri found
between philosophy and science.

Only with epistemological reflection do we arrive at the distinction

between the sciences of a natural sort and philosophy. Epistemological

83

ldid.P.17
128

reflection first brings to light that the sciences of a natural sort are not yet
the ultimate science of being. We need a science of being in the absolute
sense. This science, which we call metaphysics, grows out of a "critique"
of natural cognition in the individual sciences.
If we then disregard any metaphysical purpose of the critique of cognition
and confine ourselves purely to the task of clarifying the essence of
cognition and of being an object of cognition, then this will be
phenomenology of cognition and of being an object of cognition and will
be the first and principal part of phenomenology as a whole.84

The critique of cognition was 'the condition of the possibility of a

metaphysics', (p.l) Philosophy via metaphysics was made a science of being and
the critique of cognition was the subject matter. Phenomenology was primarily
about the task of the critique of cognition. It should be noted that when Husserl
mentioned philosophy as a science or phenomenology as a science of pure
phenomena, he meant 'science in a completely different sense, and with
completely different problems and methods', (p.46) While making philosophy a
science, Husserl was opposed to the attempt to model philosophy after science
methodologically.

In contemporary philosophy, insofar as it claims to be a serious science,

it has become almost a commonplace that there can be only one method
for achieving cognition in all the sciences as well as in philosophy. This
conviction accords wholly with the great philosophical traditions of the
seventeenth century, which also thought that philosophy's salvation lay
wholly in its taking as a model of method the exact sciences, and above
84
Ibid p. 18 The task of the critique of cognition was expanded later. It was 'to clarify, to cast light
upon, the essence of cognition and the legitimacy of its claim to validity that belongs to its essence;
and what else can this mean but to make the essence of cognition directly self-given, (p.25)

129

all, mathematics and mathematical natural science. This putting

philosophy methodologically on a par with the other sciences goes hand
in hand with treating them alike with respect to subject matter.
It is still the prevailing opinion that philosophy and, more specifically,
ontology and the general theory of knowledge not only relate to all the
other sciences, but also that they can be grounded upon the conclusions
of those other sciences: in the same way in which sciences are built upon
one another, and the conclusions of one of them can serve as premises
for the others.
I am reminded of the favorite ploy of basing the theory of knowledge on
the psychology of cognition and biology. In our day, reactions against
these fatal prejudices are multiplying. And prejudices they are. (p. 19)

Husserl perceived philosophy as something 'which begins in the critique

85

of cognition and which, whatever else it is, is rooted in the critique of cognition'.
He stressed twice, that 'In contradistinction to all natural cognition, philosophy
lies... within a new dimension; and what corresponds to this new dimension, even
if, as the phrase suggests, it is essentially connected with the old dimensions, is a
new and radically new method which is set over against the "natural" method.'
Finally he went even further to conclude that anyone 'who denies this has failed to
understand entirely the whole of the level at which the characteristic problem of
the critique of cognition lies, and with this he has failed to understand what
philosophy really wants to do and should do, and what gives it its own character
and authority vis-a-vis the whole of natural cognition and science of the natural
sort.'

85

Ibid.p.20-21
Ibid, p.21, see also p. 19

86

130

Husserl insisted on phenomenology as 'a system of scientific disciplines',

denoting 'a method and an attitude of mind, the specifically philosophical attitude
of mind, the specifically philosophical method'. The philosophical method, as
Husserl revealed, aimed at detachment from 'the transcendent' or the
'phenomena'. In his own words:

Consequently, the idea of phenomenological reduction acquires a more

immediate and more profound determination and a clearer meaning. It
means...the exclusion of the transcendent as such as something to be
accepted as existent, i.e., everything that is not evident giveness in its
true sense, that is not absolutely given to pure "seeing"... Inductive or
deductive scientific conclusions or facets, etc.,fromhypotheses, facts,
axioms, remain excluded and are allowed only as "phenomena"; and the
same with all reference to any "knowing" and "cognition": inquiry must
concern itself always with pure "seeing" and, therefore, not with the
genuinely immanent. It is inquiry within the sphere of pure evidence,
inquiry into essences. We also said that its field is the a priori within
absolute self-givenness. (p.7)

Phenomenological reduction, in principle, would require that 'pure

philosophy, within the whole of the critique of cognition and the "critical"
disciplines generally, must disregard, and must refrain from using, the intellectual
achievements of the sciences of a natural sort and of scientifically undisciplined
natural wisdom and knowledge.'(p. 19) In practice, 'the entire world of nature,
physical and psychological, as well as one's own human self together with all the
sciences which have to do with these objective matters, are put in question. Their
being, their validity are left up in the air.'(p.22) This was the epoche which
'presuppose[s] nothing as already given', and 'which must begin with some
131

cognition which it does not take unexamined from elsewhere but rather gives to
itself. This Husserlian epoche, I would say, shared the same spirit of the
Pyrrhonian epoche, 'a state of the intellect on account of which we neither deny
nor affirm

anything'. The appropriation of the term epoche for the

phenomenological reduction served as a starting point for phenomenological

theorizing in philosophy. Nevertheless, my discussion of the Husserlian epoche as
a methodological variant of the Pyrrhonian epoche is intended to open a
perspective, and I do not make it my task here to pursue this large matter any
further.
Looking backward, I have been depicting different conceptions of
philosophy and sketching corresponding methodologies favoured. From the
Pyrrhonist, via Wittgenstein and Strawson to Husserl, the nature of philosophy
varies and the methodology gains complexity and heads in different directions.
This is the philosophical journey. Indeed it is an on-going and lively expedition. In
one moment of time, questions are raised. Talented people of different inclination
and mentality, regardless of diversity in concern and methodology, nourish
multifarious responses to the questions and in this very way the philosophical
horizons expand, and the philosophical enquiry goes on.

132

Glossary
For the reader's convenience, a glossary of Greek words is given as below,
together with relevant definitions, as given by Liddell, Scott and Jones.
adoxastos
akatalepsia
alethes
anepikritos
anomalia
antithetikos

apangelia
aporia
ataraxia
azetetos
dogma
doxa

free

from doxa, not opining

(adverb: adoxastos)
inability to comprehend or attain conviction
unconcealed, so true, real, opp. false, apparent
not decided, indeterminate
irregularity
(plural: anomaliai)
setting in opposition
(dunamis antithetikos: the power or ability
to set in opposition)
report
(plural: apangeliai)
puzzle
(plural: aporiai)
impasssiveness, calmness
outside the scope of enquiry
that which seems to one, opinion or belief
(plural: dogmata)
notion, opinion, judgement

epoche
eudokeo

suspension of judgement
be content with
(infinitive: eudokein)

glukazesthai

be affected with a sensation of sweetness

hairesis

choice, purpose, also

system of philosophic principles, or
those who profess such principles, sect, school

koinos

common (first meaning),

also (in Logic) general, universal
(comparative neuter adjective: koindteron,
more common, more general)

pathos

experience, sensation
(plural: pathe)
appearance, imagination
cool
(passive infinitive: psuchesthai, to be cooled)

phantasia
psucho

133

skeptikos

thoughtful,
also hoi skeptikoi (also aporetikoi, ephektikoi) the Sceptics,
or philosophers who asserted nothing positively,
followers of Pyrrho

tarache
thermaino

disturbance
heat
(passive infinitive: thermainesthai,
be heated, feel the sensation of heat)
way, method, trope
(plural: tropoi)

tropos

134

Bibliography
Algra, K., Barnes, J., Mansfeld, J. & Schofield, M. (edd.) [1999] The Cambridge
History of Hellenistic Philosophy Cambridge: Cambridge University Press
Amico, R. [1993] The Problem of the Criterion Maryland: Rowman &
Littlefield
Annas, J. and Barnes, J. [1985] The Modes of Scepticism: Ancient Texts and
Modern Interpretations Cambridge: Cambridge University Press
Aristotle [1994] Posterior Analytics (translated with a commentary by J. Barnes)
Oxford: Clarendon Press
Audi, R. [1998] Epistemology: A Contemporary Introduction to the Theory of
Knowledge London: Routledge
Bailey, A. [1990] 'Pyrrhonian Scepticism and the Self-Refutation Argument',
Philosophical Quarterly vol.40 (no. 158): 27-44
Barnes, J. [1989] 'Some Ways of Scepticism in Epistemology', in Everson
[1990]
Barnes, J. [1990] The Toils of Scepticism Cambridge: Cambridge University
Press
Bett, R. [2000] Pyrrho, his Antecedents, and his Legacy Oxford: Oxford
University Press
Brunschwig, J. [1994] Papers in Hellenistic Philosophy (translated by Janet
Lloyd) Cambridge: Cambridge University Press
Burnyeat, M. F. [1982] 'Idealism and Greek Philosophy: What Descartes Saw
and Berkeley Missed', Philosophical Review 91 (no.l): 3-40
Burnyeat, M. F. (ed.) [1983] The Skeptical Tradition Berkeley: University of
California Press
Burnyeat, M. F. and Frede, M. (ed.) [1997] The Original Sceptics: A
Controversy Cambridge: Hackett Publishing Company
Chisholm, R. M. and Keim, R. G. [1972] 'A System of Epistemic Logic', Ratio
14:99-115
Chisholm, R. M. [1973] The Problem of the Criterion Milwaukee: Marquette
University Press

135

Chisholm, R. M. [1982] The Foundations of Knowing Minneapolis: University

of Minnesota Press
Clark, T. [1972] "The Legacy of Skepticism', Journal of Philosophy 69: 764-69
Cohen, L. J. [1992] An Essay on Belief and Acceptance Oxford: Clarendon Press
DeRose, K. and Warfield, T. A. (ed.) [1999] Skepticism: A Contemporary
Reader Oxford: Oxford University Press
Diogenes Laertius [1950] Lives of Eminent Philosophers (Greek text and
translatedby R. D. Hicks) Loeb Classical Library, 2 volumes, Cambridge:
Harvard University Press
Dretske, F. [1997] 'So do we know or don't we', Philosophy and
Phenomenological Research vol. 57(no.2): 407-409
Eusebius [1981] Preparation for the Gospel, Edwin Hamilton Gifford, 2
volumes, Grand Rapids: Baker Book House (Reprint of the 1903 Clarendon
edition)
Everson, S. (ed.) [1990] Epistemology, Companions to Ancient Thought
Cambridge: Cambridge University Press.
Floridi, L. [1985] 'The Diffusion of Sextus Empiricus's Works in the
Renaissance', Journal of the History of Ideas vol.56 (no.l): 63-85
Fogelin, R. J. [1983] 'The Tendency of Hume's Scepticism', in Bumyeat [1983]
Fogelin, R. J. [1994] Pyrrhonian Reflections on Knowledge and Justification
New York: Oxford University Press
Fogelin, R. J. [1997] 'What Does a Pyrrhonist Know?', Philosophy and
Phenomenological Research vol.57 (no.2): 417-425
Frede, M. [1987] Essays in Ancient Philosophy Oxford: Oxford University Press
Gaukroger, S. [1995] 'The Ten Modes of Aenesidemus and the Myth of Ancient
Scepticism', British Journal of the History of Philosophy vol.3 (no.2):
371-387
Glidden, D. [1983] 'Skeptic Semiotics', Phronesis vol.28: 213-255
Gottlieb, P. [1992] 'Escaping the Sceptic's Net', Ancient Philosophy vol. XII:
483-487

136

Grayling, A. C. [1985] The Refutation of Scepticism London: Duckworth

Groarke, L. [1990] Greek Scepticism: Anti-Realist Trends in Ancient Thought
Montreal: McGill-Queen's University Press
Hakinson, J. R. [1995] The Sceptics New York: Routledge
Husserl, E. [1973] The Idea of Phenomenology (translated by W. P. Alston and
G. Nakhnikian) Hague: Nijhoff
Kenyon, J. D. and Craig, E. [1985] 'Doubts about the concept of reason',
Aristotelian Society 1985 supp.59: 249-283
Long, A. A. and Sedley, D. [1987] The Hellenistic Philosophers, 2 vols.
Cambridge: Cambridge University Press
Mates, B. [1984] 'On Refuting the Skeptic', American Philosophical
Association vol.58: 21-35
Mates, B. (ed. & translated) [1996] The Skeptic Way: Sextus Empiricus's
"Outlines of Pyrrhonism " New York: Oxford University Press
Maudlin, T. [1986] 'The Structure of Skepticism', Ancient Philosophy vol.VI:
177-193
McEvilley, T. [1982] 'Pyrrhonism and Madhyamika', Philosophy East and
West 32 (no. 1): 3-35
McGinn, M. [1997] Wittgenstein and the Philosophical Investigations London:
Routledge
McPherran, M. L. [1990] 'Pyrrhonism's Arguments against Value',
Philosophical Studies 1990:127-42
Moore, G. E. [1993] Selected Writings (ed. Thomas Baldwin) London:
Routledge
Moser, P. K. [1997] "The Relativity of Skepticism', Philosophy and
Phenomenological Research vol.57 (no.2): 401-406
Nagel, T. [1986] The View From Nowhere New York: Oxford University Press.
Needham, R. [1972] Belief Language, and Experience Oxford: Blackwell
Oakley, I. T. [1976] 'An Argument for Scepticism concerning Justified Beliefs',
American Philosophical Quarterly vol.13 (no.3): 221-228

137

Owen, G. E. L. [1986] Logic, Science and Dialectic: Collected papers in Greek

Philosophy (ed. Martha Nussbaum) New York: Cornell University Press
Pappas. G. and Swain, M. (ed.) [1978] Essays on Knowledge and Justification
Princeton: Princeton University Press
Plato [1973] Theaetetus (translated with notes by J. McDowell) London: Oxford
University Press
Plato [1992] Theaetetus (ed. with an introduction by B. Williams, translated by
M. J. Levett, revised by Burnyeat) Cambridge: Hackett Publishing
Company
Russell, B. [1997] The Problems ofPhilosophy (with a new introduction by John
Perry) New York: Oxford University Press
Schofield, M., M. F. Burnyeat and Jonathan Barnes (edd.) [1980] Doubt and
Dogmatism Oxford: Claredon Press
Sedley, D. [1980] "The Protagonists', in Schofield [1980]
Sedley, D. [1983] "The Motivation of Greek Skepticism', in Burnyeat [1983]
Sextus Empiricus [1933-49] Sextus Empiricus (Greek text and translated R. G.
Bury) Loeb Classical Library, 4 volumes, Cambridge: Harvard
University Press
Sextus Empiricus [1994] Outlines of Scepticism (translated by Annas, J. and
Barnes, J.) Cambridge: Cambridge University Press
Sosa, E. [1988] 'Beyond Scepticism, to the Best of Our Knowledge', Mind 97:
153-188
Sosa, E. [1994] 'Philosophical Scepticism and Epistemic Circularity',
Aristotelian Society 1994 supp.68: 263-290
Sosa, E. (ed.) [1994] Knowledge and Justification Aldershot: Darthmouth
Publishing Company Limited
Sosa, E. [1997] 'How to Resolve the Pyrrhonian Problematic: a Lesson from
Descartes', Philosophical Studies, LXXXV 1997, pp.229-49
Stough, C. L. [1969] Greek Skepticism Berkeley: University of California Press

138

Strawson, P. F. [1985] Skepticism and Naturalism: Some Varieties New York:

ColumbiaUniversity Press
Strawson, P. F. [1992] Analysis and Metaphysics: An Introduction to
Philosophy Oxford: Oxford University Press
Striker, G. [1983] "The Ten Modes of Aenesidemus', in Bumyeat [1983]
Striker, G. [1990] "The problem of the criterion', in Everson [1990]
Stroud, B. [1984] The Significance of Philosophical Scepticism Oxford:
Clarendon Press
Stroud, B. [1997] 'Unpurged Pyrrhonism', Philosophy and Phenomenological
Research vol.57 (no.2): 411-416
Stroud, B. [1994] 'Scepticism, 'Extemalism', and the goal of Epistemology',
Aristotelian Society 1994 supp.68: 291-307
Taylor, C. C. W., Hare, R. M. & Barnes, J. [1999] Greek Philosophers Socrates,
Plato, Aristotle Oxford: Oxford University Press
Unger, P. [1975] Ignorance: A Case for Scepticism Oxford: Clarendon Press
Unger, P. [1984] Philosophical Relativity Oxford: Blackwell
Wedgwood, R. [1990] 'Scepticism and Rational Belief, Philosophical
Quarterly vol. 40 (no.158): 44-64
Weintraub, R. [1997] The Sceptical Challenge London: Routledge
Wittgenstein, L. [1953] Philosophical Investigations (ed. G. E. M. Anscombe
and R. Rhees, translated by G. E. M. Anscombe) Oxford: Blackwell
Wittgenstein, L. [1967] Zettel (translated G. E. M. Anscombe) Oxford:
Blackwell
Wittgenstein, L. [1969] On Certainty (ed. G. E. M. Anscombe and G. H. von
Wright, translated by Denis Paul and G. E. M. Anscombe) Oxford:
Blackwell

139