ResearchPDF Available

Challenging misunderstandings of scepticism



A paper giving a philosophical case, based on the implications of sceptical arguments themselves, that scepticism has been widely misunderstood and misinterpreted by philosophers and others.
Challenging misunderstandings of scepticism
Abstract: It is argued that (global, philosophical) scepticism has been widely
misinterpreted as a threat to justification in general, when it offers a threat only to the
justification of absolute claims. This position is argued only on the basis of the
philosophical nature of sceptical arguments themselves. Attempts to refute sceptical
argument have generally misinterpreted it by conflating the mere recognition of
uncertainty (a position requiring a psychological description) with negative assertion.
Sceptical arguments cannot involve negative claims, nor can they be treated
selectively, if they are to be compatible with the grounds of doubt they offer. Yet at
the same time, incremental claims are compatible with sceptical doubt. Fully global
scepticism thus cannot be ‘extreme’, but on the contrary has a beneficial and
moderating effect on belief.
In this paper I want to question a long tradition of the misinterpretation of scepticism
in Western philosophy. Based only on the implications of sceptical arguments
themselves, rather than on historical evidence or the unnecessary assumptions
made when scepticism is interpreted as a threat, I want to put the case for (full-
blooded, global, philosophical) scepticism as a benign and much misunderstood
movement. My main contentions will be as follows:
Sceptical arguments merely point out uncertainty. They are a threat only to
absolute beliefs, not to incremental ones that are compatible with uncertainty.
Sceptical arguments do not involve negative assertion of the kind that has been
mistakenly (and unnecessarily) challenged by Descartes, Hume, Wittgenstein,
and many others. Rather sceptical arguments must, by implication, undermine
absolute negative assertions as much as positive ones.
There can be no such thing as ‘extreme scepticism’. Global scepticism can be
applied consistently or inconsistently. It is selective scepticism that is unjustified
due to its failure to apply scepticism even-handedly. So-called mitigated’ or ‘local’
scepticism is merely selective.
My approach will be first to summarise a range of sceptical arguments, then to
survey the failure to successfully refute scepticism. Finally I will put forward the basis
of the above conclusions, drawing on this material.
Summary of sceptical arguments
The following arguments come from a variety of sources from Pyrrho onwards.
However, their sources are not significant to my argument only the fact that such
arguments can be made.
1. The ten modes of Pyrrhonism (first given by Aenesidemus: Sextus Empiricus
1996) give a range of reasons why our senses do not necessarily give us correct
information about objects. These are
a. that different animals have different sense organs, so therefore animals
perceive objects differently from humans
b. that different humans have different sense abilities (e.g. some have visual
impairments) and thus perceive objects differently
c. that different senses perceive objects differently (e.g. I may be able to hear
something I cannot see)
d. that differences in circumstances lead to different perceptions (e.g. a hand
put in hot water and then cold will find the cold colder through contrast)
e. that differences in spatial position relative to an observed object (e.g. a distant
landmark) lead to different perceptions and to perceptions that may be
f. that our perceptions of an object will be altered by what we see it with or near,
which may lead us to see it differently (e.g. camouflage)
g. that the same object will vary in the way it is perceived when in different
quantities or when composed differently, making it impossible to identify the
object with certainty (e.g. wheat grains look different from flour, but are
composed of the same substance)
h. that if objects are claimed to be absolutely existent this claim is still only
understood relative to other claims
i. that the constancy or rarity with which something appears changes our
perception of it (e.g. comets are rarer, and thus seeing one is more significant
to us, than stars)
j. that moral claims also differ between people (one person’s good child is
another’s bad).
In general, then, these arguments point out that all our perceptions are relative,
because influenced both by the specific circumstances of our perception and of
the object we are (or may be) perceiving. This means that any perception may be
in error.
2. The dream argument considered by Descartes (1641/1912) and others,
suggests that we cannot tell with certainty that we are not dreaming (or that our
whole experience is not otherwise illusory) at a given moment, and therefore that
our perceptions are not completely erroneous. This argument is problematic if
applied to all our experience through time, as it then deprives us of any contrast
between ‘dream’ and ‘reality’, but we could consistently maintain this distinction to
assert that at least some of our past experience must not have been a dream,
and yet not be certain that our current experience is not. The ‘brain in a vat’ is
another version of this argument.
3. The error argument points out that even if our whole experience at a given time
is not erroneous, particular objects that we think we perceive may still be so. Past
mistakes in perception show that mistakes are possible, and we were not aware
of those mistakes at the time we made them, so we may not be aware of our
current mistakes. This argument can be applied to current perceptions and also
to memory, to point out that with a past perception we may have made a mistake
in the original perception or in our memory of that perception.
4. The time lapse argument used by Bertrand Russell (1940, 13) suggests that we
cannot be certain of the object of perception because the conditions of that object
may have changed by the time we receive the perception (e.g. the sun may have
ceased to exist 7 minutes ago, but due to the distance from the sun and the time
it takes light to traverse that distance, we wouldn’t know about it yet).
5. The relativity of cultural background. Earlier sceptical arguments
acknowledged all the physical reasons for the relativity of perception, but more
recent psychological and linguistic research tells us more about the mental
reasons. Our cultural background may lead us to perceive objects differently: for
example, perceivers of the Müller-Lyer Illusion make a bigger misjudgement
about the relative lengths of the lines if they are accustomed to environments with
rectilinear architecture (Segall et al 1963, 769-771).
6. Problem of Induction. All generalisations based on specific observations lack
certainty, because the observations do not provide enough evidence to cover the
possibly infinite number of instances referred to in the generalisation. For
example, if I claim that all physical objects have mass and are subject to gravity, I
have not checked all the physical objects in the universe to ensure this.
7. The infinite regress of justification. There is no possible claim for which one
could not ask for further justification (i.e. there are no self-evident claims).
However, if a justification is offered, one could then ask for a justification of the
justification, and so on ad infinitum. This argument works not only for empirical
claims but also for a priori ones.
8. The relativity of linguistic categories. Even if we were able to overcome the
above sceptical arguments in other respects, the language out of which we
represent claims about objects in the universe does not have an absolute
relationship with the objects themselves, either as they may exist in themselves
or even as we experience them. We cannot be certain either that another person
understands the same as we do by a particular proposition about the world, or
even that we mean the same ourselves when we return to our previous
utterances after an interval of time. Even if we were to weaken the requirement to
one of identical representation of our experiences to ourselves after a few
seconds, we cannot be sure that our mental representation of that experience
has not changed, and thus that the language does not mean something different
from what it meant to us beforehand. Claims of certainty depend on the absolute
consistency of language used to represent those claims, otherwise any certainty
that might apply to a statement at one instant will immediately be lost at the next
instant, even for the person who made the statement.
9. The vagueness of linguistic categories. Any possible representational term out
of which a claim of certainty might be made is also inadequate for the
representation of any reality (or even any experience) because of its vague
relationship to that reality or experience. The terms used for representing objects
(even abstract ones) are nouns, and any given noun is vague in terms of the
scope of what it represents either in experience or the object of experience. For
example, if I use the word ‘pen’ to describe an object, and even if I give a precise
and unique description of that pen, giving measurements and physical co-
ordinates, what I am referring to is vague both in terms of space (some molecules
or even smaller particles may not be clearly defined as part of the pen or not) and
time (any interval of time I may specify for my statement about the pen will have
duration, and during that duration the pen may change). If, on the other hand, I
make no claims for the object which take up any space or time, my claims will be
uninformative. It might be claimed that a priori claims such as those about
numbers avoid this vagueness, but when applied to any claim about the universe
these numbers depend on counting and measurement, which are unavoidably
Numbers 1-5 apply to any empirical claim. Number 6 applies to any empirical
generalisation. Numbers 7-9, however, apply to any proposition whatsoever,
including a priori claims.
Together these sceptical arguments provide a huge over-determination of the
sceptical case. We do not need them all. Only one of them has to be successful (in
relation to the categories of claim it applies to) to prove that there can be no certainty
attached to any claim. These arguments seem to me irrefutable. However, to
establish this more fully I will now survey some of the unsuccessful attempts at
The failure of philosophical arguments against scepticism
Philosophical arguments against scepticism have come in broadly five types, as far
as I can identify:
1) The assertion of self-evident truths (e.g. Descartes)
2) Arguments that scepticism is practically unsustainable, and thus that dogmatism
is unavoidable (e.g. Hume)
3) Arguments that scepticism involves practical inconsistencies or a ‘paradox of
scepticism’ (e.g. Burnyeat, Nussbaum)
4) Arguments that scepticism is unjustified because it only offers negative grounds
of judgement (e.g. Moore and other positivists)
5) Arguments that scepticism makes invalid semantic assumptions (e.g.
I shall argue here that each of these lines of criticism itself involves assumptions that
we do not necessarily need to make in approaching the subject.
1. The assertion of self-evident truths
If self-evident truths exist then this would obviously undermine scepticism, as there
would be a foundational certainty from which other certainties might then be
deduced. Descartes’ cogito, in which the certainty of the thinker’s existence is
deduced from the experience of a thought, is the classic example of a self-evident
truth (Descartes 1912).
Let us accept for the sake of argument that there might be self-evident truths such as
that I, a thinker, exist at this instant. Since it is not empirical, this claim avoids the
first six sceptical arguments listed above, and it avoids the seventh, the infinite
regress, if its foundational claims are justified. However, this claim and any other
foundational claim are still subject to the last two sceptical arguments that point out
the relativity and vagueness of linguistic categories. “I, a thinker, exist at this
moment” is relative to each thinker because it can only be interpreted according to
the linguistic understanding of each individual thinker. If you tell me that you exist at
this moment, to me that obviously means that you exist at this moment, which means
something rather different from me existing at this moment. Unless this statement
has an absolute unchanging meaning for all who may comprehend it which it
clearly does not it can hardly have an absolute unchanging justification. The same
point would apply to mathematical or logical claims.
The ambiguity of statements supposedly offering self-evident truths also creates
contradictions in the very claims involved. “I, a thinker, exist at this moment” for
example, either means that a thinker exists over a short period of time, or at a
genuine instant of time with no duration. If the former, the thinker can have thoughts
(which always take up a certain amount of time), but by the time the thinker gets to
the end of her thoughts, she may be different from when she started them and thus
no longer “exist” in the absolute, unchanging sense required. On the other hand,
within an instant without any duration, no thoughts can take place and thus it seems
that a thinker cannot exist.
2. Arguments that scepticism is practically unsustainable
Hume’s argument about scepticism, on the other hand, attempts to adopt a no-
nonsense practical approach to it. After admitting that we cannot refute scepticism
on its own terms, Hume seems to be saying that there is no way that we can, in
practice, accept those terms. It is ‘nature’, he says, that drives us to belief, rather
than reason, because when we engage with objects in the world around us we do so
on the basis of a practical assumption of their existence and form. Scepticism is all
very well in the abstraction of a study, but there is no way we can keep it up in
ordinary life:
'I dine, I play a game of backgammon, I converse, and am merry with my friends;
and when after three or four hours' amusement, I would return to these speculations,
they appear so cold, and strained, and ridiculous, that I cannot find in my heart to
enter into them any farther.' (Hume 1978, 279)
Hume makes an unjustified assumption about the implications of scepticism here:
indeed, he gets the whole matter the wrong way round. Scepticism casts doubt on
any claims to certainty, but this does not imply that to take it seriously means that we
must be constantly straining to disbelieve what we encounter in everyday
experience. On the contrary, our everyday experience involves uncertainty, and
scepticism, far from relying on ‘cold’ and ‘strained’ calculations, uses this everyday
experience as its point of departure. It is claims of certainty, and the attempt to justify
them, that go far beyond everyday experience and become cold and strained.
This point is closely related to another: namely the distinction between denial of
claims and denial of certainty about them. If we were to assert the opposite of
everything we take for granted in everyday life, e.g. that there is not a table in front of
me, that the world does not exist etc, then this would indeed be a cold and strained
exercise. However, there is no reason why we should have to interpret sceptical
arguments in this way. Scepticism denies certainty, and thus leaves us in a position
lacking certainty, rather than asserting the opposite of our accepted beliefs. To
assert the opposite would be at least as uncertain an enterprise. Hume, however,
(along with many of his successors) seems to confuse these two positions. There will
be further discussion of this point below.
3. The supposed paradox of scepticism
Burnyeat and Nussbaum, on the other hand, respectively accuse scepticism of other
kinds of practical inconsistency. Burnyeat claims that it’s impossible to maintain the
degree of detachment from one’s views that scepticism demands (Burnyeat 1980).
Nussbaum argues that the classical sceptics are dogmatic about the value of
ataraxia, which in classical Pyrrhonian scepticism is the relaxed state of detachment
from opposing certainties (Nussbaum 1994). Both these objections could be seen as
versions of what is sometimes called ‘the paradox of scepticism’: namely, that
sceptics are certain about uncertainty. This supposed paradox can be presented
either as a direct contradiction or at least as a practical inconsistency.
Both of these thinkers are commentators on the classical sceptics and make these
remarks in the context of discussing classical Pyrrhonism. I am purposely avoiding
too much discussion of the scholarly issues about historical schools of philosophy
here, but am attempting only to isolate what we do or do not need to think about
sceptical claims considered in their own right. Burnyeat’s and Nussbaum’s
arguments may or may not be true of classical Pyrrhonism, but my argument is that
their criticisms distract us from the useful insights that can be offered by sceptical
arguments in general, regardless of their temporal context.
Like Hume, both of these objections fail to take sufficiently into account the
distinction between the denial of claims and the denial of certainty about claims. If
we assert the opposite of a given claim, we raise the same issues of certainty about
it as with the original claim. If we merely deny the certainty surrounding a claim,
however, we modify the way in which that claim may be held rather than setting up a
new claim. If we understand the modification of the way we hold a claim in
psychological terms, rather than merely in terms of opposing propositions, this
becomes clearer. A claim not held with certainty is held in a more provisional and a
more relaxed fashion, which may also affect our subsequent judgements about how
to assess it. Contrary to Nussbaum’s assumptions, the value of such a provisional
state is not one that we have to accept absolutely and all at once in the form of an
idealised state of ataraxia, but can be a matter of incremental recognition.
Burnyeat overestimates the degree of detachment required to take scepticism
seriously, because he shares the confusion between denial and provisionality with
many modern commentators. We do not need the amount of detachment that would
be required to seriously adopt a position of denying all our beliefs in order to merely
hold them provisionally. Nor does it require a certain fixed amount of detachment in
order even to hold them provisionally. If we think about provisionality in an
incremental rather than an absolute way then we can think of the process of giving
up attachment to certainty as a gradual and dynamic one. This process then
becomes practically achievable in a way that a sudden demand for massive
detachment would not.
The supposed ‘paradox of scepticism’ thus involves an absolutisation of the sceptical
position that scepticism itself challenges. It is a straw man perhaps not in relation
to historical scepticism, but in relation to the necessary implications of sceptical
argument. Given the way in which scepticism appeals only to practical judgements
and challenges all idealisation and abstraction, for scepticism itself to be accused of
impracticality involves a reversal that can only be described as deeply ironic.
4. Rejection of negative grounds of judgement
Another, positivist type of response to scepticism is to assert that only positive
justifications for belief are acceptable, and that negative doubts about a claim not
accompanied by definite evidence against the claim are inadmissible (e.g. Ayer
1956). The logical positivists and their allies in the early twentieth century saw this as
a way of protecting evidence-based scientific investigation against the
encroachments of metaphysics. Those who deny commonly accepted empirical
beliefs, after all, often do so only on the basis of speculation.
Much as I sympathise with the logical positivist attempt to distinguish metaphysics
from claims that can be justified through experience (which I will discuss further
below), the positivist route does not succeed in doing this. Positivism prevents us
from taking negative doubts seriously, and simultaneously makes its own
metaphysical assumptions unassailable. We need negative doubts in order to be
able to consider conventionally accepted beliefs from an adequately critical
perspective. The positivist dismissal of negative doubt leaves us dependent upon
conventional beliefs and unable to break out of the set of assumptions that are
currently accepted in our context. Logical positivism, and its successors in analytic
philosophy, remain dependent on analysis of conventional positions or commonly
shared intuitions, and unable to reach a justified critical standpoint beyond those
conventional positions. It leaves us defenceless against confirmation bias.
Like many of the previous criticisms, too, positivism confuses denial with the mere
acceptance of uncertainty. Negative doubts require us to accept the possibility of
currently accepted beliefs being wrong, not to accept the alternative claim that they
are definitely wrong. Speculative metaphysics puts forward new claims that are
beyond experience so sceptical argument is a crucial tool that should be used for
combating metaphysics, not discarded at the very point when it would be most
A more specific version of this positivist argument is that used by both Moore (1962)
and Wittgenstein (1969) in slightly differing ways to assert the existence of their
hands as a basic certainty. Their reason for dismissing scepticism about something
as certain as the existence of one of their hands was not just that mere negative
doubts were inadmissible, but that any evidence that could be used to support the
assertion that their hands existed would be less certain than the existence of their
hands. For positivists, then, some kinds of claim have to be taken as certain and
basic to all other discussions. Without those basic assumptions, it is argued, the
discussion could not take place.
One of the ironic aspects of this argument is that, by using comparative certainties, it
implicitly supports the sceptical rejection of absolute certainties. The ordinary
language sense of ‘certainty’ is not the target of sceptical argument (though I would
also suggest that there are other, pragmatic reasons for not over-using the ordinary
language sense of ‘certainty’ in philosophical discussion, as it denies us useful
language with which to identify and recognise philosophical uncertainty). In many
ways, then, the Moore/ Wittgenstein approach to scepticism is a straw man. The
sceptic does not deny that the existence of one’s hand is ‘certain’ in an ordinary-
language sense, only in an absolute sense.
Their approach can in any case only be directed against the error argument, and
other sceptical arguments that raise one specific doubt whilst taking a wider context
for granted. It does not apply to the dream argument, or the infinite regression
argument, or the linguistic arguments. The dream argument does not require us to
take one kind of fact for granted in order to cast doubt on others, only that there be
some unspecified factual basis to use as a ground of contrast with a current
uncertainty. Similarly, the infinite regression argument can be used against any claim
of certainty whatsoever, regardless of its relationship to other claims, and the
relativity and ambiguity of the linguistic composition of these claims remains
regardless of its relationship to other claims. Given that only one sceptical argument
needs to stand for sceptical uncertainty as a whole to stand, Moore and Wittgenstein
have failed to make any impression on it.
5. The claim of invalid semantic assumptions
Finally, Wittgenstein’s objections to sceptical argument were also based on the
alleged linguistic privacy of sceptical argument, and his objections to linguistic
privacy in the so-called private language argument (Wittgenstein 1969 §258).
However, he was mistaken on both counts. Not only is scepticism not necessarily
based on linguistic privacy (assuming that we can even make sense of the idea of
linguistic privacy), but there is also no reason to assume that language developed in
linguistic privacy is meaningless.
The Cartesian version of scepticism, in which I can entertain the possibility of being
the only real thing in the universe, does not depend on solipsistic assertions but only
on the possibility of solipsism (the same confusion we have already noted).
However, all the other types of sceptical argument mentioned above, including the
modes of Pyrrhonism and the linguistic arguments, could just as well be applied to a
publicly shared context as to a ‘private’ one. I might be wrong about my perceptions,
but we might also be wrong about our shared perceptions, for very similar reasons.
The publicity or otherwise of the language makes no substantial difference to these
kinds of arguments.
The concept of linguistic privacy, completely and absolutely distinguished from
linguistic publicity, seems dubious in the first place. We use language to
communicate with others, but we also use it to communicate with ourselves over
time (as in Wittgenstein’s example of a private diary) and perhaps even to articulate
without communicating (as when we talk to ourselves to clarify our thoughts).
Wittgenstein unnecessarily assumes that the only acceptable function of language is
communication. He then asserts that when using purely private language (i.e. a
symbol whose significance is known only to me) in a private diary, when using it later
I would have no clear criterion of meaning. However, I would have a relative criterion
of meaning based on my memory of previous experience which the symbol
represented. A falsely absolute distinction is made if it is assumed that the private
criterion is relative whilst a public one is absolute, for there is no guarantee that a
publicly used piece of language, even within a particular language game (i.e. social
context where that language is shared) is not equally ambiguous.
Like the other attacks on scepticism, then, Wittgenstein’s do not apply to all the
arguments, and also confuse lack of certainty with definite denial. Like the other
attacks, it is based on a confusion about the purpose and implications of scepticism.
A complete reversal of the assumptions behind all these attacks on scepticism is
required. Scepticism is not a dragon to be slain, but rather a knight in shining
armour. It is certainty that is the dragon.
Scepticism, knowledge and justification
Sceptical arguments are thus irrefutable. Let us proceed on that basis. If I have
omitted to discuss any objections to one or more of the sceptical arguments above,
my question will be whether the objection will apply adequately to, not one of them,
but all of them: an extremely unlikely eventuality.
The implication of scepticism’s irrefutability is often taken to be the undermining of all
knowledge. If ‘knowledge’ is understood in absolute rather than ordinary language
terms, this will indeed be the case. If we take the traditional definition of knowledge
as justified true belief, then sceptical argument undermines both the first and the
second terms. If justification is understood absolutely rather than incrementally, then
sceptical argument proves that there is no absolute justification. As far as the truth
condition goes, it may or may not be the case that a particular claim is true or false,
but sceptical argument establishes that we are in no position to ever establish
whether it is. The truth condition is thus irrelevant to our concerns, as even if there
were theoretically such a thing as knowledge independent of our awareness that we
have it (as speculated by externalist theories of knowledge), we would never be able
to establish that we were in possession of any particular example of it that fulfilled
the truth criterion. Again, then, the argument that we do not have absolute
knowledge is over-determined.
This need not prevent us from using ‘knowledge’ in ordinary language senses,
provided we do not make the mistake (common in analytic epistemology) of
assuming that the analysis of our intuitions about the ordinary language sense of
knowledge is in some sense philosophically informative. Such analysis merely
clarifies what we mean, but is irrelevant to the justification of claims, which are not
made more or less likely to be justified by whether or not we categorise them as
The rejection of philosophical knowledge in the wake of scepticism, however, does
not necessarily imply the rejection of justification, provided justification is understood
incrementally. I want to argue that a claim can be more or less justified in proportion
to the evidence or other reasons for believing it, regardless of whether absolute
claims can be justified. If incremental justification is possible, then some claims are
more justified than others, and relativism (understood as the view that no given claim
is better justified than another claim) is not the implication of scepticism.
All that stands in the way of a recognition that incremental justification is possible is
an assumption that incremental justification is in some sense logically dependent on
absolute justification. Such a claim, though widespread since Plato’s “You can’t use
the imperfect as a measure of anything” (Plato 1987 504c), seems to depend on a
fallacy of composition. What is the case for a part need not be the case for the
whole. Given that any supposed whole is also part of something else, in any case,
and that the continuing attribution of parts to wholes leads indefinitely outwards, the
wholes on which we are supposed to be logically dependent will have to be infinite.
Are we seriously going to claim that finite concepts are dependent on infinite ones,
rather than the converse, when our concept of infinity is clearly only the absence of
Epistemically, too, sceptical arguments show the inadmissibility of claims about
wholes: the ten modes of Pyrrhonism establish that we will only ever be able to
experience parts, and the problem of induction that any assumptions about wholes
on the basis of parts will be uncertain. Final wholes, then, are at best irrelevant
abstractions that, far from being supportive of incremental justification, potentially
stand in its way by providing fuel for dogma.
The implication of sceptical argument is thus not a complete failure of justification, as
seems to often be assumed, but merely the failure of whole-to-part top-down forms
of justification. It implies that instead of talking about ‘knowledge’ or ‘justified true
belief’ we should shift our understanding of epistemic goals to that of incrementally
justified belief. There is thus no such thing as ‘scientific knowledge’, but on the other
hand there can be a much more robust understanding of ‘justified scientific belief’
that accepts scepticism rather than fruitlessly resisting it. All we have to do is let go
of the truth condition, and the various hypothetical appeals to it, that have often
taken the place of incremental justification in analytic epistemology.
Negative and provisional assertions
A further common misunderstanding of scepticism has been the assumption that
scepticism involves negative assertions. We have already seen a range of Western
philosophers attempting to refute such negative assertions or question the
assumptions involved in making them. It would be possible to argue on historical
grounds that this involves a misinterpretation of the works of Sextus Empiricus and
other ancient sceptical sources that has been common since the rediscovery and
translation of Sextus Empiricus by Henri Etienne (Popkin 1992), but a historical
argument would be insufficiently conclusive, and no doubt subject to merely
historical objections. Another possible approach is to distinguish scepticism from
Pyrrhonism, taking ‘scepticism’ to make negative assertions whilst ‘Pyrrhonism’
instead involves the suspension of judgement (Kuzminski 2008). However, this
approach yields the ground of the very implications of scepticism in a way that can
obscure the central point that none of the sceptical arguments justify negative
assertion. If this is the case, a so-called sceptic who does attempt to offer negative
assertions is simply offering an ineffectual argument, whereas a sceptic who sticks
carefully to the recognition of uncertainty about both positive and negative assertions
offers, on the contrary, a well justified position. Western philosophers who have
attacked scepticism on the assumption that it makes negative assertions, then, have
been attacking a straw man.
For example, in the case of the Ten Modes of Pyrrhonism, each argument points out
that our senses do not necessarily give us correct information. It thus asserts that
any possible sense-data (whether we take sense-data as ‘raw’ or interpreted) cannot
justify any given claim y. If we turn this round and substitute a negation of y, it will be
evident that this is also not justified by any possible sense-data. Similarly, with the
dream argument, the recognition that any claims I make about my current
experience may be mistaken does not entail that such claims must be mistaken.
Central to the problem is the interpretation of what kind of claims scepticism is
making. If sceptics are taken to be denying any given proposition along the lines of
‘the book is on the table’, then the application of the excluded middle implies that
they are asserting its negation, ‘the book is not on the table’. It is not necessary to
question the excluded middle to avoid this implication, but merely to clarify the
sceptic’s claim as being ‘We do not know whether there is a book on the table’. This
is a negative claim about our certainty, not about the book on the table. The negative
claim about us notes an absence of absolute justification, not the presence of
absolutely justified beliefs even about ourselves.
It is only by construing scepticism in terms of negations that the philosophers
discussed above have been able to see it as impractical. It may well be the case
that, when I see a book on the table, I cannot negate that belief, nor can I start to
believe ‘there is no book on the table’ in response to sceptical considerations. But
scepticism is not concerned with deleting our beliefs, only with recognising their lack
of absolute justification. The practical implication of scepticism is thus neither
negative belief nor a complete suspension of judgement, but rather provisional belief.
The concept of provisionality has been under-investigated by philosophers, even
though it seems essential to successful scientific practice. I think much can be said
about it, but here I will merely sketch some suggestions to show that there are
positive alternatives to the way the implications of scepticism have been traditionally
understood. My suggestion is that provisionality can be understood as a
psychological state in which alternatives to our current beliefs can be considered and
assessed. Such a psychological state is distinguished by optionality: that is, by the
availability of options beyond the current one, by awareness of those options (which
is obviously a matter of degree) and by the integration of beliefs, by which I mean
that alternative beliefs do not necessarily remain separate or conflicting, but have the
capacity to be re-formed by contact with each other. I have given further details of
such a psychological theory of provisionality elsewhere (identificatory reference to be
inserted), but to pursue it further here is beyond the scope of this paper.
Dogmatism and belief
An interpretation of scepticism that is compatible with provisional belief must also
reject other beliefs as ‘dogmatic’. The mistaken understanding that sceptical
argument implies the rejection of all beliefs as dogmatic (e.g. Hookway 1990, 3)
presumably arises from issues in the interpretation of classical scepticism. Yet if we
focus neither on historical interpretation nor on modern assumption, but on the
implications of the arguments themselves, it becomes evident that the boundary
should be placed at a different point from that between ‘belief’ and what the
Pyrrhonists called ‘acquiescence in appearances’ (Sextus Empiricus 1996, I:23-4)
at least if the latter is interpreted as somehow not involving beliefs of any kind.
Rather it is certainty of belief that is the target of sceptical argument (whether that
certainty is positive or negative), and it is claims that are accompanied by certainty
that are dogmatic.
Dogmatism, like provisionality, is very evidently a psychological state: not so much a
particular proposition or set of propositions as a contextual state of mind and brain
that fails to entertain alternatives, and maintains a rigid and unreflective commitment
to one type of belief. As with provisionality, I think the state of dogmatism can be
understood psychologically and neurally, and have explored this in more multi-
disciplinary work elsewhere (identificatory reference to be inserted).
However, there are also some features of dogmatic belief that can be identified
philosophically: these are that dogmatic beliefs are absolute, metaphysical, and
understood in terms of an assumed representationalism. I want to argue that each of
these can be seen from the implications of sceptical argument itself.
Dogmatic beliefs are absolute in the sense that they lack incrementality or
incrementalisability. A belief that can be expressed in terms of qualities that form a
spectrum, rather than absolute quantities and absolute qualities, is incremental.
“Bambi does not exist” for example, is an absolute claim because Bambi could not
exist to some extent he either exists or does not. “The dog is black” is
incrementalisable, as it is at least possible to understand it in terms that are not
absolute, if we interpret the blackness of the dog as a matter of degree. Where there
are degrees of a quality there are also potential degrees of justification. The sceptical
arguments we have examined are justified wherever an absolute claim is made that
is only understood in terms of truth or falsity as a whole, but where claims are
incremental, those arguments are no longer a barrier to provisional assertion. For
example, in the case of the error argument, past error raises no problems when we
interpret it to indicate a probability of future error, but only becomes a problem when
it is assumed that future judgements can be made without error. In the case of the
infinite regression argument, the infinitely regressive demand for further justification
can only itself be justified if total justification is expected, and is no longer required if
the criterion of judgement that we apply is one of an incremental degree of
justification that has already been supplied before we demand further justification.
Dogmatic beliefs are metaphysical in the sense of lying beyond experience, not in
the sense of involving basic assumptions that frame that experience. This element
depends very much on contextual interpretation, and is not merely a product of the
particular proposition used. For example, ‘The book is on the table’ becomes a
metaphysical claim when interpreted noumenally, as asserting an ultimate
relationship between an intrinsically existent book and table, but not when
interpreted in a psychological state that would be open to alternatives if they were to
become relevant. Other classic metaphysical claims discussed in philosophy, such
as those of God’s existence or non-existence, of freewill or determinism, of realism
or idealism, or of ultimate value or ultimate relativity, more obviously involve dogma
in this sense, because they involve claims that are independent of any particular
experience and could be asserted regardless of such experiences (identificatory
reference to be inserted).
Sceptical arguments thus imply the avoidance of both positive and negative
metaphysical claims, because they cast doubt on any claims that cannot be stated in
terms of experience. For example, the dream argument casts doubt on any claim
that we either are or are not dreaming at this moment, as our experience would be
identical in either eventuality. This does not imply that we could not identify and
make claims about basic preconditions for experience. It is not necessarily dogmatic,
for example, to recognise that our experience is framed by an expectation of
substance or cause. This is a generalisation about our expectations that could be
challenged if we discover instances of it not occurring.
Nor should the avoidance of metaphysics be read in the terms of logical positivism
(as in Ayer 1936). No assertions are being made here about meaning, only about
justified belief, and metaphysical beliefs do not have to be asserted as meaningless
to be unjustified. Dogmatism does not have to be understood in the narrow terms of
verification or verifiability (which is continually subject to confirmation bias), but
instead can be understood in terms of experiential impact. If potential experiences
change ones understanding of a claim and its justification (not necessrarily its
verification), it is potentially provisional, without any need to introduce narrow and
formally scientific criteria.
We must also avoid the widespread assumption that sceptical arguments are
themselves metaphysical. They are metaphysical only in a critical sense, in that they
provisionally adopt the assumptions of a particular context in order to show them to
be absurd. Since neither positive nor negative metaphysical assertions are made by
sceptics, they are metaphysically agnostic rather than metaphysical in the dogmatic
sense given here. Agnostic beliefs are provisional ones, that may nevertheless be
decisive, and that reject absolute forms of belief, whether positive or negative. They
thus involve a Middle Way between positive and negative forms of absolutism
(identificatory reference to be inserted).
Thirdly, dogmatic beliefs involve an assumed representationalism. By that, I mean
that they assume that the meaning of the statements asserted depends on their
hypothetical truth or falsity. Such truth or falsity, if assumed in practice, requires that
the meaning of a claim must be considered in isolation from the possible
alternatives. On an embodied theory of meaning, on the other hand, the meaning of
a claim is understood in terms of a cognitive model in which it is embedded, which is
in turn understood as dependent on the metaphorical extensions of basic schemas
and categories, that have become meaningful to us in the course of active physical
Again, I do not have space here to discuss the full implications of this, or the
background in the debate between the representationalism about meaning still
widely assumed and the now well-developed alternative in the embodied meaning
thesis. I have discussed this elsewhere (identificatory reference to be inserted) and
refer readers to the work of Lakoff and Johnson (e.g. Lakoff 1987, Johnson 2007) for
details of the embodied meaning thesis and the evidence both for it and against
My suggestion is that the implications of sceptical argument can only be adequately
understood in a context of embodied meaning, because it is only in an embodied
model that provisional belief becomes possible. If the meanings of terms forming a
claim are already available to us prior to the conceptual formation of that claim, and
are understood in terms that offer flexibility, with other differing claims being
potentially assembled from the same or psychologically adjacent materials, then
provisionality and metaphysical agnosticism becomes possible and comprehensible.
Sceptical arguments can then be seen as devices to stimulate that provisionality and
challenge dogmatism. If we understand the meaning of claims only in terms of their
potential truth or falsity, however, the options are already restricted, and scepticism
will appear, not as a helpful tool to stimulate provisionality, but only as a bewildering
and paradoxical play on the edge of an abyss.
‘Global’ and ‘local’ scepticism
Traditional philosophical discussion of this area has distinguished between ‘global’
scepticism applied to all beliefs and ‘local’ or ‘mitigated’ scepticism applied to some
beliefs but not others: to the extent that this distinction is found in virtually every
philosophy textbook that discusses scepticism. I want to argue that this distinction is
entirely spurious. So-called ‘local’ scepticism is not worthy of the name, because
sceptical arguments depend on even-handedness. An argument does not become
sceptical merely by reason that it involves doubt, but rather by reason that it applies
doubt equally to all possible instances.
It may be objected that some of the sceptical arguments listed at the beginning of
this paper (i.e. numbers 1-6) only apply to empirical claims, and thus allow for the
possibility of scepticism that only applies to empirical claims but not a priori ones.
However, I have already argued that scepticism is over-determined by a range of
arguments, and given that arguments 7-9 all apply just as much to a priori claims,
there is no possible justification for limiting sceptical argument to empirical claims.
Other arguments for ‘local’ scepticism tend to assume that global scepticism
undermines all belief, and that some beliefs are necessary for practical operation.
Hume’s argument (as already discussed above) is of this kind when he claims that
nature drives us to believe in commonsense realities. As I have argued above,
scepticism does not imply the abandonment of all belief, but only the abandonment
of absolute justification and claims to truth. Since it is quite possible to maintain
provisional beliefs whilst accepting sceptical argument, there is no reason why we
should be selective in our use of sceptical arguments.
The partial or selective use of sceptical argument is also incompatible with the
implications of that argument, in the sense that any continuing acceptance of beliefs
that could be undermined by scepticism will be inconsistent with the use of
scepticism. For example, a rationalist who uses sceptical arguments against
empirical claims without also applying sceptical argument to a priori claims will be
guilty of inconsistency and subject to further sceptical criticism. Similarly, an atheist
who uses sceptical arguments against the existence of God, but fails to consider
sceptical arguments against the claim that God does not exist, will have no grounds
to claim the title ‘sceptic’. The appropriation of the term ‘sceptic’ by positive atheists
is thus unjustified.
For similar reasons, there can be no such thing as an ‘extreme’ sceptic. Scepticism,
in its all-encompassing even-handedness, is unavoidably moderate, because its
implication when doubting one position is not to adopt the opposite. Rather, sceptics
will avoid both absolute positive claims and absolute negative claims. On the
contrary, it will be those who use scepticism selectively, and reject one position
whilst failing to avoid the opposite, that will be guilty of extremism.
Of course, the popular use of ‘sceptic’ (to mean someone who merely rejects a
position of any kind and finds argument against it) raises ordinary language issues
here as it does in relation to ‘knowledge’ and ‘certainty’. Just as arguments against
certainty or knowledge in the full philosophical sense are not arguments against their
possibility in an ordinary language sense, arguments in favour of scepticism in the
full philosophical sense are not arguments in favour of its ordinary language sense.
On the contrary, sceptical argument in the full philosophical sense is very likely to
undermine all popular and partial forms of ‘scepticism’. In all these cases, the
philosophical sense is not impractical, but on the contrary of more practical
helpfulness than partial and dogmatic uses, because it requires us to subject all
assumptions to even-handed critical scrutiny, and thus aids us in detecting error
when more partial approaches would fail to do so.
My argument, then, is here based only on the implications of scepticism itself, even
though psychological and other scientific evidence could potentially also be brought
in to support this view of scepticism. Sceptical argument has been vastly
misinterpreted in the Western tradition, in the sense that it has been assumed to be
a threat of some kind. But it is a threat only to dogma. Since none of us is immune
from dogma, that makes scepticism not a threat, but a beneficial challenge to
stimulate re-examination of beliefs that have become rigidified. The practical value of
scepticism, entirely contrary to what has often been assumed, lies in its global
critique of all possible beliefs, and its ability to help us differentiate provisional beliefs
(which are sceptic-proof) from dogmatic ones that are not. The more globally and
even-handedly scepticism is used, the more beneficial it will be, and the more likely it
is that confusion with negative assertion can be avoided.
Ayer, A.J. (1936) Language, Truth and Logic: Penguin. London
Ayer, A.J. (1956) The Problem of Knowledge: Macmillan, London
Burnyeat, M.F. (1980) ‘Can the sceptic live his scepticism?’ from Doubt and
Dogmatism ed. Malcolm Schofield, Myles Burnyeat & Jonathan Barnes: Oxford
University Press, Oxford
Descartes, Rene, trans. John Veitch (1641/1912), ‘Meditations’ from A Discourse on
Method, Meditations and Principles: J.M.Dent, London
Hookway, Christopher (1990) Scepticism: Routledge, London
Hume, David (1978) A Treatise of Human Nature: Oxford University Press, Oxford
Johnson, Mark (2007) The Meaning of the Body: University of Chicago Press,
Kuzminski, Adrian (2008) Pyrrhonism: How the Ancient Greeks Reinvented
Buddhism: Lexington Books
Lakoff, George (1987) Women, Fire and Dangerous Things (University of Chicago
Press, Chicago)
Moore. G.E, (1962) ‘Proof of an External World’ from Philosophical Papers: Collier
Books, New York
Nussbaum, Martha (1994) The Therapy of Desire: Princeton University Press,
Plato, trans. Desmond Lee (1987) The Republic: Penguin, London
Popkin, Richard (1992) The Third Force in Seventeenth Century Thought: E.J. Brill,
Russell, Bertrand (1940) An Inquiry into Meaning and Truth: Allan & Unwin, London
Segall, Marshall H., Campbell, Donald T. & Herskovits, Melville J. (1963) Cultural
Differences in the Perception of Geometric Illusions: Science, New Series, 139: 3556
Sextus Empiricus, trans. Benson Mates (1996) The Skeptic Way: Sextus Empiricus’s
Outlines of Pyrrhonism: Oxford University Press, New York
ResearchGate has not been able to resolve any citations for this publication.
Data from 15 societies are presented showing substantial intersocietal differences of two types in susceptibility to geometric optical illusions. The pattern of response differences suggests the existence of different habits of perceptual inference which relate to cultural and ecological factors in the visual environment.
Can the sceptic live his scepticism
  • M F Burnyeat
Burnyeat, M.F. (1980) 'Can the sceptic live his scepticism?' from Doubt and Dogmatism ed. Malcolm Schofield, Myles Burnyeat & Jonathan Barnes: Oxford University Press, Oxford
The Third Force in Seventeenth Century Thought An Inquiry into Meaning and Truth Cultural Differences in the Perception of Geometric Illusions: Science
  • Richard Popkin
Popkin, Richard (1992) The Third Force in Seventeenth Century Thought: E.J. Brill, Leiden Russell, Bertrand (1940) An Inquiry into Meaning and Truth: Allan & Unwin, London Segall, Marshall H., Campbell, Donald T. & Herskovits, Melville J. (1963) Cultural Differences in the Perception of Geometric Illusions: Science, New Series, 139: 3556
The Therapy of Desire The Republic: Penguin
  • Martha Nussbaum
Nussbaum, Martha (1994) The Therapy of Desire: Princeton University Press, Princeton Plato, trans. Desmond Lee (1987) The Republic: Penguin, London
The Skeptic Way: Sextus Empiricus's Outlines of Pyrrhonism
Sextus Empiricus, trans. Benson Mates (1996) The Skeptic Way: Sextus Empiricus's Outlines of Pyrrhonism: Oxford University Press, New York