Intuition

as
Conscious Experience

Ole Koksvik

Copyright Ole Koksvik, 2020

1
Til Pappa

2
 Acknowledgements

A friend once convinced me that we can never tell what knowledge will be crucial in the future, and that
the pure research of today is the applied research of tomorrow. Taking heart from this I now (on good days)
think of the pursuit of knowledge as rooting around in the dirt till you find a tiny pebble that takes your
fancy, spending an inordinate amount of time with that grimy lump under the microscope, cleaning and
polishing as best you can, before finally and reluctantly tossing it—still stubbornly smudgy in places, but now
with some gleams, too—onto a huge and formless pile, fervently hoping that it will make a difference, some
time, some how. This book is one of my pebbles, and I hope you find it shiny, but I did not polish it alone.

Many of the ideas in this book begun their life in my PhD dissertation, and the debts of gratitude I
acknowledged there still largely remain. Since then I have benefitted from generous and warm colleagues
and friends at the Australian National University, the University of Bergen, Monash University, and the
Australian Catholic University, as well as more widely in the philosophical community, for whom I remain
grateful.

Some material in §3.6 was previously published in The Philosophical Quarterly (‘Intuition and Conscious Rea-
soning’, Vol. 63, No. 253); some in §4.2 appeared in American Philosophical Quarterly (‘Phenomenal Con-
trast: A Critique’, Vol. 52, No 4), and some in §§3.5 and 4.3 appeared in Philosophy Compass (‘The phenom-
enology of intuition’, Vol. 12, No 1). Thanks to these journals for allowing me to reuse the material here.
Thanks also to Kai Tanter for valuable research assistance.

Although my academic debts are many, I have decided to not delve deeper on this issue here, because my
personal debts have come to seem so much more important to me.

I wrote and tried to finish this book during dark times indeed, in the world generally, and for me person-
ally. I fell flat on my face so many times, but was lucky that good people saw me and loved me and helped
me get back up: Carl Brusse, Stephanie Collins, John Cusbert, Keith Ferdinands, Øystein Fossøy, Alan
Hájek, Mark and Sue-anne Jozsa, Andreas Koksvik, Holly Lawford-Smith, John Pye, Richard Rowland, An-
drew Shortridge, Nicholas Southwood, Weng Hong Tang, Eustace and Lorraine Weinman.

My closest friend, Leon Leontyev. My amazing partner, Elizabeth Clarke.

My funny, kind, loving, forgiving, playful, adventurous and outright awesome children, Mia and Oscar, my
greatest loves.

Achingly far away, the family from which I came. Kristine, who tries to call. Kari, who visits me, challenges
me, and stands by me, and the wonders she brought into my life: Tormod, Ådne, Johanne, and Mari Olivia.

Mum, my greatest fan bar none, steadfast and kind.

And dad, who walked with me like no other. I dedicate this book to him.

3
Contents
Chapter 1 Introduction .......................................................................................................................... 7
1.1 First Steps ...................................................................................................................................... 7
1.2 More Rigour .................................................................................................................................. 9
1.2.1 Representational Content ..................................................................................................... 9
1.2.2 Phenomenal Character ....................................................................................................... 11
1.2.3 Relation to Belief ................................................................................................................. 12
1.2.4 Degrees of Belief ................................................................................................................. 12
1.2.5 Justification ......................................................................................................................... 12
1.2.6 Not Knowledge.................................................................................................................... 14
1.2.7 … and Not Evidence ............................................................................................................ 16
1.3 Aim and Approach ...................................................................................................................... 16
1.4 Why This Matters ........................................................................................................................ 18
1.5 A New View ................................................................................................................................. 20
Chapter 2 Reduction? ........................................................................................................................... 23
2.1 Doxastic Views ............................................................................................................................ 23
2.1.1 Motivation........................................................................................................................... 23
2.1.2 Taxonomy............................................................................................................................ 24
2.2 The Standard Case Against Doxastic Views ................................................................................ 25
2.3 The Argument from Rational Criticisability................................................................................. 27
2.4 Partial Belief ................................................................................................................................ 30
2.5 Doxastic Attitudes with Different Content ................................................................................. 32
2.6 Intuition as a Disposition to Believe ........................................................................................... 32
2.7 Rational Criticisability Returns .................................................................................................... 34
2.8 A Lesson about the Nature of Intuition ...................................................................................... 37
Chapter 3 The Shape of the View......................................................................................................... 38
3.1 Intuition is a Conscious Experience............................................................................................. 38
3.2 The Justification Hypothesis is True ............................................................................................ 39
3.2.1 A Lower Boundary ............................................................................................................... 40
3.3 Liberalism .................................................................................................................................... 40
3.4 Phenomenalism .......................................................................................................................... 43
3.5 The Absent-Experience Challenge and ‘Overlookable’ Phenomenal Character ......................... 44
3.6 No Etiological Restrictions .......................................................................................................... 47

4
 3.7 Intuition is Wholly Permissive with respect to Content ............................................................. 49
Chapter 4 Perceptual and Intuitional Experience ................................................................................. 52
4.1 Methodology............................................................................................................................... 52
4.2 No Content-Specific Phenomenology ......................................................................................... 54
4.3 Answering the Absent Experience Challenge ............................................................................. 57
4.4 Phenomenology of Objectivity ................................................................................................... 58
4.4.1 Recognising Objectivity ....................................................................................................... 59
4.4.2 Objectivity and the Content of Perception and Intuition ................................................... 61
4.4.3 Objectivity Explains Transparency ...................................................................................... 63
4.5 Phenomenology of Pushiness ..................................................................................................... 65
4.5.1 Recognising Pushiness ........................................................................................................ 66
4.6 Valence........................................................................................................................................ 69
4.7 Belief ........................................................................................................................................... 70
4.8 Objectivity, Pushiness, and ‘Presentation’.................................................................................. 72
4.9 What Intuition Is ......................................................................................................................... 78
Chapter 5 Phenomenalism ................................................................................................................... 80
5.1 Perception singlehandedly justifies belief .................................................................................. 80
5.1.1 Perceptual justification without induction ......................................................................... 81
5.1.2 Incoherence ........................................................................................................................ 83
5.1.3 Taking Stock ........................................................................................................................ 83
5.2 Perception Justifies Belief In Virtue of Its Phenomenal Character ............................................. 83
5.3 Two Unsuccessful Attempts ........................................................................................................ 87
5.3.1 Seeming Able to ‘Just Tell’ .................................................................................................. 87
5.3.2 ‘Presentational’ Phenomenology........................................................................................ 89
5.4 Pushiness and Objectivity Explain Justification .......................................................................... 93
5.5 In Defence of the Analogy........................................................................................................... 97
5.5.1 The Disanalogy of Content-Specific Phenomenology ......................................................... 97
5.5.2 The Disanalogy of a Known Causal Mechanism .................................................................. 99
5.5.3 The Disanalogy of Valence ................................................................................................ 101
5.6 Taking Stock .............................................................................................................................. 101
Chapter 6 The FIFO Objection .......................................................................................................... 104
6.1 The Feeling-In-Feeling-Out Objection ....................................................................................... 104
6.2 What the Objection is Not ........................................................................................................ 104

5
 6.3 Significance ............................................................................................................................... 106
6.3.1 Core ................................................................................................................................... 106
6.3.2 Scope ................................................................................................................................. 107
6.3.3 Depth................................................................................................................................. 107
6.4 Initial Evaluation ....................................................................................................................... 107
6.5 A Reply ...................................................................................................................................... 109
6.6 Objections ................................................................................................................................. 111
6.7 A Verbal Dispute, Epistemic Pluralism, and Justification .......................................................... 113
Chapter 7 Too Much Justification? ..................................................................................................... 116
7.1 Defeat........................................................................................................................................ 116
7.2 Cognitive Penetration ............................................................................................................... 120
7.3 Logical Fallacies ......................................................................................................................... 122
7.4 Should’ve Known ...................................................................................................................... 123
7.5 There’s (Even) More to The Full Story, Of Course .................................................................... 125
7.6 Taking Stock .............................................................................................................................. 126
Chapter 8 Final Thoughts .................................................................................................................. 127
8.1 Two Questions .......................................................................................................................... 127
8.2 The Role of Intuition in Philosophy ........................................................................................... 129
8.3 The Role of Intuition in Everyday Life ....................................................................................... 132

6
Chapter 1 Introduction
SOCRATES: Being able to cut things up again, class by
class, according to their natural joints, rather than trying
to break them up as an incompetent butcher might.

PLATO
Phaedrus

1.1 First Steps

Is torturing an innocent person OK?

Just now something happened: it seemed to you (I shall assume) that torturing an innocent person is wrong.
This went on for a period of time, then it stopped. What kind of thing happened?

You believe many things: that Paris is the capital of France, that the Eiffel Tower is in Paris, and that the
currency in France is the Euro, for example. You also have hopes and fears, desires and preferences, you rea-
son and ruminate, and you feel, taste, see, and hear various things. These are all mental states (or events,
hereafter simply ‘mental states’). What happened is broadly speaking the same kind of thing as any of these:
for a few seconds you were in a mental state.

But what kind of mental state were you in? What is its nature? These questions are interesting in their own
right. We want to know what the world is like; what the nature of reality is. Minds occupy a particularly in-
teresting corner of reality. We inquire into the nature of beliefs, preferences, hopes, and fears. In the same
spirit, and for the same reason, we should also seek to discover the nature of the mental state you were just
in.

It is also interesting to ask what, if anything, mental states like this one can do. Perhaps you now believe that
torturing the innocent is wrong because it just seemed to you that it is. If you do believe that, that is on the
face of things appropriate. A good question is whether it really is appropriate, and if so, why and in what
way.

Furthermore, ‘seemings’ like this one appear to play various roles, both in everyday life, and in various sci-
entific and academic disciplines, including physics, maths, linguistics, law, and philosophy. It is interesting
to ask whether they really do play the roles they appear to play, and if they do, whether that is appropriate.

Quite a bit of philosophical attention has been paid to questions like these last ones. Not nearly as much
attention has been paid to discovering the nature of mental states like the one you were just in. But this
creature of the mind is an interesting one, well worth our attention. Moreover, understanding the nature of
these states will help us to answer other questions about them, including in particular whether believing
something because that’s how it seems to you really is appropriate.

7
Let’s consider some further examples. To many people—and maybe to you, now that you think about it—it
seems that if something is red it is coloured. To many it seems that a rational person can't believe both that
things are, and are not, a certain way: both that there is and is not another person in the elevator with
them, for example. To many it seems that this sentence: “The boy the man the girl saw chased fled” is un-
grammatical. To many it seems that if Anne is taller than Bob, and Bob is taller than Cliff, then Anne is
taller than Cliff. To many it seems that people generally prefer less pain to more. To many it seems permis-
sible to kill an aggressor if that is the only way to save one’s own life. And to many it seems that if events A
and B happened at the same time, and B and C did too, then A and C must also have.

All these mental states have certain things in common. Most obviously; each of them ‘says’ something,
something that is either true or false. It is true that if something is red it is coloured, that people generally
prefer less pain to more, that if Anne is taller than Bob, who’s taller than Cliff, then Anne is taller than
Cliff, and that killing in self-defence is morally permissible. It is false that the sentence is ungrammatical,
and that simultaneity is transitive.

A second commonality is that people sometimes come to believe that things are a certain way because that
is how things seem to them to be. We can easily imagine this happening for each of our examples; indeed it
has probably happened many times. All sorts of things might happen after that: the person might forget
that things seemed that way to them, they might reject the way it seemed on reflection (as you might, when
you manage to parse the sentence), or either lose or fail to form the belief for any number of other reasons.
But it is still clearly true that people often come to believe that things are a certain way because that is how
things seem to them to be.

A third commonality, already noted, is that if a person comes to believe that the sentence is ungrammatical,
or that it’s not OK to torture the innocent, because that is how things seem to him, then that is on the face
of things appropriate. In fact, we can say something stronger: it is apparently appropriate not just in any old
way, but in a particular way. Perhaps a sprinter should believe that she will win the race regardless of how
likely that is, if that encourages peak performance. By contrast, if you now believe that torturing the inno-
cent is not OK, that belief is not appropriate because it serves some further end: it is simply appropriate for
you to so believe, in and of itself.

Finally, the examples we have mentioned are similar in that being in the mental state feels a certain particu-
lar way. Exactly what it feels like is tricky to describe at first, and I will try hard to get this right later on. One
thing that immediately stands out, however, is that the mental states are not neutral with respect to what
they ‘say’. When it seems to you that if something is red it is coloured, for example, that is no longer some-
thing you can easily remain neutral about: you are pushed to believe that that is how things really are. (You
are not actually moved around, of course; think of this as a roundabout way of doing something quite diffi-
cult, namely to describe an aspect of what it feels like to be in that mental state.)

These apparent commonalities appear to bind these cases, and others, together. Here as elsewhere appear-
ances could be deceptive. But unless we find good reason to think that they actually are deceptive it will re-
main reasonable to think that there is a class of mental states worth caring about here.

So these are our first steps: a list of examples, the recognition that they all appear to have certain things in
common, and a preliminary description of these apparent commonalities.

8
1.2 More Rigour
Most of the examples of seemings above would be accepted by most philosophers as ‘intuitions’. I’ll have
more to say about this later, but for now I will simply adopt this usage.

So far, I have tried to keep things simple and non-technical. But no terminology is innocent or pure, and
we won’t get far by trying to stay clear of commitment and controversy. I will therefore now say some of the
same things again, in more detail, and using language that is a bit more technical, and which commits me a
bit more. This is useful because it makes the starting point clearer, because it shows what I mean by some
key terms, and because it brings to light some key assumptions that will play important roles in what fol-
lows.

Before that, a brief methodological point. Throughout this book I will make unashamed use of intuition
itself to characterise and discuss intuition. I can see no way around this, but also and more importantly no
reason not to. So I shall (continue to) feel free to say things like: “It seeming to an agent as if things are a
particular way seems to support her belief that things really are that way”. As we shall see, the conclusions I
reach in later chapters vindicate this practice.

1.2.1 Representational Content

I said that each intuition ‘says’ something that is either true or false. Let’s get a bit clearer on what that
means.

My belief that there is a person behind me is either true or false. For it to be true, the way things are must
meet certain conditions, which they may or may not meet. In this case, these conditions are, roughly, that a
living human being (that is not disqualified from personhood, if that’s possible) must be located quite close
to my back, if I am standing up or sitting, or behind my head, if I am lying down. These are that belief’s
truth conditions. My belief represents that the way things are is one of the ways that satisfies these conditions.
That is the representational content of my belief, or just its content, for short.

‘One of the ways’, because there are many aspects of the way things are on which my belief places no condi-
tions, and therefore a lot of room for variation consistent with my belief. My belief doesn’t specify what the
weather is like, for instance, or what the person behind me is wearing. For my belief to be true, things must
be one of the many, many ways they might be that are compatible with what the belief does specify: Tom is
behind me wearing jeans in the sun; Dick is behind me wearing a trench coat in the fog; Harry is behind
me in a Batman outfit in the rain, and so on.

My belief has truth-conditions, but what I believe is not truth conditions. What I believe is that the truth-
conditions are satisfied: that way things actually are is one of the many ways things might be while meeting
these conditions. It is tricky to state this without leaving any room for an interpretation on which the belief
represents both its own truth conditions and that these conditions are met, but to be clear, that is not the
picture.

This notion of representational content straightforwardly applies to other types of mental states as well.
Your intuition that torturing the innocent is not okay has truth-conditions. For it to be true, things have to
be a certain way, namely such that torturing the innocent is morally impermissible. Just as with my belief
discussed just above, there are many things on which the intuition places no conditions. What the intuition
represents is that the ways things are is one of the many ways things might be that is compatible with the

9
constraints that it does place. That is the representational content of the intuition, or just its content, for
short.

This notion of representational content applies to perception too, although a small adjustment is required.
Suppose I have a perceptual experience in which, among other things, I seem to see that a person is walking
down the street. 1 My perceptual experience represents a great many further things besides this: the colour of
the person’s clothing and of the houses in the background, that the sun shines on her head and torso but
not on her legs, and so on.

Because some of the things my perceptual experience represents might be true while others are false, a bi-
nary notion like truth conditions is too blunt an instrument to describe its content. Instead we say that per-
ceptual experience has accuracy-conditions: it is wholly accurate if those conditions are completely satisfied; and
otherwise it is accurate to various different degrees, all the way down to being wholly inaccurate. However,
even though perceptual experience represents many things, and therefore places many constraints on the
way things are, it resembles intuition and belief in that there are many things on which it places no re-
strictions at all. What it represents is that things are one of the many ways they might be such that those
conditions are (wholly) met. That is the representational content of perceptual experience, or just its con-
tent, for short.

This is a natural way to understand representational content, and a useful one for present purposes. 2 Other
notions may be more useful for other purposes, and may also deserve the label, but in this book the content
of intuitions, beliefs, perceptual experiences, and other mental states is understood as just outlined.

I can have a belief with the simple content that there is a cup on the table in front of me, but it is not clear
that I can see just that. If I see that there is a cup on the table in front of me I usually see many other things
as well: that it has a certain colour, size, and shape; that it is a certain distance from the edge; that it is partly
in shadow and partly lit, and so on. Compared to belief and intuition, the content of perceptual experience
is usually very rich. (This point will become important later on, because it helps us to explain why some peo-
ple mistakenly take intuition to be more different from perception than it actually is.)

Some argue that perceptual experience differs from belief and intuition not only in the richness of its con-
tent, but also in its kind: whereas belief has ‘conceptual’ content, perceptual experience has content of a dif-
ferent, ‘non-conceptual’ variety.3 In my view, the similarities between perception and intuition are signifi-
cant, so I need to say something about this potential difference between the two states.

If the content of perception is sufficiently different from that of belief, it may be that we never literally be-
lieve what we see. However, we clearly do believe things on the basis of what we see, and properly so. Within
the things we properly believe on the basis of perception, we can distinguish beliefs that involve some meas-
ure of ‘jumping to conclusions’ from those that do not. For example, when I look down the corridor and see

1
I say ‘seem to see’ since, because if I see that a person is walking down the street, it follows that a person actually is walking down
the street: the ‘seeing that’ locution is factive, to use some philosophical lingo. Many locutions are: ‘remember that’, and ‘realise
that’, for example. But just like with intuition and belief, perceptual experience can be inaccurate or false. So when we wish to talk
about the content of perceptual experience in a way that allows for this, we signal the absence of factivity by saying ‘seem to see’.
2
It is by no means my own invention, but a widespread way to understand the notion; see e.g. (Byrne 2009); (Jackson 2010: 44-50);
(Pautz 2010); (Schellenberg 2011); (Siewert 1998: 189-92); and (Siegel 2005/2010, 2010).
3
See e.g. (Crane 1988a; Crane 1988b, 1992b); (Evans 1982); (Heck 2000); and (Peacocke 1986, 1992; Peacocke 2001).

10
that nowhere is light shining out from under a door, in some sense I see that I am the only one still working.
But there is another sense in which that is not what I see: it is a conclusion I jump to based on what I do
see.

Because everyone needs a way to account for this difference, everyone needs a notion that plays much the
same role as representational content does here. 4 I think that perception has the same kind of content as
intuition and belief do, but I also think that even if this turns out to be wrong, my claims can be restated in
whatever terms ultimately turn out to be correct. The similarities I rely on between intuition and perception
will remain. 5

Finally, representational content is often usefully characterised in terms of possible worlds. I will think of a
possible world as a complete way things might be. My belief that I have some coins in my pocket leaves all
sorts of things open. A possible world leaves nothing open; it settles all details. Restated in this jargon; my
belief represents that the actual world is in the set of possible worlds in which I have some coins in my
pocket. Because there are so many things on which my belief places no constraints, there are many worlds
in that set. What my belief represents is that the actual world is in it: it is one of those worlds.

1.2.2 Phenomenal Character

At any given time, there is something particular it is like to be a given conscious being, such as a human.
Suppose that one afternoon you are standing on a beautiful beach, looking out to sea. Your bare feet are
being lapped by small waves of pleasantly cool water, the sun and breeze are on your face, and the sound of
seagulls mixes with children’s laughter and the rustling of leaves from tall gumtrees behind you. You are
chewing on an apple and enjoying the smell of a nearby barbeque. You may not stop to think about it, but
it feels like something, right at that moment, to be you. There is something it is like overall to be you, right at
that moment. Your overall, global conscious experience has a particular phenomenal character.

4
As Stalnaker notes, this notion of representational content is minimal in the sense that it is “a kind of content that everyone
should agree can be used to characterize mental and linguistic states, acts and events that can be said to have representational con-
tent of any kind” (Stalnaker 1998: 343). (See also Siegel 2012.) I think that the distinction I am drawing here largely coincides with
the distinction James Pryor draws between the propositions a perceptual experience basically represents, on the one hand, and those
that it only non-basically represents, on the other (2000: 538-9). As he notes, locutions such as ‘It looks as if …’, and ‘I (seem to) see
…’ are often applied even in the latter case. I still prefer a picture a more unitary notion of content, however; one on which a given
proposition simply is or is not in the contents of a given mental state. Many of the propositions Pryor (and others) take to be non-
basically represented by perceptual experience are in my view best regarded to not be part of those experiences’ contents at all, and
the fact that they are so easily justified by those experiences must be accounted for in some other way. What is and is not in the
content of a given experience will hinge on what the correct account of mental content turns out to be, and it may at least in part
be an empirical question. While we wait for all of that we will at least get close if we ask ourselves whether a given proposition is
part of the accuracy conditions of the state in question. The experience I have as I look down the corridor is not inaccurate if one of
my colleagues is working in the dark. It therefore does not represent that I am the only one still working, even though that proposi-
tion is easily justified by the experience. For these reasons, then, this book will not employ a distinction between basically versus
non-basically represented content, but simply discuss regard propositions as either in a state’s representational content, or not.
5
I know of no philosophical account of intuition which contests the assumption that intuition has the type of content I have dis-
cussed here. (I set aside de re intuition, which (if it exists) is a separate phenomenon. See e.g. (Pust 2012/2019) or (Parsons 1995)
for discussion.) Moreover, arguments intended to show that perception has ‘non-conceptual’ content have no bite on intuitions.
One example is Crane’s argument that one and the same experience has contradictory content, and that this shows that the content
cannot be conceptual (1988b). Another is the ‘richness’ argument discussed by Richard G. Heck Jr. (2000) among others. On the
former; though we may have intuitions that together are contradictory, it is not at all clear that there are single intuitions with contra-
dictory content. On the latter, there is no corresponding richness in intuition as in perception, and we usually have little trouble
articulating what we intuit. (Which is not to say, of course, that there can’t be disagreement about what a given intuition represents,
which of course there can be.)

11
There is also something it is like to taste an apple; to hear rustling leaves or a particular piece of music; to
be embarrassed, elated, anxious; to have a tickle, a pain, or an itch. We don’t have such local conscious ex-
periences in isolation: when I eat an apple many other things are also happening which make a difference
to the character of my overall experience. But if there is something it is like to taste an apple, then doing so
makes a difference to the character of the overall experience of the person who is tasting it. A particular lo-
cal conscious experience has a particular phenomenal character just in case it makes a particular contribution
to the character of the global experience of the person who is having it. That is how these terms are used
here.

We can now restate the earlier claim of commonality with more precision. To say that being in this kind of
mental state feels a certain particular way is to say that having an intuition makes a certain distinctive contri-
bution to the character of the overall experience of the person who is having it. This claim can be under-
stood in different ways, and not all of them render it true. But there is an important sense in which it is
true; or so, at least, I shall argue. 6

1.2.3 Relation to Belief

I take it for granted that people often come to believe that things are certain ways because that is how they
seem to them to be. A person having an intuition that p will often cause her to form the belief that p.7 The
interesting question is not whether or not this happens, but whether this is how it should be.

1.2.4 Degrees of Belief

We can talk about belief as a binary phenomenon: a person believes, or doesn’t believe, that it’s about to
rain, for example. But belief actually comes in degrees: one can believe more or less strongly that it’s about
to rain. Degrees of belief are often called credences, and denoted by numerical values in the [0,1] interval,
with 1 indicating certainty and 0 the opposite.

Just as binary beliefs can be justified, degrees of belief can be justified, too. Indeed, a natural way to opera-
tionalise the idea that justification comes in degrees is to tack it on to degrees of belief: stronger justification
justifies higher credence in the relevant proposition. That is how we will think of them here.

1.2.5 Justification
We all go around believing all sorts of things. We also routinely evaluate beliefs, our own and those of oth-
ers, along a number of dimensions. The most obvious one is truth: a belief being true is an important mark
in its favour, an important good-making feature of a belief, as one might put it.

But truth isn’t the only important good-making feature a belief can have. If a person believes that p because
it seems to her that p, that seems appropriate in the way that is specifically concerned with belief being ap-
propriate in and of itself, as opposed to believing something being instrumentally useful for some further

6
Phrases like ‘the phenomenal character of experience’ sometimes carry heavy theoretical or metaphysical implications, but beyond
what has just been outlined, they carry no further implications here. I take no stance, for example, on the question of whether phe-
nomenal facts can ultimately be reduced to physical ones; on whether (some part of) the phenomenal character of experience is
ultimately ineffable; or on whether something general can be said about the relation between the phenomenal character of a mental
state and its representational content (and if so, what).
7
‘An intuition that p’ is philosopher’s shorthand for an intuition the representational content of which is some proposition or
other, and it is used when it doesn’t matter which proposition is represented.

12
end, like that of running a sprint as fast as possible. That is another good-making feature a belief might
have, and we capture it by saying that the belief is justified.

A person can have justification to believe that p even if she is unable to defend her belief against epistemic
challenge. To defend one’s belief—to argue that it is justified in the face of a claim to the contrary—requires a
lot of sophistication (and resilience). Having justification does not: justification is much easier to acquire
than to defend dialectically, or, indeed, to account for theoretically. But justification is not mysterious: if
my trustworthy and competent friend once in a blue moon tells me a lie, and I believe it, then my belief is
justified even though it is false. A very striking fact is that even those who are wholly unfamiliar with the
concept’s use in theoretical contexts—indeed even those who are wholly unfamiliar with thinking about be-
lief, and the things that might be good or bad about belief, in a theoretical way at all 8—are often able to im-
mediately lock onto the target phenomenon from just one or two examples such as this one.

The most plausible explanation for this very striking fact is that we are all already familiar with the concept
of justification from its ubiquitous use in practice.9 It is commonplace, for example, to criticise someone for
holding an unjustified belief. When we say things like: ‘why do you say that?’, ‘why do you think that?’, or
‘you’re just guessing!’, we are challenging someone to explain their justification for a belief they hold. In pri-
vate reflection we might (on good days) ask ourselves whether we really should believe what we believe, and
revise our beliefs if the answer is no. All this suggests that the concept of justification plays an important
role in our lives, and in particular in the ubiquitous and important practice of reflection on, and evaluation
of, belief (Smithies 2015). 10 The view that we are all already familiar with the concept of justification can
explain the striking fact: the reason the concept of justification is so easily explained and understood is that
to ‘explain’ it is actually just to provide a convenient label for something that is already deeply familiar.

One might have chosen a more everyday turn of phrase instead. One could say, for example, that when it
seems to S that p, this ‘makes it OK’ for her to believe that p, or that p then has ‘the property of being wor-
thy of belief’ (Bengson 2010: 30). I see no virtue in this approach. Again, no term is free of commitment or
unfortunate connotation, and this is equally true for these would-be replacements. That the notion of justi-
fication can so easily be explained strongly suggests that it is one on which we have a solid grip inde-
pendently of theoretical engagement. This makes it an excellent notion to employ in our inquiry.

To say that justification is central to reflection on and evaluation of belief is not to claim that in order that a
person be justified in holding a certain belief, that belief must actually have been subjected to evaluation or
critical reflection. Instead, a justified belief is “in a position to pass such a test” (Alston 1989: 225-6), and
justification is “the epistemic property in virtue of which a belief has the potential to survive critical reflec-
tion” (Smithies 2015: 227-28). 11 These are modal claims, and it is important to understand them correctly.

8
I don’t mean to sound snooty; it’s of course completely reasonable for nearly everyone to not think about belief and its good-mak-
ing features in a theoretical way.
9
This claim might seem to be belied by the relative scarcity of the use of that word in every day contexts. It is not; a concept can be
in use even if a word that denotes it in certain theoretical contexts is not. My friend the engineer once told me to ‘agitate’ a fluid.
That I didn’t understand him did not show that I lacked the concept of stirring.
10
See also (Kaplan 1985: 358): “The enterprise of trying to arrive at justified belief is nothing more than the enterprise of trying to
arrive at a belief supported by reasons that will stand up under critical scrutiny. In asking, of the argument you have produced in
the case under discussion, ‘Does that argument stand up under critical scrutiny?’, I am asking a kind of question I have occasion to
ask virtually every time I engage in inquiry”.
11
As Smithies notes, this idea needs to be nuanced in various ways to escape objections (Smithies 2015: 227-28), but this does not
matter here.

13
First, that the belief is in a position to pass such a test does not mean that the person herself must be in a
position to submit it to critical reflection, or to defend it against epistemic challenge: young children and
persons with intellectual disabilities have justified beliefs. Second, such reflection takes into account that
available information is partial and limited: it is concerned with what can be expected of someone in the
subject’s epistemic position. It is no objection to the claim that a certain subject is justified in holding a cer-
tain belief that there is information out there somewhere which shows that the belief is false. It is an objec-
tion if one of the subject’s own beliefs shows this. We will return to this theme several times below.

Two final points. Earlier it seemed to you that torturing the innocent is wrong, and it seems that you there-
fore now have some justification to believe this. You have this justification whether or not you actually do
believe this. We mark this by saying that what’s at issue is propositional, not doxastic justification. A person
can have propositional justification to believe that p without believing that p. If she does believe it, she can
have, or fail to have, doxastic justification for that belief: having propositional justification to believe that p
doesn’t guarantee having doxastic justification to believe it. It is usually held that a person has doxastic justi-
fication to believe that p if i) she has propositional justification to believe it, and ii) she believes it on the
basis that provides propositional justification in the right sort of way.12 This again is an issue that we will
revisit later, but for now the point is to focus on propositional rather than doxastic justification.

Second, justification comes in degrees. One easy way to see this is to notice that several considerations can
favour believing the same thing. For example, seeing dark clouds rolling in gives you justification to believe
that it will rain before long. The weather forecast can make your justification stronger, and an inference
from your friend’s testimony that it’s already raining in the next suburb over coupled with knowledge of the
wind direction can make the justification stronger still. Belief can be justified more or less strongly.

1.2.6 Not Knowledge

In epistemology, a question with a long and venerable history is the so-called ‘value problem’. 13 You’d be
forgiven for thinking that (perhaps partial) formulations of a problem with that name might be:

What should epistemic agents value? What ought epistemic agents to strive for?

To be sure, these formulations raise questions of their own, such as ‘are the two statements equivalent?,
‘what is an epistemic agent?’, and so on. But it is clear that they get at something central to epistemology,
and something that might aptly be called ‘the value problem’.

But they don’t reflect the received way of thinking about the value problem. One is forewarned of this by
the way the field of epistemology itself is introduced in textbooks, where it is usually glossed as the study of
knowledge. 14 Similarly, the value problem is often said to be the problem of specifying in virtue of what
knowledge is more valuable than mere true belief (Fricker 2009; Haddock 2009). Notice that this presup-
poses particular answers to the above questions, namely that what we should value and strive for is
knowledge. It also presupposes a particular contrast, namely between knowledge and (mere) true belief.

12
An early deployment of the distinction is found in a paper by William Alston, who asks “whether it is enough for justification
that S have adequate grounds for his belief, whether used or not, or whether it is also required that the belief be based on those
grounds” (1985: 74). Alvin Goldman (1979: n. 17) attributes the distinction to (Firth 1978).
13
A really long and venerable history, as it turns out: the problem is raised in the Meno (Plato: 97a).
14
See e.g. (Audi 1998/2011); (Markie 2004/2017); and (Pritchard 2006/2014).

14
Knowledge is indeed widely seen as the epistemic gold standard, the thing which epistemic agents ought to
and do strive for, and the primary object of epistemic inquiry.

This is not the only possible view. For one, it is reasonable to object to the contrast: contrasting knowledge
with mere true belief stacks the deck in favour of knowledge since it is easy to see that knowledge is more
valuable than mere true belief. A better alternative says that justification is the thing that really matters: not
simply or even primarily for its role in knowledge, but for its own sake. To illustrate, Miranda Fricker argues
that the value of knowledge (over mere true belief) is its resilience; its “tendency to survive misleading counter-
evidence owing to the subject’s being in a position to weigh it against positive evidence already possessed” (2009: 129).
But justified belief is just as resilient to misleading counter-evidence as knowledge is, or at any rate as resili-
ent as we should want it to be: if too much counter-evidence stacks up and you don’t change your belief
that’s not resilience but stubbornness, which is not a good thing but a bad one. Sufficient counter-evidence
should cause us to give up a belief, even if the counter-evidence is misleading because the belief is true.15

In Knowledge and Its Limits, Timothy Williamson (2000) argues for ‘knowledge first’ epistemology: an ap-
proach which rejects that knowledge is the primary target for explanation and analysis, and instead uses
knowledge to explain other phenomena. Williamson’s argument is wide-ranging and influential, and I can-
not fully addressed it here. But a few remarks are in order.

First, Williamson claims that the stance is supported by the value we place on knowledge. “[K]nowing mat-
ters”, he says; “[e]ven unsophisticated curiosity is a desire to know” (30). But it is at least equally clear that
being justified matters, and curiosity is equally well understood as the manifestation of the desire to do one’s
best, rationally speaking; to believe ‘as best one can’. “Factive mental states [such as knowledge] are important
to us as states whose essence includes a matching between mind and world”, says Williamson (40). But it is
no less plausible that justified beliefs are important to us as states whose essence include being appropriately
responsive to one’s own epistemic position. Responding appropriately to one’s own epistemic position is
the very best one can do to match one’s mind to the world. It is the thing it makes sense to strive for.

Second, Williamson notes that a person knowing that p sometimes better explains her actions than her
merely truly believing that p (62). That is true, but does not count against the view that justification is what
really matters, since those cases fail to establish that anything is explained by a subject’s knowing that p that
isn’t equally well explained by her believing that p with strong justification. Indeed, Williamson’s cases can
be explained without reference to justification, and by reference to high and robust credence. 16 In one of
Williamson’s examples, a burglar continues to search for a hidden diamond even at high risk of detection,
and this is better explained by his knowing that it is in the house than merely by his truly believing it. True.
But it is not better explained by his knowing that the diamond is in the house than by his having a high and
robust credence to this effect; and still less better explained by his knowing that it is in the house than by
his having a strongly justified belief to this effect. If we wish to explain not only that the burglar stayed on,
but that he was (at least potentially) rational in so doing we need an epistemic notion, but knowledge and
strong justification will do equally well.

15
I’m using ‘evidence’ in a way intended to exclude mere obfuscation or noise; the situation I have in mind is one where there are
real epistemic indications that one should change one’s belief, it just so happens that those indications don’t track truth.
16
When specifying the attitude a person should hold to a proposition given some evidence, it is not sufficient to note what cre-
dence she should have. One must also note how robustly that credence should be held. Different evidence may dictate the same
credence, but different robustness. The testimony of one credible witness dictates high credence but lower robustness than that of
twenty credible witnesses to the same effect, even though the latter may easily fail to dictate a higher credence.

15
Finally, since justification remains important in knowledge first epistemology (41); even those who fully ac-
cept Williamson’s view can still find value in clarifying when justification does and does not obtain. 17

I think it better to focus not on knowledge but on justification. A convincing argument to this effect was
advocated by Mark Kaplan (1985). He argues that being able to distinguish between cases when an agent
can truly be ascribed knowledge from cases in which an agent can merely be ascribed justified belief can play
no role in the central epistemic project of “advancing or clarifying the proper conduct of inquiry” (354).
What matters is determining whether one’s reasons for belief “stand up under critical scrutiny”, a determi-
nation, as he notes, that is at “the very heart of the process of arriving at justified belief” (358). Since we
cannot distinguish between knowledge and mere justified belief from the inside, all we can do in order to
conduct inquiry as well as possible is to seek justified belief (361); to conduct ourselves with as much of
what Wright calls ‘intellectual integrity’ as possible (Wright 2004: 210; see also Huemer 2001: 20-22, 104-
5).

Conducting inquiry as well possible is, I think, what epistemic agents should strive to do, and epistemology
will in my view be most useful if it sheds light on what doing that consists in. Thus the focus in this book is
on whether intuition can provide justification, and if so, how. It is not on whether or how intuition can
provide knowledge. From the point of view of clarifying rationality and the proper conduct of inquiry the
former questions matter a great deal, and a lot more than the latter, if indeed the latter matters at all.

1.2.7 … and Not Evidence

Finally, the focus in this book is also not on the question of whether intuition provides evidence, despite the
fact that much recent discussion in philosophy has been couched in these terms (see e.g. (Climenhaga
2018), and references therein).18 The notion of evidence has been recruited to play a number of different
theoretical roles, and it is far from clear that a single notion can play them all (Kelly 2006/2016). There are
also widely divergent views of what constitutes evidence. 19 Correspondingly, the term is used with very dif-
ferent meanings in a way that invites confusion and equivocation, both in discussions of intuition and else-
where. 20 While perhaps possible, conducting the discussion in a way that would forestall all potential mis-
understandings here would be quite difficult, so I choose instead to avoid the term as much as possible, and
to focus squarely on justification.

1.3 Aim and Approach

The commonalities I have outlined between instances of intuition give us good initial reason to think that
grouping the cases together will give us valuable explanatory purchase. It seems significant that the cases ap-
parently share these features. We should seek to discover whether that really is so, but that there is a class of
phenomena worth caring about here is a natural working hypothesis.

17
Accepting my account does require rejecting Williamson’s claim that only what one knows can justify belief (or ‘E=K’), however,
since (I take it) one’s experiences aren’t known in the relevant sense.
18
See (Huemer 2001: 102) for lucid discussion of this point in the context of his phenomenal conservativism.
19
Kim says that “one thing is ‘evidence’ for another just in case the first tends to enhance the reasonableness or justification of the
second” (Kim 1988: 390). According to Smithies, evidence is “what gives you epistemic justification to hold certain beliefs and oth-
ers doxastic attitudes by supporting their contents to a fitting degree” (Smithies 2019: 196). Epistemically powerful experiences qual-
ify as evidence on these notions, but not on other notions: as the totality of what one knows, for example (Williamson 2000).
20
Some uses aim to capture things that are justification-making, in that they make it the case that you have justification to hold some
belief or credence. Others aim to capture things that are justification-showing: “propositions you have epistemic justification to be-
lieve, which you are rationally permitted to use as premises in reasoning” (Smithies 2019: 196); the distinction is from (Pryor
2005/2013). Intuition is evidence in the first of these senses, but not in the second.

16
A central aim of this book is to contribute to our understanding of mind and rationality by investigating
whether these cases really do have something important in common, and if so, what exactly that is. Neither
the list of examples nor the descriptions I have given are sacrosanct. I adopt the methodological approach of
letting these two elements jointly constitute the starting point, with the expectation that amendments in
either or both may be necessary as the investigation proceeds.

I aim to vindicate the starting-point impression by convincing you that there really is a psychological kind
reasonably called ‘intuition’. A psychological kind is simply a natural kind of the mind. A natural kind is a
category that, as Socrates said, ‘cuts nature at its joints’—a category that traces differences that really are there,
as opposed to dividing lines artificially imposed by an idiosyncratic human perspective. A psychological
kind cuts the mind at its joints—tracing differences between mental states that really are there.

To justify belief in the existence of a psychological kind I must show that the characteristics shared by mem-
bers of that kind explain how a kind as thus conceived can do sufficiently significant theoretical work. That
is what I set out to do. This parallels what is required for belief in a natural kind more generally: we have
reason to believe in the elements on the periodic table because of what our story about elements’ shared
properties allows us to explain, for example.

It is worth noting that I do not aim to investigate how the word ‘intuition’ and its cognates are used in ordi-
nary English, or within philosophy, mathematics, law, or any other academic discipline. Commentators of-
ten point out that these words are used in varied ways, and often in such a way that they clearly don’t refer
to mental states at all (Bengson 2010: 10-11; 2015b, 2014; Cappelen 2012). That is true, but poses no chal-
lenge to any claim made in this book.

First, many instances of loose usage are clearly derivative of uses that do refer to mental states: I can truly
say that a colleague has a certain intuition even when I know that she’s asleep, for example, if I know that
she has the relevant intuition whenever she considers the issue. But second, although it would certainly be
neater if people consistently used exactly one label for all and only the instances of the psychological kind
that I claim exists, the fact that they don’t does not at all suggest that it doesn’t.

Suppose that an investigation by a group of grumpy marine biologists were to find that most uses of the
term ‘clown fish’ in fact don’t refer to members of the relevant species, but instead to other small and
brightly coloured tropical fish. Such a finding would obviously not imperil the status of anemonefish as a
natural kind. To do that one would have to undermine its explanatory purchase, for example by showing
that a mutualistic symbiotic relationship between such fish and their host anemone in fact does not obtain.

In the same way, even if most uses of ‘intuition’ and cognates fail to refer to instances of the putative psy-
chological kind, this does not imperil the claim that there really is a psychological kind here. To do that one
would have to show that the noted characteristics do not in fact enable the kind to do the claimed explana-
tory work. In other words, the advocated account of the nature and epistemic status of intuition would have
to be criticised on its merits, and not on the basis of claims about word usage. This holds regardless of
whether one talks about the use of ‘intuition’ in common parlance, or restrict the cases of interest more
carefully, for example to uses within philosophy.

By far the most important and interesting question is whether there is a natural kind here, and whether it
can do the theoretical-explanatory work that I claim that it can do. Of much less interest is what we should
call that kind, assuming that its existence and explanatory properties is established. Nevertheless, that ques-
tion also has some interest to some.

17
What justifies applying the label ‘intuition’ to the kind is that the explanatory work the kind can do over-
laps sufficiently well with the core explanatory work things with that label have historically been invoked to
carry out. In the next section I argue that the core work is to be a ground-level source of justification, in ad-
dition to perception. If the story I tell succeeds in showing that mental states with the characteristics I single
out can play that role, then by this criterion it has also thereby justified labelling that state ‘intuition’.
Again, cases of loose usage, no matter how numerous or preponderant, make no difference. 21

An often noted fact about intuition is that we commonly use perceptual language to talk about it. George
Bealer says about de Morgan’s laws, for example: “you suddenly ‘just see’ it” (1992: 101). Unlike objections
from loose usage, the analogy between perception and intuition this talk suggests, is worth taking seriously.

I hasten to add that I do not wish to defend a ‘perceptual model’ of intuition. I am not sure what it would
take for an account to count as such, and I am ever keen to avoid verbal disputes. Instead, ‘taking the anal-
ogy seriously’ for me amounts to a methodological claim; namely that comparing intuition and perception
is theoretically fruitful. That this is so I hope to demonstrate by example.

1.4 Why This Matters

Why does this matter? I have already given an answer. We inquire into the nature of beliefs, preferences,
hopes, and fears. In the same spirit, and for the same reason, understanding the nature of intuition matters.

But we can say more. Not only do intuitions exist, but they play important epistemic roles, both in various
academic fields, and in our everyday lives. We form beliefs on the basis of intuition all the time, even if not
always explicitly. We shouldn’t do that if intuitions do not in fact justify belief: we should then revise our epis-
temic practice. So it is very important to find out whether or not they do.

This book thus forms part of a large, varied, and ongoing research project; namely that of examining the
ways in which human reasoning and inquiry goes right, and the ways in which it goes wrong. The aim,
broadly speaking, is self-improvement: to go right more often and to go wrong less frequently, and thereby
to increase our understanding, and live better lives. Much of this project is empirical; studying, for example,
how people actually reason or make choices in a range of situations. But not all of it can be: that people of-
ten rate a conjunction as more probable than one of its conjuncts (Tversky and Kahneman 1983) is interest-
ing only because we regard it as a fallacy, a misstep, a way in which we go wrong; and we can do so only
against the background of a priori justified beliefs about probability theory. Similarly, how we should evalu-
ate belief based on intuition depends on whether or not intuitions justify belief; again an a priori question.

To be an epistemic agent is to accept some epistemic practices and to reject others. Consider the epistemic
outrage you, a person of reason, might feel when someone suggests that a piece of quartz under your pillow
will improve your sleep, or that your flu might be cured by sleeping with an onion in your sock. Or consider
an atheist’s unwillingness to countenance religious experience as a source of justification for belief. One
needn’t be a nutritionist to lead a healthy life, and one does not require a theory of justification to form
justified beliefs. In both cases, however, theory is useful and important. The theory of intuition advocated
in this book says that what sets intuitions apart from the quartz-and-onion-peddler’s hunches, and from reli-
gious experience, is the specific phenomenal character intuition has and that those states lack. If we wish to

21
Since the word ‘intuition’ is often used loosely one must take care to not inadvertently apply to cases of loose usage the epistemic
lessons that only legitimately apply to core cases. But that is a point about how we apply the theory this book advances about the
nature and epistemology of intuition to (for example) philosophical methodology once that theory has been accepted, and not, as cases
of loose usage are often touted, a reason not to accept that theory in the first place.

18
claim that there is an important epistemic contrast to be had in these cases, it is important to know whether
that claim holds up.

Another way to bring out the importance of this inquiry is to situate it in a seminal dispute in the history of
thought, namely that between rationalism and empiricism. Simplifying somewhat we can say that rational-
ism and empiricism disagree whether all justification ultimately stems from perception. 22 Suppose I see a
bee and then a wasp, and come to be justified in believing that there are at least two insects nearby. For all
my justification to ultimately stem from perception not only must the justification for ‘there’s a wasp’ and
‘there’s a bee’ stem from perception (which it does); so too must any justification I relied on in the infer-
ence.

Empiricism is the view that all justification ultimately stems from perception. Rationalism is the view that
there are additional ‘ground-level’ sources of justification. If the argument in this book goes through, it vin-
dicates rationalism so conceived. That tells us something fundamental about the nature of rationality, about
the proper conduct of inquiry, and about the kinds of creatures that we are.

Rationalism is instead sometimes characterised as the view that these other sources of justification provide a
special kind of justification—justification strong enough to warrant certainty, for example; or justification that
cannot be overturned; or such that if you have it, what you believe must be true 23—or to hold that there are
propositions for which justification is only available from non-perceptual sources.24 But all of these things
saddle rationalism with unnecessarily strong claims. Rationalism can allow that the additional source(s) of
justification it countenances (usually, or often) fails to rationalise certainty, that the justification can be
overturned and doesn’t guarantee truth, and it needn’t say that some propositions can only be justified
non-perceptually.

The core rationalist claim, what is most fundamentally and importantly at issue, is simply that something
other than perception is a source of ground-level justification. The deepest question about intuition is
whether it is a source of justification at all. The position advocated here is that it is.

In current analytic philosophy this conclusion is radical. Scepticism about intuition is widespread and runs
deep; and derision of a philosophical method that makes use of it is often on clear display. 25 So if the con-
clusion can be established that intuition sometimes provides the intuiting subject with at least some justifi-
cation, that really is quite significant.

How the conclusion is established matters, too. In particular, it matters that it follows from a detailed ac-
count of the state’s nature. That is because scepticism about intuition’s ability to justify belief so often is (al-
leged to be) justified by the mystery said to surround its nature.

22
I put the point in terms of justification, rather than, what is more common, knowledge, for the reasons given above. As has often
been observed (e.g. by Markie (2004/2017: §1.2 ), the degree of opposition between these two camps, and indeed the extent to
which it makes sense to think of ‘camps’ here at all, can be overstated; and portrayals sometimes misrepresent what historical figures
actually believed. What’s important to us is not what people did or didn’t believe, however; but rather the distinction between two
broad approaches to the question of what sources of justification there are.
23
See (Popper 2001). Pollock (1974) holds this for a subset of intuitions.
24
See (Markie 2004/2017: §1.1); see also (Huemer 2005: §5, 111).
25
Cf., for example, Cummins’ claim that ‘[p]hilosophical intuition … is epistemologically useless’ (1998: 118). Fumerton says that
he has ‘no idea what these intuitions are supposed to be’ (1990: 5). And Mackie calls “the suggestion that moral judgements are
made … by… having an ethical intuition” a “travesty” (1977/1990: 38).

19
Timothy Williamson, for example, rhetorically asks, ‘What are intuitions supposed to be, anyway?’ (2007:
215), and concludes that there is no plausible candidate for a psychological kind answering to that term.
Philosophers should not, he says, talk about intuition, because doing so functions “not to answer questions
about the nature of the evidence on offer but to fudge them, by appearing to provide answers without really
doing so” (Williamson 2007: 220).

We shouldn’t refer to intuitions, Williamson thinks, because, in effect, there are none: there is no class of
mental states for such talk to refer to. The things we call intuitions belong to a motley class of phenomena
whose members have nothing significant in common.

Williamson is far from alone in thinking this (Ayer 1956/1964: 33; Cappelen 2012; Deutsch 2015;
Fumerton 1990: 6; Pollock 1974: 305). Here, for example, is Tara Smith:

[What] exactly is an intuition? One rarely encounters clear statements of their nature. If an intuition
is a thought, why employ a term suggesting it is anything less than that? If intuition is a particular
type of thought, what type? If an intuition is an emotion or feeling, what distinguishes intuition from
ill-founded feelings? . . . Are intuitions desires? Hunches? Stubborn convictions that a person refuses
to surrender? The point is, we cannot be sure whether we have such things, let alone what role they
play in providing moral guidance, until we know precisely what intuitions are. One suspects that the
absence of definition, keeping intuition afloat as a hazy “something” between a thought and a feel-
ing, may hide the fact that there are no such things. (Smith 2000: 23-4)

This book provides what Williamson, Smith and others take to be lacking: a clear statement of the nature
of intuition, an account which in detail vindicates the impression that the items on our list really do have
something significant in common. I will try to leave you in no doubt that you really do have such mental
states, and I will argue as clearly as I can that intuition as thus conceived really justifies belief. I hope, of
course, to convince you of the truth of every aspect of this story. But whether or not I succeed in doing so,
the reader at least will not be hindered from considering the justificatory status of intuition by the lack of a
clear grasp of its nature. There can be no further pretence of not knowing what intuitions are supposed to
be.

1.5 A New View

I have said that instances of intuition seem bound together by having representational content, causing be-
lief, apparently justifying belief, and by a distinctive phenomenal character; that a central question is
whether intuition not just apparently but in fact justifies belief; and that I will argue that it does.

But the claim isn’t merely that intuition justifies belief and has a characteristic phenomenal character. The
claim is that intuition justifies belief because it has that phenomenal character. The view that intuition has a
characteristic phenomenal character and justifies belief has been around for some time, but the claim that
the two are explanatorily connected is new.

Consider George Bealer, a prolific author on intuition, who regards intuition as a “sui generis, irreducible,
natural propositional attitude which occurs episodically” (1996a: n. 6; 1998a: 207; see also 2002: 74):

For you to have an intuition that A is just for it to seem to you that A. Here ‘seems’ is understood .
. . in its use as a term for a genuine kind of conscious episode. … [T]his kind of seeming is intellec-

20
 tual, not sensory or introspective (or imaginative). Intuition must be distinguished from belief: be-
lief is not a seeming; intuition is. . . . Similar phenomenological considerations make it clear that intu-
itions are likewise distinct from judgements, guesses, hunches, and common sense (1998b: 271-2,
emphasis mine). 26

Here and in many other places Bealer makes it clear that he thinks we can tell that intuition is its own psy-
chological kind at least in part because it has a distinctive phenomenal character. Despite this, he has next
to nothing to say about what that phenomenal character actually is. 27 He does say that ‘rational’ intuition
“presents itself as necessary: it does not seem to us that things could be otherwise” (1992: 102), but even
this turns out to not be a point about intuition’s phenomenal character, but about its content. 28 Moreover,
although he assigns intuition’s phenomenal character importance, Bealer never suggests that intuition’s
phenomenal character might be what explains its ability to justify belief, instead he advocates for ‘modal
reliabilism’; “the doctrine that there’s a certain kind of qualified modal tie between intuitions and the
truth” (2004: 13-14).

Or consider Michael Huemer, who defends a principle he calls ‘phenomenal conservativism’: “[i]f it seems
to S that p, then, in the absence of defeaters, S thereby has at least some degree of justification for believing
that p” (2007: 30; see 2001: 99 for a similar formulation). Among the seemings of Huemer’s concern are
‘intellectual’ ones; which are what I call intuitions. Huemer’s view is similar to Bealer’s in that he takes the
phenomenal character of a mental state to be among the characteristics that qualifies it as a seeming. 29 How-
ever, also like Bealer, Huemer says next to nothing about what that phenomenal character is, 30 and doesn’t
claim that the ability of seemings to justify belief rests on its phenomenal character, arguing instead that the
principle of phenomenal conservativism is self-evident and that resisting it is self-defeating (2001: 103-8;
2005: 99-101).

John Pollock (1974: 319-21) and Joel Pust (2000: Chapter 2) also hold that intuition is a mental state with a
characteristic phenomenal character, but these philosophers, too, have next to nothing to say about what
that character is, nor do they claim that intuitions justify in virtue of their phenomenal character.

On the theory I advance, phenomenal character plays a much more central role. An important historical
precursor for this view came with a couple of papers published nearly twenty years ago by James Pryor. In
‘The Skeptic and the Dogmatist’, Pryor (2000) suggested that perceptual experience justifies belief because
of “the peculiar ‘phenomenal force’” of such experiences (n.37). And a few years later he elaborated:

My view is that our perceptual experiences have the epistemic powers the dogmatist says they have
because of what the phenomenology of perception is like. I think there’s a distinctive phenomenology:

26
In an early article, Bealer identified intuitions with non-inferential beliefs, or with states “having a strong modal tie” with such
beliefs (1987: 300). Otherwise, however, Bealer’s view of the nature of intuition has been very stable. For similar characterisations,
see e.g. his (1992: 101-4; 1996a: 4-7; 1996b: 123-4; 1998a: 207-13; 1998b: 271-2; 2001: 3-4; 2004: 12-13; 2008: 190-1). In discussing
Bealer I usually give a single reference, but most of the characteristics Bealer attributes to intuition are discussed in several or most
of these places—indeed, the passages are often almost identical.
27
Bealer offers some indirect characterisation of intuition’s phenomenal character by mentioning mental states which he thinks dif-
fer from intuition in phenomenology, and cases where he thinks there is no difference (e.g. between those with synthetic and ana-
lytic content (1998a: 212), and between modal and non-modal ones (Bealer: 75)).
28
Bealer expresses uncertainty about how ‘presents itself as necessary’ should be understood, but suggests “something like this: nec-
essarily, if x intuits that P, it seems to x that P, and also that necessarily P” (1996a: 5). And that is to suggest that the content of the
relevant intuition is P & □P, and not to characterise its phenomenal character.
29
Huemer has confirmed this interpretation in personal communication.
30
In his 2001, he discusses the ‘forcefulness’ of perceptual experience, but says that this is not a matter of its phenomenal character
(what he calls ‘qualia’) (79).

21
 the feeling of seeming to ascertain that a given proposition is true. This is present when the way a
mental episode represents its content makes it feel as though, by enjoying that episode, you can
thereby just tell that that content obtains (2004: 356-57).

If the phenomenology of perceptual experience is what explains why perception justifies beliefs, mightn’t
the phenomenal character of intuitional experience furnish an argument that intuition does so too? This
idea was developed independently by Elijah Chudnoff, John Bengson, and myself, in PhD dissertations
completed in the years leading up to 2010-11 (Chudnoff 2011b, 2011a; Bengson 2010, 2015b; Koksvik
2011, 2013, 2017). Although these theories have a common point of departure there are, as we shall see,
important differences between them. In what might not be a complete surprise, I shall argue that these
other views suffer from significant structural problems.

Even though others have given some weight to intuition’s phenomenal character, they haven’t matched that
view with a detailed description of that character—indeed, they often haven’t described it at all. And they
have invariably failed to provide any kind of an argument that intuition really does have the relevant charac-
ter. Both these elements are important. Moreover, unlike some, 31 I think that significant progress can be
made here, despite some methodological challenges. A significant part of the book is therefore dedicated to
describing the phenomenal character of intuition in detail, and to arguing that intuition really has this char-
acter. We can see that intuition has the phenomenal character I say that it has not only through introspec-
tion and reflection on our own experience, but because of what it allows us to explain.

I claim not only that there are phenomenal similarities between perception and intuition, but that they are
significant: the phenomenal character perception and intuition share allows both states to provide the expe-
riencer with ground-level justification. Perception and intuition are thus in an important sense on equal
footing, epistemically speaking: there is a way to understand intuition where it is not, as is sometimes
claimed, “utterly different from our ordinary ways of knowing everything else” (Mackie 1977/1990: 38).
This conception of intuition also belies the dictum that intuitions are not “data of experience” (Bealer
2001: 3): there is, I argue, something which truly deserves the name intuitional experience. 32

A final point. Unlike all other philosophical accounts of intuition of which I am aware, I will argue that
there are no restrictions on the kinds of contents intuitions can have: anything you can believe you can also
intuit (§3.8). This opens up the intriguing possibility that, in addition to its role in philosophy and other
academic disciplines, intuition may play an important role throughout our epistemic lives.

We often have intuitions, in the sense developed here. When we do, the mere having of the intuitional ex-
perience can make the person justified in believing that things really are the way they seem to her to be.
Whether it actually does so in a particular instance depends on a number of things, not all of which will be
uncovered here. But, I will argue, it is not unlikely that intuition, as this psychological kind is understood in
this book, plays a pervasive and important role not only in various academic disciplines, but in our mental
and rational lives quite generally.

31
“[I]t would be wrong to deny the occurrence of states with such a phenomenology. While it might be nice to have a further under-
standing of it, I think that the combination of introspective ostension and distinction offered by Bealer is sufficient . . . . In fact, I’m
inclined to think that this is all that can be done and that the sort of state at issue admits, like a pure phenomenal color, of no fur-
ther analysis” (Pust 2000: 36).
32
An empiricist might acknowledge this, and yet hold that all justification ultimately stems from sense experience. However, such a
position is incompatible with the conjunction of the views that i) intuitional experience provides foundational justification (Chap-
ter 5), and ii) that intuition is not constrained by aetiology (§3.6), both of which I defend in this book.

22
Chapter 2 Reduction?
The theory advanced in this book is that intuitions constitute a psychological kind: a kind which cuts the
mind at its natural joints, and which can do significant explanatory work. In particular, I claim that careful
study of the phenomenal character of intuition and perception, and of the epistemic importance of that
character, reveals that the intuition and perception are equally good candidates for psychological kindhood,
and for justifying belief in their contents.

That perception constitutes a psychological kind which can do significant explanatory work is a dominant
view in philosophy. By contrast, the view that intuition is epistemically and metaphysically on a par with
perception is a tiny minority view.

Those who reject that view often say one of two things. As we have seen, some say that there is no mental
kind here at all. What we call intuitions are really something else: a motley class of different mental states
(Smith 2000), for example; or just the exercise of our capacity for judgement (Williamson 2007). Others say
that there is a kind nearby, but that it is one with which we are already familiar, and that the things we call
intuitions are at best a sub-class of that familiar psychological kind.

Views of the first kind are best countered by a positive theory of the nature of intuition, one which also ex-
plains how it can do significant explanatory work. This book gives such an account. But views of the second
kind are fruitfully met head on, and that is the business of the present chapter.

2.1 Doxastic Views

Some say that what we call intuitions are really just beliefs. 33 A standard argument is widely taken to show
that such doxastic views of the nature of intuition fail. But the standard argument overlooks a straightfor-
ward reply available to defenders of doxastic views. Below I outline the standard case and this reply, before
presenting what I take to be a new, fully general, and decisive argument against any doxastic view of the na-
ture of intuition, or of perception.

2.1.1 Motivation
Reductive views can seem attractive for a few different reasons. First, there is clearly a tight connection be-
tween intuition and belief (§1.2.3), and a simple way to account for this is to identify the two: intuitions
just are beliefs.

33
Views of this kind have been proposed by (Lewis 1983), (Plantinga 1993) and (van Inwagen 1997), and endorsed by (Williamson
2007), and I have often encountered sympathy with such views in conversation. See also (Cummins 1998) and (Ichikawa and Jarvis
2009).

23
Second, suppose you start out thinking that intuition is a grasp of objective reality (Bengson 2015a). That
can’t be the whole story, since a false intuition can’t be said to grasp reality. 34 A doxastic theory solves the
problem, since beliefs can be false.

Third, absent demonstration that a proposed new mental state kind can do explanatory work that otherwise
won’t get done it is reasonable to reject it. Belief and desire have passed this test: they are integral to folk-
psychological explanation and prediction, and folk-psychology is very successful.35 They are also ‘pure’ exem-
plars of directions of fit: 36 a belief is ‘successful’ if it fits the world, a desire is ‘successful’ if the world comes to
fit it. Other states are not ‘pure’ in this way: fear is actualised if the world comes to fit it, but well-founded if it
fits the world; hope is realised if the world comes to fit it, but realistic if it fits the world.

This might suggest that mental states are generally reducible to some mix of belief and desire: fear that an
avalanche will strike might be some degree of belief that it will combined with desire that it does not; hope
that stocks will rise might be a mixture of some degree of belief that they will combined with a desire that
they do. A natural thought is that intuition might be thus reducible. Since there is no obvious role for de-
sire, reduction to belief suggests itself.

A fourth motivation stems from epistemic concerns. Those who believe that intuition justifies belief would
like an explanation of why that is so. Many believe that intuition is used as evidence in philosophy, and
might wonder whether an account can be given that validates such use.37 Some think that the only mental
state which justifies belief in a way we understand is belief itself.38 If so, the natural view is again that intui-
tion is itself a belief, and that it justifies belief in the same way that belief generally does. 39

Finally, a doxastic account of intuition might also be motivated by broadly logical concerns: that a reductive
account best explains how intuition behaves, how we use it, and so forth.40

2.1.2 Taxonomy
That intuitions are beliefs seems to be a simple idea, but the view comes in a number of different varieties. We can distin-
guish between them according to whether intuition is reduced to a disposition to have a doxastic mental
state, or to the doxastic state itself; whether it is reduced to a doxastic state itself, or to the acquisition of the

34
George Bealer (1998a) argues that what he terms the ‘local’ fallibility of intuition is no bar to the hypothesis that intuition is
strongly modally tied to truth. The tie holds in rather special circumstances, however: “Human beings only approximate the rele-
vant cognitive conditions, and they do this only by working collectively over historical time” (202). Sosa (2007b: Chapter 3) dis-
cusses the fallibility of intuition and factive models. See also (Pollock 1974).
35
I regard this line of argument as decisive, but my purpose here is merely to explain a possible source of motivation for doxastic
views of intuition. For opposing views regarding folk-psychology, see e.g. (Churchland 1981). Sterelny (2003) is one of many who
argues that belief has earned its keep in this way, but he is more doubtful about preferences.
36
See Humberstone (1992) for discussion, and an account of the historical origins of the terminology.
37
For arguments that intuitions are used as evidence in philosophy, see e.g. (Pust 2000: Chapter 1); (Goldman and Pust 1998);
and (Climenhaga 2018). Bealer (1998a) argues that intuitions are part of our ‘standard justificatory procedure’. He has been
interpreted by some as referring to philosophers’ use of intuition (Earlenbaugh and Molyneux 2009: 91). However, he is in my
view much more naturally understood as claiming that intuition is part of a justificatory procedure that is standard in a wider
sense; viz. standard relative to normal human life and inquiry. For arguments that intuition is not used as evidence in philoso-
phy, see (Cappelen 2012); (Earlenbaugh and Molyneux 2009); and (Williamson 2004).
38
See Pryor’s discussion of the ‘Premise Principle’ (2005/2013). (Ghijsen 2014) also argues along these lines.
39
Timothy Williamson is clearly motivated at least in part by such considerations in his (2007). Richard G. Heck Jr. (2000: 507-8)
spells out this type of motivation for perception.
40
This motivation is operative in (Earlenbaugh and Molyneux 2009). These authors argue for a dispositional view (see below), but,
again, the motivation applies in either case.

24
doxastic state; to all-out, or partial, belief; and depending on the content of the reducing state (the same as
the intuition; or a function on that content). This yields the following taxonomy:

Figure 2.1: Doxastic Views of Intuition

I begin with views (A) through (H), which I collectively label ‘doxastic’ views of intuition. (We return to
views on the omitted branch below.) I shall argue that a single line of argument deals decisively with all dox-
astic views. But first I will show that the standard case against such views fails.

2.2 The Standard Case Against Doxastic Views

We begin with views of type (A), views according to which an intuition that p reduces to a belief that p. A
simple view of this type is:

Equivalence: □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 ↔ 𝑩𝑩𝑩𝑩𝑩𝑩)

25
Equivalence says that all and only those who intuit that p believe that p. Equivalence does not say that an
intuition that p is identical to a belief that p, that ‘intuition’ and ‘belief’ are synonymous, or that the con-
cept intuition is the same as the concept belief. But if any of these views are true, so too is Equivalence, so its
falsity establishes the falsity of all these views. 41

And Equivalence clearly is false. There are many things that I believe but that I do not intuit: that πr2 yields
the area of a circle, that the (northern) winter solstice is in December, that light travels faster than sound,
that nothing travels faster than light, and that if p, then ¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬¬ p, for example.

A natural next suggestion is that anyone who intuits that p believes that p, but not vice versa. An intuition
that p could then be taken to be a particular type of belief that p, so that an intuition that p reduces to the
conjunction of a belief that p with some other condition:

Ellipsis: □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 ↔ 𝑩𝑩𝑩𝑩𝑩𝑩 & … )

Clearly there are ways to fill in the blank that render the view false. The question is whether there are some
that render it true. Until we are told what is missing we can’t assess the view directly. But we can assess it
indirectly, via:

Entailment: □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 → 𝑩𝑩𝑩𝑩𝑩𝑩)

If Entailment is false, then (by modus tollens) Ellipsis is too, since the former is entailed by the latter.

Agents sometimes come to regard something they intuit as false. This is widely thought to show that such
simple reductive views as Entailment are false (Bealer 1992, 1998a, 1996a, 1996b, 2001, 2002, 2004, 2008;
Bengson 2010, 2015b; Chudnoff 2011b; Earlenbaugh and Molyneux 2009; Huemer 2001, 2005, 2007;
Kagan 1989; Katz 1981; Plantinga 1993; Pollock 1974; Pust 2000, 2012/2019; Sosa 2007b, 1996, 1998,
2006, 2007a; Williamson 2007). An oft-noted example is the naïve comprehension axiom of set theory:

. . . I have an intuition—it still seems to me—that the naïve comprehension axiom of set theory is true;
this is so despite the fact that I do not believe that it is true (because I know of the set-theoretical
paradoxes) (Bealer 1998a: 208).

Call this ‘the standard case’ against doxastic views. To evaluate it we need to know whether the naïve com-
prehension axiom is something we intuit but regard as false. So we need to know whether we intuit it, and
for that we need a formulation. Bealer doesn’t offer one. Moreover, on some common formulations it is
questionable whether we have the intuition. 42 However, I think most people do have the following intui-
tion:

41
Absent a reason to think that the properties of intuiting that p and believing that p could be necessarily coextensive but non-iden-
tical (a la that presented for having three sides and having three angles in (Sober 1982)) one might think that the truth of Equivalence
would justify credence in the identity of belief and intuition. But I won’t pursue this here.
42
For example: “For every predicate, there is a set of all and only the things to which the predicate applies”, or “To every intelligible
condition there corresponds a class: its members (if any) are all and only the things that satisfy the condition” (Sainsbury
1987/2003: 109).

26
 NCA: If anything which satisfies condition F satisfies condition G and vice versa, then the set of the
things which satisfy F is identical to the set of things which satisfy G 43

NCA is false, for from it is derivable the claim that for any F there is a set of all and only the things that sat-
isfy F, and from this Russell’s paradox follows. 44 What makes NCA such a good candidate for a counterex-
ample to Entailment is precisely that it is provably false. Becoming apprised of a proof that demonstrates
that a proposition is false seems very likely to cause an agent to believe that it is.

Consider therefore an agent who has the intuition that NCA is true and as a result acquires the belief that
it is, but who then learns the proof of its falsity. If NCA is to work as a counterexample to Entailment, two
things must be true:

(i) She keeps the intuition that NCA is true

(ii) She sheds the belief that NCA is true
I stipulate that a belief is shed if it is non-accidentally lost in an appropriate way. In this instance it means
that the agent loses her belief in NCA as a result of learning the proof that shows that NCA is false.

Are (i) and (ii) true? In my view, the answer for (i) hinges on considerations about the agent’s phenomenol-
ogy to which we return at some length below. I think the answer is ‘yes’, and I will assume this in what fol-
lows. But what should we say about (ii)?

That Entailment is false is usually thought to be demonstrated by the fact that agents sometimes come to
believe that p is false while still having the intuition that p. But this doesn’t yet constitute a counterexample
to Entailment, since coming to believe that a proposition is false is not the same thing as shedding a belief
that it is true. A defender of a doxastic view can simply insist that the person who learns the proof keeps her
intuition—on her view, her belief—that NCA is true, and also acquires the additional and contradictory belief
that NCA is false. She believes both NCA and its negation.

Mere reference to NCA and similar cases therefore doesn’t show that there really are cases of intuition with-
out belief. We have been given no argument for that conclusion, but merely been told to consider the cases
and come to agree. We need a stronger case: a real argument for why there must be cases of intuition with-
out belief. The next section provides one.

2.3 The Argument from Rational Criticisability

The key to demonstrating that Entailment is false is to recognise that agents who hold contradictory beliefs
are usually ipso facto—that is, for that very reason—rationally criticisable.

43
Or: if any F is a G, and any G is an F, then the set of the Fs just is the set of the Gs. In what follows I restrict the discussion
to NCA as stated here. If you find a different example more convincing—the conjunction of the premises in the Sorites para-
dox, perhaps—please substitute accordingly. Incidentally, if you think that the formulations discussed in the main text, or the
ones discussed in the previous footnote, are not accurately thought of as statements of ‘the native comprehension axiom’ of set
theory, then you may well be historically correct—thanks to David Ripley for discussion—and you should feel free to regard
‘NCA’ and the corresponding phrases as mere labels. In the present context, the phrase has come to mean something like ‘the
thing we intuit but that Russell’s Paradox shows is false’. For this reason, and because the details really don’t matter for the
point I am making—again, feel free to substitute—I stick with the traditional usage.
44
See (Koksvik 2011: p. 42, n. 64).

27
‘Usually’, because there might be exceptions. For instance, there might be cognitive ‘positions’ a person can
be in relative to a pair of contradictory propositions such that if you are in it you are not rationally criticisa-
ble for believing them. 45 There might also be other factors or circumstances that shield one from rational
criticisability. However, for NCA and its negation, one needn’t be in such a position, and no such circum-
stances need obtain. (If there are no such positions or circumstances, so much the better for this argument.)
Therefore, if Entailment were true, the agent who intuits NCA and believes not-NCA would be ipso facto
rationally criticisable. She is not. So Entailment is false, and so, too, is Ellipsis.

This simple argument is powerful. The central notion is rational criticisability. As we saw in Chapter 1, it is
highly plausible that the concept of justification is already deeply familiar to rational agents, even if they do
not label it that way. Exactly the same point holds for rational criticisability: indeed a central way of being
rationally criticisable is by believing something without justification. So applying this notion in this argu-
ment is a safe move indeed.

Regimenting the argument can help to make it apparent just how innocuous the premises are.

(1) All who concurrently believe both a proposition and its negation are either ipso facto rationally
criticisable, or shielded from criticisability by being in special circumstances

(2) Some people concurrently intuit NCA and believe not-NCA

(3) None of these are ipso facto rationally criticisable
(4) Some of these are not shielded by being in special circumstances

(5) So, some of those who intuit NCA and believe not-NCA do not thereby believe both a propo-
sition and its negation

(6) So, some who intuit NCA and believe not-NCA do not believe NCA
(7) So, it’s not the case that whoever intuits a proposition believes that proposition 46

Let’s go through the premises.

(1) is clearly true, and if we’re liberal about what counts as ‘special circumstances’, it is analytic. It presupposes
only that there are circumstances in which holding contradictory beliefs renders one open to rational criti-
cism, and that such circumstances are not too rare. One couldn’t retain rational criticisability as a useful
concept while denying this.

(2) might be more contentious. Phenomenology isn’t completely unaffected by acquisition of the belief that
NCA is false, and some are tempted to say that the intuition vanishes. I think that that is an overreaction. 47

45
Having a ‘compartmentalised’ or ‘fragmented’ mind are candidates, see (Stalnaker 1984: Chapters 4 and 5) and (Lewis 1986:
30-9; 1982). My interest here is in the core idea, not in the particular uses to which these authors put it. In particular, it is
plausible that one can be shielded from rational criticisability for believing contradictory propositions if the two beliefs reside
in different fragments or compartments.
46
For a formal version, see (Koksvik 2011: 45-6, n. 70).
47
To foreshadow, I think that what happens in cases such as this is (usually at most) that the intuition becomes less strong, without
vanishing. The account advocated in this book has a straightforward way to account for this, because on this view, one aspect of the
phenomenal character of intuitional experience comes in degrees. See Chapter 4 for all the details.

28
Moreover, (2) only requires that not all those who learn the proof lose the intuition as a result. This cannot
be denied.

(3) falls out of the notion of rational criticisability. No one is ipso facto rationally criticisable for concurrently
intuiting a proposition and believing its negation, just as no one is ipso facto rationally criticisable for a halfway
immersed oar looking bent to them while they believe that it is not. It is possible that one thereby falls short
of being rationally ideal, but that is a very different matter.

Some think that simply having the intuition that NCA is true renders one ipso facto rationally criticisable.48
I think that’s a mistake, but even if true it doesn’t show that (3) is false. From an agent being ipso facto ra-
tionally criticisable for intuiting NCA it does not follow that she is ipso facto rationally criticisable for intuit-
ing-NCA-and-believing-not-NCA.

Being ipso facto rationally criticisable for is a non-monotonic two-place relation. A two-place relation is monotonic
if, whenever A is thus related to B, then anything which entails B also stands in that relation to A; and non-
monotonic otherwise. For example, being entailed by is monotonic, since if p is entailed by q, then p is also
entailed by anything which entails q: q&r, for example. By contrast, being rationally supported by is non-mono-
tonic, since it is not true that if p is rationally supported by q, p is also rationally supported by anything
which entails q. That the partygoers have arrived might rationally support the party going well even if the
partygoers having arrived extremely drunk does not, even though the partygoers having arrived extremely
drunk entails that they have arrived.

Being ipso facto rationally criticisable for is non-monotonic. I may be ipso facto rationally criticisable for failing
to listen to a local’s advice before going for a hike in the mountains, but I am not ipso facto rationally criti-
cisable for failing to listen while wearing a bowler hat, even though the latter entails the former. (My ra-
tional, or aesthetic, criticisability for going on a hike while wearing a bowler hat is a separate issue.) I am ra-
tionally criticisable for failing to listen to the local while wearing a bowler hat, of course; but not ipso facto
rationally criticisable—I am not rationally criticisable for that very fact. Correspondingly, even if it were true
that an agent might be ipso facto rationally criticisable for intuiting that p, this would not indicate that she is
ipso facto rationally criticisable for intuiting-that-p-and-believing-that-not-p. She is not.

It is also clearly true, as (4) claims, that some cases of intuiting that NCA and believing that not-NCA occur
outside the circumstances, if there are any, that shield one from rational criticisability. For example, one
might be shielded from rational criticisability for believing a contradiction if the contradiction is very hard
to discover—perhaps you believe the premises of Peano arithmetic and also the negation of one of its theo-
rems, for instance—but this is not one of those cases. Perhaps even some believers of obvious contradictions
are not rationally criticisable. Whatever the correct account of these cases is, some notion of cognitive separa-
tion between the offending beliefs will have to play a key role: the agent is somehow barred from bringing
the two beliefs under rational scrutiny together. In our example there need be no such separation.

One might be tempted to deny (4) by claiming that one cannot help believing what one intuits. Ought im-
plies can, so it’s false that anyone who intuits NCA ought to not believe it, so they aren’t rationally criticisa-
ble for believing it.

48
See (Sosa 2007b). Sosa only aims to show that an intuition is rationally criticisable under certain conditions. To resist (3) on these
grounds one would need to show that all cases of intuiting NCA while believing not-NCA occur under these conditions.

29
But rational criticisability isn’t subject to ought-implies-can restrictions of this sort. A parent who has lost
his child may not be able to help believing that the child is still alive even though he knows full well—and so
believes—that the child is deceased. A person with a psychological illness may not be able to help believing
that her food is poisoned even though she has compelling evidence to the contrary, and so believes that it is
not. 49 If these people’s minds are not compartmentalised—and perhaps even if they are—they are rationally
criticisable for so believing, however psychologically impossible it may be to shed the beliefs. 50

Finally, if the reductionist simply insists that having an intuition shields one from rational criticisability
without explaining why this should be so, she is making a merely verbal claim. The concept of belief in play
here just does not allow for brute shielding from criticisability. One can be shielded by the contradiction be-
ing hard to discover, by being barred from bringing both beliefs under rational scrutiny together, and per-
haps in further ways. 51 But if there are further ways we require an explanation of why the shielding occurs.
To simply assert that it does is to change the subject. Premise (4) is true.

From these four premises it follows that Entailment is false: intuition does not entail belief. And from this
it follows that Ellipsis is false, too. 52

What’s the difference between my case and the standard one? The latter claims that certain cases directly
show that there is intuition without belief. Presenting no argument it simply gestures at the cases and relies
on us to accept its view about them. By contrast, I have presented a detailed argument demonstrating that
Entailment entails that people are rationally criticisable in situations in which we know they are not, and
that it must therefore be rejected.

I formulated the argument in terms of views of type (A), but it generalises immediately to views of type (E),
which say that an intuition that p is reducible to the acquisition of a belief that p. 53 If an agent who believes
that not-p intuits that p, and if she thereby acquires a belief that p, she would immediately thereafter be ra-
tionally criticisable. Since she is not, such views also fail.

2.4 Partial Belief

In the sixties and seventies David Armstrong and George Pitcher developed analogous views of perception
to the positions about intuition that we have just been discussing:

[P]erception is nothing but the acquiring of true or false beliefs concerning the current state of the
organism’s body and environment. (Armstrong 1968: 209)

49
Thanks to Weng Hong Tang for this second example.
50
See also n. 77 below. Everything I say is consistent with rational criticisability being subject to some ought-implies-can type re-
strictions. (Thanks to John Bengson for this point.) For example, it is plausible that we are not rationally criticisable for failing to
deduce all the theorems of Peano arithmetic largely because we cannot. I take the cases in the main text to show (at least) that there
is an exception to ought-implies-can restrictions to rational criticisability when it’s clear to the agent what rationality requires.
51
Gilbert Harman suggests that there may be situations where “the best response [to discovering an inconsistency in one’s beliefs]
may be to keep the inconsistency and try to avoid inferences that exploit it” (1986: 15). This claim is orthogonal to the issue at
hand, since being rationally criticisable for holding obviously contradictory beliefs is consistent with the best response all things
considered being to not revise one’s beliefs.
52
An alternative strategy says that upon learning the proof the agent doesn’t acquire the belief that NCA is false, rather she sus-
pends belief, and believes neither NCA nor its negation. However, it s very plausible that learning the proof will usually cause the
agent to believe not-NCA, and anyway all the argument requires is that some agent concurrently intuits NCA and (for whatever rea-
son) believes not-NCA.
53
Views of this type were advanced by David Armstrong (1968) and George Pitcher (1971) for perception.

30
 Sense perception is the acquiring of true beliefs concerning particular facts about one’s environ-
ment, by means of or by the use of, one’s sense organs. (Pitcher 1971: 65)

Against this view a precisely parallel argument can be mounted, based, for instance, on known perceptual
illusions, which are cases in which agents don’t believe what they see (Koksvik 2011: §2.6). In response to
this thought, Pitcher and Armstrong held that perception should sometimes be identified with a partial be-
lief instead of an all-out one, thus maintaining a correspondence between perception and a doxastic state.
One might think that a parallel move could save doxastic views of intuition, so it’s important to rule it out.
I first show how the manoeuvre fails for perception, and then make the point for intuition.

For simplicity, let’s understand a partial belief as a credence: a degree of belief specified by a real number in
the [0,1] interval, where 0 indicates certainty that the proposition is false, and 1 certainty of its truth.

In some cases when we disbelieve perceptual experience it’s because of the experience itself: a rock which
appears to undulate in the heat looks unreal. But in many cases nothing about the experience itself alerts us
to the illusion: there is nothing inherently ‘wrong-looking’ about a bent oar, for example. In these latter
cases, if perception is identified with credence we can demand that it not be very low. Indeed, a stronger
constraint would be reasonable: it should be high. But we only need the weaker version.

My credence that the oar is straight might be very high indeed. I might have run my hand up and down it
and placed rigid objects alongside it, I might understand how optics works, and so on. On the partial belief
account, and given the constraint, I would have credences in two contradictory propositions adding up to
(much) more than one. On standard views of rational constraints on credences I would thereby be ration-
ally criticisable. I am not. So perception can’t be reduced to partial belief.

Turning to intuition, the question is whether all and only those who intuit that p have a credence in p:

Equivalence (credence): □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 ↔ 𝑪𝑪𝑪𝑪𝑪𝑪)

As in the case of outright belief, it’s easy to come up with cases of having some credence in the proposition
that p without intuiting that p. But perhaps intuiting that p implies having a credence that p along with the
obtaining of some other condition:

Ellipsis (credence): □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 ↔ 𝑪𝑪𝒙𝒙𝒙𝒙 & . . . )

As before we cannot assess Ellipsis (credence) directly. But we can assess:

Entailment (credence): □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 → 𝑪𝑪𝑪𝑪𝑪𝑪)

If Entailment (credence) is false, then Ellipsis (credence) is false too, since the former is entailed by the lat-
ter. And Entailment (credence) fails in analogous ways to how Entailment fails. To bring this out we again
impose a reasonable constraint:

Correspondence: If intuition is to be identified with credence, whenever the intuition is strong, the
credence must not be very low

As before, a stronger constraint would be reasonable: the credence should be high. But we only need the
weaker version.

Consider NCA, discussed above. Many people have a strong intuition with that content. Fix on such an
agent, and assume she understands the proof showing that NCA is false. If Entailment (credence) is true,

31
she will, given Correspondence, have a credence in NCA which is not very low. She also has very high cre-
dence in not-NCA: she knows the proof, has consulted experts, and so on. She comes out as rationally criti-
cisable, since her credences add up to (much) more than one. But she is not rationally criticisable, so Entail-
ment (credence) is false.

This argument has concerned views of type (C), but it generalises immediately to views of type (G), which
say that intuition is reducible to the acquisition of a partial belief that p, for reasons given at the end of §2.3
above. Intuition is not reducible—wholly or in part—to the acquisition of partial belief, either.

2.5 Doxastic Attitudes with Different Content

What about views that identify intuition with a belief or a partial belief in a different content, or to the ac-
quisition thereof? To be even remotely plausible, such views must hold that q is some function of p: f(p).
However, regardless of what we take q to be, and therefore regardless of what the function f is, one can intuit
that p while believing that not-q without incurring ipso facto rational criticisability. Such accounts therefore
fail, with complete generality.

Consider the suggestion that an intuition that p is reducible to the belief I have some reason to believe that p.54
Suppose that for theoretical reasons I firmly believe that there are no such things as reasons at all, that I de-
duce from this that, a fortiori, there are no reasons to believe that p, and that I in this way come to believe: I
have no reason to believe that p. It is obviously compatible with this that I have the intuition that p, and
compatible without ipso facto rational criticisability. If my intuition were reducible as suggested, I would now
believe that I have some reason to believe that p and that I have no reason to believe that p, and so, since
this is an obvious contradiction, be ipso facto rationally criticisable. I am not, so the reductive account fails. 55

We can in fact always construct a case in which reductionists are committed to my being ipso facto rationally
criticisable but in which we know that I am not, because no matter what q is, I am never ipso facto rationally
criticisable for intuiting that p and believing that not-q. If intuiting that p entailed believing that q (for some
q) however, I would be. So, intuiting that p does not entail believing that q, for any q. A fortiori, intuiting
that p doesn’t entail believing that f (p), for any function f.

This shows that views of type (B) fail. Applying reasoning we have already gone through we see that intuit-
ing that p does not entail acquiring the belief f (p) (type F), or having or acquiring a partial belief that f (p)
(types D and H).

This completes the case against doxastic views.

2.6 Intuition as a Disposition to Believe

One might think that the objection from rational criticisability could be avoided if the relation between re-
duced and reducing state were loosened, and in particular if intuitions were identified with a disposition to
be in the doxastic state rather than with the doxastic state itself.

54
Such a position has been advocated by Christian Nimtz (2010).
55
It is possible that I can’t correctly believe that I have no reason to believe that p. I regret to report, however, that the fact that I
can’t correctly believe something has proven itself no bar at all to my actually believing it.

32
Such dispositional views of the nature of intuition come in different varieties. Some of the above distinctions
collapse: a disposition to acquire a belief that p is just a disposition to believe that p. But some carry across:
it might be a disposition to have an all-out, or a partial, belief, for instance. So we can complete our taxon-
omy of reductive views as follows:

A very simple view of type (I) would be the following:

Equivalence (disposition): □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 ↔ 𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫)

Where DBxp is read as saying that x is disposed to believe that p. As before it’s easy to come up with coun-
terexamples: there are many things I am disposed to believe but which I do not intuit.

Next, intuition might be thought of as a particular kind of disposition to believe:

Ellipsis (disposition): □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝒑𝒑 ↔ 𝑫𝑫𝑫𝑫𝑫𝑫𝒑𝒑 & . . . )

Until we’ve been told how to fill in the ellipsis, we can’t directly assess this thesis. But we can assess it indi-
rectly, via:

Entailment (disposition): □∀𝒙𝒙∀𝒑𝒑(𝑰𝑰𝑰𝑰𝑰𝑰 → 𝑫𝑫𝑫𝑫𝑫𝑫𝑫𝑫)

If Entailment (disposition) is false, then Ellipsis (disposition) is too, since the former is entailed by the lat-
ter.

To fix ideas, let’s consider a relatively detailed version of this latter type of view.56 In ‘Minimal Intuition’,
Ernest Sosa argues that a subject S has an intuition that p at time t iff:

a) If at t S were merely to understand fully enough the proposition that p (absent relevant perception,
introspection, and reasoning), then S would believe that p;
b) At t, S does understand the proposition that p; and

56
Timothy Williamson has made a similar proposal (2000, 2007), and dispositional accounts are also advocated by (Boghossian
2009); (Cohen 1981); (Earlenbaugh and Molyneux 2009); (Lynch 2006); and (van Inwagen 1997).

33
 c) the proposition that p is abstract (1998: 259). 57

A crucial notion here is mere understanding. As a matter of interpretation, I think that the bracket is best un-
derstood as explanation of this phrase. 58 A disposition to believe a proposition counts as an intuition just in
case mere understanding of the proposition constitutes the disposition’s conditions of manifestation.

One effect of this clause is to weed out false positives. Suppose I perceive that there’s a cup in front of me,
and that I am disposed to believe this. I understand the proposition, but I do not merely understand it, since
I have perception relevant to it. Were I to merely understand it, I wouldn’t believe it. So this doesn’t count
as an instance of intuition, as it shouldn’t.

On the other hand, we sometimes fail to believe a proposition when we ‘do more’ than merely to under-
stand it: when we reason our way to its negation, for instance. So the clause also weeds out false negatives:
cases where we would believe the proposition, except for the ‘more’ we do with respect to it. This allows the
unmanifested disposition to believe NCA to count as an instance, as it should.

Clause b) serves to avoid overgeneralisation: the implausible result that I intuit a number of propositions I
have never entertained in virtue of the fact that were I to understand them I would believe them. As for
clause c), Sosa doesn’t define what it is for a proposition to be abstract, but suggests that “abstract proposi-
tions abstract away from any mention of particulars”, though they might be “quite specific and determinate
in the properties or relations that they involve” (1998: 358).

2.7 Rational Criticisability Returns

Tom believes not-NCA on the basis of knowing the proof, but he also has the NCA intuition. According to
the dispositional account he thereby has the disposition to believe NCA in certain conditions C. Tom be-
lieves this account: he believes that were C to eventuate, he would believe NCA. He also believes, for what-
ever reason, that he could easily stamp out this disposition, but he takes no steps to do so.

Intuitively, Tom is ipso facto rationally criticisable in this situation. After all, by his own lights he is disposed
to believe something false, could easily amend that situation, but he does not.

When we describe Tom’s situation in a theory-neutral way, however, it is abundantly clear that he is not ra-
tionally criticisable. Tom has the intuition that NCA, believes that NCA is false, that he could take steps to
rid himself of the intuition, but he takes no such steps. He is not ipso facto rationally criticisable for this. So
the dispositional account is wrong.

Regimenting this we get the:

57
See also (Sosa 1996) for similar formulations. In his (2007b), Sosa says that intuitions “are not factors that attract us to assent . . . .
They are rather the attractions themselves. When such attraction is exerted by one’s entertaining a proposition, with its specific
content, then the attraction is intuitive” (Sosa 2007b: 54). Similarly, in his (2007a), Sosa argues that intuitions are conscious attrac-
tions to assent to propositions, that arise in a particular way. I’m not certain how to understand these proposals. However, on my
best understanding they constitute a variety of the dispositional view Sosa presents in the passages discussed in the main text. Chud-
noff (2011b: n. 4) also endorses this interpretation of Sosa (2007b).
58
Elsewhere, Sosa adds that the introspection, perception and reasoning is excluded “singly or in combination”, “even through the
channel of memory” (2006: 213; 2007a: 52). I take this as read throughout.

34
Second Argument from Rational Criticisability

(1) All who concurrently believe (i) that they are disposed to believe that p, (ii) that p is false, (iii) that
they can take easy steps to rid themselves of this disposition, but who (iv) take no such steps, are
either ipso facto rationally criticisable or shielded from rational criticisability by being in special cir-
cumstances
(2) Some people concurrently intuit NCA, believe not-NCA, believe they could take easy steps to rid
themselves of the intuition, but take no such steps
(3) None of these are ipso facto rationally criticisable
(4) None of these would become ipso facto rationally criticisable by coming to hold a true belief about
the nature of intuition
(5) Some of these are not shielded by being in special circumstances
(6) So, an intuition that p is not a disposition to believe that p 59

To see that (1) is true it may be helpful to think about analogous situations. Consider first a person dis-
posed to get angry with poorly dressed people. He eventually realises that he has this disposition, that his
reaction isn’t justified, and that he is harming people. He also believes that he could easily rid himself of the
disposition. But he doesn’t do that: instead he increases his efforts to surround himself only with well-
dressed people.

Second, consider two competing conceptions of freedom: as non-interference, or as absence of dominance. 60

If a slave master is benign or disinterested he may never impose on the will of a slave. He still dominates the
slave, however, since were the master to wish to so impose, he could. This is incompatible with freedom,
even if imposition is unlikely (Pettit 1997).

The lesson is that in normative matters, principle often matters even if practical effect is unlikely. Even if
our fashion-freak is unlikely to run into the poorly dressed, leaving the disposition in place when he (be-
lieves that he 61) could easily do otherwise makes him morally criticisable. Even if the master will almost cer-
tainly not impose, slavery is still an evil.

For parallel reasons, the agent who does not (attempt to) stamp out a disposition to believe a falsehood even
when he thinks he could easily do so is thereby rationally criticisable. By their own lights, those in (1) need-
lessly expose themselves to epistemic risk. This renders them rationally criticisable.

Turning to premise (2), the phenomenology associated with having the intuition that NCA is, as noted, not
wholly unaffected by coming to believe not-NCA, but the change is small enough to be consistent with the
intuition obtaining. Moreover, the account of intuition defended in this book accounts for and explains

59
Chudnoff (2011b: 12-14) and Bengson (2010: 14-15) mount arguments against the dispositional account of intuition based on
phenomenology. (And I have done the same (Koksvik 2011: §3.6), but I no longer defend that argument.) Bengson simply posits (in
the presentation of ‘the ardent physicalist’ (2010: 13)) the distinction between a conscious inclination to believe and an intuition,
so his discussion can do little to support it. Chudnoff relies on the claim that intuition has ‘presentational’ phenomenology (2011b:
17), a claim argue against below, so again this does not help. We need a separate case, like the one presented here.
60
Thanks to Nicholas Southwood for discussion.
61
If he is wrong about this he is morally criticisable for failing to make the attempt.

35
this fact. So it is very plausible that the first two conjuncts of (2) are true. Let the last two conjuncts be true
by stipulation. 62

As regards (3), above I argued that our ordinary understanding of rational criticisability dictates that no one
is ipso facto rationally criticisable for concurrently intuiting a proposition and believing its negation, just as
no one is ipso facto rationally criticisable for a halfway submerged oar looking bent to them while believing
that it is not. It seems clear that if a subject in addition believes that she can take steps to rid herself of the
intuition, but does not, she is not thereby ipso facto rationally criticisable. After all, if the subject attracts no
rational criticism for having the intuition while believing its negation, how could she be under any rational
obligation to rid herself of the intuition if she comes to believe that she can? She could not, so (3) is true.

There is room to think that a person in (2) falls short of being rationally ideal, but there is distance between
falling short of this ideal and being rationally criticisable. Second, it seems that what makes such a person
fall short of the rational ideal (if she does) is that the first conjunct is true of her: perhaps a rationally ideal
person has no false intuitions. Finally, even if, by not attempting to rid herself of the intuition she fails to
take steps towards the rationally ideal, it does not follow that she fails to take steps she is rationally required
to take: if she attracted no rational criticism for intuiting that p and believing that not-p she can’t be ration-
ally required to take such steps. The analogy with the halfway submerged oar is again helpful: a person who
doesn’t attempt to rid herself of this appearance may fail to take steps toward the rational ideal, but doesn’t
fail to take steps she is rationally required to take.

(3) is also not threatened by the possibility of being ipso facto rationally criticisable for having a false intui-
tion. I think one never is, but even if one could be it wouldn’t follow that the person in (2) is ipso facto ra-
tionally criticisable, as demonstrated above, since ‘being ipso facto rationally criticisable for’ is non-mono-
tonic.

(4) should now be unproblematic. If a person attracts no rational criticism for intuiting that p and believing
that not-p, adding a true belief about the nature of intuition can make no difference. A person who isn’t al-
ready rationally criticisable doesn’t become so when she acquires a true belief about intuition.

Premise (5) in this argument is the analogue of premise (4) in the argument discussed above. I argued at
some length for that premise there. The considerations largely transfer across, but it’s worth making two
quick notes. First, recall that a notion of cognitive separation would have to play a key role. There need be
no such separation in this case either, and typically there is none. Second, if Tom believed that the steps he
could take to stamp out his disposition partly consist in his F-ing, and if he believed himself to be prevented
from F-ing, he would plausibly be shielded. But in our example, Tom believes that he could take easily avail-
able steps to stamp out the disposition, so he is not shielded.

The premises in the argument are all plausible. If they are true, intuition is not a disposition to believe. So
the argument gives us good reason to think an intuition that p does not reduce to a disposition to believe
that p. Moreover, the argument is equally effective (mutatis mutandis) against views of types (J)–(L).63

62
The actual existence of such people is of course immaterial; what matters is that they could exist.
63
Simply change premise (1) so that, first, the subject believes that she has the disposition the account alleges that she has, and,
second, she believes that the relevant proposition—the one she would, on the account in question, end up believing if the disposi-
tion manifested—is false.

36
This completes the case against dispositional views.64

2.8 A Lesson about the Nature of Intuition

The reasoning in this chapter is, I believe, revelatory of the nature of intuition and perception, and of their
rational roles, in a way that goes beyond the mere production of counterexamples (Bratman 1987: 20). Ra-
tionality makes demands on our doxastic attitudes, inter alia on their coherence. It makes no such demands
on the combination of an experience with a doxastic attitude. There is no belief which renders a subject ipso
facto rationally criticisable when combined with an experience.65 I have argued that what holds for experi-
ences holds for intuition and perception: there is no belief such that a combination of the belief with the
state renders the subject ipso facto rationally criticisable.

I want to explain this by noting that doxastic states are rational commitment to their contents, but experi-
ences are not. This is a claim about the nature of these states: to be a belief that p is at least in part to be a
rational commitment to p. The explanation for why a person who believes that p and also that not-p—or who
hold corresponding credences adding up to more than 1—is rationally criticisable, is that she is committed
to something impossible: to things both being and not being such that p. By contrast, it is no part of the
nature of an experience to be a commitment to p. This explains the absence of ipso facto rational criticisabil-
ity where an experience combines with a doxastic state.

Dispositional accounts of intuition also falsely predict rational criticisability. I supported my argument to
this effect in part by noting that the subjects in the argument by their own lights needlessly expose them-
selves to epistemic risk. A disposition to believe that p in conditions C is a disposition to so believe if C
arise. That’s an epistemic risk if you also believe that not-p, because it then constitutes a disposition to be
committed to something impossible. No such thing is true for intuition and perception, and the explana-
tion is the same: a perceiver or intuiter does not expose herself to rational risk, because experiences aren’t
commitments to their contents.

I have argued that intuition and perception share an absence of commitment to their contents. I have also
argued that instances of intuition share a characteristic phenomenal character (§1.2.2). It is plausible that
instances of perception do, too. 66 Taken together these points strongly suggests a lesson about what percep-
tion and intuition are. Perception and intuition are experiences.

64
The argument applies with equal force to the case of perception. A person is not ipso facto rationally criticisable if an oar looks
bent to her, she believes that it is not bent, that she could take easy steps to rid herself of the relevant appearances, but doesn’t. So
the Second Argument from Rational Criticisability also shows that perception is not reducible to a disposition to believe.
65
Rationality also doesn’t require coherence between what a person supposes for the sake of argument and what she believes. I take
it for granted that to intuit or perceive that p isn’t to suppose for the sake of argument that p. For one, supposing for the sake of
argument that p doesn’t justify belief that p, not even apparently.
66
Both these points are explored in detail in Chapter 4.

37
Chapter 3 The Shape of the View

The theory advanced in this book has two main parts. First, a metaphysical claim: intuition is a certain con-
scious experience. Second, an epistemological one: intuition justifies belief. To some degree these claims are
treated separately: Chapters 2 and 4 focus on metaphysics; Chapters 5, 6, and 7 on epistemology.

It is important to realise, however, that the argument for the theory as a whole can’t be divided up in this
way. There is an overarching narrative about how metaphysics and epistemology fit together, the various
parts of which are mutually supportive. Here is that narrative in overview.

Perception justifies belief in what it represents. More specifically, perception provides a certain type of justi-
fication for belief, in a certain particular pattern. The best explanation of this is that perception justifies be-
lief because it is a conscious experience with a certain characteristic phenomenal character. Intuition is also
a conscious experience, and it shares the relevant aspects of its phenomenal character with perception.
Therefore, if nothing gets in the way, intuition also justifies belief in what it represents. Nothing gets in the
way, so intuition, like perception, justifies belief in what it represents.

This chapter fleshes out some of the elements of this narrative a bit further. The aim is to give you a better
idea of the shape of the view that I wish to advocate, so the discussion is conducted at a fairly high level of
generality, with most of the nitty-gritty set aside for later. Keeping both the shape of the view and the overall
argument in mind is useful when we get into the weeds later on, and will also allow readers who take issue
with certain parts of the story to still appreciate the value of the overall picture.

3.1 Intuition is a Conscious Experience

What kinds of things are intuitions? The previous chapter demonstrated that two popular answers to this
question are ruled out: intuitions are not beliefs, and they are not dispositions to believe. The reasoning in
that chapter also pointed to a positive lesson about their nature: just like perceptions, intuitions are con-
scious experiences. That intuition belongs in this metaphysical category is the first aspect of the shape of the
view advocated herein that I wish to draw your attention to.

What are experiences? I understand ‘experience’ to name a genus of psychological kinds that meet the fol-
lowing conditions.

First, many experiences, though in my view not all, have representational content. Perception and intuition
are both examples, and we cash out this talk in terms of the states’ truth or accuracy conditions (§1.2.1).

Second, to have an experience that p is not to be rationally committed to p (§2.8). In this respect, experi-
ences contrast sharply with doxastic states. This has various consequences, for example for how we evaluate
experiencing subjects with respect to rationality.

38
Third, content is among experiences’ identity conditions: a difference in content shows that we’re dealing
with two numerically different experiences. Fourth, phenomenal character is, too: a difference in character
is also enough to establish this (Chudnoff 2013a). 67

Fifth, experiences are such that the kind of thing they are also depends on phenomenal character: some dif-
ferences in phenomenal character suffice for a difference in kind. Experiences are thus mental states for
which phenomenology suffices for distinctness both in token and in type. Chapter 4 explores the phenome-
nal character of intuition in detail.

Some but not all experiences justify belief in their content. Perception is the go-to example of experiences
that do. Fear and hope are good examples of experiences that don’t. A central question for us is therefore
whether intuition in this respect is more like the former, or more like the latter.

3.2 The Justification Hypothesis is True

I began the book by asking you whether torturing the innocent is OK. This put you in a mental state that
has certain properties, one of which is that if you now believe that torturing the innocent is wrong because
you were in that state, that appears to be appropriate. The appropriateness at issue is the very same that is
operative in our ubiquitous practice of reflection on and evaluation of belief (§1.2.5). This amounts to say-
ing that intuition at least apparently justifies belief. But appearances might be deceptive: it is as yet an open
question whether intuition actually does justify belief.

Let’s call the claim that intuition does justify belief the Justification Hypothesis. Whether the Justification Hy-
pothesis is true or false really matters: if it is false we must significantly revise our epistemic practices (§1.4).
To assess the thesis it is useful to make it clearer.

To begin with, the Justification Hypothesis is of course not intended to say that just any intuition justifies
just any belief. Your earlier intuition about torture does not justify the belief that five plus seven equals
twelve. The hypothesis says that that intuition justifies belief in its content: an intuition that p justifies the
belief that p.

Second, intuitions aren’t free-floating phenomena; they are mental states, states of someone’s mind, and it
is persons who are, or fail to be, justified, in believing various propositions. When a subject S has the intui-
tion that p, that subject, S, is the only candidate for thereby being justified in believing that p.

Third, the Justification Hypothesis needn’t say that S’s having an intuition that p always justifies belief to be
interesting. As we saw in Chapter 1, simply establishing that intuition is a source of ground-level justifica-
tion, that it can justify belief, would itself be very significant.

Fourth, it is similarly not necessary that having the intuition that p fully justifies belief: showing that S can
be provided some justification for belief by having an intuition would still be an important result.

Taking all this into account, a more precise version of the Justification Hypothesis that remains useful for
our purposes is the following:

67
As noted, I take no stance on the relationship between representational content and phenomenal character, so for all I say there
can be differences in content between two experiences that don’t entail differences in phenomenal character, and vice versa.
Thanks to Elijah Chudnoff for discussion.

39
 Justification Hypothesis:
S’s having the intuition that p can provide S with some justification to believe that p

The second part of the shape of the view to which I want to draw your attention is that according to it, the
Justification Hypothesis is true.

3.2.1 A Lower Boundary

How much justification can the intuiter get? The Justification Hypothesis is not intended to allow just any
amount of justification, however miniscule. If it did, it would no longer be a distinctive and interesting hy-
pothesis. 68

A helpful way to think about what happens when I acquire new information, for example through percep-
tion, is that I locate the actual world more accurately in the space of all possible worlds. I do this by exclud-
ing as candidates for being the actual world all the possible worlds that are inconsistent with what I have
now learned (Jackson 2010: Lecture 2, especially 44-50; Stalnaker 1978). I am then left with a more re-
stricted set of possible worlds such that each of them may, for all I know, be the actual one.

Suppose I have a perceptual experience as of a hand in front of me, and let’s divide the possible worlds into
those in which there actually is a hand in front of me—the ‘hand worlds’—and those in which there isn’t—
the ‘no-hand worlds’. Given the way of thinking about new information I just outlined, if I can exclude
even one no-hand world as a candidate for being the actual one, then I have increased my justification to
believe that there is a hand in front of me at least a little.

As it turns out, everyone must acknowledge that having the perceptual experience excludes some no-hand
worlds as candidates for being actual; namely those in which my perceptual apparatus is working as it
should, and is appropriately connected with the world around me, but in which there isn’t a hand in front
of me. While many no-hands worlds are compatible with my experience—ones in which my perceptual appa-
ratus is malfunctioning or being interfered with, for example—those no-hands worlds are not. The perceptual
experience I’m enjoying excludes them, and by having it I increase my justification to believe that the actual
world is a hand world at least some small amount.

Similarly, everyone must acknowledge that having an intuition that p provides at least a small amount of
justification to believe that p; namely the amount which corresponds to excluding worlds in which my intui-
tional apparatus is working as it should and is appropriately connected with the way things are, but in
which it is not the case that p. While many not-p worlds are compatible with my experience—ones in which
my intuitional apparatus is malfunctioning or being interfered with, for example—those not-p worlds ones are
not. The intuitional experience I’m enjoying excludes them, and thereby increases my justification to be-
lieve that p some small amount. To remain distinctive and interesting, the Justification Hypothesis must say
that one gets more justification from having an intuition that p than the tiny amount which everyone must
allow that one gets.

3.3 Liberalism
Consider the following thesis:

68
Many thanks to Leon Leontyev here.

40
 Liberalism: For experiences of certain kinds, if certain conditions are met, then, absent defeat, S’s
having such an experience that p makes S to some non-negligible degree justified in believing that p

This thesis can be applied to intuition:

Liberalism about Intuition: If certain conditions are met, then, absent defeat, S’s having an intui-
tional experience that p makes S to some non-negligible degree justified in believing that p

Liberalism about Intuition—hereafter often just ‘Liberalism’—resembles the Justification Hypothesis, but dif-
fers from it. First, it makes the lower boundary of justification just discussed explicit. I’ll take this as read
from now on. Second, where the Justification Hypothesis says that having an intuition can justify a subject
in believing what it represents, Liberalism specifies that this happens when defeat is absent. Finally, Liberal-
ism makes a strong claim on which the Justification Hypothesis is silent, namely that having the intuition is
what makes S justified in believing what the intuition represents.

What does this last claim mean? Let’s say that a subject’s justification β to believe q is independent of her
justification α to believe p just in case S could have β even if she did not have α. A normal subject who has
an intuitional experience E that p, also has independent justification to believe a number of other proposi-
tions. An immediate consequence of Liberalism is that when relevant conditions are met, S’s having inde-
pendent justification to believe these other propositions is no part of what makes S justified in believing p. In-
stead, simply having E is what makes S justified in believing p.

Liberalism should be understood as a claim about the epistemic powers of certain types of experiences.
Some experiences are such that, if certain conditions are met, simply having such an experience can make a
person justified in believing what it represents. The experience does this all on its own, without ‘requiring
assistance’ from the justification the person has to believe any other proposition.

Here’s an analogy. It is no part of what makes me justified in believing that there are three pens on my desk
that I’m not distracted by a deafening noise. My visual perceptual experience is what makes me justified.
Still, not being distracted by a deafening noise is a necessary condition for being justified by the visual percep-
tual experience. This simply reflects a general distinction between necessary conditions, on the one hand,
and the things involved in making certain things so, on the other.

Let’s say that if Liberalism is true for experiences of a certain type, then experiences of that type singlehand-
edly justify the subject in believing the content of the experience. 69 Liberalism about Intuition says that intui-
tional experiences singlehandedly justify belief in their content.

Liberalism thus understood is separable from the claim that S doesn’t need to have justification to believe
other propositions in order to be made justified in believing that p by having E (Silins 2007). One can reject
that claim while still claiming that E singlehandedly justifies S’s belief. For example, a supporter of Liberal-
ism about Perception can hold that, when S has a perceptual experience that p, S’s having justification to
believe that she’s not a brain in a vat is among the conditions that must be met for her experience to make

69
My terminological choices here more closely follows Silins (2007) than Pryor (2004), but note that my use of the term ‘Liberalism’
differs from both Silins’ and Pryor’s uses. (It is not too far from Boghossian’s use in his (2009).) Silins adapts the term from Pryor,
but uses it for his own purposes.

41
her justified in believing that p. 70 But a supporter of Liberalism needn’t take that justification to be in any
way involved in making the subject justified.

Liberalism embodies a point made earlier, namely that justification is easier to acquire than to defend dia-
lectically. If simply having E makes S justified in believing p it follows that S may be justified in believing p
without (herself) being able to defend her belief against epistemic challenge, because the mental capacities
that are required for the latter far outstrip those that are required for the former.

A supporter of Liberalism could in addition defend:

Dogmatism: Liberalism is true, and S’s having independent justification to believe some other proposi-
tion is not among the conditions that must be met in order for S’s having an experience that p to make
her justified in believing that p

And here is that thesis applied to intuition:

Dogmatism about Intuition: Liberalism about Intuition is true, and S’s having independent justifica-
tion to believe some other proposition is not among the conditions that must be met in order for S’s
having an intuitional experience that p to make her justified in believing that p

According to Dogmatism, S’s having independent justification to believe some other proposition is not even
a necessary condition for her acquiring justification from having the experience.71 For example, when S has a
perceptual experience E that p, S’s having justification to believe that she is not a brain in a vat, or that her
experience is reliable, are not among the conditions that must be met for E to make her justified in believ-
ing that p. Dogmatists do say that S must lack justification to believe that she is a brain in a vat, however:
this and other defeaters for her justification must be absent.

Let’s say that if Dogmatism is true for experiences of a certain type, then experiences of that type immediately
justify the subject in believing the content of the experience.72 One can hold that certain experiences single-
handedly justify belief while rejecting or taking no stance on the view that they immediately do.

In Chapter 1 we saw that an important precursor to the present view of intuition is Pryor’s defence of Dog-
matism about Perception. One virtue of presenting Liberalism and Dogmatism as I have done here is that
doing so makes it very clear that Dogmatism is the stronger of the two theses, since it’s as a conjunction that
has Liberalism as its first conjunct.

I think the most fundamental insight from Pryor’s work on this topic is that certain experiences are epistem-
ically powerful. And while Pryor does defend Dogmatism (about perception), this insight doesn’t depend on
that view: it is wholly captured in the weaker thesis of Liberalism. Given that our interest is in investigating

70
Arguments for such views can be found inter alia in Cohen (2010); Davies (2000b, 2000a); Silins (2007); Wright (1985, 2000,
2002, 2007). A conservative view would on my usage say that a part of what makes an experience justify belief includes the perceiver
having some independent justification to believe other propositions (cf. Silins 2007: 111). Again, this differs from Pryor’s use
(2004).
71
Having justification to believe some other proposition might still be a necessary condition for S to be able to form the belief that p.
Consider the analogy: if having certain perceptual experiences is necessary in order to acquire certain concepts, it is not usually
thought to follow that reflection on these concepts cannot yield a priori justification.
72
It is a further question whether it being true that S is not a brain in a vat, or that her experience is reliable, are among the necessary
conditions for S’s having an experience to justify her in believing what her experience represents. Pryor: “Conservative and liberal
treatments of H may or may not also assign H a truth-requiring role” (2004: 354).

42
the possibility of extending that insight from perception to intuition it therefore makes sense to focus on
Liberalism.

For this reason I focus on the thesis that intuitional experience singlehandedly justifies belief in its content,
and bracket the further question of whether it immediately justifies such belief. The third aspect of the view
advocated here to which I wish to draw your attention is that it says that Liberalism about Intuition is true.

3.4 Phenomenalism
Liberalism is a thesis about the epistemic power of intuitional experiences. Phenomenalism accepts Liberal-
ism, but adds a further claim:

Phenomenalism: For experiences of certain kinds, Liberalism about such experiences is true be-
cause of the phenomenal character such experiences have

Phenomenalism about Intuition: Liberalism about Intuition is true because of the phenomenal
character intuitional experiences have

As before, a virtue of presenting Phenomenalism in this way is that it makes it very clear that Phenomenal-
ism is stronger than Liberalism. Phenomenalism entails Liberalism but not vice versa, since Liberalism
could be true for some other reason: one could hold that having a perceptual experience is what makes a
person justified in believing its content but that phenomenal character has nothing to do with it.

At the beginning of this chapter I outlined the overall argument this book advances. Restated in these new
terms we can say that the argument for Liberalism about Intuition is indirect, because it proceeds via an ar-
gument for Phenomenalism about Perception. That argument, in turn, is an inference to the best explana-
tion: the justification that perception provides is best explained by the phenomenal character of perceptual
experience. Since intuitional experience is relevantly similar to perceptual experience, and since none of the
differences between the two states get in the way, Phenomenalism about Intuition is true, as well.

Note that the claim is not merely that consideration of phenomenal character allows us to systematise a set
of previously established facts. Instead the claim is that consideration of phenomenal character both system-
atises a set of facts about justification many but not all of which are uncontroversial, but also that it
strengthens the case that justification in fact obtains in these cases. It is a mutually supporting structure.

There are other possible arguments one might advance in favour of the Justification Hypothesis, Liberalism,
or Dogmatism; for perception, intuition, or both, or perhaps for a larger class of seemings. One influential
line of argument says that rejection of the Justification Hypothesis faces epistemic self-defeat.73 I think this
line of argument has a lot going for it. It is compatible with and complimentary to the line of argument that
I pursue: if it works, it further strengthens the claim that intuition does justify belief, and thereby makes the
call for an explanation and theory of this fact—which this book provides—all the more urgent.

73
Bealer (1992) and Pust (2001) argue that blanket distrust in intuition faces epistemic self-defeat. Huemer argues via self-defeat for
what he calls ‘the rule of Phenomenal Conservativism’, according to which “if it seems to S as if P, then S has at least prima facie
justification for believing that P” (2001: 99; see also his 2007; and his 2009). BonJour argues that rejection of all a priori justifica-
tion, which he says only intuition can give, amounts to ‘intellectual suicide’ (1998: 4-6). For objections see DePaul (2009), Wein-
berg (2007), and Silva (2013).

43
A second line of argument in the case of intuition says that for a range of propositions we take ourselves to
be justified in believing, intuition is the only plausible explanation of why we are justified. 74 A weakness of
this line of argument is that it seems to underestimate the potential of inference to the best explanation as a
mode of argument. There are many facts to be explained, and from the epistemic position we’re currently
in—in which we are very far from having an all-encompassing understanding of the entirety of the way things
are—it is unclear that we can rule out that there are facts for which the relevant range of propositions forms
part of the best explanation. Still, I think this line too has a lot going for it, and if it works it has the same
effect as the one previously considered.

A third line of argument in the case of perception says that our concepts of everyday objects entail that
we’re justified. 75 And a fourth line says that the ‘irresistibility’ of perceptual beliefs gives us entitlement or
justification for these beliefs. 76

I think these two latter lines of argument are misguided. The former begs all the important questions be-
cause, if our concepts entail justification, the question just becomes why we are justified in believing the
propositions constituted by those concepts, instead of believing propositions constituted by concepts other-
wise exactly similar but that lack these entailments. For the latter, perceptual (or intuitional) beliefs are not
generally irresistible, as cases of known illusions make vivid for perception, and the NCA intuition and oth-
ers, discussed in Chapter 2, make vivid for intuition. Moreover, even if they were, this would at most buy a
very short reprieve from rational criticisability. Justification is much more than this. 77

At any rate, insofar as the explanatory project in this book succeeds, these latter two lines of argument are at
best superfluous. The fourth aspect of the shape of the view I defend is that the ability of perceptual and
intuitional experience to singlehandedly justify belief is explained by the states’ phenomenal character.

3.5 The Absent-Experience Challenge and ‘Overlookable’ Phenomenal Character

Consider the following commonplace example of perception:

74
Huemer (2005: 110-15) makes s this point in response to Mackie’s “queerness” objection to moral intuitionism.
75
See e.g. (Pollock 1974: 50).
76
Fred Dretske, for example, discusses whether a person is ever entitled to believe a proposition p in cases where there is no propo-
sition q which the person already accepts and to which she can appeal in support of p. He argues that we can come to realise we do
have such entitlement by focusing on the “psychological immediacy and irresistibility” of perceptual beliefs:
We have no choice about what to believe when we see (hear, smell, feel etc.) that things are thus and so. We experience
and forthwith believe. Between the experience and the belief there isn’t time to weigh evidence. The causal process . . .
runs its course before rational processes can be mobilized (2000: 598).
77
Even if Dretske is right that there’s not enough time for rational processes to come into play, not much would follow. Suppose I
have a perceptual belief about which it would be true that I ought to jettison it, were it not for the fact that I have not yet had time
to mobilise the required cognitive resources to do so; where to jettison a belief is to deliberately cause oneself to lose it. As soon as
enough time passes it becomes true that I ought to have jettisoned it. I’m rationally criticisable if I haven’t, even if at first I was
shielded by ought–implies–can style considerations. Dretske notes that we don’t directly control our beliefs the way we control our
limbs (2000: 604). That’s true but irrelevant since directness isn’t required: we can initiate processes which predictably will result in
the belief being jettisoned. Dretske acknowledges this, he agrees that we have indirect control over our beliefs (2000: 600), though,
he claims, only before having perceptual experiences. That’s a mistake: there are processes I can initiate both before and after having
an experience which will predictably lead me to fail to believe its content. So long as some such process is voluntary the simple irre-
sistibility claim is false. Finally, Dretske’s notes that there’s nothing an epistemically responsible agent can do to jettison her true per-
ceptual beliefs. That is most certainly true: I can jettison my belief, but I oughtn’t. But this is a mere restatement of the fact with which
we started, namely that perceptual experience justifies belief, and not an explanation of it.

44
 (1) There is a small cardboard box before Susan
(2) Susan has a visual experience as of a small, brown, cubical object
(3) Susan believes that a small, brown, cubical object is before her

In cases like this, it is natural to say that Susan’s experience mediates in two ways between her belief and the
way things are. First, the experience mediates causally: it is plausible that there is a causal relation between
the cardboard box in (1) and Susan’s experience (2), and between the experience and Susan’s belief (3). Sec-
ond, the experience mediates rationally: there is a rational relation between the experience and the belief. 78

We have noted that intuition appears to justify belief. Consider a paradigm example of this: Susan’s belief
that one plus one equals two. A significant challenge to the view that intuition actually does support belief
is the claim that in intuition there is no experience that is even a candidate for the role of rational mediation
between belief and the way things are:

(a) One plus one equals two

(b) . . . ?
(c) Susan believes that one plus one equals two

We might call this the Absent-Experience Challenge to intuition justifying belief in its content.79 Ernest Sosa is
among those who raise it. He argues that “no sensory experience mediates between fact and belief, nor does
anything like sensory experience play that role” (2006: 209). 80 Similarly, Timothy Williamson in several
places notes that intuition typically lacks the rich phenomenology of perceptual experience (2004: 117;
2007: 216-17). About the Gettier case he writes:

I am aware of no intellectual seeming beyond my conscious inclination to believe the Gettier prop-
ositions. Similarly, I am aware of no intellectual seeming beyond my conscious inclination to be-
lieve Naïve Comprehension, which I resist because I know better. … These paradigms provide no
evidence of intellectual seemings, if the phrase is supposed to mean anything more than intuitions
in Lewis’s or van Inwagen’s sense (2007: 216-17).

Since Lewis and van Inwagen take intuition to be belief, or a disposition to believe (Lewis 1983; van
Inwagen 1997), it is clear that Williamson takes the absence of rich phenomenology in the case of intuition
to show that there is no genuine intuitional experience. And others have raised the challenge in similar
ways. 81

The Absent-Experience Challenge to Phenomenalism is that skilful philosophers who have thought deeply
about the issue profess to not detect any intuitional experience in their own mental lives. That means that a
theory like mine, which places intuitional experience front and centre, must attribute to these thinkers a
significant mistake, namely that of overlooking a conscious experience that really is there, even while trying

78
Here I am following Ernest Sosa’s presentation (2006). The nature of the rational relation is not currently at issue; what matters is
that it is very plausible that there is such a relation. For interesting discussion see Richard Heck Jr. on McDowell (2000: 500-2).
79
(Koksvik 2011: §5.1). Chudnoff (2011b: §4; 2013a: §1.5) calls it ‘the Absent Intuition Challenge’.
80
See also Sosa (2007b: 46; 48; 54; 55; 62), and (2006: 209). In the latter he says that “there are no experience-like intuitive seem-
ings” (209). It is not always clear how important Sosa takes the phenomenology of intuition to be; see Koksvik 2011, p. 22.
81
Alvin Plantinga notes the difference between perception and intuition in his discussion of the phenomenology of the latter: “I
note nothing phenomenologically like, say, clearly seeing the color of Paul’s shirt (seeing it in sunlight, from up close, with an unob-
structed view), or seeing sunshine on the grass or water” (1993: 105). And Michael Lynch (2006: 228-30) argues that Sosa’s failing
to find an experience in introspection ought to dissuade him from taking intuitions to be attractions to believe.

45
to detect it. Because these people are skilled at their work, to so attribute is quite costly: a reader might eas-
ily think it more likely that the present theory is wrong than that these thinkers have made this mistake.

Not every challenge superficially similar to this one would be a real cause for concern. To appreciate the
force of this challenge it is worth being clear about certain aspects of the situation.

First, there are philosophers who quite generally claim not to understand what talk of phenomenology, or
of the subjective character of experience, is about.82 Now, we should all admit that it is difficult to speak
with precision about the phenomenal character of experience, and that we can often disagree about how
that character is best described. Still, this is not a challenge worth taking seriously. At any rate, it is not a
challenge I am prepared to take seriously. For one, everyone’s entitled to a starting point, and that there are
conscious experiences with phenomenal character is part of mine. But starting points aside, I think it can-
not rationally be denied that we have conscious experiences with distinctive phenomenal characters. To
throw the phenomenal-character-baby out with the difficult-to-describe-bathwater is not my idea of responsi-
ble theorising. So if the thinkers at issue in the objection currently under consideration were in general un-
willing to countenance phenomenal characters I think we should have little reason to worry: we would be
justified in concluding that the fault, whatever exactly it be, lies squarely with them, and not with us.

But that is not our situation. The force of the challenge rests in part on the contrast these thinkers point to
between the phenomenal character of perceptual experience—which they readily admit to notice and appre-
ciate—and what they take to be the absence of anything comparable in the case of intuition.

Now, the alleged absence of an experience is of course not the only reason for doubt about intuition’s abil-
ity to epistemically support belief about the way things are. Many hold that intuition cannot support belief
unless we can explain how we can be in contact with the way things are, and that realist construals of the
subject-matters of mathematics and logic, for example, make such contact unintelligible (Benacerraf 1973;
Devitt 2005; Dretske 2000). But the absent experience challenge arises regardless of what one takes the sub-
ject matter of beliefs supported by intuition to be: it is just as much a challenge, for example, to the view
that intuition provides us with justification for beliefs about our own conceptual structures (Goldman and
Pust 1998; Goldman 1999) as it is to the view that intuition provides us with justification for belief about
mathematical or logical reality. In that sense, it is a more fundamental challenge.

I do not, at this point, raise this challenge in order to answer it (I’ll get to that later), but to allow you to ap-
preciate a constraint on any Phenomenalist theory of intuition. Phenomenalism says that the phenomenal
character of intuition explains why it justifies belief in its content. But given that skilful philosophers who
have thought long and hard about the issue profess to not detect any phenomenology in these instances,
the phenomenalist can’t make reference to just any phenomenal character. The character the phenomenal-
ist claims is epistemically significant must allow us to understand how it could be that skilful philosophers
could come to overlook it. So, the fifth aspect of the theory to which I wish to draw your attention is that
according to the view on offer here, although intuition is a conscious experience in the full sense of that
word, its phenomenal character is still, at least in some intellectual contexts, ‘overlookable’: if one goes
looking for the wrong kind of thing, one can indeed fail to notice that it is there.

82
For example Daniel Dennett (1988).

46
3.6 No Etiological Restrictions
Before counting a mental state as an intuition, many thinkers place restrictions on the state’s causal history.
There are two broad types of such restrictions: the state must either have a particular history, or it must lack
one. A common positive etiological restriction is that the state must derive ‘from one’s understanding of one’s
concepts’ (Bealer 2004: 13; 2008: 191; Boghossian 2009: 119; BonJour 1998: 101). A widely accepted nega-
tive etiological restriction is that it must fail to result from conscious reasoning (See, for example Boghossian
2001: 636; Cohen 1986: 75-6; Gopnik and Schwitzgebel 1998: 77; Lynch 2006; Plantinga 1993: 106; Pust
2000: 44-5; but compare Huemer 2005: 101; see also Cappelen 2012: 33, 46).

The sixth aspect to which I wish to draw your attention is that on my view there are no etiological re-
strictions on which mental states count as intuitions: if a state has representational content and the right
phenomenal character, then it is an intuition.

Why are etiological restrictions so popular? Perhaps some think this is the only way to avoid epistemological
mystery: by tying intuition to concept-possession they try to show how intuition could play the epistemic
role they think that it does or should play. However, if the account I give is correct, that manoeuvre is un-
necessary. Intuition needn’t arise from concept possession for its epistemic role to make sense; explaining
how intuitional and perceptual experience are similar is enough. The positive account of intuition provided
here thus constitutes an argument against positive etiological restrictions generally: we have no reason to
impose such a restriction.

Similarly with negative etiological restrictions: one good way to show that there’s a psychological kind de-
serving of the name ‘intuition’ which can result from conscious reasoning is to provide a positive characteri-
sation of that kind without restricting on the basis of aetiology. If the kind thus delineated can do a suffi-
cient amount of interesting theoretical work—if it plays a sufficiently interesting epistemological role, let’s
say—then the claim has been vindicated. Exactly so, according to this book. So, just as with the positive vari-
ety, this book constitutes an argument against negative etiological restrictions generally.

One could also attack the various suggested etiological requirements directly. Since this book constitutes a
general argument against all such restrictions I won’t do that for all of them, but here is a case against the
very prevalent restriction of not having resulted from conscious reasoning.

It is of course true that one can come to believe a proposition p by reasoning one’s way to p without this in-
volving its seeming to one that p; if one believes that certain premises are true and that p follows from them,
for example.83 But why should we think that a process of conscious inference with p as its conclusion cannot
result in it seeming to the agent that p?

No one thinks that thinking about the concepts involved in p just before the mental state in question arises
disqualifies that state from being an intuition: pondering the logical connectives obviously doesn’t bar one
from then having the intuition that a double negation can be eliminated, or that one of de Morgan’s laws
holds, for example. Similarly for thinking about that law itself: as Bealer notes it is naturally most often pre-
cisely when you consider the law that it suddenly seems true to you (1992: 101).

So, one can clearly think both about the concepts involved in p, and about p itself, before having an intui-
tion; and this can be what causes the intuition to arise. Given this, it really is quite hard to see why one

83
See, for example, (Bealer 1992: 102). It is a separate and interesting question whether one must have an intuition corresponding
to each transition in a proof or argument, as Locke arguably thought; see (Locke 1689/1996: §§4.2.1–4.2.7).

47
would allow these thoughts to take any form whatever, except only the particular form of an argument. Such
a restriction would be ad hoc, so the clear presumption should be that this negative etiological restriction
does not apply to intuition. We would need an argument to deviate from this stance.

Joel Pust has presented such an argument: 84

[U]nless intuitions are non-inferential they cannot serve … as the ultimate premises in philosophical
argumentation and analysis. Philosophical practice treats intuitions as basic, as not admitting of fur-
ther inferential support, and this provides us with a reason for requiring of any genuine intuition
that it not be the result of conscious inference (2000: 45).

Similarly, L. Jonathan Cohen, to whom Pust attributes this argument, argues that “[i]f intuition is to pro-
vide the ultimate premises of philosophical argument, those premises should not themselves be the conclu-
sions of further reasoning” (Cohen 1986: 76). And it is reasonable to assume that the very widespread ac-
ceptance of the negative etiological requirement is in no small part due to (perhaps implicit) sympathy with
similar reasoning.

But let’s distinguish two senses of being ‘non-inferential’. In one sense, S’s intuition that p is non-inferential
if it is not the result of—in the sense of being caused by—conscious deliberation. In another, the intuition is
non-inferential just in case S’s justification to believe that p after having the intuition does not wholly rest
on the support p receives in virtue of being the conclusion of an argument.

To provide foundational justification, it is certainly true that intuition must be non-inferential in the sec-
ond sense. But why think it must be non-inferential in the first sense? I can think of no other reason than
the belief that the two don’t come apart.

But in fact, clearly they do. To see this, imagine that I don’t yet grasp de Morgan’s laws, and that you set out
to explain them to me:

Assume that it is not the case that p-and-q, which is to say that p-and-q is false. One way for that to happen is
if p is false. In that case, p and q are obviously not both true (we just said that p is false). And if p and q are
not both true, p-and-q is false. So one way for p-and-q to be false is for p to be false.

Naturally, another way to get the same result is for q to be false instead: the reasoning is just the same. And a
third way is if p and q are both false. But if p-and-q is false, one of these three things has to be the case: either
p is false, or q is false, or both p and q are false. There is no other way.

Now, not-p-or-not-q is true in exactly those three situations; when either one of p and q is false, or both p and q
are false. So, you see, if it’s not the case that p-and-q, then it is the case that not-p-or-not-q.

This may not be the snappiest of arguments, but it is an argument. It is valid, and it yields one direction of
one of de Morgan’s laws as its conclusion. In similar fashion you could have explained the other direction
to me. And it is surely at least possible that at the end of such patient explanations it comes to seem to me
that the transformation in question is valid—that, after all, seems to be point of the entire affair!

Had I been a little quicker I might have arrived at the point where I could ‘see’ that the transformation
holds simply by staring at ¬(p & q) ↔ (¬p ∨ ¬q) for a while. But my being able to ‘see’ this can just as

84
Pust no longer defends this argument.

48
well be the result of your explanation, a result of you arguing that it does, and of my following along in a
conscious reasoning process.

In such a circumstance, why should the value of my being able to see this depend on what took place just
before? If I really do see it, my justification does not rest wholly on the support it receives in virtue of being
the conclusion of an argument. It rests in part on the fact that I see it, that I have the intuition.

So a proposition which is the conclusion of an argument does not have foundational justification if it is jus-
tified only because it follows from justified premises, but its having foundational justification in virtue of
being the content of an intuition is not hindered if what caused the intuition to arise was a conscious rea-
soning process.85

The sixth aspect of the theory I advocate that I wish to draw your attention to is that, according to it, intui-
tion is not subject to any etiological restrictions.

3.7 Intuition is Wholly Permissive with respect to Content

The view advocated in this book is that intuition is a mental state which often causes belief, which appar-
ently and actually justifies belief, which has representational content and characteristic phenomenal charac-
ter, and that is not subject to any etiological restrictions. Moreover, I hold that intuition is wholly permissive
with respect to content, in this sense: any proposition a person can believe she can in principle intuit, and ab-
sent defeat she would thereby have some justification to believe it.

This is not a common view: I am, to my knowledge, its only advocate. 86 Instead, philosophers standardly
take intuition to be heavily restricted with respect to content, holding either a) that all intuitions can provide
justification, but that only states with certain contents count as intuitions, or b) that intuitions can have all
sorts of contents but that only a certain sub-class—for example, the ‘philosophical’ ones—provides justifica-
tion, and that the relevant sub-class is restricted by content.87

I think both kinds of views should be rejected, for much the same reason. Bealer is an example of an advo-
cate of a view of type b). He acknowledges intuitions other than the ‘rational’ ones (which are his focus), for

85
Knowing what caused the experience might defeat the justification, as Peter Singer thinks is the case for moral intuitions, for ex-
ample, but that is a different matter (Singer 2005).
86
That is, I am the only advocate of the view that intuition is permissive with respect to content and justifies belief. Proponents of
doxastic views of intuition may share the first half of this view, but they reject the second half. Thanks to a n anonymous reviewer
here.
87
For example, intuition is often thought to essentially involve necessity (Bealer 1996a: 5; Plantinga 1993: 105; BonJour 1985: 192).
Joel Pust also holds that ‘philosophical’ intuitions “involve an apparent necessity of some kind” (2000: 46), but doubts that all phil-
osophical intuitions have modal content. He suggests a weaker involvement of modality: a person counts as having a philosophical
intuition that p so long as “if S were to consider whether p is necessarily true, then S would have a purely intellectual experience
that necessarily p” (35-9). Pollock (1974, Chapter 4) seems to limit the content of intuition to logical truths. Others say that intui-
tions concern only abstract propositions (Sosa 1996, 2007b; 2007a, discussed in §2.6 above), or that the “the subject matter of intu-
ition includes abstract reality and excludes concrete reality” (Chudnoff 2013b: 11,227-28). Bengson’s view is a bit harder to pin
down. One the one hand, he criticizes the view that intuition must have necessary contents, or that it must regard explanation
Bengson (2010: §7). On the other, the selection of examples he uses to elucidate his account seems to indicate that he views intui-
tions as at least de facto limited with respect to content in fairly traditional ways (Bengson 2010: 8-9,11-13,18,24; 2015b: 707, 10-12;
2015a: 4). (I criticize ‘presentation’ accounts of intuition in §§4.8 and 5.3.2 below.)

49
example ‘physical intuitions’, such as that a house which is undermined will fall (e.g. in his (2001: 3)) 88. So
Bealer regards rational intuition as a sub-class of intuitions, a species of the genus. 89 Many other subclasses
could be singled out by content: Bealer discusses linguistic intuitions—intuitions about words and their
proper usage—and Huemer (2005: 102) delineates ethical intuitions as those whose contents are evaluative
propositions, for example.

The question is why we should think that each such subclass, or any one of them, corresponds to a psycho-
logical kind. Psychological kinds cut the mind at its natural joints. We’re justified in believing that we have
found one when we can characterise the members of that kind in a strongly unified way, and especially
when that characterisation also explains how the kind can do a sufficient amount of significant theoretical
work. But all the characteristics outlined in Chapter 1 apply to the intuition that a house that is under-
mined will fall, to linguistic intuitions, to intuitions with evaluative propositions as their content, and so
on. When it seems to a person that a house that is undermined will fall, or that “The boy the man the girl
saw chased fled” is ungrammatical, or that torturing the innocent is wrong; the mental state in question rep-
resentational content, it has the very same phenomenal character as the other examples we started with, it
seems to justify belief, and it often in fact brings about belief. This is very strong reason to reject the idea
that a sub-class of intuitions constitutes a psychological kind, in favour of the view that intuition is the psy-
chological kind in this vicinity, a kind which admits of all sorts of contents.

Again, we need an argument to deviate from this stance, and the mere existence of different contents is
clearly insufficient. We have beliefs about things natural and artificial, about things small or large, about
particular and general states of affairs, about the necessary and the contingent, about morality and language,
about well-known, everyday things as well as esoteric and far-fetched ones, and so on, and on, and on; but
all of this gives us precisely no reason to think that belief is divided into so many psychological kinds ac-
cordingly. This would commit us to psychological kinds restricted by content in absurd ways. There is no
psychological kind corresponding to the belief that libraries contain books, or to fear of bicycles, or to hope
for fair weather. 90

This reasoning applies with equal force whether the putative kind restricted by content is a sub-class, a ‘spe-
cies of a genus’, of whether it is instead the overarching kind. Just as it makes no sense to think that a sub-
category of intuitions that is demarcated by content corresponds to a psychological kind, it makes no sense

88
I think Bealer’s position is unstable: there is no room for him to coherently single out a sub-type of intuitions to which a positive
epistemic status may be adduced. One sign of this is in his (Bealer 1998a: 212) where he argues against a division based on a differ-
ent sort of difference in content, namely synthetic versus analytic. Even more tellingly, in his (Bealer 2002) he explicitly acknowl-
edges that “[t]here is no relevant phenomenological difference between modal and nonmodal intuitions” (75). He raises this point
in favour of attributing a positive epistemic status not only to non-modal intuitions but also to modal ones. But he apparently does
not see that this admission undermines his attempt at singling out ‘rational’ intuitions as a distinct subcategory, with distinct epis-
temic features, set apart, for instance, from ‘physical’ ones.
89
Bealer has confirmed this interpretation in personal communication. In what way is rational intuition singled out by content on
his account? By necessity being involved. Throughout his work, Bealer says that intuition ‘presents itself as necessary’. Elsewhere I
argue that if necessity (or ‘modal strength’, see (Koksvik 2011: n. 41)) were to be involved in intuition, it would have to be as part of
its content (Koksvik 2011: §1.5.1). Bealer himself (weakly) supports this interpretation: “I’m unsure how to analyze what is meant
by saying that an a priori intuition presents itself as necessary. Perhaps something like this: necessarily, if x intuits that P, it seems to
x that P, and also that necessarily P. But I wish to take no stance on this” (1996a: 5).
90
Michael Lynch raises a similar challenge: “If intuiting is a distinct kind of attitude, why can’t we, given the right circumstances,
take up that attitude towards almost any proposition, in the way that, given the right circumstances, we can find ourselves hoping or
fearing, or believing almost any proposition? Without argument, it is difficult to see how intuition would be restricted in a more
comprehensive way than other attitudes” (2006: 230).

50
to think that in order to count as an intuition at all, given that we’re thinking of intuition as a psychological kind,
a state can have only certain kinds of content. We have intuitions about all sorts of different things: general
and particular, natural and artificial, necessary and contingent, everyday and far-fetched, small and large;
about morality, language, aesthetics, danger, psychology, behaviour, nature, preferences, rationality, and on,
and on, and on, and on. They’re all intuitions. They should all be treated the same.

So the final aspect of the view to which I wish to draw your attention is that intuition, as that psychological
kind is conceived of here, is wholly permissive with respect to content.91

91
What is the epistemic status of my claims in §§3.6 and 3.7? I take these (and other claims in this book, in particular the claims
about the phenomenal character of intuition in Chapter 4) to be necessary truths about the nature of the natural kind that intui-
tion is. I also take these truths to be a priori. Actually having perceptual and intuitional experiences enables us to reflect upon them
and their epistemic roles, and thus to argue as I have. But this role is merely enabling, and a different possible creature might per-
fectly imagine these experiences, even if it did not experience them; and such a creature both could and should reach the same epis-
temic conclusions as I do here.
It remains possible that there be a posteriori necessary restrictions either on the aetiology of intuition, on the content of intuition,
or on both. (Thanks to an anonymous reviewer here.) But I see no reason to think that there are. There are obvious contingent
limitations on what we can in fact intuit with respect to complexity, just as there are on what we can in fact believe. But I see no
such limitations with respect to type of content, ruling out, for example, the concrete, the contingent, or the everyday, which is
what my opponents on this point need; no contingent ones, even—since on my view we manifestly have intuitions with such con-
tent—and a fortiori no (a posteriori) necessary ones. There are also contingent limits on our perceptual experiences: we do not have
vicarious perceptual experiences, for instance. That limit may even hold with nomological necessity: perhaps the laws of nature
make it impossible for me to literally see with your eyes. But no stronger necessity limits perceptual experiences to my own body: in
principle, I could see with your eyes, and the experience could still be perceptual in the fullest sense of that word.
Perhaps the strongest candidate restriction in this area is perceptual experience being restricted away from abstract content. But
first, even here our limitation might well stem simply from our contingent wiring rather than from anything deeper. And second,
even if there are a posteriori necessary limits to the kind of content perception admits, this does not give us reason to think that the
content of intuition is conversely restricted to exclude concrete content (or whatever), in the absence of independent reason to be-
lieve in a neat ‘division of labour’ between the two types of state—which we certainly do not have. And third, and to repeat myself,
we very clearly have intuitions with all sorts of different content, certainly going well beyond the abstract. So we have no reason to
believe in a posteriori restrictions here.

51
Chapter 4 Perceptual and Intuitional Experience

This book advances the view that intuition is a conscious experience which justifies belief in its content. We
have discussed content in several places already. This chapter is about phenomenal character.

According to the argument advanced in this book, the best explanation for why perception provides the
type of justification which it does provide, in the pattern that it provides it, is that it is a conscious experi-
ence with a certain phenomenal character. Since the epistemically relevant aspects of perceptual experience
are shared by intuitional experience, and since none of the differences between the two mental states get in
the way, intuition also justifies belief.

This chapter substantiates the that intuitional and perceptual experiences have aspect of their phenomenal
characters in common. Chapter 5 explains how those shared aspects are epistemically relevant.

4.1 Methodology
There is, in analytic philosophy, somewhat of an aversion to examining the phenomenal character of con-
scious experiences in detail, and an even greater aversion to placing theoretical weight on what such exami-
nations might reveal. 92 This pattern holds for intuition as well: even accounts on which phenomenal charac-
ter helps to distinguish intuition from other mental states usually have little or nothing to say about what
that character actually is. In my view, this is a mistake, but it has to be acknowledged that it is a mistake
made for good reason.

One reason is historical precedent. The introspectionist movement in psychology famously ended in irre-
solvable disagreement (Bengson 2010: 55), and present-day philosophers are understandably eager to avoid
a similar predicament. More generally, it is just really hard to improve our knowledge of the phenomenal
character of our experiences. This is due to the problem of missing methods (Koksvik 2015): we lack good and
authoritative methods for determining what the character of a given experience actually is.

In this book I use three complimentary methods in an attempt to move forward.

First, careful description: simply describing the character of intuitional experience in detail. This is challeng-
ing. Our language is underdeveloped, so the use of metaphor is often required. The aim is to assist you in
recognising the relevant phenomenal character in your own conscious experience.

The second method is phenomenal contrast. Elsewhere I have argued that phenomenal contrast arguments
consistently fail, and I do not present such arguments here. However, we can distinguish ostensive uses of
phenomenal contrast from argumentative ones (Koksvik 2011: §4.9.1; 2015: §6.3). Uses of the former kind
do not aim to rationally persuade us of anything, but, as with description, to assist in recognition: they are a

92
See §1.7 above. There are exceptions, for example, (Dorsch 2009; Horgan and Tienson 2002; Horgan and Timmons 2007;
Kriegel 2009, 2015; Nes 2012; Pitt 2004; Siewert 1998). A more general exception may be the so-called ‘feeling theories’ of emo-
tions. But even in this tradition there’s resistance: Goldie, for example, emphasizes how one’s options “are limited” when it comes
to ‘substantially characterise’ relevant phenomenal characters (Goldie 2002).

52
way of ‘pointing’ in a space where index fingers don’t exist. Unlike argumentative uses, I think that osten-
sive uses have a reasonably good chance of success.

Moreover, ostensive uses of phenomenal contrast are significantly more likely to succeed in aiding recogni-
tion of attitude-specific phenomenology than in doing so for content-specific phenomenology. The core of
my objection to phenomenal contrast arguments in favour of content-specific phenomenology of is that in
the cases these arguments use, a multitude of other contributors to overall phenomenal character invariably
also differ between the two cases. Each difference constitutes an equally good explanation of the existence of
a difference in overall phenomenal character between the two situations (the only legitimate explanandum
here), so we can never be warranted in inferring that the thinking that p and thinking that q themselves
make different contributions to phenomenal character by this method (Koksvik 2011: §§4.5-4.8; 2015).
This poses a problem for ostensive uses too: ostension can easily fail to focus the speaker’s and the listener’s
attention on the same aspect of the character of experience.

But nearly all these other differences exist because of the content of the state, and are just as likely to arise
when one considers that p for the sake of argument as when one thinks that p, hopes that p, and so on.
When two distinct attitudes with the same content are contrasted, on the other hand, there is a significantly
smaller likelihood that the true explanation of the difference in overall phenomenal character is that con-
founding contributors obtain in one case but not in the other. It is much more likely that the same ones
obtain in both cases, and so the existence of a difference in the character of overall conscious experience is
not as easy to explain away. To be sure, this only goes so far: it is probably more likely that an emotional re-
action arises if one believes that p than if one supposes that p for the sake of argument, for example. But the
difference is general and significant enough to make the use of phenomenal contrast for attitude-specific
phenomenology significantly more likely to succeed than uses aimed at content-specific phenomenology.

Putting these two points together we get that ostensive uses of phenomenal contrast for attitude-specific
phenomenology are much more likely to succeed than argumentative uses in general, and than argumenta-
tive uses for content-specific phenomenology in particular, even though the latter are, by a large margin, the
more common. Ostensive use of phenomenal contrast for attitude-specific phenomenology is what I employ
here.

These first two methods compliment each other and have the same goal: to allow recognition in your own
experience of the aspects of phenomenal character that, I claim, are there to be found.

The third method is inference to the best explanation: I will argue that perceptual and intuitional experi-
ence having the phenomenal character I say that they have best explains certain other facts. If it succeeds,
this line gives you a reason to believe my claims that is separate from and independent of recognising the
relevant aspects of phenomenal character in your own experience.

As with all abductive arguments, how well this line of argument fares depends on the extent to which the
purported facts are uncontroversial or can be established, and the extent to which a case can be made that
the explanation offered really is the only possible, or the best available. However, because the facts I explain
are widely accepted, a conscientious opponent cannot simply reply that they do not recognise in their own
experience the aspects of phenomenal character that I say are there. If the abductive arguments are success-
ful the opponent has reason to believe that the relevant aspects of phenomenal experience are there even if
they don’t recognise them. Absent a better explanation than the one I offer, the rational response would be
to put this down to inadvertently ‘looking for the wrong kind of thing’ (§3.5), at least for a time.

53
All of this notwithstanding, one could certainly wish for stronger arguments than what is available. Again,
we just don’t have strong methods that command widespread agreement. Abductive inference is accepted,
but it is seldom uncontroversial that the proposed explanation is the best one. An additional difficulty
where phenomenology is concerned is that the facts to be explained are often themselves under dispute.
The facts I claim to explain—about the transparency and content of experience—are in pretty good standing,
but no such case is bulletproof.

That does not mean that we should give up. We have excellent reason to think that understanding phenom-
enal experience is central to understanding the mind (see §6.4 below, and Koksvik 2011, §4.5; 2015, §1).
The dialectical situation is what it is. We must proceed with what we have and go as far as it can take us,
balancing intellectual humility and intellectual courage by paying heed to the challenges without despairing.
When, as in this case, distinct methods pull in the same direction, the chances for success are pretty good.

4.2 No Content-Specific Phenomenology

The phenomenal character of an experience is what it feels like to have it. In Chapter One we distinguished
between the character of persons’ global conscious experience—what it is like to be her overall—and the char-
acter of the various local experiences she is undergoing—what it is like for her to taste a fresh, crisp apple, for
example. A local conscious experience has a particular phenomenal character just in case it makes a particu-
lar contribution to the character of the global experience of the person who is having it. 93

Intuition is a local conscious experience, so the phenomenal character of intuition is the contribution that
having it makes to the character of the person’s global experience.

We can distinguish different ways in which the contribution from a certain mental state can be determined.
It might depend on the state’s content; it might depend on the type of mental state that it is; or it might de-
pend on both.

Consider perception. If perceiving something red makes a different contribution to the character of a per-
son’s overall conscious experience than perceiving something green does, perception has content-specific phe-
nomenology; and if not, it does not. On the other hand, it may be that whatever a person visually perceives,
the fact that she is having a visual perceptual experience makes a contribution to the character of her overall ex-
perience: perhaps a certain ‘visualness’ is contributed, for example (Grice 1962/1989). In that case, percep-
tion has attitude-specific phenomenology (Koksvik 2011: §4.1). A mental state type may have, and perception
actually does have, both attitude-specific and content-specific phenomenology: it is true both that perceiving
something red makes a different contribution to the character of a person’s overall experience than perceiv-
ing something green does, and that whatever I visually perceive, a certain ‘visualness’ is contributed.

The terms ‘content-specific phenomenology’ and ‘attitude-specific phenomenology’ are not perfect. For one,
the latter might make it sound like the phenomenal character suffices to distinguish one propositional atti-
tude (or mental state type) from another. As I will be using the terms this is a substantive question; indeed
it is crucial to the theory of intuition developed here that intuition and perception share aspects of their atti-
tude-specific phenomenology.

93
This claim is more permissive than it might at first seem, because it can be made true by surprisingly varied underlying metaphysi-
cal realities (Koksvik 2014).

54
The terms may also suggest that the ultimate origin of the phenomenology is in the content, or in the atti-
tude, respectively. The issues here are somewhat subtle, but on their intended reading the terms leave ques-
tions of origin open.

It is possible, for example, that a certain attitude only admits of content of a particular kind. As we have
seen, some think that perception is like this in that it only admits of non-conceptual content. Let’s suppose
that these people are correct, and that non-conceptual content always makes a different kind of contribu-
tion to the character of a person’s overall conscious experience than conceptual content does. But let us
also suppose, contrary to fact, that no-matter what the content is, the contribution is always the same. In
that case perception would have attitude-specific phenomenology but lack content-specific phenomenology,
on my usage. Thus the terms are intended to indicate variation with, rather than ultimate origin in, content
and attitude, respectively (though again, different attitudes can share aspects of their attitude-specific phe-
nomenal character).

The question I now wish to consider is whether intuition has content-specific phenomenology.

This book places intuition and perception on a par in important ways, and I take it to be beyond rational
dispute that perception has content-specific phenomenology. The conclusion that intuition does, too,
would therefore seem to fit my account most easily.

But that is just not where the arguments lead. In particular, i) intuition most likely has content-specific phe-
nomenology just in case conscious thought does, ii) conscious thought most likely does not, so iii) intuition
also most likely does not.

i) is supported by phenomenological considerations. Conscious belief is a kind of conscious thought. If in-

tuition had content-specific phenomenology but belief did not, coming to believe what one intuits should
be systematically associated with a loss of complexity in phenomenal character. Since there is no such loss, it
is unlikely that intuition has but conscious belief lacks content-specific phenomenal character. 94 Nor is
there a gain in complexity, so the opposite situation is just as unlikely. Intuition therefore most likely has
content-specific phenomenology just in case conscious belief does. Conscious belief has content-specific
phenomenology just in case conscious thought in general does for the same reasons: if this were not so,
passing the same content back and forth between different propositional attitudes should be associated with
systematic gain or loss of phenomenal complexity. There is no such gain or loss. So, intuition most likely
has content-specific phenomenology just in case conscious thought generally does.

Conscious thought most likely does not have content-specific phenomenology. First, if thought did have
content-specific phenomenology, it would have to be of a much subtler variety than in the case of percep-
tion. It is obvious that perceiving a red ball makes a different contribution to the character of overall experi-
ence than perceiving a blue ball does, but it is far from obvious that (non-iconically) thinking that Brexit is
an enormous mistake makes a different contribution to the character of a person’s global conscious experi-
ence than does thinking that the economic consequences of the CIVID-19 pandemic will be long-lasting.
Indeed, advocates of content-specific cognitive phenomenology usually explicitly acknowledge that content-
specific phenomenology of thought isn’t simply ‘there to be seen’ in the way that it is in perception.

94
The phenomenal character of intuition does differ from that of conscious belief, a difference I discuss in detail below, but the
difference is not one of complexity.

55
The elusiveness of the alleged content-specific phenomenology of thought establishes a presumption against
it. We need an argument to conclude that there is such phenomenology. If no argument succeeds, we have
good reason to believe that thought does not have content-specific phenomenology.

However, and as I argue in detail elsewhere, the two most important (Kriegel (2015) lines of argument in
favour of that view—namely that we need such phenomenology to know the content of our thoughts, and
that we can tell that there is such phenomenology through the use of phenomenal contrast arguments—both
fail (Koksvik 2011: §4.4; 2015). Overall, therefore, we have most reason to think that thought lacks content-
specific phenomenology, and, via i), to think that intuition does, too.

It might be objected that the question of whether intuition has content-specific phenomenology cannot be
addressed independently of assessing whole theories about the phenomenology of intuition. 95 In this objec-
tion we can discern two different claims. One has merit, the other does not, and neither threatens my view.

The first claim says that our view of the nature of intuition generally, and of content-specific phenomenol-
ogy specifically, must be sensitive to all the evidence that bears upon it, and that an important part of that
evidence is the relative fruitfulness of the overall theories which competing views of the nature of intui-
tional experience give rise to and support. Put differently, if a theory of intuition that in part is built on the
claim that intuition has content-specific phenomenology is significantly more fruitful than a theory built in
part on the negation of that claim, this fact counts strongly in favour of the former over the latter.

This claim has merit but does not constitute an objection to the theory defended here, since recognition of
this fact is a crucial part of my argumentative strategy in that theory’s favour: the views of the nature and
epistemology of intuition defended here constitute a mutually supporting structure.

The second claim says that one cannot consider whether intuition has content-specific phenomenology ex-
cept through such considerations of theoretical fruitfulness. That claim does not have merit. The fruitfulness
or otherwise of the resulting theory is an important part of the evidence that bears on a theory, but it does
not exhaust it. Direct phenomenological considerations, such as the ones presented here, are also relevant.
It is implausible, from direct consideration of intuitional experience, of conscious thought, and of their re-
lations, that intuition has content-specific phenomenology. This point could be trumped by general theoret-
ical considerations. It could be, but in fact is not, since it turns out that the fact that intuition lacks but per-
ception has content-specific phenomenology is no obstacle to placing intuition and perception epistemically
on a par (§5.5.1); and that the attitude-specific phenomenology the two states share explains both states’
ability to provide ground-level justification for belief (§§5.2 – 5.4). So, although they must indeed be consid-
ered in context, direct phenomenological considerations are still relevant.

Another reason to think that intuition lacks content-specific phenomenology is that it supports the most
plausible explanation of why, though intuition really is an experience, some people that sincerely look for it
still can’t find it in their own mental lives (§3.5).

Before turning to this point, a brief aside. Does the claim that intuition lacks content-specific phenomenol-
ogy commit me to the claim that what it is like to intuit that p is just the same as what it is like to intuit that

95
Thanks to an anonymous reviewer here.

56
q, for any p and q? It does. 96 That may seem implausible, but this concern relies on overlooking important
features of the situation.

Since intuition is a local conscious experience, if what it is like to intuit that p is the same as what it is like
to intuit that q, intuiting that p makes the same contribution to the character of overall conscious experi-
ence as does intuiting that q. But this is wholly compatible with the character of a person’s overall conscious
experiences when intuiting that p differing significantly from the character of her overall conscious experi-
ence when intuiting that q, both typically and in any particular pair of cases.

Our mental lives are characterised by Richness and Flux (Koksvik 2015): at any one time a number of
things contribute to the character of a person’s overall experience—occurrent, remembered and imagined
perceptions, moods, emotions and bodily sensations all play a role—and the contributions change often.

What a person remembers or imagines, and which emotional reactions she has, is almost certainly influ-
enced by whether she intuits that p or that q. A person who intuits that torturing the innocent is wrong will
imagine and remember different things than a person who intuits that if something is red it is coloured, she
will have different emotional reactions, and so on.

It is therefore very likely that the character of this person’s overall experience will differ in the two cases,
even though the contribution from intuition itself is just the same. For this reason, the view that intuition
lacks content-specific phenomenology does not commit me to the consequence that what it is like overall to
intuit that p will typically be just the same as what it is like overall to intuit that q. That is the counterintui-
tive consequence to be avoided, so there is no threat to the view here.

4.3 Answering the Absent Experience Challenge

Here are three questions that might well be asked of the theory in this book. First, if intuition lacks content-
specific phenomenology, in what sense is it really a conscious experience? Second, if it really is a conscious
experience, how can careful thinkers fail to recognise it in their own mental lives? And third, what would an
aspect of phenomenal character have to be like in order to make Phenomenalism come out true? As it turns
out, these questions are interlinked, and answering one of them will allow us to answer all three.

Let’s begin with the third question. Liberalism says that if background conditions are met, and absent de-
feat, having an intuitional experience makes the experiencer justified in believing what the experience repre-
sents. If, as Phenomenalism says, this is to be explained by the phenomenal character of intuition, then
every instance of intuition must have that character. After all, every instance of intuition has the potential
to justify belief in what it represents: whether or not it does hinges not on intrinsic differences between dif-
ferent instances of intuition, but rather on the extrinsic factor of whether defeat is present or not.

Different intuitions ‘say’ different things: they have different representational content. Phenomenal charac-
ter determined by that content would vary accordingly, as it does in the cases of perception and iconic imag-
ination, for instance. Such character therefore could not explain the epistemic power of intuitional experi-
ences. For that we need an aspect of phenomenal character that is stable between all intuitions. That is to
say, for an explanation of the epistemic power of intuitional experiences, attitude-specific phenomenology is

96
On the assumption that the two are of the same valence, and that the pushiness is equally strong in both cases. These issues are
discussed in detail in Chapter 5.

57
required. If Phenomenalism is true, and intuition singlehandedly justifies belief because of some aspect of
its phenomenal character, the phenomenal character in question must be attitude-specific.

One might object that intuition could justify belief because of its phenomenal character even if each in-
stance had a different phenomenal character; it’s just that the explanation would be different each time.
And it is true that some variation in character between instances of intuition is compatible with Phenome-
nalism; in fact such variation in my view plays an important explanatory role. But the variation cannot be
thoroughgoing. If it were, we would have to tell an entirely different story about how each instance of intuition
made the intuiter justified in believing what it represents. We would then not have explained how intuition
justifies belief in any useful sense of that word. On the intended interpretation, Phenomenalism about Intu-
ition would then be false. The only kind of phenomenology that can make Phenomenalism come out true
is the attitude-specific kind.

This brings us to the first question. That intuition lacks content-specific phenomenology might initially
cause one to question the claim that it is an experience properly speaking, but the fact that it has attitude-
specific phenomenology puts that worry to rest. Just as required by the characteristics of conscious experi-
ences outlined above (§3.1), differences in phenomenal character, depending on how significant they are,
suffice for distinctness both in token and in type.97

Finally, that intuition is a conscious experience with attitude-specific phenomenology but without content-
specific phenomenology allows us to answer the Absent-Experience Challenge to Phenomenalism (§3.5). As
we saw, Ernest Sosa and Timothy Williamson, for example, report not finding an experience in their own
mental lives when they reflect on cases of intuition. This challenge is forceful since these thinkers readily
acknowledge the phenomenal character of other conscious experience, for example that of perceptual expe-
rience.

That intuition is a conscious experience with attitude-specific phenomenology but without content-specific phe-
nomenology allows us to explain what happens in these cases. The reason these people cannot find the experi-
ence that really is there to be found is that they are looking for the wrong kind of thing. They are looking
for an experience with content-specific phenomenology, but correctly note that there is no such experience
in intuition (Sosa 2006: 209, 2013-14; Williamson 2004: 117; 2007: 216-17): there really is no experience
of comparable richness there to be found. However, to conclude on these grounds that there is no intui-
tional experience is unwarranted, since such experience could have attitude-specific phenomenology in-
stead. Exactly so, on this account.

Thus the same feature of intuitional experience that makes its epistemic power intelligible also explains its
status as a conscious experience properly speaking, and gives rise to a plausible answer to the absent-experi-
ence challenge to Phenomenalism.

4.4 Phenomenology of Objectivity

I now want to defend the following thesis:

Objectivity: Intuition and perception both have phenomenology of objectivity

97
A difference in degree of pushiness would suffice for a numerically distinct experience; but if pushiness is altogether absent the
mental state is no longer an intuition.

58
For an experience to have phenomenology of objectivity it must purport to represent objective facts, but
that is not all. That it so purports must be an aspect of the very phenomenology of the experience; and, in
particular, an attitude-specific aspect.

By ‘objective facts’ I simply mean facts that are independent of the subject: when I perceive or intuit, the
experience appears to tell me about the way things are independently of me. 98

Why accept this thesis? I give three reasons. First, because you can come to recognise the phenomenal char-
acter in your own experience, a recognition I hope to facilitate shortly. Second, because Objectivity explains
facts about the content of these experiences. Third, because it also explains another widely acknowledged
feature about perceptual experience, namely its transparency.

4.4.1 Recognising Objectivity

Consider the visual perceptual experience you are currently undergoing. The experience purports to be
about various things that surround you; one of them is this sheet of paper (or this screen). The experience
represents various of its properties: its shape, size, position, colour and so on. But your experience also rep-
resents that the world of which the paper is a part is not a figment of your imagination. It exists inde-
pendently of you, objectively. This is part of the very phenomenology of the experience. 99

Perceptual experience has content-specific phenomenology: perceiving a white sheet of paper makes a differ-
ent contribution to overall phenomenology than perceiving an orange sheet of paper does. But phenome-
nology of objectivity is attitude-specific, not content-specific, so it is not at the same ‘level’ as the sheet’s
whiteness. It is not, for instance, the phenomenology of a ‘marker’ attached to the sheet (and every other
experienced object), proclaiming it to exist objectively. It is instead an overall feature of perceptual phenom-
enology; a fact about the entirety of perceptual experience (in a modality). In the case of visual perceptual
phenomenology, the phenomenology is that of seeing an objective world, not of individual things being ob-
jective. It is a feature of the entire visual gestalt.

Phenomenology of objectivity is also present in other perceptual modalities. When touching the edge of a
table, an aspect of the tactile experience is very clearly that the perceived object is part of a world that is in-
dependent of the perceiver. In auditory perception, while determining the content of the experience is not
straightforward—perhaps that there is a source of sound nearby, perhaps something else—it is again clear
that the experience has phenomenology of objectivity: it is part of the what it is like to have it that the
sound source (or whatever) seems to be independent of the perceiving subject.

A salient contrast in the auditory case is between the beeping or humming sound sometimes experienced in
the absence of an external sound source, and a similar experience in the presence of one. (This is related to

98
Objection: this over-intellectualises, because animals and infants perceive but can’t conceive of independence from themselves.
Reply: to represent subject-independence one need not represent conditions under which this would obtain, and that they are met
(Burge 2009). Attributing that would over-intellectualise, but we don’t have to. The concept of self in adult humans is substantial,
but what’s required by the present account is quite minimal, and anyway required for different purposes. As Susanna Schellenberg
notes, for example, in the explanation of animal behaviour, a minimal concept of the self is already presupposed: “When a cat per-
ceives a chair . . . it sees something that is located in a certain relation to itself and something onto which it can jump. . . . Its loca-
tion in relation to the chair must figure in its perception for it to be able to flex its muscles so as to land on the chair” (2007: 620,
emphasis added). An animal’s concept of self may be nothing more than a relatively integrated collection of ingredients such as
these, ingredients we have independent reason to think exist, and the integration of which we can safely postulate given the evolu-
tionary advantages this plausibly yields.
99
Compare William Tolhurst: “Some seemings purport to be experiences of an object independent of the person having the experi-
ence” (1998: 300).

59
tinnitus, an affliction involving prolonged auditory experiences of the former kind.) Sometimes one cannot
tell whether the experience has an external source or not: one must check by asking others. But at other
times it is clear from the experience itself that there is no external sound source, and one feels no need to
check. In those cases the experience lacks phenomenology of objectivity.

Or consider the contrast between perception and iconic imagination; the kind which corresponds to percep-
tion in a sensory modality: visual imagination, auditory imagination, and so on. Iconic imagination is simi-
lar in various ways to perceptual experience. It has content-specific phenomenology, for example: imagining
a red cottage in a forest clearing makes a different contribution to the character of your overall conscious
experience than imagining a blue one does. But imagination lacks phenomenology of objectivity. If you im-
agine a small, red cottage in a forest, it is no part of your phenomenology that the cottage is a part of a
world that is independent of you.

You might object that visual imagination is so different from perception that we learn nothing from this
contrast. And it is certainly true that the difference between perception and imagination usually is stark.

For one, there is usually a difference in content. Visual imagination is often fleeting and indistinct: you are
more likely to be left with a vague impression of the ‘feel’ of the imagined cottage than with a sharp image
rich in detail. However, one can improve one’s capacity to visually imagine. It may be possible to imagine a
small, red cottage in a forest in as great detail as in a perceptual experience. And second, some perceptual
experiences have very simple content: with enough money (or a modern artist) to hand one could induce a
visual perceptual experience of a white, point-sized patch of yellowish light on an otherwise completely dark
surface about three metres in front, not flickering and approximately the strength of a candle. The contrast
can then be re-run between that visual perceptual experience, and the corresponding visual imaginative one.

A second difference between perception and iconic imagination is that iconic imagination is often volun-
tary, and some philosophers hold that taking oneself to perform an action voluntarily contributes to the
character of experience (Horgan, Tienson, and Graham 2003). But imagination need not be voluntary: wit-
ness how images one decidedly does not want to enjoy can be conjured up by conversation, for example. Im-
aginative experience can easily arise without voluntary effort, and so without any attendant phenomenology,
if such there be.

Suppose, then, that as I am out for a walk in the forest I spontaneously visually imagine, in detail as full and
rich as in normal visual perceptual experience, a small red cottage in a forest clearing, and then, as I crest a
small hill, I see exactly that scene. There would still be a difference in phenomenology between the two ex-
periences. This can’t be a difference in content-specific phenomenology—ex hypothesi there is none—so it
would have to be a difference in attitude-specific phenomenology. The difference would partly consist in
the absence of phenomenology of objectivity in the case of imagination and its presence in perceptual expe-
rience. The analogous point holds in the auditory case, and in the other cases.

Turning to intuition, consider a Gettier case.100 That Smith doesn’t know feels like a fact that is independ-
ent of you. The experience purports to represent an objective fact, and this is an aspect of the very phenom-

100
‘Gettier-cases’ are situations similar to those presented as counter-examples to an account of knowledge as justified true belief in
a widely cited paper by Edmund Gettier (Gettier 1963). Here is such a case. Smith is justified in believing that Jones owns a Ford,
because he has excellent evidence for that fact. He competently deduces that either Jones owns a Ford or Brown is in Barcelona
(this is a valid deduction), and is therefore justified in believing that proposition. Unbeknownst to him, however, Jones has sold his

60
enology of the experience, a part of what it is like to have it. Or take the intuition that torturing the inno-
cent is morally wrong. When it seems to you that this is so, it is part of the phenomenology of the experi-
ence that this is so objectively speaking, independently of you. If you have the intuition that people gener-
ally prefer less pain to more, it seems to you that most people really do, objectively speaking, so prefer. It is
part of the phenomenology of the experience that this is a subject-independent fact. And so on.

For a useful contrast in the case of intuition, recall that not all mental states that we refer to using the
‘seeming’ locution are intuitions. Suppose someone were in a mental state that she described as it seeming
to her that cold, bright autumn days are better than warm, overcast ones. It need not be a part of her phe-
nomenology that this is the way things are objectively speaking: this ‘better’ might not seem to be a matter
of the way things are objectively speaking, but more like a preference of her own. 101 Similarly, it seems to
me that all free-climbers recklessly endanger their lives, but this does not seem to be the way things are ob-
jectively speaking. Plausibly, this is an output of my irrational fear of heights, but one that hasn’t quite pen-
etrated deeply enough to make it seem to me that that is the way things are independently of me. Finally, it
might seem to you that siblings ought not to sleep with each other even if they cannot conceive, are both
consenting adults, and so on, without it seeming to you that that is the way things are objectively.

Of course, a person might instead intuit that siblings ought not to sleep with each other, or any of the other
things. 102 If so, this would have the phenomenology of objectivity. But it is also possible that things don’t
seem that way to begin with, and it is with that state the informative contrast holds.

Among the best reasons one can have to believe that there is phenomenology of objectivity in perception
and intuition is recognising it in one’s own experience. I hope I have now managed to facilitate this recogni-
tion in you.

4.4.2 Objectivity and the Content of Perception and Intuition

A second reason to accept Objectivity is that it explains features of the content of perceptual and intuitional
experiences: both that objectivity is part of the content, and more detailed facts about how it figures there.

That objectivity is part of the content of perceptual experience is independently plausible. 103 For a percep-
tual experience to be accurate things have to be a certain way. On the notion of content we are using, what
the perceptual experience represents, its content, is that things are that way. It is very plausible that objectiv-
ity is part of the content of perceptual experience, on this notion. For there is a very strong intuition that if
there is no objective world, if the world is but an aspect of my mind, then my perceptual experience is inac-

Ford, but it just so happens that Brown is in Barcelona. So Smith has a justified true belief that either Jones owns a Ford or Brown
is in Barcelona. But he doesn’t know that either Jones owns a Ford or Brown is in Barcelona, so knowledge is not the same as justi-
fied true belief.
101
I don’t mean to suggest that it’s marked as ‘just my preference’; rather, the absence of objectivity can make it feel that way, since
we are so familiar with intuitional experiences with phenomenology of objectivity.
102
Compare Maurice Mandalbaum, who argues that “the demands which appear to an agent to be ‘moral demands’ are seen by him
to be objective and independent of his desires” (1955: 57). And earlier: “[A] demand is experienced as a force. . . . It is my conten-
tion that the demands which we experience when we make a direct moral judgment are always experienced as emanating from ‘out-
side’ us, and as being directed against us. They are demands which seem to be independent of us, and to which we feel that we
ought to respond” (1955: 54).
103
Frank Jackson takes there to be ‘a causal element’ in the content of perceptual experience: “When I hear a sound as being, say,
behind and to the left, my experience represents the sound as coming from this location” (2003: 270). I take Jackson’s view about
the content of perceptual experience to entail, but not be entailed by, my view of it.

61
curate. I am currently having a visual experience as of a computer screen, a messy desk, and so on. If objec-
tively speaking there are no desks, no screens, and no mess, that suffices to show that my experience is inac-
curate, regardless of what else is going on.104 Objectivity is part of the content of perceptual experience.

My perceptual experience represents that things are a certain way, but also that they are this way inde-
pendently of me; objectively. This is well explained by Objectivity. Because perceptual experience has phe-
nomenology of objectivity, the experience itself ‘tells me’ that the way it represents things as being is part of the
way they are objectively. Because perception has this phenomenology, that is how it seems to me in percep-
tion, and this is reflected in the state’s content.

It is similarly plausible that objectivity is part of the content of intuitional experience. For an intuitional ex-
perience to be accurate things have to be a certain way. What the experience represents is that that is the
way things are. And there is a very strong intuition that if there is no objective way things are with respect to
the subject matter of a given intuition, then that intuition is inaccurate.

Take my intuition that if p, then not-not-p. If there is no way things are with respect to logic independently of
me, that suffices to show that the intuition is inaccurate. If there are no objective moral truths, my intuition
that torturing the innocent is wrong is inaccurate. My intuitional experience represents that things are a cer-
tain way, but also that they are this way independently of me; objectively.

This is well explained by Objectivity. Because intuitional experience has phenomenology of objectivity, the
experience itself ‘tells me’ that the way it represents things as being is part of how things are, objectively speak-
ing. Because intuition has this phenomenology, that is how it seems to me in intuition, and this is reflected
in that state’s content.

Objectivity thus explains that subject-independence is part of the content of intuitional and perceptual ex-
perience. 105 It also explains more detailed facts about how objectivity features there.

Compare two accounts of the content of a perceptual experience as of a white cup. On one account, the
content may be glossed as: a subject-independent cup subject-independently instantiates the subject-independent prop-
erty of whiteness. On the other, the gloss is, objectively: the cup is white. 106 The latter is more plausible than the
former, and this is well explained by the present account. Phenomenology of objectivity is attitude-specific:
a feature of the visual perceptual experience overall. It cannot plausibly give rise to content such as the for-
mer. But it can give rise to content such as the latter.

Similarly for intuition. On one account the content is glossed as a subject-independent person subject-inde-
pendently standing in the subject-independent relation of torturing a subject-independent person who subject-inde-
pendently counts as innocent, is subject-independently wrong. On the other, the gloss is: objectively, torturing the inno-
cent is wrong. The latter is more plausible than the former, and this is well explained by the present account.
Phenomenology of objectivity is attitude-specific: a feature of intuitional experience overall. It cannot plausi-
bly give rise to content such as the former. But it can give rise to content such as the latter.

104
Another possibility is that objectivity is merely entailed by the content, for example by being ‘built in to’ our concepts of ordinary
objects, though not a part of the content proper: perhaps something just doesn’t qualify as a table unless it’s objective. I think my
claims could be translated into these terms; but also that objectivity is really in the content proper. That we seem to see a subject-
independent world in perceptual experience is widely acknowledged, and arguably reflected in the popularity and intuitive pull of
such positions as direct realism and disjunctivism in the philosophy of perception.
105
Why is it not, then, content-specific phenomenology? Because ‘content-specific phenomenology’ indicates (potential) variation
with, and not ultimate origin in, attitude as opposed to content, as discussed in §4.2, above.
106
Thanks to Nicholas Silins for helpful discussion here.

62
4.4.3 Objectivity Explains Transparency
A third reason to believe that perceptual experience has phenomenology of objectivity is that it explains an-
other widely accepted fact about such experience, namely its transparency. Here is an accessible and repre-
sentative statement by Michael Tye:

Focus your attention on a square that has been painted blue. Intuitively, you are directly aware of
blueness and squareness as out there in the world away from you, as features of an external surface.
Now shift your gaze inward and try to become aware of your experience itself, inside you, apart
from its objects. Try to focus your attention on some intrinsic feature of the experience that distin-
guishes it from other experiences, something other than what it is an experience of. The task seems
impossible: one’s awareness seems always to slip through the experience to blueness and square-
ness, as instantiated together in an external object. . . . [I]ntrospection does not seem to reveal any
further distinctive features of the experience over and above what one experiences . . . . (Tye
1995) 107

Let’s begin by noting that this is a description of what it is like to have a visual perceptual experience: the
transparency datum, to the extent that it is a datum, concerns the phenomenal character of experience. Sec-
ond, this description makes the datum sound stronger than it actually is: numerous examples show that we
can, although we typically do not, focus on aspects of our experience. 108 I needn’t belabour the point, how-
ever, since it is clear that Tye has overlooked the possibility of attitude-specific phenomenology entirely.

Tye asks us to focus on a feature of an experience which “distinguishes it from other experiences”. But if
objectivity and pushiness (on which more below) are, as I claim, attitude-specific aspects of perceptual phe-
nomenology, they don’t distinguish one perceptual experience from another: the claim is precisely that all
perceptual experiences share these features. There may even be aspects of phenomenology which are shared
by all phenomenal experiences simpliciter, perceptual or non-perceptual.109 Either way, an inability to focus
on a feature of a perceptual experience which distinguish it from other perceptual experiences cannot possi-
bly demonstrate the truth of transparency across the board, on pain of ruling out attitude-specific phenome-
nology by fiat.

I think Tye is wrong even about content-specific phenomenology, but even were he right that we cannot be-
come aware of any content-specific features of our experience, nothing would follow about attitude-specific
phenomenology, because transparency may hold for the former but fail for the latter.

Indeed, Tye seems to presuppose the truth of Objectivity when he says that “[i]ntuitively, you are directly
aware of blueness and squareness as out there in the world away from you, as features of an external surface” (my
emphases). Again, this is a statement about the phenomenal character of experience, since that is what the
transparency datum is a datum about (to the degree that it is a datum at all). About that Tye is right: it is a
feature of perceptual experience that what is perceived seems to be ‘out there’ in the subject-independent
world. This is a fact about the phenomenal character of perceptual experience, exactly as Objectivity says.

107
See also (Harman 1990: 667).
108
See Amy Kind (2003) for forceful arguments for this conclusion. In addition, you can become aware of the ‘visualness’ of your
visual experience, and mutatis mutandis for the other perceptual modalities (Grice 1962/1989; see also Crane 2001: 144). You can
become aware of the auditory part of your experience dominating your overall experience, or the visual part dominating. If you shift
your attention to the slight humming of the computer, back to the whiteness of the computer screen, and so on, back and forth,
you can become aware that your overall experience has a ‘pulsating’ character. But you’re not aware of anything pulsating. Nothing
is, nor does it seem to you that way. And so on, and so forth.
109
Uriah Kriegel (2007: 129) suggests two candidates: phenomenology of expectancies (for which he cites Noë 2004), and ‘for-me-
ness’ (Kriegel 2004).

63
That phenomenology of objectivity is a feature of our experience of which we can become aware suffices on
its own to show that the transparency thesis is false, when stated in full generality. Tye is not in a position
to deny this, because his statement of the transparency thesis, and our ability to see the morsel of truth that
it contains, both rely on perceptual experience having this aspect of attitude-specific phenomenology.

This is no accidental feature of Tye’s account, easily avoided by a different formulation. Instead, Objectivity
is the deeper fact about perceptual experience, an underlying truth of which transparency is but a symptom.
To bring this out, let us again consider what transparency is supposed to be a fact about. It is plausibly inter-
preted as a fact about attention, 110 but many, Tye included, describe it as a fact about awareness.111

Construed the former way I take the datum to be that focusing attention on features of experiences appar-
ently doesn’t come easily, and that it often appears to us that we focus attention on features of experiences
by focusing attention on that which we experience. Construed the latter way I understand the datum to be
that awareness of features of our experiences apparently doesn’t come easily, and that we are apparently usu-
ally aware principally of features of that which we experience.

In either case, Objectivity explains the datum. In virtue of having phenomenology of objectivity, the experi-
ence itself ‘tells me’ that what I am attending to, or aware of, is part of a world which exists independently of
me.

Suppose that I am visually perceiving a chair, and that I try to attend to a content-specific feature of my ex-
perience. Since perception has phenomenology of objectivity, experience itself ‘tells me’ that the feature I am
attending to is part of a world which exists objectively. It does so irrespective of whether a) I am in fact pri-
marily attending to a feature of the experience, and only derivatively or secondarily, or even not at all, at-
tending to a feature of the chair, or b) I am in fact primarily attending to a feature of the chair, and only de-
rivatively or secondarily, or even not at all, attending to a feature of my experience, or c) I am in fact some-
how attending both to a feature of the experience and a feature of the chair.

Similarly, since perception has phenomenology of objectivity, as I try to become aware of a content-specific
feature of my experience, experience itself tells me that the feature I am aware of is a feature of a world which
exists objectively. It tells me this irrespective of whether d) I am in fact primarily aware of a feature of my
experience and only derivatively or secondarily, or even not at all, aware of a feature of the chair, or e) I am
in fact primarily aware of a feature of the chair, and only derivatively or secondarily, or even not at all, aware
of a feature of my experience, or f) I am in fact (somehow) aware both of a feature of the experience and of a
feature of the chair.

Either way, because perceptual experience has phenomenology of objectivity it will appear to me that I am
attending to, or aware of, a feature of the objective world, and not a feature of my experience. Objectivity
thus explains the datum on either interpretation. This constitutes a strong reason to believe the thesis.

Michael Huemer has suggested that transparency also characterises intuition. If this were right it would be
natural to argue in parallel fashion from the transparency of intuition to Objectivity. And again, it would be

110
Chalmers says that “the central datum of transparency is that when we attempt to introspect the qualities of our experiences (e.g.
phenomenal redness), we do so by attending to the qualities of external objects (e.g. redness)” (2004: 176).
111
The two might come apart. Suppose that whenever I perceive something, its features ‘grab my attention’ in much the same way a
sharp pain or a sudden movement does. If so, we might usually be unable to attend to features of our experiences, but we might still
become aware of them, since it’s far from obvious that one cannot be aware of things outside attention.

64
pleasing to complete the analogy between intuition and perception in this way (cf. §4.2). But again (again),
this is not where the arguments lead.

“Transparency”, says Huemer, is the view that:

. . . the way we determine the properties of our sensory experiences is by looking at the objects
we’re perceiving; when we try to look at our experiences, we just ‘see through’ them to the objects
they represent. . . . Likewise, in ethical intuition, as a point of phenomenological fact, we find our-
selves presented with moral properties and relationships, not with mental states. (2005: 121–2)

It is true that a person enjoying an intuitional experience isn’t usually attending to the properties of her ex-
perience. It is also true that when I intuit that torturing the innocent is wrong, that is the way things seem
to me to be. But there is no more substantial sense than that in which I am ‘presented with’ moral proper-
ties and relationships in moral intuition, nor a more substantial sense in which I am ‘presented with’ logical
properties and relationships in logical intuitions, and so on.

Insofar as it holds at all, transparency holds only for content-specific phenomenology. Since intuition
doesn’t have that, there is no analogue in intuition of the transparency datum for perceptual experience,
namely that it often appears to us that we focus attention on, or become aware of, features of experiences by
focusing attention on, or by becoming aware of, that which we experience. There aren’t any content-specific
features of intuitional experience to become aware of or attend to in this way.

The argument in this section strengthens the claim that perception has phenomenology of objectivity. At
first glance it might seem to do so only at the cost of undermining the claim that perceptual and intuitional
experience are phenomenologically similar.

But not so. I have already argued that intuition lacks content-specific phenomenology, so there is no further
disanalogy here. Moreover, an account that placed intuition and perception completely on a par, phenome-
nologically speaking, would not be credible, since it is obvious that there are significant differences in phe-
nomenal character between the two mental states. Theoretical contortions aimed at avoiding this datum
does not strengthen a theory, it weakens it.

The trick is to properly account for these differences—which the account in this book does by saying that
perception has, but intuition lacks, content-specific phenomenology, a significant and very noticeable fea-
ture of perceptual phenomenology—while at the same time upholding the analogy where it matters—which
the account in this book does by arguing that epistemically significant aspects of phenomenal character
must be attitude-specific, and by showing that intuition and perception are alike in these respects. That is
the right balance.

4.5 Phenomenology of Pushiness

I now want to defend the following thesis:

Pushiness: Intuition and perception both have phenomenology of pushiness

An experience has phenomenology of pushiness if, instead of representing its content ‘neutrally’, or as a
possibility for consideration, it ‘pushes’ the subject of the experience to accept its content, and this is an as-
pect of the very phenomenology of the experience.

65
4.5.1 Recognising Pushiness
One of the best reasons you could have for believing that intuition and perception have phenomenology of
pushiness is that you recognise this aspect of the phenomenal character in your own experience. My aim
now is to describe the phenomenology in a way that allows this to happen.

Perceptual experience purports to represent a world that is independent of the perceiving subject. But per-
ceptual experience doesn’t represent the objective world in just any old way. It doesn’t present a picture of
the subject-independent world neutrally, nor does it present for consideration the possibility that things
might be a certain way there. Instead, perception pushes the subject to believe that things really are that way.
That it does so is an aspect of the very phenomenology of the experience, an attitude-specific aspect.

Similarly, intuitional experience represents things as being a certain way independently of the subject. But
intuitional experience doesn’t present that content completely neutrally, or offer for the subject’s considera-
tion the possibility that things might be a certain way independently of the subject. Instead, it pushes the
subject to believe that things really are that way. That it does so is an aspect of the very phenomenology of
the experience, an attitude-specific aspect.

I am neutral about the truth or falsity of many propositions; let one of these be the proposition that p. If I
have a perceptual experience that p, a drastic change occurs. I can no longer remain neutral. I must take a
stance. That needn’t be coming to believe that p, or even coming to believe its negation. I can still suspend
belief. But I am pushed to believe that p, so not believing that p, and even suspending belief as to whether p,
involve a feeling of resisting the push from experience.

For example, I am currently neutral with respect to the proposition that a black bicycle is parked at a partic-
ular bike rack outside my office. I can check if I want, for example by visual inspection. But if I end up hav-
ing a perceptual experience as of a black bike in the appropriate place I can no longer remain neutral: the
experience pushes me to believe that there really is a black bike there.

The same holds for intuition. Consider my intuition that if my shoes are by the door, they are not not by the door,
and suppose what was until recently true, namely that I have never before considered this proposition. But
now I do consider it, and I have an intuition with that content.

Then I can then no longer remain neutral with respect to that proposition: I am pushed to believe it. Not
believing it, even suspending judgement with respect to it, involves a feeling of resisting the push from expe-
rience. When in the Gettier case you have the intuition that Smith doesn’t know, this seems to be the way
things are independently of you. It also feels like something you are pushed to believe, and this is a part of
what it is like to have the experience. Similarly for the intuitions that torturing the innocent is wrong, that peo-
ple generally prefer less pain to more, and so on. If you have these intuitions they are not neutrally represented:
you are pushed to believe that things actually are that way, and this is an aspect of the very phenomenology
of the experience.

By analogy, consider the contrast between assertions and questions in speech. 112 If I utter the same sentence
in two similar contexts and my intonation differs in the right way, one utterance can constitute an assertion

112
Richard G. Heck Jr. also uses a conversational analogy to characterise perception (2000: 507; 08 n. 26). However, Heck goes on
to say that perception and belief share this feature; both have “assertoric force” (2000: 508). By contrast, it is very important to the
account advocated in this book that phenomenology of pushiness is not shared with belief. Huemer (2001: 53–4) also takes belief
to share the crucial feature, so—even though he has recently expressed sympathy with the idea that the ‘forcefulness’ of experience

66
and the other a question.113 The assertiveness arises from the tone of the entire sentence, and is usually
something of which we are immediately aware. Similarly, phenomenology of pushiness impacts not on the
content itself, but on how (the entirety of) the content is conveyed.

Let’s return to perception and imagination. Above we imagined a situation in which, while out on a walk,
you suddenly come upon a small, red, cottage that is exactly similar to one that you have just visually imag-
ined. The two experiences would, I noted, differ in part by the presence in perception and the absence in
imagination of phenomenology of objectivity. They would also differ with respect to pushiness. Imagination
does not push the subject to accept its content, but perceptual experience does.

Another contrast is between perceptual or intuitional experience, on the one hand, and conscious belief, on
the other. In belief, there is phenomenology of objectivity. When I consciously believe that the Naive Com-
prehension Axiom is false, I believe that that is the way things are independently of me, objectively speak-
ing. This is a feature of the phenomenology of the belief. But I don’t feel a push to believe that NCA is
false. Conscious belief lacks phenomenology of pushiness, intuition has it.

For perception the contrast is complicated by the fact that conscious belief usually has far simpler content
than perceptual experience does, so that for many visual perceptual experiences it seems unlikely that there
can be a counterpart belief with the same content. As we have noted, however, there are very simple ways
thing can look: it can look to me as if there is a white, point-sized patch of yellow light on an otherwise dark
surface about three metres in front of me, for example. I can clearly consciously believe that content. There
would still be a difference in the character of the experience, and the difference would partly consist in the
presence in perception and the absence in conscious belief of phenomenology of pushiness.114

Like the claims about objectivity above, these claims about pushiness concern the phenomenal character of
perceptual and intuitional experiences. They are not claims about their functional roles. Although percep-
tual and intuitional experience often give rise to belief they are, as we saw in Chapter 2, independent of it,
and we can make claims about phenomenology without making claims about functional role.

Because our language for talking about phenomenal character is underdeveloped, those who wish to take
phenomenal character seriously are often forced to coin new terms. A common practice is to appropriate
terms from other parts of language. Another approach would be to create new terms in neutral language:
‘type 2 phenomenology’, or whatever. But more evocative terms can aid in recognition, and are easier to
remember. Terms are of course evocative precisely because of their connotations. In this case some of those
are functional, which is an unwelcome distraction. Still, stipulating a strictly phenomenal reading is possi-
ble, and that is what I am doing.

The term ‘pushiness’ must be understood metaphorically. I think the metaphor is apt, and evocative of the
correct aspect of experience, but it is not perfect. For one, it has associations of etiquette: a pushy person is

may be deemed an attitude-specific aspect of its phenomenal character (in personal communication)—I therefore still resist his view.
The phenomenal feature at issue is in my view emphatically not shared with belief.
113
That this account of the difference between assertive and inquisitive utterances is overly simplistic doesn’t matter here, since my
purposes are to illustrate.
114
It would also partly consist in the presence in perception, but the absence in belief, of content-specific phenomenology: this expe-
rience would differ from one in which the light was instead cold white, for instance.

67
rude, and this connotation is not helpful. 115 But the term’s functional connotations are useful (when the stip-
ulated phenomenal reading is kept in mind), and in any case, the point is not to pick the perfect label, but
to lock on to a real and important aspect of phenomenal character.

There are other terms we might have used instead, and noting what is right and what is wrong about them
can further aid recognition. For instance, one might have called the phenomenology ‘coercive’. But coer-
cion is a success term: if I coerce you to Φ, it follows that you Φ. This is wrong for both perception and in-
tuition: I can fail to believe what I seem to see, and I can fail to believe what I intuit. The term also has neg-
ative connotations, which again doesn’t fit. 116

Or one might call the experiences ‘insistent’. This is on the right track; it is not a success term for instance.
(I might insist that you Φ while you ignore me.) Frank Jackson uses the term ‘badgering’ to denote what
may well be the same aspect of perceptual experience that I am trying to single out. But this has the whiff of
repeated or continuous insistence: if I badger you to Φ, I use every opportunity to remind you to Φ. Percep-
tual and intuitional experience are not like this at all: not even when looking at an unchanging scene.

None of these terms are spot on exactly. I will stick with ‘pushiness’, which I see as an imperfect but evoca-
tive label for the aspect of the phenomenology on which we have now hopefully managed to home in.

Other thinkers have discussed the phenomenology of intuition or perception in ways which seem to indi-
cate that their view of the character of these experiences is at least similar to the view advocated here. I take
this to strengthen the claim that these mental states really do have this phenomenal character. 117

When Pryor notes “the peculiar ’phenomenal force’” of perceptual experience (2000: n. 37), for example,
this is reminiscent of phenomenology of pushiness: intuitively, if perceptual experience has phenomenal
force, it can push me to accept its content, and vice versa.

Similarly, Huemer describes perceptual experience as ‘forceful’:

[Y]ou would never confuse seeing a tomato with imagining one. The reason lies in what I call the
“forcefulness” of perceptual experiences: perceptual experiences represent their contents as actual-
ized; states of merely imagining do not. When you have a visual experience of a tomato, it thereby
seems to you as if a tomato is actually present, then and there (Huemer 2001: 77).

If Huemer intends to pick out the same character with this term as I do with pushiness, then I agree that a
difference between imagining and perceiving a tomato is that the latter has pushiness. However, as noted,
there is also a difference in objectivity. Moreover, Huemer cannot have exactly the same phenomenal char-
acter in mind, since he also attributes it to belief (2001: 53-4).

115
Thanks to Fiona McPherson for pointing this out to me.
116
M.G.F. Martin, who also discusses the ‘non-neutrality’ of perceptual experience, says: “it seems inconceivable that one should be
in a mental state phenomenologically just the same as such a perceptual experience and yet not feel coerced into believing that
things are the way that they are presented as being” (2002: 390).
117
As we have seen, the objection from the fact that other thinkers fail to recognise intuitional experience in their own mental lives,
what I called ‘the Absent Experience Challenge’, can be answered. Still, a theorist who, like me, puts significant theoretical weight
on an experience-type having a certain phenomenal character should probably eventually give up, if no one recognised the experience
from their description. Fortunately, this has not been my experience, but a sceptical reader may be reluctant to take my word for
this claim. The historical record might help, even if the similarity of expression is imperfect.

68
Ernest Sosa has said that in intuition one can “feel the ‘pull’ of conflicting considerations” (Sosa 2007b:
47). And William Tolhurst discusses what he takes to be a general class of seemings, which incorporates per-
ceptual and intuitional experiences: “[S]eemings are mental states in which the subject experiences a felt de-
mand to believe the content of the state” (1998: 298). This again seems closely related to pushiness: when
one is pushed to believe that p, it is plausible that so believing feels demanded. (Tolhurst even notes that a
subject can feel ‘pressured’ and ‘pushed’ by felt demands (1998: 298)).

Finally, the thought that perceptual experience has phenomenology of pushiness may not have been foreign
to David Hume. In the Treatise, he writes:

The difference [between impressions and ideas] consists in the degrees of force and liveliness, with
which they strike upon the mind, and make their way into our thoughts and consciousness . . . .
Those perceptions, which enter with most force and violence, we may name impressions . . . .
(1739/2007: 1.1.1.1) 118

I make no claim to Hume scholarship. However, the contrast between episodes 119 which enter conscious-
ness with force (and ‘violence’) and those that do so to a significantly lesser degree does bear a strong resem-
blance to the distinction I have been drawing between experiences that have the phenomenology of pushi-
ness and those that do not.

These passages do not justify certainty, nor indeed anything like it, that these authors had in mind exactly
that aspect of phenomenal character on which I have been trying to home in. However, even with disagree-
ment both at the margins and with respect to what I regard as central issues—for example, that conscious
belief does not have the phenomenal character at issue—it is plausible that there is a core phenomenon here
that has been imperfectly recognised by many, and for a long time.

4.6 Valence
We have noted three similarities between intuition and perception: both have representational content,
both have phenomenology of pushiness, and both have phenomenology of objectivity. We have also noted
an important disanalogy: perception has but intuition lacks, content-specific phenomenology. It is time now
to note another salient difference: intuition has valence, perception does not.

By this I mean that in intuition it can seem false that p, just as it can seem true that p. There is no corre-
sponding phenomenon in perception: it cannot perceptually seem false that p.

Suppose I ask you to consider the proposition that people are usually indifferent between pleasure and
pain. I assume that this seems false to you. That does not mean that what happened is that it seemed true to
you that people usually prefer pain over pleasure, or that it seemed true that it is false that people are usu-
ally indifferent between pleasure and pain. That people usually prefer pain over pleasure entails that they

118
I follow Hume Studies in using the recent Clarendon Press edition; not the older Selby–Bigge–Nidditch edition previously consid-
ered standard. Hume also mentions the ‘liveliness’ of impressions and ideas, and later discusses how it also can vary. It’s not clear
whether the liveliness of an impression or idea is a distinct feature from its force. Some later passages seem to indicate that they are
one and the same: “[It is] evident at first sight, that the ideas of the memory are much more lively and strong than those of the imag-
ination . . . . When we remember any past event, the idea of it flows in upon the mind in a forcible manner; whereas in the imagi-
nation the perception is faint and languid . . .” (1739/2007: 1.1.3.1).
119
Perceptions’ is Hume’s catch-all term for mental states; on this see e.g. Huemer (2001: 78). My point is that Hume may have
been concerned with a similar phenomenal feature as that which I’m attempting to describe; not which mental episodes he attrib-
utes these features.

69
are not indifferent between the two, of course. And you might at other times have either of these two other
intuitions. But you can also have an intuition with the content people are usually indifferent between pleasure
and pain, but with negative valence. That’s a different mental state, and any viable account of the metaphys-
ics of intuition must have the resources to mark this distinction.

4.7 Belief
The phenomenology of pushiness and objectivity can help single out perceptual and intuitional experience
from other mental states, and, in particular, from belief and conscious thought. In Chapter 2 I argued that
intuition is distinct from belief on independent grounds, but the claim about phenomenology is still im-
portant. An account which identifies intuition partly by its phenomenology, and which shows that the phe-
nomenology is had by intuition but not by belief, is in virtue of this a more fully fledged competitor account
to Doxastic views of intuition.

First, intuition cannot reasonably be thought to share attitude-specific phenomenology with beliefs gener-
ally. Standing belief has no phenomenal character; there is nothing it is like to have a standing belief. So the
question of whether intuition can be distinguished from belief by reference to attitude-specific phenomenol-
ogy is at least restricted to occurrent belief. One can question whether all occurrent beliefs are conscious
beliefs, but I bracket this issue here, and focus exclusively on the class of conscious beliefs. 120

I think it is often assumed that there is a straightforward sense in which one and the same belief can be ei-
ther standing or conscious, and that philosophers often speak and write as if the states of standing and con-
scious belief, and the connection between them, are all well understood. I think that’s a mistake, in part
because I think that what answers to the term ‘conscious belief’ is a rather multifarious collection of states,
and not always something that corresponds to a standing belief in a straightforward way.

One indication of this is that, while it is possible that ‘beliefs’ is a ‘bogus plural’ where standing beliefs are
concerned (Lewis 1994: 423; 1986: 32)—it is possible, that is, that there really is just one belief, a massively
complicated one—this does not seem to be possible for conscious beliefs. The accuracy conditions of every-
thing I believe put together—of my system of standing belief, if you will—are much more demanding than
the accuracy conditions of a single conscious belief, and the accuracy conditions of single conscious beliefs
are not identical to each other. Conscious beliefs are at least partly individuated by their contents, so there
must be many different conscious beliefs. But there might be just one single standing belief.

This doesn’t show that no notion of conscious belief is in good standing. But it does indicate that we are
well advised to be very clear about our subject matter when we discuss them. Once we are, we can see that
phenomenology of objectivity and pushiness helps to distinguish intuition from conscious belief.

If you ask what I’m thinking and I answer “that yesterday was a warm day”, that can be a truthful answer if
what I was doing when you asked was to iconically remember yesterday’s warmth. That thinking in this
sense contributes to the character of overall conscious experience is not controversial. The same holds for
iconic memory in the other sensory modalities, and for the corresponding varieties of iconic imagination;
and for both memory and imagination of moods, emotions, and bodily sensations. Only when all of this
has been ruled out are we left with an interesting question, namely that of whether non-iconic thought—iden-
tified by exclusion in this way—has content-specific phenomenology. I have argued that we should answer

120
A functionalist about belief might think that a belief counts as occurrent in virtue of the belief being deployed in occurrent rea-
soning or decision making processes. Given that not all reasoning and decision making processes are conscious, a belief might
count as occurrent without being conscious.

70
this question in the negative. For current purposes, however, the crucial point is that the mental state from
which intuition must be distinguished is non-iconic conscious thought.

Some argue that thinking in the sense at issue is not a propositional attitude, but a ‘mere holding in
mind’. 121 I do not understand what that is supposed to mean. I think that whenever I am related to a propo-
sition I take some attitude to it: I might consider it slightly more likely to be true than to be false, consider
it a strange proposition, consider it a surprising proposition, or whatever. At the very least, I vaguely wonder
whether it is true or false. To the best of my knowledge I never merely ‘hold a proposition before the mind’.

One thing I do do is to give a proposition what we might call ‘mental assent’: the analogue of ‘saying’ to my-
self—though inner speech need not be involved—“yes, that seems right”, of giving the proposition a mental
‘tick’. This, I think, is one of the phenomena that answers to the phrase ‘thinking that p’.

Such cases may be what gives rise to a temptation to say that belief shares the aspects of attitude-specific
phenomenology which I have argued that intuition has, since they answer to the phrase ‘thinking that p’,
and are also, I think, what we sometimes have in mind when we say that a person consciously believes that p.

I agree that some cases sometimes described in this way are characterised by phenomenology of objectivity
and pushiness. But this does not render me incapable of distinguishing intuition from conscious belief, be-
cause such episodes just are intuitions. When you consider whether p and have the reaction I have de-
scribed, including in particular the relevant phenomenal character, then you are having an intuition.

We have already noted that uses of the word ‘intuition’ in English are highly varied, and not of much inter-
est for our investigation (§1.3). In a similar vein, we should not expect natural language to make all the im-
portant discriminations, so it is no great surprise that ‘x consciously believes that p’ is at times used in cases
that are actually cases of x’s having the intuition that p.

Not all cases that we commonly classify as cases of conscious belief are like that, however. A person, call her
Susan, can of course consciously believe that p at a time even though she doesn’t have the intuition that p.
Susan might consciously believe, for example, that first order propositional logic is sound and complete
without having that intuition.

So far the description is compatible with several distinct states of affairs. One thing that might be happen-
ing is that Susan is conscious of having the standing belief that first order propositional logic is sound and
complete. But this is an introspective state: the sense of ‘conscious of’ is that of being aware of. This is not a
straightforward counterpart of her standing belief, not merely a case of a standing belief now having been
made conscious. In particular, the contents of the two states differ. The content of the standing belief is:
first order logic is sound and complete. The content of the introspective state is: I believe that first order logic is
sound and complete.

Another thing that might be happening is that Susan’s experience is characterised by her being committed to
first order logic being sound and complete. I do not mean that she is having another introspective state, the
content of which is I am committed . . . . Rather, the content of her state is first order logic is sound and complete,
but she is related to the content ‘committedly’. The attitude is one of being on board with, and there is, I
think, a corresponding aspect of attitude-specific phenomenology, which helps to demarcate one sense of

121
See e.g. David Pitt (2004: 2–3) (2004: 2–3).

71
‘conscious belief’. But this is also one to which there is, I think, no temptation to think that phenomenol-
ogy of pushiness applies (although the state is characterised by phenomenology of objectivity). What it is
like to be in such a state is very different from what it is like to be pushed to believe a content.

To sum up, we often use ‘conscious belief’ loosely, just as we use ‘intuition’ loosely. Some things we so
name are intuitions, but some are not, and in the latter cases, reflection on the phenomenal character of
intuition and of the states in question can help us to distinguish the two.

4.8 Objectivity, Pushiness, and ‘Presentation’

There are, as mentioned, other thinkers who emphasise the similarity between the phenomenology of per-
ception and the phenomenology of intuition. Some of them rely on the notion of ‘presentation’, where this
is at least in part a phenomenal notion. How does that notion relate to the account I have advocated?

I want to begin by acknowledging substantial agreement here. Presentation theorists and I agree that there
are important similarities between perception and intuition, in particular in phenomenal character, and
that these similarities have epistemic consequences. Moreover, the lack of well established methods for ar-
riving at precise conclusions about the nature of our experiences, and the fact that our vocabulary is often
imprecise or underdeveloped, may make the disagreement seem greater than it actually is.

That said, I think these theories have flaws that are real and, in the end, fatal. Some of these will be dis-
cussed in later sections (§5.5). In this chapter my focus is on giving the best possible account of the phe-
nomenal character of intuitional experience. In that regard, too, I find these accounts wanting.

As I have described the phenomenology of intuition, it is not an intellectual or complicated affair. The intu-
itional experience pushes me to accept what it represents, just as perceptual experience does, and this is part
of the phenomenology of intuitional experience. It is also part of the phenomenology of the experience that
what I am purportedly informed about is independent of me. Both notions are uncomplicated and straight-
forward, and experiences with this character could clearly be present in quite simple creatures.

For comparison, consider now Elijah Chudnoff’s explication of ‘presentational phenomenology’:

(Presentationality of Intuition) Whenever you seem to intuit that p, there is some q (maybe = p)
such that—in the same experience—you seem to intuit that q, and you seem to be intellectually
aware of an item that makes q true (Chudnoff 2011b: 641). 122

Perhaps it is possible to understand this talk of seeming to be aware of an item that makes the proposition p
true as saying nothing more than that it seems to the subject that p with some force. If so, Chudnoff could
be understood as saying that when I have an intuition that p I feel pushed to accept that p. In that case I re-
gard the description as correct but incomplete, since intuition has phenomenology of pushiness but also
phenomenology of objectivity (and valence).

A second interpretation says that the talk of being aware of an ‘item’, a truth-maker for the content of one’s
intuition, is to be understood as the intuition also having the phenomenology of objectivity. So perhaps it is
possible to understand Chudnoff to mean much the same as what I mean when I say that in intuition, as in
perception, I am purportedly informed about the way things are independently of me, and that this is part of

122
Compare Pollock: “There is a phenomenological difference between those intuitions that provide the initial premises of proofs
and those that merely guide the mathematician in trying to construct proofs. The latter involve seeing that something is true by
seeing why it is true, whereas the former just involve seeing that it is true” (1974: 323).

72
the very phenomenology of intuitional and perceptual experience. If so I would have no objections, other
than to say that I take my own description of the phenomenal character to be rather more perspicuous.

But there is a third interpretation, according to which Chudnoff is making a much stronger assertion. On
this interpretation, when a person has the intuition that p, it seems to her that she is in contact with—aware
of—some truth-maker for p, and moreover, of the fact that it makes p true: she’s aware of it as a truth-
maker. 123 I think this interpretation is correct, but it strikes me as so obviously inadequate that I am
tempted to simply point to it and set it aside.

First, this view is inadequate because it is wildly phenomenologically inadequate. It is just false that I seem
to be aware of a truth-maker for p when I have the intuition that p, and even more obviously false that I
seem to be aware of the truth-maker as a truth-maker. What would it be like to seem to be in contact with a
truth-maker for a mathematical claim, let alone being aware of it as a truth-maker for the claim? I have no
idea. It would depend, surely, on what mathematical claims in the end turn out to be claims about, on what
the nature of mathematical reality is. But I have no idea about that either, and I am not informed thereof by
the phenomenal character of my intuitional experiences.

The problem isn’t just that we get some information about mathematical objects in intuition, just not quite
enough to narrow down their metaphysical status completely. Instead, I think that nothing like this goes on
at all. If it seemed to me that I was in contact with ‘an item that makes q true’, I could fairly be asked to say
something about the properties that item has, on pain of rendering my claim to be aware of the item incredi-
ble. But I am not: again, and, I dare say, like most others, I have no idea what mathematical reality is like,
and yet I have mathematical intuitions aplenty.

The same point can be made with equal or greater force for all domains in which we have intuitions, which
on the view advanced in this book are just all domains simpliciter (§3.7). Consider ethics and rationality,
for example. What item am I aware of when I have the intuition that a person cannot rationally believe (p
and not-p)? When I intuit that torturing the innocent for fun is wrong? That widespread poverty coexisting
with absurd wealth is a moral catastrophe? That states must tax the wealthy to care for those that really need
it? That education is a right? That rational belief is responsive to evidence? No answers to these questions
are at all phenomenologically plausible, and yet intuitions in these domains are ubiquitous.

It is true that when it seems to me that two plus two equals four it seems that this is how things are inde-
pendently of me. Intuition does, in this limited sense only, seem to put me in touch with subject-independ-
ent reality. But not in any more substantial sense than this! It does not seem to me that I am aware of any
object when I have the intuition that if p, then not-not-p. What object would that be? What would it be like to
seem to be aware of it? Is that the same object or a different one than the one in play when I have the intui-
tion that if not-not-p, then p? Again, there are no acceptable answers. If classical logic is right, both inference
patterns are valid; if intuitionistic logic is correct, only the former is. If intuition in part consisted of item-
awareness of (say) the abstract logical structure that makes one or both of these true, shouldn’t we expect

123
Is Chudnoff really committed to this? Yes! In discussing the intuition that two circles can at most have two common points, for
example, he says that intuition represents that proposition, “[b]ut there is also something else. My intuition of [that proposition]
does not just represent it as being true, it also makes it seem as if its truth is revealed to me by my intellectual awareness of its sub-
ject matter, i.e. my intellectual awareness of items such as circularity that contribute to making the proposition true” (Chudnoff
2012: 58).

73
some indication as to which logic is right from close inspection of the truth-makers that ex hypothesi are part
of the phenomenology of intuition? We get no such thing. Nothing like this goes on at all.

In addition to being phenomenologically implausible, I think that Chudnoff’s view hyper-intellectualises

intuition. That is, his view entails that the intellectual threshold for having an intuition is (much) higher
than it in fact is; it makes demands of agents’ cognitive abilities that intuition in fact does not require. It is
therefore false.

On the theory of intuition advanced here, to have an intuition is to simply to have an experience with rep-
resentational content and a certain phenomenal character. This account places some intellectual demands
on the intuiting agent: to intuit that p one must be capable of believing that p. But Chudnoff’s account is
much more intellectually demanding, and implausibly so. 124

The ability to believe (true) propositions about truth-makers are in general more demanding than believing
first-order propositions in the same domain. For example, children can and do believe that there is a chair
by the door, but what makes this belief true is a complex set of facts about metaphysical reality that the
child has no capacity to grasp. Children can and do believe that the word ‘chair’ applies to chairs but not to
bicycles even though we sit on both, but what makes that true are complex facts about linguistic conven-
tions, of which children have no notion. And children can and do believe that no kangaroo is a wallaby,
but what makes that true are facts about the nature of natural kinds that quite plausibly no one fully grasps
as yet, and that children certainly cannot begin to comprehend. To lump the ability to grasp what makes an
intuited proposition true with the ability to have the intuition itself is to hyper-intellectualise intuition. A
theory that does this is false.125

124
Boghossian (2001: 637–8 ) raises a similar challenge against BonJour (1998), for similar reasons.
125
In personal communication Chudnoff has argued that my criticisms of his account at least to some extent miss their mark, for
two reasons. First, he thinks that I overlook aspects of the view that to some extent blunt my criticisms, and that were present in his
work from the start. Second, he argues that the account of presentational phenomenology that he develops in some of his later
works (Chudnoff 2016, 2019, 2017) escape the charge of hyper-intellectualisation.
Taking these two points in turn, what I have allegedly missed in earlier presentation of his views is that intuitional experi-
ence can have presentational phenomenology with respect to just some of its contents. But this does not help, since it is clear that
we do have intuitions about rationality, morality, mathematics, and so on; and that we are thereby justified in the corresponding
beliefs. Chudnoff is clearly committed to this, especially for mathematics, and insofar as he would be willing to retreat for any of
these domains, that would be a significant cost to the theory. But then the fact that there may be other represented propositions
that lack presentational phenomenology is neither here nor there. Presentational phenomenology is supposed to be explain justifi-
cation, and must therefore be present for any mathematical, moral, and rational propositions that the theory wants to say are justified
by intuition. And then the problems of phenomenal inadequacy and hyper-intellectualisation simply arise again for those proposi-
tions.
As regards the claim that later presentations escape the charge of hyper-intellectualisation; as usual I find much of value
in Chudnoff’s nuanced work. Although I cannot address it in full here, it think it is clear that these developments fail to dispel the
worry. Chudnoff starts with a point on which we agree (and which I discussed in §1.2.1 above), namely that we must distinguish
between what a perceptual experience can singlehandedly justify, and what one can only become justified in believing if one has
justification to hold other beliefs. On my account this is simply the distinction between what is and is not part of the representa-
tional content of the experience. True, it is sometimes not immediately clear what that is, but I agree with Pryor that this question is
to a large extent empirical. Whatever the content is, having the relevant experience singlehandedly justifies belief in that content. By
contrast, Chudnoff assumes that some content of experience is at best mediately justified. This gives rise to a need to account for
that difference somehow, and Chudnoff proposes to account for it in terms of a presence for some contents, and an absence for
other contents, of his presentational phenomenology.
A central case for Chudnoff is the visual perceptual experience of a partially occluded familiar object. According to him,
in these instances you stand in a relation to ‘visual awareness’ of certain parts of the object—the ones that are visible to you—but not
to the occluded parts of the object. In reply, I reject, first of all, the idea that an account of such a relation is required to tell the true

74
Chudnoff’s account of the phenomenal character of intuitional experience is also implausible on empirical
grounds.

In a widely discussed series of studies, Jonathan Haidt, Fredrik Björklund, and Scott Murphy 126 presented
test subjects with vignettes designed to elicit strong moral intuitions, but carefully written so as to block an-
ticipated arguments that might be given to explain or rationalise the resulting judgements. One of these
probed the widespread moral intuition is that siblings must not have sex with each other, regardless of cir-
cumstances (Haidt, Bjorklund, and Murphy 2000; Haidt 2001):

Julie and Mark, who are brother and sister are traveling together in France. They are both on sum-
mer vacation from college. One night they are staying alone in a cabin near the beach. They decide
that it would be interesting and fun if they tried making love. At very least it would be a new experi-
ence for each of them. Julie was already taking birth control pills, but Mark uses a condom too, just
to be safe. They both enjoy it, but they decide not to do it again. They keep that night as a special
secret between them, which makes them feel even closer to each other. So what do you think
about this? Was it wrong for them to have sex? (Haidt, Bjorklund, and Murphy 2000, p 18)

After being presented with this case, participants were asked whether Julie and Mark did something wrong.
Where participants said yes (the large majority) the experimenter proceeded to gently argue with them, try-
ing to “undermine whatever reason the participant put forth in support of his/her judgement or action” (p
9), by noting that no-one was harmed, and by exploiting the fact that possible negative consequences (con-
ceiving a child, signalling effects, the establishment of a pattern of behaviour, damage to Julie’s or Mark’s
wellbeing, damage to their relationship) did not obtain.

In this situation, Chudnoff’s theory predicts that participants should either resist the challenge and stick to
their guns—insisting that the siblings’ relationship was damaged after all, perhaps, if that damage was the
‘item’ of which they were apparently aware—or else drop the judgement completely. After all, if participants
at first seemed to be aware of an item that made the intuited proposition true, but then came to realise that
the relationship was not in fact harmed, there would by the participants’ own lights no longer be anything left
to make the proposition true. Moreover, this should be phenomenologically highly salient to them: indeed,
on Chudnoff’s account this cannot be other than a dramatic shift—for an integral constituent of the intui-
tional experience would now be missing. The intuitional experience would at the very least fundamentally
change, and indeed it would more likely simply vanish.

story of perceptual justification, in part because this line of thought overlooks that the central concern for a person invested in
proper inquiry is to respond appropriately to her own epistemic perspective (and a story that is properly sensitive to a person’s epis-
temic perspective cannot, I think, be relational.) More to the point, however, it is highly implausible that any such relation holds
between a perceiver and parts of objects. If, as I visually experience a partially occluded dog, I stand in a relation of visually aware-
ness (or seeing, Chudnoff uses the terms interchangeably) to anything, I stand in that relation to the object: I see, or seem to see, a dog.
To be sure, there is a difference between seeing a familiar object partially occluded and seeing that same object un-occluded, but it
cannot be accounted for in terms of an awareness or seeing-relation that in one case holds only to parts of the object. Presentational
phenomenology cannot, a fortiori, be elucidated by means analogy with that relation, on Chudnoff’s account. And even bracketing
this concern, I find no phenomenal plausibility in the claim that there is a distinction in experience “with respect to the proposi-
tion that the dog has a tail” and “with respect to the proposition that the dog continues” behind the occluding object (2017). Visual
perceptual experience does not contain such phenomenological distinctions corresponding to various parts of its content. The
charge of hyper-intellectualisation remains.
126
The work was initially based on Murphy’s honors thesis.

75
That is not what happened. Instead, participants maintained their judgement but dropped the argument in
its favour and swapped it for another. And not just once, either, but multiple times—the median was six.

This strongly tells against Chudnoff’s account. What it suggests is that intuition simply tells us that a propo-
sition is true, but not what makes it so. 127 The behaviour that ensued, namely casting about for an explana-
tion making sense of a strong intuition, is wholly consistent with the account of intuition I have advocated:
indeed it is exactly what one would expect. But it is not consistent with an account on which an item that
apparently makes the intuited proposition true is itself a part of that experience. And of course the behav-
iour observed in these studies is one we can readily reproduce and observe first-hand: when a strong intui-
tion is challenged, a perfectly standard reaction is to cast about for an argument in its favour, rather than to
abandon the conclusion, or have the intuition simply vanish.

A final worry. Above I argued that the epistemically significant phenomenal character of intuitional experi-
ence must be attitude-specific, and that focusing on this kind of character also yields a plausible answer to
the Absent-Experience Challenge (§3.5).

Chudnoff’s answer to the Absent-Experience Challenge is rather different. Proponents of the challenge go
wrong, he argues, in looking for “an experience that occupies a location of its own, distinct from those oc-
cupied by … thoughts, imaginings, intentions, beliefs, etc”. Intuition is instead co-located with thoughts, im-
aginings, etc, because it is constituted by them (2011b: 646). Chudnoff argues that we can still hold that intu-
itions are irreducible to other mental states, and about this I think he is right; I have argued elsewhere that
a good model for understanding his view on this point is the unity of conscious experience (Koksvik 2017:
4).

But the move gives rise to a different and serious challenge of its own. All intuitional experiences share an
aspect of their phenomenal character, and this character sets intuitional experiences apart from all other
mental states (Koksvik 2017: 5). These are core facts about intuitional experience, but Chudnoff’s view
can’t explain or account for either. Why should different intuitions have phenomenology in common if
they are constituted by completely different collections of mental states? 128 I can see no reason, so the evi-
dent phenomenal similarity between different instances of intuition becomes a brute, unexplained fact.129

To sum up, the claim that intuition has ‘presentational phenomenology’ might amount to the claim that
intuitional experience has phenomenology of pushiness; in which case it is true but incomplete, since it
leaves out phenomenology of objectivity. It might amount to the claim that intuitional experience has phe-
nomenology of objectivity and phenomenology of pushiness (and valence); in which case it is true but not

127
We can have intuitions about what causes what, of course, but that is just a small subset of the cases.
128
Chudnoff could say that all intuitions must be partly constituted by mental states of certain types, that such mental state types
have their own attitude-specific phenomenology, and that this explains the commonality in phenomenal character between in-
stances of intuitional experience. But then either the characteristic phenomenal character of intuitional experience is exhausted by
the phenomenal character of the state(s) all intuitions have in common, or it’s not. In the former case, there is renewed pressure
toward reductionism; in the latter, the challenge rearises.
129
Chudnoff in fact endorses 1 and 2, and has (in personal communication) indicated that he thinks “the principle of unity,”
which mental states must meet in order to constitute an intuition, accounts for their truth. But absent an explanation of how com-
pletely different conscious experiences can still result in a new, composite experience with its own distinctive phenomenal character,
this is not much more than a promissory note.

76
perspicuously formulated.130 And it might and probably does amount to the claim that intuitional experi-
ence consists in awareness of the truth-makers of the contents of intuition in a much more substantial
sense. In this latter case it is, I think, quite clearly false, since it is phenomenologically inadequate, hyper-
intellectualises intuition, yields false empirical predictions, and only answers the Absent-Experience Chal-
lenge by introducing an equally serious problem of its own.

For these reasons I think that Chudnoff’s account of the phenomenal character of intuition is untenable.

John Bengson also argues that intuition and perception share the feature of being ‘presentational’. As with
Chudnoff I find much to agree with in Bengson’s view, in particular the view that “intuition is similar to
perceptual experience in epistemically significant respects” (2015b: 495; See also Bengson 2010), and the
view that at least part of the similarity consists in similarities in what it is like to have a perceptual and an
intuitional experience.131

Bengson’s seeks to isolate the phenomenal feature he takes perception and intuition to share by contrasting
those states with what he terms ‘merely contentful’ and ‘merely representational’ mental states. The main
example of a ‘merely contentful’ mental state is imagination, and the main example of a ‘merely representa-
tional’ state is belief. On Bengson’s account, ‘presentational’ mental states like intuition and perception are
representational, but not merely representational.

Again, there is significant agreement between this account and my own on where the fault lines are. I agree
that there is an important contrast between imagination, on the one hand, and perception or intuition on
the other. By contrast to Bengson, however, I have characterised that distinction in more detail: imagina-
tion lacks both phenomenology of pushiness and phenomenology of objectivity. Moreover, iconic imagina-
tion is similar to perception, but dissimilar from intuition, in having content-specific phenomenology; non-
iconic imagination the other way around.

Bengson and I also agree that there is an important distinction between intuition and perception, on the
one hand, and belief on the other. I have characterised this by saying that intuition and perception have
phenomenology of pushiness, which belief lacks. Again, I think this is rather more elucidatory of the phe-
nomenal character of perception and intuition than it is to say that belief ‘merely represents’, whereas intui-
tion and perception also ‘present’. Intuition and perception push me to believe that things are thus and so;
belief at most reminds me of previous commitments. This, it seems to me, gets closer to the actual phenom-
enal character of perceptual and intuitional experience.

130
Understood this way I regard ‘presentation’ as somewhat of a black box in what is sometimes called the ‘boxology’ of the mind. I
think it arises in this way: we note that we have very good reason to believe that a certain function is carried out: in this case the
function of bringing about of (the appearance of) justification from perception and intuition. We then draw a box in the place of
that function, and we give it a name. In this way arises, I suspect, ‘acquaintance’, but also ‘presentation’. But the manoeuvre fails to
advance our understanding: “We are left staring at the problem with which we began, rather than feeling that we have been placed
on the path to real enlightenment” (Boghossian (2001: 637); criticising BonJour (1998).
In contrast, below I argue that Pushiness and Objectivity explain why intuitional experience can justify belief in the way
that it does. Identifying and separating these aspects of attitude-specific phenomenology thus takes our understanding further than
does just attaching the name ‘presentation’ to what’s happening. It allows, for instance, for the possibility that we can come to un-
derstand why other mental states fail to provide such support for belief in a nuanced way. Perhaps this is because the state lacks
objective phenomenology (as in wishful thinking), perhaps it is because it lacks pushy phenomenology (as in conscious belief), or
perhaps it lacks both (as in imagination).
131
That the ‘presentationality’ of perception and intuition is intended as (at least partly) a phenomenal feature is not always as clear
from the texts as one might wish, but Bengson has confirmed this interpretation in conversation.

77
Similarly, I do not find Bengson’s distinction between having and being under an impression elucidatory.
Both of these evoke a force being applied, so this talk, in my view, obscures rather than clarifies the distinc-
tion between intuition and perception, on the one hand, and belief, on the other.

To reiterate, there is important agreement, but I do find reason to complain about the view on the grounds
that it fails to distinguish two importantly different characteristics of the phenomenology of intuition and
perception, and as a result is rather less perspicuous in its description of the phenomenology than I believe
it ought to be. As I hope the discussion above has revealed, we can do better.

4.9 What Intuition Is

This leads us to the following conclusion. There is a class of mental states which have representational con-
tent, which are characterised by attitude-specific phenomenology of objectivity, pushiness, and valence, but
which do not have content-specific phenomenology. That this this class of mental states, this class of experi-
ences, is strongly unified in this way, makes it a good candidate for constituting a psychological kind, and
thus of interest to anyone with an interest in understanding the basic make-up of the mind. As we shall see,
it plays an important epistemological role, too, thus strengthening that claim considerably.

This class of experiences is a good deserver of the label ‘intuition’. It answers well to our use of that term,
certainly capturing the paradigmatic cases, and accounting for a large fraction of all uses once it’s acknowl-
edged that a lot of uses are derivative (§1.3). It also enables relevant distinctions to be drawn, for example
between intuition and conscious belief. Because the class is a good candidate for a psychological kind, it is
likely that it will serve us well to reserve the term ‘intuition’ for members of the class. For an account of the
nature and epistemology of the psychological kind, however, not much ultimately hinges on this verbal is-
sue.

The class of experiences I have singled out fits well with what we have seen that intuition is not. Intuition is
not a belief, nor a disposition to believe (Chapter 2). But although it isn’t identical to it, and although it
does not entail it, intuition nevertheless often gives rise to belief. This is well explained by an account of intu-
ition as an experience with the characteristics I have argued that it has.

When an experience has phenomenology of pushiness and objectivity, the experience purports to inform
the experiencer that things actually are a particular way, objectively speaking. If it seems to an experiencer
that she is informed that things are a certain way, objectively speaking, it is plausible that she will often
come to believe that things are that way. So the conception of intuition as an experience respects and ac-
counts for the fact that intuition and perception often lead to belief, while upholding the distinction be-
tween the two. As we shall see, the attitude-specific phenomenology of intuition and perception is also im-
portant to the epistemic features of these states. In particular, the account vindicates the appearance that
intuition and perception justify belief.

The characteristics given uniquely characterise intuition. The term conscious belief is sometimes used for
intuition itself, but when it is not, it designates a state that lacks phenomenology of pushiness. So intuition
can be distinguished from conscious belief (in this more restricted sense) by the fact that intuition has, but
conscious belief lacks, phenomenology of pushiness.

Intuition is distinguishable from imagination first because much imagination is iconic, and iconic imagina-
tion has, but intuition lacks, content-specific phenomenology. It is distinguishable second because imagina-
tion of any sort lacks both phenomenology of objectivity and phenomenology of pushiness.

78
Intuition is also distinguishable from what some term ‘memorial seemings’. Memory may be iconic or non-
iconic. Intuition is distinguishable from the former since the former has, but intuition lacks, content-spe-
cific phenomenology. Intuition is distinguished from memory of both the iconic and the non-iconic kind
since intuition has, but memory lacks, phenomenology of pushiness.

Perception, on the other hand, does have phenomenology of pushiness, and also phenomenology of objec-
tivity. But perception is still distinguishable from intuition on the present account. First, perception has but
intuition lacks content-specific phenomenology. Second, intuition has, but perception lacks, (negative) va-
lence. And third, visual perceptual experience has a certain ‘visualness’, perceptual experience in the other
modalities has corresponding features, but intuition does not.

Finally, intuition is distinguishable from wishful thinking. If I wishfully think that p, my phenomenology
may have phenomenology of pushiness. But it does not have phenomenology of objectivity: there is no feel-
ing that p is the way things are independently of me.

Is this really an answer to the question of what intuition is? There is a verbal question in the vicinity which
we have been careful to avoid: what we decide to use the term ‘intuition’ to name is not in itself of deep im-
portance. But even when we restrict our attention to the metaphysics of the interesting class of mental states
that I have singled out, a form of the same challenge can be raised. Have I really given an account of the na-
ture of these states?132

There is a sense in which my answer is incomplete. Had we found that intuition has content-specific phe-
nomenology, that what it is like to intuit that p differs from what it is like to intuit that q, it would have
been open to me to say that intuition has ‘phenomenal intentionality’: intentionality “constitutively deter-
mined by phenomenology alone” (Horgan and Tienson 2002; see also e.g. Pitt 2004). We would then have
been able to claim that intuition is a psychological kind very deeply determined by its phenomenal charac-
ter. But we should go where the arguments take us, and, I have argued, that is just not where they lead: we
have good reason to think that intuition lacks content-specific phenomenology.

This leaves us with the view that although phenomenal character is important to the nature of intuition—it is
one of its essential features (§3.1)—it does not exhaust it. We also need an account of what it is in virtue of
which an intuition has the particular content it has: of what it is in virtue of which it is the intuition that p
and not the intuition that q.

If the above arguments (§4.1) are correct, we need such an account anyway. 133 For even if some mental
states have phenomenal intentionality, many contentful mental states do not. When I think that p, there is
something in virtue of which I think that p and not instead that q. But if those arguments are correct, that
something is not the phenomenal character of that state.

To complete the picture of the nature of intuition, we need an account of representational content, which
of course we want anyway. What I have provided is an account which, when coupled with that account, will
answer the question of what the nature is of the mental states we have identified. That constitutes a signifi-
cant advance in our understanding of the nature of intuition, even if it does not take us all the way there.

132
Thanks to John Bengson for discussion here.
133
See Pautz (2013) for further objections to phenomenal intentionality.

79
Chapter 5 Phenomenalism

Above I have outlined and advocated a theory of the nature of intuition. On this account, intuition and
perception are both conscious experiences with representational content and a characteristic phenomenal
character. From now on I will assume that that account is correct.

In this chapter I argue that Phenomenalism about intuition is true: intuition singlehandedly justifies belief
in its content in virtue of its phenomenal character. Here again is that argument in outline.

Perception singlehandedly justifies belief. The best explanation of this is that it does so in virtue of being a
conscious experience with a certain phenomenal character. The relevant aspects of the phenomenal charac-
ter of perceptual experience are also aspects of the phenomenal character of intuitional experience. Intui-
tional experiences therefore also singlehandedly justify belief, unless some of the non-epistemic differences
between intuition and perception make an epistemic difference. But none of them do, so intuitional experi-
ence also singlehandedly justifies belief.

That’s the outline; now let’s get into the details.

5.1 Perception singlehandedly justifies belief

In an argument by inference to the best explanation (IBE), the exact nature of the explanandum is very im-
portant. This book doesn’t engage with scepticism: a starting point assumption is that perception justifies
belief about the way things are. But the explanandum is stronger than that: it is that perception singlehand-
edly justifies belief. This point is not uncontroversial and cannot be assumed, so I now argue in its favour.

To say that perception singlehandedly justifies belief is to say that if certain conditions are met, simply hav-
ing a perceptual experience can make a person justified in believing what it represents, without the involve-
ment of any justification the person has to believe some other proposition. We need not and should not say
that all justification arising from perception is singlehanded: the claim I advocate is that perception pro-
vides some justification singlehandedly.

Consider the first perceptual belief you form in the morning: perhaps that the light is on, or that there is an
unpleasant sound source nearby. Do you need to recruit your justification for some other belief, or any-
thing else at all, besides your visual or auditory experience, to be justified in holding it? It seems clear that
you don’t: having the experience suffices to justify you. Not only do we often in fact not recruit any other
justification; sometimes, like when we have just woken up, we couldn’t even if we tried.

In work I’ve referred to several times already, Pryor argues that we should take such appearances at face
value absent strong positive reasons not to (2000: 536); that there are no such; and thus that we should
acknowledge that perception singlehandedly justifies belief.

80
Pryor’s reasons for thinking that there are no good reasons not to take such appearances at face value in the
case of perception apply equally in the case of intuition. 134 In later chapters I respond to some of the objec-
tions to Liberalism that I take to be the most significant, thereby further strengthening the claim that there
is no good reason not to. It is not possible to outright demonstrate that there are no good objections, since a
powerful new objection might always be waiting in the wings. Still, Pryor’s simple argument lends signifi-
cant credence to the claim that perception provides some singlehanded justification for belief.135

We can also support the claim more directly. To deny that perception provides some justification single-
handedly is to claim that perception always provides all its justification with the help of the justification the
subject has for some other proposition. Which proposition, though? Not just any proposition is a candi-
date: my experience that there is a cup in front of me needs no help from my justification to believe that
there is an infinite number of primes to justify my belief about the cup. The proposition would have to ei-
ther directly concern, or at least bear on, perception itself.

One suggestion is a proposition about how the perceptual system works. For reasons I detail below (§5.5.2),
I don’t think that this will work.

A better candidate is the proposition that the perceptual system is generally reliable. Justification for this
proposition might be thought to be widely attainable. A person could note her own repeated success in act-
ing on perceptual beliefs and thereby gain inductive justification to believe that her perceptual system must
be generally reliable, lest the string of successes be unexplainable and miraculous. But then, given that there
is no denying that one could get justification to believe this proposition in this way, and that this would
strengthen one’s justification for perceptual beliefs, why not think that all perceptual justification relies on
inductive support? Why think perception provides singlehanded justification at all?

For two reasons. First, there are clear cases of perceptual justification without inductive support for the reli-
ability of the perceptual system. Second, the resulting picture is incoherent. Let’s take these points in turn.

5.1.1 Perceptual justification without induction

There are at least two types of clear cases of perceptual justification absent inductive support. The first is
human children. Our best evidence suggests that children obtain the capacity to reason inductively around
the age of four (Perrett 2015; Davidson 2019). Holding that inductive justification for the belief that per-
ception is reliable is required for perceptually justified belief would commit one to the view that children
younger than this have no perceptually justified beliefs. Since they do, (inductive) justification for the belief
that that perception is reliable cannot be required for perceptually justified belief.

Adults also have perceptual justification in cases where their cognitive capacities are so impaired that they
cannot make use of inductive evidence for the reliability of their perceptual system. Being really drunk stops
a person from reasoning competently, but it does not stop her from acquiring perceptually justified beliefs,
at least not of a simple sort. 136

134
The bulk of his argument consists in distinguishing his view from other nearby views to which there are strong objections. This
carries over. His view, and my view, are not, for example, that perceptual (or intuitional) beliefs are infallible or indubitable (Pryor
2000: 532-3), nor that the justification is indefeasible (533). It is not that the propositions believed are self-evident or self-justifying
(533), nor that they are capable of being believed all on their own, ‘autonomously’ (533-4). Because Liberalism isn’t committed to
any of these claims, objections against them do not affect it.
135
This claim is weaker than the one Pryor defends since it concerns singlehanded, not immediate justification.
136
It also blocks competent use of stored knowledge.

81
It may be objected that I am confusing propositional and doxastic justification. Clearly many people don’t
explicitly believe that the perceptual system is generally reliable and so a fortiori don’t believe so with doxas-
tic justification: most people just don’t stop to consider the reliability of the perceptual system. (Nor should
they, of course.) But while failing to hold a belief rules out doxastic justification, it does not rule out propo-
sitional justification. So the objection might be that both infants and adults have propositional justification
for belief in the reliability of their perceptual system, even if they haven’t actually formed that belief.

The relationship between propositional and doxastic justification is often understood as the latter obtaining
if i) the former does, and ii) the agent believes on the basis of that which provides propositional justifica-
tion. However, John Turri (2010) has convincingly argued that this account is subject to counterexamples,
and that a more plausible account says that whenever a person is propositionally justified this is because she
currently has a way of becoming doxastically justified.

As Turri notes, the ‘way’ is moderately idealised (324), so if a person is temporarily blocked from using her
normal abilities—because she’s drunk, say—she can still be propositionally justified. Thus the adult may have
inductive propositional justification for the belief that perception is generally reliable in virtue of normally hav-
ing a way of becoming doxastically justified in believing that proposition, even if she can’t do so at the time.

But we can’t say this for the child.

The idealisation is only moderate, and relative to “typical performance by a competent member of the
agent’s kind” (Turri 2010: 324). The relevant kind can’t be human beings generally, because this would en-
tail that young children have propositional justification to believe things they can’t yet even conceptualise.
For example, an adult may be perceptually justified in believing that there’s a possum on the roof in a situa-
tion where a child is only justified in believing that an animal is there.

Any reasonable candidate kind must be relativised to a stage of cognitive development, and, on pain of simi-
lar counterintuitive consequences, the stage must be quite narrowly construed. Inevitably, then, for some
children the kind is relativised to a stage at which humans can’t yet reason inductively. Such agents would
lack both doxastic and propositional inductive justification for the proposition that their perceptual system
is generally reliable, because they have no way of becoming doxastically justified in holding that belief. So,
at least modulo this conception of the relationship between doxastic and propositional justification, the
conclusion that children have perceptually justified belief without having inductive justification to believe
that their perceptual system is reliable, stands.

Even without that conception, however, there are clear cases of perceptual justification without inductive
support. Suppose that a person wakes up after trauma with no memory of her past. Even with an intact ca-
pacity for reasoning, she would have no data about the reliability or otherwise of her perceptual system to
draw on, and so could not possibly have inductive justification of any kind, neither propositional nor doxas-
tic, to believe that her perceptual system is generally reliable. 137 Since she clearly would still have perceptu-
ally justified beliefs, at the very least of a simple sort, we further strengthen the conclusion that perceptual

137
On the standard conception of the relationship between propositional and doxastic justification she does not have inductive
propositional justification to believe in the general reliability of her perceptual system, since she (ex hypothesi) has no data that would
warrant an inductive inference, whether or not she makes it. On Turri’s conception she also does not, since the has no way to be-
come doxastically justified. And note that this holds even given the moderate idealisation Turri discusses: she might have lost the
ability to form long-term memories, for example.

82
justification can obtain in the absence of (inductive) justification to believe in the perceptual system’s relia-
bility.

5.1.2 Incoherence
The phenomenalist claims that perception provides some singlehanded justification to believe what it repre-
sents, but leaves open that such justification may be strengthened by inductive evidence for the reliability of
perception. The objection under consideration says that in that case we might as well think that all percep-
tual justification relies on inductive support for the reliability of perception, and thus get rid of single-
handed justification altogether.

To get inductive justification to believe that her perceptual system is reliable a subject would need justifica-
tion to believe that particular beliefs justified by perception are true. Once a sufficient number has accumu-
lated, and provided a favourable ration of true to false beliefs, she would have inductive justification to be-
lieve that a later belief was true, too.

But how could she become justified in believing for any one perceptually justified belief that it was true?
The only candidate justifier would itself be perception, or trace back to perception. But then justification
couldn’t get off the ground. At the start of the process there is no inductive justification, and if all percep-
tual justification is inductively supported, one couldn’t acquire any, either.138 So the suggestion is incoher-
ent, and can be rejected.

5.1.3 Taking Stock

I have raised two mutually supportive considerations in favour of the claim that perception provides at least
some justification singlehandedly. It seems that it does, and absent good reason to believe otherwise we
should take these appearances at face value. And there are clear examples of people with perceptual justifi-
cation but who lack any justification that could plausibly be brought to bear in support.

The incoherence point discussed last shows that not all perceptual justification can have inductive support.
It can’t on its own show that justification to believe some other proposition isn’t part of what makes a per-
ceiver justified—after all, such justification could be a priori—so the other two lines of argument are im-
portant.139 The description of cases, in particular, is forceful: it is hard to deny that these are cases where no
justification to believe some other proposition is involved in making the agent justified. Instead, perception
on its own does this: it provides at least some justification singlehandedly.

5.2 Perception Justifies Belief In Virtue of Its Phenomenal Character

If having a perceptual experience is what makes a perceiver justified in believing what it represents, then to
explain how perception provides justification we must look to its nature. What must perception be like to
have the capacity to make the perceiver justified?

138
See (Smithies 2019: 75, 80). According to pure coherentism, a set of beliefs can become justified in virtue of its internal coher-
ence alone. However, there are good reasons, both formal and informal, to reject such a view (Fumerton 1999; Bovens and
Hartmann 2003: §1.4; Olsson 2005: Appendix B).
139
Or it might be the (not so easy to classify) justification an agent is sometimes claimed to simply have, not in virtue of having done
anything to achieve it, indeed, not as a result of having acquired it at all, but simply per default, ‘for free’ (Wright 2004).

83
The central contention of this book is that perception and intuition justify belief in virtue of their phenom-
enal character. More fully, they justify belief in virtue of being conscious experiences with a certain phenom-
enal character and with representational content. 140 In what follows I’ll usually stick with the shorthand.

Having representational content can’t be all there is to provide justification, since many states have content
without doing so. There must be something else to the nature of perception that enables it to justify belief
in what it represents. This already points to phenomenal character, for content and character plausibly ex-
hausts the nature of perceptual experiences: if a state has content and the phenomenal character characteris-
tic of perceptual experience, then it plausibly is a perceptual experience.

To this one might object in various ways, for example by claiming that a state isn’t a perceptual experience
unless it has a certain aetiology. Above I argued against that claim for intuition (§3.6), and I think the argu-
ment carries over. But I needn’t press this point here, because even if there is more to the nature of percep-
tion, phenomenal character is what matters for its epistemology.

One reason to accept the view that perceptual experience justifies belief in virtue of its phenomenal charac-
ter is that it gets important distinctions right while also explaining what other views get wrong. We just con-
sidered and rejected the view that perceptual justification only makes a person justified in believing what it
represents with the aid of that person having justification to believe that her perceptual system is reliable.
That view makes justification too hard to come by, because it requires too much of the perceiver. 141 Agents
who lack justification to believe some proposition about how the perceptual system works, even agents who
lack justification to believe that the perceptual system is reliable, still derive justification from perception.

There is no corresponding problem for Phenomenalism: instead that account cuts the cake exactly right.
The point at which an infant starts enjoying perceptual experiences properly speaking—with (determinate
enough) representational content and the right phenomenal character—is plausibly exactly the point on-
wards from which she acquires perceptual justification. Moreover, it is plausibly precisely because she now
enjoys such experience that she acquires it. 142 Conversely, the point at which an increasingly inebriated per-
son no longer enjoys experiences with the crucial phenomenal character is the point from which she loses
the capacity to become perceptually justified. Again this is plausibly precisely because she no longer enjoys
them.

Similarly, the account both gives the right verdict, and explains, famous challenges to reliabilism about justi-
fication, here stipulated as the view that a belief is justified if and only if it results from a reliable belief-

140
I don’t take a stance on the nature of representational content in this book (§4.9). Everyone needs an account of content, and
although it hasn’t proved easy, to put it mildly, to provide one (see e.g. (Neander 2004/2018; Adams and Aizawa 2010/2017;
Bourget and Mendelovici 2016/2019), we know that an account must be there to be had, and once we have it for some mental
states, we can apply it to all of them.
141
A possible exception is default entitlement views like the one considered noted in n. 124.
142
Even if we could settle whether an infant has an experience with the right character, uncertainty might still remain as to whether
their experience has representational content, and especially whether the content is determinate enough for a justified belief to re-
sult, assuming that there are restrictions along such lines. I set such problems aside throughout: the claim is that whatever perceptual
and intuitional experience represents (when they represent something determinate enough, if there are such restrictions), belief in
that content is justified by having the experience. Also, and to be abundantly clear, while I take there to be legitimate questions
about what is and is not part of the content of various perceptual (and intuitional) experiences, to me the question of what content
is justified ends there: perceptual and intuitional experiences justify belief in all of their content.

84
forming process.143 The new evil demon problem concerns a disembodied brain in a laboratory vat, stimu-
lated by a malignant neuroscientist to have a normal stream of perceptual experience (Lehrer and Cohen
1983: 192-3). This person’s perceptual experiences are completely unreliable, so according to Reliabilism,
none of her perceptual beliefs are justified. That’s the wrong conclusion: instead they are false but justified.
On the other hand, if the scientist’s machinery were to temporarily malfunction, and instead induce experi-
ences with the phenomenal character of visual iconic imagination, for example, the person would no longer
be justified in believing what her experience represents.

Conversely, the scientist might instead secretly install in a normally embodied person a device which relia-
bly causes true belief about the colour of the clothing of the person walking immediately behind her, beliefs
not associated with perceptual experience, and the veracity of which she never bothers to check. 144 Reliabil-
ism entails that those beliefs would be justified, which is the wrong result. But if the person started enjoying
perceptual experiences representing the person behind her, then—bracketing some scepticism she probably
should have about the sudden onset of this new capacity—she would then be justified.

In both cases the thesis that perceptual experience justifies belief in virtue of its phenomenal character ex-
plains what is going on. The BIV is justified because simply having perceptual experiences—which she does
have—is what makes a person justified. 145 The colour-believer is not, because she doesn’t. And in both cases
the verdicts are reversed when the facts about experience change, and for that very reason. So perceptual expe-
rience justifies belief because of its phenomenal character.

Another powerful argument to this effect is due to Declan Smithies. 146 Smithies takes point of departure in
the phenomenon of blindsight, a phenomenon first reported in 1974 (Sanders et al. 1974). These are cases
in which a person lacks visual experiences from certain areas of their visual field, and yet retains remarkable
abilities: when prompted to choose among alternatives they can accurately report on “position, movement,
orientation, simple shapes, colors, and emotions” (Smithies 2019: 76) from objects in these areas. Some can
even perform sophisticated actions that rely on information from these areas, such as catching a ball, or
turning a card to the correct orientation so as to ‘post’ it in a thin slot (ibid).

In what kind of epistemic situation are people vis-à-vis the states of affairs about which their cognitive sys-
tems, evidently, receive some information, but which are not reflected in their visual perceptual experience?
Smithies argues that their situation is relevantly similar to what yours would be if the following were to tran-
spire:

Suppose you wake up in the hospital feeling normal. The doctor holds her hand in front of your
face and asks you to count how many fingers she is holding up. You answer correctly and she gives
you some candy as a reward. Now she holds her hand just outside your conscious field of vision
and asks you to count how many fingers she is holding up. You tell her you can’t see her hand, but
she asks you to guess anyway. You say, “What the hell,” and hazard a guess. To your surprise, she
says you got the right answer and gives you some more candy. The experiment continues until
you’ve eaten so much candy you begin to feel sick (77-8).

143
See (Smithies 2011a, 2011b) for different variations of this view.
144
This example is adapted from BonJour’s example of Norman the clairvoyant (1985: 41).
145
This lends further support to the claim that perceptual experience isn’t subject to aetiological restrictions.
146
Smithies has developed the view that perception justifies belief in virtue of its phenomenal character over a period of years, be-
ginning with his doctoral dissertation (2006), and continuing in (2011b), (2014), and (2016). Here I focus on his most recent
presentation, from his recent book (Smithies 2019: §§3.1-3.2 & 7.3).

85
Here you have unwittingly had a camera installed in your head, wired so as to make your answers track real-
ity, but you don’t know this. The intuition is clear: your belief about the number of fingers being held up is
justified in the first case, when you have a normal visual perceptual experience as of the hand and the fin-
gers, but not in the second and subsequent cases. The same holds for the blindsighter: she has no justifica-
tion for belief about objects in her blind areas.

That conclusion can be further strengthened, Smithies argues, by noting that blindsighters typically don’t
form beliefs corresponding to their guesses. If the information from the blind areas did justify belief they
would (often) be rationally deficient in withholding judgment.147 But they are surely never rationally defi-
cient thereby: instead, by withholding belief they are doing exactly what they ought to do. So again we reach
the same conclusion: the information in the blind areas doesn’t justify belief (80-2). When a blindsighted
and a normally sighted subject are both presented with the same scene, and both form a belief about some-
thing in the former’s blind area, only the latter’s belief is justified.

Smithies aims to attribute this epistemic difference to the difference in phenomenal experience between the
two agents, but there are also other differences. There are, in particular, functional differences: not only
does the information the perceptual system receives from the blind region fail to produce conscious experi-
ence, it is also not readily available to play the role perceptual information is normally available to play. In
Ned Block’s terms, it is not ‘poised’ for direct control of thought and action (1995: 231). So perhaps the
difference in justification can be ascribed to this difference rather than to the difference in phenomenal ex-
perience.

Smithies convincingly argues against this idea (2019: 84-90). Actual blindsighted subjects do differ from
normally sighted subjects in this way, but other possible subjects do not. In fact, we can imagine augment-
ing a blindsighter in any functional way whatever, and yet the conclusion about her epistemic state remains
the same. For instance, we can imagine a ‘super-blindsighter’ who doesn’t need to be prompted in order to
form beliefs corresponding to the information in her blind spots, but who does so spontaneously. 148 Such a
person would no more be justified in her beliefs than the initial blindsighter. Thus we are forced to con-
clude that the normally sighted person is in a superior epistemic position because she has a perceptual experi-
ence in the relevant part of the visual field.

My current aim is to argue that perception justifies belief because of its phenomenal character. But this is
not the conclusion Smithies advocates here. Smithies presents this argument in defence of ‘the phenomenal
condition’, which says that “[n]ecessarily, perception justifies belief about the external world if and only if it
has some phenomenal character” (2019: 82, emphasis added). He elaborates:

I’ll argue that blindsight cannot justify belief about the blind field because it has no phenomenal
character. In contrast, normal human perception justifies belief about the external world only be-
cause it has some phenomenal character. In other words, the presence or absence of phenomenal char-
acter is what explains the justificational difference between blindsight and conscious sight (2019:
82, emphasis mine).

147
It’s not plausible that we are rationally required to form beliefs corresponding to all our justification (put differently: that we are
required to convert all propositional justification into doxastic justification), since there is much we have justification to believe but
in which we have no rational interest. But this doesn’t matter here, since we can easily modify the example so that enough is at
stake for the blindsighter to make belief formation rationally required.
148
As Smithies notes, this conception of a ‘super-blindsighter’ differs from that of Ned Block, who introduced the term. Smithies
goes on to convincingly argue against further augmentations, e.g. a super-duper-blindsighter.

86
It later becomes clear that Smithies doesn’t actually mean what he says here, but it is nevertheless worth
pausing to note just how absurd this principle is. It’s emphatically not just having some phenomenal character
or other that makes the epistemic difference! The blindsighter’s situation would not be improved, at least not
in the epistemic dimension, if the information from the blind spots were accompanied by the phenomenol-
ogy of orgasm, or euphoria, or joyful giddiness. The real takeaway from the case of blindsight is that the ep-
istemic difference stems from the particular phenomenal character that perceptual experience has.

This underscores the importance of correctly identifying the phenomenal character that makes the epis-
temic difference. As we’ve seen, the epistemically relevant phenomenal character must be attitude-specific,
since phenomenal character is supposed to explain how experience justifies belief, but no explanation is pos-
sible if the relevant phenomenal character is infinitely variable (§4.3). So our task now is to identify the best
candidate for an epistemically significant attitude-specific phenomenal character of perceptual experience.

5.3 Two Unsuccessful Attempts

To this end, I’ll consider two unsuccessful proposals for what that phenomenal character might be. This
will allow us to draw out a structural deficiency, and to formulate two criteria of adequacy for Phenomenal-
ist theories of perceptual and intuitional justification.

5.3.1 Seeming Able to ‘Just Tell’

Recall that an important historical precursor for the view of perceptual and intuitional justification devel-
oped in this book is James Pryor:

[What] explains why our experiences give us the immediate justification they do . . . [is] the peculiar
“phenomenal force” or way our experiences have of presenting propositions to us. Our experi-
ence[s] represent propositions in such a way that it “feels as if” we could tell that those propositions
are true— and that we’re perceiving them to be true—just by virtue of having them so represented
(2000: n. 37).

And:

I think there’s a distinctive phenomenology: the feeling of seeming to ascertain that a given proposi-
tion is true. This is present when the way a mental episode represents its content makes it feel as
though, by enjoying that episode, you can thereby just tell that that content obtains. … When you
have a perceptual experience of your hands, that experience makes it feel as though you can just see
that hands are present (2004: 357).

Let’s call this the phenomenology of ‘seeming able to just tell’.

To see that an explanation in terms of this phenomenal character cannot succeed, consider the following
example:

Blizzard. Anne is stationary on a flat, snow-covered plain in a blizzard. The wind is whipping snow
around in all directions. No features of the landscape are visible. Anne can barely see her own
knees, and she cannot see the tips of her skis.

87
 Someone approaches very slowly from the direction in which Anne is looking. At first she’s com-
pletely unable to distinguish the approaching person from patterns randomly forming and dissipat-
ing in the snow. As the person approaches, Ann’s perceptual experience gradually changes, and the
human figure gradually appears more and more clearly.

It is clear that there is a point at which Anne acquires singlehanded justification from her visual perceptual
experience for the belief that there is a person there. 149 Moreover, that point comes well before Ann’s expe-
rience takes on the character of seeming able to just tell that there is.

One could resist this by insisting that before Anne’s experience takes on the character of seeming able to
just tell, her experience doesn’t even represent that there is a person there: at most it represents that there is a
person-esque shape in the snow, or something of that nature.

But that response is unmotivated. Why think that all uncertainty associated with perception is contained
within in its content? It is true that there is a point at which Anne’s experience represents (at most) that
there is a person-esque shape in the snow: in the beginning Anne can’t distinguish the person from random
patterns in the snow, after all. However, her perceptual experience soon enough beings to represent that a
person is there, and well before the experience has the character of seeming able to just tell.

A second reply insists that Anne doesn’t get any singlehanded justification from her perceptual experience
before it has the character of seeming able to just tell. This, too, is very hard to believe. There’s a point at
which Anne’s perceptual experience represents that there’s a person there, well before her experience has
that character. At that point, why think she doesn’t get any singlehanded justification? Justification, after all,
comes in degrees. Even if she doesn’t get very much, she surely gets some.

Here’s another case:

Increasing Illumination Bob is led blindfolded into a room with black walls, ceiling and floor, and
in which there is a black table but no other objects. He’s left alone in the room at some distance
from the table, the light is turned off, the doors are closed, and he’s told to remove the blindfold.
He’s now standing in a completely dark room with his eyes open. He can’t see a thing.

Soon the illumination begins to increase very slowly. At first, Bob is completely unable to discern
the table against the background. As the illumination increases, his visual perceptual experience
changes: the table gradually appears more and more clearly. 150

Similarly to the previous case, in Increasing Illumination there is a point at which Bob acquires some single-
handed justification to believe that there is a table there, and this point clearly comes well before his experi-
ence takes on the character of seeming able to just tell.

What explains why Anne and Bob get singlehanded justification from their perceptual experience? It can-
not be that their experiences have the character of seeming able to just tell, since when this first happens, it
doesn’t. The explanation in terms of that character is therefore at best incomplete.

But we can say something stronger. It is very plausible that Ann and Bob get justification from their experi-
ences for the very same reason that we get justification from perceptual experience in other cases. After all, we

149
Or that a person approaches; or some similar proposition. For present purposes it doesn’t matter exactly what we take the con-
tent of the experience to be, since it is bound to be something along these lines.
150
Thanks to Alan Hájek for drawing my attention to examples involving dimmed light. He discusses such examples in his (2016).

88
have every reason to think that Anne’s and Bob’s visual perceptual experiences share central aspects of their
phenomenal character with more everyday visual perceptual experiences: they obviously share the character
of ‘visualness’, for example.

It is true that Blizzard and Increasing Illumination concern slightly unusual situations. However, we can eas-
ily construct cases that bridge the divide between these and more garden variety perceptual experiences:

Judging Distance: Carol is seated with her chin on a long table. Stretching out in front of her are
two flat rods. The left rod is fixed to the table, the one on the right can slide back and forth.
Carol’s line of sight is parallel to the rods, each of which has an indicator arrow some way down.

To begin with the indicators are clearly at different distances from Carol: her task is to move the
right rod until the two appear equidistant. As she carries this out, her experience changes. After a
while she stops: this is, the thinks, her best shot at solving the task, but it doesn’t seem to her that
she can just tell that the two markers are equidistant from her.

In all relevant respects, this case is just like experiences many of us have actually had, for example that of
looking out along a straight stretch of road towards a point at which one person waits for another, who ap-
proaches. In both cases one at some point gets singlehanded justification for the belief that the two are
equidistant from oneself, and yet neither experience has the character of seeming able to just tell.

The fact that all these experiences share significant aspects of their phenomenal characters, and the easy
bridge we can make from the initial cases to perfectly ordinary perceptual experiences, support the view that
what explains why we acquire singlehanded justification from perceptual experiences generally is just the
same as that which explains why Anne, Bob, and Carol acquire it from their experiences. It follows that the
explanation in terms of the seeming able to just tell is not just incomplete, but incorrect.

Moreover, not only does Anne get some singlehanded justification before she has phenomenology of seem-
ing able to just tell; her justification gradually gets stronger as the person gets closer (and mutatis mutandis for
the other cases). Her perceptual experience gradually changes, and her credence that a person is there gradu-
ally increases. The latter is clearly epistemically appropriate, and, equally clearly, it is epistemically appropriate
precisely because of the gradual change in the former.

The deep problem with explaining singlehanded justification by means of seeming able to just tell is that it is
a binary notion. When I have an experience with the representational content that p, either it seems to me
that I can just tell that p, or it doesn’t. But justification is a matter of degree: I can have more or less of it.
This holds true in general, and for justification from perceptual experience, in particular. This is a princi-
pled reason to think that Pryor’s notion is ill-suited to do the job.

5.3.2 ‘Presentational’ Phenomenology

Above I argued that ‘presentational’ accounts of perception and intuition are inadequate (§4.8). A further
problem with such accounts parallels that which I just raised for Pryor: justification from perception and
intuition comes in degrees, but presentational phenomenology is binary.

That the justification provided by these experiences comes in degrees is very clear in the case of intuition.
Intuitions come in all different strengths corresponding to variation in the justification they provide: strong
intuitions justify high credences, weak intuitions justify low credences, etc. However, although the point is
often overlooked for perception the claim also holds true there, as the cases I have just given demonstrate.

89
Delivering this result is a sine qua non, a criterion of adequacy on a theory of justification. Let’s call deliver-
ing the result that the justification perception and intuition provide comes in degrees the weak criterion of
adequacy on such theories.

However, for a defender of Phenomenalism a stronger criterion of adequacy should also apply: the result must
be delivered by facts about phenomenal character. Phenomenalism, after all, says that the two central facts
about certain experience types are that they justify belief in their content, and that they do so because of the
states’ phenomenal character. It is undeniable that having a weak intuition feels different from having a
somewhat stronger one, and different again from having a very strong one: there is a graded variation in
phenomenal character corresponding to the intuition’s strength. And again, while perhaps less obvious
than for intuition, this also holds true for perception, as the above examples demonstrate.

It would make little sense to hold that, although justification and phenomenal character vary together in
graded fashion, and although perceptual and intuitional experience are epistemically powerful in virtue of
their phenomenal character, the justification the experiences provide doesn’t vary because the character
does. But no presentational theory can say that it does, because presentational phenomenology is binary. All
presentational theories fail the stronger criterion of adequacy, and are thereby ruled out.

Consider again Elijah Chudnoff, who endorses Phenomenalism about intuition and about perception
(Chudnoff 2011b, 2011a, 2012, 2013b). Chudnoff applies the distinction between fact perception—seeing
that the rocket has launched—from object-perception—seeing the rocket, 151 and holds that there are corresponding
phenomenal properties: those of seeming to fact-perceive and seeming to be sensorily item-aware, respectively.
Meanwhile, the phenomenal character perception and intuition share, and which accounts for their being
epistemically powerful experiences, is, he says, their presentationality:

Presentational Phenomenology of Perception: [A] perceptual experience possesses presentational

phenomenology when in it you both seem to fact-perceive that p and seem to be sensorily item-
aware of an item that makes it the case that p (2011a: 320).

Presentational Phenomenology of Intuition: [A]n intuition experience possesses presentational

phenomenology when in it you both seem to fact-intuit that p and seem to be intellectually item-
aware of an item that makes it the case that p (2011a: 323).

As here defined presentationality is a conjunctive property, so demonstrating that one of its component
properties is binary would suffice to show that presentationality itself is. But in fact both components are:
you either seem to see that the rocket has launched or you don’t; you either seem to be aware of the rocket,
or you don’t. So this theory does not meet the strong criterion of adequacy. Since, as far as I can tell, the
theory has no other resources with which to deliver the result that intuition provides graded justification,
either, the view doesn’t even meet the weaker one. 152 This is an additional reason why the view is ruled out.

151
The distinction goes back at least to Chisholm (1957) and Dretske (1969).
152
In some places, Chudnoff says things that seem to indicate that he wishes to place all the gradability of intuition in its content,
as Bengson does. If so, my objections to this strategy—presented shortly—will apply to Chudnoff’s view, as well. For example, Chud-
noff writes: “You are in a club looking for your friend: your eyes are watery, the room is smoky, the lights are dim, and the crowd is
dense. You seem to see your friend in the corner. But your seeming awareness of the location of your friend is obscure. Something
similar can happen in intuition. You might intuit that there can be no moral difference between two actions without a non-moral
difference between them. And you might seem to be aware of a truth-make for this claim. But your seeming awareness might be
obscure” (2013b: 60). And: “Suppose you have a visual experience as of a person-like shape in bad lighting. Maybe this gives you

90
John Bengson also holds a presentational view (2010, 2015b). Recall that he distinguishes mental states that
are merely contentful, such as hopes, desires, and wishes, from those that are representational, and that he dis-
tinguishes the merely representational states—such as beliefs and ‘acceptances’, characterised by endorsing a
content, or being under the impression that something is the case—from the presentational, like perception
and intuition, which present contents as being the case, and correspond to having the impression that such
and such is the case.

Bengson and Chudnoff thus agree that the core similarity between intuition and perception is that both
have presentational phenomenology, but they give very different accounts of the phenomenal character in
question. Bengson too endorses Liberalism about perception, and intuition.

But as with Chudnoff, the important notions for Bengson are clearly binaries. You either have the impression
that such and such is the case, or you don’t, for instance (2015b: 716-19). Since these are the notions in
terms of which the phenomenal character of intuition and perception are elucidated, this account fails the
strong criterion of adequacy.

Does this account also fail the weaker one? Bengson does try to account for the gradable justification intui-
tion provides:

Presentational states are gradable: their overall quality may vary in different situations, depending
upon the manner in which they present in those situations (e.g. more or less clearly, vividly, etc.).
All else being equal, the overall quality of a presentational state such as perceptual experience or
intuition is in some sense better when, say, one is not distracted and has time to scrutinize the
scene or proposition in question than when one is distracted and rushed. In the former case, one’s
perceptual experience or intuition is likely to be clear or vivid (e.g. it is clearly or vividly presented
that there is a red apple on the table, or that Smith does not know); in the latter case, it is likely to
be hazy or fuzzy (Bengson 2015b: 721).

There’s plenty of room for interpretation here: it is not at all clear how we should understand the claim that
“the overall quality” of an intuition is “in some sense better” if the intuiter is appropriately focused and not
rushed, for example. Perhaps one could interpret the ‘manner in which [intuitions] present’ to gesture in
the direction of attitude-specific phenomenology that comes in degrees, in the way I have claimed that phe-
nomenology of pushiness does. In that case I take the same position as above (§4.8), namely that insofar as
Bengson’s account overlaps with mine I (unsurprisingly) agree, but criticise a less-than-optimally perspicuous
description.

But I don’t think that this is the best interpretation. An experience is clear, on the one hand, or hazy or
fuzzy, on the other, just in case its content is those things. So Bengson’s account is best understood as confin-
ing the variance in intuition and perception to the states’ content, and thus to not even try to meet the
stronger criterion of adequacy.

Now, there clearly is such variation: there are cases of intuition in which it seems to us that something like p is
the case, but where we can’t get clearer than that. Perhaps for some subjects that is what is going on with

weak justification for thinking there is a person there. But it is not like an experience that puts you in a position to know that there
is a person there. In this case, however, it seems to me that you have weak justification for believing a strong proposition—that there
is a person there—because you have strong justification for believing a weak proposition—that there is a person-like shape there”
(Chudnoff 2013b: 93).

91
the intuitions about incestual sex mentioned earlier, for example: perhaps it seems to the subjects in these
experiments that something is wrong here, but nothing clearer than this.

The problem is that variation in content cannot account for anywhere near all the variation in strength of
justification that there we need to account for, so the theory fails the weak criterion of adequacy, too. 153

First of all, the theory is supposed to apply to perception just as much as to intuition, and, as we have al-
ready seen in detail, perception is gradable in a way that cannot be accounted for merely by variation in the
representational content of that state.

To see the point in the context of intuition, consider Tim. Many people, Tim included, have the intuition
that in war it is at least sometimes wrong to kill innocent civilians in order to save the lives of soldiers.
Some people’s intuitions are absolute, but for Tim, the numbers matter. If the numbers of innocent civil-
ians killed and soldiers saved by an action are equal—for example, if each group has one member—Tim has a
strong intuition that the action is impermissible. But he also has a strong intuition that it’s not only permis-
sible but obligatory to take one civilian’s life to save the lives of one million soldiers.

In between are many intermediary cases. As the ratio of soldiers to civilians becomes larger, at some point
Tim has a weak intuition that the action is permissible. It is very plausible that Tim thereby gets some single-
handed justification to believe that the action is permissible, just by having the experience. Certainly, that’s
what a theorist friendly to Phenomenalism should say. True, he gets less justification than if the intuition
were strong, but he does get some. As the ratio increases, his intuition gets stronger, and so, too, does his
justification for believing that the action is permissible. And yet, no unclarity is involved.

So there is more variability to the strengths of intuitions, and to the strength of justification intuitions de-
liver, than there is to the clarity of otherwise of their contents. Even when there’s nothing unclear, hazy, or
fuzzy about the content of an intuition, it can still be weak, and then deliver only weak justification. Con-
versely, it can be both hazy and strong, as in the incest case, for example. It then plausibly delivers strong
justification, though it might be hard to tell for exactly which belief. 154 So Bengson’s account, too, fails on
both criteria of adequacy. 155

153
This objection is also effective against Huemer’s account, at least insofar as he relies on some things “more clearly seem[ing] true,
than others (Huemer, Ethical Intuitionism, 105).
154
One possibility is that the justification the intuition provides must be somehow divided between the propositions that have a
reasonable claim to be the content of the intuition.
155
Another proposal is due to Declan Smithies (2019: §3.3), whose account we have already discussed. As we have seen, Smithies
ties his argument to representationalism, the thesis that every phenomenal property is identical to a representational one. One
point of difference between his view and mine is that I reject representationalism. However, for Smithies’ version things get tricky,
since representing a content with force to him is a part of what constitutes the state’s representational property, whereas for me that
counts as attitude-specific phenomenology. But let’s bracket the question of representationalism. For Smithies, the epistemically
relevant aspect of (what in my parlance counts as) attitude-specific phenomenology is presentational force, which “to a first approxima-
tion … you have when it seems to you that you’re presented with the very thing that makes your experience true” (94). This is
strongly reminiscent of Chudnoff’s seeming item-perception, and so runs in to the problems already raised for that account, espe-
cially with respect to intuition. Smithies goes on to discuss Pryor’s account, but finds reasons, different from mine, to object to it.
The best we can do in identifying the relevant phenomenal character, he says, is to point to examples: it’s the type of character you
enjoy when you perceive or hallucinate, but not when you (visually iconically) imagine, or judge: “it’s the kind of phenomenal char-
acter that your experience has when you can just see that hands are present” (94). As I’ve argued in detail above, however, though it
is true that that character is present when it seems to you that you can just tell, those are not the only times it is present: the epistemi-
cally relevant phenomenal character is also present when it doesn’t seem to you that way. Moreover, and although this is perhaps not
completely clear from the text, it sure seems that the character Smithies wishes to afford epistemic weight is also binary, and if so,
his account falls for that reason, too.

92
5.4 Pushiness and Objectivity Explain Justification
The discussion so far underscores the importance of correctly describing the phenomenal character percep-
tual experience has, of actually arguing that it in fact has the described character, and of arguing, further-
more, that having such a character makes a perceptual experience epistemically powerful in a way that meets
the criteria of adequacy I have outlined.

I discharged the first two tasks in Chapter 4. What remains now is to argue that having this character makes
an experience epistemically powerful.

The failure of the accounts discussed above shows that the phenomenal character that explains perception’s
ability to singlehandedly justify belief must come in degrees, but there are further lessons to be learned as
well.

First, although describing the phenomenal character of conscious experience is very difficult, there still
seems to be a systematic way in which these accounts fall short. If it seems to a person that she can just tell
that things are thus-and-so, if a person seems to fact-perceive or fact-intuit that p, if a person seems to be sen-
sorily or intellectually item-aware of an item that makes it the case that p, or if a person has the impression
that p; all of these seem to be results of her conscious experience having a certain character, and not them-
selves aspects of that character. By analogy, when I have a splitting headache it is true that I can just tell that
I do, but that’s because the contribution the local experience makes to the character of my overall experience
is all too impossible to ignore. Me being able to just tell that I do is not itself a description of that character.
When in the dark I can just tell that there is a table before me, when I seem to (tactilely) fact-perceive that
this is so, when I have that impression, that’s a result of my tactile experience having a distinctive phenome-
nal character; but not itself an aspect of that character.

The phenomenology of pushiness and objectivity, as described in detail in Chapter 4, come closer to de-
scribing the phenomenal character of perceptual and intuitional experience itself. ‘Closer’ because this is a
matter of degree, because I myself made use of metaphor in various places, and because my characterisation
surely also fails to get all the way to the phenomena. Still, it is a move in the right direction.

Second, I think that the accounts don’t succeed in explaining what they set out to explain. They gesture at,
or to some degree describe, the character of experience, but don’t explain why having an experience with
that character should justify belief. This is a tall order, but I’ll attempt to make some progress here.

The two points are related. The better we home in on and describe the relevant character the better and
more responsive our theory will become, and the more likely that it will yield a good explanation of why an
experience with that character should be able to provide singlehanded justification for belief.

On the table in front of me is a blue water bottle. According to Phenomenalism, in virtue of having my cur-
rent visual perceptual experience I get singlehanded justification to believe that there is. That perceptual
experience has phenomenology of pushiness and objectivity explains this.

Phenomenology of objectivity provides one part of the explanation. An experience could represent some-
thing as merely imagined. That my perceptual experience is characterised by phenomenology of objectivity
explains why it does not. What perceptual experience purports to represent is not the way things are accord-
ing to my imagination, or something that in some other way dependent on me. What it purports to repre-
sent is something about the way things are independently of me, objectively speaking. That it so purports is
part of the very phenomenology of the experience. Because it has this phenomenology, the subject matter of

93
the experience is the objective way things are. Recalling the conversational analogy, we can describe the situ-
ation by saying that perceptual experience purports to ‘say’ something about the way things are subject inde-
pendently.

But what does it say about this? One thing one can say is that things are not a certain way. Or, still speaking
about the way things are subject-independently, I could ask you to suppose that things are a certain way. I
could say that it’s possible that they are that way, without saying anything about whether they actually are
that way. And so on.

Perceptual experience is not analogous to any of these things. By virtue of having phenomenology of pushi-
ness, perceptual experience purports to inform me that things actually are a particular way. My experience
pushes me to believe that this is how things actually are; it does not merely ask me to suppose that they are,
nor that they might possibly be that way.

Together these aspects of perceptual experience explain why we get singlehanded justification from having
it. For being pushed to accept that things actually are a certain way, objectively speaking; not by an agent—
whose intentions and sincerity one might doubt—but in virtue of one’s own conscious experience, this con-
stitutes justification to believe that things actually are that way. 156

The justification perception provides can only be explained by the phenomenal character of perceptual ex-
perience. It’s hard to see what more, or what else, could be required from that character for a person to be
justified in virtue of having the experience than that it have pushiness and objectivity. In virtue of having
this phenomenal character, a perceptual experience makes it seem to the perceiving subject that things actu-
ally are the way they are represented as being independently of her, objectively speaking. If that could not
give her justification to believe that things actually are that way, it’s hard to see what could. Both aspects are
necessary. Jointly they are sufficient.

Suppose that someone objected that some independent criterion being satisfied is, after all, part of what
makes the subject justified. But this would amount to saying that the experience does not suffice to really
make it appear that things actually are the way they are represented as being, but that the ‘joint appearance’
created by also keeping in mind that some other condition is satisfied—reliability, say—would make it seem
that way. However, the claim is precisely that experience on its own already makes it seem that things really
are a particular way subject-independently. No consideration of other factors being fulfilled is necessary.

That is precisely what the experience having phenomenology of objectivity and pushiness means. Unless the
claim about the phenomenal character of perceptual experience is challenged, the claim about justification
should be accepted.

One strength of this explanation is that these aspects of the character of perceptual experience can be pre-
sent even when a person’s perceptual experience lacks the character of seeming able to just tell. Anne’s expe-
rience can be characterised by objectivity and pushiness before it takes on the character of seeming able to

156
It has been argued (by an anonymous referee) that the phenomenology of pushiness and objectivity is at a dialectical disad-
vantage relative to ‘presentational’ phenomenology, because the latter but not the former “wears its epistemic virtues on its sleeve”.
I see no merit to this claim. The words we Phenomenalists use are all metaphors—we try to describe, as best we can, and given highly
imperfect linguistic and methodological resources—what we take to be epistemically significant aspects of experience. What matters
for how the accounts should be assessed is the plausibility of the accounts on substance, and not the degree to which the label we put
on the character has strong epistemic connotations. If there were a word denoting the character of having reality directly revealed to
you by being allowed to partake in God’s divine insight, no word could be more saliently epistemic, but applying that word for the
character of perceptual and intuitional experiences would not improve those accounts—and indeed quite the opposite.

94
just tell. On my view it must be, if she is to receive singlehanded justification from it before that point, but
the phenomenological description of the situation is also independently plausible.

If Anne’s experience lacks phenomenology of objectivity, she cannot be justified in believing that a person
is approaching, for her experience will then no more justify her than will experiences of imagining a mean-
ingful pattern in the snow—experiences we can suppose that she indulged in shortly before. If her experi-
ence lacks pushiness, it no more justifies than supposing for the sake of argument does. Both aspects of the
phenomenal character of perceptual experiences are necessary. When both are present, Anne acquires sin-
glehanded justification, whether or not the experience has the phenomenal character of seeming able to just
tell, whether or not she ‘has the impression’ that this is so, whether or not she seems to fact-perceive this,
and whether or not she seems to be object-aware of an item which makes this the case. So a further strength
of this explanation is that it allows us to respect the intuition that what gives rise to justification in Blizzard,
Increasing Light and Judging Distance is just the same as what gives rise to it in more pedestrian circum-
stances.

The explanation also allows us to respect the intuition that Anne’s justification changes as her experience
does. The phenomenology of objectivity is binary; it is there or it is not. But pushiness comes in degrees. As the
approaching person gets closer, Anne’s experience pushes her ever harder to accept that there is a person
approaching. This allows us to explain why Anne derives progressively more (or stronger) justification as her
experience changes, and to do so in a way that respects the intuition that this happens because of the gradual
change that occurs in the character of her experience. As we have seen, neither Pryor’s account nor the
presentational accounts of Chudnoff and Bengson can deliver this result. The parallel point holds for intui-
tion, as Tim’s intuitions about soldiers and civilians illustrate.

The explanation in terms of pushiness and objectivity improves on rival explanations in at least four ways.

First, it explains how a subject can acquire singlehanded justification from her experiences in cases when
they do not have the character of seeming able to just tell, or any of the other binary properties.

Second, it accounts for the strong intuition that as the phenomenal characters of Anne’s and Bob’s experi-
ences change, they progressively acquire more (or stronger) justification, and they do so precisely because of
the way the character of their experience changes.

Third, it unifies explanation between unusual cases, like those described in the first two thought experi-
ments above, and more everyday cases of visual perception.

Fourth, Pryor’s claim is that when I have a perceptual experience it seems to me that I can just tell that
things are a certain way. The question is salient, however: why should that give me justification to believe
that things actually are that way? And mutatis mutandis for the presentational accounts.

By contrast, the description I have given of the phenomenal character of perceptual experience allows us to
answer that question. Being pushed to accept that objectively speaking things actually are a certain way
simply by one’s experience constitutes justification to believe that they are that way. Given the concept of
justification at play, which takes its meaning from the activity of evaluating and criticising belief in an epis-
temic community, nothing else, and nothing more, could be required. How much justification one acquires
in this way depends on how strong the phenomenology of pushiness is.

The claim that pushiness and objectivity explain why perceptual experience singlehandedly justify belief can
be strengthened by imagining that we started at the opposite end, asking what a state would have to be like

95
in order for it to provide such justification. If any state could play that role, what characteristics would it
have to have?

A plausible answer is that it would have to have both phenomenology of objectivity and phenomenology of
pushiness. It would have to have phenomenology of objectivity, for otherwise it couldn’t justify a person’s
belief that a world independent of herself has the represented characteristics; and that is what the subject
goes on to believe (§4.4.2). And it would have to have phenomenology of pushiness, because otherwise it
couldn’t justify belief that the objective world actually does have those characteristics, rather than a belief
that it is might have those characteristics, say. When we ask ourselves what perception would have to be like
in order for it to provide singlehanded justification, the answer seems to be that it would have to have ex-
actly the features I have argued that perceptual experience does have.

I have claimed that a description in terms of pushiness and objectivity comes closer to characterising the
phenomenal character of perceptual experience proper than does the phenomenology of seeming able to
just tell, and presentational phenomenology in its various incarnations, which, I also claimed, seem to be
results or consequences of perceptual experience having a certain phenomenal character, rather than a de-
scription of this character itself. This claim is supported by the fact that a subject’s experience having this
character can in turn explain why it seems to her that she can just tell (or that the other characters obtain)
in the cases when it does seem that way.

In virtue of having phenomenology of objectivity, perceptual experience represents a blue water bottle in a
world which is independent of me. My perceptual experience also contains a perspective, a point of view,
and it represents the bottle as being a certain distance away from the locus of that perspective, and at a cer-
tain orientation from it. Neither of these facts, however, can explain why it seems to me that I can just tell
that there is a bottle there. In visual (iconic) imaginative experience, those very same features are present,
but such an experience doesn’t make it seem to me that I can just tell that a blue water bottle is there.

What is needed is that my perceptual experience has phenomenology of objectivity. In virtue of that, my
perceptual experience itself ‘tells me’ that what is being represented to be a certain way is independent of
me (§4.4.3). Because it has this character, the subject matter of my experience seems to be the subject-inde-
pendent world. Without the phenomenology of pushiness, however, I would not seem to have any reason
to believe that things actually are that way, rather than supposing for the sake of argument that they are, be-
lieve that they might be that way, etc. It wouldn’t seem to me that I could just tell that there’s a blue water
bottle in front of me. But when my experience is characterised by objectivity and pushiness, and when the
pushiness is strong enough, it does seem to me that I can just tell.

It will not always seem to a subject this way. Seeming able to just tell, seeming to fact-perceive, and having
the impression, are all, insofar as they are real psychological phenomena at all, plausibly the matter of reach-
ing a threshold, which may not always be reached. In Blizzard and Increasing Light, it is plausibly reached
eventually: when the person is sufficiently close, or the light is strong enough, respectively. But in Judging
Distance such a threshold may never be reached, and likewise in the case of two people along a road.

We can explain this by reference to the phenomenology of pushiness. When that aspect is strong enough it
seems to the subject that she can just tell. But that aspect is not always strong enough, and when isn’t, it
doesn’t seem to the subject that she can just tell (etc.). Nevertheless, so long as objectivity is also present, the
subject acquires some singlehanded justification from her perceptual experience.

96
Let’s sum up. I have argued that the view of perceptual experience I have presented—the view according to
which perceptual experience has phenomenology of objectivity and phenomenology of pushiness—comes
closer than its competitors to describing the actual phenomenal character of perceptual experience, rather
than noting a result of that character being the way that it is. I have also argued that it allows us to explain
why the relevant thresholds are reached, when they are. Most importantly, it greatly strengthens the case for
Phenomenalism because it better explains why a state with the described character can singlehandedly jus-
tify belief.

5.5 In Defence of the Analogy

A mental state’s epistemic features supervene on its non-epistemic features: there can be no change in the
former without a change in the latter. The supervenience base need not contain all the state’s non-epistemic
features, however: perhaps only some of them matter. Something like this may be the case for perception: I
think that the nature of perception is plausibly exhausted by being a conscious experience with representa-
tional content and a certain phenomenal character, but even if not, only these features matter epistemically.

Intuition is also a conscious experience with representational content and that same phenomenal character.
Given that perception singlehandedly justifies belief, this gives us very good reason to think that intuition
does, too. Let’s call the claim that because intuition shares these salient non-epistemic features with percep-
tion it also shares the epistemic feature of singlehandedly justifying belief in its content (absent defeat), ‘the
analogy’.

To resist the analogy, a critic would have to point to a difference in non-epistemic features between the two
states, and then argue that this difference gets in the way of the analogy. Here I will consider three attempts
to do just that, and argue that each of them fails.

5.5.1 The Disanalogy of Content-Specific Phenomenology

That perception has but intuition lacks content-specific phenomenology is a salient difference between the
two states. Moreover, it is in some ways an important difference: it explains why some people fail to recog-
nise intuitional experience is in their own mental lives, for example (§4.3). But this does not yet settle
whether the difference gets in the way of the analogy.

The difference will be regarded as significant by anyone who thinks that such phenomenology is necessary
for us to know the content of an experience (Pitt 2004), and by anyone who thinks that such phenomenol-
ogy is necessary for states like intuitions to even have determinate representational content at all (Horgan
and Tienson 2002; Horgan, Tienson, and Graham 2004; Horgan and Graham 2010; Strawson 2010;
Siewert 1998; Pitt 2011).

This, however, isn’t so much a challenge to the analogy as it is a challenge to the very coherence of the view
that I have been articulating. After all, we do know the content of our intuitional experiences, and yet, I
have argued, they lack content-specific phenomenology. At issue now are challenges specifically to intui-
tion’s ability to provide singlehanded justification for belief provided that we accept my account of its nature.
We may therefore now assume that something other than content-specific phenomenology fixes the content

97
of intuitional experiences, that we can know what those contents are, and that we also know what the con-
tents of perceptual experiences are.157

One possibility is that the content-specific phenomenal character of perceptual experience, together with
facts about how different mental states interact functionally, together fixes the content of all mental states
(Pautz 2013: §4; building on Lewis 1974/1983). On such a view, the content-specific phenomenology of
perception would play an important theoretical role. This (quite attractive) view also does not threaten the
analogy: a challenge arises only if we think that content-specific phenomenology is part of the explanation
of how perceptual experience justifies belief.

I have argued that perceptual experience having phenomenology of objectivity and pushiness on its own suf-
fices to explain why it provides singlehanded justification. If successful, that argument already shows that
perception’s content-specific phenomenology plays no such role, so there is no need to provide additional
arguments to that effect when the analogy is what’s at issue. Someone wishing to argue that content-specific
phenomenology is necessary to explain the epistemic role played by perceptual experience must find fault
with the argument at an earlier point. Absent that, we lack reason to think that perceptual experience hav-
ing content-specific phenomenology stands in the way of accepting the analogy.

Declan Smithies has recently argued that it is only reasonable to think that perceptual experience justifies
belief in its content because of its phenomenal character if one assumes representationalism, the thesis that a
conscious experience’s representational content and its phenomenal character is one and the same thing. If
representationalism is false, Smithies says, “there would be no nonarbitrary answer to the question why per-
ceptual experience justifies believing some contents about the external world, rather than others” (2019:
91).

There is, however, an entirely non-arbitrary and indeed obvious answer to why perceptual and intuitional
experiences justify belief in some contents rather than others. The answer, of course, is that they justify be-
lief in the actual contents of those experiences, rather than in any other contents one might dream up.
There is nothing arbitrary about this. 158

157
It is not necessary that we know exactly what the contents of perceptual and intuitional experiences are: some disagreement about
what is in the content proper of such experiences, and what is merely made obvious thereby, does not threaten the present view.
158
In personal communication, Smithies agrees that this suffices to answer the arbitrariness challenge, but says that his main argu-
ment for thinking that perception must have content-specific phenomenology to justify belief is a different one: “(i) if your percep-
tual experience justifies believing that p, then it must have features that can be known by reflection alone to justify believing that p,
but (ii) your perceptual experience has features that can be known by reflection alone to justify believing that p only if it has con-
tent-specific phenomenology, so (iii) if your perceptual experience justifies believing that p, then it must have content-specific phe-
nomenology” (p.c.). I take it that the thought is that what I am justified in believing is something I should be able to discover
through introspection and reflection alone; that I can’t do that unless I can know the contents of my experiences, judgements,
thoughts, beliefs, etc.; and that I can only know that if those states all have content-specific phenomenology. This argument raises a
number of intriguing and complex questions which I cannot fully address here. But first, I’m not convinced that I in fact always am
able to know by introspection and reasoning alone what I am justified in believing. Smithies argues that if not, I can become justi-
fied in believing ‘abominable conjunctions’, such as “p and it's an open question whether I have justification to believe p”, and that
no one can be justified in believing such conjunctions. However, his argument depends on accepting the account of propositional
justification which to which Turri (2010) has convincingly objected. (See §5.1, above.) If, by contrast, having propositional justifica-
tion amounts to having a way to becoming doxastically justified, it is anything but clear that I could have justification to believe an
abominable conjunction even if I can’t discover by introspection and reasoning alone what I’m justified in believing, because there
would be no way for me to believe this with doxastic justification: believing with doxastic justification that there’s an open question
whether I have justification to believe that p is incompatible with believing with doxastic justification that p. Second, even if this reply

98
It is true that the question remains of how mental states come to have contents. It is also true that no objec-
tion-proof theory of this phenomenon has as yet been put forth. However, for a phenomenon as complex
and puzzling as intentionality this is surely not surprising, and certainly does not rationalise doubt that
mental states do have determinate (enough) content, and that we (generally) know perfectly well what those
contents are. Those are Moorean facts—facts of which we are rationally more certain than we are of any ar-
gument to the contrary—and we should hold on to them with both hands. (Pun intended. I’m not sorry.)

If we do, we have the following: intuition and perception both have determinate (enough) contents, and we
(normally) know what they are, even if we can’t give a fully satisfying theory of either phenomenon. We also
have an argument that experiences having phenomenology of pushiness and objectivity suffices to explain
their ability to justify belief in their contents. Against this background the fact that perception has but intui-
tion lacks content-specific phenomenology is no threat to the analogy.

5.5.2 The Disanalogy of a Known Causal Mechanism

In the eyes of many, the most important disanalogy between perception of intuition is that the former but
not the latter results from a causal process we understand well. This means, they think, that it is rational to
trust what perception delivers, but anything but to do so for intuition.

Let me note in passing that this could not be the basis of an objection to the analogy for Dogmatism. Ac-
cording to Dogmatism I need not be justified in holding any belief in order to be justified by my perceptual
experience that p in the belief that p. A fortiori I need not be justified in holding any beliefs about the system
that produces that experience.

On Liberalism, on the other hand, it may be a necessary condition for experience to justify belief that I have
justification for other beliefs. That the experience is produced by a well-understood mechanism might be
one such. One could then deny Liberalism for intuition while accepting it for perception on the grounds
that the mechanism is well understood for perception but not for intuition. I think that many people are
convinced by this reasoning. But should they be?

It is certainly true that we know a lot about how the perceptual system works. We know how light reflects
off objects, how the eye is composed, and we know how information is transmitted through the eye and the
optical nerve to the brain. One might even think that we have the beginnings of an understanding of how
the brain processes information.

But there was a time when none of this was known, and, of course, many people currently living know no
part of this story. Fixing on these people, there is a strong intuition that they nevertheless acquire(d) single-
handed justification from perceptual experience in exactly the same way that we do. If that is right, under-
standing how the perceptual system works cannot be a necessary condition for perceptual experience to sin-
glehandedly justify belief.

fails, Smithies later argues for what he calls the ‘simple theory’ of introspection, according to which “when you have an introspec-
tive reason to believe that you're in a certain kind of mental state, you have that reason just by virtue of being in that mental state”
(154). And, even though Smithies’ own reasons for accepting this view are tied to his view on content-specific phenomenology, I
think I am equally entitled to that theory, since its intuitive plausibility does not rest on any such claims. If so, then I can block the
route to the abominable conjunctions in the very same way that Smithies himself does, though without attributing content-specific
phenomenology to states which, in my view, don’t have it.

99
True, people of 900 B.C. did have access to the coherence of their experiences, even if they knew nothing
about the perceptual system. So one could claim that they had justification for their perceptual beliefs in
virtue of some explicit or implicit awareness of the harmonious integration between their experiences, and
on no other grounds. But that’s very implausible. They didn’t need coherence to derive justification from
their perceptual experiences: simply having the perceptual experience was enough.159

Let’s grant that the connection between the way things are and intuitional experience is currently ill-under-
stood. 160 This places us now in a parallel situation with respect to these intuitional experiences as that
which those who lived long ago were in with respect to perceptual experience. If their lack of understanding
of perception didn’t then stop them from acquiring singlehanded justification from perceptual experience—
and it didn’t—then our lack of understanding of intuition doesn’t stop us from acquiring such justification
from intuitional experience now.

It is fully compatible with this to think that those who lived long ago got some additional justification from
the coherence of their experiences, just as it is compatible with this that we get additional justification from
the coherence of our intuitions. So long as they got some singlehanded justification merely from having a
perceptual experience then so do we, from our intuitional experiences. Similarly, Liberalism is consistent
with the claim that we now get some additional justification from our understanding of the perceptual sys-
tem, compared to what those who lived long ago enjoyed. None of this threatens the claim that simply hav-
ing an intuitional experience provides singlehanded justification for belief.

Finally, it is not clear that a phenomenalist must accept that there even is a relevant disparity in understand-
ing here. Phenomenalism says that having perceptual experience is what matters epistemically. While the per-
ceptual process is in many ways well understood, why and how conscious experience arises, 161 or why our ex-
perience has the phenomenal character that it does have, 162 is not, to put it mildly. A forceful response is
therefore that there is, after all, no significant disparity in our understanding of the epistemically relevant
aspects of the two mental states: in neither case is the aetiology of the epistemically powerful mental state,
namely the conscious experience, well understood. And yet, justification accrues.

I have argued that perceptual experience having phenomenology of objectivity and pushiness suffices to ex-
plain how it yields immediate justification. Perhaps we get additional justification from knowing what we
do about perceptual processes. But since we don’t need that knowledge to derive justification from percep-
tual experience, we don’t need it to acquire justification from intuitional experience either. 163

159
See also n. 138. This isn’t a mere reaffirmation of Pryor’s intuition. One might’ve thought that in considering our own case we
confuse the intuition that we are justified when enjoying perceptual experiences (surely true), with the distinct intuition that our
perceptual experiences singlehandedly justify us (perhaps false). Reflection on the historical case removes this possibility.
160
For such claims see e.g. Boghossian (2000: 231; 2001: 635); Chihara (1982: 215); Devitt (2005: §§3–4, though especially p. 144);
Goldman and Pust (1998: 184-6); Hintikka (1999) and Mackie (1977/1990: 38-9). This criticism needn’t be general; one could say
that the connection is only mysterious in some cases. For an attempt to account for the connection in terms of constitution, see
Bengson (2015a).
161
The literature on this subject is extensive, but the essential point is just that facts about the way things are with respect to phe-
nomenal experience do not logically supervene on facts about the way things are with respect to physical or functional facts. The
recent locus classicus is Chalmers (1996).
162
A central research project in this vicinity is the search for a so-called Neural Correlate of Consciousness. For the many different
things this might mean, and for some significant challenges facing this research project, see e.g. (Chalmers 1998, 2000).
163
For related discussion, see e.g. Cummins (1998); Goldman (1987); Grundmann (2007); Harman (1977); Pust (2001,
2012/2019).

100
5.5.3 The Disanalogy of Valence
I have argued that intuition is a mental state with representational content, which lacks content-specific
phenomenology, and which has attitude-specific phenomenology of objectivity, pushiness, and valence. Va-
lence, recall, is the phenomenon that it can, in intuition, seem false that p, just as it can seem true that p;
and that whether p seems false or true in intuition is reflected in the attitude-specific phenomenal character
of that experience. There is no corresponding phenomenon in perception.

According to Liberalism about Intuition, an intuitional experience that p with positive valence can provide
singlehanded justification to believe that p, and an intuitional experience with negative valence can provide
singlehanded justification to believe that not-p. There is, then, a direct correspondence between the valence
of the experience and the justification the subject acquires. Given this it is hard to see why the fact that
some intuitions have negative valence should get in the way of the analogy.

5.6 Taking Stock

I have argued that perception provides singlehanded justification because it is a conscious experience with
representational content and a certain phenomenal character; that intuition also is a conscious experience
with representational content and that same phenomenal character, and that none of the non-epistemic dif-
ferences between the two states make an epistemic difference. This gives us strong reason to believe Phe-
nomenalism both for perception and for intuition: both perceptual and intuitional experiences justify belief
in virtue of their phenomenal character.

This view has a number of benefits, both relative to close rival views, and simpliciter. A number of these
have been touched on already. The account I have given improves the description of the epistemically rele-
vant phenomenal character. It gets closer to describing the actual phenomenal character (rather than some-
thing that results from its having that character); it argues (rather than just claims) that perceptual and intui-
tional experience actually have that character; and it improves the explanation of why an experience with
that character can justify belief. Humility is certainly called for here: describing the phenomenal character
of experience is difficult, and so is arguing that having that character enables an experience to singlehand-
edly justify belief. Further improvements no doubt remain to be made. Still, the present account is a step in
the right direction.

The account also has significant structural advantages.

First, one of the phenomenal characters that on this view is epistemically relevant comes in degrees. This
allows the present account, but no competitor view, to explain and account for the facts that justification
from perceptual and intuitional experiences come in degrees, and that this justification varies in concert
with corresponding variation in those states’ phenomenal character. In so doing, this account also signifi-
cantly strengthens the case for Phenomenalism, because it makes the connection between the nature of in-
tuitional and perceptual experiences, and those states’ epistemology, stronger and more intelligible.

Second, when we have intuitional experiences, we often but not always acquire justification to believe the
content of those experiences. This happens when that content seems true. But in intuition, the content can
also seem false, and we then get justification to believe that content’s negation. Competitor accounts have no

101
way to account for this fact. 164 This account does, and again in a way that closely marries the nature and
epistemology of intuition: this happens when and in virtue of the intuitional experience having negative
valence, which is another aspect of its attitude-specific phenomenal character.

Another strength of the account developed here is that the phenomenal character this account deems to be
epistemically significant is non-demanding: both children and animals can clearly have the relevant experi-
ences. The account therefore avoids the charge of over-intellectualisation which can fairly be raised against
competitor accounts, and gives a plausible and coherent story about the conditions under which human
infants and animals receive justification for perceptual beliefs from their experiences.

Some arguments for the claim that intuition provides justification are transcendental: how else could we
explain why we are justified in believing moral, or logical, propositions? 165 A weakness of such arguments is
that they rely on the impossibility of using inference to the best explanation to conclude that we are so justi-
fied. But there are many facts to explain, and it is hard to rule out from the start that the best story entails,
via IBE, that we have such justification. A positive story, such as the one provided here, is preferable.

Another significant advantage of the present account, to my knowledge unique to it, is that it delivers on a
demand forcefully advocated by Timothy Williamson (2007). We should not pursue, he argues, an account
of philosophical methodology which commits us to ‘philosophical exceptionalism’, the thesis that philoso-
phy—perhaps along with (certain parts of) other disciplines, such as linguistics and mathematics—is in pos-
session of and can legitimately use an epistemic method that’s not otherwise widely available. “In general,”
says Williamson, “the methodology of much past and present philosophy consists in just the unusually sys-
tematic and unrelenting application of ways of thinking required over a vast range of non-philosophical in-
quiry” (2007: 3).

I entirely agree that philosophical exceptionalism is irresponsible, and that the methods we use in philoso-
phy are widely available. However, I propose to deliver that result in exactly the opposite way to how Wil-
liamson delivers it. Williamson argues that we philosophers do not rely on intuition, and that when it looks
like we do, we are instead just using our ‘ordinary capacity for judgement’. By contrast, I hold that we do use
intuition in philosophy—the denial of this claim is to my mind not even worth taking seriously. But we do
so not only in philosophy—and not only in a wider set of academic disciplines, either. We use intuition all the
time, also in everyday life.

Bealer was right when he claimed that intuition is a part of our ‘standard justificatory practice’, but he was
wrong to restrict (the relevant kind) of intuition by content. Intuition is completely permissive with respect to con-
tent. Anything you can believe, you can, at least in principle, intuit.

We have intuitions with all sorts of content: I shouldn’t walk that way. He is lying. That branch won’t hold
my weight. That sentence is ungrammatical. People generally prefer to avoid pain. Most people try to do
what they think is right most of the time. Justice is hard to come by but worth having. For this construction

164
Indeed, not only do presentational accounts such as those of Chudnoff and Bengson (and Smithies) lack the resources to
account for the facts that in intuition, the content can seem false as well as true, and that the intuiter then acquires justification to
believe the negation of the representational content. No, these accounts seem to be in principle barred from adding to their ac-
counts so as to accommodate these facts. This is particularly clear for Chudnoff’s view, given his reliance on awareness of a truth-
maker, but I think it holds in the other cases as well. If this is correct, it serves to sharpen this challenge.
165
E.g., Huemer (2005: 111-15).

102
not to fall, we’ll need add a bunch of diagonals. Torturing the innocent for profit is wrong. He’ll fall asleep
if I just keep singing and patting. I should marry this person. And on, and on, and on.

Intuition is deeply engrained in all aspects of our epistemic lives: in philosophy and other disciplines with
significant a priori components, in science, and everywhere else in our everyday lives. The present account
of the nature and epistemology of this state uniquely delivers an account which explains why we can truly
say that intuition is everywhere, and appropriately so, since having an intuitional experience can make the in-
tuiter justified in believing its content, regardless of what that content is.

103
Chapter 6 The FIFO Objection

Phenomenalism says that intuition singlehandedly justifies belief in its content in virtue of its phenomenal
character. In the previous chapter I advocated for a version of this view, and drew out what I take to be seri-
ous structural problems for other accounts.

In this chapter, the fine print is no longer at issue. Here I want to discuss a challenge to any phenomenalist
epistemology, be it about intuition, perception, or any other type of conscious experience; a challenge to the
very idea that a conscious experience’s character can have the epistemic impact these theories say that it has.

6.1 The Feeling-In-Feeling-Out Objection

I’ll call the objection I have in mind the feeling-in-feeling-out or FIFO objection to Phenomenalism. 166 The
objection says that if what ‘goes in’ is what it feels like to have a certain experience, what you ‘get out’ is at
most a fact about whether that person feels justified in believing something. You don’t get facts about the
person in fact being justified in believing that thing.

Here’s an analogy. A car’s shape, spoilers, and paint job can make a difference to whether it looks like a fast
car, but they can’t make a difference to whether it is a fast car. What makes one car fast and another car
slow are substantial differences under the hood, not superficial differences in paint and shape. The FIFO-
objection holds that the phenomenal character of an experience is, at most, a good paint job and a nice
chassis: no matter on which character we hone in, having a conscious experience with that character can’t
make us justified in believing anything.

In my experience, some version of this objection almost always comes up when Phenomenalism is under
discussion, but it has received little attention in print. The one exception of which I am aware is due to Har-
men Ghijsen. For the phenomenalist’s project to be plausible, he argues, its advocates must identify a char-
acter had only by the experience types they say justify belief, and they must explain what’s so special about
that character (2014: 1554). “[W]hy”, he asks, “should [having that experience] provide you with justifica-
tion for believing” the represented proposition? “How could the phenomenology somehow add justificatory strength
to the represented propositions?”(1560, emphasis added). I regard this as the driving force behind the FIFO-
objection: a deep-seated incredulity that something as ‘light-weight’ or ‘surface-level’ as the phenomenal
character of experience could make a difference to such a ‘weighty’ matter as justification or epistemic sta-
tus.

6.2 What the Objection is Not

It is worth being as clear as possible on exactly what the objection is, and to that end it is useful to get clear
on what the objection is not.

166
Thanks to David Chalmers for suggesting this name.

104
First, the FIFO objection says that phenomenology can make a difference to whether people feel justified,
but not to whether or not they actually are. This might make one think that the phenomenalist denies that
the two come apart, and that the objection at issue amounts to a reassertion of this distinction. Not so, on
either count. There are cases where the appearance-reality distinction collapses—pain, for example—but Phe-
nomenalism doesn’t say that this is so for justification. The phenomenalist claims that having an experience
with a certain carefully described phenomenal character makes a person justified; not due to appearance-
reality collapse, but because experiences of that sort are epistemically powerful. The FIFO objection denies
that phenomenal character is capable of doing that sort or thing.

Second, at least as I shall understand it, the FIFO objection doe not essentially concern evidence. As Pryor
(2000: 519) notes, it is misleading to talk about experience as evidence, since this suggests that the subject
uses a belief about her experience as a premise in an inference, or that she adduces it in response to epis-
temic challenge. According to the phenomenalist no such thing needs to be happening. But even bracketing
this, it is preferable to avoid stating Phenomenalism in terms of evidence, since this invites needless confu-
sion (§1.2.7). The FIFO objection is an objection to Phenomenalism, and Phenomenalism does not rely on
the notion of evidence, so the objection should also not be understood in those terms. 167

Third, the objection worth taking seriously is also not that the phenomenal character at issue could poten-
tially be attributable to a different mental state or event: perhaps a higher-order belief (Ghijsen 2014), or a
standing belief becoming conscious (Hanna 2011).168 The phenomenalist presents careful description, anal-
ysis, and argument to the effect that the experience type in question really has the character he says that it
has. Given this, merely raising the possibility of error does not suffice to rationalise rejecting her claim.

This is a general point. The mere possibility of error does not rationalise doubt: instead one must probabil-
ise that there has actually been an error in a given instance. Just as the mere possibility of evil neuroscientists
causing your current perceptual experience is not to be taken seriously absent reason to believe that it is ac-
tualised, the mere possibility that the relevant character stems from a mental state other than perceptual ex-
perience should not be taken seriously in this context. The description, analysis, and argument must be con-
fronted head on.169

Even bracketing this, it is anyway wholly implausible that higher-order beliefs account for the phenomenal
character in question, not least because perception (and intuition) routinely takes place absent such belief,

167
See Huemer (2001: 102) for lucid discussion of this point in the context of his Phenomenal Conservativism.
168
Hanna’s theory differs slightly from Ghijsen’s, but the crucial claim is the same, namely that the epistemically powerful phenom-
enal character could come from something other than intuition or perception, so the reply in the text applies with full force. Here
is an additional point worth noting. Hanna says: “When I intuit or introspect, it seems I can be conscious of the relevant content
without being conscious of the fact that I am intuiting or introspecting it” (2011: 218). That is of course true, but it poses no threat
at all to Phenomenalism (or to Huemer). The phenomenalist claim is not, recall, that my awareness of intuiting that p makes me justi-
fied in believing that p. Instead the claim is that the fact that I am intuiting that p—that I am having that experience—makes me justi-
fied. There is a fact of the matter: either I have the intuition, or I do not. If I do I get justification, irrespective of my awareness.
169
In §3 of his paper, Ghijsen runs together a few different arguments. One strand, also pursued elsewhere (Ghijsen 2016: §3.4.1,
and p.c.), is that the possibility of the phenomenal character arising from higher-order belief is supposed to be problematic because
(‘perceptual’) justification would be too easy to come by: one could get such justification “just by adding” further (possibly unjusti-
fied) higher-order beliefs (p.c.). Ghijsen says that this objection “is similar to” the cognitive penetration objection. That’s an under-
statement: in the context of an evaluation of Phenomenalism the two are equivalent. The cognitive penetration objection says that a
belief causes the perceptual experience, and Ghijsen holds that it causes the phenomenal character of that experience. Since the phe-
nomenalist holds that the experience is epistemically powerful in virtue of its phenomenal character, the outcome is the same. I
reply to this objection in §7.2, below, and to the related ‘bootstrapping’ objection in my (2011: §A.4).

105
and indeed even absent the ability to have such belief, for example because one lacks the required concep-
tual apparatus. As I argued in detail in Chapter 2, having an intuition that p cannot entail having a belief
that q, for any q, since this would yield rational criticisability in cases where we know that there is none. So
the phenomenal character of intuition (or perception) may not be attributed to a belief-state of any kind,
since intuition (and perception) can always occur without it.

Fourth, the FIFO-objection is not that it is implausible that we know that we are perceiving on the basis of the
character of experience, as Ghijsen has also argued. Phenomenalism is a view about justification of belief
corresponding to the content of perceptual and intuitional experience. It is not a view about how we come
to be justified—if we do—in believing that we are perceiving, or that we are intuiting. Phenomenalism is compati-
ble with a range of views here. 170

Finally, the FIFO objection is also not that the wrong phenomenal character has been picked out. I myself
raised such objections above, both to Pryor’s Dogmatism, and to other phenomenalist views. These are sep-
arate from the FIFO objection. The objection at issue says that no phenomenal character can play the epis-
temic role which the phenomenalist says that it can: it’s just the wrong kind of thing to do that job.

6.3 Significance
Why address this objection? First, because it very often comes up in discussion. The incredulity to which it
gives expression clearly comes naturally to many, regardless of the objection’s ultimate merits.

But there are further reasons too, which I discuss here under the banners of core, scope, and depth.

6.3.1 Core
A crucial skill that any student of philosophy must develop is to distinguish central aspects of a position
from peripheral ones. A central commitment of non-cognitivism about ethics, for example, is that moral
statements are ‘not in the business’ of being true or false: they are not truth apt. One could develop that
view by saying that moral statements express approval or disapproval instead of purportedly stating facts; so-
called emotivism (Barnes 1933; Stevenson 1944; Ayer 1952). This claim is more peripheral than the claim
that moral statements are not truth-apt. By developing the view in a different way one could reject emoti-
vism but remain committed to non-cognitivism.

The FIFO objection is important because it targets the most central feature of Phenomenalism. The objec-
tion says that the character of a conscious experience cannot play an epistemic role at all. It’s not that the
wrong character has been chosen; one that can’t play an epistemic role, although a different one can. It’s
not that the character is claimed to do epistemic work in a wide range of conditions when it can only do so
in a narrow range; or that it’s claimed to do so unaided when it can only do so when helped along by other
features. The FIFO-objection says that the character of conscious experience is the wrong kind of thing to play
an epistemic role. But the central commitment of Phenomenalism is that the character of conscious experi-
ence is precisely the right kind of thing to play that role.

If the objection works, no phenomenalist view can be saved. There is no way to tinker with the position, to
change it slightly while keeping it fundamentally intact; or to develop the same type of position in a differ-
ent direction. The FIFO-objection strikes Phenomenalism at its core. If successful, it is devastating.

170
See n. 158.

106
6.3.2 Scope
A second reason the FIFO-objection is so important is that, if successful, it would capture more views than
just the one at which it is targeted: the objection has wide scope. Any view which holds that the character of
a type of conscious experience has epistemic import would fall prey to it. There are many views of this type.
There are, as we have seen, various phenomenalist views of intuition and perception (Chudnoff 2011b,
2011a, 2013b; Bengson 2010; Koksvik 2011, 2013, 2017). There are also various further liberalist and dog-
matist views of perception (e.g. (Pryor 2000, 2004; Silins 2007; Smithies 2019; Huemer 2001, 2007, 2013;
Tucker 2010; Skene 2013). And there are natural ways to develop views of this kind for other mental states,
too, for instance imagination 171 and memory. All such views stand in danger of falling prey to the objection.

This reason to care about the objection is related to the first. The objection will capture any variety of Phe-
nomenalism about intuition because it strikes at the core of that view. It will capture any other view which
assigns epistemic significance to the phenomenal character of a mental state for the same reason.

6.3.3 Depth
A final reason to consider the FIFO objection is that it seems to raise a genuine explanatory challenge for
the view. It is true that all explanation must stop somewhere. After carefully describing the phenomenal
character in question there is a limit to how much more the phenomenalist can do: at some point it is both
necessary and legitimate to simply rely on the recognition that a state of that kind just is the right kind of
thing to play an epistemic role.

But what matters isn’t just how far the phenomenalist explanation can be pushed, but, as Ghijsen insight-
fully argues, the contrast between that explanation, and those provided by competing views (2014: 1560).
Reliabilism, for all its flaws, has at least this much going for it: reliability is an indisputable epistemic virtue.
Reliability has a necessary connection to truth: tending to lead to true belief is just what it is for a belief-
forming mechanism to be reliable. Truth is an indubitable epistemic good, so tending to lead to truth is
also beyond dispute as such.

The FIFO objection is a challenge worth addressing in part because it demands to be told why having a cer-
tain phenomenal character should be seen as capable of constituting an epistemic good at all, against the
background of other, competing explanations for which there is no such doubt.

6.4 Initial Evaluation

Let’s begin by noting two considerations intended to begin blunting the force of the objection.

First, the FIFO-objection clearly captures many people’s imagination, but I think we should view it with
some scepticism from the start. It is widely held that the two central features of the mind which most loudly
cry out for explanation and understanding are intentionality—the fact that mental states are about things
other than themselves—and conscious experience (Chalmers 2004: 153; Smithies 2019). How can some-
thing be about something else? Why, when we represent things as being certain ways (or are in non-repre-
sentational mental states), does it feel the particular way that it feels? Why does it feel any way at all?

Philosophical work on conscious experience in the analytic tradition has tended to focus less on its phe-
nomenal character than on other aspects, such as its representational content or metaphysical status. In

171
In her (2001), Kind argues that imagining must in part be characterised by its phenomenal character, and in later papers (2018,
Forthcoming) she argues that imagination can give rise to knowledge.

107
these lines of inquiry the specific character of our various conscious experiences holds little importance.
This is clearest for the latter, which targets conscious experience as a monolith, seeking to discover how this
strange phenomenon fits in an otherwise apparently thoroughly physical world. But it is also true for the
former: in discussions about Representationalism, for example, attention to the particular phenomenal
character of conscious experiences is usually limited either to pumping the intuition, for very simple experi-
ences, that their characters are exhausted by their representational contents, or to arguing, in objection to
Representationalism, that there are cases of phenomenal differences without a representational one.

But if conscious experience really is one of the two central features of the mind, and given the incredible
richness and variety of such experiences—which analytic philosophers of course do notice—would it not be
incredible if that variety meant nothing at all? If it really didn’t matter for anything that experiences feel the
particular way that they feel, instead of some other way among the infinitude of other ways that they might
have felt? An astounding idea, once made explicit, and yet, that’s what the lack of attention to the details of
the character of conscious experience seems to suggest.172

I submit that the FIFO-objection can arise and seem plausible only against a somewhat skewed intellectual
tradition: one in which we acknowledge both the centrality of conscious experience to the mind, and its in-
credible richness and variety, and yet have somehow come to regard it as plausible that the particular char-
acter of conscious experiences have no bearing on anything else, and in particular not on the experiencer’s
epistemic situation. A much more plausible starting point is that the character of conscious experience will
turn out to matter to a whole host of other things, including in particular to matters epistemic.

The concept of justification at issue in this book takes its meaning from the everyday practice of reflection
on and evaluation of belief. Striving for justification, striving to comport ourselves with ‘intellectual integ-
rity’, in Wright’s words, is an activity necessarily sensitive to the limitations of our point of view, to the na-
ture of our epistemic universe. Conscious experience is “at the very centre of our epistemic universe”
(Chalmers 1996: 169), so it would be very surprising if its character didn’t matter for justification.

Here is the second consideration. The FIFO-objection says that phenomenology can at most make a differ-
ence to whether a person feels justified, but not to whether or not she actually is. But if we take an objec-
tion to p to be a reason to not believe that p, then FIFO can’t be regarded as a strong objection to Phenome-
nalism. Indeed, to the extent that the FIFO objection isn’t a reason to disbelieve Phenomenalism but
merely a statement of its negation, it doesn’t count as an objection at all. 173

If p is inconsistent with q, the mere statement of p is no objection to q, although a solid argument in favour
of p might be. Characteristically, however, the FIFO-objection is not accompanied by much in the way of
argument, but merely by the rhetorical question: how could phenomenology play a role in justification? In
response to this challenge surely no more can reasonably be expected of phenomenalists than that they pre-
sent their theory that shows that it can. ‘How could this possibly be?’ – asks the objector. ‘Here is how!’ –
says the phenomenalist, and gestures to a detailed theory. And then, one can reasonably claim, she is done.

I am not suggesting that these two considerations fully dispel the force of the FIFO objection. They do,
however, show that the objection is on rather less secure footing than one might have initially thought, and
that we can reasonably require a careful enunciation of what exactly the objection is before giving it weight.

172
There are of course exceptions. Some of the views discussed earlier in this book are among them, as are Dainton (2000), Kriegel
(2009), (Siewert 1998; Siewert 2011), and Smithies (2019).
173
Thanks to John Bengson here.

108
6.5 A Reply
As noted, one reason the FIFO-objection is worth addressing is that it strikes at the very heart of Phenome-
nalism by claiming that the character of experience is the wrong kind of thing to make an epistemic difference.
Given this, a natural way to interpret the objection is the following:

FIFO-1 The character of a person’s experiences is irrelevant to her epistemic state

FIFO-1 entails:

FIFO-2 The character of a person’s experiences makes no difference to the credence she is justified
in having in any proposition

FIFO-1 entails FIFO-2 because the credences a person is justified in having in different propositions is a
(proper 174) part of that person’s epistemic state.

FIFO-2 in turn entails:

FIFO-3 For all propositions p, agents S, characters C, and credences c: the credence c which S is
justified in having in p given that the actual character of her conscious experience is C, is
equal to the credence c' she would be justified in having in p if her experience instead had
character C', for any C'.

This entailment holds because, if the phenomenal character of a person’s conscious experience makes no
difference to the credence the person is justified in having in any proposition, there can’t be a single cre-
dence-proposition pair such that a subject’s justified in having that credence to that proposition, but would
be justified in having a different credence if her experience had a different character. So it must be that, for
all phenomenal characters, the credence S is justified in holding in any proposition p is the same for any C'.

Given that these entailments hold, if we can find a single example which shows that FIFO-3 is false, we
would, by modus tollens, also have shown that FIFO-1 is false. Since FIFO-1 is a reasonable interpretation
of the FIFO-objection, this would at least indicate and perhaps show that the objection fails.

Consider a well-rested, alert, and focused mechanic who specialises in vintage American cars, and who is
intently listening for a car she expects to arrive at any minute. She is (justifiably) certain that it will be one
of only three models—perhaps because there were only three entrants in the relevant car race—all of which
she knows very well, but she doesn’t know which one it will be. On hearing a characteristic sound she be-
comes justified in holding credence .8 that it’s the 68 Mustang.

174
A proper part because other things characterise that state, too: the person’s epistemically relevant dispositions, and perhaps her
epistemic virtues (if there are any; see e.g. (Sosa 2007b)), for example.

109
There are ways to alter the character of the mechanic’s experience which would be unhelpful in replying to
the FIFO-objection. In particular, altering the content-specific character of the auditory experience doesn’t
help, since it’s then plausible—at any rate to someone sympathetic to Representationalism—that one thereby
changes the representational content of the experience. 175 And almost everyone agrees that doing that alters
what she is justified in believing. Therefore, on the most charitable interpretation the FIFO-objection says
that phenomenal character can’t possibly make a difference to the experiencer’s epistemic state over and
above the difference it makes by altering the representational content of that experience (if it does—a propo-
nent of the objection needn’t take a stance on Representationalism).

However, it is plausible that one could alter the credence the mechanic would be justified in holding by al-
tering the phenomenal character of her experience in other ways. Her experience is characterised by sharp,
well-rested alertness: that is what we might call the phenomenal context in which the auditory experience oc-
curs. A different overall character experience can have—a different phenomenal context in which the audi-
tory experience might occur—is the drowsy, sluggish, and confused character characteristic of being suddenly
and prematurely woken up from deep sleep, especially after prolonged sleep-deprivation. I take it as given
that if the overall character of the mechanic’s experience had been characterised by the latter phenomenol-
ogy instead of by the former she would not have become justified to the same degree by having the auditory
experience that the approaching car was the 68 Mustang.

At any given time many local conscious experiences contribute to the phenomenal character of a person’s
overall conscious experience. It is possible, indeed likely, that the very same local experience itself makes
different contributions depending on what else is going on, phenomenally speaking—depending on the phe-
nomenal context (Koksvik 2014). But even if not, phenomenal context clearly matters for justification. The
very same contribution to overall character from auditory perceptual experience cannot make the mechanic’s
belief justified to the same degree in the phenomenal context of dizzy, sluggish confusion as it can in the
phenomenal context of rested, sharp alertness. Ditto for many other phenomenal contexts, such as drunk-
enness, joyful giddiness, sexual arousal, and deep sadness: the mechanic would not be justified to the same
degree if the auditory experience had occurred in any of these phenomenal contexts.

We can make the same point with a different phenomenon. A person is lucidly dreaming if she is aware,
while dreaming, of the fact that she is dreaming. A person engages in motivated reasoning if she performs

175
How Representationalism should be understood is itself a matter of significant controversy, but for our purposes it is useful to
start by saying that on this view phenomenal properties—the properties that determine or constitute what it’s like to have conscious
experiences—are identical to representational properties—the properties of representing certain contents (Chalmers 2004: 155-6). If
so, then changing the character of a person’s conscious experience would entail changing what the experience represents too—it
would be to do that—which on almost any view would entail a difference in justification for some proposition or other. FIFO-3
would be false, but without this fact showing what I want it to show. Stated this way Representationalism is pretty obviously false,
however, since the mind can represent contents without any conscious experience at all, for example in standing belief, and since
representing the same contents visually and auditorily (say) gives rise to (or: are) experiences with different phenomenal characters.
A more plausible version says that representing a certain content in a certain way is identical to certain phenomenal properties (160).
Once it’s acknowledged that a difference in phenomenal character—such as that between a visual and a tactile experience, say—can
fail to be reflected in the content of that experience—which must be the case if there’s even one pair of a visual and a tactile experi-
ence with the same content, or any other pair of mental state types with different attitude-specific phenomenology—then there is no
remaining problem with my use of counterexamples to FIFO-3; because a restatement of Phenomenalism’s core commitment is that
certain specific phenomenal ways of representing a content can make an experience epistemically powerful, and a natural way of un-
derstanding the FIFO-objection is as the negation of that claim.

110
inferences, or in other ways reasons, in unsound ways because she is motivated to reach a certain conclu-
sion (Hahn and Harris 2014). Let’s say that a person engages in lucid motivated reasoning if she engages in
motivated reasoning while being aware that that is what she is doing.

Whether motivated reasoning is compatible with full awareness is perhaps doubtful—at any rate, we might
then be talking about a different phenomenon (deliberate self-deception, or what have you). But awareness
comes in degrees, and I think it’s a safe bet that motivated reasoning routinely takes place while the agent is
at least dimly aware of it. I assume without argument—though with some first-personal justification, I ad-
mit—that such awareness usually changes the character of the subject’s overall conscious experience, for ex-
ample by causing it to take on a tinge of doubt or ‘guilty conscience’, a feeling of slight cognitive disso-
nance, or the like. If so, the agent would again not achieve justification to the same degree—she would not
become justified in holding the same credence—in the conclusion as she would if the fact that she was en-
gaging in motivated reasoning had no effect on her experience at all.

FIFO-3 says that for all propositions p, agents S, characters C, and credences c, the credence c which S is
justified in having in p given that the actual character of her conscious experience is C is equal to the cre-
dence c' she would be justified in having in p if her experience instead had character C', for any C'. How-
ever, we have now considered a number of overall phenomenal characters C and C' such that the person is
justified in holding one credence in p in C and a different credence in p in C'. We may therefore conclude
that FIFO-3 is false. By modus tollens, so is FIFO-2, and FIFO-1: it is false that the phenomenal character of
a person’s experiences is irrelevant to that person’s epistemic state. Since FIFO-1 is a reasonable interpreta-
tion of the FIFO objection, that objection is itself thrown into serious doubt.

6.6 Objections
It may be objected that I have only established that conscious experience can negatively influence an agent’s
epistemic state, and not the converse. In the phenomenal context of dizzy, sluggish confusion one gets less
justification from the very same (local) auditory experience than in the phenomenal context of wakeful,
sharp alertness. One might therefore be tempted to think that a degraded phenomenal context can nega-
tively impact a person’s justification, but that no positive contribution can take place.

But no such asymmetry can be upheld. For one, there is no non-arbitrary baseline from which we can find
deviations in only one direction. The best candidate is ‘normal wakefulness’. Now, it is certainly true that
one gets justification from perceptual or intuitional experience in the phenomenal context of normal wake-
fulness, and that justification is diminished relative to this baseline in other phenomenal contexts, for ex-
ample in those described above. But it is also true that one gets more justification in a phenomenal context
characterised by particular sharpness or clarity than one does in normal wakefulness. And anyway, since
clarity and sharpness of mind clearly come on a spectrum, we would be hard pressed to justify a particular
point to count as the normal. For these reasons, the claim that phenomenal character can influence an
agent’s epistemic state only in the negative direction cannot be sustained.

Someone enchanted by FIFO style reasoning might also ask us to provide a positive case for the conclusion
that the character of phenomenal experience influences justification in a positive way, and fair enough, too.
Fortunately, that is just what the previous chapters have provided. The phenomenalist theory of perception
and intuition advocated in this book constitutes a positive case to that effect.

Such a theory should certainly be as deeply explanatory as possible. There should be careful description of
the phenomenal character that is alleged to be epistemically significant, and argument to the effect that the

111
experience-types in question actually have it (Chapter 4). There should be as careful as possible an argument
that the character really does give rise to an epistemic good (§5.4), and it should be made clear what that
good is (§1.2.5). However, once these things have been done, what more can the objector ask? As David
Lewis famously quipped, one cannot argue with an incredulous stare. As he might have added, there is no
rational obligation to even try. The role of this chapter, by contrast, is to complement that earlier case by
dispelling the idea that phenomenal character is the wrong kind of thing to play an epistemically significant
role at all.

Another objection says that FIFO-1 is not a reasonable interpretation of the objection, and that my reply
therefore misses its target. In reply I want to begin by insisting that it is a reasonable interpretation. Recall
the incredulity expressed by Ghijsen’ question “[h]ow could the phenomenology somehow add justificatory
strength to the represented propositions?”. 176 To me only two interpretations suggest themselves: phenome-
nal experience is just the wrong kind of thing to affect epistemic matters at all (“how could phenomenology
somehow add justificatory strength”)—which is what the FIFO-1 interpretation captures, and which my argu-
ment addresses directly—or else that although it is the right kind of thing to affect matters epistemic in the
negative direction, it could not possibly affect them in a positive one (“how could phenomenology some-
how add justificatory strength”)—a claim which I just addressed. So my reply does not miss its target.

A related objection holds that the core of the FIFO objection is not whether the phenomenal character of
conscious experience is the right kind of thing to affect epistemic matters at all, but instead whether it is the
right kind of thing to suffice for justification. 177 Such an objector accepts that phenomenal character is nec-
essary for justification, but disputes that it could ever be sufficient.

First, and in passing, it is again worth reminding ourselves that such a line cannot possibly be advanced for
phenomenal character in general. It makes no sense, as we have seen, to think the presence or absence of some
phenomenal character or other is what is required for, or what suffices for, justification (§5.2). Instead we
must always think of these matters in terms of specific phenomenal characters.

That said, it is hard to see how to support a stance which, on the one hand, accepts both that the phenome-
nal character of conscious experience is the right kind of thing to affect epistemic matters, and that it is the
right kind of thing to affect such matters in a positive direction, while on the other hand still maintaining
that phenomenal character could not possibly suffice—along with the experience in question having deter-
minate enough representational content, let’s not forget—to provide the experiencing subject with some jus-
tification. At any rate, and at risk of repeating myself, I myself do not see what more such an objector could
reasonably ask of the phenomenalists once they have argued that phenomenal character is the right kind of
thing to affect epistemic matters (§6.5); that it can affect them in a positive direction (just above); and, ex-
plicitly, that the specific character at issue (§§4.4 – 4.5) in fact does suffice to provide justification (§5.4). 178

176
Not only is this the incredulity of the only person I know of that discusses the FIFO objection in print; it is, moreover, a senti-
ment immediately recognisable to me from many conversations about this matter with colleagues.
177
Thanks to an anonymous reviewer here.
178
A related objection asks how the argument in this chapter could possibly establish that having an intuitional experience that p is
what makes the intuiter justified in believing that p—that it is the thing in virtue of which she is so justified—as opposed to its being a
merely sufficient condition thereof. (Thanks to an anonymous reviewer.) But this objection is misdirected, because that argumenta-
tive burden is not one that is intended to be carried by the present chapter. Again, the point of this chapter is to defend against the
charge that the phenomenal character of experience is the wrong kind of thing to play the role that the theory advocated in this book

112
Note that my claim is not that no argument of this general form could ever succeed. It is perfectly cogent to
argue, for example, that while an intention to kill is necessary for murder, it could not ever suffice. But
what murder consists in is much better understood. Murder is a particular type of killing, and a killing is
(something like) the ending of a (certain type of?) life through one’s action (or, perhaps, inaction). (I said
murder is better understood, not that our understanding is all that great!) Against this background the claim
about intent makes perfect sense.

By contrast, the metaphysics of justification—the question of what justification consists is—is a theoretical
territory at a much earlier stage of exploration. There is no background agreement that can be leveraged
against the idea that having an experience with (representational content and) a particular phenomenal
character suffices for justification, comparable to what can be leveraged against a thesis of the metaphysics
of murder that leaves out the killing. As we have seen, nearly all theses in this area are contested, including,
for example, what role if any reliability—a property with a necessary relation to truth, no less—should play in
the theory of that which epistemic agents should value and pursue.

In this book and in other works phenomenalists present detailed arguments for the conclusion that having
an experience with representational content and a certain phenomenal character is enough for justification
to obtain. Many truths are surprising, and no doubt that truth will be surprising to many. But surprise is no
argument, and the possibility that we outline must be taken seriously, and considered on merit.

6.7 A Verbal Dispute, Epistemic Pluralism, and Justification

I argued above that perception justifies belief in virtue of its phenomenal character (§5.2). As part of that
argument I considered (and endorsed) Declan Smithies’ argument for this conclusion, an argument which
trades on the difference in justification between blindsighted and normally sighted subjects.

I want to end this chapter by briefly considering a close ideological relative of the FIFO objection, namely
the claim that the difference between the blindsighter and the normally sighted person, whatever it is, is not
one of justification. To be justified, says the objector, a belief must have the objector’s favourite epistemic
feature; it must be formed by a reliable belief-forming process, let’s say.

My reply—which owes much to Smithies—proceeds in two steps: one careful and one ambitious.

The careful step starts by saying that whatever we chose to call it, the difference between the blindsighter
and the normally sighted person is clearly epistemic. If a blindsighter of any variety forms a non-inferential
belief about something in a ‘blind’ region, she is thereby rationally criticisable. If the normally sighted per-
son forms a belief corresponding to that same region, she is not. Rational criticisability is, as we have seen,

says that it does. The argument for the claim that intuitional experience does so in virtue of having the phenomenal character it does
have is different, and has been outlined several times above—for example in the introductory paragraphs to Chapters 3 and 5.
Yet another possible objection says that it is not strictly speaking enough to establish that there are credence-proposition-
phenomenal context triples such that a subject is justified in having that credence to that proposition in that phenomenal context,
but would instead be justified in having a different credence to that proposition in a different phenomenal context. That is because,
as the same reviewer noted, some such triplets can be accommodated by the FIFO-objector. In particular, the objector can accom-
modate beliefs about the character of experience. It is plausible, for example, that I am more strongly justified in believing that I am in a
phenomenal context of sharpness and clarity when I am in that context than when I am not. So the FIFO objector can accept that
FIFO-1 is false, on the grounds of limited exceptions such as these, but hold that exceptions are tightly circumscribed. However, my
case against the FIFO objection did not depend on such cases: all of the cases were examples of phenomenal character affecting a
person’s justification for believing what the experience represents. Thus the counterexamples to FIFO-1 are in fact not circumscribed in
the way that the objector would need them to be, and this objection does no damage to my argument.

113
like justification in being a concept with which normal epistemic agents are intimately familiar from their
everyday practice of criticism and evaluation of belief. It is an epistemic concept par excellence. Since the con-
cept applies to one and not the other, the difference between the two is epistemic.

The careful step continues by noting that we should adopt epistemic pluralism (Alston 2005): we should
acknowledge that there are multiple epistemically good-making features that a belief might have. A reliable
link to truth is one of them, but it is not the only one: for clearly a brain-in-a-vat which is a phenomenal du-
plicate of me has something epistemically good in common with me even though we differ radically in relia-
bility; and, equally clearly, whatever the BIV and I have in common is not something I share with a colour-
believer (§5.2), even though her belief-forming process is, we may stipulate, exactly as reliable as mine.

The careful step concludes by saying that, given that there are multiple epistemic good-making features that
a belief might have, and given that what the phenomenal character confers is clearly among them, the objec-
tor here risks lapsing into a merely verbal dispute: a dispute not about the facts on the ground, but about what
labels to affix to their various parts (Chalmers 2011). The phenomenalist agrees that reliability is an im-
portant epistemic good-making feature which a belief might have; she just adds to this that so, too, is being
formed in response to conscious experiences with a certain phenomenal character. Since the latter claim is
backed up by argument, and since the (imagined) objector here isn’t purporting to rebut any part of that
argument, the objector is merely quibbling about labels. But we shouldn’t quibble about labels. What mat-
ters is, first, that in virtue of having intuitional and perceptual experience, a subject’s beliefs clearly have a
good-making epistemic status, and, second, what this entails about the roles beliefs with that status can play.
Whether we call this justification just doesn’t matter very much.

At this point we can imagine the objector falling back on the claim that, although it must be admitted that
the good-making features phenomenal experience confers is epistemic, it is not an important epistemic good-
making feature. 179

This sets the stage for the ambitious part of the reply. In Chapter 1 I argued that the ‘gold standard’ for
epistemology should be justification, not knowledge. Justification, I said, is a concept easily introduced to
the novice, and this is plausibly because we are all already familiar with the concept’s application in practice.
We know what justification is because we are familiar with the role justification plays in the epistemic prac-
tice we all participate in, namely that of critically evaluating our own beliefs, and those of others.

I now want to insist again that that role—the role that is played in our epistemic lives by whatever the con-
cept of justification refers to—that role can be played by a person’s having an intuitional or perceptual experi-
ence. Having one of these experience types gives us exactly the kind of epistemic good-making feature to
which we refer in our everyday, ubiquitous practice of evaluating and responding to challenges to our be-
liefs. To many such challenges an entirely adequate response is that I believe that things are a certain way
because in either perceptual or intuitional experience, that is how things seem to me to be. So the epistemic
good-making feature that intuitional and perceptual experience bestow on our beliefs is not lower-grade or
less important: it is the central one, the one which plays the most central epistemic role there ever was.

179
Thanks to Leon Leontyev here.

114
115
Chapter 7 Too Much Justification?

In this chapter I want to address a family of objections which say that my view provides too much justification,
in that it entails that a person has justification in cases where we know that she does not.

As we have seen, plenty of people take intuition to be hopelessly mysterious, so any theory according to
which intuition at least sometimes justifies belief will face objections of this sort. The theory advanced in
this book is more strongly exposed to this line of attack, however, since it says not only that having an intui-
tion can justify belief, but that intuition is wholly permissive with respect to content. Thus there will be
many more cases thought to be problematic for my theory than for competitor views.

There are two basic strategies for defence here. For any given case I can either say (i) that my theory is cor-
rect to say that there is justification, perhaps contrary to initial appearances; or (ii) that the objector is cor-
rect to say that there isn’t, but that my theory doesn’t have to say that there is. In my view, a combination of
these two strategies can successfully manage all apparently problematic cases, so no ‘too much justification’
case ultimately constitutes a damaging objection to the view.

7.1 Defeat
I want to begin by discussing defeat, because doing so will provide a framework for thinking fruitfully about
both actual and future objections in this family. Phenomenalism says that absent defeat having the intuition
that p makes the intuiter justified in believing that p. If in an apparently problematic case there is a defeater,
according to the theory no justification is provided, so there can be no ‘too much justification’ type prob-
lem. Defeaters are thus a resource for the theory under strategy (ii). As we shall see, consideration of the
landscape in this vicinity will also reveal resources for the theory under strategy (i).

It is common to distinguish two different types of defeaters: undermining (or ‘undercutting’) defeaters, and
rebutting ones.180 Undermining defeat happens when the rational connection between something that
would otherwise justify belief, and that belief, is weakened. If you enter a room and some cups on a table
look pink to you, having that visual perceptual experience justifies you in believing that they in fact are
pink. Once you learn that there’s a strong red light overhead, that connection is weakened. If a trustworthy
expert tells you that the rock you’re showing her is quartz you’re justified in believing that it is, but if you
acquire justification to believe that she was paid to lie, the connection is weakened. And so on.

By contrast, a rebutting defeater for your belief that p is simply evidence that not-p. Such evidence changes
the subject’s epistemic status with respect to the belief that p, but it doesn’t bear on the relation between
experience and belief. 181

180
The distinction was introduced by Pollock (1987: 485). Pollock talks in terms of reasons, which he takes to be propositions. The
present theory requires a more permissive notion, so the my presentation differs somewhat from Pollock’s.
181
At least not directly. I set aside the question of whether evidence that not-p also weakens the connection between justifier and
belief, and if so under what circumstances.

116
The language of defeat was introduced in the context of binary concepts of belief and justified belief. In
that context sufficient evidence that not-p would indeed render (outright) belief that p unjustified, so one
could say that the subject’s justification was defeated. Here, however, we are operating with degrees of be-
lief. In that context the language of defeat is misleading for rebutting ‘defeaters’, since evidence that not-p
does not affect the connection between justifier and belief—the justification isn’t actually weakened in any
way—but merely adds weight on the other side of the scale.

Moreover, rebutting ‘defeaters’ are not resources for responding to ‘too much justification’ type objections.
The objector’s concern isn’t alleviated if the subject fails to end up with all-things-considered justification
for outright belief that p because she also has sufficiently strong evidence that not-p. Instead the objector’s
claim is that it’s a mistake to think that intuition provides justification at all in the cases at issue—that it
adds any weight on one side of the scale—whether or not it is ultimately outweighed. 182

For these reasons I will focus exclusively on undermining defeaters.

Let E be S’s intuitional experience, p be the intuited proposition (E’s representational content), and q the
proposition that E is caused by an omniscient and omnipotent being G intent on deceiving S. G is omnisci-
ent, so she knows all truths. G is intent on deceiving S, so she’ll do whatever she can to induce false beliefs.
G is omnipotent, so she’ll succeed. q is thus an undermining defeater par excellence.

The concept of justification at play in this book is the one with which we are all already familiar from evalu-
ation and criticism of belief. This point informs us now, because it shows that even if q is true, if S has no
idea that it is, her justification to believe that p remains undefeated. Given the concept of justification at
issue in this book, for q to defeat S’s justification, q must be somehow connectible to S’s epistemic state. Ex-
actly how is an important question.

To systematise our thinking, let us ask the following:

(i) Does S have doxastic justification to believe that q?

(ii) Does S have propositional justification to believe that q?

If the answer to (i) is no but the answer to (ii) is yes—if S has propositional but lacks doxastic justification to
believe that q—we can go on to ask:

(iii) Is the fact that S lacks doxastic justification something for which S is rationally criticisable?

For simplicity, I will assume that S believes that q if and only if she has doxastic justification to believe it.
Given this, to guide our thoughts about (iii) we can ask whether S ought to believe q. This yields the follow-
ing options for S’s justification to believe q:

182
That said, it’s important to keep the possibility of there being a wealth of such ‘defeaters’ available to a particular epistemic agent
in cases where the proposition for which she gets justification is morally troubling, on which issue more forthwith.

117
I assume it to be agreed by all that if option (1) obtains—if S has doxastic justification to believe that E is
caused by G—then S has an undermining defeater for her justification to believe that p. After all, to have
doxastic justification is to justifiably believe, and if S justifiably believes that her intuitional experiences is
caused by an omnipotent agent intent on deceiving her, the connection between that purported justifier
and her belief is decisively severed. This may be bad news for S but it’s good news for Phenomenalism:
there can be no ‘too much justification’ challenge in cases of this kind.

A fully ideal epistemic agent would presumably believe all and only the truths. On the other hand, the dis-
tinction between propositional and doxastic justification requires it to be possible to have propositional jus-
tification while lacking doxastic justification. This points to a moderately idealised agent: one who believes
with justification everything the current agent has propositional justification to believe without believing all
and only the truths. This recalls another notion of moderate idealisation discussed earlier. On Turri’s sug-
gestion, when a person has propositional justification to believe that p that’s because there is a moderately
idealised way for her to come to believe that p with justification. We can think of the moderately idealised
agent as having availed herself of all such moderately idealised ways.

How much idealisation it would take for an agent to believe with justification all the things the current
agent has propositional justification to believe, and along which dimensions the agent would have to be ide-
alised, are important questions for understanding justification. Does she merely have to carefully carry out
reasoning of a kind which she already masters, even if a lot of it, or does she in addition have to learn and
carry out reasoning which she doesn’t currently master? Must she carefully introspect and catalogue her inner
life? Must she undergo some process to make all her standing beliefs accessible to her (assuming with Freud
and friends that there can be inaccessible standing beliefs)? I don’t know the answer to these questions, but
I will assume that they can be settled one way or the other. For ease of exposition I will also assume that rea-
soning is all that is required: no introspection or psychotherapy is needed, for instance. Nothing hinges on
this; everything can be restated if required.

Even with all these questions left open there is still a strong restriction on the idealisation: if S has proposi-
tional but not doxastic justification for belief in q, there must be a way for her to come to believe that q with
justification. So the reasoning that would bring about that belief must be available to her. Conversely, if S
lacks propositional justification to believe that q, then the reasoning that would take her to belief in q is

118
most likely unavailable to her. It doesn’t follow that it is—there might be other explanations for why she lacks
propositional justification—but it’s the most likely scenario.

All of this means that challenges to Liberalism from cases of type (4)—cases where the subject doesn’t even
have propositional justification to believe that her justification is undermined—are unlikely to succeed. The
concept of justification at play is tied to everyday evaluation and criticism of belief. I think it’s clear that S is
not blocked from acquiring justification in this sense to believe that p from having E if the reasoning that
would take her to justified belief in q is not even available to her. In such cases the right conclusion is that
she does have justification. This, then, is an instance of strategy (i).

Thus an objection from a case of type (4) would have to be one in which reasoning to q is available to the
intuiting subject, in which she somehow nevertheless lacks propositional justification to believe q, and
where we should still be confident that having the intuition does not justify her in believing what it repre-
sents. Because I can’t see how to construct such a case I will assume, in what follows, that this can’t be
done, and consequently that no case of type 4 poses a ‘too much justification’ type threat to the view.

The interesting cases are of type (2) and (3). In cases of type (2), S doesn’t believe q, but she has proposi-
tional justification to believe it, and she is rationally criticisable for lacking doxastic justification to believe q.
Put differently, the way she has to come to believe that q with justification is one of which she ought to have
availed herself.

I think the reasonable stance here is to say that S does have an undermining defeater for justification of p
by E. Although a defeater must be connectible to the subject’s epistemic state, it is surely not the case that all
defeaters must be recognised by her as such (connected to her epistemic state): we can demand of an agent
that she does at least some work to put two and two together. If that is right, then cases of type (2) also pose
no challenge for Liberalism. Liberalism is a claim about justification arising absent defeaters, and in cases of
type (2), a defeater is present. For such cases, strategy (ii) bears fruit.

What about cases of type (3)? In such a case, S doesn’t believe that E is caused by G, but she does have prop-
ositional justification to believe this. She is not, however, rationally criticisable for not ‘converting’ the
propositional justification to doxastic justification. Whatever she would have to do to arrive at a justified
belief that q, she is not rationally criticisable for not doing so: she is not rationally criticisable for not avail-
ing herself of the ‘way’ of coming to believe with justification that is open to her.

In such cases I think the right answer is that S does get justification for her belief that p from having the in-
tuition that p. When we considered case (4) we said, roughly, that if there is nothing the agent could do to
discover q—if the reasoning to q is not even available to her—then there is no block to justification. It is
equally reasonable to say that if there is nothing the agent should do, epistemically speaking, then there is no
block to justification. In cases of type (3), then, the agent really does acquire justification merely from hav-
ing the experience, and, again, there can be no threat to views like the one advocated here. In these cases,
strategy (i) is again the winner.

Some might at this point accuse me of having dealt myself a ‘get out of jail free card’; of having given myself
a way to wiggle out of any objection to the view, thus making the view I advocate unfalsifiable. Not so.

119
What I have done is to set out a framework for thinking about ‘too much justification’ type objections; and,
in outline, explain how I’d like to answer to cases of the various types. But that doesn’t inoculate the theory
from objection. It remains quite possible, for instance, to present a case of type (3) or (4) and to argue that
I’m wrong to say that in such cases the intuiter gets justification; or to provide a case of type (2) and to ar-
gue that, despite what I argued above, there is in fact no defeat in such cases (and also no justification). So
the theory remains falsifiable. It just hasn’t been falsified yet.

7.2 Cognitive Penetration

Having outlined what I think is a useful framework for thinking about intuition and defeat, I now want to
consider a particular type of case, to see how the framework fares. As with the FIFO objection, this type of
case is quite frequently raised in objection to Liberalism and Phenomenalism. It takes on different forms in
different contexts, of course, but here I’ll focus on a representative instance.

Racism. Tom is a liberal white man. Just like everyone else, Tom has a large number of standing
beliefs. Among them, unfortunately, are some deeply racist ones. These beliefs were inculcated in
him when he was very young, and haven’t surfaced to the level of consciousness for decades. They
entail that a young black male out walking late at night is significantly more likely to be dangerous
than a young white male is. Tom has no evidence to support this belief nor did he ever: these stand-
ing beliefs are not justified.

As he is walking home one night, Tom sees two males approaching, one on either side of the street.
The person on his side is black, the person on the other side is white, otherwise he can discern no
relevant differences between them. Tom has an intuition to the effect that the person on his side of
the street is more likely to be dangerous than the other person is, forms a belief to that effect, and
crosses the street.

I think it’s useful to distinguish two different lines of thought that can lead up to the objection I have in
mind here. Some people seem to think that the conclusion that Tom’s intuitional experience justifies his
belief is somehow morally bad. It is not clear to me whether they think that advocating a theory with this
consequence is bad; that we can know that the conclusion is wrong because it is bad to be racist; that we
can know that the conclusion is wrong because a world in which that racist belief is justified is a bad world;
or something else. But be that as it may, I propose to set this line of thought aside.

It most certainly is morally bad to be racist. However, and although I’m not convinced that moral and epis-
temic normativity are completely separated, I don’t think we can conclude quite so quickly that intuition
fails to justify belief in this case. It is very unfortunate indeed that racist beliefs are sometimes justified, even
though they are false. But many unfortunate things are true, and false justified beliefs, even bad false justi-
fied beliefs, are entirely commonplace. Some people think that testimony is a source of justification and
some think it’s not, but it is not a good objection the former kind of view that if it were, people would
sometimes come to justifiably hold morally bad beliefs. Ditto, mutatis mutandis, for mnemonic seemings—
cases of seeming to remember—which again may or may not be sources of justification, but where, again, it’s
no objection to the view that they are that if so people will sometimes be justified in holding morally bad

120
beliefs. 183 Similarly, it can’t be an objection to a view that says that intuition can justify belief that if the
view is true, bad beliefs will sometimes be justified.

The objection I think is worth taking seriously is that Tom is being justified is the wrong conclusion, epistem-
ically speaking. Understood this way, this objection is known as that of cognitive penetration (of experience,
by belief).184 Moreover, seen in this way there is nothing distinctive about the moral case: the case is equally
challenging for the theory regardless of content.

The framework established above is now useful. For S’s belief in the content of her intuitional experience,
the proposition that that experience was caused by an omniscient being intent on deceiving her was a de-
feater par excellence. Moreover, it ‘wore defeat on its sleeve’ in that it would be obvious to S, were she to
come to believe it (and other things equal), that this defeated her justification from intuition. In Tom’s
case, a corresponding proposition might be that the intuition was caused by an unjustified standing belief.
In any case, let q be some corresponding defeater par excellence that also wears defeat on its sleeve for Tom,
and let’s begin with case (1), in which Tom has doxastic justification to believe q. Because he has a defeater
and consequently gets no justification, there is no challenge to Phenomenalism here.

On the other hand, if Tom doesn’t even have propositional justification to believe q I take it that Tom does
get justification to believe what his intuitional experience represents. To some this seems an unacceptable
result, but I think that it is correct. The argument above has shown that having an experience with the phe-
nomenal character that intuitional experience has justifies belief in its content absent defeat. There is no
defeat here, and we should accept the theory’s output.

In case (2), Tom doesn’t believe q but he has propositional justification to believe it, so he has a way of com-
ing to believe it with justification, and he is rationally criticisable for not availing himself of it. As before,
this strikes me as a clear case of defeat, and again, this aligns well with the concept of justification at issue,
grounded as it is in the practice of evaluation and criticism of belief. We can imagine indicating to Tom the
line of reasoning that was open to him but of which he did not avail himself, and showing him, moreover,
that he was criticisable for not so doing. If Tom were convinced of this, and if he managed to react ration-
ally instead of emotionally, he would be convinced that he wasn’t justified in believing as he did even before
being led through this reasoning. That is what rational criticisability does.

In cases of type (3), by contrast, there is reasoning open to Tom that would take him to justified belief in q,
but he is not rationally criticisable for not availing himself of it. If an interlocutor in a superior epistemic
position were to raise this with him, she would, if Tom reacted rationally, be able to lead him to justified
belief in q, in which case we would be back in a case of type (1). But what she wouldn’t be able to do is to
get Tom to agree that he wasn’t justified before he was shown this. That is what the absence of rational criti-
cisability does. The conclusion that he does get justification from having the intuition is correct.

183
As Sarah Moss (2018) emphasises for her own case, so too must I emphasise in mine, that although the discussion is limited to
epistemic matters surrounding the beliefs in question, there are many other problems with racist beliefs. Nothing in this text should
be taken to imply that it is in the slightest morally acceptable to hold or perpetuate racist beliefs. It is not.
184
The objection was to my knowledge first raised by Susanna Siegel in her (2012), where she raised it against Pryor’s Dogmatism.

121
To be strictly accurate, the above discussion would have had to be refined in various ways. It is not clear
that only justified belief or credence counts as a defeater, for instance, and even if so, there is if course not
just one proposition that counts. For Tom to get justification from his intuitional experience, what we have
already said would have to apply to all of them. But there will be a range of cases in which subjects get justi-
fication from intuition to believe a proposition p even though there is a true proposition q such that if the
subject came to believe it with justification, her justification to believe p would be undermined. 185 Some of
those will be cases where the intuiter gets justification to believe propositions we really would rather she
didn’t. Unfortunate though that is, it is nevertheless true.

7.3 Logical Fallacies

Consider this sequence:

(1) If it’s raining I’ll get wet when I step outside.

(2) It’s not raining.
(3) Therefore, I won’t get wet when I step outside.

To reason like this is to commit the fallacy of ‘denying the antecedent’, which is one of many ways to reason
invalidly. A reasoning pattern is invalid if it’s not guaranteed to take you to a true conclusion from true
premises. By contrast, a valid reasoning pattern does guarantee that if the premises are true, the conclusion
will be true also.

The problem with this inference pattern is obvious on reflection. Other things than rain can make me wet
when I step outside—the sprinkler, for example—so I can’t validly infer that I won’t get wet from the fact
that it’s not raining. Yet this can be a pretty convincing form of reasoning nevertheless; it’s an easy mistake
to make. 186

The theory advanced in this book concerns what happens when a person intuits that p. It is not committed
to a particular stance about how justification accrues in reasoning. However, the theory does say that intui-
tion is wholly permissive with respect to content, and that anything you can believe you can intuit. So I am
committed to the possibility of someone intuiting: ‘if it’s raining I’ll get wet when I step outside, but it’s not
raining, so I won’t get wet when I step outside’, and thereby, absent defeat, getting justification to believe
that proposition. 187 (A similar possibility, perhaps a bit more tractable, is that the person might intuit ‘the
facts that I’ll get wet if it’s raining, and that it’s raining, together support that I’ll get wet’, and thereby, ab-
sent defeat, getting justification to believe that. 188)

This is sometimes raised as another ‘too much justification’ challenge to the view I advocate. “We do not
think”, says David Christensen, “that someone who reasons in accord with, say, the fallacy of denying the

185
Perhaps some would wish to raise a ‘too much justification’ type objection for cases where there is no defeater, i.e., no true prop-
osition that shows that the represented content is false. I can’t see that such an objection can have force separate from engagement
with the argument in previous chapters, so I set this possibility aside.
186
I speak, unfortunately, from experience: (Koksvik 2011: 278).
187
Of course, the possibility that a person can have an intuition with this content (or the one that follows) also doesn’t commit me
to a view on the role intuition might play in reasoning, but I bracket this point here.
188
Chudnoff suggests something like this in his (2014).

122
antecedent attains rational belief that way, even if the conclusion he adopts strikes him as following conclu-
sively from his premises” (2010: 205).189

We don’t? More to the point, shouldn’t we? As before, a mere statement of the negation of a view, or some-
thing which entails that, is not an objection to it (cf. §6.4).

The bad result is supposed to obtain when the subject becomes justified in believing that the conclusion
follows from having an intuition with that content (or to that effect). This is supposed to be a bad result be-
cause the inference pattern is invalid. But is there anything wrong with replying in the same way as with the
case of cognitive penetration; with replying, that is, by making the outcome dependent on the intuiter’s ep-
istemic position with respect to defeaters?

Consider the NCA intuition discussed above (p. 20). I take it to be a fact about the history of philosophy
that a number of thinkers, some of whom were giants in the development of logic and set-theory, have had
this intuition (or near enough), and have formed a belief on its basis. Were these people not thereby justi-
fied? NCA is an intuition that’s (usually taken to be) 190 false, because it leads to contradiction. But it is not,
I submit, at all plausible that this fact—the fact that it is false—is any reason at all to think that the intuition
with that content doesn’t provide justification, provided that the intuiters didn’t believe that it did (lead to
contradiction) with justification, and nor did they have propositional justification for this that they were
criticisable for not having ‘converted’ into justified belief. 191

If that is right, then that result applies equally to cases of denying the antecedent, and other simple logical
fallacies. It doesn’t apply if the agent believes with justification that the true logical theory reveals that the
intuited proposition is false since this, I take it, decisively undermines the justification. 192 But if the agent
has no way of believing this with justification, not even a moderately idealised way—and so not even propo-
sitional justification for this (or any other) defeating proposition—then it seems entirely appropriate to say
that she does get justification from having the intuition. For then the logic novice is in a similar position
vis-à-vis the proposition she intuits as were the great pioneers of logic and set theory vis-à-vis NCA, and justi-
fication is sensitive to a person’s epistemic position.

In short, then, I think that the same strategy as above applies in this case: whether the intuiting subject gets
justification from having the intuition depends on her epistemic position vis-à-vis defeaters in the systematic
way that I have outlined. And this strategy generalises, I take it, from the two cases we have discussed to fu-
ture ‘too much justification’ cases, as well.

7.4 Should’ve Known

I’ve argued that when a subject S would normally get justification from her experience E for belief that p,
but where there’s an undermining defeater q, we must consider how S is epistemically related to q in order
to reasonably evaluate the output of the present theory. As part of this discussion I considered cases of type
(4), where, although there’s a defeater ‘out there’—q is true—S doesn’t even have propositional justification

189
Christensen is not addressing Phenomenalism or Liberalism about intuition specifically here, but it seems clear from context
that he would be happy to level the objection to these views. The objection has been pressed on me by Leon Leontyev.
190
Though not by dialethists; see (Priest, Francesco, and Weber 1998/2018: §3.3).
191
Some evidence for the claim that Frege was not in this position is recounted by Russell, who says that Frege responded “with
intellectual pleasure clearly submerging any feelings of personal disappointment” (Irvine and Deutsch 1997/2016: §2).
192
It is possible to see this as providing decisive outweighing justification instead—a ‘rebutting defeater’—but I think it’s more plausi-
ble to think of this in terms of undermining. Similarly for the conjunction fallacy in probability theory.

123
to believe it. For these cases I said that the right result is that S does get justification to believe that p from
having E. This, I agreed, is an unfortunate result in some cases, but it is still true.

One way of blunting the force of the objection (for those who find it persuasive)—or, to put it differently, of
providing more nuance to the picture here—that is still overall in keeping with the view I advocate, is to say
that there may be cases where a person should have known that q is true—or, perhaps, that q has a good
chance of being true—even when she lacks propositional justification to believe q, and that these should also
count as cases of defeat.

In a recent paper, Sandford Goldberg argues that there are cases in which it is true that the agent “should
have known that [q], even though, given her state of evidence at the time, she was in no position to know it”
(2017: 2863). I assume that if, given her evidence at the time, a person is in no position to know that q, then
there is no (moderately idealised) way for her to come to believe that q with justification, and so, the way
these concepts are used here, she lacks even propositional justification to believe it. The reason the person
should have known even though, given her evidence, she wasn’t in a position to, Goldberg argues, is that
other people are entitled to expect her to know. They have legitimate expectations to this effect, grounded either
in specific practices in which they both participate, or, more generally, simply in “the institutions of moral-
ity and epistemic assessment (2017: 2879).

First, a tiny adaptation: insofar as it’s reasonable to say that a person should have known that q, it is just as
reasonable to say that she should’ve had a high justified credence that q, or that she should’ve known, or had a
high justified credence, that it is likely that q. Given this, we can apply the view to the objections at hand.

Consider again the case of Tom and his racist standing beliefs. It is quite plausible that Tom, along with
many others, should have know that he is quite likely to have racist standing beliefs (or should have a high
justified credence to that effect). And from there it is not too big a step to say that he should have known (had
credence, etc.) that the intuitional experience resulted from these beliefs, which would, on this suggestion,
entail defeat. If so, what I said about cases of type (4) must be amended: some such cases are, after all, cases
of defeat.

Presumably that would be a welcome conclusion for those who raise this style of objection. I won’t take a
stance on Goldberg’s view here. My point is simply that insofar as we think there are legitimate cases of the
should have known phenomenon which he describes, they seem to relatively easily plug in to the framework I
have laid out in this chapter.

This point raises deeper issues about the nature of justification. When we engage in the social practice of
evaluation and criticism of belief, I think it is clear that we are deploying a bona fide epistemic concept:
hence my earlier insistence that the epistemic good-making feature which having an intuitional experience
bestows on belief is an important epistemic one. But it is not anywhere near as clear that we are operating
with a concept that is exclusively epistemic.

Issues in this vicinity are often discussed under the banners of pragmatic and moral encroachment: en-
croachment, that is, of other factors into epistemic normative space. Pragmatic encroachment occurs for
example when whether a person’s belief has a particular epistemic good-making feature—e.g. that of being
justified, or amounting to knowledge—is taken to depend on pragmatic matters beyond what evidence she
has for that belief, for example on what’s at stake, practically speaking, if she gets it wrong. Moral encroach-
ment is a view cast in the same mould: this occurs when it depends on moral matters instead. And both
phenomena, although presented here in binary terms, extend straightforwardly to degrees of belief.

124
If we follow the path outlined in this section, then we allow for encroachment. Again, I don’t want to say
that we should. What the point clearly illustrates, however, is that there are many more avenues than is
commonly appreciated for avoiding the consequences of the present theory that, at least for some, seem dis-
qualifyingly troubling.

7.5 There’s (Even) More to The Full Story, Of Course

Two points that sometimes seem to be overlooked when the ‘too much justification’ objection is at issue,
and which suggest that the consequences of accepting the view I’m advocating are (even) less dire than they
might, still, to some, appear, are the abundantly obvious ones that there is more to epistemology than the
theory advocated in this book, and that there is more to the theory of permissible speech and action than
what epistemology can give you.

First, there is more to the theory of justification then just that part of it which concerns how justification is
acquired by having certain experiences. Justification can also be lost, and the present theory is compatible
with all sorts of views about the conditions under which this happens. For example, it is compatible with
the view advocated here that a) a subject who receives some justification to believe that p, and who realises
that her holding this belief is likely to have significant practical or moral consequences, must subject that
belief to further scrutiny, for example by seeking out others people’s opinion on whether p, seeing further
evidence that bears on p, actively considering reasons to believe that not-p, and so on; and b) if she does not,
she is thereby rationally criticisable; and even that c) she thereby loses (some) justification to believe that p.

Above I argued that the output of the theory in cases of type (3) and (4)— the latter possibly modified by
Goldberg cases—namely that Tom does get justification from having intuitional experience to the effect that
the person on the other side of the street is less likely to be dangerous, the racist aetiology of that experience
notwithstanding—should be accepted. The point I have just now tried to illustrate is that saying that is
wholly compatible with a range of accounts of how epistemic agents must conduct themselves, the further
inquiries they must, in some circumstances, undertake, and what happens to the epistemic status of their
various doxastic states if they do not.

Second, many people think that some binary concepts will inevitably play a significant role somewhere in
the conglomerate of theories covering epistemically or morally justified belief and action. 193 A view naturally
(though not inevitably) combined with that commitment is that thresholds play an important role. For ex-
ample, someone who thinks that a binary concept of belief is needed might well think that for binary belief
to be justified is a matter of justification reaching a certain threshold. And thresholds may shift. Subject-
sensitive invariantism, for example, says that where the epistemically significant threshold is depends on,
among other things, what’s at stake for the subject in question (Fantl and McGrath 2002; Hawthorne 2004;
Stanley 2005). 194

193
See e.g Buchak (2014). For a thorough argument that no binary concept of belief is needed, see Tang (2009).
194
“The advocate of IRI [interest-relative invariantism] … proposes that, in addition to whatever one's favored theory of knowledge
says about when x knows at time t that p, there is a further condition on knowledge that has to do with practical facts about the
subject's environment. One could, therefore, combine IRI with any number of widely differing views about the nature of the
knowledge relation. For example, according to probabilistic strength of evidence IRI, practical facts about a subject’s environment at
time t might make it the case that that subject must have stronger evidence than usual in order to know a proposition p at that time
than she must possess in order to know that proposition at other times, where strength of evidence is measured in probabilistic
terms” (Stanley 2005: 85). Note that subject-sensitive invariantism needn’t hold that it is justification that must reach a certain
threshold.

125
It is not a big leap to say that the stakes may instead be moral (Fritz 2017; or see Basu and Schroeder 2019
for a threshold account in terms of justified belief). Alternatively, the binary epistemic concepts at issue—
belief, acceptance, knowledge—may be sensitive to moral (or practical) concerns in ways that don’t rely on
thresholds, for example because they impact on the risk of a mistake, and thereby whether the agent can ex-
clude certain possibilities from consideration (Bolinger forthcoming; see also Moss 2018). As I hope is evi-
dent, all of these are possibilities for exploring how, even though having an intuition that p absent defeat
provides the intuiter with some justification to believe that p, that person may still fail to be justified out-
right in believing that p, or he may fail to attain some other important binary epistemic status—at least for
those who believe in such.

Third, when we leave the epistemic domain behind, the account of when a person may permissibly speak or
act in certain ways must surely take in more than just the epistemic component, especially if one holds a
view on which the epistemic domain is not itself encroached upon by either pragmatic or moral concerns
(Foley 2000; e.g. Lehrer 2000). 195 Whether a person may act on the basis of her credence depends not only
on whether that credence is justified, but on what the proposed action is, and its consequences. To take but
one exceedingly obvious example, a person who justifiably believes that p may not be in a position to speak
or act accordingly, if to do so would seriously hurt another person’s feelings or expose them to great danger.

Finally, there is also room to slightly re-interpret the theory itself. In type (2) cases I said that the agent has
an undermining defeater for justification of p by E, because she ought to believe q with justification. In that
context the ‘ought’ was understood purely epistemically. But it is also possible to say either that whether the
epistemic ought-claim is true can hinge on moral concerns—similarly to how type (4) cases were modified by
Goldberg’s ‘should have known’—or that that even in cases where the ought-statement is false on an epis-
temic reading, it is true on a moral one, and that if so, justification is still undermined. The former option
may turn out to be a notational variant on one of the above accounts; the latter would amount to quite sig-
nificant moral encroachment. Both are compatible with the core commitments of the view defended in this
book.

7.6 Taking Stock

In this chapter I have considered a style of objection that is based around the idea that the theory which
this book advocates yields justification in cases where there is none. My strategy has in large part been to
divide and, I hope, conquer: by drawing a more nuanced sketch of the landscape I have tried to show that
in some cases, that the subject receives justification from her intuitional experiences is indeed the right con-
clusion, initial appearances possibly notwithstanding; whereas in others there are resources available to ar-
gue that there are defeaters and so no justification. Finally I have tried to show that accepting the theory’s
verdict in the cases when it says that justification is accrued may not be as unpalatable as it to some at first
seems, since the account of how intuition provides justification is but a small part of the full epistemic
story, and a smaller part still of the full theory of how a subject may permissibly speak or act.

195
Foley writes: “The standards that one must meet if one’s beliefs are to be responsible (or non-negligent) slide up or down with
the significance of the issue” (2000: 185).

126
Chapter 8 Final Thoughts

If the account of the nature and epistemology of intuition given in earlier chapters is accepted, what fol-
lows? Which questions emerge as important, and what are the consequences for how we should think of the
role of intuition in our epistemic lives? I end the book with some reflections on these issues.

8.1 Two Questions

The account I have advocated says that if certain conditions are met, then, absent defeat, S’s having an intu-
itional experience that p makes her to a non-negligible degree justified in believing that p. Given this formu-
lation two questions stand out: what are these conditions, and how much justification?

Regarding the former, it is especially important whether having justification to believe some other proposi-
tion is among the necessary conditions. In particular, is having justification to believe that all non-intuiting
hypotheses are false one of them?

Recall again that an important historical antecedent to the view developed here is James Pryor’s Dogmatism
for perception. That view denies that having justification to believe any other proposition is among the nec-
essary conditions for perceptual experience to make a person justified in believing what it represents. A forti-
ori, the subject needn’t have justification to believe that no non-perceiving hypothesis is true. 196 A non-perceiv-
ing hypothesis is a hypothesis the truth of which is incompatible with perceptual experience constituting
successful perception of the world (Pryor 2004: 355), either because the world isn’t as it is represented as
being, or because the experience fails to be connected with the world in the right way (as in cases of veridi-
cal hallucination, for example). Correspondingly, a non-intuiting hypothesis about the intuition that p says
that either p is false, or p is true but the intuition is not connected with the way things are in the right sort
of way.

In Chapter 5 I noted Pryor’s argument that it seems that nothing but having a perceptual experience is nec-
essary for perceptual belief to be justified, and that absent strong reason to the contrary, we should take
such appearances at face value. I used this reasoning as part of my argument that having a perceptual experi-
ence singlehandedly justified belief—it is what makes the subject justified. Much of the strength of that argu-
ment transfers across, not only from Liberalism to Dogmatism for perception, but also to Dogmatism for
intuition. No less than in the case of perception does it seem that having the intuitional experience is all
that is required to be justified. In particular, having justification to believe that all non-intuiting hypotheses
are false does not seem to be required. Moreover, I take it that the considerations in Chapters 6 and 7 signif-
icantly strengthen the claim that there are no good reasons not to take these appearances at face value. So I

196
Pryor earlier expressed this point in terms of bad scenarios. For a discussion of why he switched, see (Koksvik 2011: §6.3).

127
think we have good reason to believe that having justification to believe that all non-intuiting hypotheses
are false is not among the necessary conditions mentioned in Liberalism. 197 Beyond this I tend to think that
the necessary conditions are pretty minimal, and that they largely boil down to the necessary conditions for
even holding the purportedly justified beliefs. I will attempt no argument for this conclusion here, however.

For those who tend to think the opposite, a further question is whether, in order that having an intuitional
experience may justify belief in what it represents, a person must really be justified in believing that all the
non-intuiting hypotheses are false in the sense that has been at issue in this book, or whether a lesser epis-
temic status might suffice. One alternative is what Crispin Wright calls ‘entitlement’ (Wright 2004). Entitle-
ment, Wright says, is not the mark of a cognitive achievement, and it “does not require the existence of evi-
dence (2004: 174). We can get such ‘warrant for nothing’, Wright argues, for a range of reasons. For exam-
ple, some cognitive projects that are either indispensable or very valuable to us rely on presuppositions such
that I couldn’t undertake to check them except in a way that committed me to new presuppositions no
more secure than the first (190-1). In such cases, he says, we are “rationally entitled to … trust” that the pre-
suppositions are met (192). There may be a case to be made for a rational entitlement to trust that the non-
intuiting hypotheses are false along similar lines.198

As regards the second question, it is strategically tempting to remain agnostic on how strong justification
intuition can provide; and it is certainly open to readers to take the above account on board while rejecting
what I am now about to say. That said, I think we have very good reason to think that intuition can supply
us with justification of all different strengths, including very strong justification. Here are some considera-
tions in favour of such a view.

First, at least among those who think that intuition can justify belief at all, it is widely recognised that intui-
tion can overturn previous consensus. Many describe the publication of Gettier’s famous paper in this way, for
example (Gettier 1963). But even though there is disagreement about which cases constitute the best exam-
ples of this happening, it strikes me as hard to deny that this in fact takes place. This point is often, and
rightly, used to argue that intuition isn’t plausibly thought of as the mere output of previously held beliefs.
But it is at least equally convincing as a consideration in favour of intuition justifying strong credence. For
if intuition only justified weak credence it wouldn’t be able to play the consensus-overturning role that it
evidently does play, since the propositions held in consensus are presumably held with quite strong justifica-
tion.

Secondly, a similar point can be made for anyone who thinks that intuition can substantially constrain the-
ory development, in the sense of being part of an argument that sets limits for what subsequently developed
theories may deliver.

197
One important question is whether this view is compatible with Bayesianism. Several authors have argued that it is not (White
2006; Hawthorne 2004; Silins 2007). In his (2013) Pryor argues that the challenge depends not merely on the formalism of that
framework but on certain interpretive assumptions that can be challenged.
198
This is probably not a case Wright himself would endorse. Interestingly, Wright does not think that the rationale for entitlement
extends to belief in an external world. The closest he gets to an argument for entitlement to believe in an external world is that it
may, he think, be a precondition for thinking about oneself as a part of an objective world, in the sense of one which extends tem-
porally and spatially beyond oneself. This, of course, is especially suggestive to me, given the importance I have placed on the phe-
nomenology of objectivity in both perception and intuition. The argument envisaged in the text is related to the argument dis-
cussed in n. 73 above.

128
For example, Frank Jackson and Philip Pettit argue that it just can’t turn out that there are no beliefs and
desires, and that we know that things can’t turn out that way because of the success of folk psychology
(Jackson and Pettit 1990). This argument pretty clearly relies on intuition in crucial places. For example, it
relies on the intuition that (it is true that) “to believe (or to desire) that so-and-so is to have a state in one
playing the role definitive of that belief between inputs, outputs, and other functionally specified states”
(33)—a formulation they derive from Paul Churchland (1981). Even more obviously, they rely on the intui-
tion that our widespread success in predicting other people’s behaviour can’t ‘be chance’ (Jackson and Pettit
1990: 35).

Jackson and Pettit thus argue that there are certain results that a future scientific theory is just not ‘allowed’
to deliver: it just doesn’t get to say that there are no beliefs and desires. If one thinks that an argument that
relies on intuition can legitimately play this role—and I do—then one is committed to the view that intuition
can provide strong justification indeed. For the argument is supposed to limit future science, and we can
imagine that a future scientific theory can have quite strong justification on its own.

In a similar vein, Terence Cuneo and Russ Shafer-Landau argue that there is “a battery of substantive moral
propositions” that “must find a place in any system of moral norms that applies to beings like us, in worlds
similar to ours” (2014: 399). Among candidates for such ‘moral fixed points’ they mention the propositions
that it is wrong to torture others just because they have inconvenienced you and that it is wrong to rape a
child solely to indulge one's lust (400). If one thinks that belief in such propositions can be justified by intu-
ition, and that they can while so justified play the role the authors wish them to play, namely to “fix the
boundaries of moral thought” (401)—and I do—then again one is committed to intuition being able to pro-
vide very strong justification indeed, for reasons parallel to those just given. 199

So we have good reason to think that intuition sometimes provides very strong justification for belief.

8.2 The Role of Intuition in Philosophy

If the account I have presented is accepted, how should we think about the role of intuition in philosophy?

One obvious and immediate consequence of accepting this account is that intuition must be recognised as
having a legitimate place in philosophical methodology. Clearly we often have intuitions with philosophically
relevant content; content such that it matters for some philosophical theory or other whether the intuited
proposition is true or false. If intuition provides justification for belief absent defeat, and provided defeat is
not omnipresent,200 intuition is certainly a legitimate part of the philosopher’s toolkit.

The theory I have advocated also entails that the strength of intuitions will matter a lot. In particular, it
rules out an approach on which one takes all beliefs justified by intuition as equals, and attempt to systema-
tise them in a way that accounts for as many of them as possible without regard to the strength of justification

199
The authors’ view is that these truths are knowable “simply by adequately understanding its constituent concepts and their rela-
tions to one another” (Cuneo and Shafer-Landau 2014: 408). I don’t think this is a promising account of a priori knowledge but I
needn’t belabour this point here: my claim is that it’s plausible that they can play that kind of role while justified by intuition.
200
Cf Chudnoff (2013b: §3.3).

129
for each belief.201 There is an unfortunate but understandable reluctance in contemporary philosophy to
pay much attention to the strength of intuition. We say things like: ‘Yes, p seems to be true. But there’s also
an intuition that q, and the two seem to be in tension.’ And we often do so without regard to the relative
strength of the two intuitions—presumably because we find it hard to get precise about these strengths.

If the theory I have advocated is correct, we should try to move away from this tendency. It is unlikely that
we will be able to assign precise numerical values to the strength of our intuitions. But it is not unrealistic
that most of us can roughly assign strengths to their intuitions that reflect not only the ordinal facts about
those strengths (‘this intuition is stronger than that one’), but also some cardinal facts (this is much stronger
than that). This won't be exact, but we will at least get the broad structure of the ordinal and cardinal prop-
erties of the strengths of our intuitions. We can expect that different people will differ a bit about this, but
it is plausible that there will be enough overlap between individuals that we can aggregate many intuitions
into a general, if partial, consensus on how intuitions should be ordered by strength, and roughly how
much stronger one intuition is than another. That would already take us a long way.

For the role of intuition in philosophy it will also obviously be important how thick on the ground defeat-
ers are. I want to make two points about this: one ambitious, the other humble.

Consider the conjunction fallacy, mentioned earlier. When a person learns the relevant part of probability
theory, the justification from the particular intuition she had is defeated. But that is not quite all. Defeat
spreads. It spreads, in this instance, at the very least to other cases that are immediately recognisable as in-
stances of the same fallacy. There will be a number of such cases: we get some by substituting the name of
the protagonist bank teller for another name, for instance; but we can get others by changing the conjoined
propositions entirely. Perhaps at some point it becomes sufficiently unobvious that the case is an instance of
the same fallacy, so that defeat fails to spread to such cases. The hopeful and ambitious point is that there is
a field of inquiry here: that which maps out the scope of defeat. Getting clearer on the contours of this
landscape would be revealing in considering the role of intuition in philosophical methodology. And it may
well be that a significant contribution can be made by the meta-theorist: perhaps there are interesting and
domain-independent rules to be found.

Now for the humble point. Even if there are such rules, the more fundamental question is what counts as a
defeater to begin with. And I want to suggest that this may be a question which the meta- theoretician must
hand the issue straight back to the disciplines and subdisciplines themselves.

The dominant view of theory-choice in science is that it is a matter of balancing different desiderata. Candi-
dates include simplicity, realism (i.e., the absence of idealising assumptions, a la the frictionless plane), fit
with existing data, prediction of future observations, generality, elegance, and others. 202 We can’t maximise
all of them, since different desiderata pull in opposite directions. Moreover, Thomas Kuhn famously argued
that there’s “no natural algorithm for theory choice” because different theorists, even if they agree on the

201
Such an account seems to be suggested by F.M. Kamm: “Consider as many case-based judgments of yours as prove necessary. Do
not ignore some case-based judgments, assuming they are errors, just because they conflict with simple or intuitively plausible princi-
ples that account for some subset of your case-based judgments. Work on the assumption that a different principle can account for
all of the judgments” (2007: 5).
202
Whether fit with existing data and prediction of future observations are distinct is one of many hotly debated topics in this area;
see e.g. Hitchcock and Sober (2004).

130
desiderata, may weigh them differently (Kuhn 1962/2009: 200). A vivid illustration of this is given by Lev-
ins (see e.g. Levins 1966), who notes that in population biology, the desiderata he calls realism, generality,
and precision are balanced differently by different theorists. A special instance of this general phenomenon
is when different disciplines weigh the desiderata differently.

In psychology, a widely used standard for a result being publishable in a scientific journal is that the p-value
is less than or equal to 0.05. Such results are standardly reported as ‘statistically significant’. This threshold
being met is often understood as indicating that the probability that the hypothesis being tested is false,
given the data that was gathered, is less than or equal to five percent. It does not, in fact, show this (see e.g.
Wasserstein and Lazar (2016)), but the details don’t matter here.203 For us the important point is that this
standard differs sharply from those operative in other fields. In particle physics, for example, to qualify as
‘evidence’ the probability that the data is a statistical fluctuation must be less than or equal to 0.0013, and
to qualify as an ‘observation’ it must be less than or equal to 0.000000287. This doesn’t outright demon-
strate that the two disciplines weigh theoretical desiderata differently, but it is certainly evidence to that ef-
fect. Moreover, when different scientific disciplines weigh theoretical desiderata differently, they plausibly
do so reasonably: fit with the data seems less crucial in a field like sociology than in chemistry, for example.

Closer to home, I take it that the theoretical desiderata in philosophy are different than in law, and differ-
ent again between different sub-disciplines of philosophy. Some parts of philosophy of language, for in-
stance, take fit with the data of felicitousness-intuitions very seriously indeed, and much simplicity is sacri-
ficed in its favour. By contrast, in metaphysics it is not unheard of for the elegance or simplicity of a theory
to outweigh such apparently egregious conflicts with intuition as that tables and chairs don’t exist. So intui-
tion is treated quite differently in different subdisciplines of philosophy. And this is not obviously im-
proper. Those of us who don’t work in the area can sometimes feel a certain exasperation when felicitous-
ness-intuitions are solicited for strange and unusual sentences, but it would be unreasonable, I take it, to
suggest that sub-discipline as a whole is completely on the wrong track here.

In analogous fashion it is likely that what counts as a defeater will vary with sub-discipline. We get some evi-
dence for this claim from thinking about widespread intuitions from different areas altogether. Almost eve-
ryone has the intuition that simultaneity is transitive, for example, and are thereby, on my view, justified in
holding that belief. But that justification is entirely defeated when one learns that strongly confirmed theo-
ries of physics entail that this is false, because simultaneity is a physical phenomenon.

A simple, elegant, general, and explanatory theory in metaphysics may with reason be taken to undermine
our everyday intuitions about tables and chairs. Whether this is so will depend on what metaphysics is all
about. And that is a question for that branch of inquiry itself: for metaphysics, or, perhaps, meta-metaphysics.

Similarly, supposed it turned out that we can adequately explain why we have (a sub-set of) our moral intui-
tions without invoking their truth.204 (I don’t think the somewhat common claims that we are already at

203
Instead it indicates “how incompatible the data is with a specified statistical model” (Wasserstein and Lazar 2016: 131). Since the
model is constructed out of both a null-hypothesis—e.g. the hypothesis of no correlation between two variables (such as smoking and
cancer)—and a set of assumptions, the data can cast doubt on either one.
204
See e.g. Street (2006) for a recent account to this effect.

131
this point are warranted, but let’s set that aside.) Whether this counts as a defeater for moral intuitions de-
pends in part on what ethics is all about. And that is a claim for ethics and its meta-discipline, meta-ethics;
and not for the more general meta-theoretician.

To sum up, accepting the theory I have been advocating about the nature and epistemology of intuition
commits one to accepting a role for intuition in philosophical methodology, and to the importance of pay-
ing attention to the different strengths of different intuitions. But it is plausible that an important part of
the story must be passed right back to the theorists themselves, since how important it is for a theory to fit
with intuitions, and what it takes for the justification intuitions provide to be defeated, are plausibly ques-
tions for the sub-disciplines themselves, or for their meta-disciplines, or for both. This moderate meta-theo-
retical humility is, I take it, both a plausible and a pleasing result.

8.3 The Role of Intuition in Everyday Life

Finally I want to very briefly comment on what the theory entails for the role of intuition in our everyday
epistemic lives.

A distinctive feature of the account that’s been advocated here is that, since intuition is not restricted by
content, it can in principle provide justification for any content the agent can believe. But if intuition can
justify all sorts of everyday beliefs, why, you might ask, does intuition play so much more of a prominent
role in philosophy (and perhaps other academic disciplines) than in our everyday epistemic lives?

For many everyday beliefs, defeaters and outweighing evidence are thick on the ground. It follows from my
view that you can become justified (to some degree) in believing that there is a person behind you by having
an intuition to that effect. But in any normal circumstance what you should then do is to turn your head,
in which case you will often get stronger justification to believe the opposite. To the extent that intuition
plays a smaller role in belief formation for everyday contents than does perception, this can fairly easily be
explained by our (obviously sound) cognitive habits of not forming beliefs that are routinely defeated.

But second, it is not, I think, in fact all that obvious that intuition doesn’t play an important role in our eve-
ryday epistemic lives. We form a great many beliefs every day, with all sorts of different contents. Many of
them are not easily explained as perceptual, but seem justified all the same. When we form beliefs about
other people’s trustworthiness, aggressive or benign intentions, likelihood to cooperate, and so on, for ex-
ample, the justification isn’t straightforwardly perceptual, since it doesn’t seem that our perceptual experi-
ence represents that the person is any of these things. (Perceptual experience wouldn’t be inaccurate if the
person were untrustworthy.) And yet such beliefs seem justified in the central sense at issue in this book. So
it is, I think, quite plausible that intuition, as this psychological kind has been conceived of here, and what-
ever its role in philosophical, theoretical, and scientific inquiry more generally, plays a pervasive and im-
portant role in our everyday mental and rational lives. And, I insist, appropriately so; for absent defeat intui-
tion justifies belief in its representational content, whatever that turns out to be.

132
133
REFERENCES

Adams, Fred, and Ken Aizawa. 2010/2017. "Causal Theories of Mental Content." In The Stanford
Encyclopedia of Philosophy, edited by Edward N. Zalta.
Alston, William P. 1985. 'Concepts of Ppistemic Justification', The Monist, 68: 57-89.
———. 1989. Epistemic Justification: Essays in the theory of knowledge (Cornell University Press).
———. 2005. Beyond Justification: Dimensions of Epistemic Evaluation (Cornell University Press).
Armstrong, David M. 1968. A Materialist Theory of the Mind (Routledge).
Audi, Robert. 1998/2011. Epistemology: A Contemporary Introduction to the Theory of Knowledge
(Routledge).
Ayer, A.J. 1952. Language, Truth and Logic (Dover Publications).
———. 1956/1964. The Problem of Knowledge (Penguin).
Barnes, W.H.F. 1933. 'A Suggestion about Value', Analysis, 1: 45-46.
Basu, Rima, and Mark Schroeder. 2019. 'Doxastic Wronging.' in Brian Kim and Mark Schroeder (eds.),
Pragmatic Encroachment in Epistemology. (Routledge).
Bealer, George. 1987. 'The Philosophical Limits of Scientific Essentialism', Philosophical Perspectives, 1:
289-365.
———. 1992. 'The Incoherence of Empiricism', Aristotelian Society Supplementary Volume, 66: 99-138.
———. 1996a. 'On the Possibility of Philosophical Knowledge', Philosophical Perspectives, 10: 1-34.
———. 1996b. 'A Priori Knowledge and the Scope of Philosophy', Philosophical Studies, 81: 121-42.
———. 1998a. 'Intuition and the Autonomy of Philosophy.' in Michael DePaul and William Ramsey
(eds.), Rethinking Intuition: The Psychology of Intuition and Its Role in Philosophical Inquiry
(Rowman & Littlefield).
———. 1998b. 'A Theory of Concepts and Concept Possession', Philosophical Issues, 9: 261-301.
———. 2001. 'A Theory of the A Priori', Pacific Philosophical Quarterly, 81: 1-30.
———. 2002. 'Modal Epistemology and the Rationalist Renaissance.' in Tamar Szabó Gendler and John
Hawthorne (eds.), Conceivability and Possibility (Oxford University Press).
———. 2004. 'The Origins of Modal Error', Dialectica, 58: 11-42.
———. 2008. 'Intuition and Modal Error.' in Quentin Smith (ed.), Epistemology: New Essays (Oxford
University Press).
Benacerraf, Paul. 1973. 'Mathematical Truth', The Journal of Philosophy, 70: 661-79.
Bengson, John. 2010. 'The Intellectual Given ', University of Texas.
———. 2014. 'How Philosophers use Intuition and "Intuition“', Philosophical Studies, 171: 555-76.
———. 2015a. 'Grasping the Third Realm.' in Tamar Szabó Gendler, and Hawthorne, John (ed.), Oxford
Studies in Epistemology (Oxford University Press).
———. 2015b. 'The Intellectual Given', Mind, 124: 707-60.
Block, Ned. 1995. 'On a Confusion about a Function of Consciousness', Behavioral and Brain Sciences, 18:
227-87.
Boghossian, Paul A. 2000. 'Knowledge of Logic.' in Paul A. Boghossian and Christopher Peacocke (eds.),
New Essays on the A Priori (Oxford University Press).
———. 2001. 'Inference and Insight', Philosophical and Phenomenological Research, 63: 633-40.
———. 2009. 'Virtuous Intuitions: Comments on Lecture 3 of Ernest Sosa's A Virtue Epistemology',
Philosophical Studies, 144: 111-19.
Bolinger, Renée Jorgensen forthcoming. 'The Rational Impermissibility of Accepting Racial
Generalizations', Synthese: 1-17.
BonJour, Laurence. 1985. The Structure of Empirical Knowledge (Harvard University Press).
———. 1998. In Defence of Pure Reason: A Rationalist Account of a Priori Justification (Cambridge
University Press).

134
Bourget, David, and Angela Mendelovici. 2016/2019. "Phenomenal Intentionality." In The Stanford
Encyclopedia of Philosophy, edited by Edward N. Zalta.
Bovens, L, and S Hartmann. 2003. Bayesian Epistemology (Clarendon Press).
Bratman, Michael E. 1987. Intentions, Plans, and Practical Reason (Harvard University Press).
Buchak, Lara. 2014. 'Belief, Credence, and Norms', Philosophical Studies, 169: 285-311.
Burge, Tyler. 2009. 'Perceptual Objectivity', Philosophical Review, 118: 285-324.
Byrne, Alex. 2009. 'Experience and Content', Philosophical Quarterly, 59.
Cappelen, Herman. 2012. Philosophy Without Intuitions (Oxford University Press ).
Chalmers, David. 1996. The Conscious Mind (Oxford University Press).
———. 1998. 'On the Search for the Neural Correlate of Consciousness.' in Stuart R. Hameroff, Alfred W.
Kaszniak and A. C. Scott (eds.), Toward a Science of Consciousness II (The MIT Press).
———. 2000. 'What is a Neural Correlate of Consciousness? .' in T Metzinger (ed.), Neural Correlates of
Consciousness: Empirical and Conceptual Questions (The MIT Press).
———. 2004. 'The Representational Character of Experience.' in Brian Leiter (ed.), The Future for
Philosophy (Oxford University Press).
———. 2011. 'Verbal Disputes', Philosophical Disputes, 120: 515-66.
Chihara, Charles. 1982. 'A Gödelian Thesis Regarding Mathematical Objects: Do They Exist? and Can We
Perceive Them?', The Philosophical Review, 91: 211-27.
Chisholm, Roderick. 1957. Perceiving: A Philosophical Study (Cornell University Press).
Christensen, David. 2010. 'Higher-Order Evidence', Philosophical and Phenomenological Research, 81:
185-215.
Chudnoff, Elijah. 2011a. 'The Nature of Intuitive Justification', Philosophical Studies, 153: 313-33.
———. 2011b. 'What Intuitions Are Like', Philosophical and Phenomenological Research, 82: 625-54.
———. 2012. 'Presentational Phenomenology.' in S Migues and G Preyer (eds.), Consciousness and
Subjectivity (Ontos).
———. 2013a. 'Gurwitsch's phenomenal holism', Phenomenology and the Cognitive Sciences, 12: 559-
78.
———. 2013b. Intuition (Oxford University Press).
———. 2014. 'The Rational Role of Intuition.' in Anthony Robert Booth and Darrell P. Rowbottom (eds.),
Intuitions (Oxford University Press).
———. 2016. 'Epistemic Elitism and Other Minds', Philosophy and Phenomenological Research, 96: 276-
98.
———. 2017. 'The epistemic significance of perceptual learning', Inquiry, 61: 520-42.
———. 2019. 'Experience and Epistemic Structure: Can Cognitive Penetration Result in Epistemic
Downgrade?' in Timothy Chan and Anders Nes (eds.), Inference and Consciousness (Routledge).
Churchland, Paul M. 1981. 'Eliminative Materialism and Propositional Attitudes', The Journal of
Philosophy, 78: 67-90.
Climenhaga, Nevin. 2018. 'Intuitions are Used as Evidence in Philosophy', Mind, 127: 69-104.
Cohen, Laurence Jonathan. 1981. 'Can Human Irrationality Be Experimentally Demonstrated', Behavioral
and Brain Sciences, 4: 317-70.
———. 1986. The Dialogue of Reason (Clarendon Press).
Cohen, Stewart. 2010. 'Bootstrapping, Defeasible Reasoning, and A Priori Justification', Philosophical
Perspectives, 24: 141-59.
Crane, Tim. 1988b. 'The Waterfall Illusion', Analysis, 48: 142-47.
———. 1992b. 'The Nonconceptual Content of Experience.' in Tim Crane (ed.), The Contents of
Experience (Cambridge University Press).
———. 2001. Elements of Mind (Oxford University Press).
Crane, Tim 1988a. 'Contents in Perception', Analysis, 48: 150-53.

135
Cummins, Robert C. 1998. 'Reflection on Reflective Equilibrium.' in Michael DePaul, and Ramsey, William
(ed.), Rethinking Intuition: The Psychology of Intuition and Its Role in Philosophical Inquiry
(Rowman & Littlefield).
Cuneo, Terence, and Shafer-Landau. 2014. 'The Moral Fixed Points: new directions for moral
nonnaturalism', Philosophical Studies, 171: 399-443.
Dainton, Barry. 2000. Stream of Consciousness: Unity and Continuity in Conscious Experience
(Routledge).
Davidson, Janet E. 2019. 'Inductive Reasoning.' in Robert J. Sternberg and Joachim Funke (eds.), The
Psychology of Human Thought: an introduction (Heidelberg University Press).
Davies, Martin. 2000a. 'Externalism and Armchair Knowledge.' in Paul A. Boghossian and Christopher
Peacocke (eds.), New Essays on the A Priori (Oxford University Press).
———. 2000b. 'Externalism, Architecturalism, and Epistemic Warrant.' in Crispin Wright, Barry C. Smith
and Cynthia Macdonald (eds.), Knowing Our Own Minds (Oxford University Press).
Dennett, Daniel. 1988. 'Quining Qualia.' in A Marcel and E Bisiach (eds.), Consciousness in Modern
Science (Oxford University Press).
DePaul, Michael. 2009. 'Phenomenal Conservativism and Self-Defeat', Philosophical and
Phenomenological Research, 78: 205-12.
Deutsch, Max. 2015. The Myth of the Intuitive (MIT Press).
Devitt, Michael. 2005. 'There Is No a Priori.' in Matthias Steup and Ernest Sosa (eds.), Contemporary
Debates in Epistemology (Blackwell).
Dorsch, F. 2009. 'Judging and the Scope of Mental Agency.' in L O’Brien and M Soteriou (eds.), Mental
Action (Oxford University Press).
Dretske, Fred. 1969. Seeing and Knowing (Routledge & Kegan Paul).
———. 2000. 'Entitlement: Epistemic Rights without Epistemic Duties?', Philosophical and
Phenomenological Research, 60: 591-606.
Earlenbaugh, Joshua, and Bernard Molyneux. 2009. 'Intuitions are Inclinations to Believe', Philosophical
Studies, 145: 89-109.
Evans, Gareth. 1982. The Varieties of Reference (Clarendon Press).
Fantl, Jeremy, and Matthew McGrath. 2002. 'Evidence, Pragmatics, and Justification,' The Philosophical
Review, 111: 67-94.
Firth, Roderick. 1978. 'Are Epistemic Concepts Reducible to Ethical Concepts?' in Alvin Goldman and
Jaegwon Kim (ed.), Values and Morals: Essays in Honour of William Frankena, Charles Stevenson,
and Richard Brandt (D. Reidel).
Foley, Richard. 2000. 'Epistemically Rational Belief and Responsible Belief', The Proceedings of the
Twentieth World Congress of Philosophy: 181-88.
Fricker, Miranda. 2009. 'The Value of Knowledge and the Test of Time', Royal Institute of Philosophy
Supplement 64: 121-38.
Fritz, James. 2017. 'Pragmatic Encroachment and Moral Encroachment', Pacific Philosophical Quarterly,
98: 643-61.
Fumerton, Richard. 1990. Reason and Morality: A Defense of the Egocentric Perspective (Cornell
University Press).
———. 1999. 'A Critique of Coherentism.' in Louis P. Pojman (ed.), The Theory of Knowledge: Classic and
Contemporary Readings (Wadsworth/Thomson Learning).
Gettier, Edmund. 1963. 'Is Justified True Belief Knowledge', Analysis, 23: 121-23.
Ghijsen, Harmen. 2014. 'Phenomenalist Dogmatist Experientialism and the Distinctiveness Problem',
Synthese, 191: 1549-66.

136
———. 2016. The Puzzle of Perceptual Justification: Conscious experience, Higher-order Beliefs, and
Reliable Processes (Springer).
Goldberg, Sanford C. 2017. 'Should Have Known', Synthese, 194: 2863-94.
Goldie, Peter. 2002. 'Emotions, Feelings and Intentionality', Phenomenology and the Cognitive Sciences,
1`: 235-54.
Goldman, Alvin. 1979. 'What is Justified Belief?' in George Sotiros Pappas (ed.), Justification and
Knowledge: New Studies in Epistemology (D. Reidel).
———. 1987. 'Cognitive Science and Metaphysics', The Journal of Philosophy, 84: 537-44.
———. 1999. 'A Priori Warrant and Naturalistic Epistemology', Philosophical Perspectives, 13: 1-28.
Goldman, Alvin, and Joel Pust. 1998. 'Philosophical Theory and Intuitional Evidence.' in Michael DePaul,
and Ramsey, William (ed.), Rethinking Intuition: The Psychology of Intuition and Its Role in
Philosophical Inquiry (Rowman & Littlefield).
Gopnik, Alison, and Schwitzgebel. 1998. 'Whose Concepts Are They, Anyway? The Role of Philosophical
Intuition in Empirical Psychology.' in Michael DePaul and William Ramsey (eds.), Rethinking
Intuition: The Psychology of Intuition and Its Role in Philosophical Inquiry. (Rowman & Littlefield).
Grice, H. P. 1962/1989. 'Some Remarks about the Senses.' in, Studies in the Way of Words (Harvard
University Press).
Grundmann, Thomas. 2007. 'The Nature of Rational Intuitions and a Fresh Look at the Explanationist
Objection', Grazer Philosophische Studien, 74: 69-87.
Haddock, Adrian, Millar, Haddock and Pritchard, Duncan (ed.)^(eds.). 2009. Epistemic Value (Oxford
University Press).
Hahn, Ulrike, and Adam J. L. Harris. 2014. 'What Does It Mean to be Biased: Motivated Reasoning and
Rationality.' in Brian H. Ross (ed.), Psychology of Learning and Motivation (Academic Press).
Haidt, J, F Bjorklund, and S Murphy. 2000. "Moral Dumbfounding: when intuition finds no reason." In.
University of Virginia.
Haidt, Jonathan. 2001. 'The Emotional Dog and its Rational Tail: a social intuitionist approach to moral
judgment.', Psychological Review, 108: 814-34.
Hájek, Alan 2016. 'Philosophical Heuristics and Philosophical Methodology ' in Herman Cappelen, Tamar
Szabó Gendler and John Hawthorne (eds.), The Oxford Handbook of Philosophical Methodology
(Oxford University Press).
Hanna, Nathan. 2011. 'Against Phenomenal Conservatism', Acta Analytica, 26: 213-21.
Harman, Gilbert. 1977. The Nature of Morality (Oxford University Press).
———. 1986. Change in View: Principles of Reasoning (MIT Press).
———. 1990. 'The Intrinsic Quality of Experience.' in Ned Block (ed.), The Nature of Consciousness (The
MIT Press).
Hawthorne, John. 2004. Knowledge and Lotteries (Oxford University Press).
Heck, Richard G. Jr. 2000. 'Nonconceptual Content and the „Space of Reasons’’ ', Philosophical Review,
109: 483-523.
Hintikka, Jaakko. 1999. 'The Emperor’s New Intuitions', The Journal of Philosophy, 96: 127-47.
Hitchcock, Christopher, and Elliot Sober. 2004. 'Prediction Versus Accommodation and the Risk of
Overfitting', The British Journal for the Philosophy of Science, 55: 1-34.
Horgan, Terrence, and George Graham. 2010. 'Phenomenal Intentionality and Content Determinacy.' in
R Shantz (ed.), Prospects for Meaning (de Gruyter).
Horgan, Terrence, and John Tienson. 2002. 'The Intentionality of Phenomenology and the
Phenomenology of Intentionality.' in David Chalmers (ed.), Philosophy of Mind: Classical and
Contemporary Readings, (Oxford University Press).

137
Horgan, Terrence, John Tienson, and George Graham. 2003. 'The Phenomenology of First Person
Agency.' in Sven Walter and Heinz-Dieter Heckmann (eds.), Physicalism and Mental Causation:
the Metaphysics of Mind and Action (Imprint Academic).
———. 2004. 'Phenomenal Intentionality and the Brain in a Vat.' in R Shantz (ed.), The Externalist
Challenge: New Studies on Cognition and Intentionality (de Gruyter).
Horgan, Terrence, and M. C. Timmons. 2007. 'Moorean Moral Phenomenology.' in Susana Nuccetelli and
Gary Seay (eds.), Themes from G.E. Moore (Oxford University Press).
Huemer, Michael. 2001. Scepticism and the Veil of Perception (Rowman & Littlefield).
———. 2005. Ethical Intuitionism (Palgrave Macmillan).
———. 2007. 'Compassionate Phenomenal Conservatism', Philosophical and Phenomenological
Research, 74: 30-55.
———. 2009. 'Apology of a Modest Intuitionist', Philosophical and Phenomenological Research, 78: 222-
36.
———. 2013. 'Phenomenal Conservatism Über Alles.' in Chris Tucker (ed.), Seemings and Justification:
New Essays on Dogmatism and Phenomenal Conservatism (Oxford University Press).
Humberstone, I. Lloyd. 1992. 'Direction of Fit', Mind, 101: 59-83.
Hume, David. 1739/2007. A Treatise of Human Nature (Clarendon Press).
Ichikawa, Jonathan, and Benjamin Jarvis. 2009. 'Thought-Experiment Intuitions and Truth in Fiction',
Philosophical Studies, 142: 221-46.
Irvine, Andrew David, and Harry Deutsch. 1997/2016. "Russell’s Paradox." In The Stanford Encyclopedia
of Philosophy, edited by Edward N. Zalta.
Jackson, Frank. 2003. 'Mind and Illusion.' in Anthony O’Hear (ed.), Minds and Persons (Cambridge
University Press).
———. 2010. Language, Names, and Information (Wiley-Blackwell).
Jackson, Frank, and Philip Pettit. 1990. 'In Defence of Folk Psychology', Philosophical Studies, 59: 31-54.
Kagan, Shelly. 1989. The Limits of Morality (Oxford University Press).
Kamm, F. M. 2007. Intricate Ethics: Rights, Responsibilities, and Permissable Harm (Oxford University
Press).
Kaplan, Mark. 1985. 'It’s Not What You Know that Counts', The Journal of Philosophy, 82: 350-63.
Katz, Jerrold J. 1981. Language and Other Abstract Objects (Rowman & Littlefield).
Kelly, Thomas. 2006/2016. "Evidence." In The Stanford Encyclopedia of Philosophy, edited by Edward N.
Zalta.
Kim, Jaegwon. 1988. 'What Is "Naturalized Epistemology?"', Philosophical Perspectives, 2: 381-405.
Kind, Amy. 2001. 'Putting the Image Back in Imagination', Philosophical and Phenomenological Research,
62: 85-109.
———. 2003. 'What’s so Transparent about Transparency', Philosophical Studies, 115: 225-44.
———. 2018. 'How Imagination Gives rise to Knowledge.' in Fabian Dorsch and Fiona Macpherson (eds.),
Perceptual Imagination and Perceptual Memory (Oxford University Press).
———. Forthcoming. 'What Imagination Teaches.' in John Schwenkler and Enoch Lambert (eds.),
Becoming Someone New: Transformative Experience (Oxford University Press).
Koksvik, Ole. 2011. 'Intuition', Australian National University.
———. 2013. 'Intuition and Conscious Reasoning', The Philosophical Quarterly, 63: 709-15.
———. 2014. 'Three Models of Phenomenal Unity', Journal of Consciousness Studies,, 21: 105-31.
———. 2015. 'Phenomenal Contrast: A Critique', American Philosophical Quarterly, 52: 321-34.
———. 2017. 'The Phenomenology of Intuition', Philosophical Compass, 12.
Kriegel, Uriah. 2004. 'Consciousness and Self-Consciousness', The Monist, 82: 182–205.
———. 2007. 'The Phenomenologically Manifest', Phenomenology and the Cognitive Sciences, 6: 115–
36.

138
———. 2009. Subjective Consciousness: A Self-Representational Theory (Oxford University Press).
———. 2015. 'The Character of Cognitive Phenomenology.' in T Breyer and C Gutlands (eds.),
Phenomenology of Thinking (Routledge).
Kuhn, Thomas S. 1962/2009. The Structure of Scientific Revolutions (The University of Chicago Press).
Lehrer, Keith. 2000. 'Sensitivity, Indiscernibility and Knowledge', Philosophical Issues, 10: 33-37.
Lehrer, Keith, and Stewart Cohen. 1983. 'Justification, Truth, and Coherence', Synthese, 55: 191-207.
Levins, Richard. 1966. 'The Strategy of Model Building in Population Biology', American Scientist, 54:
421-31.
Lewis, David. 1974/1983. 'Radical Interpretation.' in, Philosophical Papers.
———. 1982. 'Logic for Equivocators', Noûs, 16: 431-41.
———. 1983. Philosophical Papers, Volume 1 (Oxford University Press).
———. 1986. On the Plurality of Worlds (Blackwell).
———. 1994. 'Reduction of Mind.' in Samuel Guttenplan (ed.), A Companion to the Philosophy of Mind.
Locke, John. 1689/1996. An Essay Concerning Understanding (Hackett).
Lynch, Michael P. 2006. 'Trusting Intuition.' in Patrick Greenough and Michael P. Lynch (eds.), Truth and
Realism (Oxford University Press).
Mackie, J., L. 1977/1990. Ethics: Inventing Right and Wrong (Penguin).
Mandelbaum, Maurice. 1955. The Phenomenology of Moral Experience (The Free Press).
Markie, Peter. 2004/2017. "Rationalism vs. Empiricism." In The Stanford Encyclopedia of Philosophy
edited by Edward N. Zalta.
Martin, M. G. F. . 2002. 'The Transparency of Experience', Mind & Language, 17: 376–425.
Moss, Sarah. 2018. 'Moral Encroachment', Proceedings of the Aristotelian Society, 118: 177-205.
Neander, Karen. 2004/2018. "Teleological Theories of Mental Content." In The Stanford Encyclopedia of
Philosophy, edited by Edward N. Zalta.
Nes, A. 2012. 'Thematic Unity and the Phenomenology of Thinking', Philosophical Quarterly 62: 84-105.
Nimtz, Christian. 2010. 'Saving the Doxastic Account of Intuitions', Philosophical Psychology, 23: 357-75.
Noë, Alva. 2004. Action in Perception (The MIT Press).
Olsson, Erik J. 2005. Against Coherence: Truth, Probability, and Justification (Clarendon Press).
Parsons, Charles. 1995. 'Platonism and Mathematical Intuition in Kurt Gödel’s Thought', The Bulletin of
Symbolic Logic, 1: 44-74.
Pautz, Adam. 2010. 'Why Explain Visual Experience in Terms of Content.' in Bence Nanay (ed.),
Perceiving the World (Oxford University Press).
———. 2013. 'Does Phenomenology Ground Mental Content.' in Uriah Kriegel (ed.), Phenomenal
Intentionality (Oxford University Press).
Peacocke, Christopher. 1986. 'Analogue Content', Proceedings of the Aristotelian Society, 60: 1-17.
———. 1992. 'Scenarios, Concepts and Perception.' in Tim Crane (ed.), The Contents of Experience
(Cambridge University Press).
Peacocke, Christopher 2001. 'Does Perception Have a Nonconceptual Content', The Journal of
Philosophy: 239-64.
Perrett, Patrick. 2015. 'Children’s Inductive Reasoning: Developmental and Educational Perspectives',
Journal of Cognitive Education and Psychology, 14: 389-408.
Pettit, Philip. 1997. Republicanism: A Theory of Freedom and Government (Clarendon Press).
Pitcher, George. 1971. A Theory of Perception (Princeton University Press).
Pitt, David. 2004. 'The Phenomenology of Cognition Or What Is It Like to Think That P?', Philosophical
and Phenomenological Research, 69: 1-36.
———. 2011. 'Introspection, Phenomenality and the Availability of Intentional Content.' in T Bayne and
M Montague (eds.), Cognitive Phenomenology (Oxford University Press).
Plantinga, Alvin. 1993. Warrant and Proper Function (Oxford University Press).

139
Plato. 2005. "Protagoras and Meno." In.: Penguin.
Pollock, John. 1974. Knowledge and Justification (Princeton University Press).
———. 1987. 'Defeasible Reasoning', Cognitive Science, 11: 481-518.
Popper, Karl R. 2001. 'Facts, standards, and truth: a further criticism of relativism.' in Paul Moser and
Thomas Carson (eds.), Moral Relativism: A Reader (Oxford University Press).
Priest, Graham, Berto Francesco, and Zach Weber. 1998/2018. "Dialetheism." In The Stanford
Encyclopedia of Philosophy edited by Edward N. Zalta.
Pritchard, Duncan. 2006/2014. What is this Thing called Knowledge? (Routledge).
Pryor, James. 2000. 'The Skeptic and the Dogmatist', Noûs, 34: 517-49.
———. 2004. 'What’s Wrong with Moore’s Argument', Philosophical Issues, 2004: 349-78.
———. 2005/2013. 'There is Immediate Justification.' in Matthias Steup and Ernest Sosa (eds.),
Contemporary Debates in Epistemology (Blackwell).
———. 2013. 'Problems for Credulism.' in Chris Tucker (ed.), Seemings and Justification: New Essays on
Dogmatism and Phenomenal Conservatism (Oxford University Press).
Pust, Joel. 2000. Intuition as Evidence (Garland Publishing).
———. 2001. 'Against Explanationist Skepticism Regarding Philosophical Intuitions', Philosophical
Studies, 106: 227-58.
———. 2012/2019. "Intuition." In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta.
Sainsbury, R. M. . 1987/2003. Paradoxes (Cambridge University Press).
Sanders, M. D., Elizabeth K. Warrington, John Marshall, and L. Wieskrantz. 1974. '"Blindsight": Vision in a
Field Defect', The Lancet, 303: 707-8.
Schellenberg, Susanna. 2007. 'Action and Self-Location in Perception', Mind, 116: 603-31.
———. 2011. 'Perceptual Content Defended', Noûs, 45: 714-50.
Siegel, Susanna. 2005/2010. "The Contents of Perception." In The Stanford Encyclopedia of Philosophy,
edited by Edward N. Zalta.
———. 2010. The Contents of Visual Experience (Oxford University Press).
———. 2012. 'Cognitive Penetrability and Perceptual Justification', Noûs, 46: 201-22.
Siewert, Charles P. 2011. 'Phenomenal Though.' in T Bayne and M Montague (eds.), Cognitive
Phenomenology (Oxford University Press).
Siewert, Charles P. . 1998. The Significance of Consciousness (Princeton University Press).
Silins, Nicholas. 2007. 'Basic Justification and the Moorean Response to the Skeptic.' in Tamar Szabó
Gendler and John Hawthorne (eds.), Oxford Studies in Epistemology (Oxford University Press).
Silva, Paul. 2013. 'Epistemically self-defeating arguments and skepticism about intuition', Philosophical
Studies, 164: 579-89.
Singer, Peter. 2005. 'Ethics and Intuitions', The Journal of Ethics, 9: 331-52.
Skene, Matthew. 2013. 'Seemings and the Possibility of Epistemic Justification', Philosophical Studies,,
163: 539-59.
Smith, Tara. 2000. Viable Virtues (Rowman & Littlefield).
Smithies, Declan. 2006. 'Rationality and the Subject's Point of View', New York University
———. 2011a. 'Attention Is Rational-Access Consciousness.' in Christopher Mole, Declan Smithies and
Wayne Wu (eds.), Attention: Philosophical and Psychological Essays (Oxford University Press).
———. 2011b. 'What Is the Role of Consciousness in Demonstrative Thought?', Journal of Philosophy,
108: 5-34.
———. 2014. 'The Phenomenal Basis of Epistemic Justification.' in M Sprevak and J Kallestrup (eds.),
New Waves in Philosophy of Mind (Palgrave Macmillan).
———. 2015. 'Why Justification Matters.' in David Henderson and John Greco (ed.), Epistemic
Evaluation: Point and Purpose in Epistemology (Oxford University Press).
———. 2016. 'Perception and the External World', Philosophical Studies, 173: 1119-45.

140
———. 2019. The Epistemic Role of Consciousness (Oxford University Press).
Sober, Elliot. 1982. 'Why Logically Equivalent Predicates May Pick out Different Properties', American
Philosophical Quarterly, 19: 183-89.
Sosa, Ernest. 1996. 'Rational Intuition: Bealer on Its Nature and Epistemic Status', Philosophical Studies,
81: 151-62.
———. 1998. 'Minimal Intuition.' in Michael DePaul and William Ramsey (eds.), Rethinking Intuition: The
Psychology of Intuition and Its Role in Philosophical Inquiry (Rowman & Littlefield).
———. 2006. 'Intuitions and Truth.' in Patrick Greenough and Michael P. Lynch (eds.), Truth and Realism
(Oxford University Press).
———. 2007a. 'Intuitions: Their Nature and Epistemic Efficacy', Gtazer Philosophische Studien, 74: 51-
67.
———. 2007b. A Virtue Epistemology: Apt Belief and Reflective Knowledge (Oxford University Press).
Stalnaker, Robert. 1978. 'Assertion', Syntax and Semantics, 9: 315-32.
———. 1984. Inquiry (MIT Press).
———. 1998. 'What Might Nonconceptual Content Be?', Philosophical Issues, 9: 339-52.
Stanley, Jason. 2005. Knowledge and Practical Interests (Oxford University Press).
Sterelny, Kim. 2003. Thought in a Hostile World (Blackwell).
Stevenson, Charles Leslie. 1944. Ethics and Language (Oxford University Press).
Strawson, G. 2010. Mental Reality (The MIT Press).
Street, Sharon. 2006. 'A Darwinian Dilemma for Realist Theories of Value', Philosophical Studies, 127:
109-66.
Tang, Weng Hong. 2009. 'Belief and Credence', The Australian National University.
Tolhurst, William. 1998. 'Seemings', American Philosophical Quarterly, 35: 293-302.
Tucker, Chris. 2010. 'Why Open-Minded People Should Endorse Dogmatism', Philosophical Perspectives,
24: 529-45.
Turri, John. 2010. 'On the Relationship between Propositional and Doxastic Justification', Philosophy and
Phenomenological Research, 80: 312-26.
Tversky, Amos, and Daniel Kahneman. 1983. 'Extensional versus intuitive reasoning: The conjunction
fallacy in probability judgment', Psychological Review, 90: 293-315.
Tye, Michael. 1995. Ten Problems of Consciousness (The MIT Press).
van Inwagen, Peter. 1997. 'Materialism and the Psychological-Continuity Account
of Personal Identity', Noûs, 31: 305-19.
Wasserstein, Ronald L., and Nicole A. Lazar. 2016. 'The ASA's Statement on p-Values: Context, Process,
and Purpose', The American Statistician, 70: 129-33.
Weinberg, Jonathan M. 2007. 'How to Challenge Intuitions Empirically Without Risking Skepticism',
Midwest Studies in Philosophy, 31: 318-43.
White, Roger. 2006. 'Problems for Dogmatism', Philosophical Studies, 131: 525-57.
Williamson, Timothy. 2000. Knowledge and its Limits (Oxford University Press).
———. 2004. 'Philosophical 'Intuitions' and Scepticism about Judgement', Dialectica, 58: 109-53.
———. 2007. The Philosophy of Philosophy (Blackwell ).
Wright, Crispin. 1985. 'Facts and Certainty', Proceedings of the British Academy, 71: 429-72.
———. 2000. 'Cogency and Question-Begging: Some Reflections on McKinsey’s Paradox and Putnam’s
Proof', Philosophical Issues, 10: 140-63.
———. 2002. '(Anti-)Sceptics Simple and Subtle: G. E. Moore and John McDowell', Philosophy and
Phenomenological Research, 65: 330-48.
———. 2004. 'Warrant for nothing (and foundations for free)?', Aristotelian Society Supplementary
Volume, 78: 167-212.

141
———. 2007. 'The Perils of Dogmatism.' in Susana Nuccetelli and Gary Seay (eds.), Themes from G. E.
Moore: New Essays in Epistemology and Ethics (Oxford University Press).

142