_
M-,

\xéo—y 3". . fl..-
. T

A‘ "‘ 1. -v—.»
'-—‘-4—- - .. <.n ,

 

{7. :0-

u...
. .115?

'-

‘H -‘Lwn-. -«v.uu an - fi'...’ rd...» '1 a,

“- <1-—
7“ -

 

.K

     
  

 

 
  

 
  
  

  
  
 

     

     
   
  
 
 

               

     
 
 
  
 
  
  
  
 

 

      

       
     
 
     

   
    
  

 

   
 
  
  
  
     
  
 
 
  
  

         

   
     
   
   
 
 
 
    

  
  

 

    

 
 
   

' " . -.':. .9 35,14 ..-- m
.r 33,3:3‘331‘3'33 third“: I 3 31:3? {“1" {I‘m}, IL“? IIqII-‘Ii: 4:; 51;” ":15. I: I 'I‘E-I!":i
;.III‘3§:"{ 3‘? VIII; IN; I”; I’IIi I, ’3“: ("In III pm". 3] . 3! 11"”; [V ' '16. Y: ‘3.
"pl, ‘3 . LII 1 u‘ ~ - '" fw- . 33:3" I “ lids ‘" '3 ‘ ' ' "
3333333‘33 ”3““ ’ “'-3333 s."~3f"3 s-fn“aflwnaasz' 3?‘
' ." ‘~ " 933'? 3333-5 I» 7.3: 3,335}? .135‘ ;' W .1" -
I: ”I" . L333: 333‘, . {-I .h: “Lift? ‘3‘ .‘ {H'I'J‘i‘ 3‘!
'L. :33? In" W'- ,3" -‘: "3:33:15 3 - ‘ w ’
11:33:33‘31.l‘;":"V-'-I "3‘33" " 2:333 ”‘3“ at“:
3.3} .I - , {'33 ”III Him} ‘szlgii -, '3’.
I' - , 9 m}: 'h‘ - '342- . “I" val
fm .N‘mh hm» 'm, ,m%
' {'33 ’3: MI}: “If!” m III .‘If'. ;H:‘ ' d}: “333:4
. it": I I, . 3:. .

     
 
  
 
 
 

  

  
    
 
 

‘1 "W939.
‘ "I 3 2t - W 2“ 0 :q;
‘13.!!!”33‘hmwmt‘ NI!» 3 1'55?!
_ 1. h "u. .:..;‘a?
u"; II 533433“; 33"“ ‘1 .2 355;:53;
.f. .‘ :I .I ,.I .. 3‘ , .2 \.§,« up;
. -, :1 .3 J“ 3 .«
1-. :5 -' t ' .3,
II: 5'. I. ’ I .
. ,. N

    

7'

  

E”

..._ ,- 517' ‘5
, I

15‘?
.. .~.. .2;
"a ,..
2 ' 97,91
“a:

“:3:-

..-J.. -3...
v
--0~—.—~

  
  
   

w

      

w... u. g..—

    
     
    
    
   

   
  
  
        

  
  
 
   

  

 

   
  

 

   
 
   
   
 
 
 

~ 33"9'1; ‘1'3 .
33.233!“ ‘3' :3 ‘35,}
WW# WI.
3,33,: ,3. 3333-33 4* E
figfifl kit." vi! :ggr . .:I I: 3‘ .' '. .I
3% i“w WM 'WK.«
3-! -; {4,5 . 5‘3? ' ' '3 435333133.“
333" ' I: ,gt:!§a,I,5.; II. . v

,1

1., ,4:
..
9; 'L
.3:
z.
b

H...
_J
43.
...
..
2‘.
.
.v.“‘
_t';
-..
-.....-..
‘33.. -w

  
 
   

' ' n‘- v I w

2 ~ 91"- 2 ‘ 3‘ l.
fi3fifiwfifi
Iaifiigo I" :I 3.1:” '9'}?-
. 3:1:.-L‘lu.1'tt:,x,1 9;
33"}! "’ '71:! :y: :n I I .
szfi‘i‘g‘mwffiaifi
II?“;~.I~.,:I-.~.;3 q.-..«;;-,
mfififlm$3
wwwfiwflm
fg'fiaékziifiiili '1 ‘ LII! I.
“£333; ‘33:: . ‘ "u‘

9’ t9.

_..4
an
.

v‘

"_'I§'?n7:~ ,
I
3?;

        
  

  

"f:

    
 

u}:
‘¥.":.

.uli-d- o

_.--.

.-.3-.

  

an“

  
   

    

-.. . .
. ..4

.1- -E
1‘.--

_-.~
¢<

  
  

J Lo-..

    
 
 
 

 
    

 
 
  

I,»-
’x' i, :43“

It’v
'21:} ‘

--..‘, ,

» u-..'..‘-'. .:

.-.»....

   

 

.-.‘..<

 

   
  
   

   

4 u.
. «--. -

 

 

     
 
  
  

mtf‘r
I?" '3 3:.3

  
 

. (In I

‘3“ .“ l

. . 3, {I

.5" I .II 33:..3.‘ 1"- LINK. 314‘;
’ APP:

    
  
  
 
 

   
  
 
 
  

  
 

 
       
     

W e .15 -3w-3 ““5
1‘3 ' ‘.- - I - M333“ 3‘31334.‘
IIIIIIIII . - 3" E: 33‘ ”3,3313% I‘IJI if V; 331%.”! 3 I I II IIIIII
fl WW$wflwWfifi $33 flux '53?
3' 3-waws-Ium‘ ‘ ,; 33w
. 3333333333 ..€fi3 @313

3333,33 2333*: m3: : 3 33“.?

1M3 h 33. mmflJéfifi

 
 

THESE

MIC CH IGAN STATE

ll!!! lllll IllIllIIHJHHHIHllllllllll

1293 01571 1538

ll

 

 

 

 

 

 

 

 

 

 

 

 

This is to certify that the

dissertation entitled

Doxastic Voluntarism and Epistemic Responsibility:
The Ethics of Belief Revisited

presented by

Charles R. Johnson

has been accepted towards fulfillment
of the requirements for

Ph . D . degree in PhilosophL

 

 

MW /_/-(/

M a Jg’professor

 

Date April 21, 1997

MS U it an Affirmative Action/Equal Opportunity Institution 0-12771

 

LIBRARY

Mlchlgan State
Unlverelty

 

 

 

PLACE IN RETURN BOX to remove We checkout from your record.

TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

ll

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative ActIorVEquel Opportunity lnetltutlon
Wane-m

DOXASTIC VOLUNTARISM AND EPISTEMIC RESPONSIBILITY:
THE ETHICS OF BELIEF REVISITED

By

Charles Roderick Johnson

A DISSERTATION

Submitted to
Michigan State University
in partial fiilfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Philosophy

1997

ABSTRACT

DOXASTIC VOLUNTARISM AND EPISTEMIC RESPONSIBILITY:
THE ETHICS OF BELIEF REVISITED

By

Charles Roderick Johnson

At least since the time of the ancient Greeks the question of whether human beings
have freewill has been debated in both religious and philosophical circles. During this time,
though, it has been thought that this is primarily a debate about the possibility of moral or
perhaps legal responsibility. That this debate might concern other “types” of responsibility,
for instance epistemic responsibility, has not been thoroughly considered. In a recent article
on this subject it has been lamented that the “ideals of free will and free thought are not
usually connected” (see Pettit and Smith: “Freedom in Belief and Desire”). In fact, many
theorists concerned with the idea of epistemic responsibility have argued that free thought or
doxastic voluntarism (the thesis that belief is under voluntary control) is conceptually
impossible and phenomenologically improbable. In my dissertation I challenge this separation
of free thought and {ice will by employing a conceptual argument against doxastic non-
voluntarism (the thesis that thought or belief is not free), an argument that was originally
employed by Immanuel Kant as a proof for fiee moral agency. Hence I employ an argument
that has been primarily relegated to the moral sphere, i.e., the sphere of freewill, in order to
attempt to prove the conceptual necessity of freedom in the epistemic sphere, i.e., the sphere
of free thought. In this way I hope to show the parallels between ethics and epistemology in

terms of the positions that are maintained, and the commitments or assumptions that underlie

them. This I believe will help to inform nonnative theory or inquiry in general by showing
that the two major foci of this inquiry, epistemology and ethics, conduct themselves along the

same basic lines.

T0 Lauren,
my reason to believe.

iv

ACKNOWLEDGMENTS

I would like to thank those members of the Michigan State Philosophy Department
who assisted me throughout the writing of this dissertation. These persons are Professors
Richard Peterson, Nicholas Smith, Charles McCracken, and Richard Hall. I am especially
grateful to Professor Hall who served as my dissertation chair. “Without his patience,

guidance, and support, this project would never have been completed.

TABLE OF CONTENTS

INTRODUCTION ........................................ 1
Epistemic Responsibilities
What is Doxastic Voluntarism?
The Different Types of Voluntarist Positions

CHAPTER ONE ......................................... 13
HISTORY

The Stoics

The Skeptics

Descartes

Locke

Clifford

James

Conclusion

CHAPTER TWO ........................................ 47
NON-VOLUNTARI SM

The Inconceivability of Non-Voluntarism

A Defense of Non-Voluntarism

Conclusion

CHAPTER THREE ....................................... 67
INDIRECT VOLUNTARI SM

The Basis for Indirect Voluntarism

The Question of Viability

Conclusion

CHAPTER FOUR ........................................ 81
DIRECT VOLUNTARI SM

The Definitional Argument

The Conceptual Argument

The Phenomenological Argument

Direct Voluntarism and the Norms of Rationality

Conclusion

CHAPTER FIVE ........................................ 100
DIRECT VOLUNTARISM

Naturalism

Teleology and Deontology: Virtue, Consequences, and Duty

Normative Goals and Conduct

Deterrninism and Responsibility

LIST OF REFERENCES ................................. 121

vii

INTRODUCTION

We evaluate people's beliefs all the time. Teachers evaluate the cogency and
reasonableness of their students’ beliefs. Judges and juries evaluate the beliefs of the accused
(e.g., in negligence cases where it is determined whether or not the defendant's beliefs were
reasonable). This scrutiny can even be self-imposed, such as when judges and jurors are
asked to apply the standard of reasonable doubt to their own beliefs. This evaluative process
suggests that persons can be responsible for the beliefs that they hold. This in turn
presupposes that persons have some control over their beliefs; for, following Kant‘s dictum
that “ought” implies “can,” persons cannot be responsible for something that falls outside
their power.1

In many cases, though, it appears that persons do not have power over their beliefs.
Many argue that perceptual beliefs, in particular, are involuntary -- that “seeing is believing.”
Similarly, early childhood beliefs may not be subject to self-control. Wittgenstein in On

Certainty writes, “The child learns by believing the adult. Doubt comes afier belief.”2 Lastly,

 

' John Heil, Bernard Williams, H.H. Price, and Louis Pojman, among others, hold this
view that in our ordinary talk about beliefs we often attribute responsibility to the believer
and that this attribution implies beliefs are subject to conscious control.

2 “Wittgenstein is certainly not alone in holding this position. Locke in the Essay asserts
that “There is nothing more ordinary than that children should receive into their minds
propositions (especially about matters of religion) from their parents, nurses, or those
about them” (Locke: 43 7). These propositions he argues insinuate themselves into the

1

2

many argue that evidence can compel belief, that the evidence for some propositions is so
persuasive that they simply cannot be doubted.

Given this view of belief is the kind of freedom requisite for epistemic responsibility
possible? This is the question I will address in this dissertation. I will examine what is termed
in the literature “doxastic voluntarism,” the thesis that beliefs are subject to our voluntary
control. I will examine this thesis with an eye to epistemic obligations or responsibilities,
obligations or responsibilities concerning what we should believe and how we should come
to believe. The concept of rationality is intimately connected to this notion of epistemic
responsibility. Rational believers are, at least in part, responsible believers, believers who
fulfill their epistemic responsrbilities. During this examination, then, I will invoke the concept
of rational agency as a basis for discussing this issue. As such, the central question of this
dissertation becomes: relative to our conception of ourselves as rational agents is the kind of
fi’eedom requisite for epistemic obligations possible, and if so, what must be the extent of this
fi'eedom?

In addressing this question the dissertation is divided into five chapters. In the first
chapter, I explore the history of doxastic voluntarism by examining a specific evolution of
thought on this topic. In the second chapter, I discuss non-voluntarism, and use a Kantian-
style argument to show that voluntarism cannot be denied in that the attempt to do so is self-
contradictory. In the next two chapters, I examine the remaining alternatives to non-

voluntarisrn - indirect and direct doxastic voluntarism. In these chapters I argue that direct

 

“unwary as well as unbiased understandings” of children (Locke: 437). In the Principles,
William James likewise adheres to the position that “we begin by believing everything”
(Principles of Psychology Vol. II hereafter PP: 319).

3

voluntarism is the more viable of the two alternatives. In the fifth chapter, a substantial
appendix, I examine parallels between ethics and epistemology which illuminate certain issues
and positions within these two disciplines. The remainder of this introduction will be devoted

to laying the groundwork for the discussion that is to follow.

I. Epistemic Responsibilities

In this dissertation I work from the assumption that there are epistemic obligations,
duties, or responsibilities. This assumption is certainly commonplace enough, at least in the
history of western philosophy. Descartes clearly thought there were such duties or
responsibilities. In the Meditations, he discusses how we may set about “behaving correctly
and avoiding error” through carefully regulating belief (Descartes: 103). Locke, in the Essay,
discusses the idea that a rational mind or person has an obligation to inquire. He states that
“the mind, if it will proceed rationally, ought to examine all the grounds of probability”
(Locke: 406). James, of course, is well known and cited for his formulation of our two “first
and great commandments as would-be knowers” (“Will to Believe” hereafter WTB: 508).
Lastly, in more recent years epistemologists such as Keith Lehrer and Roderick Chisholm
have formulated accounts of our epistemic responsibilities, or what is worthy of our belief
(see Chisholm’s Perceiving and Lehrer’s Theory of Knowledge). I wish to work from within
this tradition, then, and assume that we are confronted with certain duties, obligations, or
responsibilities cont the formation of belief. This does not mean, though, that my work
is done. I have instead assigned to myself the task of trying to make sense of these

responsibilities. I examine whether it is conceptually plausible to assert that we have the kind

4

of doxastic control requisite for epistemic responsibility.

I am also assuming that these epistemic responsibilities are relevantly similar enough
to moral or ethical responsibilities that our intuitions concerning the latter may be used to
shed light on the nature of the former. In Chapter Five, I discuss in more detail this issue of
the analogies and disanalogies between ethics and epistemology, but here let it suflice for me
to say that I hold these domains to be using the notion of obligation or responsibility in the
same way, though, ultimately these domains orient this notion in terms of different ends or
goals. Thus, in epistemology the end or goal toward which our epistemic responsibilities aim
is truth, whereas in ethics it is goodness or rightness. Given, then, that both domains make
use of this notion of responsibility in the same way, comparisons between these two domains

become possible.

I]. What is Doxastic Voluntarism?

Doxastic voluntarism would seem to be, quite simply, the thesis that we have some
amount of fi'eedom or liberty in regards to our beliefs or belief states, i.e., that our beliefs are
under our free or voluntary control. But this definition is incomplete, for we now need to
know what we mean by the notion of freedom that we are employing. How we define this
notion will in turn structure our entire outlook as to whether doxastic voluntarism is a tenable
thesis. Throughout this discussion of freedom, I assume that belief is an action relevantly
similar to physical actions, so that the views of freedom used in discussions of the latter can
be correctly applied to the former. It may be the case that these species of “actions” are not

relevamly similar, and that therefore the language of physical actions cannot be applied to

5

belief. This question, though, is beyond the scope of this dissertation.3

There are a number of rather standard or stock definitions of freedom. Let us call the
first definition that we will look at the Desire Thesis. This position holds that we are free
when we can do what we want. On this account, a belief would be fi'ee in so far as it was the
product of the believer's wants or desires. Another standard definition of freedom, which
opposes that of the Desire Thesis, is what we will term the Counterfactual Thesis. This thesis
argues that one has acted freely only in so far as one could have done otherwise. Thus
freedom on this view rests in the idea that one has the power or capacity to act in an
alternative manner. According to this view, then, a given belief would be free or voluntary
only ifit were possible for the believer not to form the belief.

Proponents of the Desire Thesis often try to make their view consistent with that of
the Counterfactual Thesis. The former attempt to do this by arguing that persons in their
view firlfill the requirement of being able to do otherwise. Desire theorists argue that any
person could have done otherwise if the desires of the person had been different. This,
though, fails to meet the intention of the Counterfactual Thesis which is that the agent be able

to do otherwise everything being the same. A person is free, in the Counterfactual view if

 

3 It may not be essential to this thesis, though, that this assumption be true, i.e., that belief
is an act relevantly similar to physical acts. Both John Heil (1983) and H.H. Price argue
that belief is not an action, and yet both conclude that epistemic agents can have
responsibilities or duties in terms of belief. Thus, though they argue that belief is not an
act, they do not from this conclude that we have no control over belief, and hence that it
cannot be the proper subject of responsibility. The control or freedom that they attribute
to belief is indirect. On their respective views, we control belief through engaging in
certain “procedures” or “methods”, and it is for our engagement in these methods that

we are ultimately responsible. For more on this issue of belief and action see Price: 15,
Heil (1983): 358, and Montmarquet: 79-82.

6

in the exact same situation, desires and all else being identical, the person could have done
otherwise. The Desire Thesis carmot meet this requirement. In this latter view, if a person’s
desires remain the same (in conjunction with the surrounding circumstances), then so will
their actions.

The view of freedom put forward by the Counterfactual Thesis is surely inadequate,
however. It does not fit our intuitive notions of freedom and responsibility. Consider the
following example: Jones is trying to decide whether or not to rob a convenience store.
Unbeknownst to Jones, an Evil Genius has placed a small computer chip in his head. This
meclmnism will allow the Evil Genius to make Jones rob the store if Jones opts not to. Jones
decides to rob the store and acts accordingly. Should we in this case claim that Jones was not
free in his action, and hence not responsible? In this case, he could not have done otherwise.
Given the Evil Genius's mechanism, there were no alternative actions open to Jones; if he had
chosen otherwise the Genius would have made him rob the store. And yet, doesn't it seem
that because Jones decided to rob the store, he was in fact free and responsible, even though
he could not have done otherwise (I am indebted to Harry G. Frankfurt's “Alternative
Possibilities and Moral Responsibility” for this exarnple).‘

The Desire Thesis definition of freedom could account for this example. Jones's

 

‘ If one finds this example too outlandish, simply imagine that the Evil Genius will control
Jones’ behavior not through a computer chip, but rather through his henchmen who will
physically compel Jones to rob the store if he opts not to. I do not think that this
modification will alter the conclusion of the example, which is simply that freedom does
not require alternative possibilities in the sense outlined by the Counterfactual Thesis.
Dennett concurs with this latter point. In Elbow Room he argues that the question of
whether a person could have done otherwise is not our central concern in situations where
we are assessing responsibility (Dennett 1984: 135-136). I am indebted to Prof. Richard
Peterson for pointing out this possible difiiculty.

7

action is free because he did what he wanted, and therefore he is responsible. There is
another definition of fi'eedom, though, that could equally well account for this example. Let
us call this alternative the Will or Choice Thesis. According to this view, an act is free if it
is the product of the will of the actor. It is free if the agent brings it about through his will,
self, or faculty of choice. Acts would not be free on this account if they were the result of
some kind of external cause or compulsion -- if the self or will of the agent was not the sole
author of the action.5 This view is able to explain our intuitions concerning the above case
rather nicely. Jones is responsible because he decided to rob the store. This act was the
result of his own will, and as such, it was fiee.

The Will Thesis is much like the view of CA. Campbell. Though at times he talks as
if he were an adherent of the view that freedom requires the capacity to do otherwise, in his
more careful moments he states that what is actually requisite is that the agent be able to
choose otherwise (Campbell: 393). Jones certainly meets this latter condition. Though he
could not have done otherwise, he certainly could have chosen to do otherwise (such a choice
then resulting in the Evil Genius compelling him to act against his will). Campbell, in
agreement with this alternative view, states that fi'ee acts are those in which the self or will
of the agent is the sole author of the action (Campbell: 393). Jones meets this condition as

well. In choosing to rob the store he is the sole cause or author of his action (though if he

 

5 Sole cause is probably too strong. We could certainly weaken the requirement to that of
sufficient cause, but this may be too strong as well. Issues of causation being too complex
and far afield for this dissertation, let us consider an act on this view to be fine when the
will is at least a substantial cause, a substantial cause being one whose absence would
significantly alter the causal picture of the event to which it is connected (I am indebted

to Prof. Nicholas D. Smith for suggesting this definition)

8

had chosen otherwise another causal mechanism would have brought the action about).

I believe the erl Thesis is superior to the Desire Thesis for two reasons. First, even
though the Desire Thesis does well in terms of the example we discussed, if we alter this
example slightly we can see that this view has a serious shortcoming. What if Jones decided
not to rob the store even though he wanted to? Perhaps he just lacked the courage necessary
to act on his desire. The Evil Genius would then make Jones rob the store. Would Jones be
responsible for this act? I believe that on the Desire Thesis he would be, for he would have
acted as he desired or wanted to. Intuitively, though, I think we would not hold Jones
responsible. We may think him reprehensible for wanting to rob the store, but not responsible
for he did not choose to do so.

Secondly, the Desire Thesis has the difficulty that it does not completely agree with
our phenomenology of action. As Campbell indicates, we often experience the effort of our
will as it conflicts with desire (Campbell: 395). In such situations we frequently choose to act
contrary to our desires. We none the less view these actions as free (in fact, Campbell views
only such acts as free). According to the Desire Thesis, these acts would not be free. This
is because free acts on this view are those done in accordance with our wants or desires. It
could, of course, be argued that we only act contrary to our desires when we have a stronger
or higher order, want, or desire that we wish to firlfill. As such, any conflict with desire is
only superficial. If one probes deeper, a stronger desire can be found to explain the action.
But I find this response unsatisfactory. It would seem that in our ordinary experience we
often fi'eely choose to do things that we do not want to do. I do not agree with Campbell,

though, that freedom always requires us to act contrary to our desires. I think that we can

also freely choose to do what we want.

A belief will be free, or voluntary, according to the Will Thesis in so far as it is
brought about by the will of the believer, in so far as the self or will of the believer is the
substantial author or cause. Also requisite is the notion that the believer could have chosen
to believe otherwise (this condition being firlfilled even when this contrary choice would not
have been successful). Thus it is not requisite that the free believer be able to believe
otherwise, any more than it is requisite that the free agent be able to do otherwise. What is
requisite is that the fiee agent or believer be able to choose otherwise. Just as moral
responsibility is not altered by moral luck -- situations where one's intended action has
unintended morally significant results, epistemic responsibility is not nritigated by epistemic
luck -- situations where an intended belief state does not come about. One may be subject
to epistemic blame for choosing to believe a falsehood, even where this choice is
unsuccessful.

This account of freedom is the one I employ in the examination of doxastic
voluntarism that is to follow. A free or voluntary belief will be one that satisfies the two
following conditions:

1) it is the result of the believer's choosing, deciding, or willing to believe it; and
2) the believer could have willed or chosen otherwise.
The question that will be addressed, then, is whether beliefs can be the product of the will of

the believer, i.e., whether we can suitably choose or will to believe.

10
III. The Different Types of Voluntarist Positions
Control over belief would seem to admit of variation both in terms of the degree of
the controL and the manner in which it is exercised. With this variation there are criteria by
which voluntarist positions can be classified. Voluntarist positions can be distinguished by
whether they assert this control to be direct or indirect, complete or incomplete, and negative

or positive.

a. Direct and Indirect Voluntarism

A direct voluntarist maintains that it is possible to directly control or influence beliefs.
James Montmarquet holds this position. He argues that through the exercise of certain
epistemic virtues, such as conscientiousness, it is possible for one to directly influence one's
belief states (whether Montmarquet is actually successful in showing that his position is one
of direct voluntarism will be discussed in Chapter Three). In his work Epistemic Virtue and
Doxastic Responsibility, Montmarquet states:

the exercise of these qualities [the epistemic virtues] in belief formation or retention (e.g.,
by way of being “carefirl” or “conscientious” in what we believe) is not a mode of
indirect, but a mode of direct influence over belief (Montmarquet: viii).

An indirect voluntarist asserts that it is not possible to influence or control beliefs
directly. Beliefs, according to this view, can only be influenced through certain indirect
means, such as gathering evidence or even hypnosis. One person who holds this position is
John Heil. In his article "Doxastic Agency" he states:

It is not that one has a choice in the beliefs that one forms, but that one has a say in the

procedures one undertakes that lead to their formation. The notion of ‘episternic
responsibility’ attaches to the undertaking of appropriate procedures (Heil, 1983: 363).

11

These positions, though, need not be in complete opposition. Weaker versions of
these positions would maintain only that some beliefs are, or are not, subject to direct control.
Such weaker versions would be compatible. Thus one could maintain that some beliefs are
subject to direct control, whereas others can only be influenced indirectly. On this reading,
one would be a weak direct voluntarist concerning the former beliefs, and a weak indirect
voluntarist concerning the latter. Descartes, in the Discourse on Method, seems to hold a
position like this in regard to simple or self-evident propositions. Although Descartes is
typically seen as asserting that we have some kind of direct control over all of our beliefs, in
the Discourse he seems to argue that in terms of simple propositions all we can do is prepare
the mind properly to receive them. Thus he states:

As for simple propositions, the only rules we provide are those which prepare our
cognitive powers for a more distinct intuition of any given object and for a more

discenring examination of it. For these simple propositions must occur to us
spontaneously; they cannot be sought out (Descartes: 17).

b. Complete and Incomplete Voluntarism

A complete voluntarist argues that the power of our will over belief is absolute, i.e.,
that there is no proposition such that we cannot choose or will to believe it, or not to believe
it. An incomplete voluntarist holds that our control is limited, i.e., that there are certain

propositions or subjects of belief that are outside the range of our will or choice.

12

c. Negative and Positive Voluntarism

A negative voluntarist maintains that one's control over belief consists in one's being
able to withhold it. Though this is not his overall position, Descartes in the Meditations
seems to describe what such a position would be like when he states:

even if it is not in my power to know any truth, I shall at least do what is in my power,
that is, resolutely guard against assenting to any falsehoods (Descartes: 79).

A positive voluntarist, on the other hand, maintains that our power over belief rests in our
actually being able to produce beliefs.

These positions are quite compatible. One could hold limited versions of these two
positions that would not be in conflict. Thus one could maintain that while our principal
control over belief is negative, there are some beliefs that we can actually bring ourselves to
have. One could even hold complete versions of these positions simultaneously and argue
that on all occasions we have the power either to believe or suspend judgment at will.

Overall, these three distinctions concerning voluntarism can overlap. Thus one could
be a complete, negative, direct voluntarist. A person holding this position would maintain
that we have an absolute or unlimited power to withhold belief, and that this power is directly
exercised (that it is not brought about by indirect means such as avoiding evidence, or

avoiding the consideration of certain propositions).

CHAPTER ONE

HISTORY

In this chapter I seek a better understanding of voluntarism by examining a particular
progression of thought on this subject, a progression which involves many of the most
prominent thinkers on this subject. I begin with the voluntarist positions of the Stoics and
Skeptics. I then go on to the position of Descartes, who was influenced by these earlier
thinkers. Next, I discuss the view of John Locke, since so much of his work in The Essay
Concerning Human Understanding was a response to Descartes. Lastly, I examine the
positions of W.K. Clifford and William James. Clifford puts forward a Cartesian account of

the proper role of will in belief, and James (most famously) replies to it.

I. The Stoics

To these perceptions accepted by the senses
Zeno joins the assent of the mind, which he
considers to be placed in us and voluntary.

Cicero: SVF I, 61
T o the mind all is indrflerent except its own
activities, and all its own activities are

within its control.

Marcus Aurelius: Meditations BK 6, 32

13

14

The Stoics, like Plato and Aristotle before them, were concerned with the question
of the good life - namely, what living well consists of or what is the appropriate life for
human beings. Though there was some disagreement about the answer to this question, the
Stoics believed the general goal of life to be living in accordance, harmony, or agreement,
with nature. This was sometimes called the smooth or virtuous life (Saunders: 113). Thus
according to Diogenes, Zeno held that “the goal was to live in agreement with nature”
(Gerson: 136). Likewise, Cicero states that:

since the goal is to live consistently and in agreement with nature, it follows necessarily
that all wise men always live happy, perfect, and fortunate lives, that they are impeded by
nothing, hindered by nothing and in need of nothing (Gerson: 150).
Living in accordance with nature involves a submission to the universal rule of fate or natural
events. Epictetus advises us to “Ask not that events should happen as you will, but let your
will be that events should happen as they do” (Saunders: 135). Such a submission to, or
acceptance of, natural events he argues will bring about peace or happiness. It will result in
the smooth or virtuous life, life in agreement with nature (Saunders: 135, 113).

The Stoics often used analogies in order to help explain what the virtuous life was
like. These analogies usually appealed to situations or circumstances where an agent is
subject to forces outside of his control. Thus Zeno speaks of the dog and the wagon:

Suppose a dog to be tied to a wagon. If he wishes to follow, the wagon pulls him and he
follows, so that his power and necessity unite. But if he does not wish to follow, he will
be compelled to anyhow. The same is the case with mankind also. Even if they do not
mg; to follow, they will be absolutely forced to enter into the fated event (Saunders: 109-

Marcus Aurelius speaks of the pig at sacrifice and the runaway slave:

He who flees from his master is a runaway slave. The law is our master and he who

15

transgresses against it is then such a runaway too. The man who is grieved or angry or
afraid wishes that something had not happened, were not happening, or would not happen,
of those things which were ordered by that which rules over all, that is, the law which
gives to each man what befalls him (Grube: 104).
Picture to yourself every man who is pained or dissatisfied with anything as being like a
pig kicking and squealing when sacrificed; and so is the man who laments silently, alone
in his bed, that we are bound by fate (Grube: 105).

Finally, Epictetus, like Shakespeare, speaks of life as being a stage:
Remember that you are an actor in a play, and that the Playwright chooses the manner of
it: if he wants it short, it is short; if long, it is long. If he wants you to act a poor man you
must act the part with all your powers. . . For your business is to act the character that
is given you and act it well; the choice of the cast is Another's (Saunders: 137).

The Stoic view, then, seems quite fatalistic, and yet within fate or nature there lies
room for human freedom through our capacity for judgment. Chrysippus tries to explain this
by likening us to a cylinder that is knocked down a hill. The initial blow that sets the cylinder
moving is like the hand of fate or nature. It is outside the cylinder's control. It is an
irresistible impulse that sets it in motion. But, once the cylinder is moving, it moves in
accordance with its own nature. Its movement is no longer subject to fate (Saunders: 109,
Gerson: 133). Likewise, we as human beings are subject to the impulses or irnpetuses of
nature which come in the form of sense impressions or presentations. Our fi'eedom lies in our
capacity to judge these impressions. This capacity or faculty of judgment, assent, or belief,
lies within our power. Assent may require external stimulation, but this faculty itself is still
within our control. As Cicero states, recounting the argument of Chrysippus:

an object strikes our sense and as it were stamps its image in the soul, but the assent is in
our power, which, as has been said in the case of the cylinder, while put in motion from

without, moves for the rest by its own force and nature (Saunders: 109).

The basic idea, then, is that life, nature, or fate, presents us with certain impressions

16
or irnpetuses for judgment. These are outside our control. What lies within our power is our
judgment, assent, or belief, concerning them. For example, we cannot help the fact that we
are being bumed or hurt in some manner, but we can help our belief or judgment concerning
this situation. Marcus Aurelius states:
Whether a thing is bad for you does not depend upon another man's directing mind, nor
upon any turn or change in your environment. Upon what then? Upon that part of you
which judges what is bad. Let it make no such judgment and all is well. Even when that
which is closest to it, your body, is cut, bumt, suppurating or festering, let the judging part
of you keep calm. (Grube: 33)
To live in harmony with nature, then, is to accept or submit to the will of nature or
fate. Marcus Aurelius tells us that this capacity to submit willingly distinguishes us as rational

creatures. He states:

Realize too that it is granted only to the rational creature to submit willingly to the course
of events; merely to submit is inevitable for all creatures. (Grube: 105)

Living in harmony also requires good or true judgment. Disharmony and unhappiness, an
unsmooth life, results from making false judgments. Thus Epictetus tells us that:

‘Ifhe threatens me with death,’one says, ‘he compels me.’ No, it is not what he threatens
you with which compels you, but your decision that it is better to do what you are hidden
than to die. Once again it is your own judgement which compels you - that is, will puts
pressure on will. . . if you will, you are free: if you will, you will blame no one, you will
accuse no one: everything shall be in accordance with your own mind and the mind of God
(Oates: 255).

Marcus Aurelius likewise advises us to

Remain then within the limits of your actual perceptions; do not add to them from within
yourself, and you are not afi‘ected. (Grube: 83).l

 

' This admonition is in some ways reminiscent of Descartes’ theory of error in the Rules
for the Direction of our Native Intelligence where he states that “we can go wrong only
when we ourselves in some way compose the objects of our belief” (Descartes: 16).

17

For the Stoics, it is our capacity to control judgment that allows us to avoid error and live in
harmony with nature.

Though there seem to be some exceptions, such as that persuasive propositions lead
to assent or that some presentations come with a yielding or assent (Gerson: 86, 81), the
Stoics on the whole appeared to maintain that our control over belief, judgment, or assent,
is absolute. It is this faculty that serves as our only response to, and exception from, the rule
of fate. But some seemed to believe that this control is only fully realized through the study
of dialectic. Thus some Stoics argued that

Wrthout the study of dialectic the VVrse Man cannot guard himself in argument so as never
to waver, for it enables him to distinguish between truth and falsehood (Saunders: 74).

Dialectic, moreover,
conveys a method that guards us from giving assent to any falsehood or ever being
deceived by specious probability, and enables us to defend the truths that we have learned
(Saunders: 123).
In temrs of our eadier taxonomy, then, the Stoics could be classified as complete, positive and
negative, direct voluntarists. Those who allow the exceptions cited above would be
incomplete voluntarists. Lastly, if one considers the discussion of dialectic to imply that this
discipline is a means or tool for controlling judgment or assent, they could be classified as
indirect voluntarists. I am inclined to believe, however, that the Stoics were on the whole
complete, positive and negative, direct voluntarists. They seemed to maintain that judgment

was completely in our control, and that as such, in the words of Marcus Aurelius, “Everything

is as you think it be, and the thinking is within your control” (Grube: 126).

18
II. The Skeptics
Skepticism is an ability, or mental attitude, which
opposes appearances to judgments in any way whatsoever,
with the result that, owing to the equipollence of
the objects cmd reasons thus opposed, we are
brought firstly to a state of mental suspense and
next to a state of “unperturbedness ” or quietude.

Sextus Empiricus: Outlines of Pyrrhonism BK I, IV, 8

The Skeptics viewed themselves as occupying a quasi-therapeutic role. They saw
themselves as putting forward a cure for the illness of dogmatism. Sextus states that the
Skeptic, “because he loves humanity, wishes to cure dogmatists of their opinions, and
rashness with reasoning, so far as possible” (Gerson: 183). The goal of Skepticism was
peace, quietude, or “unperturbedness.” Dogrnatism or dogmatizing (asserting the truth of
that which is non-evident or non-apparent) counted as an “illness” in relation to this goal,
because the Skeptic believed the concern to assent or judge truly, which accompanies
dogrnatisrn, to be the source of mental distress. Thus Sextus writes of “men of talent” who
were “perturbed by the contradictions in things and in doubt as to which alternatives they
ought to accept” (Bury: 19). The Skeptics sought to attain quietude, or find relief from the
disquiet of dogmatisrn, through complete suspension of judgment. Freedom from anxiety
would follow this suspension, the Skeptics claimed, like a shadow follows a body (Gerson:
189)

This suspension of judgment could be achieved through the opposing of arguments.
Sextus argues that:

The main principle of the sceptical system is that for every argument another argument of

19

equal [weight] is opposed. As a result of this we seem to arrive at a cessation of
dogrnatism (Gerson: 186).

At other places he writes that:
The sceptical ability is the ability to set in opposition appearances and ideas in any manner
whatsoever, the result of which is first that, because of the equal force of the opposed
objects and arguments, final suspension of judgment is achieved, and then freedom from
disturbance (Gerson: 185).
There were various ways or “modes” in which arguments, ideas, or presentations, could be
opposed in order to bring about suspension of judgment. One mode, for example, centers
on the difi‘erences among the senses. Through this mode one could achieve suspension of
judgment on such matters as whether honey is pleasant. To the tongue honey appears
pleasant in some instances, but unpleasant to the eye, and as such “it is impossible to say
whether it is, all on its own, pleasant or unpleasant” (Gerson: 203). Through the opposition
of arguments, or appearances in this case, one could come to suspend judgment regarding
the “true” or “essential” nature of honey. This suspension of judgment is induced or
compelled by the opposing of arguments or modes. Thus Sextus writes in regards to the
seventh mode that “it is evident by this mode too we are compelled to suspend judgment
about the nature of things” (Gerson: 206).

The range of this method of suspending judgment, however, was limited. The
Skeptics argued that it could not extend to appearances which force our assent, or bring us
to assent involuntarily. Sextus writes that Skeptics do not “overthrow the affective sense
impressions which induce our assent involuntarily; and these impressions are ‘the

appearances’ ” (Bury: 21). It is argued that the Skeptic “assents to the states forced on him

by presentation” (Gerson: 186). The Skeptic, then, would be willing to admit that honey

20

aopeared sweet to him; he would suspend judgment though as to whether this was its actual
nature. Hence Timon says, “That honey is sweet I do not posit; that it appears so I concede”
(Gerson: 182).

In that assent to appearances was inevitable, these became the “criterion” for Skeptics
in the sense that they became guides for living. Sextus argues that Skeptics

follow a rational procedure based on appearances, this procedure indicating a life in accord
with traditional customs and laws and practices and private states (Gerson: 187).

Adherence to Skepticism, then, did not mean inaction or even abstaining from the customs
and traditions of the society in which one was situated.

The Skeptics were consistent in their approach. Their goal of suspension of judgment
extended even to themselves or their own assertions. Thus, Sextus warns us not to mistake
the statements of Skeptics as being dogmatic assertions. He tells us that the Skeptic, when
he makes claims, merely states how things appear to himself. The Skeptic is not making a
judgment or assertion concerning how things really are. Sextus maintains that:

the sceptic offers his utterances as implicitly self-canceling, and so he should not be said
to dogmatize in offering them. The most important point of all is that in making these
utterances he expresses that which appears to himself and he reports the state he is in, but
he does so undogrnatically, and he commits himself to nothing about any underlying
external things (Gerson: 187).
Even the Skeptical method itself is not exempt fi'om this claim. It would seem that the
method of opposing arguments involves the implicit assumption that there can be no valid
demonstration, for every argument is said to have an opposing argument of equal weight. Yet

this is not the case with the Skeptic. Sextus argues that the Skeptical method is like fire

which, after destroying its firel, ends by consuming or destroying itself. The Skeptical method

21

is not a demonstration against the validity of all demonstration; it is not given a privileged
position. Sextus argues that “the argument against demonstration is able to wipe itself out
after having destroyed all demonstration” (Gerson: 194). Still trying to make this point clear,
Sextus employs an analogy that Wittgenstein would later use in the T ractatus. Sextus argues
that the Skeptical argument against demonstration is like a ladder that, once it has been used
to reach the height one wants to achieve, is kicked down (Gerson: 194).

There are several illuminating differences between the Stoics and the Skeptics. The
Skeptics considered the Stoics to be dogmatists. Recall that the Stoics asserted such things
as that the goal of life was to live in accordance with nature; the Skeptics would not assert
any such universal goal (Gerson: 192). In addition, the Stoics believed that it was possible
to judge correctly, so long as one did not insert oneself into one's judgment; the Skeptics
believed that such insertion was unavoidable. Thus the fourth or circumstantial mode argues
against the validity of judgments concerning the reality of external objects, on the grounds
that such judgments will always have to occur while the judge is in some disposition, and
hence “if he is going to judge the presentations while in some disposition, then he will be
party to the disagreement” (Gerson: 204).

Moreover, the position of the Skeptics is clear on a number of issues where the Stoic
position is difiicult to discern. There was some doubt, for instance, as to whether the Stoics
believed that at least some presentations or appearances compel assent (whether there were
some exceptions to our control over belief or judgment). The Skeptics address this issue
quite clearly: they argue that appearances do force assent and, as such, our control over

belief; assent, or judgment, does not extend to them. For the Skeptics, we can then control

22

only our beliefs or judgments concerning our appearances. One cannot, for example, control
the belief or judgment that “he is appeared to redly.” One can, however, control any
judgments or beliefs as to the validity or truth of this appearance.

It was also unclear as to what exactly was the role of Dialectic or argument in the
Stoic view of belief formation. Thus it was shown that some Stoics maintained that training
in Dialectic was necessary at least for one to have firll control over belief. The Skeptics are
once again clear on this issue. They argue that one's control over belief, one's capacity to
suspend judgment, arises fi'om the application of the modes (or opposing arguments) to the
matter at hand. Our control over belief, according to their view, is indirect. It is through
employing the method of opposing arguments that we are induced or compelled to suspend
judgment.

Given these differences I classify the Skeptics as incomplete, indirect, negative
voluntarists. They are incomplete voluntarists in that they claim appearances force belief.
They are indirect voluntarists in that they maintain that our control over belief rests in the
employment of the modes or opposing arguments. They do not hold that we can affect belief
or judgment directly, independent of this method. Lastly, they are negative voluntarists in
that their discussion of control always centers on the suspension of judgment. According to
their account, our control over belief consists in our capacity to withhold it through the
method of opposing arguments. This is yet another difference between the Skeptics and the

Stoics, in that the latter maintained that we have some positive control over belief.

23
III. Descartes
I have alreaay amply demonstrated that everything of
which I am clearly aware is true. And even if I had not
demonstrated this, the nature of my mind is such that

I cannot but assent to these things.

Descartes: Fifth Meditation, Sec. 65

The goal of Descartes’ philosophy was to find a secure or firm foundation on which
to ground human knowledge. In the Meditations Descartes writes:
I realized that it was necessary, once in the course of my life, to demolish everything
completely and start again right from the foundations if I wanted to establish anything at
all in the sciences that was likely to last (Descartes: 76).
The primary duty that he identifies in relation to this goal is that of withholding assent from
any proposition which is less than certain. In the Discourse, this duty is the first maxim by
which he chooses to regulate his conduct during his quest for certainty. This maxim states
that one is not to accept as true anything for which one does not have “evident knowledge
of its truth” (Descartes: 29). In the Meditations, Descartes cites the same primary duty when
he claims that:
If; however, I simply refiain from making a judgment in cases where I do not perceive the
truth with sufficient clarity and distinctness, then it is clear that I am behaving correctly
and avoiding error (Descartes: 103).
Descartes acknowledges within his writings the influence that the Skeptics and Stoics
had on his work Descartes attempt to emulate the Stoics is evidenced by the third regulative
maxim he cites in the Discourse. In this maxim he, like the Stoics, indicates that thought is

completely within our control, and therefore that it is through thought that we can control our

own happiness or peace of mind. Descartes writes that:

24

My third maxim was to try always to master myself rather than fortune, and change my
desires rather than the order of the world. In general I would become accustomed to
believing that nothing lies entirely within our power except our thoughts. . . This alone,
I thought, would be sufficient to prevent me fiom desiring in future something I could not
get, and so to make me content (Descartes: 32-33).

The Skeptics, on the other hand, seemed to have had a negative influence in that
Descartes sought to show how his philosophy substantially differed from theirs. Both
Skeptical and Cartesian philosophy concentrate on inducing doubt, but Descartes argued that
his philosophy stood apart in that doubt itself, or the suspension of judgment, was not the end
goal of his work. In the Discourse, he writes that:

I kept uprooting from my mind any errors that might previously have slipped into it. In
doing this I was not copying the sceptics, who doubt only for the sake of doubting and
pretend always to be undecided; on the contrary, my whole aim was to reach certainty
(Descartes: 34).
In the Meditations, Descartes reconfirrns this difference when he indicates that the usefulness
of extensive or hyperbolic doubt lies in its eventual result which is to eliminate all doubt, “to
make it impossible for us to have any firrther doubts about what we subsequently discover to
be true” (Descartes: 73).2
Despite Descartes’ own explicit assertions on this subject of influences, his position

concerning the type of control we have over belief borrows almost equally from both the

Skeptics and Stoics. From the Skeptics he seems to borrow the notion that suspension of

 

2 Perhaps this difference between Descartes and the Skeptics in terms of their use of doubt
is indicative of a greater difference between them in terms of their respective goals. One
could argue that the goal of the Skeptics, that of quietude or unperturbedness, is an ethical
or practical goal, whereas Descartes’ goal of certainty is epistemic. As such, not only do
they differ in terms of their employment of doubt, but also in terms of the kind of goal this
employment is meant to achieve. I am indebted to Prof. Richard Hall for pointing out this
difference.

25

judgment, or the inducing of doubt, is brought about indirectly. Just as the Skeptics induced
doubt through the various modes of opposing arguments, Descartes likewise contends that
doubt requires reasons or grounds, and toward this end he seeks to employ the “suppositions”
of the Skeptics.

In the Discourse, Descartes describes all the arguments he used to induce doubt, and
we can see that they clearly resemble the modes employed by the skeptics (Descartes: 36).
For instance, Descartes both here and in the Meditations seeks to induce doubt by questioning
the reliability of the senses and the distinction between dreams and reality. These correspond
to aspects of the third and fourth modes cited by Sextus (Gerson: 203). For Descartes, like
the Skeptics, all that was requisite to reject a belief was to find “at least some reason for
doubt” (Descartes: 76). Where he differs from the Skeptics is that he did not view the
available modes as being sufficient for his task. Descartes believed that a new reason must
be found to invoke the kind of hyperbolic doubt that he viewed as requisite for his
foundationalist project. This new reason was provided by his assumption of the evil or
malicious demon (Descartes: 79).

Also like the Skeptics, Descartes maintained that we cannot be mistaken about the
way things appear to us and that, as such, appearances cannot be doubted (nor can they be
subject to skeptical attack) -- we cannot doubt our appearances or impressions themselves,
but only whether they are true. The Skeptics, as we saw, argued that appearances or
impressions were not open to attack by the modes or opposing arguments. Descartes, in the
Meditations, likewise states: “I certainly seem to see, to hear, and to be warmed. This cannot

be false; what is called ‘having a sensory perception’” (Descartes: 83). Like the Skeptics,

26

Descartes distinguishes between impressions, or appearances, and judgments. It is in the
latter that we err. In the Meditations he gives us the example of our perceiving ‘men’ clothed
in coats and hats. He argues that in fact our perception is not of men. The notion of their
being men is rather a judgment. Descartes states that:

Yet do I see any more than hats and coats that could conceal automatons? I judge that

they are men. And so something which I thought I was seeing with my eyes is in fact

grasped solely by the faculty of judgment which is in my mind (Descartes: 85).

Like the Stoics, Descartes seems to maintain that we do have some direct control over
belief as well. Descartes distinguishes between two faculties, that of the will and that of the
understanding. The former faculty is that by which we judge or affirm things. He sometimes
calls it the faculty of choice or freedom of the will. It is through the understanding or intellect
that we intuit or know things (Descartes: 14, 101). Descartes maintains that although the
understanding, when it clearly and distinctly perceives, can lead the will to afirm something,
still the understanding is finite whereas the will is infinite (Descartes: 102-104). The fact that
the will is infinite means that we can exceed the bounds of the understanding in making our
judgments -- we can seek to affirm or deny things that we do not know or understand.
Descartes argues that in such situations where we apply the will to a proposition that is not
understood, we find that we are indifferent; our judgment or assent can go either way. In
these situations, then, we seem to have a kind of direct control over our belief or judgment.
We seem to be able to choose what we will believe. Descartes states that when the

intellect has not yet come upon any persuasive reason in favor of one alternative rather
than the other. This obviously implies that I am indifferent as to whether I should assert

or deny either alternative, or indeed refiain from making any judgment on the matter
(Descartes: 103).

27

Where exactly, then, does Descartes fit in terms of our earlier voluntarist taxonomy?
Descartes’ position is complicated. This is in part due to a visual analogy that to some extent
seems to dominate his discussion of intuition, or the apprehension of ideas. Descartes likens
mental intuition to vision or sight. He often speaks of the mind's eye or gaze when he is
discussing intuition (Descartes: 106, 88, 10, 6). In the Rules, Descartes is quite explicit in
maintaining that intuition can best be understood through analogy with sight. Thus, he states,
“We can best learn how mental intuition is to be employed by comparing it with ordinary
vision” (Descartes: 10). Descartes further believed that this mental vision or intuition can be
strengthened through the cultivation of certain epistemic habits or virtues. This would make
Descartes appear to be a kind of indirect voluntarist, who believes that control over belief can
be indirectly gained or modified through one’s engaging in certain practices or exercises. For
Descartes, the practices or exercises consist of such things as studying mathematical and
geometrical deductions and breaking problems down to their simplest components (Descartes:
10-1 1).3 Further, Descartes believes the cultivating of these intellectual or epistemic habits
allows us to combat the negative effects of unreliable epistemic habits or tendencies. Thus

Descartes writes that:
Admittedly, I am aware of a certain weakness in me, in that I am unable to keep my

attention fixed on one and the same item of knowledge at all times; but by attentive and
repeated meditation I am nevertheless able to make myself remember it as often as the

 

3 Descartes’ position, though, may be like that of Montmarquet, a current direct
voluntarist who maintains that we can directly control belief through the cultivation of
such virtues as conscientiousness (Montmarquet: viii). Descartes similarly maintains that
we can sharpen our mental gaze, and in doing so perhaps directly affect our judgments,
through developing the habit or virtue of perspicacity (Descartes: 10, 30). This virtue
Descartes believes to be analogous to the sharpening of the visual gaze that allows
craftsmen to make perfect distinctions among minute things (Descartes: 10).

28

need arises, and thus get into the habit of avoiding error (Descartes: 104).

This visual analogy does help to explain one interesting feature of Descartes’ view,
namely, his claim that when we clearly and distinctly perceive or intuit something we cannot
help but assent or believe. Thus, in the Rules, he argues that the conception of a clear and
attentive mind is such “that there can be no room for doubt about what we are understanding”
(Descartes: 3). In the Meditations, he says that the nature of his mind is such that he “cannot
but assent to these things [clear and distinct ideas]” (Descartes: 106). In fact, the argument
for the existence of a non-deceiving God is motivated by this aspect of our minds, for
Descartes wants to show that our forced assent in these situations will not lead us into error
(Descartes: 104-105).

Using the visual analogy this aspect of his view seems to make sense. Taking the
intuition of an idea or proposition as a kind of seeing, we can view the Skeptical arguments
as a way to poke holes in the perception so as to avoid a false judgment. With sight such
arguments might attack the lighting, or the angle of vision, or the distance. But nonetheless,
when we view something in perfect lighting, fi'om a perfect angle, at the perfect distance,
there can be no room left for doubt. When something is cleariy and distinctly perceived in this
way, seeing is believing. Thus Descartes states that when he turns to things that he perceives
or intuits very clearly, he is convinced of them (Descartes: 88).

This aspect of our minds is not a limitation on our freedom for Descartes. According
to his view, freedom involves the absence of external compulsion (Descartes: 102). In these
cases, where our assent is determined by clear and distinct perception, there is no external

force that compels our assent. Our assent is rather brought about by our own inclination

29

toward truth or goodness. Thus, in such situations we are in fact doing what we want or are
inclined to do, in the absence of external compulsion and, hence, are free. In fact, Descartes
believes that situations of indifference, where we are not inclined in one direction over
another, are actually instances of imperfect freedom, because we are ignorant of the good or
true alternative. This ignorance is what brings about our indifference. We are, according to
Descartes’ view, perfectly flee when we are completely determined by inclination for in such
situations we are completely knowledgeable (we clearly and distinctly perceive the truth or
goodness of one of the alternatives). Descartes writes:

I could not but judge that something which I understood so clearly was true; but this was

not because I was compelled so to judge by any external force, but because a great light

in the intellect was followed by a great inclination in the will, and thus the spontaneity and

freedom of my belief was all the greater in proportion to my lack of indifference

(Descartes: 102).
Of course, in the view of fi'eedom that is being employed in this dissertation such belief would
not be free, for one cannot choose or will to believe in a manner contrary to this inclination.

Where does Descartes’ view fall? He may fill all the categories. Thus, in regard to
the inducing of doubt, at least in some instances, he appears like the Skeptics to be an
indirect, negative voluntarist. In regard to cases of indifference he appears to be a direct,
positive and negative, voluntarist. Finally, in regard to clear and distinct ideas, he appears
given our definition of freedom, to be a non-voluntarist.
Despite its resistance to classification, Descartes’ view departs from that of his

influences, the Stoics and Skeptics, in that he does not view either direct or indirect control

as being even close to absolute or complete. Both types of control are, for Descartes, limited,

direct control being restricted to cases of indifference, while indirect negative control finds

30
its limit in the case of clear and distinct ideas. Given these limitations on control, Descartes’
primary epistemic duty or maxim to ‘refrain from assenting in cases where one does not
clearly and distinctly perceive’ could be rephrased as a duty to ‘withhold assent in all cases
where this is possible,’ or ‘to believe or assent only in those cases where one cannot help but

do so."

IV. Locke
When the agreement of any two ideas appears to our ,
minds, whether immediately or by the assistance of
reason, I can no more refitse to perceive, no more
avoid knowing it, than I can avoid seeing those objects
which I turn my eyes to, and look on in daylight; and
what upon full examination [find the most probable
I cannot deny my assent to.

Locke: Essay BK IV, XX, 16

Though Locke’s Essay is in many ways a critical response to Cartesian philosophy
(with its emphasis on innate ideas), Locke’s own view on belief and voluntarism bears much
similarity to that of Descartes. First, both Locke and Descartes believe that custom or habit
is a major source of cognitive error. As we saw earlier, Descartes viewed his own method
(regarding the development of certain intellectual virtues) as a way to counteract certain
negative epistemic habits which lead us into error. Locke likewise believes that custom, habit,
or what he terms the association of ideas, is an aspect of our intellectual lives that must be

closely checked. In fact, both Descartes and Locke use much the same language in discussing

 

‘ I am indebted to Prof. Charles McCracken for this observation.

31
the nature of habit and its potentially damaging effects. Thus they both liken habit to a path
worn smooth by continuous travel. In regards to the attractiveness of customs Descartes
writes:
it is almost always easier to put up with their imperfections than to change them, just as
it is better to follow the main roads which wind through mountains, which have gradually
become smooth and convenient through frequent use (Descartes: 27).
Although Descartes may seem to place habit or custom in a positive light here, he states
elsewhere that:
My habitual opinions keep conring back, and, despite my wishes, they capture my belief,
which is at it were bound over to them as a result of long occupation and the law of
custom. . . I think it will be a good plan to turn my will in completely the opposite
direction. . . until the weight of preconceived opinion is counterbalanced and the
distorting influence of habit no longer prevents my judgment from conceiving things
correctly (Descartes: 79).
In the Essay, Locke writes the following concerning habit:
Custom settles habits of thinking in the understanding, as well as of determining in the will,
and of motions in the body; all which seem to be but trains of motions in the animal spirits,
which, once set a-going, continue in the same steps they have been used to, which, by
often treading, are worn into a smooth path, and the motion in it becomes easy
(Locke: 251).
Locke goes on to say of habit that:
This wrong connexion in our minds of ideas. . . has such an influence, and is of so great
force to set us awry in our actions. . . that perhaps there is not any one thing that deserves
more to be looked after (Locke: 252). ‘
Descartes and Locke also believe that truth, self-evidence, or cleamess and
distinctness, in some way compels us to believe or assent. As we saw earlier, Descartes

maintains that when an idea is clearly and distinctly perceived we cannot help but assent to

it. Locke's position on this matter seems quite similar to that of Descartes. Locke makes a

32

distinction between belief or judgment, on the one hand, and knowledge, on the other. Both
of these cognitive attitudes, for Locke, deal with assenting or dissenting to the perception of
the agreement or disagreement of ideas. The difference between them is that with knowledge
this perception is clear or certain, whereas with judgment it is merely probable (Locke: 402-
403).

Regarding knowledge, Locke maintains that the clearness or distinctness of the
perception involved not only confers certainty, but also compels one to assent to the
perception. Locke argues that the ideas involved are known or perceived so distinctly and
unconfusedly that:

all such afiirnmtions and negations are made without any possibility of doubt, uncertainty,
or hesitation, and must necessarily be assented to as understood (Locke: 366).

Locke maintains that with knowledge there exists “intuitive evidence which infallibly

determines the understanding” (Locke: 406). With regard to judgment or belief, Locke

argues that we have the duty to apportion the degree of our assent to the evidence, and to

withhold assent until this apportionment can be properly determined. Thus, Locke states:
if it will proceed rationally, [the mind] ought to examine all the grounds of probability and
see how they make more or less for or against any probable proposition, before it assents
to or dissents fi'om it; and, upon a due balancing the whole, reject or receive it, with a
more or less firm assent, proportional to the preponderancy of the greater grounds of
probability (Locke: 406).

This would suggest that the degree of evidence in cases of judgment, as it falls below

certainty, is such that it does not determine assent, and this is the reason that we can properly

be obligated to regulate our assent in such cases. Locke, however, does not hold this position

in regard to instances of judgment. Unlike Descartes, who believed that the pull of the merely

33

probable could be resisted, Locke argues that probability can compel assent. Thus, whereas

Descartes writes in regards to the probable that:
although probable conjectures may pull me in one direction, the mere knowledge that they
are simply conjectures, and not certain and indubitable reasons, is itself quite enough to
push my assent the other way (Descartes: 103),

Locke writes that:
where the proofs are such as make it highly probable, and there is not sufficient ground
to suspect. . . there, I think, a man who has weighed them can scarcely refirse his assent
to the side on which the greater probability appears. . . I think, assent [judgment] is no
more in our power than knowledge...what upon full examination I find the most probable
I cannot deny my assent to (Locke: 439-440).

Given this position, how can Locke maintain that we have any obligations concerning
how we apportion our assent to the evidence? It would seem that if the preponderancy of the
evidence always determines assent there cannot be the freedom or control requisite for
obligation or duty. Locke, however, is not a non-voluntarist - he does not maintain that we
have no control over belief. Rather, he argues that given the way in which evidence
determines assent, our control over belief is almost exclusively indirect: we can control belief
through our control over inquiry or the gathering of evidence. Locke, in fact, believes that
it is only by virtue of this power over belief via inquiry that any praise or blame in regard to
cognitive error is possible. Thus, he states that:

though we cannot hinder our knowledge where the agreement is once perceived, nor our
assent, where the probability manifestly appears upon due consideration of all the
measures of it; yet we can hinder both knowledge and assent, by stopping our inquiry, and
not employing our faculties in the search of any truth. If it were not so, ignorance, error,
or infidelity, could not in any case be a fault (Locke: 440).

For Locke, our control over belief is almost exclusively indirect, in that he concedes

that there are situations which afford greater control. Like Descartes, Locke holds that in

34
cases of indifference, where the evidence is uncompelling on both sides (or insufiicient to
determine assent), we possess more liberty in regards to belief. Thus, Locke states that:
[In] less clear cases I think it is in a man's power to suspend his assent, and perhaps
content himself with the proofs he has, if they favor the opinion which suits his inclination
or interest (Locke: 43 9-440).
Like Descartes, Locke admits some direct control (albeit negative) in cases of indifference,
He contends that in such cases suspension of judgment is within our power. But unlike
Descartes, Locke believes that our negative control over belief is limited not only by certainty,
but by probability as well.

We saw earlier that Descartes’ view of freedom was such that the compelling nature
of clear and distinct, or certain, propositions, did not constitute a limitation on our freedom
or liberty. This is because Descartes accepts the Desire Thesis under which we are free in so
far as we can do what we want, and for Descartes we want to believe what is clear and
distinct. Locke, on the other hand, holds the view of freedom I earlier identified as the Will
Thesis: Locke believes we are free in so far as we are acting under our own control or choice.
Liberty or Freedom for Locke thus involves “the person having the power of doing, or
forebearing to do, according as the mind shall choose or direct” (Locke: 167). For Locke,
then, the compelling nature of certainty and probability (of evidence in general) does entail
a real limitation on our freedom.

Given this, the attempt to impose our taxonomy of voluntarism on Locke's view has
at least the possibility of being in agreement with the language of his own theory. In cases

of indifference, Locke would seem to be a direct negative voluntarist. Regarding all other

cases, he would seem to be an indirect, negative and positive, voluntarist. In both cases,

35

however, Locke, like Descartes, is certainly an incomplete voluntarist. He limits direct
control to cases of indifference, and he limits our indirect control by our perception of the
evidence. Our indirect control seems is limited by our perception of both probability and
certainty in that both of these compel assent. Thus, Locke argues that we cannot hinder
knowledge or judgment once the evidence has been perceived (Locke: 440).

We can see now that the epistemic positions of Locke and Descartes bear a striking
disanalogy to most contemporary ethical theories. This is because, in their view, the major
epistemic goal (truth) compels epistemic action (belief or assent). This is in contrast to most
ethical theories where the ethical goal (goodness or rightness) does not compel ethical action
or conduct (physical action). Locke himself may have been aware of this difference for in
regards to physical conduct he writes that we always have the power to suspend action or
suspend the satisfaction of any desire (Locke: 175), but as we have seen, he certainly does
not argue that we have a similar power in regards to assent. In this regard, in Locke's View,
our physical liberty or fieedom exceeds our epistemic freedom. In fact, our physical fieedom

serves as the primary basis for what indirect control he admits.

V. Clifford

he had no right to believe on such evidence as was before
him. He had acquired his belief not by honestly earning

it in patient investigation, but by stifling his doubts... [and]
inasmuch as he had knowingly and willingly worked himself
into that state of mind, he must be held responsible for it.

Clifford: The Ethics of Belief

36

Like Descartes and Locke before him, Clifford holds that we have certain duties or
obligations pertaining to belief. The principal duty that he recognizes is that of ensuring that
one's belief is based upon suflicient evidence. Clifford sums up this duty in the following way:
“it is wrong always, everywhere, and for anyone, to believe anything upon insufiicient
evidence” (Clifford: 505).

Also like Descartes and Locke, Clifford views one motivation for this duty as being
the avoidance of bad epistemic habits. Thus, the observance of this duty will ensure that we
protect ourselves from habits which may lead us into error, and which may undermine our
control over belief. Accordingly, Clifford states that:

Every time we let ourselves believe for unworthy reasons, we weaken our powers of self-
control, of doubting, of judicially and fairly weighting evidence. . . [and] a habit of
believing for unworthy reasons is fostered and made permanent (Chfiord: 504).

Clifford’s position is especially close to that of Descartes in that, like the latter,
Clifi’ord views his epistemic duty as entailing a correlative duty to question what one believes.
Recall tint, for Descartes, our primary duty is to avoid error, to believe only what is clearly
and distinctly perceived. This, in turn, leads him to employ skeptical arguments against what
he had previously assented to, seeking not merely to induce doubt but rather to achieve

certainty. Likewise, Clifford maintains that his duty to believe only on the basis of sufiicient

evidence leads to the “universal duty of questioning all that we believe” (Clifi‘ord: 504).s

 

5 Descartes’ position, though, may not be quite as strong as that of Clifford in this regard.
Thus Descartes does seem to admit that certain practical necessities may require us to
believe, or at least act as if we believed, in the absence of certainty. In the Discourse on
Method Descartes states that “since in everyday life we must often act without delay, it is
a most certain truth that when it is not in our power to discern the truest opinions, we
must follow the most probable” (Descartes: 32). He goes on to assert that “in practical
life it is sometimes necessary to act on opinions as if they were indubitable, even when one

37

Lastly, like Descartes, Clifford seems to maintain that we have an almost complete or
absolute negative control over belief. This assumption of near absolute control in Clifford’s
position is implied by the exceptionless nature of the duties or obligations that he discusses.
Thus, in terms of the duty of questioning all that we believe, Clifford argues that “No
simplicity of mind, no obscurity of station,” can permit one to escape this universal duty
(Clifford: 504). In terms of the duty to believe only on sufficient evidence, he argues that its
violation is “wrong always, everywhere, and for everyone” (Clifford: 505). This latter duty
seems to imply a powerful negative control over belief (the power to withhold belief in the
face of insufficient evidence). Without such a strong negative power, the universal duty to
believe only when the evidence is suflicient would be nonsensical. Without such a power we
could not avoid believing in the face of insufiicient evidence. This negative control is perhaps
exercised indirectly through our power to test statements, which is in turn implied by our
universal duty to question. This would also bring Clifford’s position closer to that of
Descartes, for the latter believed that doubt or suspension of belief is primarily induced
indirectly, through skeptical arguments or questioning.

There is in fact some evidence within Clifford’s writing which suggests that he viewed
our control over belief as being primarily indirect. In the examples of wrong belief that he
discusses, the belief is usually the product of the agent intentionally ignoring relevant and
undermining evidence. Thus he discusses the following case:

If a man, holding a belief which he was taught in childhood or persuaded of afterwards,
keeps down and pushes away any doubts which arise about it in his mind, purposely avoids

 

knows they are quite uncertain” (Descartes: 35). In this respect Descartes’ position is
similar to that of James in “The erl to Believe”.

38

the reading of books and the company of men that call in question or discuss it, and

regards as irnpious those questions which cannot be easily asked without disturbing it - the

life of that man is one long sin against mankind (Clifford: 505).
For Clifford, it is through our examination of evidence, investigating and questioning what
we believe, that we exercise our doxastic control. This control is therefore indirect, and can
be both positive and negative. By carefirlly regulating evidence, either seeking it out or
avoiding it, we can both bring about and suspend belief. As such, I would argue that
Clifford’s position is that of an indirect, negative and positive voluntarist. In addition, like
Descartes and Locke, he is probably not a complete voluntarist, at least in terms of negative
control. Where doubt cannot be induced through questioning or skeptical argument, where
our belief is certain or unshakable, we cannot suspend our belief. For Clifford, it is only in

the absence of doubt (brought about either by suppression or the conclusive of the evidence)

that belief can be wrongfully, or rightfiilly, maintained.

VI. James
Our passional nature not only lawfully may, but
must, decide an option between propositions,

whenever it is a genuine option that cannot by its
nature be decided on intellectual grounds.

James: “The Will to Believe” CH IV

James’ position, at least in “The Will to Believe,” is a response to the epistemic view

of Clifford which he sees as emphasizing the avoidance of error at the expense of truth.6 Of

 

6 In Chapter Five (Sec. IId) the disagreement between Clifford and James will be discussed
in more detail. It will be shown that James’ response to Clifford is motivated by

Clifford, James writes that he exhorts us to

Believe nothing...keep your mind in suspense forever, rather than by closing it on
insufficient evidence incur the awful risk of believing lies (WTB: 510).

James puts forward two epistemic duties which he believes are independent, and to some
extent in conflict. These duties are to “know the truth” and to “avoid error,” these are “our
first and great commandments as would-be knowers” (WTB: 509). James argues that the
intellect or reason does not dictate which of these duties is primary. He argues that emphasis
of one over another is merely an expression of our passional nature. As such, it is possible
for one to choose which is primary. One can decide whether to be guided by the “horror of
being duped” or the “chase of truth” (WTB: 509-510). Rather than follow the example of
Clifford, who he views as emphasizing the second of these duties, James wishes to argue for
an epistemic position which privileges the first. James wants to defend a position which
emphasizes the pursuit of truth, and which will allow us under some circumstances to risk
error in its name. James writes that:
Clifford's exhortation has to my cars a thoroughly fantastic sound. It is like a general
informing his soldiers that it is better to keep out of battle forever than to risk a single
wound. Not so are victories either over enemies or over nature gained (WTB: 510).
Like Descartes and Locke, James pays special attention to cases of indifference,

instances where our intellect or reason is unable to determine belief due to insufficient

evidence, or where the arguments or evidence on both sides are about equally strong. As

 

misunderstanding. James mistakenly views Clifford as an epistemic consequentialist, i.e.,
as weighting the avoidance of error more heavily than the attainment of truth. It will be
argued, though, that Clifford is an epistemic deontologist. That his view arises not fiom
the assessment of epistemic consequences, the weighting of the benefit of true belief
against the detriment of false belief, but rather fi'om strict adherence to duty.

40
with his two predecessors, James recognizes that in such cases the will has a powerfirl and
perhaps direct influence over belief. In fact, in other cases James seems to deny any doxastic
power or potency to the will. He writes:
Can we, just by willing it, believe that Abraham Lincoln's existence is a myth...Can we, by
any effort of our will, or by any strength of wish that it were true, believe ourselves well
when we are roaring with rheumatism in bed. . . We can say any of these things, but we
are absolutely impotent to believe them; and of just such things is the whole fabric of the
truths that we do believe made up (WTB: 507).
James argues that from one point of view “believing by our volition seems...simply silly”
(WTB: 508).

But some qualifications need to be made here. In the “Will to Believe” James is
arguing against the commandment of agnosticism, or the scientific credo, which he contends
is represented by the position of Clifford. This Credo maintains that one “shall not believe
without coercive sensible evidence” (Essays on Faith and Morals hereafter EFM: 25). Given
the promptings of this commandment, suspension of belief, or neutrality, is required in cases
where this evidence is lacking. James wants to show that neutrality in every such situation
is unattainable, that under certain circumstances inaction is a form of action (EFM: 24). He
thus focuses on cases where the credo would advise us to suspend judgment, cases of
indifl‘erence where the evidence is inconclusive, and attempts to show that certain situations
which meet this description do not admit of neutrality - that neutrality would amount to a
choice of one of the alternatives, and that as such we are permitted to “will to believe” (to
choose actively rather than by default). In these cases of indifference, then (which I will go

on to define very specifically in a moment), James seems to contend that we do have a kind

of direct control over belief, that we can will or choose to believe. In the Principles,

41

concerning the choice between belief in either determinism or freewill, James writes:

But when scientific and moral postulates war thus with each other and objective proof is
not to be had, the only course is voluntary choice (PP: 573).

Outside of such cases where we are faced with warring alternatives and lack any
objective means to decide between them, James seems to maintain that we have a kind of
indirect control over belief. He argues that we may not be able to believe abruptly, but that
we can employ methods by which we may gradually be led to belief (PP: 321). One such
method is to is to act “as if the thing in question were real,” to act as if we believed (PP: 321).
Another method by which we can indirectly control belief involves our attention to evidence.
In a conversation with Chauncey Wright, James is said to have argued that:

Belief is only a matter of choice, and therefore a moral duty, so far as attending to
evidence is a volitional act...and attention to all accessible evidence [is] the only duty
involved in belief (The Thought and Character of William James hereafter TC: 531).
But there are forces to be battled against in this regard. As such, our indirect control, like our
direct control, is limited. These forces consist of such things as emotions (PP: 308, 562) as
well as custom or habit (PP: 568). For James, then, both our indirect and direct control are
incomplete. The former is limited by certain forces that oppose the will, and the latter to
specific circumstances of indifference.
James’ position, then, concerning our control over belief is not as strong as it may
appear in “The Will to Believe.” Though his position is at times unclear, James seems to
maintain that an indirect control over belief exists outside of cases of indifference. It is, none

the less, only in cases of indifference, where the intellect cannot determine belief, that the will

has direct influence. Given his emphasis on the pursuit of truth, James wishes to argue for

42

the position that in cases of indiflerence, the will not only has the capacity to determine belief,
but the right to do so (WTB: 509). In these such cases we have permission to “will to
believe.”

James is more specific than either Locke or Descartes are in defining the cases or
instances of indifference in which the will has direct power over belief. He begins by arguing
that the choice between the hypotheses, or candidates of belief, in these cases must not only
be undetermined by the intellect or evidence, but that the choice must also be living,
momentous, and forced (WTB: 510, 513). By a living choice or option, James means that
the possible candidates for belief, or hypotheses, are both living. This in turn means that they
both appeal as a real possibility to the person to whom they are proposed. As such, liveness
or deadness is not an intrinsic property of an hypothesis, but is rather relative to the agent
who is confronted by it (WTB: 506). To illustrate through example, for most persons today,
the notion of a geocentric universe would not be a living hypothesis, because it would not
present itself as a real possibility. A forced option or choice is one that is unavoidable, one
about which we cannot remain neutral (WTB: 507). For instance, if I tell you to “either leave
the room or stay within it” your choice or option is forced, you cannot avoid choosing one
of the alternatives. Finally, a momentous option or choice is one in which the opportunity is
unique, the stakes are great, and the decision is irreversible (WTB: 507). A momentous
option would fall under what we commonly call today “a once in a lifetime chance.”

From the standpoint of the “right to believe” that James is attempting to establish, the
forced and momentous nature of the option or choice is perhaps most important. The option

must be forced, for if it is not, we can, and perhaps should, suspend judgment until our

43

intellect can decide the matter - until, for example, more evidence can be gained (WTB: 510).
And if the option is not momentous, we will not suffer an irreversible loss through such a
suspension of judgment; we will not miss “a once in a lifetime chance.”7 From the standpoint
of voluntarism, though, the living nature of the option or choice is essential, for in the case
of dead hypotheses our will is impotent. Our will cannot bring them to life again (WTB:
509)

For James, then, at least our direct control over belief extends to very specific
circumstances. The situation must not only be one of indifference. The option or choice
involved must also be living. We must regard the candidates for belief involved as being real
possibilities. In this regard, perhaps James is merely refining the insight of Descartes and
Locke from the vantage point of his pragmatism. The phenomenology of Descartes’ and
Locke’s position seems essentially correct. We do appear to experience a greater degree of
control over our belief in cases where the evidence on both sides is essentially even. In such

instances, we do feel as if we could choose or will to believe either way, or perhaps even

 

7 James is not alone in asserting that in certain circumstances, given the importance of the
matter at hand or the need for decisiveness, we must, and hence are permitted to believe,
regardless of the insufficiency of the evidence. As we saw earlier (footnote 6), Descartes
seems to hold a position quite like this. Locke as well discusses the “necessity of
believing without knowledge, nay often upon very slight grounds, in this fleeting state of
action and blindness we are in” (Locke: 409). But unlike Locke and Descartes, James
puts forward epistemic as well as practical justification for this. Thus in addition to
discussing the forced and momentous nature of the options involved, he discusses
examples where belief in such instances will in essence create its own justification after the
fact (where justification was in some sense lacking prior to belief). He tries to show us
that in some cases the truth is nicely adjusted to our needs and powers (WTB: 511). Thus
he discusses cases where belief will give the agent access to evidence that will confirm the
belief and which would be inaccessible without it (WTB: 512-513), and cases where belief
will actually create or bring about the fact believed (WTB: 511).

44

choose to remain neutral. For some of us, one or perhaps more of these choices may be more
painful than the other(s), but we do none the less feel as if the choice is in our power. James,
however, goes on to point out that mere indifference is not enough. The cases in which we
feel we can choose, the cases of indifference with which we actually concern ourselves, are
those in which the candidates are living or real possibilities.

I am indifferent about whether Buddha is a real deity; I cannot, however, apply my
will to this case and chose to believe one way or the other. This is because the candidates
involved are not living; neither is a real possibility for me. Because they are not living options
for me, each seems “absurd” (WTB: 513). As a pragrnatist, James goes beyond the account
of Descartes and Locke to examine in greater detail when we actually can and do apply our
will to belief. From the standpoint of our taxonomy, James’ position is that of an incomplete,
positive and negative, direct voluntarist, as well as an incomplete, positive, indirect
voluntarist. On his account we can directly apply our will in specific conditions of
indifference and “will to believe” even when the option is unforced or unmomentous. We can
also choose to follow the scientific credo and suspend judgment, or exercise direct negative
control, when we lack coercive evidence. James, in fact, argues that where the choice is not
momentous or forced we ought to follow the scientific credo (WTB: 511). Thus, our direct
control seems limited by the evidence. It seems to be restricted to cases where the evidence,
or our intellect, does not determine belief. James also holds that we can indirectly influence
belief in some instances by either acting as if we believed, or through our attention to

evidence.

45

VII. Conclusion

What can we learn from this examination of voluntarist positions? One conclusion
that can be drawn concerns the relationship between evidence and belief. Almost all of the
voluntarist thinkers we examined contend that once evidence reaches a certain degree or level
it compels or coerces belief. For the Skeptics, the only evidence seen as reaching this degree
was our sensory evidence, in that they argued appearances cannot be made subject to doubt.
For Descartes, appearances as well as clear and distinct ideas were able to reach this level.
Locke added probability to this list.8 He contended that even merely probable evidence
compels belief. Both Clifford and James also seemed to recognize the existence of coercive
evidence. The question for them was whether it is permissible to believe in the absence of
such evidence. The question we need to address in the chapters to come is whether this
acknowledged relationship between evidence and belief is causal or normative. Is our feeling
that we are somehow constrained or compelled to believe once the evidence reaches a certain
degree the result of evidence having a causal influence on belief, or is it rather due to our
adherence to certain norms which prescribe belief in such instances?

Another conclusion concerns the plausibility of direct voluntarism. For Locke,
Descartes, and James, the feeling of control that we have in cases of indifference seems to be

the principal ground for their contention that we have some direct control over belief. James

 

8 Heil argues that voluntarists focus primarily on perceptual beliefs. That it “is with
respect to beliefs of this sort that voluntarists seem most inclined to offer prescriptions and
advice” (Heil, 1983: 360). Our historical examination of voluntarism suggests, though,
that Heil is incorrect. Of the voluntarists we examined, almost all were willing to concede
that perceptions or appearances compel belief or assent. In Heil’s terms, almost all would
seem to agree that “seeing is believing.”

46

is the most clear in terms of specifying the conditions that must be present for the exercise of
this direct control. He argues that it must not only be the case that the evidence is
inconclusive, or does not suficiently favor one side over the other; the options or hypotheses
entertained must also be live or real possibilities for the agent who is confi'onted by them.
James, then, seems to go firrther than Locke and Descartes by specifying the actual practical
circumstances under which we feel we can will to believe. This intuition that we are capable
of direct doxastic control, which arises from our experience of belief in instances of
indifference, would seem to be a powerfirl one. Given that it was able to convince these
thinkers of the possibility of direct voluntarism, perhaps we too should remain open to this

possibility and not allow the current authority of determinism to overwhelm our judgment.

CHAPTER TWO

NON-VOLUNTARISM

It is none of the naturalized epistemologist ’s business
to assess whether, and to what degree, the input “justifies ”
the output, how a given irradiation of the subject ’s
retinas makes it “reasonable ” or “rational ” for the subject
to emit certain representational output. His interest is
strictly causal and nomological: he wants us to look for
patterns of lawlike dependencies characterizing the
input-output relations for this particular organism and
others of a like physical structure.

Jaegwon Kim: “What is ‘Naturalized Epistemology’?”
Mechanism... presents a harsh, and perhaps insoluble,
antimony to human thought.

Norman Malcolm: “The Conceivability of Mechanism”

The goal of naturalized epistemology is to trace out the relationship between what
Quine called the “meager input” of experience and the “torrential output” of theory or science
(Quine: 83). This relationship between experience and theory, or the world and belief, is
seen in causal terms. Surely Quine saw it this way, for he wished to replace questions of
epistemic priority with those of causal priority (Quine: 85). More recent naturalizing
epistemologists, such as Alvin Goldman, view this relationship in causal terms as well.

Goldman in much of his work attempts to trace out the kind of “appropriate” causal networks

47

48

that could produce justified true belief, or knowledge (see “A Causal Theory of Knowledge”).
It is not by looking at belief in strict causal terms that Quine differs fiom someone like
Goldman. It is, rather, in abandoning the project of justification, which Goldman does not
do. Quine wishes to put aside questions Quid Juris in favor of those Quid Facti: the role of
epistemology is no longer to show in what manner justification can be conferred, but rather
to show how theory, or belief, actually arises from sensory experience (Quine: 7 5-76).

With its emphasis on placing belief within a completely deterministic or causal
framework, resulting in the abandonment of justification, strong or Quinean naturalized
epistemology can be seen as the paradigm of a current non-voluntarist position. As Louis
Pojman argues, this view’s stance in regards to justification entails a particular view on the
subject of doxastic voluntarism, and hence on whether we can be responsible for our beliefs.1

Pojman writes that nattualists such as Quine reject the notion of justification in that, for them,

 

' Richard F eldman disagrees with this assumption that some degree of voluntarism is
requisite for responsibility or obligation. Feldman argues that there can be “obligations
concerning involuntary behavior”, or rather “obligations and requirements that obtain in
the absence of our ability to firlfill them” (F eldman: 547). As examples Feldman puts
forward the notion that teachers hold students responsible for the work assigned in class
even if it becomes apparent that they lack the ability to complete'or master it, as well as
the notion that one may be held responsible for paying a mortgage even if one lacks the
capacity to pay (Feldman: 547-548). These examples, though, do not seem to be
instances of obligation without voluntarism. The reason that we hold persons responsible
in these cases is that they voluntary entered into these relationships, e. g., that of a student
or a debtor. Thus even though the persons in these cases lack the power to firlfill the
obligations imposed by these relationships, it was nonetheless a voluntary decision at the
outset to enter into them. In this way these cases are much like those of strict liability as
found in legal theory. In such strict liability cases we hold persons legally responsible for
certain events, even if these events were outside their control, simply because they
voluntarily entered into a certain relationship or profession, e.g., milk distributors may be
held legally responsible for tainted milk even if the contamination is not the result of any
action or negligence on their part.

49
knowledge
is a matter of having beliefs caused in the proper way. The idea of justification implies
normativity, but we are not responsible for our beliefs; so the idea of evaluation is
misplaced (Pojman, A Theory of Knowledge: 290).

Jaegwon Kim has already put forward a powerful objection against this view of
naturalized epistemology. He argues that epistemology is an inherently normative endeavor,
and thus he questions whether Quinean naturalized epistemology, with its abandonment of
normative questions, is epistemology at all. Kim maintains that the agenda of epistemology,
at least in the Western tradition, has been to establish the criteria by which one can discern
what propositions are “worthy of belief" (Kim, 1988: 329). Hence, the central goal of
Western epistemology has been to uncover the conditions under which belief is justified. Kim
argues that this notion of justification is inherently normative. Like the idea of justification
in normative ethics, the idea of justification in epistemology deals with the notion of what is
permissible, reasonable, or responsible, conduct (Kim, 1988: 330). Kim states that “if a belief
is justified for us, then it is permissible and reasonable...for us to hold it, and it would be
epistemically irresponsible to hold beliefs that contradict it” (Kim, 1988: 330).

Kim believes, however, that Quine is calling for us to abandon this notion of
justification and to replace it with a “purely descriptive,” causal model of belief acquisition.
Quine, he believes, “is urging us to replace a normative theory of cognition with a descriptive
science”; epistemology is “to go out of the business of justification” and is rather to become
“a chapter of psychology” (Kim, 1988: 333). For Kim, this rejection of normativity is
especially troubling in that it not only results in a rejection of the Western epistemic tradition,

but also undermines the very idea of belief itself. Kim argues that the notion of belief is

50

intimately connected to certain norms of rationality, and that, as such, the concept of belief
is itself normative (Kim, 1988: 335). He states that unless the output of a given cognizer “is
subject to evaluation in accordance with norms of rationality, that output cannot be
considered as consisting of beliefs” (Kim, 1988: 336); (see footnote 5, Chapter Four). Kim
questions, then, the status of the Quinean project on two grounds, which both center on its
rejection of normative inquiry: first, that this rejection entails an abandonment of the tradition
which informs the very idea of Western epistemology, and second, that it rules out the central
epistemic concept of belief.

In this chapter, I examine a different criticism of naturalized epistemology which
focuses more directly on the non-voluntarism of this position. This criticism questions the
coherency of naturalized epistemology. I shall ask whether naturalized epistemology, due to
its adherence to doxastic non-voluntarism, is compatible with our conception of ourselves as

rational beings.

I. The Inconceivability of Non-Voluntarism

At least as early as Kant we see traces of an argument that attacks the conceivability
of determinism or mechanism, and hence by implication, that of non-voluntarism2 - an
argument which, at heart, attempts to show that mechanism is inconsistent with our
conception of ourselves as rational beings, beings capable of deliberation or judgment

(capable of assessing, judging, evaluating, believing, and finally asserting, reasons and

 

2 For the purposes of this dissertation I will be using the terms “mechanism” and
“determinism” interchangeably.

51

doctrines). Kant argued for the impossibility of our conceiving of our reason as being
directed from outside, as being directed by what he called alien causes or influences. For if
this were the case, we could no longer consider our judgments to be those of reason, they
would rather be the product of impulse. As such, according to Kant, we can never rationally
assent to mechanism, in that this view divorces our judgments from reason. Kant concludes,
then, that as rational beings we must consider ourselves free, we must operate under the Idea
of fieedom. He states:
But we cannot possibly conceive of a reason as being consciously directed fi'om outside
in regard to its judgments; for in that case the subject would attribute the determination
of his power of judgment not to his reason, but to an irnpulsion. Reason must look upon
itself as the author of its own principles independently of alien influences. Therefore as
practical reason, or as the will of a rational being, it must be regarded by itselfas free
(Kantl 16).

It is not just acts of judging, believing, and asserting that require freedom. Even acts
of doubting do. To doubt is to give reasons. One cannot doubt without reasons or grounds
(Brett: 189). As Wittgenstein indicates, in On Certainty, rational suspicion must have
grounds; doubt must find a “foothold” in reasons (Wittgenstein: 41c, 46c).

Given this connection of doubt to reasons, it can then be argued that the activity of
doubt necessitates epistemic freedom. This is because the giving of reasons is an evaluative
enterprise. One must be able to select reasons on the basis of their cogency, and for this
selection to be valid it cannot be externally determined.

Doubt is possible only where some sort of reasons can be given for not committing
oneselfto the claim in question Now, an assumption of the whole enterprise of reasoning
is that claims can (and should) be selected on the basis of their coherency, their
compatibility with other accepted claims and considerations which provide evidence for

or against them. Any thesis which implies that we cannot make such selections or that the
arrival at these selections does not have its real basis in the deliberations involved, is thus

52

at odds with the possibility of doubt...But the exercise of the capacity for rational selection
among alternative claims about the world is the exercise of epistemic freedom. One
cannot argue that this supposed rational selection is rigged or an illusion without engaging
in a process of rational selection which he takes to be neither rigged nor spurious
(Brett: 193-194).
Thus, one cannot attack or doubt one’s own epistemic or doxastic freedom, in that the
activity of doubting itself implies this fi'eedom. Such an attack would amount to an attempt
to doubt one’s capacity for doubt.

Norman Malcolm puts forward a somewhat similar argument attempting to show the
inconceivability of determinism or mechanism and, hence, non-voluntarism. Malcolm argues
that mechanism denies the very existence of intentional states (e.g., wants, desires, beliefs)
in that it severs the necessary connection between such states and action or behavior.
Malcolm claims that the connection between intentional states and behavior is a priori. He
maintains that intentional states are conceptually connected to behavior or action (Malcolm:
142). Malcolm believes that mechanism severs this connection between intentional states and
behavior, by rendering such states irrelevant to the explanation of behavior. He argues that
a comprehensive mechanistic theory concerning human behavior or bodily movements (e.g.,
a comprehensive neurophysiological theory), would seek to provide complete causal
explanations of behavior in terms of external and internal physical forces. In such a theory,
then,

in no cases would desires, intentions, purposes be necessary conditions of any human
movements. It would never be true that a man would not have moved as he did if he had
not had such and such an intention (Malcolm: 136).

Through breaking the a priori connection between intentional states and behavior by

rendering the former explanatorily irrelevant to the latter, Malcolm argues that mechanism

53

eliminates the very idea of intentional states. He states:
Purposes and intentions are, in concept, so closely tied to behavioral effects that the total
absence of behavioral effects would mean the total absence of purposes and intentions
(Malcolm: 142).
In denying the existence of intentional states, according to Malcolm, mechanism renders itself
incoherent. This is because the assertion of a doctrine is itself intentional and, hence, the
assertion of mechanism becomes self-contradictory. Malcolm states:
In particular, stating, asserting, or saying that so-and-so is true requires the intentional
uttering of some sentence. If mechanism is true, therefore, no one can state or assert
anything...Speciflcally, no one can assert or state that mechanism is true. Ifanyone were
to assert this, the occurrence of his intentional ‘speech act’ would imply that mechanism
is false (Malcolm: 145).
From this, he concludes that mechanism is in this sense inconceivable “in that no one can
consistently assert (or state, or say) that mechanism is, or may be, true” (Malcolm: 146).
Finally, even proponents of mechanism seem to accept the idea that there is some sort
of difliculty in the conceivability of this doctrine. John Watling, in a critique of Hampshire’s
view of fi'eedom, concedes that belief must be the product of inquiry (Watling: 27). He
concedes that “a person cannot hold a belief for reasons other than reasons for the truth of
the proposition in which it is a belief” (W atling: 28). In the mechanist view, however, belief
is the product of causes, and not reasons which proclaim truth or could serve as the basis for
the justification of the belief. Likewise, Daniel Dennett, in an early article on this subject,
argues that “we, as persons, cannot adopt exclusive mechanism” (Dennett, 1974: 181),
meaning that we cannot completely abandon a picture of ourselves in which we explain our

behavior on the basis of reasons rather than external causes (Dennett, 1974: 181). In a later

work, Dennett continues in this vein, arguing that:

54
still there is a feeling that there must be something absurd about soldiering on with one’s
deliberating after reaching the opinion that that very deliberating is determined...There is
a feeling that anyone who happened to continue deliberating under such circumstances
would have to be deluded about something (Dennett, 1984: 107).
He goes on to conclude that we “cannot help acting under the idea of fieedom, it seems; we
are stuck deliberating as if our futures were open” (Dennett, 1984: 108). In fact, as we will
see, Dennett maintains that belief in determinism must involve a kind of schizophrenia (or
irrationality?) in which we move between two incompatible images of ourselves (Dennett,
1984:115)

We see, then, in the positions of both opponents and defenders of mechanism, or non-
voluntarism, the notion that there is something paradoxical about this doctrine. The paradox
arises in the attempt to believe, judge, assert, or even doubt this doctrine, in that to do so
appears to presuppose the very freedom that mechanism denies. As such, mechanism comes
into conflict with our conception of ourselves as rational beings - beings capable of engaging
in these activities of belief, judgment and assertion.

It is important to note, however, that none of these arguments are attempting to
attack directly the rationality or justification of particular propositions believed but, rather,
the justification or rationality of the epistemic agents who believe, assert, argue for, or doubt
these propositions. When we discuss the rationality of belief we may be concerned with the
rationality of one or all of the following: (1) the propositional content of the belief, (2) the act
of believing that content (that is, we may be concerned with whether belief in that proposition
is rational), or (3) the believer. It may be that the first of these concerns is parasitic on the

second. To discuss the rationality of a given proposition it may be the case that we must

55

place it within the context of its actually being believed by some epistemic agent. We may
lmve to ask the question, “would it be rational for someone to believe this?” None the less,
this first concern is not being attacked directly by these arguments against determinism. Ifit
is being attacked, it is only via the second and third concerns which are the primary focus of
these arguments. The principal claim is that the act of belief in determinism cannot be rational
or that belief in determinism conflicts with the idea of a rational believer, but this does not
contest directly the rationality of the propositional content of this belief.

This distinction can perhaps be best brought out through analogy with ethics. Thus
in ethics, we might, even in reference to an agent whose act was completely determined,
evaluate the rightness or wrongness of his act in an “objective” sense. Let us assume that the
completely determined agent performs an act of a type that is morally wrong. We may quite
consistently judge that acts of this type are wrong, though we may not judge that it was
wrong of him to do the act - that he is morally responsible or blameworthy for his particular
token of the act in question. This latter attribution of responsibility is ruled out by the
determined or involuntary nature of the act.3 Similarly, we may say of a completely
determined believer that a “belief” or proposition he holds is justified, that there is suficient
evidence in favor of this belief considered as a type, though we may not say of the believer

that he himself is justified or rational in holding this “belief,” that he is a rational believer or

 

3 Thomas Nagel argues for this position when he states: “we feel that the appropriateness
of moral assessment is easily undermined by the discovery that the act or attribute, no
matter how good or bad, is not under the person’s control. While other evaluations
remain, this one seems to lose its footing” (Nagel: 280). These other evaluations include
the “evaluation of something as a good or bad thing, or state of affairs” (Nagel: 280).

56

that his particular instance or token of believing is justified.‘ This is because this latter
judgment, like its moral cousin, implies responsibility. It implies responsibility on behalf of
the believer, which is ruled out by the determined or involuntary nature of the “belief.”

In these arguments above and in those that follow, then, when it is said that one
cannot rationally accept, believe, judge, or assert a given doctrine, I do not mean to draw a
direct conclusion conceming whether the proposition believed, asserted, etc., can be justified
in itself (objectively or as a type); rather I mean only that the epistemic agent cannot perform

these activities in a rational or justified manner.

 

‘ My main concern is with the connection of rationality and rational belief to voluntary or
non-determined cognitive activity. There is this other notion of rational belief, though,
which hinges on the evidential relationship. Roderick Firth holds a position which
explicates this evidential view of rational belief. He distinguishes between what he terms
propositional warrant and doxastic warrant. Firth argues for the separation of “the
‘logical content’ of a belief fiom the psychological state of believing” (Firth, 1978: 218).
For Firth the “assessment of propositional warrant is a judgment about the evidential
relationship between certain psychological states and the proposition” (Firth, 1978: 218).
These psychological states consist of the recognition of evidence, perhaps not by the
believer but by others in the community to which the believer belongs, which supports the
proposition in question (Firth, 197 8: 219). In this way the notion of propositional
warrant, justification, or rationality, can be seen in terms much like Putnam’s division of
linguistic labor. Putnam argues that the reference of terms can be fixed by experts within a
linguistic community, and as such, all individuals within the community can employ terms
properly even in cases where they are not aware of the information or “criteria” by which
reference is fixed (Putnam, 1973: 126-128). Similarly, we may argue that propositions of
a given type, of which a token is believed by an individual, can be justified or rational by
virtue of there being experts within the individual’s community who possess evidence
which confers justification on this type. This could be a way of making sense of the idea
of “there being evidence out there which supports the proposition”. This does not mean,
though, that the individual possesses doxastic warrant for his particular token of the
proposition, or is rational in believing it. Thus the individual may not possess this
evidence which the “experts” hold, or he may not recognize it as evidence (e.g., those
persons who did not recognize the axioms of mathematics as supporting Fermat’s last
theorem).

 

57

II. A Defense of N on-Voluntarism

Most theorists recognize this apparent conflict or tension between non-voluntarism
and rational belief and deliberation (and hence rationality), and in doing so attempt to ease or
eliminate it in some way (for example, through the strategy of multiple explanations which
will be discussed below). Bernard Williams, like some naturalizing epistemologists such as
Alvin Goldman, does not employ this defensive strategy - he does not recognize any inherent
conflict or tension between rationality and determinism. Instead, Vtrrlliams attempts to define
rational or justified belief in terms of causality or external and internal causal processes. He
discusses “ratio ” creatures as having beliefs that are causally produced, and he argues that
if the causal connections broke down, they would cease to be rational creatures” (Williams:
143). erliarns argues that when we view a person’s belief as being rational because he holds
certain evidence, “because ” is generally used in such situations to indicate a causal
connection (Williams: 142). Thus according to Williams’, account rational belief is typically
the result of there being a causal connection between the belief in question and the evidence
for that belief. In regard to this reliance on causality in defining rational or justified belief,
erliams’ view is much like that of Goldman’s. Goldman argues that “correct principles of
justified belief must be principles that make causal requirements” (Goldman, 1979: 297).
Goldman argues, however, that it is the conditional reliability of the causal process, the
tendency of the process to produce true rather than false outputs or beliefs (when given true
inputs), and not mere causal connection to evidence, which confers justification (Goldman,
1979: 300). Goldman prefers reliability as the criterion of justification in that he is concerned

about cases where belief is causally connected to proper evidence but only accidentally. He

58

maintains that the causal connection which confers justification should not be the result of
chance or accident but, rather, the result of a process which reliably connects belief to
confirming evidence or states of affairs (Goldman, 1979: 297). Where Goldman and Williams
most sharply part ways, however, is on the notion of whether rational belief should involve
the believer having some awareness of the causal process which produced his belief. Williams
argues that such awareness is necessary for rational belief. He argues that belief produced
by a “pure causal connection” - one in which the believer has no such awareness - is not
rational. Even though the person may have the belief because of proper evidence, in that he
cannot recount this, the belief is irrational (Williams: 141). Goldman, on the other hand,
argues that no such awareness is requisite for rational or justified belief. He argues that one
“can have justified belief without knowing that it is justified (or believing justifiably that it is
justified)” (Goldman, 1979: 301). Ironically, Williams’ view is quite similar to that of an
earlier work of Goldman’s on knowledge. In this work, Goldman argues for a causal theory
of knowledge which requires that the knower be able to reconstruct the causal process which
leads to the belief in question (Goldman, 1967: 144).

Goldman and Williams in their respective views can, on the face of it, come up with
an account of both rational and irrational acts of belief, as well as rational believers. Thus, a
rational act of belief, according to their accounts, would be one that is caused in the
appropriate manner, and a ratioml believer would be one who arrived at beliefs in this manner
the majority of the time. Their use of the idea of rational belief, I contend, does not accord
with our common understanding. Thus, as argued above, our conception of rational beliefs

and believers seems to involve the notions of assessment and deliberation which presuppose

59

doxastic fieedom. Their causal theory of belief would seem to run counter to this notion of

freedom, and thus conflict with our conception of rationality to which it is attached.

a. Multiple Explanations

Dennett, among others (e.g., Watling), attempts to resolve this conflict by
distinguishing between two different kinds of explanation - causal, and purposive or
rationalizing explanations. The former kind of explanation is, of course, mechanistic, whereas
the latter is intentional and provides the basis of reasons for belief that our conception of
rationality requires. This strategy, to some extent, amounts to having one’s cake and eating
it too. According to this view belief assent, etc., can be explained on both levels and, hence,
a given belief can be shown to be both causally determined and rational or based in reasons.s

Watling argues that the rationalizing explanation or reasons a person may have for
holding a given belief can be separated fiom an explanation of why he actually holds the belief,
the latter being expressed in causal terms. Remember that Watling conceded that rational
belief must be based in reasons, or be the product of enquiry. In the face of this adnrission
he attempts to defend mechanism through the above distinction, arguing that the reasons for

holding a belief can be separated from the causal explanation of its occurrence. Watling

 

5 Some proponents of this dual explanation view may wish to argue that the same “thing”
is not being explained in the two explanations. When this view is related to action

theory, it is often argued that the causal or mechanistic explanation deals with bodily
movements, whereas the purposive, intentional, or rationalizing explanation deals with
actions. This attempt, though, to distinguish the objects of the two kinds of explanation
seems artificial. It would seem more plausible to assert not that we have two different
objects or events being explained, but rather one event being explained under two different
descriptions. Thus both explanations seek to explain the same “thing”, albeit under
difi’erent descriptions.

60

states:

Since the explanation involves only that belief is the outcome of enquiry and that enquiry

is an attempt to arrive at the truth, it has implications only concerning the reasons a person

may have for holding his beliefs and none concerning the explanation of his holding the

beliefs he does in terms of causal factors (W atling: 28).
He then goes on to argue that the attempt to use the requirement that belief be based on
reasons as an argument for voluntarism, and against determinism, arises from a failure to
distinguish these two types or modes of explanation, a failure

to separate the explanation a person gives of his beliefs when he gives his reasons for them

and the explanation which can be given of them in terms of causes. Those observations

[concerning the relationship between belief and reasons] do set limits to the reasons a

person can have, but not on the type, or existence, of explanation (Watling: 29).

Dennett similarly argues that what he terms the intentional stance, or rationalizing

explanation, is just one mode of explanation that we may apply to beings or systems. We may
also take what he terms a physical stance. Such a stance would amount to a mechanistic
explanation of the behavior of the system (Dennett, 1974: 162-164). He firrther argues that
these two stances, or modes of explanation, are not incompatible, that the mechanistic or
physical stance does not preclude the rational or intentional (Dennett, 1974: 173). For
Dennett, we need not interpret the system that we are attempting to explain as being perfectly
rational, that is, as being explainable strictly in terms of the intentional stance. We may find
that on particular occasions the physical stance is more appropriate or fi'uitful in terms of
explaining the system’s behavior. These modes of explanation are compatible, then, in that
they may be alternatively applied to a given system depending upon the circumstances of the

explanation Just because a system admits of rationalistic or intentional explanation, this does

not mean that we cannot, at times, seek to explain its behavior in physical or mechanistic

6]

terms (Dennett, 1974: 173). According to Dennett, the primary motivation that we have for
moving from the intentional to the physical stance is a breakdown or failure in terms of the
rationality of the system, resulting in a failure of intentional explanation (Dennett, 1974: 170).
The intentional stance is usually assumed when the complexity of the system examined
precludes physicalistic explanation (Dennett, 1974: 166).

Jaegwon Kim has already argued against this strategy of multiple explanations (Kim,
1989). I will argue against this strategy as well, but for reasons that go beyond those Kim
puts forward. Kim is concerned, as I am, with whether a single instance of behavior can be
given both a mechanistic and a purposive or rational explanation, but for Kim this issue of
compatibility leads to more general questions about the possibility of rendering multiple
explanations for a single explarrandum (Kim, 1989: 77). As we saw earlier, Norman Malcolm
argues against the possibility of multiple explanations (both rational and mechanistic) for the
same behavioral event. Kim wishes to side with Malcolm, but given his more general
concerns about multiple explanations, he takes a different route. The difference between Kim
and Malcolm becomes most clear in discussing instances of behavior. Malcolm, as we have
seen, wishes to argue that explanations involving beliefs or intentional states cannot be causal;
this is his primary argument against multiple explanations. Kim, on the other hand, wants to
construe such rational explanations causally in that he believes doing so will more clearly
bring out the difficulties he finds inherent in cases of multiple explanations (Kim, 1989: 80).
Thus, Kim is more concerned with the possibility of multiple causal explanations, and not just
multiple explanations simpliciter.

Kim argues that when we are confionted with an instance of multiple explanations,

62

each apparently complete and independent of the other, we find ourselves faced with an
epistemic instability which compels us to resolve the situation by either reducing or
eliminating one of the explanations, or by showing the explanations to be incomplete (Kim,
1989286). Thus we may, for example, attempt to identify the two causes involved and hence
construct a single explanation. In the case of rational and mechanistic explanations one might
do this by identifying beliefs with certain brain or physical states. One might also attempt to
reduce one of the causes to the other, thus showing one of the explanations to be more basic.
In the case of mechanistic and rational explanations, this is often done by attempting to reduce
the latter to mechanistic explanations in neurophysics and then physics. Quite simply, Kim
argues that when confionted with multiple explanations we must ask the question: would the
event have occurred if one of the competing causes were absent? If the answer is “no” then
we do not have two independent explanations; if the answer is “yes” then they are not both
complete (contra Hempel, Kim maintains that in cases of over-deterrnination all the causes
are necessary to form a complete causal explanation or picture); (Kim, 1989: 90, 92). Kim
argues for an epistemic principle of explanatory exclusion. He states that “No one may accept
both explanations unless one has an appropriate account of how they are related to each
other” (Kim, 1989: 95). Kim holds that multiple explanations bring a sort of incoherence into
our belief system for they defeat the goal of explanation which is simplification and unification
(Kim, 1989: 96-97). It is an incoherence which, he believes, must ultimately be resolved.
It may indeed be the case, as Kim suggests, that multiple explanations for a single
event present an instability or epistemic tension that must be resolved by the elimination or

reduction of the excess explanations so that only one “true” explanation remains (Kim, 1989:

63

86-87, 92). In the present case, however, there is a conceptual incompatibility between
mechanistic and intentional explanations that renders dual explanation impossible. This is
because the mechanistic explanation undermines the rational or intentional, and for this reason
they cannot co-exist as explanations of the same event. That mechanistic explanations
undermine rationalizing explanations we saw earlier. The rationalizing explanation seeks to
show the basis of belief in terms of reasons, whereas the mechanistic looks to explain the
event in terms of external or foreign causes. These explanations cannot co-exist. Once belief
is explained in terms of external causes and neurophysical events, it can no longer be shown
to be the product of reasons. As such the belief is no longer rational, and the rationalizing
explanation must be abandoned.‘5

This does not mean that one cannot move between the two modes of explanation as
Dennett suggests (using the mechanistic mode in cases where the intentional fails or the
intentional when the physical is too complex). What it does mean is that these two modes
cannot be employed simultaneously to explain the same doxastic event. Accordingly, one

cannot cohereme put forward a rational argument for a completely causal account of belief,

 

" It would seem that one could avoid the undermining of rational explanations by
mechanistic explanations simply by identifying the reasons for a belief with the
neurophysical events that are called upon in mechanistic explanation. As such, it would
appear that the mechanistic explanation would not be inconsistent with belief being

based on reasons. The mechanistic explanation would rather discuss these reasons under a
different description. This strategy, of course, would no longer be a multiple explanations
strategy. Identifying the reasons with neurophysical events undermines the assertion that
there are really two distinct explanations being put forward. It would rather appear that a
reduction of the explanations could take place, due to the identity of the events involved,
leaving us with only one “true” explanation. Kim argues that this is one of the approaches
that we find ourselves taking when we seek to relieve the epistemic tension resulting fi'om
multiple explanations (Kim, 1989: 89-91). The trouble with this strategy is that it
confiises reasons with causes. This particular distinction I will discuss in Chapter Four.

64

a completely non-voluntarist view. The rational acceptance of, or belief in, such an account
would require that at least one intentional or rationalistic explanation be the sole explanation
of one doxastic event, namely, the belief in determinism. The non-voluntarist or mechanistic
account would have to exempt the belief in, assertion of, itself from the causal picture. As

such, multiple explanations cannot be a defense of complete determinism or non-voluntarism.

b. Metaphysics Vs. Epistemology

In a later work, Elbow Room, Dermett puts forward a different defense of mechanism
in light of its apparent incoherency with our conception of rationality. This defense argues
that the claim of conflict between mechanism and rationality is an epistemic claim, and as such
it cannot be used to establish the metaphysical claim that mechanism is false. Dennett is
willing to admit that the notion of deliberation is in some sense incompatible with mechanism
in that it requires open or multiple possibilities (Dennett, 1984: 103). He admits that “our
capacity to engage in real time deliberation - including the deliberation required to engage in
scientific research - depends on our manifest image,” an image of ourselves as outside the
causal fi'amework established by mechanism (Dennett, 1984: 114). Dennett argues, however,
that this reliance on the manifest image, on a rationalistic or intentional explanation of our
own behavior, is “a sort of illusion born of cognitive miserliness” (Dennett, 1984: 114). What
Dennett attempts to show, then, is that “elbow room” for deliberation exists within
mechanism due to our own ignorance or lack of information (the fact that our cognitive
systems are limited); (Dennett, 1984: 112-113). It is the unpredictability of our own decisions

and most external events which provides room for the “illusion” of ourselves as rational

65

deliberators. Thus Dennett attempts to resolve this epistemic tension within mechanism
epistemically, he attempts to show how the limitations of our own cognitive faculties allow
us to act under the idea of freedom, and thus fulfill our conception of rationality or
deliberation, even though the universe is mechanistic or determined. For Dennett, it is our
lack of knowledge about ourselves and the world that renders determinism coherent.7
Dennett seems to view arguments which attempt to show that determinism conflicts
with our conception of rationality, as attempting to show the falsity of determinism itself.
Even proponents of doxastic freedom, though, recognize that the freedom we feel we must
possess to be rational may be just an illusion. Thus, Malcolm admits that the inconceivability
of mechanism “does not establish that mechanism is false” (Malcolm: 149). Even within
Kant’s view, the presupposition of our fi'eedom may just be an “unavoidable illusion”
(Sullivan: 83). As such, this defense by Dennett misses the point. The thrust of these anti-
mechanistic arguments is not to show that determinism is false. The idea is to show that
mechanism or non-voluntarism violates our conception of ourselves as rational beings or
believersand, as such, it cannot be rationally defended. Dennett himself seems to admit this
point when he argues that adherence to mechanism necessitates a kind of schizophrenia where
we move between two incompatible images of ourselves, a rational and a mechanistic
(Dennett, 1984: 111, 113-114). He goes so far as to argue that the very image of the world

that we employ in science, that of mechanism, must be abandoned by those who engage in this

 

7 Paul d’Holbach makes this point in his System of Nature (1770). He states that “It is,
then, for want of recurring to the causes that move him; for want of being able to analyze,
from not being competent to decompose the complicated motion of his machine, that man
believes himselfa fi'ee agent: it is only upon his own ignorance that he founds the
profound yet deceitful notion he has of his fiee agency” (d’Holbach: 367).

66

research - that they as researchers or rational beings cannot consider themselves to be a part

of the causal picture that they impress on their subject(s) (Dennett, 1984: 114).

[11. Conclusion

The major concern of this dissertation is to establish the possibility or conceivability,
relative to our conception of ourselves as rational agents, of the kind of doxastic freedom
requisite for epistemic obligations or responsibilities. In this chapter, I have argued that
doxastic freedom is not only conceivable given the assumption that we are rational beings, it
is conceptually necessary. Now it remains to be seen whether there are any conceptual
constraints or limitations on the scope of this freedom consonant with our conception of
ourselves as rational beings and the notion of epistemic responsibility. This is the task of the

next two chapters.

CHAPTER THREE

INDIRECT VOLUNTARISM

though we cannot hinder our knowledge where

the agreement is once perceived, nor our assent,
where the probability manifestly appears upon due
consideration of all the measures of it; yet we can
hinder both knowledge and assent, by stopping our
inquiry, and not employing our faculties in the
search of any truth. If it were not so, ignorance,
error, or infidelity, could not in any case be a fault.

Locke: Essay BK IV, XX, 16
It is not that one has a choice in the beliefs that
one forms, but that one has a say in the procedures
one undertakes that lead to their formation. The
notion of ‘epistemic responsibility’ attaches to the

undertaking of appropriate procedures.

John Heil: “Doxastic Agency”

Now that we have established that some form of doxastic voluntarism is necessary
relative to our conception of ourselves as rational agents, it remains to be seen what is the
most viable of the voluntarist positions that are open to us, and whether there are any

conceptual constraints or limitations on voluntarism.

67

68
l. The Basis for Indirect Voluntarism
As we saw earlier, Locke was, for the most part, an indirect voluntarist. Like
Descartes, Locke held that clear and distinct ideas, or self-evident truths, compel assent. He
argued that “all such affirmations and negations are made without any possibility of doubt,
uncertainty, or hesitation, and must necessarily be assented to as understood” (Locke: 366).
But unlike Descartes, Locke held that mere probability compels assent as well. He argued
that one “can scarcely refiise his assent to the side on which the greater probability appears”
(Locke: 439-440). As such, Locke was left with the view that we see quoted above, namely
tint the source of epistemic blameworthiness, the source of our responsibility for belief, lies
in the procedures or methods that we can employ to help bring it about - that responsibility
lies in the indirect control that we have over belief (our control via inquiry).
More recently, John Heil argues that the phenomenology of belief suggests that it is

not under our direct control. He states:

How often do we decide to adopt a certain belief? To what extent are our beliefs under

our direct control at all? In contrast to the picture painted by the voluntarist, our beliefs

seem mostly forced on us. Or, if that is too strong, they come to us unanticipated and

unbidden...The phenomenology of belief, then, as distinct from its epistemological

conceptualization, looks distinctly non—voluntary (Heil, 1983: 357).
Heil contends that this fact about the relationship between our will and belief is contingent.
He argues that the “there seems to be no a priori reason why a belief could not be created by
‘directly’ willing it” (Heil, 1983: 358). Focusing on the example of perceptual beliefs, Heil
simply maintains that belief is the product of, or caused by, the interaction between external
stimulation and background beliefs or theory, and that within this interaction there is no room

for “an additional act of will” (Heil, 1983: 361). Heil, none the less, does not wish to deny

69

the notion of epistemic responsibility. He argues that although we may not be able to control
belief directly, there are certain activities that we can perform, such as information-gathering,
and these activities can promote or inhibit the formation of belief (Heil, 1983: 359). These
activities, in turn, may be carried out in an appropriate or inappropriate manner, and it is here
that responsibility attaches itself. Heil argues that “the ‘responsible’ epistemic agent is,
roughly, one who goes about the activity of information gathering in a suitable fashion” (Heil,
1983: 362). The notion of what is suitable he believes is dependent on a variety of factors
such as the subject matter and the circumstances surrounding the investigation (Heil, 1983:
362). For Heil, it is our control over the “business” of belief gathering that makes us
responsible, albeit indirectly, for the beliefs we have.

Other recent writers, though sharing Heil’s conclusion that belief is subject only to
indirect control, dispute his claim that this is merely a contingent fact. Bernard Williams
argues that our lack of direct control over belief constitutes an a priori truth; it is not a
contingent fact about us. Williams puts forward, like Heil, a causal account of belief that
could motivate a contingent phenomenological argument against direct voluntarism (Williams:
141-143), but none the less his major argument against such control is conceptual or a priori.
Williams contends that the idea of direct control over belief conflicts with the very concept
of belief itself. He maintains that such direct control conflicts with the truth-oriented nature
of belief, which is an essential aspect of it (Williams: 136, 148). Williams states that

it is not a contingent fact that I cannot bring it [belief] about, just like that...Why is this?
One reason is connected with the character of beliefs that they aim at truth. If I could

acquire a belief at will, I could acquire it whether it was true or not; moreover I would
know that I could acquire it whether it was true or not (Williams: 148).

70
Thus, for Williams, direct control over belief is a conceptual impossibility, because he views
such control as severing the necessary connection between belief and truth.

Another recent writer, Louis Pojman, agrees with Heil that direct control over belief
seems phenomenologically unlikely, but like Williams he argues in addition that such direct
control is also conceptually or logically impossible. Thus Pojman, like Heil, argues that
“phenomenologically speaking, acquiring a belief is a happening in which the world forces
itself on a subject” (Pojman, 1985: 528). Pojman also holds a causal view of belief, like Heil,
which helps in part to motivate this phenomenological account. Pojman claims that “in a
sense, belief that p seems to imply the thought of a causal chain stretching back from the
belief to a primary relationship with the world and so faithfully representing the world”
(Pojman, 1985: 534). Phenomenologically speaking, Pojman maintains that direct control
over belief is “abnormal and bizarre” (Pojman, 1985: 531). But he admits that this
bizarreness does not rule out the possibility of direct voluntarism. This is where the
conceptual argument comes in for Pojman. Like Williams, Pojman contends that direct
voluntarism conflicts with the truth-oriented nature of belief. This truth-oriented nature of
belief is spelled out, in part, in the evidential requirement that Pojman views belief as
possessing. Thus he argues that “Believing is evidential, in that to believe p is to presuppose
that I have evidence for p or that p is self-evident or evident to the senses” (Pojman, 1985:
534). Direct voluntarism then conflicts with this view of belief in that it is said to involve
believing for non-evidential reasons. Thus Pojman states that:

volitional believing involves a conceptual confitsion; it is broadly speaking a logical

mistake. There is something incoherent in stating that one can obtain or sustain a belief
in full consciousness simply by a basic act of the will, that is, purposefiilly disregarding

71

the evidence connection (Pojman, 1994: 275).

Thus like Williams, Pojman argues that direct control over belief is conceptually or logically
impossible because it violates the truth or evidential aspect of belief - direct voluntarism, it
is argued, must anive at belie “independently of evidential or truth considerations” (Pojman,
1985:533).l

Like Heil, Williams and Pojman do not conclude from their arguments against direct
voluntarism that epistemic responsibility is impossible. Both are adherents to indirect
voluntarism as a way to salvage some notion of responsibility in terms of belief. Thus
Williams allows that “there may be room for the application of decision to belief by more
roundabout routes” (Williams: 149). These methods or routes include such things as hypnosis
and drugs. Williams focuses on such methods because he discusses indirect voluntarism
chiefly within the context of self-deception (Mlliams: 149). He also believes, however, that
evidence is a causal factor in the production of belief (“Williams 141-143). Given this, there
would seem to be room within his account for rational, or epistemically responsible, indirect
control over belief through the conscientious pursuit of evidence. Pojman argues that “It is
prinmrily because we judge that our beliefs are to some significant degree the indirect results

of our actions that we speak of being responsible for them” (Pojman, 1985: 535). The chief

 

‘ It is interesting to note that Pojman, like Locke, seems to primarily rule out only positive
direct voluntarism. Thus Pojman allows for the possibility that we may have negative
direct control, or some sort of “veto” power over belief. He admits that the will may have
“a negative role to play in the preventing of beliefs from fixing themselves in us”

(Pojman, 1985: 532). Locke, as we saw in Chapter One, held that the will may have such
a “veto” power. Thus Locke argues that in less clear cases “it is in a man’s power to
suspend assent” (Locke: 35). Pojman in fact attributes his view, in part, to Locke. This
point will be discussed further in the next chapter.

72

manner in which we exercise this indirect control, according to Pojman, is through the
exercise of attention (Pojman, 1985: 536).
lastly, H.H. Price argues against direct control over belief on the grounds that belief
is a disposition and not an act. Thus Price argues that:
It is true that we do sometimes use volitional words in describing the acquisition of beliefs
(“I decided that p”, “I made up my mind that p”). But we must not allow ourselves to be
confused by the fact that something rather like preferring or choosing does quite often
occur as a stage in the process by which a belief is formed, especially when we acquire our
belief in a reasonable manner, afier careful consideration of the evidence pro and con.
Believing a proposition is, I think, a disposition and not an occurrence or “mental act”,
though the disposition is not necessarily a very long-lived one and may only last a few
seconds (Williams: 15).
This dispositional view of belief is much like the account of belief put forward by William
James. Thus James argues that “there is some believing tendency wherever there is
willingness to act at all” (WTB: 506). And like James, Price maintains that “Beliefs can be
gradually cultivated, though they cannot be instantaneously produced, or abolished, at will
(Price: 16—17, 21). The primary means that he recognizes for exercising this indirect control
are attention, as well as focusing on propositions, and acting as if one believed (Price: 17-20).
For Price, as with the other indirect voluntarists that we have discussed, belief can be
influenced by “a longish course of voluntary effort, though not by a mere fiat of the will here
and now” (Price: 21-22).2

We are now in a position to see about Montmarquet’s view that, contrary to his

assertions, his position seems to be that of an indirect rather than direct voluntarist. As we

 

2 These phenomenological, conceptual, and definitional arguments against direct
voluntarism will be critically examined in Chapter Four. They are discussed in this chapter
uncritically in keeping with the goal of illustrating some possible reasons motivating the
adherence to indirect voluntarism.

73

briefly saw in the introduction, Montmarquet maintains that we can control belief by
cultivating certain epistemic habits or virtues, and that the exercise of these virtues in relation
to belief is a form of direct control (Montmarquet: viii). Montmarquet states that to be good
or virtuous epistemic agents the epistemic virtues, habits, or dispositions we cultivate should
be conducive to the epistemic goal of truth. Such virtues would include conscientiousness,
intellectual sobriety, etc. (Montmarquet: 20). To cultivate these virtues would seem, to
borrow Heil’s language, to be roughly equivalent to going about the business of information
gathering in a suitable or responsible manner. But unlike Heil, Montmarquet argues that the
exercise of this control over belief is direct. This is because he views the exercise of these
virtues as a “modality of the belief-forming process” and not a way of indirectly “causing
oneself to have or not have a certain belief” (Montmarquet: 45). In an analogy with carving,
Montmarquet argues that the exercise of care is not an indirect way of controlling the
outcome, “it is not a separable action by means of which one brings about careful movements
of one’s limbs,” it is rather something that is expressed in the work of carving, it is a modality
of this activity (Montmarquet: 46). Similariy, he wants to argue that the exercise of the virtue
of conscientiousness, for instance, is not separable from the act of believing, but is rather a
mode of believing, and hence that it constitutes direct rather than indirect control
(Montmarquet: 46).

His argument, however, is unconvincing. It would seem that we can distinguish the
exercise of these dispositions or virtues fiom belief. Thus the exercise of epistemic
conscientiousness, the careful investigation of a matter with attention payed to all relevant

information, would not seem to be a kind or mode of believing, but rather a way of going

74

about arriving at certain beliefs. These activities are not a way of believing, but rather a way
of “causing” or “producing” certain beliefs, or perhaps a certain kind of belief (e.g., justified
or true belief). As we will see in the next chapter, if Montmarquet maintained like Price that
beliefs were dispositions, then the cultivation of certain virtues or dispositions might be a form
of direct control. This is because on such an account there would no longer be a distinction
between the cultivation and exercise of virtue and certain beliefs; they would both be

identified with the same dispositions.

II. The Question of Viability

It now remains to be shown whether indirect voluntarism is actually an attractive or
tenable position. Some philosophers, as we have seen, accept this view in order to avoid
certain perceived conceptual and phenomenological difficulties with direct voluntarism, but
does indirect voluntarism in turn create new difficulties of its own? There does seem to be
a regress problem with indirect voluntarism that is potentially vicious if one wishes to hold
a substantive account of epistemic responsibility. This regress problem arises from the
indirect voluntarist assertion that belief is not subject to direct control, but only indirect
control (that our control over belief consists of our ability to do things like pay attention, or
employ certain other belief-acquisition methods such as gathering evidence). Our selection
of these methods, though, is directed and motivated by our background beliefs. For example,
one chooses the method of the double blind test for forming beliefs about the efiicacy of a

certain drug, because one has certain background beliefs about how one should arrive at

75

conclusions about empirical matters. Investigation, like observation, is theory-laden.3

How then can we be responsible for the beliefs that arise from our paying attention,
or our employment of other acquisition methods, if we are not responsible for, or cannot
control, the background beliefs that direct and motivate this indirect control? Even if some
of these background beliefs themselves are the result of earlier employment or exercise of
indirect control, at some point early on there must have been some beliefs that started ofi‘ our
indirect control which we were not responsible for (remember the claims of \Vittgenstein,
James, and Locke, that the child begins by believing). This is where the regress begins.
Either we are responsible for the background beliefs which determined a particular method
by being responsible for employing some earlier method, and are then responsible for that
earlier method by being responsible for some still earlier method which determined the later
one’s background beliefs, ad infiniturn, or there are some beliefs which are forced upon us,
some volitionally foundational or basic beliefs, outside our control and hence responsibility,
which break the regress, but in doing so undermine or mitigate our responsibility for the
methods we employ, and hence the beliefs that result.

James Montmarquet clearly recognizes this problem.‘ He discusses how indirect

 

3 Clifford seems to recognize this effect that background beliefs can have on judgments,
especially judgments concerning what to believe. Thus Clifford argues that “No real
belief, however trifling and fragmentary it may seem, is ever tnily insignificant; it prepares
us to receive more of its like, confirms those which resembled it before, and weakens
others; and so gradually it lays a stealthy train in our inmost thoughts” (Clifford: 503).

‘ Another thinker who seems to clearly recognize this regress problem is Thomas Nagel.
In discussing the issue of epistemological skepticism he states that one source of this
skepticism is the “consideration of the respects in which our beliefs and their relation to
reality depend on factors beyond our control. External and internal causes produce our
beliefs. We may subject these processes to scrutiny in an efl‘ort to avoid error, but our

76

voluntarists wish to argue that our responsibility for belief resides in certain acts or omissions
on our part that inhibit or produce belief. The problem with this view, Montmarquet claims,
is that our omissions or actions are often the result of certain background beliefs that we have.
He takes the example of a person whom we wish to hold culpable, on an indirect voluntarist
account, for not gathering more evidence:
Notice that typically in such an instance she will believe that no more evidence is required.
Thus...if we are to hold her culpable for omitting to look for more evidence, we must
hold her culpable for believing that she needn’t so look. Moreover, if the latter
culpability is to be made out in terms of some further action or omission - she failed to
check to see whether her belief that further checks were unnecessary was really justified -
I will argue that we are on the road to regress (Montmarquet: viii).

This difficulty is also apparent in Pojman’s account. He argues that “evidence is
always relative to a person’s individuating background beliefs, capacities to interpret data, and
expectations” (Pojman, 1985: 537). Pojman goes so far as to argue that this in turn
relativizes the notion of justification for belief. He maintains that justification has to do with
what it is reasonable for a person to believe given such factors as their noetic structure and
background beliefs (Pojman, 1985: 53 8). These ideas Pojman employs primarily to motivate
his arguments against direct voluntarism. He asserts that background beliefs can compel
assent, thus eliminating the possibility of direct control. He argues that “Given a whole

network of background beliefs, some views or theories are simply going to win out in our

noetic structure over others” (Pojman, 1985: 529). But though he may put these claims

 

conclusions at this next level also result, in part, fiom influences which we do not control
directly. The same will be true no matter how far we carry the investigation. Our beliefs
are always, ultimately, due to factors outside our control, and the impossibility of
encompassing those factors without being at the mercy of others leads us to doubt
whether we know anything. It looks as though, if any of our beliefs are true, it is pure
biological luck rather than knowledge” (Nagel: 281).

77

forward to argue against direct voluntarism, they have ramifications for the possibility of
attributing responsibility to our exercise of indirect control.

The case is roughly analogous to that of mitigated responsibility in ethical situations.
When assessing ethical or moral responsibility we will often take into account the background
beliefs of an agent. In situations where an agent performs a wrong act, we may still not hold
the agent responsible if the act was motivated by false beliefs which were none the less
reasonable for him to hold. I would claim similarly, that we may not hold an agent
responsible for a false belief, even if this belief is the result of indirect control that he
exercised, if this indirect control was motivated and directed by beliefs over which the agent
had no control - beliefs for which the agent is not properly responsible.

To draw another parallel with ethics, this problem is like that faced by Aristotle’s
account of moral responsibility or voluntary action. Aristotle allows that the perception of
the goodness or badness of a course of action may be relative to the constitution or character
of the perceiver (Aristotle: 30). This being allowed, he recognizes it has the consequence that
one may choose what is bad simply because, given a defect in one’s character, one perceived
the thing in question to be good. It would seem, then, that responsibility is completely
relative to one’s character or perception. One is only responsible for choosing what is bad
if one actually perceives it that way. Aristotle, however, disagrees. He argues that we are
in fact responsible for these misperceptions that arise fiom our character in that we are
responsible for the fomiation of our character in the first place. He argues that our character
is the result of our prior voluntary choices, and that as such we are responsible for it, and

therefore for the misperceptions it generates and the wrong conduct that then results

78
(Aristotle: 32).

Character plays the same role in Aristotle’s account of moral responsibility as
background beliefs do in the indirect voluntarist’s account of epistemic responsibility. For
both positions, responsibility stands or falls with whether we have control over these
background motivators or guides. Aristotle fully acknowledges this fact. He states that “if
each man is somehow responsible for his state of mind, he will also be himself somehow
responsible for the appearance; but if not, no one is responsible for his own evildoing”
(Aristotle: 32-33). Some argue that given the effects of one’s community and family on the
formation of character it is perhaps hasty to say that we are responsible for it (see Martha
Nussbaum’s discussion of “Luck and Ethics” in her work The Fragility of Goodness).
Similarly, given the largely forced nature of our background beliefs (as Pojman and others
maintain), it seems unlikely that we are responsible for all or even most of these beliefs. As

such, the argument for responsibility fails.’

III. Conclusion

Indirect voluntarism attempts to offer us an account of our control over belief that
allows for epistemic responsibility. It would seem, however, that responsibility on this
account is non-existent. On this account, whether an epistemic duty or responsibility has been
violated will always have to be determined relative to the agent’s background beliefs which

motivate and direct this control. But given the limited picture of voluntarism that this view

 

5 I am indebted to Prof. Richard Hall for pointing out this similarity between Aristotle’s
view of moral responsibility and the indirect voluntarist account of epistemic
responsibility.

79

presents, these background beliefs will be outside the agent’s control. As such, we will not
be able to argue that an epistemic agent should have performed difl‘erently, should have
exacted his indirect control in a different manner, due to his lack of control over the
background beliefs that directed his activity. We will not be able to say that the agent should
have seen or believed certain evidence to be relevant, or should not have believed some
method to be reliable, for he will not be responsible for the background beliefs concerning the
evidence or the method.

With this in mind it may be questioned whether indirect voluntarism is even
voluntarism at all. It would seem to be like the case with soft determinism where the attempt
is made to define freedom in such a way that it is consistent with causality. This attempt often
resulted in a definition of “fieedom” that was out of touch with how we ordinarily use or
apply this term, and which subsequently did not agree with our concept of responsibility (as
I argued with respect to the Desire Thesis in the Introduction). Similarly, the indirect
voluntarists have so limited our ‘yoluntary” control over belief in order to make it consistent
with their view of the causation of belief, that they have rendered it non-existent, and hence
incapable of supporting any substantive notion of epistemic responsibility.

Hence indirect voluntarism like non-voluntarism conflicts with our concept of a
rational agent. As we saw earlier, essential to this concept is that the agent be able to engage
in the activities of deliberation, reflection, assessment, etc., and these activities imply fi'wdom.
These activities also imply responsibility for they can be done better or worse, responsibly or
casually. Because indirect voluntarism rules out such epistemic responsibility, it conflicts with

our concept of rational agency. As such, our concept of rational agency would seem to

80

demand some direct control over belief so as to allow for epistemic responsibility. This

notion of direct voluntarism we will examine next.

CHAPTER FOUR

DIRECT VOLUNTARISM

It is only the will, or freedom of choice,
which I experience within me to be so great
that the idea of any greater faculty is beyond

my 87W;
Descartes: Fourth Meditation, Sec 57
the capacity so to govern himself: in thinking
and believing, and in the doing which is consequent
upon it, is the distinctive capacity of the human

animal.

C.I. Lewis: “The Philosopher Replies”

In the previous chapter I examined several arguments against direct doxastic
voluntarism which were meant to motivate adherence to indirect voluntarism. I showed that
indirect voluntarism falls short of advancing a robust account of our epistemic responsibilities.
I argued that some form of direct voluntarism is necessary in order to remedy this problem.
But in order to put forward such an account of direct voluntarism, it now becomes necessary
to reply to the definitional, conceptual, and phenomenological arguments raised against this
view by adherents to indirect voluntarism. This is the major task of this chapter. In this

chapter, I begin with a response to the definitional argument of HR Price, which argues

81

82

against direct voluntarism on the grounds that belief is a disposition and not an act. I begin
with this argument because it is the easiest to counter. Next, the conceptual or logical
argument of Pojman and VVrlliams will be examined. This argument is the most formidable
as it attempts to show the inconceivability of direct voluntarism. The phenomenological

argument will be addressed last.

I. The Definitional Argument

There are presently in circulation many different views or definitions of belief,
including representational, propositional, behavioral, and other accounts. What is the correct
view? Price argues that belief is a disposition (Price: 15), most likely a disposition to act or
behave in certain ways. This argument is probably motivated by the desire to define belief in
terms of some externally identifiable or testable criterion, such as a disposition to act, while
avoiding the difficulties associated with crude behaviorism which directly defines belief or
other mental states in terms of certain overt behaviors or actions.1 This attempt to define
belief in terms of certain dispositions creates new difficulties, however, in that it renders the
content of belief states indefinable. This is because the disposition that a particular belief state
is to be identified with could have an infinite number of manifestations given the peculiarities

of the circumstances one finds oneself in. As such, each belief state in this view would stand

 

‘ For instance, crude behaviorism has the difficulty that persons incapable of bodily
movement (e. g., those who are completely paralyzed) are excluded fiom having a wide
range of mental states associated with such movements, for in this view to have these
mental states just is to behave or act in certain ways. A dispositional view like that of
Price, though, avoids this problem in that it argues such persons still have the disposition
to behave in the appropriate ways, and thus meet the conditions requisite for having these
mental states.

83

for an infinite number of manifestations, and hence could not be completely understood to the
extent that our understanding depends on definition. This problem is similar to that faced by
those logical positivists who attempted to define empirical statements in terms of their
experiential ramifications. Every empirical statement was seen to entail an infinite number of
experiential statements, and as such became indefinable. This view also suffers from the odd
consequence that it renders beliefs unknowable via introspection. Given that belief is a
disposition, one would have to discern one’s beliefs by examining the ways in which one is
disposed to behave. One would thus discern one’s own beliefs in much the same way that one
would discern the beliefs of others; namely, by examining behavior. Doing this may indeed
be necessary at times when we are uncertain what we believe, but it does not seem to be
required at all times. It does seem that in general we can have an immediate introspective
awareness of our beliefs. Given these difliculties there would not seem to be much favoring
acceptance of Price’s view over the competition.

In addition, contrary to Price’s argument, a dispositional view of belief need not be
inconsistent with direct voluntarism. As we saw in the previous chapter, at the heart of the
distinction between direct and indirect voluntarism is the notion of methods, procedures, or
activities, that are distinct from belief. Indirect voluntarists maintain that such methods are
requisite to control belief and direct voluntarists maintain that belief can be directly willed
independent of such methods. The distinction between these positions depends, then, on
there being a distinction or difi‘erence between willing a belief and willing something other
than a belief (e. g., some physical action that will result in belief). If belief is propositional,

then there certainly is a difference between believing and willing certain physical actions. If

84

belief is dispositional, however, this is no longer clearly the case. Ifbelief is a disposition to
act, the way that one would will oneself to believe is by cultivating the relevant disposition.
In this case one is not cultivating the disposition in order to arrive at some belief, rather the
disposition just is the belief, and hence its cultivation is the only way that one could will the
belief. For example, the only way that I could will belief in God on this account is by
cultivating in me the disposition that this belief is identified with. This disposition would not
then be a means to believing in God, such as with Pascal where cultivating the disposition to
pray or attend church is a means to belief; rather the disposition would be the belief (belief
in God would just be the disposition to pray, etc). As such, even if we accept Price’s
dispositional account of belief, this would not force us to abandon direct voluntarism. If
anything, such a view of belief would seem to render direct voluntarism more plausible, for

most of us at one time or another have cultivated or willed a disposition.2

II. The Conceptual Argument

As we saw in Chapter Three, the conceptual argument is based on the idea that direct
voluntarism involves the severing of the necessary or conceptual connection between belief,
on the one hand, and truth or evidence, on the other, and in doing so renders itself incoherent.
Pojman states that direct voluntarism “is conceptually confused because it neglects the
evidential aspect of conscious belief acquisition and sustainment” (Pojman, 1985: 53 5).

Williams in turn states that, because of the truth-oriented nature of belief, “With regard to

 

2 I am indebted to Prof. Richard Hall for pointing out this way of reconciling Price’s view
and direct voluntarism.

85

no belief could I know - or, if all this is to be done in fiill consciousness, even suspect - that
I had acquired a belief at will” (\Villiams: 148). In the discussion to follow, I will focus
primarily on Pojman’s argument for this claim, as his is the most detailed.
Poj man repeatedly argues that direct voluntarism requires that beliefs be formed
“without regard to truth considerations,” that the believer must purposely disregard the
evidence (Pojman, 1985: 533, 535). He argues that the notion of choosing or willing to
believe would have to involve the confusion that “willing makes it so,” for such voluntary
control over belief would amount to believing something true simply on the basis of its being
willed (Pojman, 1985: 533, 534). Finally, he argues that in a direct voluntarist account,
because of the disregard for the evidence or truth that it involves, belief would not be about
states of affairs but about our desires (Pojman, 1985: 534). Pojman’s view of direct
voluntarism is dominated by one particular image, that of a mental finger being placed on the
epistemic scales. He sees direct voluntarism as having to involve the altering of evidence to
produce the desired belief, and as such it must involve a disregard for evidence, or more
specifically, a disregard for the actual evidence (Pojman, 1985: 526-527). He states that
In the standard model of belief acquisition, the judgment is not a separate act but simply
the result of the weighing process. It is as though the weighing process exhibited the state
of evidence, and then the mind simply registered the state of the scales. In the volitional
model, the judgment is a special action over and above the weighing process. It is as
though the mind recognized the state of the scales but were allowed to choose whether
to accept that state or influence it by putting a mental finger on one side or the other,
depending on desire (Pojman, 1985: 526—527).

Pojman’s position is puzzling. His discussion of the epistemic scales and mental finger makes

it seem as if he is describing a kind of indirect voluntarist position in that the will does not

seem to be able to effect belief directly, but rather can effect it only by altering the evidence.

86

It is the evidence that seems to produce the desired belief. But this confusion aside, let us
look more closely at Pojman’s claim that direct voluntarism must in some sense involve a
disregard for truth or evidential considerations.

Why must direct voluntarism involve the will producing belief independently of
evidence? I think that this position is the result of conflating reasons and causes. Pojman’s
(as well as Williams’) causal account of belief views evidence or reasons as a causal factor in
belief production (\Villiarns: 141-143/Pojman, 1985: 530, 534). One then need only go on to
consider what are the two general possibilities concerning the relationship between the will
and evidence in this production - either the will regards evidence in the production of belief
or it doesn’t. If the will fails to regard the evidence then we arrive at the conceptual
confiision discussed above. But if the will regards evidence in the production of belief, then
it would be considering a causal factor. Through this consideration, the belief produced
would then become susceptible to the causal influence of this factor, and hence, the belief
would be the product of multiple causes. The will, then, could no longer be considered the
sole cause in the production of belief. Thus, at most, the will could be counted as a
contributing causal influence. This might be enough for responsibility except that Pojman
views the causal role of evidence as being decisive (Pojman, 1985: 534). As such, any direct
influence of the will becomes unnecessary. He states that the “rational believer, in full
consciousness, would see that there must be a truth connection between states of afl‘airs and
the belief by virtue of which the belief is true, so the will is essentially unnecessary for the
belief” (Pojman, 1985: 534). On this causal model, then, if the will regarded evidence we

could no longer speak meaningfully about direct voluntarism. We could no longer speak

87

about the will as having a significant direct causal role in the production of belief, a role such
that the absence of its direct influence would significantly alter the causal picture.

But isn’t this confusion of reasons and causes misguided? I would like to argue, as
does Jaegwon Kim, that reasons qua reasons do not have a role in causal explanations.
Reasons are justifiers, they are not causes (Kim, 1984: 314). Reasons serve a role in
normative explanations, not causal explanations. Thus, just as explanations involving reasons
in ethics do not purport to describe existing states of affairs and a nomological or causal
connection between them, but rather attempt to show how a given action is justified in light
of the beliefs and desires of the agent and the rules or norms of morality, similarly, reasons
in epistemology do not serve a causal role, but rather serve the role of showing how beliefs
(or perhaps even actions) are justified given certain norms of rationality (Kim, 1984: 314).3
In so far as a causal view of reasons conflicts with our conception of rationality, it would
seem that this non—causal or normative view should be favored. If one does reject the causal
view in favor of the normative or non-causal, as I do, then the notion that direct voluntarism

must involve a disregard for evidence falls away.

III. The Phenomenological Argument
The phenomenological argument is perhaps the least formidable in that it hinges on

the question of what is an appropriate interpretation of our subjective experience of belief

 

3 Richard Taylor puts forward such a non-causal view of reasons when he states that “In
the case of an action that is both free and rational, it must be such that the agent who
performed it did so for some reason, but this reason cannot have been the cause of it”
(Taylor: 384).

88

formation. First, this weakens the argument because there seems to be no unanimous
agreement as to the exact nature of this experience. If the thinkers we examined in the earlier
historical overview of voluntarism can be seen as drawing their positions from a certain
perspective on the phenomenology of belief acquisition, then we can see from this overview
that there is much disagreement as to the exact extent of our control over belief. Even
persons who hold the same type of position disagree. Thus Locke, Heil, Price, Williams, and
Pojman, are all indirect voluntarists, and yet they disagree as to the nature of our control over
belief. Pojman and Locke, as we saw in Chapter Three, argue that our control is primarily
indirect, but that we also experience a kind of direct negative control, or “veto power,” over
belief. Heil, Price, and Williams, on the other hand, contend that no such negative control is
present. What is interesting to note is that the participants in this disagreement even share
many of the same premises. Thus as we saw in this chapter, Williams and Pojman both adhere
to the notion that evidence is a causal factor in the production of belief, and that because of
this direct voluntarism must involve a disregard for evidential considerations.

Secondly, the phenomenological argument is weakened by its reliance on subjective
experience in that this forces its proponents to claim that any experience which may suggest
some direct influence of the will on belief must involve some confusion, illusion, or mistake
on the part of the believer. Thus Pojman, in discussing cases of indifi‘erence, where we feel
that we may actively choose what to believe, wishes to explain this experience by appealing
to unconscious processes ‘Vvhich play a decisive role in belief formation, but... are not things
we have direct control over” (Pojman, 1985: 530). This call for re-interpretation is similar

to that made by determinists concerned with action, such as d’Holbach, who likewise suggest

89

that the “fee ' g” of free will is based on some confusion or ignorance. D’Holbach states that:
As the mechanism of these sensations, of these perceptions, and the manner they engrave
ideas on the brain of man, are not known to him; because he is unable to unravel all these
motions; because he cannot perceive the chain of operations in his soul, or the motive
principle that acts within him, he supposes himself a free agent; which literally translated,
signifies, that he moves himselfby himself, that he determines himself without cause; when
he rather ought to say, that he is ignorant how or why he acts in the manner he does
(I-Iolbach: 366).
Just as there are many experiences which suggest freedom of action, there are many doxastic
experiences which seem to suggest direct control of belief. Certainly in the case of many of
our theoretical beliefs, where these beliefs are the product of deliberation or reflection, we
often feel that we have, on the basis of the evidence, chosen or decided what to believe. Even
in the case of perception we often feel that we can directly control belief, or at least the
suspension of belief. Thus when we encounter perceptions that appear odd or incoherent,
we can, through the exercise of our will, override belief. An example of this would be the
water mirages that one often encounters while driving along a paved road on a sunny day.

When the conditions are totally unsuitable for such large puddles to be in the road, one can

decide in these circumstances not to believe one’s eyes.‘ Given the lack of unanimity in terms

 

‘ Lawrence Bonjour may provide a framework that can make sense of this voluntarist
aspect of perception, along with our experience that most of the time seeing does just
seem to be believing [e.g., Berkeley: “whatever power I may have over my own thoughts,
I find the ideas actually perceived by sense have not a like dependence on my will”
(Berkeley: 99-100)]. Bonjour appears to suggest that as rational believers we accept a
kind of meta-belief which attributes a high degree of reliability to our perceptual beliefs
under specified conditions. This meta-belief, then, renders even perceptual beliefs
inferential in that they are in part justified or warranted by virtue of their being subsumed
under this meta-belief (Bonjour: 233). Given this meta-belief; perceptual beliefs which fall
under it are accepted unproblematically unless they come into conflict with other beliefs
that we hold, i.e., unless the perceptual beliefs are incoherent with our background beliefs.
Such an instance of incoherency would be like that discussed above - a puddle

appearing on a flat road on a dry day. In such situations we must then decide whether to

90

of the nature of our experience as believers, given the fact that we often do feel as if we have
direct doxastic controL and given the conceptual argument against direct voluntarism can be
shown to be invalid (i.e., if one accepts a normative rather than causal account of reasons or
evidence), what motivation do we have to re-interpret our experience as the
phenomenological argument suggests? With the failure of the conceptual argument one can
no longer claim that such re-interpretation is requisite in order to render our experience

coherent.

IV. Direct Voluntarism and the Norms of Rationality

Let us examine, then, a different picture or image of the role of the will in belief, a
picture which is far different than that of Pojman’s mental fingers and epistemic scales. We
might argue that the human mind finds itself endowed with certain norms of rationality.
These norms we might say constitute its epistemic conscience, the analogue of the moral
conscience which consists in certain norms of morality. These epistemic norms may be the
result of socialization, or they may be hardwired. The question of the origin of these norms

would be almost exactly analogous to the question concerning the origin of moral norms, e.g.,

 

reject enough of our background beliefs to resolve the incoherency or to reject the
perception; we must in such instances decide what to believe. This model would seem to
imply a great deal of voluntarism in regard to perceptual beliefs. Perceptual beliefs are
evaluated in light of this meta-belief as well as background beliefs, and are either accepted
or rejected on the basis of them. This model seems to explain the notion of seeing being
believing as well. Given the meta-belief concerning the reliability of perception, in the
absence of incoherency, perceptual beliefs are readily accepted. Roderick Chisholm
attributes such an inferential view of perception to CI. Lewis, where this inferential nature
of perception surfaces in the defense of challenged perceptual beliefs (see Chisholrn’s
“Lewis’ Ethics of Belief”). I am indebted to Prof. Nicholas D. Smith for this connection
to Bonjour, and to Prof. Richard Hall for the Berkeley reference.

91
whether these norms are in some sense innate or socially produced. I believe that an account
similar to that which Kant puts forward concerning our relationship to the moral law would
provide an interpretation of our relationship to these epistemic norms that would be most
consistent with doxastic voluntarism. This is the avenue that I will now explore.

Kant wishes to reconcile freedom or autonomy with the idea of being governed by
law. In doing so he settles on an account of freedom which involves agents operating under
laws that are self-imposed. Consequently, when Kant discusses the freedom or autonomy of
the will in the Groundwork he describes it as “the property the will has of being a law to
itself” (Kant: 108). Later in the Groundwork, Kant argues that the freedom which reason
necessarily attributes to itself involves the notion that it “must look upon itself as the author
of its own principles independently of alien influences” (Kant: 116). As one commentator on
Kant’s position has suggested:

to say that a moral agent acts freely does not mean that such an agent acts lawlessly. In
a lawless (i.e., completely ruleless) world, anything would follow fi'om anything, and that
would make the notions of causality in general and of free, moral agency in particular
totally meaningless...To say that a moral agent is one who acts freely must therefore mean
that such an agent can exercise causal power on the basis of a law or laws given by his
own reason alone” (Sullivan: 47-48).

Applied to the norms of rationality this view takes the following form: these norms
are self-imposed in that we fieely accept them as constraints on our epistemic agency. As we
have just seen this is much the same with moral norms. No matter what psychologists or
sociologists tell us about the origin of these norms (i.e., how they have come to find their way

into our consciousness), as moral agents we must view ourselves as fieely submitting to their

constraint. As moral agents we must view ourselves as capable of rejecting the norms which

92

govern us, and not as being strictly determined to hold them. The process of moral reflection
often consists in an evaluation and selection of these norms. Similarly, as epistemic agents
much of our reflection is devoted to trying to discern what rationality requires of us. Our
reflection here centers on the question of what rational norms we should submit to. It may
be that our conception of rationality limits our possible choices of rational norms, just as our
concept of morality limits our possible choices concerning moral norms. We are limited by
these concepts in that they do, to some extent, dictate exactly what we can recognize as being
a moral or rational norm. Thus, our concepts of morality and rationality are not without their
bounds beyond which “norms” become unrecognizable as moral or epistemic. This does not
mean, though, that we cannot still fieely impose these concepts and norms on ourselves. As
we saw earlier, freedom does not necessitate multiple alternatives; one can freely choose a
given course of action even if it is the only one open. It may, after all, be the thing one wishes
to do. Though space prohibits me fi'om discussing this here, I do believe that there is within
these concepts much room for alternatives. In the epistemic case alone there is much
discussion over whether our norms are all synchronic or whether some are diachronic or
fiiture oriented.’

In discussing how we hold these norms, however, one must not lose sight of their
function which is to constrain the will nonnatively in its choice or production of belief. These

norms serve to attune the will to evidential or truth factors in this choice.6 The will, then, in

 

5 See Hall and Johnson: “The Epistemic Duty to Seek More Evidence”.

6 Kim argues that a rational being is one “whose cognitive ‘output’ is regulated and
constrained by norms of rationality” (Kim, 1988: 33 5). Kim views this constraint as
evidential. He argues that beliefs are constrained in terms of their contents, and that a

93

its deliberation and choice of belief does not ignore evidence; it in fact employs evidence as
a way to justify this choice, as a way to fit its influence within the norms of rationality that in
a sense regulate it.

As with the norms of morality, one may feel a conflict between the norms of
rationality, on the one hand, and desire on the other. We may feel a conflict between what
we want to be believe and what we ought to believe, given our rational norms, just as we may
feel a conflict between what we want to do and what we ought to do, given our moral norms.
The latter conflict is often discussed. Kant, in the Groundwork, states that: “Man feels a
powerful counterweight to all commands of duty presented to him by reason as so worthy of
esteem - the counterweight of his needs and inclinations” (Kant: 73). Similarly, Campbell
discusses situations where “desire and duty are at odds” (Campbell: 398). In the epistemic
case, this conflict is perhaps less often discussed, but it is discussed none the less. Clifford,
in his “Ethics of Belief,” examines at length the danger of giving into to desire and believing
something simply because one wants to (see Clifford’s discussion of the shipowner, Clifford:

502). Richard Feldman, in his article “Epistemic Obligations,” likewise discusses how what

 

belief “has the content it has in part because of its location in a network of other beliefs
and propositional attitudes; and what at bottom grounds this network is the evidential
relation, a relation that regulates what is reasonable to believe given other beliefs one
holds” (Kim, 1988: 335). He goes on to maintain that without this constraint we cannot
intelligibly interpret a cognizer’s outputs as beliefs. He states that “there is a sense of
‘rational’ in which the expression ‘rational belief is redundant; every belief must be
rational in certain minimal ways” (Kim, 1988: 335-336). He states that “unless the output
of our cognizer is subject to evaluation in accordance with norms of rationality, that
output cannot be considered as consisting of beliefs” (Kim, 1988: 336). CI. Lewis seems
to have held a similar view. Roderick Chisholm argues that in Lewis’ view the rationality
of belief arises fi'om it conformity with the norms of the “ethics of belief.” He states that
in Lewis’ account, “when a man fails to conform to the ethics of belief he is, ipsofacto,
behaving irrationally” (Chisholm, 1968: 227).

94

we ought to believe from a prudential standpoint may conflict with what we ought to believe
from the standpoint of epistemology (F eldman: 545). Examples also abound in our everyday
experience. We have probably all encountered persons, perhaps even ourselves at times, who
believe against the evidence because it conflicts with what they want to believe - for example,
persons who contrary to strong evidence that they are ill none the less choose to believe that
they are fine, persons who persist in believing their spouse to be faithfiil even in the presence
of damning evidence, persons who choose to believe one side of an issue simply because they
cannot bear to sit on the fence.

The will, then, must struggle against desire in order to will belief in accordance with
these norms of rationality, just as it must struggle in order to will actions in accordance with
the norms of morality. In the moral case, Campbell discusses the freedom of the will in
regards to such effort. He states:

There is X, the course which we believe we ought to follow, and Y, the course toward
which we feel our desire the strongest. The fi'eedom which we ascribe to the agent is the
freedom to put forth or refrain from putting forth the moral effort required to resist the
pressure of desire and do what he thinks he ought to do (Campbell: 397).
Similarly, in the epistemic case, the will must exert a degree of force. However, in this case
the force exerted is to some extent the product of the resistance, if any, presented by the
norms of rationality or epistemic conscience, and the temptation or pull of desire. Thus,
where the evidence is clear or evident, there is no resistance fi'om the norms of rationality,
belief is easily justified, and in the absence of desires to the contrary, belief can be easily

willed. It is probably cases like these that make us view clear or “compelling” evidence as

simply producing belief, because so little effort of the will is required. But where desire is

95

strong, and the evidence less clear, we find ourselves struggling with belief. As with the
analogous case in ethics, we find ourselves struggling with what we ought and what we want
to believe. It may, in these cases, take great effort to believe, or suspend belief, as rationality
requires. In cases of indifference we may feel the power of our will more clearly in that we
are able to put forward equally strong reasons on either side; as such, neither side requires a
greater effort of will to be believed (assuming of course an absence of desires favoring one
side over the other). This equality of efi‘ort in turn makes us feel that belief is up to us. The
fact that neither choice is more difficult truly places our will at center stage.

These norms may also have a greater or lesser hold on us, thus resulting in differences
in terms of the amount of effort required to will belief. A person who is a devout scientist,
a person whom we might think strongly adheres to these norms, might require little efi‘ort to
combat desire. Such a person would be analogous to an extremely moral person for whom
doing the right thing comes almost naturally. Conversely such an epistemically “good”
person would need to exert a great deal of effort to believe contrary to the dictates of
rationality. This would be like their moral counterpart, who might need to exert a tremendous
effort to do something that is morally wrong (e. g., such a person may find it very difiicult to
lie even when doing so may help someone they care about). Of course, the looser the grip
these norms of rationality have on a person, the more irrational the beliefs of this person will
be. Such an irrational person would be like Pojman’s description of the direct voluntarist:
they would believe simply on the basis of desire with a total disregard for truth considerations.
A person on whom these norms have some hold, however, does not disregard evidence in the

production of belief. Such a person uses evidence to provide reasons in order to ground,

96

justify, or fit his choice of belief within the norms of rationality. For such a person Pojman
would be right in asserting that the will alone cannot cause belief (Pojman, 1985: 534), for
such a person will find themselves constrained as rational agents to provide reasons to justify

their choice of belief.7

V. Conclusion

The position outlined above may seem too fantastic. It may be argued that direct
voluntary control over belief suggests a kind of mind/body dualism that has long been
abandoned. Here, it may seem, is presented an image of the epistemic agent as completely
outside the causal order, as having a fiee will in the sense that it is able to act independently
of any external causes. The ramifications of this would seem to be serious indeed: the
violation of causal closure, the ruling out of any possibility of constructing a neuro-
physiological theory of mind or of reducing mental phenomenon to physics, and so on.

I wish now to allay these fears. My arguments are not aimed at making any
metaphysical assertions. They are rather aimed at mapping out the implications of our
concept of rational agency for doxastic voluntarism and epistemic responsibility. I leave open
the question of whether we are in fact subject to and determined by causal laws. What I wish
to argue is that given our concept of ourselves as rational agents, we cannot coherently assert

or believe that all belief is determined by causal laws. As rational agents we must conceive

 

7 Matthias Steup seems to concur with this intuition concerning rationality, evidence, and
voluntarism. Steup argues that “No matter how grim the circumstances are, if an agent
holds a belief contrary to the evidence, it is within his power, given that he is a rational
agent, to reflect upon his belief and thereby find out that he had better withhold it, or even
assent to its negation” (Steup: 78).

97
of ourselves as outside the causal order, we must conceive of ourselves as capable of
complete, direct doxastic voluntarism.

In some respects my position is like that of Kant, or even, for that matter, like Dennett
in his later work (see Chapter Two, Sec. Hb). Both argue that we hold two incompatible
views of ourselves that we move between at different times.8 For Dennett, these views are
the manifest and the scientific, and for Kant, they are the intelligible and the sensible. I would
term them “the normative” and “the causal.” In the Groundwork, Kant argues that depending
on which view we take, we will see ourselves as being subject to different laws. When we
view ourselves as sensible beings we see ourselves as subject to natural or causal laws, as
subject to the same concept of causality as all other appearances or members of the sensible
realm (Kant: 120). But when we view ourselves as intelligible beings this changes. Now we
must view ourselves as free, as not governed by the causal laws of the sensible realm. Kant
argues that

As a rational being, and consequently as belonging to the intelligible world, man can never
conceive the causality of his own will except under the Idea of freedom; for to be
independent of determination by causes in the sensible world (and this is what reason must
always attribute to itself) is to be free (Kant: 120).
Similarly, Dennett discusses how we waver “back and forth between a practical - even,
perhaps, optimally practical - way of thinking of the world, and an impractical but still

rationally endorsed vision” (Dennett, 1984: 114). The former is the scientific image in which

 

’ Campbell also seems to hold this view that we are subject to two incompatible images of
ourselves. For Campbell they are the practical and the theoretical. He discusses the
notion that there is “a radical conflict between the theoretical and practical sides of man’s
nature” - the theoretical side being that which adheres to determinism, and the practical
being that which asserts man’s freedom (Campbell: 396).

 

WWW—“‘-

98

we view the world in causal temrs, the latter, the manifest image in which we view ourselves
as rational beings capable of deliberation, and hence as free or outside the causal order. As
rational beings, Dennett argues, we must conceive of determinism as being false, we must “act
as if the world really does have an open firture” (Dennett, 1984: 115). This, of course, does
not mean that determinism actually is false. The question of the truth of determinism is left
open. All that this means is that as rational beings we must conceive of ourselves as free.
We employ these different views of ourselves in the explanation of our beliefs.
Sometimes we explain our beliefs rationally and other times causally. We can see the
employment of both of these views in the explanation of religious belief. Thus some persons’
religious beliefs are seen by both themselves and others as the product of inquiry or reflection.
These persons examined different religions and then eventually decided which to believe.
Others’ religious beliefs, however, are seen as the product of their upbringing, and so the
beliefs were never chosen. The explanation of religious belief in these two cases would
correspond to the two views discussed. In the first case, religious belief would be explained
rationally. The person would discuss how, given their view of the world, they chose to
believe a certain religion because it made sense or because it “revealed” certain truths to them.
In the second case, the person would indicate that the religious beliefs in question were the
product of their childhood, that they were not chosen, but rather caused by their socialization.
The person would not consider these beliefs to be rationally held. The beliefs would be seen
as forced, as the remnants of their upbringing. As Dennett suggests, then, we recognize
ourselves as not being perfectly rational agents. We recognize that some of our beliefs are

better explained causally. These causally explained beliefs, however, we do not consider to

 

 

 

99

be rationally held, for to be rationally held, our beliefs must be the product of deliberation and
choice.

We are subject to a kind of schizophrenia. We employ two incompatible images of
ourselves and the world at difl’erent times. When viewing the world through our normative
lens, we see ourselves as self-determining, as having the kind of fi'eedom that is requisite for
legal, moral, and epistemic responsibility. When viewing the world through our causal lens,
we see ourselves as subject to the same causal laws as the rest of the universe, and seek to
explain our behavior in terms of antecedent conditions or causes. In this latter image, we are
no longer rational or moral agents; we are no longer free. The irony is that, as such, we
cannot coherently assert the position that this image maintains. By being removed fi'om the
normative sphere we no longer possess the kind of freedom that such an assertion would
require. As members of the normative realm we cannot assert the position of the causal view
without undermining the very freedom that is the ground for such an assertion. These views
must remain separate. The incompatibility of these views is established, their truth remains

a different issue, an issue which I will leave for others.

 

If '.
.‘l

 

CHAPTER FIVE

EPISTEMOLOGY AND ETHICS

The principles of right believing are like the
principles of ethics - or, rather, like the other
principles of ethics.

Roderick Chisholm: “Lewis’ Ethics of Belief”

Throughout this dissertation I have drawn examples from ethics in order to illuminate
certain epistemic positions and problems. I did this because, I believe that for the most part,
our intuitions in ethics are much stronger than those in epistemology. This would seem to
be especially true with the concepts of responsibility and obligation. Our sense of these
concepts is mostly the product of our ethical or moral experience, and hence it is no surprise
that it should be strongest in this domain. This does not mean, however, that the study of
epistemology has nothing to offer to the clarification of certain issues in ethics. In this
chapter, I would like to examine the ways in which ethics and epistemology can be used to
illuminate one another. I will do this by exploring parallels in the terrains of these two

disciplines.

100

10]

I. Naturalism

One important parallel between ethics and epistemology is the application of
muralism to these realms. Naturalists hold that the furniture of the world is made up of only
those firings recognized within physical or natural science, and that as such this firrniture can
be properiy described using only “natural” or “factual” terminology. Hence, naturalists strive
to define normative terms without recourse to normative language. In ethics, a prime
example of this position would be classical utilitarianism where goodness is defined in terms
of pleasure or happiness. Here, then, the normative term “good” is defined without the use
of other normative terms. It is defined using only non-normative or natural terms - terms
which designate certain factual or natural states of afi‘airs, such as “pleasure” or “happiness.”
Another example of this approach would be Gilbert Harman’s identification of statements
employing the normative term “ought” with factual statements concerning motivational
attitudes. Harman argues that to make a person the subject of an “ought” statement or
judgment is to ascribe certain motivational attitudes to him. Thus to say the “X ought to do
Y”, is to say that “X has motivation or reason to do the prescribed action” (Harman, 1975:
34-3 5).

With G.E. Moore’s attack on ethical naturalism, or what he termed the naturalistic
fallacy, naturalism in ethics lost much of its popular appeal (although not all, as the Harman
example illustrates). Moore showed the inadequacy of attempts to define the normative or
ethical term “good” in naturalistic language. In the many variations of his “open question
argument,” Moore showed that such naturalistic definitions failed to capture the complete

meaning of the normative terms which constituted their definiendum. He asked in one

 

 

102

variation: if we were confronted with the definition “the good is pleasure,” would we consider
the following proposition “pleasure is not good” to be a contradiction like the proposition
“the good is not good”? If the answer is no, and Moore certainly thought it was, then the
naturalistic definition failed. It did not produce a synonym or capture the complete meaning
of the normative term involved. Moore thought that all such naturalistic definitions would
fail this test. He believed that the ethical concept “good” was simple or unanalyzable, and
hence indefinable with any temrs. Moore states:

The most important sense of ‘definition’ is that in which a definition states what are the

parts which invariably compose a certain whole; and in this sense ‘good’ has no definition

because it is simple and has no parts” (Moore: 360).

The difficulty with the naturalistic endeavor, as Moore saw it, was that it attempted
to discover what ‘good’ means, to state its essence or necessary and sufficient part(s), which
is impossible given its simple nature. Moore states that “no difficulty need be found in my
saying that ‘pleasure is good’ and yet not meaning that ‘pleasure’ is the same as ‘good’, that
pleasure means good” (Moore: 362). With the emergence of this view in ethics, two principal
responses arose. Ethical theorists either came to doubt whether ethical statements had any
factual or cognitive content (developing non-cognitivism as a response to Moore’s non-
naturalism), or they came to see questions concerning the definition of ethical terms as
concerning the notion of priority within the circle of normative or ethical terminology. With
the latter response, the principal question became whether “the right” is prior to “the good” --
whether it would be defined in terms of “the good” or vice versa. For example, Rawls in his
A Theory of .histice argues that the priority or primacy of the right over the good is a central

feature of his conception (Rawls: 32). In general, attempts to define normative terms in

 

103
naturalistic or non-normative language were all but abandoned in ethics.
Ifnaturalisrn came to wilt in ethics, it has come to flourish in epistemology. Here, the
post-Moore naturalistic enterprise, however, has taken a new tack. There is, in a sense, an
admission of the success of Moore’s challenge to naturalism -- naturalistic analysis would no
longer attempt to capture the essence or complete meaning of normative terms, but would
rather work to clarify our understanding of the very properties, or in some cases processes,
that these terms were meant to refer to. The attempt to capture meaning completely, to
analyze normative concepts into their necessary and sufficient part(s), no longer takes center
stage. Goldman’s preface to his own attempt to put forward a naturalistic analysis of
“justified belief’ illustrates well this shift in emphasis. Rather than attempting to analyze the
concept of epistemic justification, to define this notion using only naturalistic terms, Goldman
opts for what he calls an “explanatory theo ”of epistemic justification (Goldman, 1979: 292-
293). Such a theory would not seek to produce a synonym, or capture the complete meaning
of the concept of epistemic justification, but would instead attempt to specify the natural or
physical conditions that must obtain for a given belief to be justified. He states:
The term “justified”...is an evaluative term, a term of appraisal. Any correct definition or
synonym of it would also feature evaluative terms. I assume that such definitions or
synonyms might be given, but I am not interested in them. I want a set of substantive
conditions that specify when a belief is justified (Goldman, 1979: 293).

The naturalist’s goal, then, is not to analyze the meaning of a given temi, but rather to

investigate empirically the actual nature of the property or process to which the term purports

to refer. This constitutes a departure fi'om the idea of capturing essences or complete

 

 

104

meanings, as Moore saw the naturalistic endeavor to be.1

This shift in emphasis does seem to protect naturalism in epistemology fiom the burnt
of Moore’s challenge. This does not mean, however, that epistemological naturalism can be
carried out unimpeded. The position suffers from its point of focus - epistemological
naturalism is agent-focused unlike ethical naturalism. By first examining ethical naturalism,
this difference in focus becomes clear.

The focus of ethical naturalism is not on ethical or moral agents, but on the analysis
of certain moral properties. The main question is whether these properties are capable of
naturalistic definition, that is, whether they can be properly identified with certain natural or
(in the weaker version) non-moral properties. In this respect moral properties are treated
much like physical properties - as entities which exist independently of agents. Along these
lines, Moore even draws a comparison between the moral property of goodness and the
physical property of yellowness. He argues that good is a property like yellow, in that both
are indefinable, the difference between them being that the latter is a physical or natural
property unlike the former (Moore: 362).

This focus on ethical properties, and the treatment of these properties as existing

independently of agents, has led to problems for ethical non-naturalism, the major problem

 

' Hilary Putnam, in his article “Psychological Predicates,” nicely characterizes this shift in
emphasis that post-Moore naturalism involves. Putnam argues, contrary to those who
would adhere to a Moorean style position (where the analysis of psychological predicates
aims at the production of synonyms), that we should “allow statements of the form

“pain is A,” where ‘pain’ and ‘A’ are in no sense synonyms. . . to see whether any such
statement can be found which might be acceptable on empirical and methodological
grounds” (Putnam, 1967: 226). The shift, then, is away from the analysis of meaning,
toward empirical analysis or investigation.

105

in this regard is that of interaction. David Brink discusses this problem in his “Argument from
Queemess.” Brink argues that if moral properties or facts are non-natural “entities” which
exist independently of us, it becomes difiicult to discern how we as physical or natural beings
can interact with these non-natural properties so as to gain knowledge of them, and be
properly motivated by them (Brink: 425-428). Thus if ethical naturalism is plagued by
Moore’s open question, ethical non-naturalism suffers from problems with interaction that are
reminiscent of Cartesianism.

Episternological naturalists, on the other hand, have the agent as their primary focus.
In epistemological naturalism it is the epistemic agent and his relationship with the world that
is at center stage. This position seeks to identify the relationships between agent and world
that will result in justified belief and knowledge. As we saw earlier, this relationship is seen
in causal terms. Naturalistic epistemologists attempt to identify justified belief and knowledge
with the causal processes or mechanisms in the agent that produce belief. This is a logical
step. As Quine admonishes us, wouldn’t it be best to examine how belief actually comes
about? - “why not just see how this construction really proceeds?” (Quine: 75). Similarly,
Goldman in his pursuit of an explanatory theory of justified belief wants to explore the
question of why a given belief is justified (Goldman, 1979: 293). For him, this necessitates
inquiry into why a given belief is actually held, and this means an examination of what causally
initiates and sustains belief (Goldman, 1979: 297).

Now we are in a position to see why this difference in focus poses difiiculties for
epistemological naturalism. As was discussed in the previous chapter, this emphasis on a

causal account of belief is in conflict with our concept of rational agency. Rational agents are

106
agents who are capable of fulfilling certain epistemic responsibilities in regards to belief. The
deterministic nature of naturalistic explanation in epistemology renders this impossible. Thus
epistemological naturalism, like determinism, becomes a position that cannot be rationally
asserted or held. Epistemological naturalism, then, sufi’ers not from Moorean open questions,
but rather from what it holds at center stage. By attempting to view the relationship between
epistemic agents and the world in terms of natural or causal processes, naturalistic

epistemology renders itself incoherent in the sense that it cannot be rationally asserted or held.

11. Teleology and Deontology: Virtue, Consequences, and Duty

Though there is currently a great deal of variety in terms of ethical positions, the
majority of these positions are still classifiable as either consequentialist, deontological, or
virtue based. Positions in epistemology are usually not classified in this manner, but there are
none the less epistemic positions that would correspond to this taxonomy, and classifying

these positions in this way could shed light on old conflicts.

a. Virtues

Virtue based ethical theorists hold that the rightness or moral appropriateness of
actions is a fimction of their being the result of the proper exercise of certain moral virtues
or habits. In this view, one might say that the act of giving to charity is morally appropriate
if it arises out of the proper exercise of the moral virtue of benevolence, rather than the
exercise of the vice of greed (e. g., to get a tax write off). Virtue-based ethical theories are

usually teleological in that something is commonly identified as being a virtue by its being

107
conducive to a certain end, telos, or good, that is deemed to be ethically or morally valuable.
In Aristotle’s view of the moral virtues, for example, the virtues are identified in relation to
the end or good of human flourishing or happiness (Aristotle: 29).

Montmarquet is an example of a virtue-based epistemologist. As we saw earlier,
Montmarquet maintains that we can control belief by cultivating certain epistemic habits or
virtues (Montmarquet: viii). Montmarquet identifies three general classes of epistemic
virtues, or “ways of being conscientious,” that we as responsible epistemic agents should
strive to cultivate. These are the virtues of impartiality (e. g. openness to ideas and the
willingness to exchange them), intellectual sobriety (e. g., the restraint to not embrace what
is unwarranted simply because it is new and exciting), and intellectual courage (e. g., the
willingness to conceive and examine new alternatives) (Montmarquet: 23). These,
Montmarquet claims, are the qualities that are necessary to guarantee that one has

a proper orientation toward one’s own or other’s beliefs, and this is why the
qualities...seem so necessary to intellectual inquiry (and integral to our notion of a virtuous

inquirer) (Montmarquet: 25-26).
In keeping with the teleological bent of virtue-based theories, Montmarquet shows how these
virtues can be identified through their conduciveness to what he takes to be our primary
epistemic goal - namely that of gaining truth while avoiding error. Montmarquet argues that
what any “contemporary account of epistemic virtue can and should take from Aristotle, at
least as its starting point, is his association of the epistemic virtues with trut ” (Montmarquet:
20). He goes on to state more specifically that the “epistemic virtues are those personal
qualities (or qualities of character) that are conducive to the discovery of truth and the

avoidance of error” (Montmarquet: 20).

108
b. Consequences

Consequentialists in ethics believe that the tightness or wrongness of an act is strictly
a product of the goodness or badness of its consequences. As such, consequentialist theories
in ethics are teleological, for the moral appropriateness of actions is identified through their
conduciveness to a certain end or good, namely, the end or good in relation to which “good
consequences” are identified. Where consequentialists in ethics differ is typically in terms of
what they take this end or good to be, and in terms of who they believe must be considered
in the examination of consequences. Egoistic hedonists identify the good as pleasure, and
maintain that only the agent need be considered in the examination of consequences. Persons
holding this view maintain that good actions are those that produce the most pleasure for the
agent of any alternative action open. Most contemporary utilitarians, on the other hand,
identify the good as happiness or self-actualization, and not mere pleasure. In addition, they
maintain that all those affected by the act in question must be considered in the examination
of its consequences. Someone holding this position, then, believes that good actions are
those, of any alternative action open, that produce the most happiness for all concerned, not
just the agent.

It is interesting that in epistemology there is no analogue to this dispute in ethics over
the nature of the ethical goal or telos. In epistemology, the goal or telos in relation to which
epistemic consequences are assessed is truth, and this goal does not allow for variation in
terms of the scope of one’s consideration (it does not allow for egoistic vs. more altruistic
interpretations). Whereas in ethics it makes sense to distinguish between what is pleasurable

for me and what is pleasurable for others, thereby raising the question of whose pleasure or

109

happiness should be considered in deciding on a course of action, in epistemology there is no
such distinction. The truth of statements (excepting those involving indexicals) does not vary
in this way. Truth is independent of agents in a way that happiness or pleasure is not.

One more distinction remains within consequentialism: one can be a positive or
negative consequentialist. Where persons holding these positions differ is in terms of the
weight that they give to good consequences as opposed to bad. A negative consequentialist
places more weight on the avoidance of bad consequences than the attainment of good. In
ethics, a person holding this position might be an egoist of an Epicurean stripe. Such a person
would argue tlmt one should concentrate on the avoidance of pain or negative consequences,
rather than on the attainment of pleasure or positive consequences. It is argued that in this
way one will attain a higher overall net balance of pleasure over pain. A positive
consequentialist, on the other hand, weights the attainment of good consequences more
heavily. In ethics, such a person could be a utilitarian theorist who emphasized our positive
duty to help others over our negative duty to avoid doing harm. Such a theorist would be
emphasizing the attainment of good consequences over the mere avoidance of bad.

I would argue that “filliam James is an example of an epistemological
consequentialist. This is because James attempts to justify the choice of believing in the face
of insufficient evidence (where the option is genuine, forced, and momentous - see Chapter
One, Sec. VI), by appealing to the goodness of the consequences of doing so. For James, the
principal epistemic goals of attaining truth and avoiding error would seem to provide a
criterion by which to evaluate the goodness or badness of epistemic consequences.

Accordingly, belief in the face of insufficient evidence is justified for James, under specific

”0

conditions, on the basis of its likelihood to result in good epistemic consequences, or true
belief. James discusses the risk of such belief, and how its perrnissibility depends on how one
weighs the consequences. He states:

You, on the other hand, may think that the risk of being in error is a very small matter

when compared with the blessings of real knowledge, and be ready to be duped many
times in your investigation rather than postpone indefinitely the chance of guessing true

(WTB: 510).
In terms of how he weighs consequences James appears to be a positive consequentialist. He
does not seem to weigh the avoidance of error more heavily than the attainment of truth.
James writes that
our errors are surely not such awfirlly solemn things. In a world where we are so certain
to incur them in spite of all our caution, a certain lightness of heart seems healthier than
this excessive nervousness on their behalf (WTB: 510).

Descartes, given his preoccupation with avoiding error, might be characterized as a
negative epistemic consequentialist. Such a characterization would seem to be consistent with
his position in the Discourse on Method, where the first law that Descartes cites for the
direction of his epistemic conduct, is that he include nothing in his judgments that is not
presented to his mind so clearly and distinctly that no reason or occasion could be found to
call it into doubt (Cottingharn: 29).

Clifford, with his contention that “it is wrong always, everywhere, and for anyone,
to believe anything upon insumcient evidence” (Clifford: 505), might likewise seem to be a

negative epistemic consequentialist. We will now see, though, that such an interpretation of

Clifford’s position is quite inaccurate. Clifford is, instead, an epistemic deontologist.

 

 

111
c. Duty

Deontological ethical theories do not hold that the tightness or wrongness of actions
is a product of their being conducive to some end or good. In more direct contrast to ethical
consequentialism, as opposed to other teleological theories, deontological theories in ethics
do not view the consequences of actions as being relevant to the assessment of their moral
worth. A prime example of a deontological view in ethics would be the duty-based theory
of Kant. Kant argues that the tightness or wrongness of an act is not determined by an
examination of its natural or empirical contingencies or accidents, such as its consequences,
but rather by the motive of the agent who performs the action. Kant states:

in morals the proper worth of an absolutely good will, a worth elevated above all price,
lies precisely in this - that the principle of action is free from all influence by contingent
grounds, the only kind that experience can supply (Kant: 93).
The criterion of rightness or wrongness is whether the agent performs the act out of a sense
of duty, and not whether it produces more pleasure than pain.

A prime example of a deontological position in epistemology would be that of C .K.
Clifford. Clifford, much like Kant, argues that when “an action is once done, it is right or
wrong forever”, no accidental failure of its good or evil fiuits can possibly alter that” (Clifi‘ord:
502). In discussing a particular example of belief he states that

The question of right or wrong has to do with the origin of his belief, not the matter of it;

not what it was, but how he got it; not whether it turned out to be true or false, but

whether he had a right to believe on such evidence as was before him (Clifford: 502).
Clifford, then, in contrast to the consequentialism of James, argues that the likelihood of the
belief being true can in no way justify believing in the face of insuficient evidence. In his eyes

the goodness of the consequences, or the truth of the resultant belief, cannot render an

112

epistemic “act” permissible.

d. Clifford vs. James

Applying this taxonomy of ethical positions to epistemology sheds light on the debate
between James and Clifford. It can now be seen as another instance of the general debate
between consequentialists and deontologists -- as another instance of the debate over the
relevance of consequences in determining the appropriateness of certain “actions” or
“conduct.” James wishes to argue that consequences, at least in some instances, matter.
Clifford argues that consequences are irrelevant. What is important is the way in which the
belief is arrived at. The central question for Clifford is whether the agent violated his duty
by believing in a situation where he knew that he shouldn’t: did the agent believe in a situation
where he knew, or should have known, that he had “acquired and nourished a belief, when
he had no right to believe on such evidence as was before him?” (Clifford: 503).

It is interesting to note that James did not see the disagreement between himself and
Clifford as one of deontology vs. consequentialism. Rather, James viewed Clifford as a fellow
consequentialist. The source of their disagreement, in his eyes, concerned the evaluation of
consequences -- a debate between positive and negative consequentialism. James saw
Clifl‘ord as emphasizing the avoidance of error over the gaining of truth, as emphasizing the
avoidance of negative consequences over the reaping of good ones. James states that one
may “treat the avoidance of error as more imperative, and let truth take its chance” and that
“Clifford...exhorts us to the latter course” (WTB: 510). Given Clifi’ord’s assertion, though,

that in discerning the permissibility of belief it only matters how one got it and “not whether

 

i. -

 

rl'. ——..
I

113

it turned out to be true or false” (Clifford: 502), James’ characterization of their disagreement

seems to miss the mark.

111. Normative Goals and Conduct

There are strong parallels between ethics and epistemology in terms of the role that
goals play in the evaluation of conduct. In both ethics and epistemology, what might be
termed “ideal goals” are put forward as the end toward which responsible conduct aims. In
ethics, the ideal goal is tightness or goodness. The responsible ethical or moral agent strives
to conduct himself in such a way that his actions are right, or maximize the good. In
epistemology, the ideal goal is truth. The responsible epistemic agent conducts himselfin
such a way that he believes only those things that are true. An interesting dissimilarity
between these domains, however, is that in epistemology the ideal goal of true belief can be
plausibly interpreted as the goal to have all and only true beliefs.2 In ethics, the ideal goal
cannot be analogously interpreted as the goal to perform all and only good or right actions.
In ethics the “all” falls out. The goal is rather to perform only right or good actions.
Evidently, the notion of believing all truths can be entertained unlike the obviously

overwhelming task of performing all good or right actions.3

 

2 As we saw earlier (Sec. Ila), Montmarquet adheres to such a formulation of our
epistemic goal in relation to which he identifies the epistemic virtues. James as well seems
to adhere to such a formulation, though by presenting it as two competing goals he makes
explicit our need to weight the desire for truth against the dread of falsehood. Others who
put forward such a formulation of our epistemic goal include Alston (1988), Steup,
Feldman, and Foley (1987), as well as Johnson and Hall in “The Epistemic Duty to Seek
More Evidence.”

3 I am indebted to Professor Richard Hall for pointing out this difference.

 

114

In both epistemology and ethics, practical or intermediate goals are postulated to help
us both regulate our own conduct, and evaluate the conduct of others. These practical goals
are parasitic on the ideal. They represent our limited capacity to discern when the ideal goal
has been reached. Thus in epistemology, it may be argued that while the attainment of truth
is our ideal goal, given that it may be dificult to discern when truth has been reached, we
should strive always to have at least justified or rational beliefs. This goal of justified belief,
then, becomes the standard for blameworthiness. One is always blameworthy for having
unjustified beliefs, though one may not always be blameworthy for having false ones (when,
for example, they are nonetheless justified). But this practical goal is not separate from the
ideal goal; it has its status as a regulative goal by virtue of its relationship to the ideal. It is
because rational or justified belief is conducive to true belief that it can be an epistemic goal.
Similarly in ethics, the practical or intermediate goal is morally justified action. It can, just
as with truth, be diflicult to discern if our actions are morally right or good in the ideal sense.
Thus, as agents, the standard of blarneworthiness is whether our actions are morally justified.
One is always blameworthy for performing an action that is morally unjustified, though one
may not be blameworthy for performing an act that is morally wrong in the ideal or objective
sense if the action is none the less morally justified. And, just as with the practical epistemic
goal of rational belief, the practical or intermediate ethical goal of justified action is parasitic
on the ideal goal of right or good conduct. It is because it is conducive to good or right

action, that morally justified action can be a regulative ethical goal.

 

 

 

115
IV. Determinism and Responsibility
In terms of the epistemic debate over freedom we have examined three major
positions: non-voluntarism, indirect voluntarism, and direct voluntarism. This taxonomy can

be applied to ethics, and, when it is, the landscapes of these two domains can be made clearer.

a. Non-Voluntarism

Epistemological non-voluntarists hold that belief is completely determined - that there
is no sense in which we can control, and hence be responsible for, belief. As we saw in
Chapter One, Quine is an excellent example of someone who holds this position. Quine
argues that the application of normative concepts to belief is misguided in that this application
implies a sense of responsibility, and hence control in regards to belief, which is completely
absent. Quine rather opts for a naturalized view of belief where normativity completely falls
out. Belief is no longer the subject of responsibility.

Paul d’Holbach is an example of an ethical non-voluntarist. He argues that man “is
good or bad, happy or miserable, wise or foolish, reasonable or irrational, without his will
being for anything in these various states” (d’Holbach: 363). Man, he argues, “always acts
according to necessary laws from which he has no means of emancipating himself”
(d’Holbach: 363). As such, there is no sense of responsibility for one’s conduct. D’Holbach
states that “crimes are witnessed on the earth only because everything conspires to render

man vicious and criminal” (d’Holbach: 365).

 

 

.1;

 

116

b. Indirect Voluntarism

Epistemological indirect voluntarists maintain that belief can be controlled only
through indirect means, namely through the performance of various actions such as the
gathering of more evidence. As we saw earlier, Locke and Heil are good examples of this
position. Both hold that our responsibility for belief arises from our capacity to control it
through indirect means - through the implementation of certain procedures or methods.

As we saw in the critique of this position, indirect voluntarism is based on the idea of
there being different “layers” to human action, the first layer - background beliefs - being
completely determined, the second layer - procedures for influencing belief - being one of
“free” action. But since this second layer is guided by the first, it was argued that indirect
voluntarism in the epistemic case leads to a vicious regress of responsibility. Epistemic
indirect voluntarists argue belief is controllable only via indirect means, and yet these indirect
means will be ultimately guided by background beliefs that are themselves determined. If the
layer of background beliefs is determined, and hence is not the subject of responsibility, then
the second layer of free action, in being guided by this determined layer, must not be the
proper subject of responsibility as well (for a more detailed discussion of this regress see
Chapter Three, Sec. H).

This regress of responsibility is much clearer in the epistemic case than in ethics. In
epistemology it is very difiicult to understand how one could have fiee choice with
determined beliefs. In ethics, on the other hand, where the influence of desire on action is the
primary focus, it seems more plausible that one could have determined desires and yet still

have free choice, for it would seem that one could choose not to fiilfill or act on one’s desires.

117

This difference is due to the different degrees of control that belief and desire have over action
or choice. It would seem that belief determines epistemic choice much more readily than
desire determines ethical choice.‘ What is necessary, then, to close the gap between ethical
and epistemological indirect voluntarism, and thus produce an analogous regress in ethics, is
to maintain that desire is a determiner of action - that one must always act on one’s strongest
desire. Such an ethical position would be like that of Thomas Hobbes. Hobbes is a soft-
determinist, or ethical indirect voluntarist, who adheres to a position much like what we
earlier called the Desire Thesis. He argues that one possesses freedom in so far as one is able
to do what one wants or wills (Hobbes: 171). Freedom is just the freedom from external
constraints on one’s actions. Hobbes goes on to discuss, however, how this “layer” of
freedom is guided by an initial layer that is wholly determined - he goes on to discuss how

liberty and necessity are consistent.’ Hobbes argues that

 

‘ Ultimately, it may be that it is the combination of background beliefs and desires that
determines our conduct in most situations. It may be that desire and belief almost always
work in concert. It would seem, though, that in epistemology consideration is primarily
given to the influence of belief on choice, and in ethics the influence of desire. Thus one
of the primary concerns in ethics is the apparent conflict between the demands of morality
and those of self-interest - or the conflict between moral responsibility and inclination or
desire. In epistemology, though the conflict between what one rationally ought to believe
and what one wants to believe is not completely absent, the role of desire in epistemic
misconduct and mistake does not seem to be at center stage. Rather, more attention seems
to be payed to explaining false or unjustified belief in terms of cognitive defect, problems
with the evidence, or the presence of false background beliefs which are viewed as
corrupting the system.

5 In this respect the view of Bernard Williams concerning rational belief may be an
analogue to an ethical soft deterrninist position like that of Hobbes. For, just as Hobbes
attempts to reconcile fieedom and necessity, Williams tries to reconcile rationality and
necessity. Both attempt to do so by distinguishing between different types of causal
chains. Thus for Williams, certain types of causal chains are viewed as producing rational
beliefs, while others are viewed as producing beliefs that are not rational. Similarly, for

118

the actions which men do. . . because they proceed from their will, proceed fi'om liberty,
and yet - because every act of man’s will and every desire and inclination proceeds from
some cause, and that from another cause, in a continual chain whose first link is in the
hand of God, the first of all causes - proceed from necessity (Hobbes: 171-172).

c. Direct Voluntarism
The final position is that of direct voluntarism. The epistemological direct voluntarist
holds that one can control belief directly through an act of will. The Stoics are perhaps the
strongest adherents to this position. As was seen in Chapter One, they seemed to maintain
that belief is almost entirely within the realm of direct willing or choice. C.A. Campbell is
probably closest to an ethical direct voluntarist. Campbell argues that the will is capable of
directly controlling action. In contrast to the ethical indirect voluntarist, he maintains that it
is possible for one to will or choose actions in opposition to one’s desires or inclinations - that
freedom extends beyond doing what one wants, and is in fact exclusively illustrated by those
cases where the will struggles against the pull of desire. Thus Campbell states that
Here, and here alone, so far as I can see, in the act of deciding whether to put forward or
withhold the moral effort required to resist temptation and rise to duty, is to be found an

act which is free in the sense required for moral responsibility (Campbell: 395); (for a more
detailed discussion of Campbell see Introduction, Sec. II).

 

Hobbes certain types of causal chains, those progressing through the desires of the agent,
are considered as producing fi’ee actions, while others, those circumventing the agent’s
desires, are considered to produce actions that are not free. I am indebted to Prof.
Richard Hall for pointing out this similarity.

119

d. Determinism In Ethics and Epistemology

I have examined the relationship between determinism and responsibility in both
epistemology and in ethics. In epistemology, I showed indirect voluntarism to be inadequate
in that it was shown to be incapable of supporting a robust account of epistemic
responsibility. In the ethical case I have argued that ethical indirect voluntarism, or soft
determinism, fails to give a satisfactory account of responsibility as well. There is, however,
another important similarity between ethics and epistemology in terms of determinism -
specifically, the response that is raised in these two domains to the notion of complete
determinism or non-voluntarism.

1 have spent much time trying to show the incoherency of any attempt to rationally
assert non-voluntarism or determinism. It have argued that any such attempt is incoherent
because it presupposes the very freedom or control over belief that determinism denies. To
be a rational agent is, at least in part, to be an agent who is capable of evaluating or weighing
evidence and then choosing on the basis of this evaluative process what to believe or assert.
Determinism denies the control over evaluation, and the capacity for free choice, that this
aspect of rational agency implies.

In that my discussion of the coherency of asserting determinism has centered on our
conception of ourselves as rational agents, it should perhaps be construed as epistemic given
the central place of this conception in epistemology. In ethics, however, theorists have also
discussed the incoherency of asserting or holding the position of determinism. Certainly Kant
is one such ethical theorist. But Kant’s discussion of the possibility of coherently asserting

determinism seems to be primarily epistemic as well, for it focuses on the entailments of our

 

 

 

120
conception of ourselves as rational agents. He uses this conception of rational agency, with
the freedom or voluntarism that it presupposes, as a basis for refiiting determinism in the
moral sphere so as to establish the possibility of free moral agency.

C.A. Campbell, another ethical thinker, views the question of the coherency of
asserting determinism as an ethical question - as one centering not on our conception of
ourselves as rational agents, but rather on our conception of ourselves as practical or moral
agents. Campbell seems to argue that determinism departs so strongly from our
phenomenology of moral experience, a phenomenology which he sees as establishing freedom
and hence responsibility, that to hold or assert determinism is more than merely counter-
intuitive. Campbell discusses the awkward predicament that arises fi'om the theoretical
arguments raised in favor of determinism. He contends that if these arguments are found
convincing, one may have “to deny as a theoretical being what he has to assert as a practical
being” (Campbell: 396). This arises from the fact that Campbell believes “no one while
fimctioning as a moral agent can help believing that he enjoys fi'ee will”:

Theoretically he may be completely convinced by Deterministic arguments, but when
actually confronted with a personal situation of conflict between duty and desire he is quite
certain that it lies with him here and now whether or not he will rise to duty”
(Campbell: 396).
From this, Campbell concludes that the position of determinism illustrates “a radical conflict
between the theoretical and practical sides of man’s nature, an antinomy at the very heart of
the self” (Campbell: 396). Campbell’s position nicely illustrates, then, that in ethics, just as

in epistemology, the notion that it is somehow incoherent or self-refitting to assert

determinism can emerge.

LIST OF REFERENCES

LIST OF REFERENCES

Alston, \Villiam. 1985. “Concepts of Epistemic Justification.” Monist.

. 1988. “The Deontological Conception of Epistemic Justification.” Philosophical
Perspectives, 2, Epistemology.

Ammerman, Robert. 1965. “Ethics and Belief.” Proceedings of the Aristotelian Society.

Aristotle. 1925. Nicomachean Ethics. In The Spectrum of Responsibility, ed. Peter A.
French. New York: St. Martin’s Press.

Aurelius, Marcus. 1983. The Meditations. trans. G.M.A. Grube. Indianapolis: Hacket
Publishing Company.

Berkeley, George. 1993. Philosophical Works Including the Works on Vision. ed. Michael
R. Ayers. London and Vermont: Everyman Library.

Bonjour, Laurence. 1976. “Holistic Coherentism.” In The Theory of Knowledge, ed. Louis
P. Pojman. Belmont, California: Wadsworth.

Brandt. RB. 1985. “The Concept of Rational Belief.” Monist.

Brink, David. 1984. “Moral Realism and the Skeptical Arguments from Disagreement and
Queemess.” In Ethical Theory, ed. Louis P. Pojman. Belmont, California: Wadsworth

Brett, F. 1980. “Doubt and Descartes’ Will.” Dialogue.
Bury, KG. 1990. Outlines of Pyrrhonism. New York: Prometheus Books.

Campbell, CA 1957. “Has the Self ‘Free Will’?” In Reason and Responsibility (6th edition),
ed. Joel F einberg. Belmont, California: Wadsworth Publishing Co..

Chisholm, Roderick M. 1957. Perceiving. Ithaca, New York: Cornell University Press.

. 1968. “Lewis’ Ethics of Belief.” In The Philosophy of Cl. Lewis, ed. Paul Aurthur
Schilpp. La Salle, Ill: Open Court.

 

121

122

Clarke, Murray. 1986. “Doxastic Voluntarism and Forced Belief.” Philosophical Studies.
Classen, H.G. 1979. “Will, Belief, and Knowledge.” Dialogue.
Clifford, WK. 1879. “The Ethics of Belief.” In The Theory of Knowledge, Supra.

Dennett, DC. 1974. “Mechanism and Responsibility.” In Essays on Freedom of Action,
ed. Ted Honderich. London and Boston: Routledge and Kegan Paul Ltd..

 

. 1984. Elbow Room: The Varieties of Freewill Worth Wanting. Cambridge, Mass.
and London England: MIT Press.

Descartes, Rene. 1994. Selected Philosophical Writings. trans. John Cottingharn, Robert
Stoothoff, and Dugald Murdoch. Cambridge: Cambridge University Press.

Evans, J .L. 1968. “Error and the Will.” Philosophy.
Feldman, Richard. 1988. “Epistemic Obligations.” In The Theory of Knowledge, Supra.

Firth, Roderick. 1959. “Chisholm and the Ethics of Belief.” In The Philosophy of G]. Lewis,
Supra.

_. 1978. “Are Epistemic Concepts Reducible To Ethical Concepts?” In Values and
Morals, eds. A.I. Goldman and J. Kim. Holland: D. Reidel Publishing Co..

Foley, Richard. 1983. “Epistemic Conservativism.” Philosophical Studies.

.1987. The Theory of Epistemic Rationality. Cambridge, MA: Harvard University
Press

. 1991. “Evidence and Reasons for Belief.” Analysis.

Gale, Richard. 1980. “Mlliam James and the Ethics of Belief.” American Philosophical
Quarterly.

Gerson, Loyd P. and Brad Inwood. 1988. Hellenistic Philosophy. Indianapolis, Indiana:
Hacket Publishing Company.

Goggans, Phil. 1991. “Epistemic Obligations and Doxastic Voluntarism.” Analysis.

Goldman, Alvin I. 1967. “A Causal Theory of Knowledge.” In The Theory of Knowledge,
Supra.

123

_. 1979. “Reliabilism: What is Justified Belief?” In The Theory of Knowledge, Supra.
Govier, Trudy. 1976. “Belief, Values and the Will.” Dialogue.

Grant, Brian. 1976. “Descartes, Belief, and the W111.” Philosophy.

Harman, Gilbert. 1975. “Moral Relativism Defended.” In Ethical Theory, Supra.
Harvey, Van. 1969. “Is There an Ethics of Belief ?” Journal of Religion.

Heil, John. 1982. “Seeing is Believing.” American Philosophical Quarterly.
_. 1983. “Doxastic Agency.” Philosophical Studies.

Hobbes, Thomas. 1987. Leviathan - Parts One and Two. New York: Macmillan Publishing
Company.

Holbach, Paul. 17 70. System of Nature. In Reason and Responsibility (8th edition), ed. Joel
Feinberg. Belmont, California: Wadsworth Publishing Co..

Hoyler, Robert. 1983. “Belief and Will Revisited.” Dialogue.
James, Mlliam. 1897. “The Will to Believe.” In The Theory of Knowledge, Supra.
. 1950. The Principles of Psychology Vol. 11. U. S.A.: Dover Publications, Inc..

. 1967. Essays on Faith and Morals (4th edition). ed. Ralph Barton Perry. Cleveland
and New York: The World Publishing Company.

Kant, Immanuel. 1956. Groundwork of the Metaphysics of Morals. trans. H.J. Paton. New
York: Harper & Row, Publishers.

Kapitan, Tomis. 1989. “Doxastic Freedom: A Compatibilist Alternative.” American
Philosophical Quarterly.

Kauber, Peter and Peter Hare. 1974. “The Right and Duty to Will to Believe.” Canadian
Journal of Philosophy.

Kim, Jaegwon. 1988. “What is ‘Naturalized Epistemology’?” In The Theory of Knowledge,
Supra.

. 1989. “Mechanism, Purpose, and Explanatory Exclusion.” Philosophical
Perspectives, 3.

 

 

124
_. 1984. “Self-Understanding and Rationalizing Explanations.” Philosophia Naturalis
Kornblith, Hilary. 1985. “Ever Since Descartes.” Monist.
Lehrer, Keith. 1990. Theory of Knowledge. Boulder and San Francisco: Westview Press
, and Nicholas D. Smith. 1996.“Human Wisdom: An Epistemological and

Pedagogical Inquiry.” In Knowledge, Teaching, and Wisdom, eds. K. Lehrer, B. Lum, B.
Slichta, and N. Smith. Kluwer Academic Publishing.

 

Lewis, CL 1968. “The Philosopher Replies.” In The Philosophy of CI. Lewis, Supra.

Locke, John. 1974. An Essay Concerning Human Understanding. ed. A.D. Woozley.
Harmondsworth, Middlesex, England: Penguin Books Ltd..

Malcolm, Norman. 1968. The Conceivability of Mechanism.” Philosophical Review.
Meiland, Jack. 1980. “What Ought We to Believe?” In The Theory of Knowledge, Supra.

Montmarquet, James A. 1993. Epistemic Virtue and Doxastic Responsibility. Lanharn,
Maryland: Rowman & Littlefield Publishers, Inc..

Moore, GE. 1903. “Non-Naturalism and the Indefinability of the Good.” In Ethical Theory,
Supra.

Nagel, Thomas. 1979. “Moral Luck.” In Ethical Theory: Classical and Contemporary
Readings, ed. Louis P. Pojman. Belmont, California: Wadsworth Publishing Co..

Nelson, John O. 1984. “How and Why Seeing is not Believing.” Philosophical Research
Arch..

Newman, J.H., 1947. A Grammar of Assent. New York, Toronto, London: Longrnans
Green and Co..

Nussbaurn, Martha. 1986. The Fragility of Goodness. Cambridge, New York: Cambridge
University Press

Oates, Whitney J. 1940. The Stoic and Epicurean Philosophers (8th edition). New York:
Random House.

O’Hear, Anthony. 1972. “Belief and the Will.” Philosophy.

125

Pears, David. 1974. “Rational Explanation of Actions and Psychological Determinism.” In
Essays on Freedom of A ction, Supra.

Perry, Ralph P. 1935. The Thought and Character of William James Vol. 1. Boston:
Little, Brown, and Company.

. 1938. “The Right to Believe.” In In the Spirit of William James. New Haven:
Yale University Press.

Pieper, Josef. 1963. Belief and Faith. New York: Pantheon Books.

Pojman, Louis P. 1985. “Believing, Willing, and the Ethics of Belief.” In The Theory of
Knowledge, Supra.

. 1986. Religious Belief and the Will. London and New York: Routledge &
Kegan Paul Ltd..

. 1994. What Can We Know?. Belmont, California: Wadsworth Publishing Company.

Price, H.H. 1954. “Belief and Will.” Proceedings of the Aristotelian Society, Supplementary
Volume.

Putnam, Hilary. 1967. “Psychological Predicates.” In Art, Mind, and Religion, eds. W.H.
Capitan and DD. Merrill. Pittsburgh: University of Pittsburgh Press.

. 197 3. “Meaning and Reference.” In Naming, Necessity, and Natural Kinds, ed.
Stephen P. Schwartz. Ithaca and London: Cornell University Press.

Quine, W.V. 1969. Ontological Relativity and Other Essays. New York: Columbia
University Press.

Rawls, John. 1971. A Theory Of Justice. Cambridge, Massachusetts: The Belknap Press
of Harvard University Press

Saunders, Jason L. 1966. Greek and Roman Philosophy after Aristotle. New York: The
Free Press.

Steup, Matthias. 1988. “The Deontic Conception of Epistemic Justification.” Philosophical
Studies.

Sullivan, Roger J. 1991. Immanuel Kant 's Moral Theory. Cambridge, New York: Cambridge
University Press.

126

Taylor, Richard. 1974. “Freedom and Determinism.” In Reason and Responsibility (8th
edition), Supra.

Watling, John. 1973. “Hampshire On Freedom.” In Essays on Freedom of Action, Supra.

Williams, Bernard. 1973. “Deciding to Believe.” In his Problems of the Self. Cambridge:
Cambridge University Press.

Winters, Barbara. 1979. “Believing at Will.” Journal of Philosophy.

Wittgenstein, Ludwig. 1972. On Certainty. New York: Harper and Row.

 

 

"I11111111111111111111"