[Return to Dmitry Sepety’s
Personal Page]
Comprehensively Critical
Rationalism: Rationality with Revisable Foundations*
by Dmitry
Sepety
1. Bartley’s comprehensively critical
(pancritical) rationalism
as solution to the problem of limits of
rationality
In his book “The Retreat to
Commitment” William Bartley proposed comprehensively critical (pancritical)
rationalism (CCR) as solution to the problem of limits of rationality. The
problem arises from the following facts:
1) according with traditional
concept of rationality (the view Popper dubbed uncritical or comprehensive
rationalism), rational attitude demands that we accept those and only those
views which are rationally justified;
2) this demand can’t be implemented
neither for rational attitude itself nor for any other view.
In a bit more details.
1) It is generally acknowledged that
justifying some view by reference to itself is fallacious, – well-known type of
argumentative error called “begging the question”. (Direct form: “Why X?
Because X.” Indirect form: “Why X? Because Y. Why Y? Because X.”) Any attempt
to justify something rationally already presupposes rational attitude. So,
attempt to justify rationally rational attitude is attempt to justify rational
attitude by rational attitude, – direct begging the question. So, it is fallacious.
This is the situation Popper
described in “The Open Society and Its Enemies” when he wrote about uncritical or comprehensive rationalism:
“We can express this … in the form of the principle
that any assumption which cannot be supported either by argument or by
experience is to be discarded. Now it is easy to
see that this principle of an uncritical rationalism is inconsistent; for since
it cannot, in its turn, be supported by argument or by experience, it implies
that it should itself be discarded.” (Popper 1977, 230)
2) To justify some view we need some
other, justifying views which themselves have to be justified. Now to justify
these views we need yet other justifying views which themselves have to be
justified. And so on. So, any attempt of justification gets us
either into logical vicious circle
of indirect begging the question, which makes justification fallacious
or into infinite regress which never
achieve justification.
A ← B ← C ← D
← …………
So, no view can be properly
justified.
Popper in “The Open
Society and Its Enemies” stated this as follows:
“Since all argument must proceed from assumptions, it
is plainly impossible to demand that all assumptions should be based on
argument.” (Popper 1977,
230)
So, rationalism, if it is to be
identified with the demand to accept those and only those views which are
rationally justified, is untenable.
Thus far Bartley agrees with
Popper’s treatment of the problem in “The Open Society and Its Enemies”. But he
disagrees with Popper’s proposed solution of the problem (CR-1) and severely
criticizes it.
Poppers proposition is to admit that
rationalism can’t be comprehensive, that it is itself “irrational faith in
reason” and to designate such position "critical rationalism" – in
opposition to both irrationalism and comprehensive (uncritical) rationalism.
Bartley shows that this view is as unacceptable
as comprehensive (uncritical) rationalism. Popper’s concession to irrationalism
is too much, as it is unneeded, and too little, as it doesn’t really solve the
problem.
Its gravest fault is that as far as
we retain the demand of justification we can’t get out of trouble just by
making one exception – for rational attitude itself. For we have already seen
that it is not only rational attitude itself that can’t be properly rationally
justified, – no view at all can be rationally justified!
It follows, that justification is
possible only relative to some
framework – some set of basic axioms and principles which constitute rules and
references of justification. These basic axioms and principles themselves can’t
be justified. Moreover, they can’t be criticized too, – for, being basic, they constitute
rules and references of criticism. So, all we can do is just to accept some
framework of basic axioms and principles and work within it. As these can’t be
properly justified, they are exempt from demand for justification; as they
can’t be properly criticized, they are exempt from criticism.
If it is so, then any arbitrary
framework of basic axioms and principles can be accepted just as well as any
other. (At least – any which isn’t self-contradictive – which stands it own
rules.) There is no rational way to compare and select what is best; no
possibility for rational preference of one framework over another; no rational
way to judge what basic axioms and principles are true and what are false. You
just make commitment to some framework, – make irrational decision to hold to
it. Really, usually you don’t make real decision, – you just uncritically
absorb some framework from your childhood social environment or from
professional environment (science). And you can’t be critical toward it. The
only way you can change your framework is irrational leap of faith, – which is
making new commitment. If you make such irrational leap, you are in the same
attitude toward new framework as you were toward old, – its axioms and principles
are exempt from criticism.
Such commitment-theory became philosophical
fashion of XX century. To name only three most influential
commitment-theorists: J.P.Sartre, L.Wittgenstein, T.Kuhn.
Where is rationality to find its
place? Only inside this or that framework. On this view rationality can be
nothing but keeping to axioms and principles (which constitute rules of
justification and criticism) accepted inside some framework. There is no real
difference between frameworks which call themselves rational and those which
don’t. This is merely a matter of word-preference. You can introduce any set of
basic axioms and principles and call it “rationality”. As far as you keep to
these axioms and principles, you are “rational”. There is no one comprehensive
rationality, there are as many “rationalities” limited to frameworks as you
want.
This is tu-quoque argument which
Bartley undertakes to refute.
Popper in The Open Society and Its Enemies proposed, as solution to similar
problem, to distinguish two forms of rationalism:
1)
uncritical,
comprehensive rationalism which demands that “any assumption which cannot be supported either by
argument or by experience is to be discarded”;
and
2)
critical
rationalism which identifies rationality with critical attitude and “frankly admits its limitations, and its origin in an
irrational decision (and so far, a certain priority of irrationalism)”, “irrational faith in reason”.
Popper admitted that uncritical,
comprehensive rationalism is self-defeating, but contended that critical
rationalism is tenable.
Bartley criticized this purported solution
as fideistic appeal to decision or commitment. Popper rejected charge of
fideism. I think that Bartley’s criticism wasn’t quite right, but it suggests improvements
to Popper’s description of CR:
1) the real difference between two
forms of rationalism is to be described not in terms of comprehensiveness/limitations
but in terms of justificationism/critical attitude;
2) critical rationalism is to admit
comprehensiveness of rationality understood not as justification, but as critical
attitude, openness to criticism.
Bartley locates the source of the
problem which makes traditional rationalism self-defeating not in
comprehensiveness, but in demand for justification. This demand is unrealizable
– whether it is comprehensive or not. And this demand is needless for
development of knowledge. All we really need is openness to criticism,
willingness to consider seriously critical arguments when presented. And this
openness, critical attitude needs no exception, even for itself! The term
'rationality' is traditionally associated with both – demand of justification
and critical attitude. We can well discard the first and retain the second in
full comprehensiveness! So, Bartley proposes to identify rationality with the attitude
of openness to criticism. Views and theories are neither rational nor
irrational. What is rational (or irrational) is our attitude towards our views.
Rational attitude is nothing but critical attitude.
Bartley proposes to call this
conception of rationality comprehensively
critical rationalism or pancritical
rationalism.
Now a bit more about
comprehensiveness.
Can rationalist really hold all his
views, without exception, open to criticism? There are two often doubted points
about the question?
First. If everything is
criticizable, then all statements used in criticism are criticizable too. If
so, how criticism is possible?
Bartley explains:
“I don’t suggest that all
assumptions be abandoned, only that
they be opened to criticism… pancritical rationalist, like other people, holds
countless unexamined presuppositions and assumptions, many of which may be
false. His rationality consists in his willingness to submit these to critical
considerations when he discovers them or when they are pointed out to him...
When one belief is subjected to criticism, many others, of course, have to be
taken for granted – including those with which the criticism is being carried
out. The latter are used as the basis of criticism not because they are
themselves justified or beyond criticism, but because they are unproblematical at present. These are, in that sense alone and during that time
alone, beyond criticism. We stop criticizing – temporarily – not when we
reach uncriticizable authorities, but when we reach positions against which we
can find no criticisms. If criticisms of these are raised later, the critical
process then continues. This is another way of saying that there is no theoretical
limit to criticizability – and to rationality. …the fact that most of a man’s
beliefs are beyond criticism at any one time does not mean that any of them has
to be beyond criticism all the time...” (Bartley 1984, 121-122)
Second. Can rationalist really apply
this critical attitude even to rationalism itself? The fact is, he can.
Really, there were quite a many
cases when rationalists renounced rationalism exactly because they judged
criticisms against rationalism sound. But their judgement was mistaken, for
they identified rationalism with the demand for justification. There is really
crushing criticisms against rationalism so understood – justificationist rationalism. So, justificationist rationalism is
to be renounced. But rationalism can be retained in its (pan)critical variety.
Those former rationalists converted to irrationalism didn’t see this
alternative. Nevertheless, their cases have some positive value: they prove
that rationalist can genuinely hold rationalism open to criticism.
Pancritical rationalist holds open
to criticism all his views, including CCR itself. Attitude of pancritical
rationalist towards CCR can be stated as follows:
“I accept CCR, for, as far as I see
it, it is the best view, and it is consistent. As far as I know, there are no sound
criticisms showing CCR unsatisfactory, and there is no satisfactory alternative
to it. But, as pancritical rationalist, I keep all my views open to criticism
and make no exemptions, even for CCR itself. So, if new criticisms or
alternatives will be proposed, I am willing to consider them seriously, and if
I see that criticisms are sound or that alternatives proposed are better, I would
renounce CCR.”
Précis. No position (theory, statement,
belief etc.) can satisfy justificationist demand. Justificationist rationalism
is self-defeating. But any position can be held open to criticism. Openness to
criticism can be comprehensive. There are no limits of rationality understood
as openness to criticism. And there is no need for commitment.
2. Strong and Weak CCR
Bartley had given different
descriptions of CCR which suggest two different interpretations. I will call
them strong CCR and weak CCR.
Strong CCR
Strong CCR is the interpretation
based on Bartley’s statement (possibly the most often quoted in discussions of
CCR):
“The position may be held rationally without needing justification at
all – provided that it can be and is held
open to criticism and survives severe examination”. (Bartley 1984, 119)
(By “position” Bartley mean any
statement, theory, attitude etc.)
The statement may be interpreted in
two different ways.
The first is to consider it as stating
sufficient, but not necessary conditions of rationality. In fact, Bartley does not say that position is held
rationally if and only if it fits
both conditions described. If so, this statement, taken literally, does not
exclude the possibility for a position to be held rationally even if one of the
conditions is not satisfied!
But usually this statement is taken
(both by supporters and critics of Bartley) in the stronger meaning, as the
definition of rationality from CCR’s viewpoint which states sufficient and necessary conditions of
rationality. In other words, it is understood in the sense: “The position is
held rationally if and only if it is
held open to criticism and survives
severe examination”. I will call this view “strong CCR”.
So, strong CCR defines, as necessary
for a position to qualify as rationally held, two conditions:
1) it is held open to criticism;
2) it survives severe criticism.
Now, CCR states that it is possible
for a person to hold all his positions rationally. So, for strong CCR to be
true, it must be possible for a person (1) to hold all his views open to
criticism and (2) to hold only those views which survive severe examination.
Surely, for a view to survive severe
examination, there must be severe examination of it. But what is to be counted as
“severe examination”? If pancritical rationalist is forbidden to hold any view
which wasn’t severely examined, he needs some general criteria to distinguish
what is to be counted as “severe examination” from criticisms and checks which
aren’t severe enough. Bartley proposes nothing of the sort.
Now, some most usual criticisms
against CCR point out different varieties of statements which any rational
person would consider true, though it seems that there can be no criticism or
check of them which would count as “severe examination”.
The most obvious criticism of strong CCR is that for any person it is
absolutely impossible to subject all his/her views to “severe examination”,
whatever this means. (Surely, it is not to deny that critical examination, the
severer the better, is always desirable and makes for rationality.) The great
majority of views every person holds are unexamined, and it can’t be otherwise.
No one can examine all – or majority – of his/her views. If attempt was made,
it would lead to that same infinite regress which Bartley tried to escape by proposing
CCR. So, if Bartley’s proposed solution to the problem of limits of rationality
is strong CCR, he doesn’t solve the problem, but falls into the same trap as
justificationaists. (For more detailed discussion of this point see section “How
Bartley Have Let in the "Troyan Horse" of Justificationism” in my
article “Critical Rationalism and the Problem
of Reasonable Beliefs”.)
The interesting thing is that
Bartley is well aware of this and, when criticized, answers:
“…pancritical rationalist, like other people, holds countless unexamined
presuppositions and assumptions … His rationality consists in his willingness
to submit these to critical considerations when he discovers them or when they
are pointed out to him...” (Bartley 1984, 121-122)
But this statement is a direct
refutation of strong CCR.
So, strong CCR is untenable. The tenable interpretation of CCR is weak CCR.
It may be proposed that the phrase
about the survival of severe examination is to be interpreted in a weaker sense of
the demand to reject (disbelieve) positions
that have not survived criticism. (Surely, this demand may apply only to
those positions which were, in fact, criticized.) Really, we may meet the
attribution of such a demand to CCR in some discussions. But I doubt that Bartley
ever stated such demand. In any case, it makes no sense. If a position has not survived criticism, it does mean that it was
rejected (disbelieved); if it was not rejected (disbelieved), it has survived
criticism. There is no such event (action) as rejection (disbelieving) a
statement over and above its failure to survive criticism.
It is a consequence of the fact that
no conclusive proof of falsity, except for self-contradiction, (as well as no
conclusive proof of truth, except for tautology) is possible. Whenever you are
presented with a critical argument intended to demonstrate the falsity of some
position P, the most that can be said is that it logically follows from the
premises of the argument (if the argument is valid) that P is false. But this
does not mean that P is false. This does mean that either P is false, or one
(some) of the premises of the argument is (are) false. So, you always have a
choice; it is always for you to judge; logically, the only thing you are
forbidden is “to sit on both the chairs”. That is why there is no other sense
in which a position may fail to survive criticism except that the criticism has
convinced someone that the position is false, i.e., has made someone to reject
(disbelieve) the position.
Weak CCR
Weak CCR says that to hold position (statement, theory,
attitude etc.) rationally is to hold
it open to criticism. And it says that it is possible for a person to holds all
his/her views open to criticism. Openness to criticism is the only condition of
rationality; no further conditions are required.
On my view, weak CCR is the same
with what Popper repeatedly (with some variations) said to be the essence of
critical rationalism:
“I may be wrong and you may be right,
and by an effort, we may get nearer
to the truth”. (Popper 1977, 238)
“…perhaps I am wrong and perhaps you are right.
But we could easily both be wrong.
… we can nearly always come closer
to truth
in a discussion which avoids
personal attacks…” (Popper 1992, 199)
“I may be wrong, and you may be right,
but let us sit together and discuss
matter critically,
and in the end we may not agree
but we will both have learnt
something.” (Artigas 1999, 30)
The last two rows are to be
emphasized.
Now the question is: isn’t weak CCR
too weak? The answer is: it is strong enough to be a sound solution to the
problem of limits of rationality.
One may say that weak CCR is too
weak on the other considerations: it does mean that the kind of arguments or criticisms one heeds to is irrelevant to
the distinction of rational and irrational. Let us take one extreme example. It
is logically possible for some religious fundamentalist to be comprehensively
critical rationalist. Suppose for John it just seems very plausible that the
Holybook (whatever it is) is the most reliable source of true knowledge. But,
as a fallibilist and comprehensively critical rationalist, John admits that he
may be mistaken, and he is ready to consider all relevant arguments and
criticisms. And now whatever arguments and criticisms are adduced (and however
strong they seem to us), John doesn’t
find them convincing; for him it still seems more plausible that the Holybook
is the most reliable source of true knowledge than that any of the premises on
which the criticisms are based are true. Now, from the viewpoint of weak CCR
there is nothing irrational in John’s attitude; he is perfectly rational.
The supporter of weak CCR may say in
its defence that though such John is logically possible, he is hardly really
(psychologically) possible or, if possible, is an exception rather than a rule.
Generally, the supporter of weak CCR must admit that openness to critical
discussion can’t exclude as irrational any sort of foolishness, and gives no
guarantee of achieving truth or agreement. In a sense, weak CCR needs a sort of
“faith in reason” – an unjustified belief that openness to critical discussion
is, by and large, conductive to truth and, in practical matters of human
relationships, to agreement or worthy compromise.
In other words, Kant’s famous motto
“Dare to use your own intelligence!” is not meant as a panacea. To be a
rationalist, a person does not need to believe that using one’s reason is always
successful. Also, he/she does not need to believe that reason is the same with
all persons; that all persons, when they use their reasons, need arrive to the
same results. Also, he/she does not need to deny that there are fools who,
while using their reason, will very often arrive at foolish beliefs, and that
there are demented whose use of their reason is of no good. All that a person
need to believe to be a rationalist is that using one’s reason is, by and
large, a good thing.
And the supporter of weak CCR can
claim that it is sufficiently strong for the purpose it was intended for – the
solution of the problem of limits of rationality. It is strong enough for
Bartley’s major task, which was to refute “the contention … that there is an
essential logical limitation to rationality: the rational defence and
examination of ideas must, for logical reasons, be terminated by an
arbitrary and irrational appeal to what can be called dogmas or absolute
presuppositions”, and to show that “there are no limits to rationality in
the sense that one must postulate
dogmas or presuppositions that must be held exempt from review…” (Bartley 1984,
221), that “criticism can be carried out successfully and satisfactorily
without … any resort to dogmas or authorities”, “that it is not necessary, in
criticism, … to declare a dogma that cannot be criticized”, “that it is not
necessary to mark off a special class of statements, the justifiers, which do the justifying and criticizing but
are not open to criticism…” (Bartley 1984, 223)
3. Attempted Criticisms of CCR
Some usual criticisms of CCR deny
possibility of comprehensive critical attitude, along the lines discussed – and
shown unsound – at the end of section 3.
Imre Lacatos
in footnote of the article “Popper on Demarkation and Induction” formulates CCR
as the theory according to which “all propositions accepted by a rational
person must be open to criticism”. Lacatos dismisses it offhandedly by the remark:
“the basic weakness of this position is emptiness. There is not much point in
affirming the criticisability of any position we hold without concretely
specifying the forms such criticism might make”. (Lacatos 1974, 264) Lacatos
had to know better. It is his appraisal of CCR which is empty and pointless.
For it is empty and pointless to appraise the theory without considering the
problem it is meant to solve. To appraise any theory we need to answer
questions: What is the problem this theory attempts to solve? Is this problem
important? Is the solution proposed by the theory sound? On answering these
questions we will see that Bartleys’ theory is not empty (pointless) – it deals
with important philosophical problem
and proposes its solution. Besides (as far as I see it), the solution is sound.
Surely, it would be highly desirable to supplement the basic general principle
with concrete specifications, and this lays down a program for further
elaboration. But clear statement and rational defence of the basic general
principle is valuable by itself when this principle is widely denied (or its
opposites affirmed) on the supposedly firm logical reasons, and there is no
other tenable solution known.
There is also interpretation of CCR as
the theory about rationality of beliefs,
as a doctrine according to which “a belief is rational if it is criticisable”. (Musgrave 1991, 30) This
misunderstanding is characteristic also for criticisms of CCR by John Watkins
and John Post. Bartley explains:
“…I declare that all
statements are criticizable…” (Bartley1984, 223) (Accentuation mine.)
“It is … one of the merits of pancritical rationalism … that it presents a
theory about people, not statements. It is, quite explicitly, an account of the
essense of being a rationalist. It is
an account of how a rationalist or critical person might behave. It is not an
account … of rational statements or of rational belief. … I doubt the merit of
discussing the problem of rationality in terms of rationality of statements…
For statements are intrinsically neither rational nor irrational. …rationality
is not a property of statement but is a matter of the way in which a statement
is held, and also … of the way in which the statement has been examined.”
(Bartley1984, 233-234)
So, according to CCR, attributes
'rational' and 'irrational' are to be applied not to beliefs, but to our attitudes toward our beliefs. The
rational attitude consists in holding our
beliefs open to criticisms, – admitting the possibility of these beliefs
being false and being ready to consider criticisms proposed and to reassess our
beliefs. (Bartley 1984, 233-234)
John Watkins mentions Bartley’s
explanation that “CCR is a theory about people, and the attitude they should
adopt if they seek to be rational, not about statements” (Watkins 1993, 275) and
contraposes it with another Bartley’s statements saying that CCR concerns
“so-called logical limitations of
rationality” (Bartley 1964, 5). So, Watkins remarks: “That doesn’t sound to me
as a theory about people”. (Watkins 1993, 275) But this is just picking on
words without considering what they are about! CCR is the theory that it is logically possible for a person to hold all his views open to criticism, that there
are no logical limitations which
would make it impossible.
Watkins and Post proposed criticisms
to demonstrate that CCR is untenable because applying the principle that
everything is criticisable to the principle itself leads to contradictions or
logical paradoxes of “the liar” kind. These arguments are at length discussed
by Bartley (in appendix “On Alleged Paradoxes of Pancritical Rationalism” to
2-nd edition of “The Retreat to Commitment”).
When I have read Bartley’s defence I
judged it successful and criticisms of Watkins and Post unsound because they
are based on confusing different meanings of word “criticizable”. After reading
Watkins’ article I have slightly reassessed this judgment. Now I think that
confusion mentioned is present in some Bartley’s own descriptions of CCR. Criticisms
of Watkins and Post are partially sound
as criticisms of strong CCR. But
as such they are superfluous, unneeded sophistications, – for strong CCR is
refuted by much simpler arguments. If taken as criticisms of weak CCR, Watkins’ and Post’s arguments
are unsound. Bartley’s defence of CCR against Watkins’ and Post’s arguments is
successful as defence of weak CCR.
Watkins in his latest discussion of
CCR (1987) explains what first raised his doubts:
“…I became uneasy about this generalisation
of Popper’s idea of rational acceptance within a scientific context to much
wider contexts, and the generalization of the idea of empirical testability
into that of criticizability. We are sometimes in a position to say
unequivocally of two propositions that one is, and the other is not, testable.
But can we ever say of two theories that one is, and the other is not
criticizable? Bartley once proposed that the important demarcation is not
between falsifiable and unfalsifiable but between criticizable and
uncriticizable theories. But would such demarcation effect a cut? How could the
criticizability of a theory be established?” (Watkins 1993, 271)
I share these doubts of Watkins, but
think they have little to do with CCR on its adequate understanding (as weak
CCR).
Really, CCR is not logical extension of Popper’s theory of scientific testability
as falsifiability into general theory of criticism.[1] Also, I agree with Watkins that
there is no criteria of criticizability; so no demarcation between criticizable
and uncriticizable theories is possible.
Watkins writes that “Bartley once
proposed that the important demarcation is not between falsifiable and
unfalsifiable but between criticizable and uncriticizable theories”. I guess
Watkins misunderstands and misrepresents what Bartley really proposed. I have looked
through Bartley’s mature works, including “The Retreat to Commitment”, 2-nd
ed., and found no such proposition. Instead, I have found something that seems
similar, but have really quite another meaning:
“…the
evaluatory problem is less to demarcate scientific from nonscientific theories
as to demarcate critical from uncritical theories or from theories that are
protected from criticism – particularly pseudo-critical theories.” (Bartley 1982)
This quotation is preceded by
discussion of Popper’s attempt “to exclude from science those
theories (often ones which claim or aspire to scientific status) which have built-in devices for avoiding or deflecting critical arguments – empirical or otherwise.” (Bartley 1982)
So, Bartley was not proposing demarcation between criticizable and uncriticizable
theories. Instead, he discussed Popper’s attempt to demarcate critical from uncritical theories, where 'uncritical' means not
'uncriticizable', but “protected from criticism”, “which have built-in devices for avoiding or deflecting critical arguments”.
As for “demarcation between
criticizable and uncriticizable theories”, Bartley’s view is that no such
demarcation is possible: “…I declare that all statements are criticizable…”
(Bartley1984, 223). There is no way to determine
criticizability of a theory, except in comparative terms – that theory X is
more/less criticizable than theory Y (for example, in some sense, empirical
theories are more criticizable than metaphysical, metaphysical statements more
criticizable than analytic; clearly stated theories are in some sense more
criticizable than obscurely stated).
Then, Watkins discusses Bartley
statement (which I discussed above as definitive
for strong CCR) that unjustifiable position can be held rationally provided
that it is held open to criticism and survives severe criticism. As CCR states
that such rationality can be comprehensive and is applicable to CCR itself, the
question is: “how might one set about trying to criticize it?” Watkins, 272)
This question contains implicit
assumption which is characteristic also for Post’s argumentation. As Bartley
describes it:
“…essential to Post’s discussion, as
he has pointed out, is his claim that criticizability is a semantic property of
statements in the sense that for a statement to be criticizable we must be able
to specify its “potential criticizer”.” (Bartley 1984, 234)
Bartley answers that it is unclear
what is meant by “specification” and how it can be used to demarcate
criticizable statements from the uncriticizable. To identify rationality with
ability to produce such specification – if you can’t specify potential
criticizers for some view you hold, this means that you hold it irrationally
and uncritically – would be unreasonable. In any case, it is not the view of
rationality/irrationality which is meant by CCR. On CCR view, inability to
specify a potential criticism (whatever it is) have nothing to do with
irrationality:
“Lack
of interest in potential criticism, or hostility
to potential criticism, would be quite another thing: that would indeed
probably mark a dogmatic or at least uncurious attitude. What I had in mind
when writing of pancritical rationalist was one who holds his claims open to
review even when – and particularly when – he is unable to imagine, let alone
specify, what would count against them.” (Bartley 1984, 234)
Generally, outside science, it is
often impossible to say what kind of argument would do as sound criticism
before the argument is produced (and if such argument is produced, the view it
criticizes is refuted):
“It is, of course, valuable and
important, whenever possible, to
specify in advance what sort of things would count against a theory. It is also
characteristic of the evaluation of scientific theories that one can quite
often make such specifications in advance… But I was in no way restricting
myself to science… I was concerned with a broad range of ideas, with religion,
ethics, theory of value, and metaphysics, as well as with science. In this
broader domain there is not the slightest reasonable hope of always being able
to specify potential criticisms in advance, although one may try even here to
specify the sorts of things that
would be critically effective. Yet there is all the more reason, in such
circumstances, to continue to hold such theories as open to criticism.”
(Bartley 1984, 235)
Pancritical rationalist is not
obliged to specify the kind of argument which would count as sound criticism of
CCR. Really, he presupposes that there is no such sound criticism, while admitting
that he can be mistaken:
“the onus is on the critic to
produce it.” (Bartley 1984, 120)
It is possible to propose some
general lines of possible criticisms for CCR, but they are very general. Some
of them were outlined by Bartley. These lines reflect the structure of CCR,
admissions it contains. You can try to criticize each of these admissions. For
example:
– you can try to show that there are
some views that can’t be hold open to criticism;
– you can try to show that
justification and criticism are inseparable in principle;
– you can try to show that basic rule
of logic – rule of non-contradiction, which is presupposed in any rational
discussion, – is unsound;
– you can try to show that the
attitude of CCR is not conductive for truth;
– you can try to criticize
CCR-attitude on ground of some other ethical values.
Next line of criticizing CCR,
proposed by Watkins, suggests that CCR contains inbuilt anti-critical
strategies, so it is not really self-critical.
In his earlier article Watkins presented
it as general criticism of CCR as using “dictatorial strategy”. He contended that CCR can be defended against
any criticism by very simple trick: stating that, as there is a criticism of
CCR, CCR is criticizable.
This is indeed very strange
criticism. It is rude misinterpreting CCR into the theory that every view is
criticizable in the sense that someone can try to criticize it. CCR is not such a theory. CCR is the theory
that a person (pancritical rationalist) can held open to criticism each of his
views. There is no logical inference from “Someone criticizes X” to “I hold X
open to criticism”.
So, Bartley assesses
this Watkins’ criticism as “a deplorable argument”:
“If someone were to come forward
with a cogent argument against pancritical rationalism; and if I were then to
reply: "Oh, you see, that just goes
to show that I was right in saying that my position is open to criticism",
I would be laughed at. And we all know from Charlie Chaplin that one thing that
dictators cannot stand is to be laughed at.” (Bartley 1984, 242)
In his latest (1987) discussion of
CCR Watkins omits this older criticism but proposes similar, though more
limited objection against one Bartley’s suggestion about possible line of
attempts to criticize CCR. Bartley wrote that CCR would be refuted if someone
produced “an argument showing that at least some of … standards necessarily
used by a pancritical rationalist were uncriticizable …” (Bartley 1984, 120). Watkins
objects that “this line of criticism is not available to a would-be critics of
CCR”. Argument is as follows:
To criticize X is to show that X
have some undesirable property. On view of CCR uncriticizability is undesirable
property. So, any argument which intends to show that X is uncriticizable,
states that X have that undesirable property of “uncriticizability”, and so it
criticizes X and demonstrates its criticizability. (Watkins 1993, 272)
Taking into account objections to
this argument proposed by Agassi, Jarvie, Settles and Keke, Watkins agrees that
his argument doesn’t exclude the possibility that “there may be other lines of
criticism than one indicated by Bartley in his book”. (Watkins 1993, 273)
So, it seems that Watkins consider
his argument as sound against possibility of “line of criticism indicated by
Bartley in his book”. I think it isn’t. My objection is: Watkins takes word
“criticism” in too wide meaning referring to “some undesirable property”. Moreover,
it is not some indisputably undesirable property, but some undesirable property
from CCR viewpoint. If defenders of
CCR used word “criticism” in such meaning, they would be objected that this is obviously
unallowable, for it is direct begging the question (suppose, a would-be critics
of CCR consider uncriticizability desirable
property). Really, “line of criticism indicated by Bartley” means criticism in
much narrower meaning – referring only to truth/falsity properties: to
criticize theory X is to try to show that it is false.
Really, Bartley’s meaning it is yet
more narrow and definite. When Watkins quoted Bartley he omitted important part
of the sentence which explains what Bartley means by "uncriticizable"
“… that there too, something had to
be accepted as uncriticizable in order to avoid circular argument and infinite
regress.” (Bartley 1984, 120).
Taking this omission into account,
it is clear that what Watkins criticized was really not “line of criticism
indicated by Bartley in his book”, but something rather different.
Watkins’ next argument, proposed as
refutation of CCR, refers not to formulation definitive for strong CCR, but to
Bartley’s characterization of rationalist “as one who holds all his beliefs … open to criticism”. (Watkins
1993, 273) This is formulation of weak CCR, so assessment of Watkins’ criticism
is of importance.
Watkins begins with explanation that
his argument
“was based on two explicitly made
assumptions… The first was that, just as scientist cannot genuinely hold open
to falsification a proposition that is in fact unfalsifiable, so a rationalist
cannot genuinely hold open to criticism a proposition that is in fact
uncriticizable…” (Watkins 1993, 273)
I deny this assumption. Before
discussing it, let us make some convention to avoid confusion of different
possible meanings in which word 'criticizable' can be used. Let us designate
'criticizable-1' – held open to
criticism;
'criticizable-2' – can be successfully
criticized (its sound criticism exists).
Now let us observe that CCR (weak
CCR) is the theory about criticizability-1, not about criticizability-2.
Watkins’ assumption says that
criticizability-2 is necessary condition of criticizability-1. I think it is
mistaken. Supposed analogy between falsification and criticism doesn’t hold.
I must confess that I really don’t understand
what can be the meaning of statement that a proposition (statement, theory,
attitude) is “in fact uncriticizable”. In some sense, statements can be more or
less criticizable, but they are never uncriticizable. (And higher/lower
criticizability is not to be considered directly as index of goodness/badness.
It depends on the nature of statement, problematic area to which it pertains.
So, empirical theories are more criticizable that metaphysical or complex mathematical
ones; I am at loss how to compare criticizability of metaphysical and complex mathematical
statements, but statements of both kinds are more criticizable than simple tautologies.
But this doesn’t mean that we are not to hold any metaphysical views or even
tautologies. On the other side, in the same problematic area, more criticizable
theories are preferable. For example, more clearly formulated theories are more
criticizable than more obscurely formulated.)
Now what can “in fact
uncriticizable” mean? I can guess for possible meanings:
(A) in fact no criticism was proposed.
(B) in fact no sound criticism was proposed.
(C) no sound criticism exists.
(D) it is provable that no sound criticism exists.
Whatever is the option, positions (statements,
theories, attitudes etc.) which are “in fact uncriticizable” can be held open to criticism.
Really, suppose
(A) There is a position P
for which no criticism was proposed. Surely, it is possible that sound
criticism of P exists. So, P can be held open to criticism.
(B) There is a position P
for which no sound criticism was proposed.
Really, rationalist can’t hold P if he knows of some sound (on his
judgment) criticism of it! But rationalist is to admit that though he knows of
no sound criticism, such criticism can exist. So, he can hold P
open to criticism.
(C) Whenever rationalist holds some position
P,
he holds it on supposition that no sound criticism of P exists. Really, if such
criticism exists then P is false! If rationalist supposes
that V is true, he supposes that no sound criticism of V exists. But, at the
same time – and this is just what fallibilism and critical-pancritical
rationalism are about – rationalist
admits that it is possible that he is mistaken, that despite his belief
that P
is true it is really false and there is sound criticism of it. So, rationalist (1)
holds P on supposition that no sound criticism of P
exists, and (2) he holds P open to criticism as he admits
possibility that he is mistaken in (1).
(D) Now is it possible to hold some position
P
open to criticism, if it is logically proven by the argument A
that no self-consistent criticism of P exists? Yes, it is possible, and
easily indeed. (Pan)critical rationalist can say:
“Well, as far as I can judge, A proves that no self-consistent
criticism of P exists. But I admit the
possibility that I am mistaken and that really A is mistaken. I don’t
see mistake, but it can be there. So, if you propose some criticism of A,
I am willing to consider it. Now, as I admit that it is possible that A
is mistaken, I admit that it is possible that, despite A, self-consistent
criticism of P can exist. So, I hold P, as well as A, open to criticism.”
(C) and (D) are interesting in that
they reveal dual attitude of (critical) rationalist towards his views and
towards proofs (arguments) he judges sound. (C) shows that critical rationalist
(1) holds V on supposition that it
is true and no sound criticism of it exists;
(2) admits the possibility that V is
false and sound criticism of it exists.
(D) shows the same about proofs which
critical rationalist judges sound and, as consequence, to views they allegedly
prove.
This dual attitude is not to be
confused with logical contradiction. Without this dual attitude no criticism
could be effective: we could criticize only views we don’t hold, and so, no one
could change his views as result of criticism.[2]
Now Watkins’ and Post’s sophisticated
arguments are varieties of (D). If they are directed against weak CCR, they are
unsound. If they are directed against strong CCR, they are unnecessarily
sophisticated, for it is refuted by much simpler arguments. Let us consider
Post’s argument.
Post's argument
changed many formulations, but it can be presented, in most general form, as
follows:
The main principle
of CCR is:
All statements, theories, attitudes etc. are criticizable.
(1)
As (1) is a
statement, it follows that
(1) is criticizable. (2)
Now, argument goes,
(2) is a statement and, according with (1) should be criticizable. But as (2)
logically follows from (1) then any criticism of (2) would be also criticism of
(1). And so, any criticism of (2) would demonstrate that (1) is criticizable.
But this is exactly what (2) states. So, any attempt to criticize (2) would
demonstrate that (2) is true. But if it is so, it means that (2) is
uncriticizable.
As you can see, the
argument is supposed to prove that (2) can’t
be successfully criticized, i.e. that (2) is uncriticizable-2.
Now the answer to
this argument is that it doesn’t affect weak CCR at all. Weak CCR isn’t the
theory about criticizability-2, it is the theory and attitude about criticizability-1,
i.e. about holding statements, theories, attitudes etc. open to criticism.
To remove confusion
of criticisability-1 with criticisability-2, Bartley reformulates Post’s
argument changing “criticizable” for “open to criticism”. I think this reformulation still isn’t
enough, for statement.
All statements, theories,
attitudes etc. are open to criticism (1B)
is still misleading. “Openness
to criticism” isn’t something inherent in statements, theories, attitudes
etc. All statements, theories, attitudes
etc. are neither open nor closed for criticism. It is we who hold them this or
that way. Now the adequate reformulation of (1), to express the principle of CCR,
is:
It is possible to
hold all statements, theories, attitudes etc.
open to criticism (1S1)
or
Pancritical
rationalist holds all statements, theories, attitudes etc.
open to criticism (1S2)
Now if you reformulate (2) accordingly as
It is possible to hold (1S1) open to criticism. (2S1)
or
Pancritical rationalist holds (1S2) open to criticism (1S2)
these formulations won’t engender logical
paradoxes.
To emphasize, the attitude of
openness to criticism is as follows:
“I hold this view, for, in the light
of all arguments I know, I judge it tenable and the best among alternatives I
know. But I admit that I can be mistaken. So I am willing to consider your
arguments, and possibly you will succeed in persuading me to change my view.”
There is nothing in Watkins’ and
Post’s arguments which would demonstrate that pancritical rationalist can’t
apply this attitude comprehensively, – to every statement, theory, attitude
etc., including CCR itself.
The next misunderstanding is
especially relevant to our discussion. We can find it in Mariano Artigas’ description of the problem CCR is designed to
solve, and in his objection to CCR. These are as follows:
“His theory could help the
rationalists fight off the relativists, the sceptics and the fideists with
their reproach that ultimately all of us depend on some fundamental convictions
that cannot be proven and that all of us are guided ultimately by irrational
decisions.” (Artigas 1999, 95) “Bartley does not want to hear anything about
commitment, because he identifies commitment with blind faith in the sense of
fideism. There exist however, types of faith and of commitment that are
reasonable and have nothing to do with the blind faith of fideism.” (Artigas
1999, 99)
Now compare this with Bartley’s
explanation:
“I was confronting the contention …
that there is an essential logical limitation to rationality: the rational
defence and examination of ideas must,
for logical reasons, be terminated by
an arbitrary and irrational appeal to what can be called dogmas or absolute
presuppositions. …I argued: (1) that nothing of any interest can be
justified in the way required…; (2) criticism is nonetheless possible provided
one unfuses justification and
criticism…; (3) there are no limits to rationality in the sense that one must postulate dogmas or presuppositions
that must be held exempt from review…” (Bartley 1984, 221)
“What I did … was to
show that no authorities or justifiers in
this sense were needed in criticism. I separated the notions of
justification and criticism (for the first time explicitly) and showed that
criticism can be carried out successfully and satisfactorily without … any
resort to dogmas or authorities. That is, when I declare that all statements
are criticizable, I mean that it is not necessary, in criticism, … to declare a
dogma that cannot be criticized…; I mean that it is not necessary to mark off a
special class of statements, the justifiers, which do the justifying and criticizing but are not open to criticism…” (Bartley
1984, 223)
We can see that Artigas’ description
is much fuzzier than Bartley’s and can be interpreted in different ways. It fails
to capture the main point – that Bartley’s theory is opposed to arguments which
seem to prove that “the rational defence and examination of ideas must, for logical reasons, be terminated by an arbitrary and irrational
appeal to what can be called dogmas
or absolute presuppositions.” And “commitment”
against which Bartley fights is exactly decision to hold some views as dogmas or absolute presuppositions, exempt from criticism. So, there can’t be
commitments in this meaning which are
“reasonable and have nothing to do with the blind faith of fideism”. Surely, if
you chose to use word 'commitment' in
another meaning, then Bartley’s statements and arguments won’t be
applicable to it!
Really, the word “commitment” is often used in the
different meaning – that of deep devotion. You can be committed to some values
in this meaning, without being committed in Bartley’s meaning. To say for
myself, with respect to my dearest views I am still willing to know and
understand arguments of their opponents or enemies (at least, those that are
influential enough), and in the process I run the risk (however small I now
estimate it) to be persuaded by them. So I don’t consider my dearest views as
dogmas exempt from criticism, though I hope there is no sound criticism of them.
As for Bartley’s view on this point, may be
this quotation will do:
“...the claim that a rationalist need not commit himself … is no
claim that he will not or should not have strong convictions an which he is
prepared to act. We can assume or be convinced
of the truth of something without being committed
to its truth.” (Bartley
1984, 121)
4. Foundations Firm (Final) and “Swampy” (Revisable)
Joseph Agassi describes fideism (commitment-theory,
in the meaning criticized by Bartley) as follows:
“…there is no practical difference
between the fideist mode of arguing and other, except for one: whenever what
they happen to consider their initial basic axioms happens to be challenged,
they refuse to discuss it. …there is a basic difference between the fideist and
others: we all exempt some dogma from the demand to examine it rationally, but
the advocates of fideism exempt their own dogma consciously and systematically
and for good. … tu-quoque argument …
is used as an excuse to exempt from rational scrutiny some unspecified view, on
the basis of the observation that usually we all do so… But we all do so now
and then, whereas advocates of fideism do so come-what-may. This is why the
suggestion to keep one’s own initial axiom floating is so deadly to fideism…”
(Agassi 2012, 46-47)
I think this is quite a good
description, except that term 'dogma' is not applicable properly to those views
we don’t exempt from criticism “systematically and for good”; and if “axioms”
are “floating” (if I rightly understand the meaning of this metaphor), they
aren’t axioms.
Now we can see that fideism departs
from what “usually we all do” in that it considers some views as dogma, exempts
them from examination “consciously and systematically and for good”. What we
all usually do seems similar in that we don’t examine the great multitude of
our views; but there is principal difference in that we don’t exempt them from
criticism “consciously and systematically and for good”. If criticisms are
produced we can assess them and possibly reconsider our views. Surely,
sometimes we are unwilling to heed to criticisms, but this unwillingness is
easier surmountable while it is not “conscious and systematical and for good”. It
can be nearing to dogmatism, but is not dogmatism yet. In a sense, fideism seems
to excel usual way in self-awareness and consistency. But self-awareness and
consistency is not always a merit, it can be a great vice. I think (and suppose
all critical rationalists will agree) that this is just such case. Self-aware and consistent dogmatism is not to
be praised.
Now there is another – opposite,
adogmatic – way of higher self-awareness and consistency. It consists in keeping
all our views “floating” – in the meaning: open
for revision (if some arguments against them are produced) “consciously and
systematically and for good”. There is no dogma; all “foundations” are as
“floating” as everything. Surely, they are not to be changed without reasons we
judge good enough, but they are to be changed whenever such reasons are
produced.
But what are those “reasons good
enough”? How are we to judge if reasons proposed are good enough? The answer
is: there is no exhaustive, final, unrevisable answer. Any account of reasons
is as “floating”, revisable as everything.
Finally everyone judges for himself. In a sense, we never know how do we know.
******
What Bartley criticizes as commitment is
exactly decision to hold some explicitly marked off dogmas exempt from
criticism “consciously and systematically
and for good”, “come-what-may”
(in Agassi’s terms, fideism).
And defenders of
commitment-theory state that we all do so, but the way of higher awareness is
to make it consciously, with awareness, explicitness about what are those
dogmas you make commitment to.
The meaning of Bartley‘s theory, on my understanding,
is that commitment-theory misdescribes the logical situation. Really, everybody
holds great multitude of views without critical examination, and this is
inevitable; it is impossible to make critical examination on all our views and the attempt would lead
to infinite regress. But this doesn’t mean that we are logically forced to
commitment (fideism), in the meaning: “to mark off a special class of
statements” (Bartley 1984, 223) as basic dogmas (framework), and decide to hold
them exempt from criticism “consciously
and systematically and for good”, “come-what-may”.
If we don’t do this “consciously and systematically and for good”, we don’t do this at
all. I can hold the great multitude of my views without their critical
examination until I have run into strong criticism of it. And if I have run
into it, I am to consider criticism seriously, try to evaluate it as unbiasedly
as I can, and judge if it is strong enough to refute the view criticized. It is
always my judgement, there is no logical forcedness in it (except in rather
rare situations when it is persuasively demonstrated that the view is
self-contradictive).
Commitment-theorists describe the situation as
if there is no escape from fideism (commitment). The only difference is that
there are fideists with full awareness of it, and fideists unawares. And surely
it is better to be the first than the second. Now Bartley’s argument shows that
this is misdescription: there are really three possibilities.
The first is common, not philosophizing way; it
is not to be properly described as
“fideism unawares”, – for you can move from it to higher awareness (with
respect to "commitment" or "fideism") and consistency in
two opposite direction – 1) that of fideism ("commitment"), if you
dogmatize some views “consciously and
systematically and for good” or 2) pancritical rationalism, if you
consciously reject fideism ("commitment"). Despite arguments of
commitment-theorists, the second, consistently non-dogmatic way is logically
possible.
So,
pancritical rationalism can be properly described as
1) rejection of justificationist theory of rationality, renouncing the
search for final justification and authoritative source of knowledge, identifying
rationality with critical attitude
plus
2) conscious
rejection of the proposition of commitment-theorists (fideists) – conscious
decision not to make some views
exempt from criticism “consciously and
systematically and for good”, “come-what-may”.
******
In e-mail correspondence with me,
J.Agassi raised an objection against the word “comprehensively”: “I have yet to meet a person whose openness
to criticism does not meet with some exception, proviso, stubbornness etc.”
I think this objection doesn’t hit the mark. The word “comprehensively” is used by Bartley
(and by me) extensively, not intensively, – not in the meaning “perfectly”, but
in the meaning “applicable to everything”. Stating that critical rationalism can be
comprehensive Bartley means only that there is no logical necessity to hold some
views exempt from criticism “consciously and systematically
and for good”, “come-what-may”;
in this meaning we can held each of our views open to criticism; this openness
to criticism can be comprehensive (but not perfect).
CCR doesn’t mean that our openness to criticism
with respect to each view can meet some highest standards – really, I have no
idea as to what those standards may be.
If Smith produces some argument which seems
persuasive to Smith and John doesn’t accept it, Smith can judge John stubborn,
while John judges Smith’s argument unpersuasive.
Really, as a matter of psychology of
rationality, I think that some amount of what can be called “stubbornness” is a
good thing, – in the sense that you mustn’t be too easily renouncing your views, or views you intuitively feel interesting and promising; even if at the
moment you can’t answer well some criticisms. Especially with entrenched moral
views, you need time to rethink things better, to estimate total significance
of the possible change under consideration. Also about scientific hypotheses
which you feel promising. What is important – and what is proper meaning of
openness to criticism – is not to shun criticisms, try to understand them
correctly in their real meaning, without their distortion and picking on words
(which are means to avoid understanding). And then, taking into proper account all criticisms, you judge if it
still seems advisable to stick to the moral view under consideration, or if the
scientific hypothesis still looks promising, – or further sticking to them is just stubbornness. If you do this way, you make your mind open to
criticisms; indeed, you have already let them in. If criticisms are really strong and you are
willing to be open and fair about them, they will gnaw their way in proper
time.
For me, there is no judge besides myself. But
there is the choice for me (and for everyone):
either
1) I can be willing to consider arguments seriously, to understand them
in their real meaning and judge their soundness as unbiasedly as I can. This
means that I risk my views: if arguments against them will turn out persuasive
for me, those views they criticise will be abandoned.
or
2) I can decide to hold some views exempt from criticism “consciously and systematically and for good”,
“come-what-may”. On this choice, if
criticisms are produced, I shun them, or reject without real attempt to
understand them in their real meaning and judge their soundness unbiasedly.
These are just two opposite
conscious personal attitudes, – not algorithms of behaviour in concrete cases
(though surely, your consciously taken attitude greatly influences your
behaviours in concrete cases).
******
Critical rationalism is sometimes
described as philosophy of “rationality without foundations” (Gattei 2009). I
would rather describe it as philosophy of rationality without firm, final foundations – philosophy of
rationality which takes all "foundations" as conjectural, open to
criticism, revisable.[3]
It is just what Popper has written
about science:
“The empirical basis of objective
science has thus nothing "absolute" about it. Science does not rest
upon solid bedrock. The bold structure of its theories rises, as it were,
above a swamp. It is like a building erected on piles. The piles
are driven down from above into the swamp, but not down to any natural or "given"
base; and if we stop driving the piles deeper, it is not because we have
reached firm ground. We simply stop when we are satisfied that the piles
are firm enough to carry the structure, at least for the time being. My term
'basis' has ironical overtones: it is a basis that is not firm.” (Popper 2002, 93-94)
5. What Justification Is Denied?
“The principles that form the basis
of every rational discussion, that is, of every discussion undertaken in the
search for truth… 1. The principle of fallibility: perhaps I am wrong and
perhaps you are right. But we could easily both be wrong. 2. The principle of rational discussion: we
want to try, as impersonally as possible, to weigh up our reasons for and
against a theory… 3. The principle of approximation to the truth: we can
nearly always come closer to truth in a discussion which avoids personal
attacks…” (Popper 1992, 199)
One rather usual misunderstanding of
Popper’s and Bartley’s non-justificationism is interpreting their denial of
possibility of justification too widely, as denial of the possibility of anything
that can be called “justification”, in whatever weak sense, as denying any supporting,
favourable arguments, any “positive” (in the sense “favourable”, “pro”) reasons.
This understanding of the thesis
that no view (theory) can be justified is mistaken. It stems from the confusion
of two meanings in which it can be said that theory is (or is not) justified.
The first one is the meaning of absolute
– conclusive or probabilistic – justification, i.e. logically valid
demonstration that a statement under consideration is true or that its truth is
highly probable (in the sense of probability
calculus). This demonstration must be a logically valid argument without unjustified premises. But such
argument is impossible because of infinite justificatory regress. So, absolute –
conclusive or probabilistic – justification is never possible.
The second one is what can be called
justification in fuzzier common language meaning. In this meaning, whenever we
formulate some reasons why we think the statement (theory) is true or
preferable to alternative statements (theories), we justify the statement
(theory). In this fuzzier meaning, justification is often possible, and all our
reasoning is such fuzzy, tentative justification, justification on some
tentative (unjustified but plausible for us) assumptions. (In this sense, any
criticism of the statement S is tentative justification of the statement not-S.)
For example, when Bartley argues in
favour of CCR (and when he criticizes its alternatives), he surely justifies it,
in the meaning of fuzzy, tentative justification. And he doesn’t justify it, in
the meaning of absolute (conclusive or probabilistic) justification.
Surely, we can reserve the word
“justification” for the strict logical meaning, and use some other word – for
example “support” or “commendation” – for fuzzier common language meaning. But
to avoid confusion and misinterpretation, we need to be clearly aware of this
reservation.
One probable reason of
misunderstanding described above is mistaken interpretation of CR(CCR) as
generalization of Popper’s falsificationism. This mistake is made by Bartley
himself. To quote: “in 1960, I proposed to contrast justificationist and
nonjustificationist theories of criticism as a generalisation of his own
distinction between verification and falsification” (Bartley 1990, 237)
Roughly it looks like this. Popper
argues that scientific theories can’t be verified (confirmed) by observations so
that it would establish their truth or high probability, but can be falsified so
that it would establish their falsity. CR(CCR) generalizes this into arguments
that no positions can be justified, but any can be criticized. This interpretation
of CR(CCR) is mistaken: there is no such generalization; two cases are quite non-analogous.
In important sense, relation between
criticism (refutation) and support (justification) in general case is very different
from relation between falsification and verification in special case of empirical
sciences. So, analogy between falsificationism and critical (pancritical)
rationalism is only partial. To understand this, we need to recollect why, on
Popper’s account, falsification plays so decisive role in development of
empirical sciences, while verification plays no.
First, we need to distinguish three
types of statements with which empirical sciences work:
1) universal statements – scientific
theories;
2) singular statements which
logically follow from universal statements – predictions;
3) singular statements which are
obtained as result of properly made repeatable observations – observational
reports.
Falsificationism is the theory that there can be no logically valid verification
(confirmation) of the theory by observational reports, while there can be
logically valid falsification (refutation) of the theory by observational
reports. Scientific progress is achieved by discarding falsified theories and
replacing them by new, which stand attempts of falsification and, so, are
possibly true or near enough to truth.
Now this is so because of two
factors:
(1) Observational reports have in
empirical science sort of privileged status – that of “basic statements”. Though
“basic statements” aren’t final, though they are as hypothetical as everything,
they are given sort of preference in science: if properly obtained
observational reports conflict with predictions, observational reports “by
default” are considered true and predictions falsified; if truth of
observational reports is doubted, the burden of proof belongs to doubters.
(2) There is logical retransmission of ‘false’ value in
direction from predictions to theories, while there is no such logical retransmission
of ‘true’ value. If some predictions which logically follow from some theory
when checked happened to be false,
then it logically follows that the
theory is false. But if all
predictions which logically follow from some theory and which were checked until now happened to be true, then it logically does
not follow that the theory is true.
This is so, because the number of predictions which logically follow from the
theory is infinite, while the number of checked predictions, however large, is
always finite. So, infinite number of predictions always remains unchecked.
Observational reports (basic
statements) can falsify predictions, and falsity-value is transmitted from
predictions to theories from which they follow, – and so theories are
falsified. On the other side, if observational reports (basic statements)
verify predictions, truth-value is not transmitted from predictions to theories
from which they follow, – and so the theories can’t be verified.
Now how about general case of
criticism? Of two factors discussed above, in general case but one remains.
(1) There is no analogue of privileged
“basic statements”.
(2) Retransmission of ‘false’ value and
absence of retransmission of ‘true’ value remains. If you have statement or
theory S and multitude of statements which logically follow from it {s1,
s2, … sn, …}, then falsity of any of these means falsity
of S, while to establish truth of S on the base of its inferences you need that
all these inferences were true, and usually it is impossible to make such
assessment, for their number is infinite. If Y logically follows from X and we
judge Y false, then we are to judge X false too (falsity of X logically follows
from falsity of Y). But if Y logically follows from X and we judge Y true, then
we still have no reason to judge X true (truth of X logically doesn’t follow
from truth of Y).
Absence of analogue of scientific
“basic statements” in general case makes “asymmetry” between negative and positive
argumentation weaker than in science. One of two main logical reasons pertaining
to the theory of falsification is lacking in the general case of criticism.
Nevertheless, one which remains makes negative reasoning more efficient than
positive. It is the way how we can discover and eliminate errors in our
theories. In this respect, there is asymmetry both in specific case of
scientific theories and in general. But in another important respect, asymmetry
doesn’t hold.
That stronger sense is particular
for theories (universal statements) of empirical sciences: they can be empirically
falsified, but can’t be empirically verified. This asymmetry is not
generalizable into statement about all positions (statements, theories,
attitudes etc.) that they can be properly criticized (refuted) but can’t be
properly supported (justified).[4] In this respect, relation between criticism
and argumentative support is symmetrical. If both are taken in strong sense of
strict, conclusive refutation and strict, conclusive justification, they both
are logically impossible. If both are taken in weak sense of fuzzy,
inconclusive criticism and fuzzy, inconclusive support, they both are possible.
No wonder, for any criticism of X equals to support of not-X.
Bartley’s arguments – that
justification is impossible while criticism is possible and can be comprehensive
– are sound. But this result has nothing to do with asymmetry between
favourable and unfavourable arguments which would be analogous to asymmetry
between verification and falsification. In general, there is no such asymmetry between
favourable and unfavourable arguments. The asymmetry reflected in the statement
that justification is impossible while criticism is possible is much less
interesting, for it is verbal; it is the asymmetry between use of word "justification"
in the strong meaning (of absolute, conclusive or probabilistic, justification)
and use of word "criticism" in the weak meaning of tentative
unfavourable argument.
CCR doesn’t mean denying possibility
of justification in all possible meanings of the word – it means
1) negation of justificationism – rejection
of unfeasible justificationist demand to
accept no position without justification; acknowledgement that no position
can be justified in compliance with this demand, that we can’t avoid accepting
some (great many) unjustified positions and use them in our arguments
and
2) negation of “commitmentism” or
fideism – refutation of the view that since it is impossible to avoid accepting
some unjustified positions, we must accept them dogmatically, as exempt from
criticism; explanation that all our positions can be held open to critical
discussion and revision.
There is quite a clear explanation
in Popper’s book “Realism
and the Aim of Science”, chapter 1, section I. Here Popper explains that he
denies possibility of 'justification' and 'positive reasons' in the meaning: providing-certainty-or-high-probability reasons. Surely, this doesn’t mean
impossibility of 'positive
reasons' in the meaning of arguments in
favour, reasons for tentative acceptance. Really, Popper admits that we can often
provide such tentative reasons, reasons
for preference and proposes to call them 'critical reasons'. Also, Popper admits:
“Giving reasons for one’s preference
can of course be called a
justification (in ordinary language). But it is not justification in the sense
criticized here. Our preferences are 'justified' only relative to the present
state of discussion …” (Popper 2005, p.20)
I think it is appropriate to defend
some sort of positive (in the meaning 'in favour') reasoning against some kind
of pseudo-critical attitude. We are not to criticize for criticizing sake;
criticism is (efficient) instrument in search for truth and in making rational
preference between alternative theories. To make such preference we are to be receptive
to “pros” as well as to “contras”. Both – purely negative (captious, ungenerous
criticism which fails to admit merits of the view under discussion) and purely
positive reasoning (apologetics) – are not conductive to truth. The hindrance
is that most people usually tend to positive reasoning when views they already
hold are concerned, and to negative reasoning when discussing views which
contradicts those they hold.
So, the most reasonable way is to
try to consider, as unbiasedly as you can, all known alternatives, all ‘pro’
and ‘contra’ you can find for each of them, and then make your judgement as to what
seems more truthlike (if you consider theories or other factual statements) or
better (if you consider attitudes) in the light of these reasons. And consider
your judgement tentative, open for reassessment whenever new arguments or new
alternatives are proposed. This is how I understand critical rationalism, in
most general terms. And I think Bartley was right that it can be held
comprehensively. Critical (pancritical) rationalism is comprehensive
fallibilism.
On the other hand, there can be more
to critical (pancritical) rationalism than mere open-mindedness, willingness to
consider and weigh unbiasedly known pros
and contras to known alternatives. There
is important active component to be added: search for mistakes to be eliminated
and for discovery of new alternatives. And here criticism plays crucial role
while supports are useless.
To summarize:
With respect to task of making
rational choice between available (known)
alternatives, arguments ‘pro’ (criticisms) and ‘contra’ (supports) are on equal
logical footing. But criticisms do more. They are necessary means and drives for discovery. They show
faults in the best of available (known)
theories and push the search for new, better theories.
PS. Mark Notturno on ‘refutation’, ‘justification’ and ‘criticism’
After finishing the draft of this
article I have read Mark Notturno’s book “Science and the Open Society”, where I
have found a good elucidations. In particular, Notturno explains:
“There are senses of ‘refutation’, for example, in which someone refutes
a statement not by proving that it is false, but by contradicting it, or simply
by denying that it is true. And there are senses of ‘justification’ in which
someone justifies a statement not by proving that it is true, but by showing
that it is a logical consequence of other things that he asserts.” (Notturno
1999, 126)
And Notturno remarks that because of
this ambiguity Popper’s statements about ‘justification’ and ‘refutation’ are
easily misunderstood and were in fact misunderstood…
To make things clearer Notturno proposes
to understand ‘criticism’ not as refutation or argument against, but as “the
setting of problems” and “forcing us to choose”:
“We set problems by showing that our beliefs
are contradictory and cannot all be true. And we try, in this way, to force
ourselves to reexamine our beliefs and to choose between them. Criticism can
never force us to reject any particular belief as false or to accept any
particular belief as true. But it can force us, if it is effective at all, to
recognize that our beliefs are contradictory, and to reexamine them, and to try
to decide how we should revise them in order to remove the contradiction. This,
in my view, is a large part of what the learning process is all about.” (Notturno
1999, 58-59)
This proposition shifts the meaning of the term
'criticism' from negative (argument
against) to neutral, fit to encompass both arguments pro and contra. On such
understanding, there is no use for 'justification', except in the sense which
is already incorporated by 'criticism':
“Critical thinkers do not justify
their beliefs; they criticize them.
They question and test the beliefs that others take for
granted. In so doing, they oftentimes clarify how some of their beliefs are
based upon others. But this basing of beliefs one upon another must ultimately
end. And if you are a critical thinker, then you will, somewhere in the course
of your tests, inevitably come upon statements that you believe for no other
reason than that they seem true – to you.
In such a case, it would be more accurate to say not that such statements are
justified, but that they seem, in your
judgment, to be true.” (Notturno 1999, 147)
Bibliography
Agassi,
J., 2012. Critical Rationalism,
Comprehensive or Qualified: The Popper-Bartley Dispute. In G.Franco, ed.,
Der Kritische Rationalismus als Denkmethode und Lebensweise: Festscrhift zum
90. Geburtstag von Hans Albert.
Artigas, M. 1999. The Ethical Nature of Karl Popper’s Theory
of Knowledge.
Bartley, W.W., III. 1964.
Rationality versus the Theory of Rationality. In M.Bunge, ed., The Critical Approach to Science and
Philosophy.
Bartley, W.W., III. 1982. The Philosophy of Karl Popper. Part III.
Rationality, Criticism, and Logic. Philosophia
11 (1-2):121-221.
Bartley, W.W. 1984. The Retreat to Commitment.
Bartley, W.W., III. 1990. Unfathomed Knowledge, Unmeasured Wealth.
Gattei,
S., 2009. Karl Popper’s Philosophy of
Science. Rationality without Foundations.
Gellner,
E. 1994. Conditions of
Lacatos,
Musgrave, A. 1991. What is Critical
Rationalism? In A.Bohgen, A.Musgrave, eds., Wege
der Vernunft: Festchrift zum siebzigsten Geburtstag von Hans Albert.
Tübingen: Mohr.
Notturno, M. 1999. Science and the Open Society.
Popper, K.
1977. The Open Society and Its Enemies. Vol.2.
Popper, K. 1992. In Search of a Better World.
Popper K., 2002. The Logic of Scientific Discovery.
Popper, K. 2005. Realism and the Aim of Science.
Watkins, J., 1993. Comprehensively Critical Rarionalism: a Retrospect. In
G.Radnitzky, W.W.Bartley, III, eds., Evolutionary
Epistemology, Rationality, and the Sociology of Knowledge,
* I thank Joseph Agassi, Mark Notturno and Alan
Musgrave for helpful comments and discussions on earlier draft of this article
[1] I discuss some aspects of
this question in section
“This concept is, however, itself
left unspecified and hence is so vague as to be practically useless. Moreover,
since all sorts of theories are in principle capable of functioning as
potential criticizers … and since we cannot know in advance what theories we
may get, the concept of potential criticizer is not only vague but also appears
to be empty. I conclude that the attempt to generalize Popper’s methodological
concept of potential falsifier into the
concept of potential criticizer is a cul-de-sac.
It does not help us to present Bartley’s position more clearly: that the
rational way of dealing with any position is to hold it open to criticism –
including this principle itself.” (Radnitzky 1993, 304) “…it is not possible to
paraphrase the basic principle of pancritical rationalism with the help of the
concepts of criticizability and potential criticizer.” (Radnitzky 1993, 309)
[2] In similar meaning, Ernst Gellner writes about “Liberal
Doublethink”. Term “doublethink” comes from G.Orwell’s famous anti-utopia “
[3]surely, all these
"foundations", "buildings" etc. are just loose metaphors –
apt for some purposes but not for others
[4] This
ungeneralizability is directly connected with ungeneralizability of Popper’s
concept of “potential falsifier” into the concept of “potential criticizer”,
pointed out by G.Radnitzky, – see footnote 2.)