Raz. 98. Disagreement in Politics

DISAGREEMENT IN POLITICS
J. RAZ1
A. The Setting
Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

In what ways should the structure of states and their policies be affected
by disagreements among their inhabitants on the foundations of political
authority, and the principles of the constitution, including principles of
restraint in the face of disagreement?2
Two reactions to our question should be avoided. The first is common to
those who think that the question is misconceived. Politics to them is about
the way collective action is generated given people's conflicting interests.
The search for an appropriate response to disagreements over principles is,
so far as they are concerned, neither here nor there. We should consider
other people's interests, not the principles they, rightly or wrongly, uphold.
But the answer to the question about the proper response to conflicting
interests has to await an answer to our question about the implications of
disagreement about principles. For plainly, collective action should be just
and fair to those affected by it. Given people's conflicting interests this is
not easy, and it requires taking a position on ethical issues. And that
requires, among other things, a position regarding our question: What is the
ethical implication for political action of the fact of disagreement about
principles?
In recent years, disagreement about principles was much discussed by
political theorists. It sometimes seems as if many writers assume that this is
the key question regarding political conflicts. No one supposes that all
conflicts are disagreements about principles, but perhaps such disagreements
lie at the root of the most intractable conflicts, or are the hardest to know
what to do about. This is the second, opposite, mistake that I wish to warn
about at the outset, especially as it is not a trivial mistake.
1. Shorter versions of this paper were presented at a seminar I taught jointly with J.
Finnis at Oxford and as lectures at McGill University and the University of Notre Dame. I
am grateful for questions and comments from participants on those occasions, in particular
from Sarah Stroud and Cathleen Kaveny. I am also grateful to Thomas Pogge for penetrating
and helpful written comments.
2. For the sake of brevity I will refer to disagreements on these issues as disagreements
over principles.
25
26 THE AMERICAN JOURNAL OF JURISPRUDENCE (1998)
Political conflicts are often conflicts about the allocation of scarce

resources among people who share belief in the same political principles.
They may dispute the way shared principles apply to the parties, or the facts
which determine their application. There can also be disputes where the
conduct of the parties is the ground on which, according to the shared
principles, their fate is determined. The allocation may be made sensitive to
the respective votes of different parties, or to the intensity of their feelings,
manifested in demonstrations, petitions, strikes, and the like. In other words,
political and ethical principles may make the just outcome depend on the

outcome of conflicts, or of aspects of conflicts. Principles of justice may
determine procedures for allocating benefits and creating entitlements whose
just allocation depends on the outcome of conflicts. The winners in the
contest win control of the prescribed procedure, and may then follow their
own goals, and allocate themselves the benefits they fought for.
If Scotland will find itself in conflict with England in years to come, this
is not necessarily because the Scots and English believe in different political
principles, nor will the absence of such disagreement on principles guarantee
that the dispute will be easy to solve, or that its consequences will not be
disastrous to both parties.
Yet the tendency among writers on theory to overemphasize the centrality
of disagreements over principles is significant, both in its effects and in its
presuppositions. Not uncommonly, people in dispute seek to reinforce their
position by appropriating any argument which appears to support their
cause. The result is a regrettable escalation of the conflict with each side
becoming further entrenched, fortified by self-righteous conviction in the
superiority of its claim. This process of mutual entrenchment is particularly
vicious when moralized, when high principles are enlisted in the cause of
the conflicting parties. Principles are notoriously hard to compromise, and
the results can be really disastrous.
The Western democracies are currently going through a moralizing phase,
manifested not only in the growth of fundamentalist religious tendencies and
in the extremes of armed libertarianism, but also in the spread of political
correctness, militant feminism, and the popularity of zero-tolerance policies.
Is the growing faith in some academic circles in the power of theory to
provide answers to all questions part of this trend? It may well be.
That leads to the other significance of exaggerating the importance of
disagreements about principles. It is fed by a general surge of faith in theory
which is sweeping philosophy, and not only moral and political philosophy.
Those who are old enough to remember pre-Chomskyian times may well
marvel at the extent of the transformation in mood, and at the slenderness
of the reasons for belief in its soundness. Marvel is no substitute for
J. RAZ 27
argument, no firm ground for well-founded doubt. But I will have to leave
it at that and return to my subject.
B. What Is Special about Our Problem?

That having been said, I hope that my exclusive concentration on this
occasion on disagreements about the principles of politics will not be
misunderstood. The question of the ethical implications for political action
of the fact of disagreement about principles is real and important. The
question does not invite an escape from judgement. Yielding to consensual

views is not a way to avoid relying on one's own views on hard moral
issues. To come to the view that one should rely only on consensual
principles or that one should not take any notice of disagreement, to give
but two extreme examples, is to come to a moral view and to rely on it.
There is no other way. So long as we are rational in our actions, each one
of us can act only for reasons we believe to justify our action. Deferring to
the consensus is no exception. If it is something we should do, to some
degree, or on some issues, then we will be able to do so3 only if we come
to believe that that is the right thing to do. In deferring to the judgement of
others we are still acting on our own judgement. Moreover, it is likely that
the decision whether to defer, and when and how much, will involve many
of the considerations which figure in discussions of the justification of
authority, democracy, judicial review, of the nature of justice, and all other
ethical issues that politics raises.
In answering the question of the proper reaction to disagreements, we are,
inescapably, providing an answer from the perspective of one point of
view,4 one which we hold because we believe it to represent things as they
are. That is the crucial point: we rely on our answer to the question not
because it is our view, but because it is, as we believe, true. If our reply to
the question of the significance of disagreement is that we should refrain
from this action or that because it is controversial, i.e. because others
believe it to be wrong, or unfair, or unwise, then we refrain from that action
not because the views of those others are true, but simply because they are
held by them, because they are their views, be they true or false. To the
extent that that is our answer we give other people's views a weight which
we do not give our own. Or, to be more precise, we give the fact that other
3. Reliably and with understanding. Obviously, we may be able to do so accidentally

and without understanding in some circumstances, and possibly we should cultivate them. I
will not explore such avenues.
4. "Point of view" here means simply a body of beliefs. It carries no relativistic or
perspectival implications.
people hold whatever views they have some weight, whereas we give none
to the fact that we hold any view.5
The difference has significant implications. That we have certain views
is not, for us, a reason for anything. Our reason is that, as we see it, things
are so and so. Naturally, we may be wrong. If things are not so and so, then
we are wrong and we have no reason for the action we contemplated. If our
beliefs are false, we believe that we have reasons, but we do not. We must
admit this even while we believe them to be true. If I believe that abortion
is murder, then I believe that there is a reason for prohibiting abortion. That

reason is that it is murder. If that is not so, then there is no reason to
prohibit abortion (unless there are some other independent reasons for doing
so). I know that even while I believe, in my example, that abortion is
murder. But if you believe that abortion should be prohibited out of
deference to the general belief that abortion is murder, then, if you are right
that politics should defer to the common view, there is a reason to prohibit
abortion if that is the common view, regardless of whether it is true,
regardless of whether abortion is murder. There is an asymmetry here
between the role that my beliefs and those of others play in my reasoning.
Giving weight to the views of others is not treating them the same way I
treat myself. It is giving them, i.e. their views, greater weight than the
weight I give myself. That is why deferring to others is not the obvious
thing to do.
True, in deferring to others I am also implying that others should defer
to me. If I think that I should refrain from acting on my views simply
because others disagree, I am committed to thinking that others should
refrain from following the truth simply because I, among others, disagree.
In this respect, I am not favoring others over myself. But that does not make
deferring to a consensus any more plausible. It merely suggests a new way
of stating the difficulty: If I do not think that the fact that I hold a view is
a reason for me to follow it, why should I think that it is a reason for
others? The answer to the question of the right response to the existence of
widespread disagreement over principles is far from obvious.
C. The Minimal-State Solution

One answer is to remove politics from areas of disagreement. How can
that be done? A strategy which recommends itself to many is to reduce our
5. As will be obvious, my discussion concerns reasons for actions, and not epistemic
reasons. That, having carefully considered the problem last year, I believe that so and so, may
be reason (for me and others) to believe that so and so. Epistemic reasons will be considered
in section J below.
J. RAZ 29
expectations of the state. The more we expect of the state the more acute the
problem becomes, the argument goes. If we expect the state to provide us
with education, health care, including the banning of life styles and of
conditions of work which involve health risks; if we expect the state to
guarantee the welfare of the poor, the unemployed or unemployable, the ill,
and the elderly, to protect people's privacy and their reputation; if we expect
it to regulate construction and traffic, the environment, and the civility of
its people, then we multiply the political issues on which people disagree,
and we aggravate the problems of the proper response to disagreement. The

solution is to roll back the frontiers of the state. There will be little room for
disagreement if all the state is expected to do is to protect people's property
from wilful damage and theft, and their bodily integrity from assault.
Perhaps even on such matters there will be disagreement. This is a matter
where opinions are liable to differ from country to country. So let us
stipulate that the expression "the minimal state" will refer to a state whose
jurisdiction is limited to the application of principles which are not in
dispute among its own people.6 We could respond to disagreement by
rolling back the frontier of the state to the borders of the minimal state,
whatever they may be in this country or that.
Yes, indeed, we will reply. If people's convictions are thus limited, then
political disagreement will be greatly reduced. But they are not so limited.
People disagree about the advisability of rolling back the state whether in
order to avoid disagreement over principles, or for other reasons. Therefore,
as a global answer to the question of the right response to disagreement this
answer fails. It says: the state should avoid taking action whose justification
rests on disputed principles. Therefore, it should confine itself to the
implementation of principles regarding which there is no dispute. This
answer fails to notice that it is itself a disputed principle. Therefore,
following it will not be right in its own terms. By its own light it should not
be followed until there is no dispute about its correctness, which in effect
means not in the remotely foreseeable future.
The minimal-state response to disagreement is contingently self-defeating.
If the principle is true, men it dictates that it should not be followed when
its truth is in dispute. Given that its truth is in dispute it follows that to
comply with it we should ignore it. This does not mean that the minimal-
state ideology is not attractive for other reasons, nor does it mean that
avoiding controversial principles may not be right in some circumstances.
It may be right for the state to avoid certain policies because they rest on
6. And this covers not merely substantive principles like "thou shall not steal," but
jurisdictional principles, like "the state may punish for theft" as well.
disputed principles, even though the decision to avoid them because they are
in dispute is itself disputed. That second dispute (about the decision to
avoid) may matter less in some ways than the first (the dispute about the
policies). All that my argument showed was that the minimal-state solution
is contingently self-defeating when taken to be an absolute and comprehen-
sive response to disagreement. Furthermore, given how things are, i.e. given
that this solution is in dispute, it does in fact defeat itself. Since its wisdom
is indeed disputed it recommends its own rejection.
Even if the improbable happens and the principle is not in dispute it

cannot be relied upon. States can avoid certain issues. But there are some
matters from which it is absurd to exclude states. Such, for example, is the
prohibition of murder. But so long as that is within their jurisdiction states
cannot avoid deciding about war and peace, about abortion and euthanasia,
about in vitro fertilization and embryo experimentation, and so on.
Should those who believe in excluding controversial state action conclude
that where the program of the minimal state is in dispute no state can be
legitimate? Strictly their argument is not that no state is legitimate, but that
no state is entitled to take any action because the tightness of any action is
disputed. This conclusion is itself in dispute, and therefore cannot, according
to the assumption on which it is based, be a basis for state action. There
cannot be a justified decision to take no more action. So far as this
argument goes, the fact that the principle (contingently) defeats itself cannot,
therefore, be a ground for any action. It can only be a lament for the
circumstances of societies where political principles tend to be in dispute.
These conclusions do not depend on any special features of minimal state
principles, and can be generalized. All absolute and comprehensive
disagreement-avoiding principles, political or other, are contingently self-
defeating. Absolute, comprehensive, disagreement-avoiding principles, as I
use the term here, are those which give lexical priority to the avoidance of
any decision whose justification is in contention. All such principles have
the consequence that if they are themselves in contention then they require
political decision makers to disregard them. In the circumstances of the
Western democracies, it is safe to assume that any such principle will be
disputed and therefore that it will defeat itself.
D. Neutrality
Principles which seek to avoid decisions based on disputed grounds need
be neither absolute nor comprehensive. They can seek not to avoid disputed
principles at all costs, but to avoid them so long as the cost is not too great.
They can elaborate and articulate what sacrifices would and what would not
be justified to avoid disputed principles. Whether or not they are absolute,
J. RAZ 31
they may be less than comprehensive. They may exclude disputed principles
of certain kinds, rather than all of them.
A much-favored example of noncomprehensive principles of restraint in
the face of disagreement is principles of political neutrality. Like minimal-
state principles, various neutrality principles have found support, and the
reasons for their support are not necessarily the desire to avoid decisions
whose justification is in contention. For our purposes we could take as an
example of the type a principle which calls for the constitution of the state
to be neutral between conceptions of the good, and does so on the ground

that conceptions of the good are controversial. Such a principle will require
the state to react in a way yet to be specified, i.e. neutrally, to disagreements
about conceptions of the good when constitutional decisions are at stake,
and only then.
The problem is that noncomprehensive absolute principles for avoiding
disagreement may also be self-defeating. For example, since the principle
of neutrality in my example is part of the justification of the constitution,
if it is itself controversial, as it is in every Western democracy, it rules itself
out, and can be disregarded with impunity for the simple reason that it is not
neutral between different conceptions of the good.7
Self-defeat is not easy to avoid by stipulation. One can, of course,
advocate a principle saying that the justification of the constitution, its
reliance on this principle excepted, should invoke only noncontroversial
principles. This principle is not contingently self-defeating, for it does not
apply to itself. The question is: Is its self-exclusion arbitrary? Is there a
good reason why the controversiality of other principles excludes them, but
the controversiality of this principle does not exclude it? It is not clear to me
whether such a reason can be found. Rather than pursue that question, I will
turn to another difficulty with this principle, a difficulty which applies to all
neutrality principles. In determining the political effects of disagreements,
the principle does not take account of the importance people assign to
different values and principles where the matter is controversial.
Controversy about conceptions of the good leads, according to neutral
principles, to the exclusion of any reliance on the value of this or that
conception of the good from the justification of some or all political
decisions. But the controversiality of the principle of neutrality does not lead
to its self-exclusion. Some people, myself being one, think that, given the
degree and nature of disagreement on some aspects of the good which
7. It is not clear that the conclusion is that it is right to disregard it. After all, it also
rules out every other controversial constitution. But as it defeats itself, there is no reason to
heed its implications in other regards. They only apply if there is reason to have regard to
the principle, and as it is self-defeating, there is no such reason.
actually exists in our societies, it is more important that the state should take
a decision on certain evaluative issues than that it should steer clear of
controversy in these domains. We see the good in steering clear of
controversy (or at any rate we believe that we see some reasons for doing
so). But we do not believe that that good outweighs the loss that excluding
central concerns from politics involves.
We know, of course, that if the state takes a decision on such matters,
then our view may well be defeated, and a mistaken view relied on.
Nevertheless, we believe that the chance is worth taking, or we may believe

that as a matter of principle people acting collectively have a right and a
duty to decide such matters taking notice of all relevant considerations,
however controversial. Supporters of neutrality may agree with all this, but
for them the relative importance people believe the excluded principles have
does not matter. It is not one of the considerations which determine the
scope of the exclusion.
When we reject principles because they are false the importance people
assign them does not matter. Since they are false they should not be relied
upon. But neutrality principles exclude principles not because they are false
but because not everyone agrees with them. It is less plausible to exclude
the importance people attach to those principles when this is the ground for
exclusion.
E. The Reasonable-View Condition

One approach to the question of the proper political reaction to disagree-
ment which shares this disregard, and which is often relied on by neutralists
among others, requires some consideration. According to it, there are
principled reasons to avoid relying on disputed principles only if the dispute
is among people whose views are reasonable. Unreasonable views should be
taken into account for pragmatic reasons only, such as the bad consequences
to which pursuit of the controversial measure is likely to lead. (Unreason-
able opposition to a measure may make it counterproductive even to try to
pursue it, or it may make legislation obtainable but unenforceable, or it may
lead to corruption of the law enforcement agencies, to active resentment,
and widespread law-breaking on the part of the people against whom the
measure is enforced, etc.8)
8. Making reasonable disagreement the focal point of a doctrine of the limits of

legitimate political action has become central to the contemporary Kantian contractualism of
Rawls, Nagel, Scanlon, and others. It can be incorporated in theories of legitimacy based on
consent, or hypothetical consent, into doctrines of political neutrality, and others. My remarks
attempt to capture one central variant of that approach, without considering its precise
attribution or use.
J. RAZ 33
To focus the discussion let us imagine a specific principle using the

notion of reasonable disagreement. Imagine it to state that a political
authority is legitimate only if its authority is established by a principle
which all people are committed to accepting by their current views,
whatever they are, provided only that they are reasonable.9 The question
arises, If the legitimacy of authority depends on agreement to the principles
which govern its constitution and delimit its powers, why should the
agreement only of those with reasonable views be required? It is obvious
that if the principle is not so restricted it will most likely fail to vindicate

the authority of any government at all. In any contemporary state, it is likely
that the views of some people are at odds with the authority of their existing
government. Moreover, it is likely that given any possible alternative
government some people will turn out to hold views inconsistent with it. But
the inference to draw from this fact may be that no government is
legitimate, or that legitimacy does not depend on agreement at all. It is not
evident why if legitimacy does depend on agreement it should depend only
on the agreement of those with reasonable views.
A prima facie objection to restricting the required consent in that way is
that every person counts. The life and well-being of those with unreasonable
views are just as likely to be affected by the actions of political authorities
as the life and well-being of other people. Moreover, their life and well-
being are of moral consequence. They cannot be ignored, and if the other
people's agreement is required, so should theirs be. The objection is
reinforced by the fact that having reasonable views is not a moral virtue, nor
is having unreasonable ones a moral blemish.
It is true of course that moral character defects can make it likely that one
will adopt unreasonable views.10 But there are other reasons for the
prevalence of such views, as will be seen below, and in any case not every
result of a moral defect is a moral defect. Unreasonable views may be
morally repugnant, or may lead one to adopt views which are morally
repugnant, and they may lead to morally repugnant action. If that is so it is
9. Or at any rate that their views on the legitimacy of political authorities are
reasonable. Note that the principle under discussion is not a principle of consent, which
makes authority a matter of binding precommitment, but of agreement, which has to be
continuous. The legitimacy of a political authority comes to an end when the principles on
which it is based no longer enjoy the agreement of all the reasonable. This makes the
principle very demanding. On the other hand, while the agreement required is actual and not
hypothetical, it is agreement only to principles from which the justification of the authority
can be derived, whether or not the people involved actually realize that.
10. The consistent holding, and acting on, some unreasonable moral views may constitute
a moral defect. For example, those who believe that no one should be treated with mercy are
morally defective people. But these are special cases.
because, whether reasonable or not, they are erroneous and may lead one
into further errors, and sometimes the errors may lead to immorality. None
of that makes the holding of unreasonable views a moral defect.
Defenders of the view that one need not secure the agreement of those
with unreasonable views could reply to the objection, saying that it misses
the point of the exclusion of the unreasonable. It is usually supported by
people who favor contractualist approaches. These are based, in the opinion
of many," on the fact that people are rational self-directing agents who
ought to be treated as such. One consequence is supposed to be that people

cannot be ordered about. As a result, the legitimacy of authority is
conditional on the fact that its subjects have reason to accept it as binding.
But the unreasonableness of unreasonable people means that they do not
respond to reason as they should and therefore the reason for treating people
as rational self-directing agents does not apply to them. This does not mean
that just principles of politics do not take the interests of the unreasonable
into account. They do. But it does mean that the legitimacy of the
government is not conditional on their consent, nor on their having views
which commit them to consent.
One may take exception to every step of the outlined argument. However,
for present purposes, suffice it to say that it is invalid simply because the
notion of reasonableness employed by contractualists when they define the
range of the exclusion is not the one needed to make the argument valid.
The argument I outlined here uses our ordinary notion of unreasonableness,
as applied primarily to people. It marks an intellectual defect. Unreasonable
people are unyielding people. They persist with their views or intentions in
the face of the evidence. They stubbornly fail to see what the reasonable
see. The argument of the previous paragraph says that the moral imperative
of treating people as agents who should be reasoned with rather than
manipulated, ordered about, or coerced does not apply to those who are, in
this sense, unreasonable.
The unreasonableness the contractualists fix on, on the other hand, is not
of people but of beliefs. Beliefs are usually judged unreasonable if they are,
in relation to the evidence available to the experts, patently false. That
means that experts do not usually disagree about these beliefs. For the
experts, the holding of such beliefs falls beyond the margin of common
error. At times not the experts but some other group will be the implied
11. But not all. Contractualists may base their approach on considerations of reciprocity,
or the moral merit of the views ("reasonable" views may be views consistent with being a
good neighbor, etc.), or regard it as constitutive of practical rationality (communication
theorists, for example, may take such a line). As with minimal-state and neutralist
approaches, my criticism here is not of the solutions, but of certain arguments for them.
J. RAZ 35
reference group for such judgements. But all judgements of the reasonable-
ness of a certain view abstract it from any particular believer.12 It follows
that perfectly reasonable people can have unreasonable beliefs. Creationist
science is committed to many unreasonable beliefs, but if you were taught
creationist science at school and at home and came across no other science,
then your reasonableness may be more in doubt if you reject creationist
science than if you believe it. So the reasonableness of views measured by
reference to the common view of the experts is irrelevant to the argument
I sketched above.

There is a second, derivative, sense of unreasonable belief, or rather of
unreasonable believing, which is obviously more suitable for use in the
context of this argument. If the holding of a certain view by a person
manifests an unreasonable attitude of that person (i.e. because it is held in
the face of readily available evidence, etc.), then the holding of that view
is itself unreasonable. Naturally, this does not show that the view is
unreasonable, only that it is unreasonable for this person to hold it.
Unfortunately, the unreasonableness which figures in the writings of
leading contractualists is not of this kind. Rather, in their hands, reasonable-
ness or its absence is measured by the content of the views held, not the
rationality of holding them. For Rawls, the reasonable accept that securing
willing cooperation of those who share this view is a social goal of
paramount moral importance, for the sake of which the pursuit of any value
may have to be sacrificed. The unreasonable think otherwise. They do not
accept the unconditional and absolute value of securing willing cooperation.
He explains:
Persons are reasonable in one basic aspect when, among equals say, they
are ready to propose principles and standards as fair terms of co-operation
and to abide by them willingly, given the assurance that others will
likewise do so. . . .
Reasonable persons, we say, are not moved by the general good as such
but desire for its own sake a social world in which they, as free and
equal, can cooperate with others on terms all can accept.. . . 13
12. Though it is possible to restrict the scope of such judgements to individuals: "It is
unreasonable for Jones to believe that so and so" means roughly the same as "Given his other
beliefs, Jones's belief that so and so is unreasonable." It is his believing which is unreason-
able, not necessarily the belief itself.
13. John Rawls, Political Liberalism (New York: Columbia University Press, 1993),
49-50. In describing the unreasonable, Rawls confines himself to people who refuse to abide
by principles of cooperation and are willing to violate them when this would suit their
interests. This leaves out all those who would violate these principles for the sake of others
when they conflict. But the inevitable implication of Rawls's theory is that they, too, are
unreasonable. I will not explore in this article Rawls's own reasons for the exclusion of what
In this passage he presents his explanation as an explanation of a

character trait or a disposition. But it is not one which has any bearing on
the intellectual virtues. It is simply a disposition to act as one would from
a belief that securing willing cooperation of those sharing this view is
lexically superior to any other goal. Even if this view were true, and even
if it were the case that its negation is unreasonable, it may turn out that a
person who accepts it is unreasonable and one who rejects it is reasonable.
That will be the case if, given their personal histories, their education,
experience of social and political life, and so on, the one accepts the

supreme importance of willing cooperation even though everything in his
life points the other way, and the other rejects it because everything in his
life points the other way.
Nor is Rawls alone in introducing one notion of reasonableness when
another is needed. For Nagel, the reasonableness is in the reasonableness
of accommodation between impartial and partial concerns.14 He aims to
rely only on conclusions that no one can reasonably reject.15 But if this aim
is credible, it must be interpreted conditionally, i.e. no one who followed
Nagel that far could reasonably reject the conclusions he points to. I think
that Nagel would agree that those who have not accepted his argument to
that point may not have been unreasonable people and their rejection of his
conclusions does not show that they are. They are, in the sense which
matters, i.e. the sense that would show people to be rational agents
responsive to reason, reasonable.
Formally, these remarks only show that Rawls and Nagel have misapplied
their own test (assuming that they were motivated to accept it for the
reasons I stated, and that is open to dispute). They do not show that the test
is misconceived, nor that others have not applied it correctly. The remarks
have, however, more far-reaching implications. They show that one cannot
take the rejection of any proposition as in itself strong evidence that the
agent is unreasonable (in the sense in which this is a cognitive vice). The
implication of this is that the exclusion has no political teeth. Let me
explain.
Rejecting a belief would testify to the unreasonableness of the rejecters
only if it is one which no rational agents, whatever their beliefs, could
rationally reject, or if given the rejecters' other beliefs and their situation,
they could not rationally reject. I will disregard the first half of the test, on
the ground (or in the hope) that there are no propositions with direct
political implications which are current in contemporary societies and which
he regards as "unreasonable" views. Note that Rawls has a separate discussion of "reasonable
comprehensive doctrines" (Political Liberalism, 59), which does not fit well with the passage
cited above.
14. Thomas Nagel, Equality and Partiality (New York: Oxford University Press, 1991),
39-40.
15. Cf. ibid., 50-52.
J. RAZ 37
meet this test. It therefore does not seem to have political use in contempo-
rary societies. What of the second part of the test? In homogeneous societies
it may have political implications. In such societies it may turn out that only
the unreasonable could reject the authority of their government. That
authority may be illegitimate, it may perpetrate moral atrocities, but they
know no better, and those who reject it do so irrationally. But the pluralism
and diversity of contemporary Western societies disable the test. There is no
proposition which has currency in Western societies and which some people
in them could not reasonably accept.

F. Agreement and Legitimacy
So far I have argued against the unreasonableness-exclusion in tests for
the legitimacy of political authority. I argued, in effect, that if legitimacy
depends on agreement and if that means that the principles of the constitu-
tion should be such that all reasonable people are committed, by views that
they already hold, to accept them, then any of the views, ideologies,
philosophies, religions, or what not which have currency in that society
should vindicate the principles of the constitution, or else they lack
legitimacy.
The hard question is whether the requirement that agreement to the
principles of the constitution be a condition of its legitimacy is a reasonable
response to disagreement over principles in a country.16 Most writers do
not claim that actual agreement in judgement is necessary. That would show
that no government is legitimate, nor any alternative government likely to
be legitimate in any of the Western countries—not a conclusion those
writers are willing to accept. The claim is that a condition of legitimacy is
that the principles on which the constitution is founded can be justified to
the people who are supposed to be subject to them. Does that mean more
than that the principles of the constitution are justified? Yes, you may say,
it means that those subject to the constitution can in principle come to
realize that the principles of the constitution are justified.
That does not seem to me to add anything to the thought that the
principles are justified. Justifications are in principle publicly available.
There can be contingent reasons why this person or that will find it very
difficult to come to realize that the principles are justified. But there is
nothing inherently private about justification. Some people are not in a
position fully to understand some justification or another, their senses may
be impaired, or their powers of understanding limited. For the most part,
even they are capable, however, of realizing that the principles or beliefs in
16. Notice that I am considering the case for some agreement as a condition of
legitimacy made necessary by the fact of disagreement, i.e. the fact that some people are in
error. Some writers think that agreement, or some forms of it, is constitutive of evaluative
truth or validity. I have nothing to say about such views here.
question are justified. They would be relying in part on testimony, as we all

do in justifying most of our beliefs for which we have any justification.
Some people's mental capacities are so limited that it is in principle
impossible for them to comprehend the (correct) justification of the
principles in question. If that is a problem for the legitimacy of government,
it is not solved by taking account of any mistaken ideologies they have, or
of their misguided religious beliefs. If their mental capacities are too limited
to understand correct justifications, then they are also too limited to
understand any plausible incorrect one. The problem they raise, if any, is
not one of accommodating disagreement over principles.

Regarding any justified principle, people of normal capacities are in
principle able to understand that it is justified. The requirement that the
principles on which the constitution is based be justified already includes the
requirement that every potential subject of the constitution be in principle
capable of understanding that they are justified.
Some people would say that adding "being capable of being justified to
its subjects" to "being justified" makes a difference, namely, it states that
only a justification which can be publicly and comprehensively and
explicitly articulated and communicated can lend legitimacy to a regime.17
But this is a mistake. Justification does not have to be articulated to be
known and understood. To know what is in the kitchen I do not need to be
told what is there. I can go and see for myself. When I see it for myself I
know what is there without telling myself what is there, i.e. without listing
the items for my information. I just see them. Nothing need be articulated.
Could it be that such examples are irrelevant for no justifications of political
principles can be available to people except through being fully stated to
them? That seems highly unlikely. Our understanding of politics derives in
large part from reflection on historical events we lived through or read
about. We (often reasonably) derive conclusions from reflection on concrete
events, without being able to state fully all the features of the events which
influenced us, nor all the ways they affected us, nor the precise reasons why
they led us to the conclusions we reached. Hence, the requirement that the
principles of the constitution be capable of being justified to their subjects
does not amount to a requirement that there be a complete articulation of the
justification of the principles which can be communicated to all potential
subjects.
17. It is not easy to give precise meaning to this condition. In one sense every
justification can be articulated, even though its articulation may include propositions like the
following: "the truths revealed to Moses on Mount Sinai show that..." where the content
of those alleged truths is not otherwise specified. The general idea behind the requirement
of complete and explicit articulation is that the propositions figuring in the justification are
not only identified, but their content is stated, and in ways which enable an average or
normal member of that society to understand them—i.e., it is a standard which is relative to
some societal norm.
J. RAZ 39
Let it be agreed, then, that an authority can be justified, and its justifica-
tion known even though it has not been articulated, and therefore not been
publicly communicated. Nevertheless, it may be claimed that at least in this
day and age, i.e. in an age of bureaucratic states governing very large
numbers of people, political authority is legitimate only if it can be justified
in fully articulated principles and arguments. This claim should be rejected
since it bars us from important information which is vital to political
reflection. I am not saying that the justification of authority may rest on
knowledge that in principle cannot be articulated. My point is that while
each item of knowledge can be articulated it is impossible to articulate all

that we know. In any case, my claim is that full articulation of the
justification of authority is impossible in practice, in the conditions of our
life.
What, then, does the condition that authority is legitimate only if it can
be justified to its subjects add to the obvious condition that it is legitimate
only if justified? One line of thought goes like this: The justification of a
legitimate authority is highly likely to be in principle within the reach of its
subjects. But this availability in principle is compatible with it being highly
unlikely that many of the people whose current views imply a rejection of
the justification will ever see the error of their views. Respecting the fact
that people are self-directing agents requires bridging the gap between what
people can in principle come to know, and what they are likely to come to
know or to believe. This bridge is built by making it a condition of
legitimacy that at least some justification of authority is available to people
as they are, i.e. given the beliefs they actually have, even if they are false
beliefs.
This thought, plausible as it may sound, is deeply puzzling. Does it mean
that actual agreement about the justification of the principles of the
constitution is required? This will mean that no authority is legitimate. What
is normally claimed to be required is that the justification follows from, or
is at least consistent with, views that people already have. If this means that
people must be capable of seeing the cogency of the justification without
abandoning any of their current views, then this is still too strong18 and
will most likely lead to the conclusion that no Western government is
legitimate. It is likely that in every country there are some who hold beliefs
which are incompatible with any plausible justification of the authorities
which these countries have, or any legitimate alternative to them.
Suppose that we moderate the condition to exclude only changes in
people's fundamental beliefs. For the purpose of my argument it does not
18. In calling this and other suggested conditions top strong I do not mean to beg the
question against anarchism. The argument is ad hominem. It is part of the contractualist
ambition to provide an account of legitimacy by which there can be legitimate governments
in pluralistic societies. That is why the condition is too strong. It defeats the purpose of the
account.
matter how "fundamental beliefs" is understood. It can be taken to refer to

any subset of a person's beliefs which, it is suggested, should be privileged
in the principle of legitimacy." In this form it says that a government is
legitimate only if its legitimacy can be justified by considerations and
arguments that all those living under it can come to accept without
abandoning any of their fundamental beliefs. It still follows that no Western
government is legitimate since it is highly likely that in all Western
countries there is at least one convinced anarchist. Whether or not this is so,
the condition gives any convinced anarchist, or indeed anyone else who has
a principled objection to the current government, a veto on its legitimacy.

Even if the conclusion establishes only a personal exclusion, so that those
who have principled objections are exempt from the authority of the
government whereas everyone else is subject to it, this is a troubling
conclusion. People with principled objection to the government of the day,
or some of them, such as Nazis and religious extremists, are an important
part of the reason why the rest of us think that governments have authority.
They are the people we need a government to control.
This problem is far-reaching. Let it be admitted that it is legitimate to
stop would-be murderers from perpetrating murders. It would therefore be
legitimate to set up authorities which would organize our defense against
murderers—appoint police officers, grant and define the limits of powers of
self-defense, of citizen's arrest, of the doctrine of necessity, and so on. At
least it would be legitimate to do so unless having such authorities is bad in
other ways which undermine their legitimacy. Principles of the rule of law
dictate that to be justified such laws presuppose another. They presuppose
a legal prohibition on murder. But by the principle of legitimacy we are
examining, that prohibition may not be addressed to a fanatic who rejects
the authority of the state. Therefore, if the principles of the rule of law are
valid, then it is not merely that the state has no authority over the fanatical
murderer. It does not even have authority to defend people against him.
G. Respecting People through the Propagation of False Beliefs

There is another problem with the agreement condition. It says that an
authority is legitimate only if the principles of its constitution can be
justified in a way which is compatible with the fundamental beliefs of any
of the subjects of the authority. This can be achieved in one of two ways.
First, one may be able to produce a sound justification of the authority,
which as it turns out is compatible with the fundamental beliefs of everyone
in that country. If that is so, the problem of justifying authority to those who
19. Alternatively, the condition can be moderated by including a reference not to a subset
of the beliefs of each person, but to socially familiar beliefs. For example, it may be claimed
that a political authority is legitimate only if it can be justified in a way which is consistent
with the moral or religious outlooks with significant following in that society. The comments
below apply equally to socially relativized conditions of this sort.
J. RAZ 41
disagree with us does not arise. Justifying the authority to them turns out to
be the same as justifying the authority. Where the two differ we have to
resort to the second method. It is invoked when it turns out that no sound
justification of the authority is compatible with the fundamental beliefs of
some. In that case some alternative justification is needed, a justification
which relies on some of the false beliefs those people already have, or
which involves getting them to accept some new false beliefs20 which are
compatible with their existing fundamental beliefs, and from which the
validity of the principles of the constitution can be inferred.
Where the first route is available the agreement condition does not add

anything to the principle that only justified authorities are legitimate
authorities. The point of the agreement condition is to make legitimacy in
effect depend on use of the second route. But does the second route yield
a justification at all? It relies on the fact that it is possible to derive true
conclusions (the principles of the legitimate constitution) from false
premises. But no such derivation can count as a justification of anything
since it relies on falsehood.
Supporters of the agreement condition may say that this argument merely
catches them misstating what they really mean. They did not mean that it
is possible to justify the constitution without contradicting people's
fundamental beliefs. They had in mind a dual condition: (a) the principles
of the constitution are justified and (b) they are derivable from various sets
of premises such that everyone in the country can accept at least one of
these sets consistently with their fundamental beliefs, whatever they may be.
If this is the proposed condition of legitimacy, why should one accept it
as such? Presumably, because while such derivations (as mentioned in (b))
do not necessarily justify the principles of the constitution, they make people
who (mistakenly) believe in their soundness believe that the principles of the
constitution are justified. But it is a strange doctrine which regards the
inculcation of false beliefs as a condition of respecting the rationality of
people as self-directing agents.
I may be moving too fast. Condition (b) does not require that anyone has
false beliefs. It merely says that if some people already have false beliefs
which are inconsistent with all sound justifications of die principles of the
constitution, then the legitimacy of the authority of the state depends on the
fact that the principles of the constitution can be derived from those false
beliefs, or from some other false beliefs which are consistent with them. The
legitimacy of the government does not depend on anyone actually believing
in these derivations, or in their (false) premises.
20. It is possible to have a more restrictive condition, requiring derivation of the

legitimacy of the government from premises including no new false beliefs. This will make
the condition very hard to satisfy, especially if it is taken to exclude even new false beliefs
which are derivable from those people's current false beliefs (in combination with other of
their new or old beliefs).
Formally, the point is valid. Yet its relevance is moot. For if the
underlying idea is that a government is legitimate only if all the citizens can
believe that it is legitimate, then the underlying thought is not that the
satisfaction of (b) is a condition of the legitimacy of government, but that
the promulgation of (b) is politically desirable. For surely the reason why
the possibility of people believing in the legitimacy of government is a
condition of its legitimacy is that it is desirable that they will actually
believe in its legitimacy. That much is compatible with the weaker principle
which says simply that to be legitimate the constitution must be justified.
The agreement condition goes beyond that. It is committed to the view that

it is desirable that people without abandoning any of their fundamental
beliefs should believe in the legitimacy of the government. It follows that
it is committed to the view that it is desirable to propagate false beliefs or
unsound inferences.21
Perhaps it is true that it is desirable to cultivate such false beliefs in
arguments for the legitimacy of the government. Doing so may secure
voluntary compliance with the law, and in general, it will encourage people
to behave as good citizens should. But it is strange to uphold the cultivation
of false beliefs in people as a requirement of respect toward those very
people.
To a certain extent the objection can be avoided by modifying the
condition of legitimacy to make it more restrictive. Let condition (b) above
be modified to read: "The principles of the constitution are derivable from
various sets of premises such that everyone in the country is committed to
accepting at least one of these sets by their fundamental beliefs whatever
they may be." This modified condition basically states that there is an ad
hominem argument against every person in the country to the effect that his
current fundamental beliefs, be they true or false, commit him to the
legitimacy of the constitution. It can be further adjusted in various ways to
meet further difficulties in bringing out this point. Needless to say, this
modified condition (b) is more restrictive than the original (b), thus making
it less likely that the legitimacy of the government can be established. Does
it, however, reestablish a contractualist reading of the agreement condition
which is free of the charge of inculcating falsehoods? In part it does, but not
entirely. There is no warranty that in reaching the true conclusion (that the
constitution is legitimate) from the false premises people would not be
21. In passing, it is worth noting that advocates of the principle treat the validity of
inferences and the truth of their premises differently. They insist that legitimacy depends on
valid derivations (even if people do not believe in their validity), even if from false premises.
Why not treat both components in the same way and make legitimacy depend (among other
things) on the availability of some derivations (valid or invalid) from some premises (true
or false), so long as the validity of the derivations and the truth of the premises are consistent
with the fundamental views of people regarding validity of derivations and truth of
propositions?
J. RAZ 43
required to accept further falsehoods which, while justified by their existing

beliefs, were not in fact accepted by them until they tried to justify the
constitution without reviewing and revising their fundamental beliefs.
Can it be argued that we must respect people's false beliefs, not of course
because they are false, but because people have come to have them by the
exercise of their rational powers? This will raise questions about when do
people's beliefs result from the exercise of their rational powers. Some will
say that the very having of beliefs is a manifestation of rational powers.
Others will exclude beliefs induced by hypnosis or subliminal manipulation.
Yet others will go much further and will rule out beliefs induced by

demagoguery. Many difficult questions arise, e.g., how to judge beliefs
which are received because they are the common view in one's family or
society, and are never examined?
Let me put all these questions on one side, for I wish to deny the basic
claim which gives rise to them. Respecting people as rational self-directing
agents does not require desisting from following true beliefs which those
people dispute. The suggestion that it does have this implication confuses
respect for people, because they have rational powers, with respecting their
currently held views. That people have rational powers means that they are
not stuck with the views they have at any given time, that they can examine
and revise them. We are considering the response to the fact that they have
false beliefs. Given that they are rational we expect them to examine and
revise such beliefs, and if we have any duties in this matter it is to
encourage such reexamination. In politics these duties include respecting
freedom of expression, explaining policies to the population, and being open
to arguments and to contrary suggestions. It does not include a duty to desist
from following the truth, when it is disputed.
Remember that people have a duty to respect their own rationality. That
includes the duty not to take the fact that they hold a belief as a reason for
anything.22 There is no reason to think that others must take such beliefs
as reasons out of respect for the people who hold them. To be sure,
respecting people's rationality is inconsistent with systematically preventing
them from relying on their own beliefs. But it is a far cry from letting
people rely on their own views, to denying the legitimacy of the constitu-
tion, even though it is based on sound principles, because it is mistakenly
disputed by some of its subjects.23
22. The exceptions are cases where they seek therapy because they have certain beliefs,
etc.
23. It is important not to confuse the view I have criticized above with other,
superficially similar ones. Some, like Scanlon, hold moral justification in general to be
relational. According to them, to be morally justified is to be justified to people who share
certain motivation. Others may have a similar view about the justification of constitutions,
and their underlying principles. I find such views implausible, but I do not consider them
here. As they do not endorse any distinction between being justified, and being justified to
H. Waldron on the Right of Rights

Recently Jeremy Waldron has revived the case for participatory
democracy and against bills of rights backed by American-style judicial
review with an argument resting on the fact of disagreement over principles.
Entrenched bills of rights backed by judicial review are often seen as the
mechanism through which restraints due to political disagreements should
be implemented. Protected constitutional rights, we are told, are first and
foremost there to protect minorities. Waldron rejects this view for two
reasons. It relies on a form of right-instrumentalism, which he finds uncon-

vincing, and it ignores the right of rights, the right to participation.24 Both
reasons are suspect.
Waldron's main claim is that individuals are entitled to participate in
decisions affecting their lives, and that includes decisions about the prin-
ciples which should govern the life of the country in which they live.
[T]he idea of rights is based on a view of the human individual as
essentially a thinking agent, endowed with an ability to deliberate
morally, to see things from others' points of view, and to transcend a
preoccupation with his own particular or sectional interests.... [Rjights
typically provide an individual with a protected choice on an issue which
remains morally significant. . . . The faith in the right-bearer's choice
evinced by the attribution of the right is certainly not confidence that he
will unerringly make the right choice; nevertheless it is borne of a
conviction that he has the wherewithal to ponder responsibly whatever
moral issues the choice involves.25
Relying primarily on Locke, he observes:
It is impossible, on this account, to think of a person as a right-bearer and
not think of him as someone who has the sort of capacity that is required
to figure out what rights he has. This means that arguing about a person's
rights is not like arguing about the rights of animals or about the
preservation of a building. When we argue about someone's rights, the
subject of the conversation is likely to have a considered view on the
matter. And since the point of any argument about rights has to do with
the respect that is owed to this person as an active, thinking being, we are
hardly in a position to say that our conversation takes his rights seriously
if at the same time we ignore or slight anything he has to say about the
matter. Yet again, I emphasise that this does not mean a person has
whatever rights he thinks he has . . . But it does show that there is
something appropriate about the position we are considering—that the
. . . they are not open to the criticism above.

24. Jeremy Waldron, Law and Disagreement(OxfoT&. Oxford University Press, 1999),
chapter eleven.
25. Ibid., 250.
J. RAZ 45
right-bearers should be the ones to decide what rights they have, if there
is disagreement about that issue—and something unpleasantly inappropri-
ate and disrespectful about the view that questions about rights are too
hard or too important to be left to the right-bearers themselves to
determine, on a basis of equality.26
The argument is meant to establish a procedural response to disagreements
over principles, namely a participatory majoritarian resolution of such
disputes. But the argument is too weak to be given much credence. Since
people have the capacity to figure out what rights they have, including the

capacity to see things from others' point of view, it is appropriate for each
person to be able to decide what his or her rights are—this seems to be the
drift of Waldron's thoughts. But why not reverse the argument: Since people
have the capacity to figure out what rights other people have, including the
capacity to see things from others' points of view, it is appropriate for each
person to be able to decide what rights other people should have. I will
decide the scope of your rights and you will decide mine.
Both arguments are equally invalid, and which one, if any, one finds
more unconvincing seems to me to be a matter of personal taste. A better
argument, improved by taking motivations seriously into account, would
suggest that it is best if every person has the power to determine the scope
of rights belonging to others who have the power to determine his or her
rights. But I am not suggesting that this is anything other than a very flawed
argument as well.
Perhaps the fate of Waldron's argument does not matter, for he does not
rely on it himself. He acknowledges the undesirability of each person
deciding his or her own rights, and retreats to an equal participatory right.
The least persuasive argument for participatory democracy is the one which
regards it as the nearest we can get to self-determination, at least if one
follows Waldron in thinking of determining rights as a process of applying
moral truths. Arguing one's case before a learned and impartial tribunal
gives one more power over the determination of one's rights than being one
among the multitude in a participatory democracy, with equality of political
power to all.
A natural way to proceed is to assume that the enforcement of fundamen-
tal rights should be entrusted to whichever political decision-procedure is,
in the circumstances of the time and place, most likely to enforce them well,
with the fewest adverse side effects. So to argue is, in Waldron's eyes, to
succumb to the mistake of right-instrumentalism.
He claims27 that if the justification of a constitutionally entrenched right
is that its enforcement is made more secure by following the constitutionally
provided procedure than it would be if made subject to a democratic
26. Ibid., 251-52.

27. Ibid., 253.
procedure, then the argument assumes knowledge of the content of the

right—knowledge which is controversial.28 How could one justify the claim
about the likelihood of it being respected if one did not know what it is?
The answer is that there may be a whole variety of reasons which could
justify such a conclusion even absent knowledge of the content of the right.
It all depends on the circumstances. For example, we may have sufficient
reason to believe that the legislature will not even try to establish what
rights people have, or what restraint it should exercise, given the fact of
disagreement over principles, whereas the courts, who are in charge of
enforcing constitutional rights, will try honestly to do so. Absent an

argument that the correct content of rights is more likely to be revealed if
one does not try to discover it, this would be a good instrumentalist
argument in favor of constitutionally entrenched rights.
This example is not as far-fetched as it sounds. In many countries there
are ample reasons to suspect that members of the legislature are moved by
sectarian interests to such a degree that they are not likely even to attempt
to establish what rights (some) people have. The same considerations may
apply to courts as well. But it is possible for them not to apply to courts,
even while they apply to the legislature. This argument is a close cousin of
another. We may know that certain factors are likely to cloud people's
judgements. They may be, for example, liable to be biased in their own
interest. We may therefore prefer a procedure in which those charged with
a decision are not affected, or not directly affected, by their own decision.
There are other factors known to bias judgement, and their nature and
presence can be established even without knowledge of the content of the
rights concerned.
These points require careful handling. Sometimes, for example, there are
reasons for thinking that those whose interests are not going to be affected
by a decision are unlikely to try honestly to find out what is just in the
circumstances. Sometimes one may be unable to appreciate the plight of
classes of people unless one belongs to the same class oneself, and therefore
rather than entrusting the decision to those not affected by it, it should be
given to those who are so affected. Moreover, it is fair to say that knowl-
edge of factors of the kinds I mentioned does require if not knowledge of
the right concerned then some knowledge of the kind of right it is: that it
concerns the effects of disagreement, that it relates to social or religious
matters, etc. But then such general knowledge is not hard to come by. It
may be controversial, but that does not undermine the case for an instrumen-
tal approach to the choice of mechanisms for enforcing whatever measures
of restraint or other measures disagreements over principles necessitate.
28. Assuming that we are concerned with rights which constitute the right response to
political disagreement, this is in a way a second-order controversy—controversy about what
is the right response.
J. RAZ 47
Waldron is aware of the considerations advanced above and finds no

solace in them. Given that there may be reasons both for and against any
decision mechanism he concludes that it is almost impossible "to defend an
impartial account" of the best means available in the circumstances for
defending rights, and that given existing disagreements "we are not in
possession of any uncontroversial moral epistemology."29 The second claim
is true, but irrelevant. That sound moral epistemology is controversial does
not mean that we cannot know what it requires. From the fact, if it is one,
that its central precepts are controversial, it only follows that avoiding
controversy is not a goal to be pursued. At least that follows unless there is

an alternative and at least sufficiently attractive goal which is not subject to
the same problem. Waldron's right of participation is not such an alterna-
tive. It is equally controversial. Having declared that right instrumentalism
does not take seriously the problem of disagreement, Waldron seems
oblivious to the fact that if valid the same charge can be levelled against his
solution, as (quite likely) against all others.
/. The Moral Standing of Historical Settlements

My remarks, being critical of various proposed responses to the fact of
disagreement over principles, may create the impression that I do not believe
that such disagreements call for any restraint in political action. That is far
from the truth. There are many possible responses to disagreement. Only a
few were criticized. The objections related to the rationale of the different
principles rather than their scope. It is worth noting, however, that the
rejection of the unreasonableness restriction opens the way to taking full
account of the existence of "unreasonable" positions. Their "unreasonable-
ness" should not, in my view, limit the way public action should be
modified in light of their currency in the country. At least to this extent my
response to political disagreement is more extensive than that of advocates
of some of the views here criticized.
Yet it differs from many responses in abandoning the attempt to occupy
a noncontroversial high ground. Sound restraints are part of a sound
normative theory of politics. Such a theory allows for policies to be
responsive to circumstances, economic, social, cultural, and others. One such
circumstance is the existence of disagreement over principles. The result is
that however extensive the accommodation with others may be, its rationale
is firmly based in a sound theory of politics, and may well not appeal to
those who disagree with it.
Some find this paradoxical: the party of government modifies the policies
it would otherwise pursue because some in the country disagree with them,
but it does so in the name of principles which they, the objectors, also
disagree with. There is, however, nothing paradoxical here. It is natural that
29. Ibid., 254.

if I accede to your view in spite of my disagreement with you I will not

accede to it for your reason. You are moved by what you perceive to be its
merit. I am moved by reasons for respecting your view. I started with this
asymmetry of reasons, and it is definitive of principles of restraint in the
face of disagreement.
This does not mean that the response to disagreement over principles is
bound to be a matter of controversy. As mentioned above it is likely to be.
But disagreement about the response to disagreement over principles is not
an inevitable consequence of that disagreement itself. For the most part the
rival moral, religious, and political outlooks in any country disagree only in

part. There is a good measure of agreement among them.30 With some
outlooks the agreement may include the right way to respond to disagree-
ment. When this is the case it makes political life much easier. When it is
not the case it is still possible for a large measure of convergence on agreed
compromise solutions and on agreed policies of mutual restraint to obtain,
even though adherents of different groups have different reasons for
accepting these measures.
History plays a big part in establishing convergence around measures of
self-restraint in the face of disagreement over principles. I mean more than
to point to the obvious fact that the boundaries of toleration in even the
most tolerant countries were determined in the course of their history, often
as a result of bitter struggles, sometimes ending with mutual exhaustion. I
mean that contrary to much philosophical opinion that fact is not morally
neutral. Other things being equal, we have reason to respect historically set
boundaries. From a normative standpoint, historical settlements are often
acknowledged as mere modi vivendi, belittling them both for being unstable
and liable to be easily unsettled, and because they are unprincipled, based
on might not on right. Therefore, while they may represent the best
compromise available for a time, they are not true "solutions" to the
problem of political disagreement.
The stability of historical settlements is contingent and varies a good deal
over time, but the claim that they lack stability because they are unprinci-
pled and not morally based is misguided. It arises out of the same exagger-
ated belief in the power of theory I referred to at the beginning. It expresses
the same exaggerated faith in the power of reason and morality I adverted
to before. This is of course a controversial issue, and I cannot do more than
be controversial about it. Any reasons for this position will have to await a
different occasion.31 Speaking very roughly my suggestion is that at the
most abstract level morality provides roughly defined parameters such that
30. A fact which became the foundation stone of Rawls's "political liberalism."
31. Needless to add, these remarks on the moral standing of the historical settlement in
any given country apply not only to principles of response to disagreements over principles.
They apply to other political principles as well. Equally obviously, they apply only to the
extent that there is a historical settlement generally accepted in a country.
J. RAZ 49
what falls outside them is unjust or immoral, and what falls within them is
just or moral. Many different arrangements fall within them, and morality
provides that if the institutions and procedures in place in our society
conform with the abstract principles then they are binding and flouting them
is unjust, even though it may involve conduct that in another time and place,
where other institutions and procedures prevail, may have been acceptable,
or even obligatory.32
Up to a point the practical implications of this view are not far removed
from common views. For the reasonableness of different institutional
arrangements, and the prevalence of different applied (i.e. less abstract)

principles in different countries is commonly acknowledged. Most often it
is thought to be a result of differences in the application of a normative
(perhaps a moral) theory to nonnormative social, economic, technological
or other conditions between different countries. In a way that is my view
too, except that in my view we cannot regard the concrete principles which
should prevail in any country as principles whose content derives from
applying abstract principles to nonnormative circumstances. Rather, they are
contingent products of history, valid because they fall within what is
permissible by abstract principles, and are the principles prevailing in fact
in that country.33
The difference between these positions may appear too rarefied to matter.
In fact it has far-reaching practical consequences, for the hope to derive
applied concrete principles from abstract ones coupled with nonnormative
facts encourages a false belief in the power of reason or morality and a
tendency for impatience with tradition. It encourages the wrong attitudes to
politics, the wrong assumptions in political debates, and a misguided belief
in the skills required of politicians. It tends to be disdainful of constraints
which are institutional, or political in nature, and which do not appear to be
derivable from first principles. It encourages dogmatism and lack of
tolerance in the practice of politics.
32. The constraining view of most political principles is equivalent to belief that they
judge many different political arrangements as incommensurate in value (while ruling out
others as unacceptable, or as less than ideal), provided that it is coupled with a principle
justifying following common principles in any given country. This additional principle means
that one may not deviate from the historical settlement simply on the ground that an
alternative political arrangement would have also been acceptable.
33. Questions rush in at this point: Does it mean that there is something wrong with
trying to change the way a country runs its affairs even though morally there are other
acceptable ways of doing so? No is my answer, provided the alternative one favors is itself
morally acceptable, and one has good, even though not necessarily moral, reason for pursuing
change. What of divided countries where no agreed way of doing things is accepted as
legitimate by all the population? It may be that in such countries there is no binding concrete
principle, and one's duty is not to breach the abstract principles and to help establish by
legitimate means common ways of organizing collective life.
J. Some Additional Pointers

The discussion of the appropriate response to disagreement over
principles in recent times bristles with new ideas, new principles unknown
to political thought or practice of the past. It is none the worse for this.
However, in my concluding remarks, I wish to mention some familiar
traditional factors which indicate how to think of disagreements over
principles and other matters. First come two pragmatic considerations.
Disagreement over principles, if deeply felt, and involving many people,
indicates that attempts to impose the disputed principles will meet serious

opposition, thus leading to a whole array of undesirable side effects. These
in themselves often justify restraint and compromise.
A second pragmatic consideration weighs with many people more than
any of the considerations of justice and legitimacy we examined earlier.
These people worry that disagreement over principles tends to undermine the
stability of the constitution and the state. They seek compromise with
dissenting voices for fear that otherwise those in power may find the
principles of the constitution overthrown when others come to power in
favor of unjust and abhorrent principles.
It is tempting to say that there is nothing we can do against such danger.
After all, that you or I say to each other that governments should exercise
restraint when faced with disagreement will have no effect on what a wicked
government will do. But this answer misconceives the worry. What we say
here may not have much influence on how things are in the world. But we
are trying to find out what principles would be sound principles for a
government to have. The government we envisage is of course a just and
wise government, not a wicked one. Therefore, we tend to feel that it should
simply follow just and wise principles. The worry about the principles which
a wicked government will follow is misplaced, we are inclined to say. By
definition its principles will be wicked, and the fact that the just and wise
government will follow moderation in all things will have no effect on the
wicked government. But this is not quite right. The principles of a just state
should take account of the possibility of a wicked government gaining
power. A just government will be based on principles which will make it
less likely that such a change will occur. In other words, part of what we
expect of the principles governing a just regime is that it should make its
own perpetuation more likely. This includes its ability to inculcate respect
for the principles on which it is based, and this may require finding
accommodation with dissenting groups to avoid alienating them altogether.
The advisability of compromise and accommodation does not mean that they
are always possible, but they are worth working for.
A second group of principles are epistemic in nature. The simplest and
most elementary takes the fact of disagreement as a reason to double check
one's own views. Disagreement is proof that at least one of the parties is
wrong, and one must wonder whether it is not oneself who is in the wrong.
Here compromise may be simply a way of hedging one's bets, of taking out
J. RAZ 51
protective measures in case one is wrong. Naturally, sometimes reexamina-

tion will explain why the other side is wrong and reaffirm one's principles
beyond doubt or question. Such epistemic doubts will have a limited effect
and a modulated one: sometime they are serious, at other times negligible.
A lot depends on whether the disagreement can be assigned to bias or
ignorance on the part of the other party, or, on the contrary, whether it may
be due to the government acting against the best expert advice, or in matters
which affect communities or groups of people whose life experiences are
foreign to those in power.
A second epistemic consideration has been discussed since Aristotle.

Some collective, public ways of reaching agreement can give one greater
confidence in the principles which emerge with general endorsement. By
openly debating public issues people can pull their wisdom and experiences
together, learn from each other, and reach agreement on views which
perhaps none of them held at the outset, views which are supported by the
collective experience and knowledge of everyone in the group.
Finally, let me mention briefly two familiar considerations of principle.
First, if possible one should avoid creating situations in which people are
forced to act against their conscience. In its purest form this concern
manifests itself in rights of conscientious objection. Such rights, however,
have undesirable aspects. They involve public authorities adjudicating on the
sincerity of people's avowals of belief, as well as forcing people to declare
their beliefs in public. It is therefore desirable to avoid the circumstances
giving rise to rights of conscientious objection. This can sometimes be
achieved; in particular it can be achieved when it is possible, instead of
imposing duties to act in certain ways and exempting objectors from them,
to maintain the service the duty would have provided by rewarding those
willing to engage in the desirable course of conduct.
The offense to people of being forced to act, as they see it, immorally
should be avoided whenever it is reasonable to do so. In many cases it is
similar to, or even associated with another offense, that of being treated as
a second-class citizen. As has often been pointed out, people have a public
as well as a private life. Their well-being is tied up with their success in
love and friendship, and in many private pursuits. But it is also wound up
with their public life, as full citizens of their own country, respected by the
authorities, and by other people. This respect includes allowing people to
engage in their religious practices, or act on their fundamental personal
beliefs and preferences without being insulted or victimized for doing so.
Respecting people by not forcing them to act in a way they believe to be
immoral, and by not taking actions which will condemn them to second-
class status and alienate them from their society are two moral concerns
requiring restraint in the face of disagreement over principles. Like the
pragmatic and epistemic concerns I mentioned, they differ from doctrines
like neutrality, consent, or minimal-state which I criticized earlier. They do
not recommend radical restraint in the face of disagreement over principles.
Rather, they are considerations which provide general guidelines toward

compromise solutions in some cases of such disagreements, or indeed any
others, for they are all relevant to any kind of disagreement over political
measures.
While being general considerations, and thus, if you like, matters of
principle, their application depends on detailed evaluation of the facts. Even
the application of the last two moral considerations I mentioned last are
highly sensitive to contingent facts, for it is a mistake to think that what
counts as state action which alienates its members is a matter of first
principles. It is in fact highly contingent. I will give but one example: if in

some countries one were to introduce a state religion and decree that the
head of state must be a member of that religion, and that the high-ranking
priests of that religion will have nonelected seats in the parliament of that
state, I would not be surprised if all other religions as well as atheists will
find the measure insufferable, and decry it as condemning them to second-
class citizenship. But for a very long time no one felt that of Britain, where
such measures are the law. So what alienates groups in the society and what
does not cannot be determined as a matter of first principles. It is to a
considerable degree dependent on contingent facts.
In their dependence on contingencies, the six considerations I mentioned,
and others like them, are typical of most considerations which inform
normal politics. The doctrines I criticized, those of the minimal state, of
consent, and of neutrality, are among the considerations which seem more
suitable to be incorporated in bills of rights, and adjudicated upon by
constitutional courts. This, I suspect, is part of their appeal in the United
States and Canada.
I would not wish to suggest that the nature of the considerations I listed
dictates that these matters should be left to be decided through the process
of parliamentary politics. There is no direct correlation between types of
moral-political consideration and the institutions appropriate for their
enforcement. The only conclusion I would draw is that the institutional
question should be decided in a way sensitive to the traditions and the
conditions of different countries at different times, and that the institutions
which are most likely to best implement the correct considerations be
entrusted with the task. As you see, I am inclined toward a position which
is informed by moral concerns but regards their application as a thoroughly
nuanced and pragmatic matter.

Raz. 98. Disagreement in Politics

Uploaded by

Copyright:

Available Formats

Raz. 98. Disagreement in Politics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Raz. 98. Disagreement in Politics

Uploaded by

Copyright:

Available Formats

DISAGREEMENT IN POLITICS

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Political conflicts are often conflicts about the allocation of scarce

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

B. What Is Special about Our Problem?

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

3. Reliably and with understanding. Obviously, we may be able to do so accidentally

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

C. The Minimal-State Solution

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

E. The Reasonable-View Condition

8. Making reasonable disagreement the focal point of a doctrine of the limits of

To focus the discussion let us imagine a specific principle using the

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

In this passage he presents his explanation as an explanation of a

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

question are justified. They would be relying in part on testimony, as we all

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

matter how "fundamental beliefs" is understood. It can be taken to refer to

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

G. Respecting People through the Propagation of False Beliefs

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

20. It is possible to have a more restrictive condition, requiring derivation of the

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

required to accept further falsehoods which, while justified by their existing

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

H. Waldron on the Right of Rights

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

. . . they are not open to the criticism above.

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

26. Ibid., 251-52.

procedure, then the argument assumes knowledge of the content of the

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Waldron is aware of the considerations advanced above and finds no

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

/. The Moral Standing of Historical Settlements

29. Ibid., 254.

if I accede to your view in spite of my disagreement with you I will not

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

J. Some Additional Pointers

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

protective measures in case one is wrong. Naturally, sometimes reexamina-

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

Rather, they are considerations which provide general guidelines toward

Downloaded from http://ajj.oxfordjournals.org/ at Univ of Southern California on April 6, 2014

You might also like