The University of Sydney Law School

Legal Studies Research Paper Series

No. 19/64

October 2019

Rule of Law v. the Rule of Men and Women:

What’s in the Distinction?
A very short essay for Martin Krygier
Wojciech Sadurski

This paper can be downloaded without charge from the

Social Science Research Network Electronic Library
at: [Link]
 Rule of Law v. the Rule of Men and Women: What’s in the Distinction?
A very short essay for Martin Krygier
Wojciech Sadurski

Martin Krygier’s enormous contribution to rule-of-law scholarship is both original and

compelling – two characteristics which rarely go hand in hand in a scholarly world in which
almost everything has been said about anything of importance (and also so much about things
of no importance at all). It is, first, original, because his main thesis – that the rule of law should
be interpreted teleologically rather than anatomically – goes against much of the conventional
wisdom in legal scholarship and practice which privileges a checklist approach. Longevity of
rules? Publicity? Non-retrospectivity? Judicial independence? etc. – when a sufficient number
of items are ticked, the system is validated as rule-of-law compatible, under a conventional,
“anatomical” conception. Second, it is compelling because once one considers Krygier’s
argument that what really matters for our concerns about the rule of law, and what animates
our “checklist”, is a deeper purpose having to do with the tempering of power, and with curbing
arbitrary uses of power, it is impossible to deny the wisdom of the thought. As Krygier says,
in one of the most recent articulations of his theory, “Over generations, indeed millennia,
thought about the rule of law has been haunted by the spectre of arbitrary exercise of state
power …. Where arbitrariness is available … despotism is not likely to be far away….”. 1 In
different times and places, different tools may serve – for better or worse – this underlying
purpose of preventing arbitrariness and despotism, so why would one be concerned with the
tools rather than the ends?
I cannot hope to add anything of interest to the field of enquiry dominated by the towering
figures of Krygier, as well as Joseph Raz, Jeremy Waldron or Friedrich Hayek. What I want to
do in this short essay is to offer some observations on the opposite of the ideal of the rule of
law, i.e. the much-maligned concept of the “rule of men [and women]” (as in the conventional
contrast: “the rule of law v. the rule of men”). By reflecting upon this counter-ideal, I discern
some aspects of the ideal of the rule of law itself – very much like the way an inquiry about
unfreedom may reveal something of interest about the value of freedom. I will proceed in the
teleological spirit of Krygier’s rule-of-law oeuvre, and will attempt to ascertain what it is that
is truly objectionable about the counter-ideal. Or perhaps it is not that objectionable, after all
(Part 1)? This aspect of my argument will be modelled on “devil’s advocate” reasoning. Having
discerned the only interpretation of the counter-ideal of the “rule of men and women” which is
serious and objectionable at the same time (Part 2), I will propose a way of addressing the harm
discerned in this counter-ideal, and in this way, contribute to our thinking about the ideal of the
rule of law itself (Part 3).

1
Martin Krygier, “The Rule of Law and State Legitimacy”, in Wojciech Sadurski, Michael Sevel &
Kevin Walton, eds., Legitimacy: The State and Beyond (Oxford: OUP 2019) 106-136 at 114-115.

1
1.
On its face, the hostility to the counter-ideal of the “Rule of Men and Women” (RMW) appears
nonsensical, both as a descriptive, and as a normative matter. In these two senses, The RMW
triggers what may be called a dilemma of the de-personification of law. As a description, the
RMW (understood as the opposite of the idea of Rule of Law, RoL) is completely misleading:
laws are being made (enacted, issued, amended, or repealed) by men and women, not by
impersonal forces such as a deity, nature, the spirit of the people, or history. All these
phenomena (and more) may certainly affect or precede the creation of law, but the law itself,
both as specific rules and as more general principles, are legal by virtue of human decisions.
To de-personify the law risks creating a fundamental epistemological confusion in our thinking
about the origin of legal rules and principles: we will be searching for them in the wrong places,
and attributing them to false makers.
At a second, normative level, de-personification is deeply objectionable: it leads to absolving
specific individuals or groups of individuals (such as political parties, legislatures,
constitutional courts, etc.) of the responsibility for bad (unjust, unfair, oppressive) laws. If we
object to a given rule or principle, besides wanting to remove it from the lawbooks, we may
also want to inflict liability upon its authors, especially if we believe that the creation of the
bad law was not an innocent error but an act of abuse or misuse of power – for instance, with
self-serving or corrupt intentions. De-personification of law absolves those guilty of abuse of
persons of their responsibility, as they can hide behind an allegedly impersonal “law”.
So the idea of wrongness of RMW is neither descriptively nor normatively compelling, if it is
understood as a negative, unattractive contrast to the RoL. And yet, considering the
fundamental significance of this opposition to our understanding of the RoL and the
stranglehold it has on our thinking, we may try to retrieve the rational and persuasive kernel in
this opposition – but we need a better understanding of what the RMW might consist of, and
why it would be unattractive. We need a fresh start. Very much in the spirit of Krygier’s
enterprise, we should ask ourselves about the telos of the RMW, and then construct it in a way
which will lay its unattractiveness bare. We need an interpretation of the RMW which reveals
what is invidious about it, and in this roundabout way, discern what is attractive about RoL
when understood as an antinomy of the RMW so interpreted.

2.
I suggest that a useful place at which to start is Friedrich Hayek’s understanding of what is
wrong about the RMW (though he has not used this term). Rule of Law for Hayek was
functionally equivalent to his conception of liberty, as sketched in the first pages of
Constitution of Liberty, and its paramount value lay in the fact that under a regime of liberty
no one person’s will can constrain the range of choices of a specific other person, and in this
way coerce her to act in accordance with the will of the coercer. This understanding actually
resonates well with the antipathy to the RMW, and preference for the rule of (impersonal) law.
“Whether [a person] is free or not does not depend on the range of choice but on whether he
can expect to shape his course of action in accordance with his present intentions, or whether
somebody else has power so manipulate the conditions as to make him act according to that

2
person’s will rather than his own”.2 This is his theory in a nutshell, but it should do for our
purposes. RoL understood in contradistinction to RMW in this Hayekian perspective best
protects individual liberty because it renders it impossible for anyone to subject anyone else to
the coercer’s will, and serve her (the coercer’s) interests.
There are many problems with this conception of liberty, and consequently of the RoL, but let
us arguendo adopt it here as a platform from which to ascertain the disvalue of the RMW. That
disvalue is not obvious. In a democracy, in particular, persons in power often enjoy the
legitimacy of their constituency, most obviously through their electoral mandate, and we may
assume that even if they do coercively shape specific citizens’ rights and duties (a factor which
matches Hayek’s condition for non-liberty), they do so in a benign and non-oppressive way.
And yet, it is one thing for them to shape their citizens’ bundles of rights and duties (and hence,
their scope of liberty) in a paternalistic fashion and another thing for them to try to track and
reflect, as well as they can, the citizens’ true preferences and values. We may venture a
terminological proposition that the best way of understanding the (unattractive) RMW is that
the governors shape their subjects’ situations without trying to ascertain whether the maxim
forming the basis for authoritative directives is acceptable to their subjects. In contrast, the
value of the (attractive) RoL, understood as an antinomy of the concept of the RMW and of its
disvalue, is characterized by the sensitivity of rulers to the principles already held by the
citizens. If the citizens can be plausibly represented as in some way the co-authors of rules
applied to them, they cannot protest by saying the rules produce an objectionable coercion
serving the ends of a coercer, which is an evil targeted by Hayek’s conception of liberty. They
(the rulers) do not need to make sure that all authoritative decisions are acceptable to all
subjects to whom they apply – this would be an impossible ideal – but, more modestly, they
may ask themselves whether the principles presupposed by those authoritative decisions are
rationally acceptable – or at least non-rejectable – by those to whom they apply. As Philip Pettit
put it, such interference by the “regime of equal, effective control over the state” is non-
dominating, and it does not “involve the imposition of an alien will”.3 If non-arbitrariness is an
indicator of a proper understanding of the RoL (as it is, for Krygier), and if non-arbitrariness
requires that the authorities attempt to track the actual preferences, values, and interests of the
addressees of the rules (pace Pettit), then the connection between the RoL and the directive
that those who wield public power, supported by coercion, should attempt to track actual
preferences of those to whom the rules apply, becomes syllogistically evident.

3.
The “rational acceptability” standard (or, in a negative formulation, that of non-rejectability)
should incline our attention to the value of the justification of legal coercion, and it must be a
justification in both senses of the word: being grounded in proper reasons (as in, “this is a
justified proposition”), and also being properly and convincingly explained to those to whom
the rules apply (as in: “This law had been properly justified by its proponents to the

2
Friedrich A. Hayek, The Constitution of Liberty (Henry Regnery: Chicago 1960) at 13.
3
Philip Pettit, Just Freedom (W.W. Norton: New York 2014) at 112.

3
legislature”). As Thomas Nagel put it, “when we force people to serve an end that they cannot
share, and that we cannot justify to them in objective terms, it is a particularly serious violation
of the Kantian requirement that we treat humanity not merely as a means, but also as an end.
The justification of coercion must meet especially stringent standards”.4 But what are the
grounds of justification which are likely to mitigate the pain of coercion? It is useful here to
recite Nagel’s four-fold classification of grounds of coercion, in his Equality and Partiality,
based on different degrees of acceptance or acceptability of its grounds to a coercee:
(1) [G]rounds which the victim would acknowledge as valid; (2) grounds which the
victim does not acknowledge, but which are nevertheless admissible because he is
grossly unreasonable or irrational not to acknowledge them; (3) grounds which the
victim does not acknowledge, without being irrational, but which are admissible under
a higher-order principle which he does acknowledge, or would be unreasonable not to;
and (4) grounds which the victim does not acknowledge – either reasonably or even
somewhat unreasonably – and which are such that he cannot be required to accept a
higher-order principle admitting them into political justification even if most others
disagree with him.5
As one can immediately see, there is decreasing force to the grounds justifying coercion in this
list. The first is a no-brainer: even though there is coercion, it is consistent with the actual
preferences of the coerced persons; the necessarily coercive resolution of a collective action
problem (eg the Prisoner’s Dilemma) is an example. The use of force is necessary to actually
implement the individuals’ actual preferences about the creation of public goods, so the point
of coercion may be seen to enable a person achieve her goals (congruent with public goods)
rather than to restrict her autonomy. A person may display a (psychologically familiar) gap
between her purposes and her incentives to act towards them, or else suffer from a weakness
of will, or imperfect knowledge, or defects in reasoning from the goals to the means, etc.
An example of the second category is any reasonable criminal law prohibition: it is patently
unreasonable for the offender to disagree with the grounds of the use of coercion towards him
though, as an empirical-psychological matter, he may disagree with the actual act of coercion.
His liberty is restricted but the justification of coercion towards him is easy. As Arthur Ripstein
put it, “the fundamental question of punishment is not whether the criminal already accepts the
punishment, but whether it is justified anyway. To suppose that coercion is illegitimate unless
the wrongdoer accepts the standard by which he or she is judged is to give up on the idea of
fair terms of interaction…”6
The third category is the most difficult problem for liberals. Here, a strong justification is
required, and it will not always be available, but once it is, it will largely reduce the

4
Thomas Nagel, “Moral Conflict and Political Legitimacy”, Philosophy & Public Affairs 16 (1987):
215-40 at 238. See also, similarly, Thomas Nagel, Equality and Partiality (Oxford University Press:
New York 1991) at 159.
5
Nagel, Equality and Partiality, at 159-60.
6
Arthur Ripstein, Equality, Responsibility, and the Law (Cambridge University Press: Cambridge
1999) at 8.

4
oppressiveness of coercion, and the sense that it flows from the RMW rather than the RoL. If
a “higher-order principle”, of which Nagel speaks, is such that it is unreasonable for a “victim”
(a coerced one) to acknowledge, and the grounds for the use of coercion flow from this higher-
order principle, the sense that the act of coercion is an arbitrary use of power by men and
women against a “victim” is greatly minimized.
The fourth ground is the least acceptable, from the point of view of reducing the coerciveness
of the RMW: for instance, the enforcement of comprehensive moral or religious conceptions
which a coerced person does not espouse belongs to the fourth category. No matter how high
up in the justification chain we go, we are unable to identify the “grounds” for coercion which
are rationally acceptable to a person coerced. In such circumstances, the law mandating an act
of coercion cannot really be seen to match the RoL and smacks of the dreaded RMW. This
evokes a situation in which “a person cannot reasonably be expected to agree to a system which
authorizes the use of political power to enforce or promote values opposed to his own…”7 Here
public reasons, or any other impartial, neutral justifications for coercion are not available, and
so the restriction of freedom is the most palpable, and difficult, to justify. This, I argue, is the
harm targeted by the idea of the RoL, and exemplified by the RMW.
Now if we understand the RMW in this way, the puzzle depicted earlier in this essay (Part 1)
disappears. The puzzle, let it be reminded, consisted in the fact that the RMW understood at
face value could not reveal what is wrong about it (and consequently, what is right about the
RoL, if the latter is seen as a direct antinomy of the RMW): both empirically and normatively,
the RMW seemed correct, and resistance to it – misplaced. However, if we dig behind the
textual formulation of the RMW and see it as a proxy for a particular harm about the bad law
– namely, the harm of producing Hayekian coercion – the badness of the RMW becomes clear,
and so does the rightness of the RoL, as an ideal serving the purpose of limiting arbitrary uses
of power, and to the clarification of which Martin Krygier made such a lasting contribution.

7
Nagel, Equality and Partiality at 161.