The dualistic Grotian conception of international society

Edward Keene
According to Martin Wight, Grotians believe in a dualistic or concentric
conception of international society (Wight 1977, 128). He described
this position in terms of the idea that there are two circles of
international society in the world: a universal outer circle that
comprises the whole community of mankind and is governed by natural
law, founded on Gods will or human reason; within which there is an
inner circle, composed only of Christian, European or civilised states,
and regulated by positive legal rules based on their consent. In most
current theories of international society such as Hedley Bulls
influential account of the anarchical society (Bull 1977) -- this dualism
tends to be understood in broadly chronological terms. Employing an
historical narrative borrowed from the opponents of the French
Revolution, theorists like Bull emphasise the practical importance of the
inner circle of the society of states in the modern world, and treat the
universalist, natural law strand in the Grotian tradition as a hang-over
from medieval Christendom or as a prematurely hypothetical vision of
the future. The latters value is not so much that it describes a pattern
of order that actually exists today, but rather that it provides a guide for
possible reforms and a moral standard against which to assess the
prevailing norms, rules and institutions of the modern society of states.
Wight noted, however, that not all Grotian thinkers have accepted
this picture of the relationship between the two circles of international
society; some prefer to think of Grotian dualism in geographical, rather
than chronological, terms. Instead of locating Grotian natural law
arguments in the context of the middle ages or a forward-looking
utopianism, for example, Charles Alexandrowicz contended that one
might rather understand them in terms of early modern international
politics beyond Europe (Alexandrowicz 1967). In his view, European
relationships with non-European peoples originally constituted a highly

egalitarian and universal international society where Europeans,

following the prescriptions of natural lawyers like Grotius, recognised
non-European rulers as full members of the family of nations. This
universal international society, Alexandrowicz went on to say, gradually
contracted during the eighteenth and nineteenth centuries as positivist
legal doctrines pushed the naturalist themes of the Grotian tradition to
the margins of European thinking, and as Europeans began to acquire a
significant commercial and military edge over non-European peoples.
Non-European rulers were, in effect, evicted from the universal society
of the natural lawyers, which shrank to become merely a society of
European states; they were then forced to apply for (re)-admission into
the society of states, and were obliged to accept more European criteria
for entry as they did so.
I want to pursue this line of thinking about the geographical
limits of international society here, but along a very different path from
that taken by Alexandrowicz. In my view, there are two main problems
with Alexandrowiczs account. First, his analysis of the sixteenth and
seventeenth-century law of nations beyond Europe is far too generous
to the early modern Europeans; really, his argument strikes me as an
attempt to manipulate the evidence so as to defend the moral
superiority of the natural ius gentium over the positive ius inter gentes.
The crucial gap in his work here is his too abrupt dismissal of the
colonialist and imperialist practices that were already a prominent
feature of European activities beyond Europe by the seventeenth
century; if we include these in our analysis, we will obviously come
away with a very different, and far less egalitarian, picture of the
norms, rules and institutions governing relations between Europeans
and non-Europeans in the early modern period. Secondly,
Alexandrowicz overstates the extent to which positive law and the
society of states erased the normative foundations of the naturalist
universal family of nations. Since he sees natural law theory through
the rose-tinted glasses of universal fellowship and egalitarianism, he

exaggerates the extent to which the exclusionary and oppressive

treatment of non-European peoples in the eighteenth and nineteenth
centuries was in fact antithetical to the earlier practices of Europeans
in the sixteenth and seventeenth centuries. The most important point
here, as I see it, is that if we adopt a somewhat more balanced
assessment of the norms governing early modern relations between
Europeans and non-Europeans we can see how those norms persisted
into the later modern period as well. The norms that governed extraEuropean international politics were not simply abandoned or pushed
aside in the contraction and expansion of the European society of
states; they persisted and continued to inform relations between
Europeans and non-Europeans during the twentieth century; evidence
of their lingering importance can even, I would argue, be found in the
contradictions that mark world political order today.
These two criticisms should not detract from the fact that
Alexandrowiczs thesis on the geographical limits of modern
international society opens up an unorthodox, but interesting and
enormously fruitful, line of enquiry into the normative, legal and
institutional structure of the modern international society. If we amend
Alexandrowiczs substantive analysis in the ways hinted at above, this
approach allows us to consider the possibility that there was not one
but two modern international societies: a society of states in Europe,
and a colonial international society beyond Europe. These
international societies emerged at roughly the same time, and it is,
generally speaking, difficult to make a clear chronological distinction
between them. Moreover, in geographical terms, contra Wight and
Alexandrowicz, they reflected the existence of separate patterns of
international order rather than concentric ones; each involved its own
characteristic normative principles, legal rules and institutional
arrangements. (Nor, it should be pointed out, can the distinction
between them be understood simply in terms of the dichotomy between
natural and positive law). On the whole, but with some important

reservations, I would accept the conventional view that the European

society of states was designed so as to resist potential hegemons and to
maintain peaceful coexistence in the context of cultural, political, legal
and religious pluralism; to this end, its fundamental norm was that
states should reciprocally recognise each others territorial sovereignty.
The colonial international society, however, was dedicated to the
provision of good government and the civilisation of non-European
peoples, including such issues as the defence of non-European peoples
against harsh or arbitary indigenous rulers, the cultivation of trade, and
the protection of Europeans and their interests overseas. Here, the key
norm was that the sovereign prerogatives of non-European rulers
should be divided between themselves and the relevant European power
as dictated by the requirements of good or civilised government.
Non-European rulers were thus rendered, in the international legal
terminology of the time, semi or demi-sovereigns, and European
rulers took upon themselves a few specific sovereign prerogatives -such as the authority to determine systems of property rights and
taxation -- and claimed the wider competence to decide how such
prerogatives should be distributed.
The appeal of this version of the dualistic conception of
international society for me is that it helps us to make sense of a
number of phenomena that appear as anomalies or irregularities
from a more orthodox perspective (see Bull 1977, 274-75). Indeed, Bull
himself openly acknowledged that his conception of the anarchical
society of states needed to be supplemented in order to cope with the
phenomena involved in relations between European and non-European
peoples prior to the granting of independence to the latter:
In the gradations of independence recognized by the
European powers in the extra-European world, the
spectrum
of
positions
intermediate
between
full
sovereignty and the status of a colony (spheres of
influence, protected states, protectorates, subjection to
imperial paramountcy), there could be seen the survival,
alongside the concept of a society of equally sovereign

states, of the older and historically much more ubiquitous

concept of international relations as the relations between
suzerains and vassals (Bull 1984, 126).

I strongly suspect that Bull would not have agreed that suzerain-vassal
relations should be described as an international society at all. But, if
we think of the concept of international society as simply the
proposition that international relations are norm-governed, then to deny
that a conception of extra-European politics in terms of imperial
suzerainty is a conception of international society -- although obviously
a very different conception from Bulls -- one would have to show either
that imperialism did not involve international relationships as such; or
that colonial and imperial practices were not norm-governed. I concede
that both of these points are arguable, but it seems to me, in the first
place, that common sense dictates that we should consider imperialism
as a facet of modern international politics; indeed, Bull seems to tacitly
accept this when, in the above comments, he treats ideas about
suzerainty as a conception of international relations. On the second
point, Bull and others have maintained that imperialism does not
represent an international society because it involved relationships that
were not conducted in an appropriate way. According to Bull, for
example, beyond Europe
A pattern of economic, military, and political interaction
had grown up in which not only European states and their
colonies of settlement but Asian, African, and Amerindian
rulers and peoples were involved -- in different ways and to
different degrees. But they were not united by a perception
of common interests, nor by a structure of generally agreed
rules setting out their rights and duties in relations to one
another, nor did they co-operate in the working of common
international institutions (1984, 117).

The trouble with this objection, however, is that it presupposes

precisely the positivist conception of international society that is at
issue here. To the extent that the colonial international society involved
norms that were unilaterally imposed by Europeans and rationalised
through natural law arguments, of course it would not have the

characteristic features of a positivist system of international law where

states volition, consent and willingness to participate is crucial.
As one might expect, modern international lawyers were
somewhat vexed by the fact that two different sets of guiding principles
were followed by Europeans in their international affairs, depending on
the specific context of the relationships at hand. By and large, the
positivist lawyers dealt with the problem in the most straightforward
way possible: they ignored it. More exactly, they either denied that
imperial governance involved international relations at all, but operated
somewhere in an odd twilight world between the international and the
domestic; or they dismissed the semi-sovereigns as anomalies that no
reasonably simple doctrine of international law should be expected to
explain. Instead, as Alexandrowicz pointed out, they relied on an
historical account of the modern law of nations that was narrowly
focussed on the consolidation of territorial sovereignty in Europe to the
exclusion of other phenomena. But the more authentic Grotian thinkers,
like Henry Sumner Maine or Arthur Berriedale Keith, did not adopt that
solution. On the contrary, they asserted that the divisibility of
sovereignty was a basic principle of international law, and they
employed qualitative distinctions between Europeans and nonEuropeans to rationalise the different codes of conduct in the two
international societies. It was a toss-up as to which of these lines of
argument should be labelled the mainstream view in later nineteenthcentury international law; the positivists probably shaded it, but the
Grotians were certainly not lacking in influence, especially when it
came to the difficult task of explaining and justifying the assertion of
imperial paramountcy.
I do not have the space here to cover this entire argument.
Instead, I will reduce my discussion to an overview of two key points:
the importance of ideas about divisible sovereignty in the Grotian
tradition; and the way in which sovereignty was divided in practice in

the colonial international society. I want to highlight these two areas

because I think that a necessary first step here is to recover a sense of
the theoretical and practical anomalies that are generally ignored or
forgotten in international society theory. Simply put, Bulls account of
the European society of states tells half the story; here, I will sketch the
main themes in the other half, and we can then begin to consider how
the two halves might be put together.
1. Grotianism and the theory of divisible sovereignty
Andrew Linklater has argued that a key element of modern
international society is the unitary conception of sovereignty developed
by Bodin and bolstered in the subsequent literature (Linklater 1996,
95). The part of this claim that I find most significant is the idea that
Bodins conception of sovereignty was bolstered in the subsequent
literature, since that suggests why we should not treat Bodins
justification of princely absolutism simply as a sixteenth-century
curiosity. After a brief exposition of Bodins theory, I will examine the
reception of his ideas by subsequent political theorists, pointing
especially to the way in which Grotius and his followers developed a
completely different conception of sovereignty.
Bodin and theories of indivisible sovereignty
In essence, Bodins theory was a rejection of the Aristotelian conception
of the mixed constitution: a political arrangement combining elements
of democracy and oligarchy, but inclining towards the former. Such a
constitutional arrangement required a certain distribution of the
deliberative, executive and judicial powers of government, such that
parts of these powers would be held by all the members of the society,
while some would only be held by a few (Aristotle 1958, Bk 4). Bodin
held that this implied the sharing or division of the prerogatives of
sovereignty (Bodin 1992, 47), which he regarded as both illogical and

dangerous. It would be illogical because, on Bodins definition, a

sovereign could not be subjected to any other human power, and
therefore the prerogatives of sovereignty have to be of such a sort that
they apply only to a sovereign prince. If...they can be shared with
subjects, one cannot say that they are marks of sovereignty (Bodin
1992, 49). Bodin recognised that this implied that it was an error to
treat all the powers of government as prerogatives of sovereignty;
instead, he treated only the power to make and repeal law as a mark of
sovereignty, properly understood, and contended that law-making
includes all the other rights and prerogatives of sovereignty, so that
strictly speaking we can say that there is only one prerogative of
sovereignty, inasmuch as all the other rights are comprehended in it
(Bodin 1992, 58). Moreover, Bodin argued, Aristotle s idea of the mixed
constitution would be dangerously unstable, since [s]overeignty
will...be tossed up and back between two parties, and sometimes the
people, sometimes the prince will be the master (Bodin 1992, 27). He
warned that, in such a situation, it must always come to arms until
such time as sovereignty resides in a prince, in the lesser part of the
people, or in all the people (Bodin 1992, 104). Thus, no such state [as
the mixed polity] has ever existed andnone can be made or even
imagined, because the prerogatives of sovereignty are indivisible
(Bodin 1992, 104).
While Bodins additional arguments in support of princely
absolutism were frequently contested, his argument for the indivisibility
of sovereignty was widely influential even among his political
opponents. In large part, this was because it could be adapted to give a
potent justification of resistance against princely absolutism through
the simple device of transferring the locus of the absolute and
indivisible sovereign power from the prince to the people. Bodins
argument was therefore sufficiently flexible to allow a major early
modern debate about political authority and obligation to be conducted
entirely within its terms of reference. For example, it provided a

framework for Johannes Althusius s argument for indivisible popular

sovereignty (see Gierke 1967); and it permitted a strong justification of
resistance on the basis of inalienable popular sovereignty by the French
Huguenot monarchomachic theorists (who also used an argument to the
effect that inferior magistrates -- the nobles on whose support they
relied -- could act against the prince in defence of the peoples
sovereignty). Attention is sometimes paid to this resistance-theoretical
use of Bodins theory by historians of international thought (see Wight
1991, 9), but more often post-Bodinian thinking about sovereignty is
understood through the work of Thomas Hobbes and Jean-Jacques
Rousseau (see, for example, Camilleri & Falk 1992 or Knutsen 1992).
Hobbes, of course, found Bodins warnings about the dangers of
dividing sovereignty to be especially pertinent in the context of the
English civil war:
rights [of sovereignty] are indivisible....a kingdom divided
in itself cannot stand: for unless this division precede,
division into opposite armies can never happen. If there
had not first been an opinion received of the greatest part
of England, that these powers were divided between the
King, and the Lords, and the House of Commons, the
people had never been divided and fallen into this civil war
(Hobbes 1962, 139-40).

For his part, Rousseau reformulated Bodins logical objections to

divided sovereignty, through the concept of the general will: [f]or either
a will is general or it is not; it is either the will of the whole people or of
only a part of it.... [A]nyone who thinks that he sees sovereignty divided
is always mistaken...all the rights taken for parts of it are subordinate to
it and always presuppose supreme wills which they merely put into
execution (Rousseau 1974, 25-26).
In themselves, though, Hobbes s and Rousseau s arguments already
raise some interesting questions about the extent of Bodins influence,
as depicted by Linklater. For example, it is intriguing that about one
hundred years after Bodin wrote the Six Books of the Republic, Hobbes

still had to confront an opinion received of the greatest part of

England that sovereignty could be divided. This suggestion is
confirmed by Julian Franklin, who has analysed how the confusion
created by Bodin s theory of indivisible sovereignty was cleared up in
the course of the debate on the locus of sovereignty in the German
Empire (Franklin 1991, 299). Here, the debate among academic jurists
was quite different from the debate between princely or popular
absolutists. The basic problem was that [t]he German princes
claimed...virtual sovereignty over their principalities, yet still
acknowledged the superiority of the Holy Roman Emperor
(Koenigsberger 1997, 62). Bodin s doctrine was enormously difficult to
apply to this situation, since either it would have meant denying the
constitutionally-guaranteed prerogatives of the German princes, or it
would have implied that the Empire was a monarchy at all. Ironically,
the Peace of Westphalia exacerbated this problem: it simultaneously
affirmed the rights of the German princes, while endorsing the
international personality of the Emperor, who was one of the most
important signatories of the treaties. What status, then, would the
treaties have had, if Bodin s unitary conception of sovereignty was to be
accepted?
As historians of international thought are well aware, there were efforts
to solve this problem, most notably in Samuel Pufendorf s description of
the Empire as a special kind of constitutional arrangement -- pregnantly
labelled as a states-system -- within which sovereignty was indivisibly
held by the individual states, all of which were bound together by
special ties of comity. However, while there were some who followed
Pufendorf, there was also an unbroken succession of theorists who
adopted the contrary, and much more straightforward, line that Bodin
had simply been wrong, and that the sharing of sovereignty as well as
the distribution of its parts are...modes of mixture that are ultimately
consistent with the coordination of governmental functions (Franklin
1991, 328). In his survey of the debate on sovereignty in the Empire,

10

Franklin identifies Christoph Besold as the originator of a theoretically

decisive accont of the divisibility of sovereignty in a mixed constitution
(Franklin 1991, 328). Another prominent figure in this literature was
Gottfried Wilhelm Leibniz, who answered Hobbes s warnings about the
dangrs of civil war with the (rather weak) argument that when the
supreme power is divided, many dissensions can arise; even wars, if
everyone holds stubbornly to his own opinon. But experience has shown
that men usually hold to some middle read, so as not to commit
everything to hazard by their obstinacy (Leibniz 1972, 119).
The Grotian conception of divisible sovereignty
The idea of divisible sovereignty was not just applied to the lingering
curiosity of the German empire, however. Crucially, not all resistance
theorists accepted the Bodinian framework of indivisible sovereignty
and invoked it as the basis for popular sovereignty. Some, especially
those associated with the rather less democratic and more aristocratic
or republican forms of resistance theory stuck to the classical
Aristotelian ideal of the mixed polity and turned it -- here Grotiuss role
was pivotal -- into a defence of resistance as a form of just public war.
At first glance, however, Grotiuss arguments seem like a blunt rejection
of the possibility of resistance. For example, he does not even see that
one could reasonably dispute the point that arms may be taken up
against subordinates by those who are armed with the authority of the
sovereign power (Grotius 1925, 138). He simply accepts it as given that
sovereigns have the authority to engage in war against their subjects in
this way. Furthermore, in the reverse case, he argues that as a general
rule rebellion is not permitted by the law of nature, nor by Hebraic law,
nor by the law of the Gospel, nor by early Christian practice (Grotius
1925, 139-46). Beyond these very general and, as we will see, carefully
qualified comments, the primary reason why Grotius is so often
regarded as an absolutist political theorist is his outspoken criticism of

11

monarchomachic resistance theory in De Jure Belli ac Pacis. Certainly,

Grotius attacked both of the planks of monarchomachic theory: the use
of the idea of inalienable natural liberties to argue for popular
sovereignty, and the constitutionalist argument for inferior magistrates
right of resistance. However, although Grotius was definitely opposed to
monarchomachism, he should not, by that token alone, be regarded as
an absolutist; he developed a different kind of constitutionalist
argument for resistance.
Most famously, he denied the monarchomachic claim that the condition
of natural liberty endows people with an inalienable set of rights for
self-government. Thus, he denies that the condition of natural liberty
necessarily leads to an ineradicable core of popular sovereignty, and so
cuts himself off from the Althusian line of response to Bodinian princely
absolutism. It is, so Grotius argues, legally possible for a peoples
natural freedoms to be entirely alienated to a prince, leaving them
without any claim to a residual popular sovereignty: To every man it is
permitted to enslave himself to any one he pleases for private
ownership.... Why, then, would it not be permitted to a people having
legal competence to submit itself to some one person, or to several
persons, in such a way as plainly to transfer to him the legal right to
govern, retaining no vestige of that right for itself?(Grotius 1925, 103)
This question leads to a version of the Aristotelian natural slavery
argument, that some men are by nature slaves...so there are some
peoples so constituted that they understand better how to be ruled than
to rule (Grotius 1925, 105) According to this proposition, not only is it
legally possible for a people to give away their entire right to govern, it
is indeed natural for this to happen in certain cases. Although this adds
force to the argument, it does not substantially change the basic
proposition that it is legally possible for a people to give away all of
their rights to self-government, resulting in the creation of a
patrimonial sovereign authority. Thus, for Grotius, merely to assert the
existence of a condition of natural liberty is not to provide an

12

unanswerable defence of the right of resistance in the face of all claims

to absolute sovereign authority, as the monarchomachs supposed.
Grotius follows this attack on theories of popular sovereignty with a
strong rejection of the monarchomachic proposition that inferior
magistrates have a constitutional right of resistance. To such a claim,
Grotius gives the reply that from the point of view of those possessing
higher authority [inferior magistrates] are private persons. All
governmental authority possessed by public officials is in fact so
subordinated to the sovereign power that whatever they do contrary to
the will of him who holds it is divested of authority and is, accordingly,
to be considered as a private act (1925, 146). In other words, there is
nothing so special about the legal status of inferior magistrates that it
permits them to engage in resistance. This indicates Grotiuss shrewd
understanding of the intentions of the monarchomachs in adapting
constitutionalist arguments to their own purposes. Supposing that the
natural liberty argument on its own was insufficient, the Huguenots
sought to bolster their position by suggesting that resistance was not
really a mixed war at all, but was rather a kind of public war that can
only be engaged in by public officials. Grotius recognises that this is a
plausible interpretation of the category of public war, but nevertheless
contends that if the word public is understood in a higher sense as
characterizing that which is done with due formality, as beyond
question the word often is, such wars are not public, for the reason that
both the decision of the sovereign power and other conditions are
necessary for the fulfilment of the legal requirements involved (1925,
98-99). Therefore, he concludes that a public war ought not to be
waged except by the authority of him who holds the sovereign power
(Grotius 1925, 100-101).
Obviously, this is a rejection of monarchomachic resistance theory.
However, it is not sufficient reason to label Grotius as an absolutist,
who denies a right of resistance altogether, since all Grotius has said up

13

to this point is that the particular monarchomachic justifications of

mixed and quasi-public wars are ineffective: the monarchomachic
version of the natural liberty argument assumes that which has to be
demonstrated, while the constitutionalist argument rests on a
misunderstanding of the concept of public war itself. These are
criticisms of monarchomachic arguments, but they are not
endorsements of the contrasting absolutist position. Indeed, having
dispensed with monarchomachism, Grotius goes on to construct an
account of natural law and the law of nations which justifies resistance
in a quite different way from the monarchomachic argument
One of the most interesting aspects of Grotiuss work here is his
attempt to show that some instances of resistance, including the Dutch
Revolt, are, in fact, just public wars. This argument is based on a theory
of the divisibility of sovereignty, which lies, as Peter Borschberg has
pointed out, between theories of absolute princely and absolute popular
sovereignty; that is to say, between Bodinian and Althusian thinking
(1994, 126). Thus, while Grotius does not accept the constitutionalist
argument that inferior magistrates have a right of resistance against a
ruler, he nevertheless recognises that this does not translate into a
general constitutional ban on resistance. Indeed, here his qualifications
of the general rule are so extensive that, as Hirsch Lauterpacht has
observed, they render the major proposition [that resistance is
unjustified] almost theoretical (1946, 45). Grotiuss arguments on this
point can be divided into two categories.
The first kind of qualification is theoretically unexceptional, and
concerns cases where the king has abdicated the sovereign power or
shown himself to be the enemy of the people (Grotius 1925, 157-58).
These are, for Grotius, in practice the same thing, since the will to
govern and the will to destroy cannot coexist in the same person. The
king, then, who acknowledges that he is an enemy of the whole people,
by that very fact renounces his kingdom (1925, 158). Although, in a

14

practical sense, this proved a very important way of justifying

resistance, it is unexceptional in terms of its consequences for
constitutionalist theory, because here the king simply reduces himself to
the status of a private person, and the question of resistance against an
authority does not really arise.
The second category of qualification of the general rule against
resistance is more interesting. The point here is that legally there are
several different ways in which a people may alienate some or all of its
rights of self-government to a ruler, making some or no reservations on
the absoluteness of that alienation. Grotius specifies three instances of
this: where rulers are responsible to the people; where sovereign
power is held in part by the king, in part by the people or senate; and
where in the conferring of authority it has been stated that in a
particular case the king can be resisted (1925, 156 & 158). In all of
these cases it is clear that the kings sovereign authority is not absolute.
On the contrary, either certain sovereign powers have been reserved by
the people, and the king is thus accountable to them (as in the first
case); or some sovereign powers have been granted by the people to
other representative institutions than the king (as in the second case);
or some sovereign powers have not been granted in an absolute way,
but only conditionally (as in the third case). In each of these cases,
resistance is justifiable as a form of just public war, since sovereignty is
divided within the political community, and in which case whoever
possesses a part of the sovereign power must possess also the right to
defend his part (Grotius 1925, 158). Thus, even though, as we saw
earlier, Grotius denied the monarchomachic argument about inferior
magistrates on the grounds that a public war ought not to be waged
except by the authority of him who holds the sovereign power, this did
not rule out all resistance, because of the divisibility of the sovereign
power itself.

15

Grotius gives these three points a very cursory treatment in De Jure

Belli ac Pacis, although even there he makes it very clear that
sovereignty is potentially divisible, and notes the relevance of its
division for the justification of resistance as a kind of public war. There
is a more extended discussion of this line of argument, however, in the
earlier Commentarius in Theses XI, where Grotius uses the second
qualification (that sovereign power is divided between different
representative institutions) explicitly to justify the Dutch Revolt. In the
Commentarius, as in De Jure Belli ac Pacis, Grotius states that in
principle it is possible for some marks [of sovereignty] to reside
with...persons or assemblies, while others do not (1994, 227). Further,
his example of such an instance is the right of the States of Holland to
raise taxes (Grotius 1994, 219). This consideration is then used to
justify the Dutch Revolt, on the grounds that the Duke of Alba unjustly
removed this mark of sovereignty by levying taxes himself, and
therefore the Dutch are entitled to defend that mark by waging a just
public war (Grotius 1994, 281-83). The complete independence of the
United Provinces is then justified by arguing that the States General
legitimately acquired the other marks of sovereignty from Philip II
through the successful prosecution of this just war (Grotius 1994, 15960).
We can therefore go beyond Lauterpachts cautious suggestion that [i]t
is unlikely that one or more of [Grotiuss] exceptions [to the general
principle of non-resistance] did not recall to the mind of the reader the
various articles of the Dutch Act of Abjuration against Philip II of Spain
(1946, 45-46). In fact, these exceptions clearly reflected the way in
which Grotius had earlier deliberately set out to produce an account of
resistance as just public war that was more acceptable than the
logically dubious Huguenot argument from the quasi-sovereign status of
inferior magistrates, and which was part of the project of justifying the
Dutch Revolt. This account was predicated on the claim that it was
legally possible for the various powers of sovereign authority to be

16

divided amongst several different institutions within a political

community, which is, as Borschberg argues in his introduction to the
Commentarius, directed against Bodin, as well as other exponents of
princely, or royal absolutism, who see full princely sovereignty as an
inalterable and therefore universally applicable foundation of all
political power (1994, 122).
Nor was this Grotian conception of divisible sovereignty merely a
curiosity of seventeenth century thought. It reappears in revolutionary
American political thought. As Bernard Bailyn shows, the dominant
Whig doctrine of absolute parliamentary sovereignty posed an
exceptionally difficult problem for those colonial political theorists who
wanted to justify the revolution. If Parliament was indeed the repository
of the theoretically ultimate supremacy of the people, how could the
right of Parliament to impose taxation in the colonies be denied?
Consequently, [h]ow to qualify, undermine, or reinterpret this tenet of
English political theory was the central intellectual problem that
confronted the leaders of the American cause (Bailyn 1967, 202). As in
the Dutch Revolt, the de facto circumstances of political authority in
America offered a possible solution to the problem faced by the political
theorists. The actual situation was that the powers exercised by the
formally sovereign British Parliament and Crown
were far from total powers; together they did not constitute
governance in depth, nor did they exclude the exercise of
real power by lesser bodies or organs of government. They
touched only the outer fringes of colonial life; they dealt
with matters obviously beyond the competence of any
lesser authority; they concerned the final review of actions
initiated and sustained by colonial authorities. All other
powers were enjoyed, in fact if not in constitutional theory,
by local, colonial organs of government (Bailyn 1967, 203).

Bailyn summarises the situation that actually existed in the colonial

societies as extreme decentralisation of authority within an empire
presumably ruled by a single, absolute, undivided sovereign (1967,
204).

17

Given the similarities in the decentralised nature of political authority

in the sixteenth-century Netherlands and eighteenth-century North
America, it is perhaps unsurprising that political theorists of the
American revolution began to treat the concept of sovereignty in much
the same way as had Grotius in his theory of resistance as just public
war. Their key move in this respect was to propose a qualitative
distinction between the nature of sovereignty over a single nation, and
that of sovereignty in an empire. In the case of the latter, it was argued
that the sovereign body need not be supreme everywhere and in all
matters in the territory it controlled, but only on some issues and in
some ways, and that other, lesser bodies might exercise absolute and
arbitrary powers, sovereign powers in effect, within spheres specifically
allotted to them (Bailyn 1967, 216). As with the concept of
appropriation, the principal difference between the Grotian position and
American political thinking about sovereignty was that American
theorists took the idea of the divisibility of sovereignty much further
and elaborated it into an idea of an imperial federation of sovereign
states (Bailyn 1967, 224), in a way that had not been anticipated by
Grotiuss much more humble conception of a mixed republican
constitution. However, the similarities of these two anti-absolutist
conceptions of sovereignty are far more significant than the differences.
It makes sense to see the theory of sovereignty used in the American
revolution as an extension or consolidation of certain themes that were
present in Grotiuss conception of the law of nations. This runs counter
to the position of those who prefer to use William Blackstones
definition of sovereignty as indivisible, and argue that to see
sovereignty as divisible is merely to confuse sovereignty with authority
(see, for example, Deudney 1995, 198). However, I think that this
misses the point that both the Grotian and American revolutionary
conceptions of sovereignty were worked out in conscious opposition to
the absolutist or unitary doctrine asserted by Blackstone, inter alia.

18

The Grotian conception of sovereignty as divisible was also maintained

outside of America, and well into the nineteenth century. Maine, for
example, used his 1887 Whewell Lectures to chide the legal positivist
John Austin for having mistakenly argued that sovereignty is
synonymous with independence because it is only conceivable in
Hobbesian terms -- as absolute, unitary and indivisible (Maine 1888,
chapter 3). In Maines conception of sovereignty, we can see a quite
different understanding of the classical position from that evident in
Bulls or others work. Not coincidentally, Maine also paid careful
attention to the question of how to understand what Bull regarded as
the anomalous situation of the Indian Native States under British
imperial paramountcy.
It is necessary to the Austinian theory that the all-powerful
portion of the community which makes laws should not be
divisible, that it should not share its power with anybody
else, and Austin himself speaks with some contempt of the
semi-sovereign or demi-sovereign states which are
recognised by the classical writers on international law. But
this indivisibility of sovereignty, though it belongs to
Austins system, does not belong to international law. The
powers of sovereigns are a bundle or collection of powers,
and they may be separated one from another. Thus a ruler
may administer civil and criminal justice, may make laws
for his subject and for his territory, may exercise power
over life and death, and may levy taxes and dues, but
nevertheless he may be debarred from making war and
peace, and from having foreign relations with any authority
outside his territory. This in point of fact is the exact
condition of the native princes of India (Maine 1888, 58;
see Austin 1879).

John Westlake, who devoted a whole chapter of his widely used book on
The Principles of International Law to The Empire of India, was
uncertain that the Native States possessed international personality,
and therefore that their situation should be described in international
legal terms at all; but even he was prepared to entertain the possibility
that it could be, and was quite happy to treat semi-sovereignty as a
useful and valid concept (1894, chapter ten). He certainly made a
strong distinction between sovereignty and independence on the

19

grounds that, while [i]ndependence, like every negative, does not admit
of degrees....sovereignty is partible (Westlake 1894, 87). Arthur
Berriedale Keith, yet another prominent British international lawyer of
the period, began his analysis of the circumstances of the British
Dominions with the observation that sovereignty can be divided, and
that in any country both internal and external sovereignty may be
shared by various authorities (1929, 1).
It is odd to reflect that Maines criticisms of Austinian positivism were
made over one hundred years ago, especially when one finds
contemporary theorists claiming that there is a novel disjuncture in
world politics today because, within the European Union, sovereignty
is now...clearly divided: any conception of sovereignty which assumes
that it is an indivisible, illimitable, exclusive and perpetual form of
public power -- embodied within an individual state -- is defunct (Held
1995, 112-13). It is also strange to think that these opinions were held
by extremely senior and influential members of the international legal
establishment. Maine, to take just one example, was Corpus Professor
of Jurisprudence in the University of Oxford, and far from being cast out
because of his opinions, he was applauded by his successor for his
criticisms of Austinian positivism and his defence of a more
comparative-historical method (Vinogradoff 1904 & 1920). Nowadays,
though, from the point of the orthodox conception of the society of
states, Maines point of view would probably be regarded as heresy.
Witness, for example, Alan Jamess opinion that even the distinction
between internal and external sovereignty is a dangerous terminology,
for it can all too easily be taken to mean that sovereignty can, as it
were, be split down the middle, enabling one of its halves to exist
without the other. In fact, a sovereign state is all of a piece (1999, 464).
2. Imperialism and the practice of divisible sovereignty

20

James would no doubt argue that he is talking about how sovereignty

actually works in practice; Grotians like Maine may well have had a
theory of divisible sovereignty, but, to put it crudely: so what? If this
had no grounding in the actual practice of modern international politics,
why should we waste time reading these archaic texts on international
law? To see why such a criticism is unfair, let us look at just a few
examples from of European colonialism and imperialism. To begin with,
let us we consider Dutch colonisation in the East Indies (an area of
considerable importance to Alexandrowiczs argument, and obviously
relevant to Grotiuss work). Here, we find here a strikingly familiar
strand in the Dutch East India Companys (VOC) thinking about the
sovereignty of indigenous rulers, such as Raja Sinha, the King of Kandy
(Ceyloin). The Dutch had originally acquired various ports and
territories in Ceylon largely through their manipulation of the
provisions of a treaty signed with Raja Sinha in 1638 relating to a
Dutch-Kandyan alliance against the Portuguese (note: the treaty was
crucial to subsequent Dutch claims; not everything depended on natural
law arguments, although those remained important). In this respect, the
Dutch faced a tricky legal problem. Throughout all of their colonising
activities, the Dutch continued to observe, if only formally, Raja Sinhas
sovereignty over the whole of Ceylon, however, they needed to acquire
as strong a legal title as possible to the lands under their control, in
order to uphold it both against the intrusions of other European rivals,
and the claims of the indigenous authorities (Arasaratnam 1958b, 105).
The representatives of the VOC in Ceylon suggested that they should
secure a new contract with Raja Sinha, where he would recognise the
new changes [after the final expulsion of the Portuguese in 1658] and
give a de jure sanction to what was already a de facto sovereignty of the
Dutch over the lowlands (Arasaratnam 1958a, 5). However, as far as
the Directors of the VOC were concerned, such a new contract was
unneccesary, and in the end the Company simply fell back on the
argument that the lands were held in trust from Raja Sinha, who

21

remained the formal sovereign, until such time as their expenses had
been fully paid (Arasaratnam 1958a, 5-6). This ingenious device could
be extended at the Companys desire more or less indefinitely by
concoting new expenses, but as a legal argument it fell well short of
having full authority over the land. Consequently, Dutch power found it
difficult to shake off the stigma of usurped authority over the lowlands
(Arasaratnam 1958b, 110). As pressure on the cinnamon trade
increased from other European powers, notably the English and the
French, so the Dutch began to introduce other arguments, such as
length of occupation and conquest, to justify their control, and they
began to assert their claims in an increasingly exclusive manner,
denying Raja Sinha any rights in the coastal regions. Although these
arguments were much more far-reaching than had hitherto been the
case, they still stopped short of declaring Dutch sovereignty. This
continued well into the eighteenth century. Arasaratnam cites Baron van
Imhoff, Governor-General in Ceylon from 1736-1740:
The great number of years during which we have been in
exclusive possession gives us the right to maintain our
right, if necessary even by force. Our rights have been
legalised by the undisputed exercise of them by the
Companys possession of West and east, and this will serve
also as proof of proprietorship, of other parts, so far as it
concerns a third party, although the King is and remains
the sovereign (cited in Arasaratnam 1958b, 117).

In fact, it was not until 1776 that the Dutch managed to secure a treaty
that formally recognised their sovereignty over the coastal lands. The
point to notice is that, even well into the eighteenth-century, the Dutch
did not claim sovereignty over the coastal regions, but they
nevertheless claimed certain rights, including the right to dispose of
unoccupied cultivable waste land as they saw fit. These prerogatives
were, on any early modern understanding, part of the sovereign power
over the lands. In effect, then, the VOC were arguing that they
possessed certain marks of sovereignty, but simultaneously maintained
that this did not compromise the Kings sovereignty. This position could

22

only have made sense within a legal framework that permitted

sovereignty to be so divided.
The Dutch did not merely assert their rights as creditors of the
sovereign lord of Ceylon, however, nor as long-standing occupiers.
Another important feature of the Dutch strategy was that the coastal
lands that they occupied were vital to communications between the
Kingdom of Kandy and the outside world, and it was their intention to
channel all foreign intercourse and commerce of the Kandyans through
Dutch sources and maintain the Kandyan Kingdom in subordinate
isolationa sovereign entity but deprived of the right of foreign control
(Arasaratnam 1960, 114, emphasis added). Thus, overall, the attitude of
the Dutch was to accede to the formal sovereignty of the Kandyans, but
chip away at the marks of sovereignty contained therein, until they
were in a position to begin to claim complete sovereignty for
themselves. Arasaratnam gives a good overall summary of the policy:
The indigenous state claimed total de jure sovereignty and
this claim was not challenged. But it was not allowed the
full exercise of all attributes of sovereignty in view of the
superior power of the Dutch. In actual practice, the
Kandyan Kingdom though exercising all aspects of internal
sovereignty within its own dominions was externally
dependent on the Dutch. It could not carry on freely its
trade, its foreign relations or its foreign contacts without
the tacit consent of the Dutch power (Arasaratnam 1958b,
121).

Once again, as we saw in the case of the political institutions in the

medieval Netherlands, the marks of sovereignty were divided between
different institutions. This could be used to justify waging war against
the Kings of Kandy, since, as we saw in chapter four, Grotius had
pointed out that, according to the law of nations, whoever possesses a
part of the sovereign power must possess also the right to defend his
part (Grotius 1925, 158). This constitutes an important revision of
Alexandrowiczs claim of the equality of sovereign rights granted to
East Indian rulers under the law of nations. While they may have

23

initially been treated as equals with European sovereigns, and while

their sovereignty was, broadly speaking, recognized, they were
nevertheless vulnerable -- as was also the case for European rulers -- to
the division of their powers with respect to different issues. It is no
exaggeration to say that the conception of sovereignty as divisible was
the centerpiece of the legal regime that eventually allowed European
imperialists to rationalise the gains that they initially won through
commercial pressure and military superiority.
Something similar happened with the British in India, where roughly six
hundred Native States remained in existence, and occupied an
extremely unclear constitutional and legal position with respect to both
the British government and the government of British India. The East
India Company had concluded treaties with most of the larger or more
important of these states, and increasingly relations with the rest were
framed against a broad doctrine, based in part on customary ideas
about Mughal suzerainty and in part on the force of British arms
demonstrated in the Mutiny, that the British called paramountcy.
Although the treaty-based relationships were a more secure footing to
which indigenous rulers could appeal, paramountcy was taken to mean
that all foreign relations and military decisions of the Native States
were in the hands of the British. Rather interestingly, the position of the
Native States with respect to the Empire was held to be analogous to
the States of the United States of America, which had also surrendered
certain external powers (e.g. defence and foreign affairs) to a central
government, but which nevertheless retained residual sovereign
powers (Sever 1985, vol. i., 26).
Paramountcy was obviously flexible and, ominously for the rulers of the
Native States, could be defined and redefined by the British more or
less at will. An activist Governor General or Viceroy, like Dalhousie or
Curzon, could interpret it as a permission to be extremely
interventionist with regard to the domestic affairs of the Native States,

24

or even to conduct a policy of annexation. If it is possible to generalise

about paramountcy, however, we might say that there was a profound
disagreement between liberals and conservatives about its use. Many
liberals were extremely concerned about allowing the rulers of Native
States to retain what often amounted to quite full rights of internal
sovereignty: Convinced that western civilisation was superior and
inspired by the belief that Britain had a moral obligation to reform
Indian society, the reformers were appalled to learn that British policy
encouraged princely mismanagement. Despite their anti-imperialist
sentiments, they thus advocated the termination of princely rule and
became committed to a policy of annexing the states (Ashton 1982, 1213). Conservatives, on the other hand, were obviously less inclined to
castigate the rulers of the Native States merely on the grounds that
they represented an archaic and illiberal feudalism. Moreover, in a point
of view that gained some reinforcement after the Indian Mutiny, they
believed that respecting the (limited) sovereignty of the Native States
would allow them to act as safety valves in order to provide for the
security of British rule (Ashton 1982, 14).
I think that little would be gained here from a discussion of the
practical operation of paramountcy during the nineteenth century.
However, it is important to make two points about the official doctrine.
First, it was legally eclectic. It derived partly from custom and treaty,
but it also depended on natural legal principles: the paramount power
took upon itself the task of supressing inhuman practices.... there
existed a universal prohibition throughout the sub-continent against
suti, slavery and infanticide (Sever 1985, vol. i, 24). Perhaps more
explicitly than when dealing with an independent sovereign over whom
they did not claim paramountcy, the British were prepared to assert the
natural and universal character of their values, and vigorously imposed
them. Secondly, irrespective of the political disagreement between
liberals and conservatives about the application of paramountcy, the
doctrine itself had profound implications for conceptualising

25

sovereignty. Henry Sumner Maine, who combined (not insignificantly)

the dual roles of being an expert on international law and on AngloIndian relations, asserted in an 1864 minute on the issue of the
sovereignty of the Native States:
Sovereignty is a term which, in international law, indicates
a well-ascertained assemblage of separate powers or
privileges.... A sovereign who possesses the whole of this
aggregate of rights is called an independent sovereign, but
there is not, nor has there ever been, anything in
international law to prevent some of those rights being
lodged with one possessor and some with another.
Sovereignty has always been regarded as divisible (cited in
Sever 1985, vol. i, 25, emphasis original).

Certainly, the Native States were not independent, but nevertheless,

they retained some aspects of sovereignty. According to Maine at least,
there was nothing about their situation that was all that peculiar from
the point of view of international law; after all, sovereignty had always
been regarded as divisible. And the theory was not at all an airy or
hypothetical exercise in intellectualism; it was, one could argue, more
closely rooted in practice than the Austinian theory of indivisible
sovereignty.
Concluding remarks
I confess that in this paper I have done little more than skate over the
surface of this argument about the dualistic Grotian conception of
international society. Most obviously, I have only told part of the story;
but I have concentrated on the theory and practice of divisible
sovereignty because this is far and away the most neglected theme
within the Grotian tradition of thought in international relations theory
today. I suspect, indeed, that many students of Grotianism are barely
aware of its existence. My purpose here, then, has been to recover this
line of argument and to give some sense of its practical context in extraEuropean international politics. To take this further, one would need to
look at how Grotians coped with the contradictions between the

26

different normative logics contained in the two patterns of international

order. For now, I hope merely to have given a clearer indication of the
exact problem that this task presented.
Edward Keene
edkeene@hotmail.com

27