“Trust” in Hobbes’s Political Thought

• Author: Deborah Baumgold
• Published date: 01 December 2013
• Date: 01 December 2013
• DOI: http://doi.org/10.1177/0090591713499764

499764PTX41610.1177/0090591713499764Political TheoryBaumgold
research-article2013
Article
Political Theory
41(6) 838 –855
“Trust” in Hobbes’s
© 2013 SAGE Publications
Reprints and permissions:
Political Thought
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0090591713499764
ptx.sagepub.com
Deborah Baumgold1
Abstract
“Trust” is not usually considered a Hobbesian concept, which is odd since
it is central to the definition of a covenant. The key to understanding
Hobbes’s concept of trust is to be found in his account of conquest—
“sovereignty by acquisition”—which is a heavily revised adaptation of the
Roman justification of slavery. Hobbes introduces a distinction between
servants, who are trusted with liberty, and imprisoned slaves. The servant/
master relationship involves mutual trust, an ongoing exchange of benefits
(protection for service and obedience), and performance monitoring. In
contrast to Quentin Skinner’s and Philip Pettit’s shared concentration on
the contrast between slavery and freedom, I argue that the salient analogy is
between servants and subjects.The trust relationship between subjects and
the sovereign involves defined roles, limited absolutism, and accountability
in the early-modern form of licensing subjects to switch political allegiance
should a regime fail.
Keywords
trust, Hobbes, absolutism, Skinner, Pettit
1University of Oregon, Eugene, OR, USA
Corresponding Author:
Deborah Baumgold, Department of Political Science, University of Oregon, 1284 University of
Oregon, Eugene, OR 97403-1284, USA.
Email: baumgold@uoregon.edu

---

Baumgold
839
“Where’s no trust, there can be no Contract.”
Thomas Hobbes, De Cive, 8.3
Trust is commonly thought to be a Lockean concept and not a Hobbesian
one.1 But the estimation would be revised if we paid more attention to
Hobbes’s account of “sovereignty by acquisition,” the “realistic” corollary to
the hypothetical tale of “sovereignty by institution.” On its face, “sovereignty
by acquisition” involves an odd, implausible consent argument: those con-
quered in war—whether individuals or countries—legitimize the victor’s
authority via consent. Doesn’t their lack of choice preclude anything resem-
bling genuine consent? Hobbes’s answer hardly improves the argument. It
seems that consent is signified by a choice on the part of the victor whether
or not to allow the conquered to go free. “Every one that is taken in the War,
and hath his life spar’d him, is not suppos’d to have Contracted with his Lord;
for every one is not trusted with so much of his naturall liberty, as to be able,
if he desir’d it, either to flie away, or quit his service, or contrive any mischief
to his Lord.”2 It appears that it is the victor who decides whether their rela-
tionship is or is not consensual!
But this dismissive interpretive construction is not the last word. More
kindly interpreters, notably Kinch Hoekstra, have given a sympathetic read-
ing to Hobbes’s account of consent in the conquest situation. Addressing the
textual question of whether the principle of a “mutuall Relation between
Protection and Obedience” is a novelty of Leviathan,3 Hoekstra argues it was
a consistent feature of the theory from the beginning. Hobbes was always
both “a consent theorist and a de facto theorist of authority. If tacit and attrib-
uted consent count as consent, then he may be considered a thoroughgoing
consent theorist. If they do not, then he is, after all, a de facto theorist of
authority of a particular kind.”4 Hoekstra saves Hobbes by substituting con-
sent for agency as the core issue in the story of conquest and consent: the
identification of the victor as the active agent fades into the background in
favor of a notion of Hobbesian tacit consent.
The general thrust of Hoekstra’s interpretation is, I believe, correct:
Hobbes is both a consent theorist and a de facto theorist of authority. However,
my purpose here is to show that the key concept connecting these dimensions
of the theory is trust and not, as Hoekstra thinks, tacit consent. The Hobbesian
covenant institutionalizes a trust relationship in which both sides have a con-
sensually defined part to play—government’s part is to supply protection and
subjects’, to be obedient. Versus Hoekstra’s complicated notion of attributing
(tacit) consent to parties in situations in which it would be necessary for con-
sent to be given, trust, as Hobbes uses the idea, refers to the commonplace

---

840
Political Theory 41(6)
experience of trusting others to carry out their part in a defined relationship.
It is the case, to be sure, that Hobbes’s theory includes a concept of tacit con-
sent, which is defined in the usual way as “Signes of Contract by Inference”:
consent given “by other sufficient signes of the Will” beside express words.5
However, I see this as but a subsidiary element in the larger account of con-
sent, whereas the relationship between ruler and ruled is constituted and
defined by trust. This relationship, we will see, explains away the problem-
atic agency issue and links the conquest model of sovereignty by acquisition
to the theory of sovereignty by institution.
While unfamiliar as an interpretation of the relationship between ruler and
ruled in Hobbes’s theory, the concept of trust has figured in recent years in
game-theoretic analyses of the state of nature. Rather than representing a
prisoner’s dilemma in which the dominant strategy (in a one-shot game) is to
defect, Hobbes’s narrative, it is argued, corresponds to an assurance game.6
The two differ in that assurance games are ones in which it is rational to
cooperate when others can be trusted to do so too, which corresponds to
Hobbes’s second law of nature (i.e., “a man be willing, when others are so
too, . . . to lay down this right to all things”7). In an assurance game, “once
mutually advantageous agreements become possible, individuals can com-
plain . . . whenever others fail to uphold their part of the agreements they have
made.” Within the cooperative equilibrium, “failing to do one’s fair share
becomes reprehensible.”8 The present interpretation argues that the relation-
ships between master and servant and sovereign and subject center, in similar
fashion, on trust between the parties.
Hobbes’s thinking about trust grew out of the neoRoman account of slavery
which the English inherited via Brackton and Coke and which he adapted for
his own political and philosophical purposes.9 My interpretation will begin
here: with the justification of slavery in the sixth-century codification of Roman
law, the Digest of Justinian, and Hobbes’s adaptations of it. The root idea that
he took over and elaborated was the rationalization that slavery involves an
exchange of benefits between master and servant. The Digest explained that
“Slaves (servi) are so-called, because generals have a custom of selling their
prisoners and thereby preserving rather than killing them.” Explicitly rejected
in the Roman work was the Aristotelian view that some humans may have a
slave’s nature: “Slavery is an institution of the jus gentium, whereby someone is
against nature made subject to the ownership of another.”10
Against this background, I turn, in the second section, to the main subject:
how Hobbes transformed a neoRoman view of service into an account of
trust in which, crucially, life and liberty are seen as connected goods. The
final sections take up the political significance of the account, which played
out in Hobbes’s definition of the political covenant and, in an idiosyncratic
explanation of great immediacy in the Civil War, the nature of political

---

Baumgold
841
servitude and freedom. In recent years, Quentin Skinner and Philip Pettit
have made the last a subject of note in discussions of republican thinking
about freedom. Disagreeing with the views of both, I offer an alternative
“take” on Hobbes’s message concerning the freedom of subjects. This leads
into a concluding re-assessment of the nature of Hobbesian absolutism.
War Captivity: Hobbes’s NeoRoman Reasoning
From the start, Hobbes imported the Roman rationalization of slavery into his
political theory. Even in The Elements of Law, though, he also criticized their
failure to conceptualize alternative outcomes to war captivity:11
when a servant taken in the wars, is kept bound in natural bonds, as chains, and the
like, or in prison; there hath passed no covenant from the servant to his master; . . .
This kind of servant is that which ordinarily and without passion, is called a
SLAVE. The Romans had no such distinct name, but comprehended all under the
name of servus; whereof such as they loved and durst trust, were suffered to go at
liberty . . . the rest were kept chained, or otherwise restrained with natural
impediments to their resistance.12
The subsequent versions develop the point into a full-fledged distinction
between servants, who are allowed freedom, and slaves, who are not: “after
such Covenant made, the Vanquished is a SErVAnt, and not before: for by the
word Servant . . . is not meant a Captive, which is kept in prison, or bonds, . . .
for such men, (commonly called Slaves,) have no obligation at all.” A “servant,”
Leviathan further explains, is a captive “that being taken, hath corporall lib-
erty allowed him; and upon promise not to run away, nor to do violence to his
Master, is trusted by him.”13 Invoking the authority of the Greeks, Hobbes
inserted a digression about the distinction into a chapter on dæmonology (Part ...