Voluntary Slavery1
La esclavitud voluntaria
DANNY FREDERICK
dannyfrederick77@gmail.com
Abstract: The permissibility of actions depends upon facts about the
flourishing and separateness of persons. Persons differ from other
creatures in having the task of discovering for themselves, by
conjecture and refutation, what sort of life will fulfil them. Compulsory
slavery impermissibly prevents some persons from pursuing this task.
However, many people may conjecture that they are natural slaves.
Some of these conjectures may turn out to be correct. In consequence,
voluntary slavery, in which one person welcomes the duty to fulfil all
the commands of another, is permissible. Life-long voluntary slavery
contracts are impermissible because of human fallibility; but fixedterm slavery contracts should be legally enforceable. Each person has
the temporarily alienable moral right to direct her own life.
Keywords: Aristotle, compulsory slavery, critical rationality,
enforceable slave contracts, essential personal task, natural slave,
voluntary slavery.
Resumen: la permisibilidad de las acciones depende del hecho acerca
del florecimiento y separación de las personas. Las personas difieren de
otras criaturas por tener la tarea de descubrir por sí mismas, mediante
conjetura y refutación, qué tipo de vida les resulta plena. La esclavitud
voluntaria evita inaceptablemente que algunas personas lleven a cabo
esta tarea. Sin embargo, muchas personas pueden conjeturar que son
esclavos por naturaleza. Algunas de estas conjeturas pueden resultar
correctas. En consecuencia, la esclavitud voluntaria, en la que una
persona acepta el deber de satisfacer todos los deseos de otra, es
permisible. Los contratos de esclavitud voluntaria de por vida son
inaceptables a causa de la falibilidad humana; pero los contratos de
esclavitud por un tiempo determinado deberían ser legalmente
1
My thanks to William Glod for comments and queries on an early draft, which
helped me to improve the paper substantially.
115
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
ISSN: 2255-3827
www.lastorresdelucca.org
116
DANNY FREDERICK
reconocidos. Cada persona tiene el derecho moral temporalmente
alienable de dirigir su propia vida.
Palabras clave: Aristóteles, esclavitud voluntaria, racionalidad crítica,
contratos de esclavitud legales, tarea esencial de la persona, esclavo por
naturaleza, esclavitud.
Recibido: 16/04/2013.
Aprobado: 23/09/2013.
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
117
Introduction
I
n one important sense of the word ‘free,’ which is the sense employed
here, a free person is one who has the legal right to direct his own life.
As a consequence, he is legally entitled to use force, or to invoke the
force of the state or of an agent acting on his behalf, to prevent others
from interfering with his direction of his life. Kidnapping a person is a
clear-cut case of interfering with that person’s direction of his own life.
Reading a book of one’s own in one’s own home is a clear-cut case of not
interfering with another person’s direction of his own life. However, while
there are other clear-cut cases on each side of the boundary, the
boundary is somewhat fuzzy. For example, does playing loud music in
one’s own home interfere with one’s neighbour’s direction of his own life
(if, for example, he needs peace to study)? However, articulating the
content of the right to direct one’s own life is a job for another paper, or a
book. An intuitive grasp of the notion will suffice for our purposes here.
In contrast to a free person, a slave is a person, x, for whom there is
another person (a slave-owner), y, who has the legal right to direct the life
of x. A compulsory slave is one who has been enslaved against his will. A
voluntary slave is one who has agreed to become a slave of another
person; that is, someone, x, who has, with the consent of another, y, given
y the legal right to direct the life of x. Contemporary philosophers are
unanimous that compulsory slavery is morally impermissible. However,
while most contemporary theorists think that voluntary slavery is morally
impermissible, there are some dissenters. Robert Nozick (1974, 331) and
Judith Jarvis Thomson (1990, 283-84) opine, though without argument,
that voluntary slavery is morally permissible. Hillel Steiner (1994, 232
including footnotes 4 and 5) argues that the right to ‘self-ownership’ is,
like other rights, waivable, and it is possible that a person’s ends will best
be achieved by waiving his right to self-ownership and becoming a slave.
Steiner does not say what he thinks the relevant ends might be. Walter
Block (2003) and Stephen Kershnar (2003) defend the moral permissibility
of voluntary slavery on the ground that it can be a lesser evil in some
circumstances, for example, where a parent sells himself into slavery in
exchange for resources paid to his needy family which will enable them to
buy food or medical treatment, or where a prisoner chooses slavery in
preference to life imprisonment or execution. Such arguments are
controversial; I will not consider them here. Instead, I consider whether
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
118
DANNY FREDERICK
the possibility that there are natural slaves makes some form of voluntary
slavery morally permissible.
To save on words, henceforth ‘permissible’ always means morally
permissible. Something which is normally permissible may be
impermissible in some exceptional circumstances in which it generates
grave consequences; and something which is normally impermissible may
be permissible in some exceptional circumstances in which it is the only
way to avoid grave consequences (how grave is ‘grave’ depends upon
context). Thus, all general statements of permissibility or impermissibility
must be understood to hold ceteris paribus (see my forthcoming).
Throughout, when I speak of persons I should be understood to be talking
of normal adult humans. This is not to suggest that other humans are not
persons; it is just to avoid encumbering the discussion with profuse
qualifications.
In section I, I explain what voluntary slavery is by describing a
paradigm of it. In section II, I argue that it is possible that there are
natural slaves of at least five different types. In section III, I explain how
persons differ from other creatures and how the human predicament
grounds an alienable right of each person to direct his own life, which
makes compulsory slavery impermissible and a non-paradigm form of
voluntary slavery permissible. In section IV, I conclude.
I. Voluntary Slavery
A common confusion about slavery is exhibited by Steiner:
[S]elf-enslavement…cannot be incurred by a self-owner’s transferring (selling or
donating) that right [of self-ownership], since such transfers entail that
transferrors thereby acquire duties to their transferees, whereas slaves, as
things wholly owned by others, must lack duties as well as rights (1994, 232
footnote 4).
However, slavery is essentially a relationship between persons: a
mere thing, or even a living non-person such as a cat, cannot literally be a
slave. A person is a moral agent with the duties to respect the rights of
others that all moral agents have. A slave cannot, therefore, lack duties.
On the assumption that compulsory slavery is wrong, one might wonder
what duties a compulsory slave can have to her slave-owner; but even a
compulsory slave has duties to other people. In all societies which had an
institution of compulsory slavery, slaves were held responsible for their
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
119
actions and punished for committing crimes (Patterson 1982, 196).
Steiner’s error is to take a metaphor literally. A slave is in some ways
analogous to a mere thing that is owned; but to be analogous to a mere
thing is not to be a mere thing. If a slave, who is essentially a person, had
to be a mere thing, a non-person, then slavery would be metaphysically
impossible and there could never have been any slavery in the world. 2
Voluntary slavery is slavery by agreement. Various types of
relationship might aptly be described in that way, but it seems that the
paradigmatic type would be a legally recognised agreement along the
following lines. One free person, x, makes a legally binding contract with
another free person, y, that, from some specified future time,
(a) x has the legal duty henceforward to fulfil all the commands of y;
(b) x cannot release x from that duty;
(c) y can release x from that duty;
(d) y may legally use or hire force to compel x to fulfil the commands
of y.
From the specified future time, the person, x, ceases to have the
legal right to direct her own life; she instead becomes a slave to her slaveowner, y, whose commands x has a legal duty to obey, which gives y the
legal right to direct the life of x. It may help to clarify this arrangement if
we compare and contrast it with employment.
With regard to (a), in both voluntary slavery and a contract of
employment, one person accepts the duty to fulfil the commands of
another: the employer or the slave-owner is responsible for directing the
employee or the slave; and the employee or the slave is responsible for
executing the directions appropriately. However, in employment this holds
only during working hours, and only with respect to commands which are
both consistent with organisation policies and relevant to the job as
specified in the job description. In voluntary slavery, there are no such
restrictions: the slave has no rights against the slave-owner. However,
there are some restrictions on an employer’s control over an employee
which also apply to a slave-owner with regard to a slave. For example,
becoming an employer or an employee does not give one authority to
break the law, so it does not give the employer authority to command an
employee to break the law and it does not divest the employee of her legal
2
This sort of confusion goes back at least to Rousseau (1762, book I, ch. iv, 10) and
occurs in Ellerman’s (2010) objections to employment and Gabriel’s (2012)
objections to professional soldiers.
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
120
DANNY FREDERICK
duty to obey the law. The same applies to becoming a slave-owner or a
slave. Any command to break the law is legally void, and the slave has a
legal duty not to fulfil it.
In connection with (b), although an employee accepts the duty to
fulfil the commands of her employer, she can release herself from this
duty at any time by leaving the job. The employment contract normally
specifies a notice period; but the employee may refuse to work her notice,
though she may then be liable to pay compensation to her ex-employer.
The slave, in contrast, cannot release herself from her duty to fulfil the
commands of her slave-owner; otherwise it would simply not be slavery.
In connection with (c), while an employer must give an employee notice of
termination, and may also have to satisfy some other legal requirements,
before she can terminate an employee’s employment, a slave-owner can
release the slave from her slavery immediately. With respect to (d), an
employer may not legally use force to compel an employee to perform her
duties. The employment contract is enforceable against the employee only
in the sense that, if the employee defaults on her duties, the employer
may terminate the relationship and may in some circumstances be able to
exact financial compensation for damage or loss. In contrast, a slaveowner is legally entitled to use or hire force to compel the slave to fulfil the
slave-owner’s commands, and to restrain or recapture a slave who tries to
get free of the relationship. In paradigmatic slavery there are no limits to
the severity of the force that a slave-owner may employ against her slave;
in consequence, the slave-owner is legally entitled to kill the slave.
A legally binding contract normally requires consideration on both
sides. To meet this condition, the prospective slave-owner must make
some kind of payment for the services to be received from the prospective
slave. A contract for slavery is made at a time when the two parties to it
are free persons who thereby acquire contractual obligations to each
other; but the contract stipulates a later time at which one of the parties
ceases to be free and becomes a slave, and the other party ceases to have
any obligations, contractual or otherwise, to the slave. Consequently, in a
slave contract, the contractual duties of the party who becomes the slaveowner must be completed before the other party becomes a slave. If
payment for the slave is made to the prospective slave, the latter must
consume or transfer this payment before the slavery begins, otherwise the
slave-owner could command the slave to hand it back; if payment is made
to a third-party beneficiary, the payment must be made before the slavery
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
121
begins, otherwise the slave-owner’s obligation to the slave to make the
payment would lapse.
A legally binding agreement normally requires the absence of
coercion. It might be suggested that only coercion or coercive
circumstances could explain a person’s consent to slavery. I argue against
that suggestion in the next section.
II. Natural Slavery
Aristotle claimed that some humans are natural slaves while others are
natural slave-owners (2007, I, iv, v), and that the latter are entitled to
enslave the former against their will (2007, I, xiii). Aristotle thought that
this difference in moral entitlement depends upon a difference in
psychological capacities: ‘he who participates in rational principle enough
to apprehend, but not to have, such a principle, is a slave by nature’
(2007, I, v). It appears that what Aristotle meant by this was that, while
the natural slave is capable of practical reasoning, he is not capable of
competent moral reasoning; so, left to himself, he will often act in a way
that even he later sees to be amiss (Heath 2008, 244-53). Thus, although
he is enslaved for the slave-owner’s benefit, a natural slave is better off
being enslaved (2007, I, ii, iv, v, xiii; 2006, VIII, x). Aristotle thought that
non-Greeks, the majority of humans, are natural slaves (2007, I, ii, vi; III,
xiv).
Aristotle’s theory of natural slavery is defective. First, there seems to
be no pronounced contrast between Greeks and non-Greeks with respect
to the capacity for competent moral reasoning. Second, it does not follow
from the fact that a person is incapable of competent moral reasoning
that the person would be better off as a slave. A person might live a
better, more fulfilling, life by directing his life less than competently than
he would live by being a permanent drudge on twenty-four-hour call to
another person. Still, Aristotle may have been right that there are some
people who are incapable of living well if left to decide things for
themselves; though the reasons may be unconnected with a deficiency in
reasoning capacity. Let us say that a natural slave is a person who is such
that the most fulfilling and worthwhile life that he can lead is one in which
he is a slave to another person. I do not know whether there are any
natural slaves; but it does seem possible that there are. I show this by
describing five types of person who would find fulfilment in being a slave.
For each type I point to familiar kinds of actual people who exhibit
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
122
DANNY FREDERICK
tendencies toward that type. These tendencies are stronger in some people
than they are in others. Indeed, it seems possible that there are some
people in whom the tendencies are so pronounced and pervasive that they
would find fulfilment in being a slave. The following fivefold categorisation
of natural slaves is neither exclusive nor exhaustive.
The dutiful slave derives fulfilment from being under another
person's total control. His satisfaction comes from being at the service of
another. He wants to take on the duty to serve another completely and to
discharge that duty in an exemplary manner. He takes pride in his service
and even in his subservience; indeed, his self-respect depends upon doing
these things well. He is not incapable of directing his own life; but a life of
subservience is more valuable to him. Indeed, he might even supervise
others on his slave-owner’s behalf, in which case even these exercises of
dominance will be suffused by his submission to his owner. A tendency to
dutiful slavery often finds expression in religion. St. Paul says that
Christians are slaves to obedience (Romans 6:16) and slaves to God
(Romans 6:22). That is a metaphor; but it seems that many religious
people have wanted to take it literally. If natural slaves exist, it seems
likely that some of them will be found in monasteries, nunneries and
cults. Similarly, though perhaps to a lesser extent, the tendency to dutiful
slavery seems to be exhibited by some who join the armed forces or
domestic service. A fictional example, which will be true to life in some
respects, is Stevens the butler, in The Remains of the Day, whose
fulfilment comes from service to a gentleman of moral worth. He is in
charge of seventeen staff and carries out his duties stoically whatever the
circumstances, striving for a ‘dignity in keeping with his position’ which is
worthy of the admiration even of the gentry (Ishiguro 1989).
The weak-willed slave yearns to be a dutiful slave who fulfils every
command of his superior; but he knows that in practice he will often
default on his duties because of his lackadaisical nature. The most
worthwhile life he can lead is as a slave to a slave-owner who will not only
command but also exhort and punish, to goad him into doing his duty, or
as much of his duty as he can do. He knows that if he has the option of
terminating his servitude, he will do so when the going gets tough and
thereby forsake his fulfilment, so it is essential for him that he is a slave
rather than just a servant. If dutiful slaves are possible then so are weakwilled slaves. They are analogous to someone who submits to a
detoxification programme during which his pleas for drugs, alcohol or
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
123
tobacco will be ignored and his attempts to escape will be frustrated. The
weak-willed slave’s self-respect is fragile, not because he lives as a slave,
but because he often falls short of his slave duties and disappoints his
slave-owner.
The refractory slave, unlike the previous two types, avows no duty to
obey his slave-owner. When he complies with his owner’s instructions, he
does so to avoid punishment or for some ulterior motive. Although the
slave-owner has the legal right to control him, the slave denies this. He
acknowledges that he is kept in slavery but he deems this situation
unjust and he recurrently rebels against the slave-owner. There is an
element of fiction here, if he is a voluntary slave, because he has given the
slave-owner the right to control him. But the fiction is necessary for him
because he finds his fulfilment in rebellion and his self-respect comes
from resisting an oppressor (just as his owner may find his fulfilment in
subduing refractory slaves or, at least, in attempting to subdue them, for
the chase may be better than the catch). For slaves of this type, if
oppressors do not exist, they have to invent them (to adapt Voltaire). The
tendency to this type of slavery is commonly exhibited in political life,
where the scene is littered with people striking revolutionary poses
against all manner of more or less imaginary oppressors (evil conspiracies
of various kinds) or looking for a reason to martyr themselves.
The anxious slave may be competent to make good decisions for
himself but he thinks, correctly, that the disadvantages of the anxiety he
suffers in making choices outweigh the advantages of making good
choices. It therefore makes sense for him to hand over the decisionmaking to someone else, even though the choices will be poorer, perhaps
even much poorer, simply in order to be disburdened of the anxiety that
he finds unendurable. He will want a reasonably virtuous slave-owner, to
avoid having to cope with the dilemma of a command to break the law. He
suffers from ‘the strain of civilisation’ (Karl Popper’s phrase). He just
wants to be told what to do, so that he does not have to think for himself.
The tendency to this form of slavery is evident in the widespread
attraction to tribalism (Popper 1945, chapter 10).
The masochistic slave finds his fulfilment in pain, humiliation or
other torment. He could be a slave of any of the preceding types, but he
needs a slave-owner who is sadistic. Some slaves of this type might find
self-respect in bearing the suffering they endure. But for others a loss of
self-respect may be an important part of the suffering. Indeed, some may
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
124
DANNY FREDERICK
accept the duty to obey the slave-owner because they recognise that it
demeans them. Some refractory slaves, who avow no duty to obey the
slave-owner, may derive satisfaction from the humiliation they feel in
submitting to the slave-owner’s demands. The tendency to this form of
slavery is commonplace and is exhibited in sadomasochistic practices,
lifestyles and media.
Thus, tendencies to each of these types of natural slavery appear to
be common. I conjecture that a tendency to natural slavery exists in each
of us to some degree, and that evolutionary biology may be able to provide
an explanation for it. Indeed, the elements of such an explanation may
already exist, since Christopher Boehm (1999) describes our natural
proclivity for dominance hierarchies, and Joseph Henrich and Francisco
Gil-White (2001) offer an account of the evolution of voluntary deference.
Since the tendency toward one or another kind of natural slavery is
stronger in some people than in others, we cannot rule out the possibility
that there will be people in whom the tendency is so strong as to make
them natural slaves, people whose vocation was expressed by Bersyenev:
‘I feel that one’s whole destiny in life should be to make oneself number
two’ (Turgenev 1859, 30).
The possibility of natural slave-owners seems easier to explain. Let
us say that a natural slave-owner is a person whose flourishing would be
enhanced if he were a slave-owner. We all have chores which a slave could
do for us, thereby permitting us to spend much more of our time and
energy on more satisfying activities. So long as we would not find directing
a natural slave so irksome that it outweighed those benefits, our
flourishing would be enhanced by being a slave-owner. Further, many
people find managing others intrinsically fulfilling, so for some of those
people directing the life of a slave could be a substantial part of their
fulfilment. It seems not only possible but highly likely that there are
natural slave-owners, as well as people with tendencies to be such. Many
marriages and friendships appear to resemble, to a greater or lesser
extent, one or more of the types of slavery outlined above.
This account of natural slavery assumes that some activities or
forms of life are objectively valuable for some people; but a value
subjectivist or anti-realist might reformulate it in terms of preferences.
III. Permissible Slavery
In the case of an animal which is not a person, the best life that it can
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
125
lead is determined by its biology; and the animal normally tries to live a
life of that kind by acting in accord with its biological instincts and the
culture, if any, that it acquires from its local conspecifics. In contrast, a
person is a creature who has language and self-consciousness. That
enables a person to formulate in words her beliefs, thoughts, desires,
expectations, emotions and such like. Once a person objectifies the
contents of her subjective states in this way, she can distance herself
from them, view them sceptically, consider alternatives, and formulate
critical arguments and tests to decide between alternatives (Popper and
Eccles 1977, 57-59, 108-12, 144-46). This capacity for critical rationality
enables a person not only to evaluate ways of living exemplified by
existing persons (including herself), but also to discover new possibilities.
The sort of life that a human person will find fulfilling will be related to
that human’s biology, but the exercise of critical rationality can make that
relation tortuously indirect. For example, people today can live sorts of
lives that would have been inconceivable or thought physically impossible
a few centuries ago, such as an astronaut, a transsexual model, a genetic
engineer, or a web-site designer. Some people find their fulfilment in living
lives of these new kinds, while others find fulfilment in living more
traditional types of lives. Therefore, unlike those animals which are not
persons, a person is faced with the question: what sort of life will be most
fulfilling for me?
There is an abundance of material which is relevant to answering
that question. There are studies by psychologists, anthropologists and
other social scientists concerning different ways of life. There are
biographies, autobiographies, novels and dramas, as well as lifestyle
discussions throughout the popular media, containing accounts of how
different people have fared in different kinds of life. However, what suits
some people does not suit others, so the person also needs to know about
herself. She may be able to learn about herself from family, friends,
teachers and other acquaintances, since other people sometimes know a
person better than she knows herself, in at least some respects. However,
while such research will enable the person to form some conjectures
about the sort of life she will find fulfilling, those conjectures need to be
tested. Even if another person’s knowledge of the sort of life that will fulfil
a specific person is better than the knowledge that the person has herself,
the other person’s knowledge is still conjectural and needs to be put to
the test. The crucial test of whether a type of life will fulfil a person is that
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
126
DANNY FREDERICK
person’s own experience of living that type of life. This is especially so if
the kind of life she conjectures will fulfil her is one that no one has lived
before. Therefore, in order to answer the question of how to live, a person
has to form a hypothesis about what type of life will fulfil her, and then
test that hypothesis by living that type of life, or at least some
approximation to it. If she finds that the life she has chosen does not fulfil
her, her hypothesis is refuted. If she is to find an answer to her question,
her next steps must be to try to learn from her mistake, think up another
theory about who she is and then, insofar as she can extricate herself
from the circumstances of her current life, set out to test that new theory.
The essential personal task of each person is to discover for herself,
through conjecture and refutation, what sort of life will constitute her
fulfilment. (This is not to suggest that every person apprehends her
essential personal task and attempts to pursue it: for example, some
people just follow convention or drift from one lifestyle to another under
the influence of circumstances.)
A natural slave is a person whose most fulfilling and worthwhile life
is one in which she is a slave to another person. If there are any natural
slaves, what is good for them is slavery; it is through slavery that they will
realise themselves and flourish. It is possible that there are natural
slaves. Further, tendencies to natural slavery are endemic; so, many
people may conjecture that they are natural slaves. None of these people
can discover whether they are natural slaves except by experimenting
with slavery, or some approximation to it. If they are to accomplish their
essential personal task, they must have the opportunity for such
experiments. If they thereby discover that they are natural slaves, their
fulfilment depends upon them choosing slavery for themselves. Therefore,
voluntary slavery is permissible. That is a sketch of the argument that will
be developed in the rest of this section.
Few, if any, theorists would deny that morality is centrally
concerned with the flourishing of persons. Given the nature of the
essential personal task, human flourishing can be achieved only if people
are able to try out ‘experiments in living’ (a term from John Stuart Mill,
1859), subject to the proviso that such experiments do not undermine the
flourishing of others. The interconnectedness of people in society means
that almost every such experiment by one person will have some adverse
impact on some other people, so it is only where its adverse impact is
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
127
substantial that a specific experiment is impermissible. 3 For instance, a
man may conjecture that he will find his fulfilment in driving on busy
roads while he is drunk; but it is impermissible for him to test that
conjecture by living it because the threat to the flourishing of others is too
great. However, there is a limit to the permissible trade-offs between the
flourishing of one person and that of others. Most theorists insist that
morality is centrally concerned with respect for the separateness of
persons (see Zwolinski 2008 for useful discussion): a standard objection
to act-utilitarianism and some other consequentialist theories is that they
fail to take account of this separateness. We can illustrate the point with
one of Thomson’s examples (1990, 135). An expert surgeon has five
patients who will die if they do not get new parts: two need one lung each,
two need one kidney each, and one needs a heart. A healthy young man
who is visiting the hospital has the right blood type and can be cut up to
supply the parts for the five patients. The surgeon asks the young man
whether he would like to volunteer his parts; but the young man declines.
If the surgeon nevertheless kills the young man and cuts him up, this
will, let us assume, enable five people to flourish at the cost of the
flourishing of only one person. However, despite the net gain for human
flourishing from curtailing the young man’s experiment in living, the
surgeon’s action is impermissible because it uses the young man simply
as a means to the flourishing of others.
We must pause a moment to consider what it means to use someone
simply as a means. Kant (1785, 4: 429-30) says that it is permissible to
use a person as a means where she shares the end of the action. This
suggests that she must consent to be used as a means with a view to
some potential benefit for herself. For example, when I hire a plumber I
use her as a means to the end of getting my leaking pipe fixed; but I do
not use her simply as a means because she agrees to fix my pipe in return
for a payment. It is not necessary that she actually derives some benefit
from the transaction: the trial-and-error nature of our search for
fulfilment means that we will often make mistakes about which actions
are beneficial. However, it does seem that conjectured potential benefit is
necessary. For example, suppose the young man in the transplant
example had consented to be cut up for the sake of the other five people.
Would that have made it permissible for the surgeon to proceed? Some of
3
The notion of a substantial adverse impact needs to be spelt out carefully; but we
do not need to address that issue here.
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
128
DANNY FREDERICK
us may think not, because we cannot see what is in it for the young man.
However, if we could see how, given the facts about his life, it would
enhance his flourishing to make that sacrifice, we may think that his
consent then makes it permissible for the surgeon to proceed. It seems,
though, that it must be possible that the conjectured potential benefit will
arise. For example, if the young man agrees to the transplants because he
mistakenly believes that, as a reward for his altruism, his body will grow
replacement organs that will be better than the original ones, then it
would be impermissible for the surgeon to proceed.
It also seems that consent is necessary. For example, even if I am
convinced that the plumber’s flourishing will be enhanced if she fixes my
pipe for a payment, it is not permissible for me to compel her to fix my
pipe, even if I pay her handsomely, and even if it turns out in fact that the
experience and the payment are genuinely beneficial for her. How can we
explain this? Here is my attempt. Let us consider two arbitrarily chosen
people whom we can label ‘Alf’ and ‘Betty.’ No one can know for sure what
will benefit another person. So even where Betty knows better than Alf
what is good for Alf, it may still turn out that both Betty and Alf are
mistaken and that Alf learns more from his mistaken experiment than
Betty could ever teach him. Further, it is not the case that everyone who
thinks she knows better actually does know better. Indeed, in the sorts of
cases we are considering, in which Betty wants to use Alf as a means to
the flourishing of Betty, it will be almost inevitable that self-interest will
give Betty a jaundiced view of what will be good for Alf. As a consequence,
if Betty were to use Alf as a means whenever Betty thought that Alf would
benefit, Alf would often have his flourishing undermined. It must be Alf
who conjectures that there is a potential benefit for him, and also that the
potential benefit is greater than that obtainable from alternative courses
of action open to him. The test of that is that he consents. It is therefore
impermissible to use a person as a means to the flourishing of others
without that person’s consent.
We can summarise the preceding discussion in the following (not
very precise) principles:
(i) it is permissible for a person to try out whatever kind of life she
conjectures may fulfil her, if such experimentation on her part is not
impermissible according to (ii) or (iii);
(ii) it is impermissible to use a person as a means to the flourishing
of some others, unless
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
129
the person agrees to be so used because she conjectures
that there is a potential benefit for her, and
•
it is possible that a person should obtain the conjectured
benefit from being so used;
(iii) it is impermissible for a person to experiment with a kind of life
she conjectures may fulfil her, if doing so has a substantial adverse
impact on the flourishing of others.
There seems to be general, though not universal, agreement that if
we could save, say, a million or a billion people by killing just one, that
would be an increase in overall flourishing sufficient to make it
permissible to use the one as a means to the flourishing of the others
without the one’s consent. That is consistent with (ii) because the implicit
ceteris-paribus clause in all general statements of permissibility and
impermissibility allows for exceptions (though it need not commit us to
accepting this particular type of exception, since there is room for debate
about exactly when other things are not equal). It does not follow from (i)
that any kind of experimentation is permissible if it is not impermissible
according to (ii) and (iii), because (i) concerns only experimentation
conducted by a person to test conjectures about what kind of life will fulfil
her. It does follow immediately from (ii) that compulsory slavery is
impermissible, because the slave-owner uses the slave as a means to the
flourishing of the slave-owner without the slave’s consent. I now show
that principles (i) - (iii) make one kind of voluntary slavery permissible.
Slavery is a relationship between two people; so, for voluntary
slavery to be permissible, it must not only be permissible for a person to
be a slave, it must also be permissible for a person to be a slave-owner
(Vallentyne 2000, 3-4). Compare: it is permissible for Jane to marry, but
there are no eligible bachelors. We have seen that there are widespread
tendencies to slavery and to slave-ownership. Consequently, many people
may conjecture that they are natural slaves and many may conjecture
that they are natural slave-owners. From (i) it follows that it is permissible
for those people to try out slavery or slave-ownership, respectively, if such
experimentation on their part is not impermissible according to (ii) or (iii).
If two such people agree to experiment with a relationship in which one is
slave-owner and the other slave, each in order to further her own
flourishing, the possibility that there are natural slaves and natural slaveowners means that such an experiment in voluntary slavery is not
impermissible according to (ii). Each uses the other as a means to
•
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
130
DANNY FREDERICK
discovering what sort of life will fulfil her, but neither uses the other
simply as a means. It has been argued that voluntary slavery would
generate costs or harms for people outside of the slavery relationship,
since the costs of monitoring or enforcing slavery agreements would be
borne by the public, and third parties may suffer distress, or become
insensitive to suffering, if they are aware of the existence of voluntary
slaves (Feinberg 1986, 75-81). However, the costs of monitoring or
enforcing slavery contracts can be charged to the people who participate
in them (see below). Further, interracial affection may generate distress in
some third parties, and the promotion of political ideas like nationalism
may reduce some people’s sensitivity to the suffering of others, but such
third-party effects are not considered substantial enough to make those
activities impermissible. Therefore, (iii) does not render experiments in
voluntary slavery impermissible (see Archard 1990, 456, and Kershnar
2003, part II, section B, for further discussion). Consequently, it is
permissible for some people to experiment with voluntary slavery and
voluntary slave-ownership.
The conclusion that it is permissible for some people to experiment
with voluntary slavery falls short of the conclusion that voluntary slavery
is permissible, because experiments typically come to an end. Someone
who has experimented with voluntary slavery, or something
approximating it, and has found the life fulfilling, might conclude that she
is a natural slave. However, it is not possible for her to know for certain
that she is a natural slave. The fact that she has found a life of slavery, or
something close to it, fulfilling so far, does not preclude her from
discovering in future that it ceases to fulfil her, either because there is
only so much of such a life that she can enjoy, or because she changes
over time. Someone who finds slavery fulfilling at a younger age may find
it unbearable at a later age. However, voluntary slavery in its
paradigmatic form does not permit the slave to terminate her relationship
with her slave-owner. She may appeal to her slave-owner to release her,
but the slave-owner may refuse. She may then find herself trapped in an
unfulfilling, probably intolerable, form of life for the rest of her days.
Consequently, if paradigmatic voluntary slavery is permitted, many people
(perhaps all) who enter voluntary slavery may eventually discover that
they have made a terrible mistake which they are unable to correct. Their
continued slavery enables their slave-owners to continue to experiment
with a kind of life that they conjecture may fulfil them, but only by
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
131
prevented the slaves from finding fulfilment. That is impermissible
according to (iii). Therefore, voluntary slavery in its paradigmatic form is
impermissible.
It may seem that the trouble with paradigmatic voluntary slavery is
its condition (b), according to which the slave cannot release herself from
the duty to follow the commands of her slave-owner. However, if the slave
can release herself from that duty, the slave-owner cannot enforce it, so if
condition (b) goes, so does condition (d), which means that the slaveowner can do only that to which the slave willingly submits. Such a
relationship would not count as slavery at all; nor would it fulfil a natural
slave. Even a dutiful slave who never wants to terminate the relationship
and who willingly submits to all the slave-owner’s demands and abuses
may find her slavery ersatz if she knows that she can pull out of it at any
time and that her owner is constrained by that consideration. The weakwilled slave and the refractory slave would be unable to realise
themselves, for each needs an arrangement in which she is forcibly
restrained from ending her slavery and compelled to do things she is
disinclined to do. The anxious slave, somewhat similarly, will find it a
great relief to be spared any agonising over whether to terminate the
slavery. The masochistic slave may be denied her peculiar satisfaction
because the slave-owner is reluctant to punish her severely for fear that
she might end the relationship. Indeed, all types of slave may be denied
the more extreme forms of treatment, which they may need. Therefore, a
relationship without condition (b) would not be fulfilling for natural
slaves; it would be just play-acting at slavery.
However, there is a non-paradigmatic form of voluntary slavery the
permissibility of which is consistent with (ii) and (iii) and which also has
the following benefits:
•
it allows people the opportunity to experiment with
something approximating paradigmatic slavery,
•
it permits people who are convinced that they are natural
slaves to live their lives in a state that approximates
paradigmatic slavery.
We obtain this form of non-paradigmatic slavery if we amend
condition (a) of paradigmatic voluntary slavery by replacing ‘henceforward’
with ‘until the end of the contract period.’ For, a fixed-term contract will
enable people to experiment with a good approximation to slavery for a
limited time, so that they can test whether they are natural slaves; and it
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
132
DANNY FREDERICK
will enable people who are convinced that they are natural slaves to renew
the contract as soon as it expires. The term of the contract can be agreed
by the parties, up to a period of, say, five years. Such voluntary slavery is
non-paradigmatic for two reasons. First, although the contract may be
terminated at any time by the slave-owner, the contract terminates at the
end of the fixed term independently of the slave-owner’s wishes. Second,
this fact is likely to inhibit somewhat the slave-owner who wants the
relationship to continue beyond the end of the fixed term: she may be
wary of being too severe in case the slave does not renew the contract. So
the time-limited nature of the contract makes it significantly weaker than
paradigmatic voluntary slavery.
It is, though, the time-limited nature of the contract that makes it
permissible. Fixed-term voluntary slavery is consistent with (ii) because
slave-owner and slave have agreed to the contract, each in order to
further her essential personal task; and it is possible that there are people
who are fulfilled by slave-ownership and slavery, respectively. It is also
consistent with (iii), because a slave in such a relationship who discovers
that she is not a natural slave will be able to resume her experimentation
with other kinds of life at the end of the contract period, or sooner if her
slave-owner releases her. Of course, she might not be able to resume her
experimentation as soon as she concludes that she is not a natural slave.
In such cases, there will be some adverse impact on her flourishing
because she will be preventing from pursuing her essential personal task
for a time. However, it will be only for a relatively short period; and this
drawback is unavoidable if people are permitted to experiment with
slavery, which is in turn necessary if natural slaves are to live fulfilling
lives and if people who mistakenly think that they may be natural slaves
are to pursue their search for fulfilment and obtain information that
improves their chances of finding fulfilment in the future.
In paradigmatic voluntary slavery, the slave-owner is entitled to
acquire the slave’s property and to kill the slave. The former entitlement
will attract people looking for a ‘licence to rob,’ who want to take on a
slave, acquire her property and then immediately release her from slavery;
while the second entitlement will attract people, such as serial killers,
who see it as a ‘licence to kill.’ In both sorts of case, the slave-owner
would be using the slave simply as a means to enhance her own
flourishing. This is impermissible according to (ii): the prospective slave
has not agreed to be used as the prospective slave-owner intends to use
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
133
her. The prospective slave-owner is attempting a kind of fraud and is not
a permissible slave-owner. Thus, fixed-term voluntary slavery will be
permissible only if such people are excluded from being slave-owners.
This can be achieved by making two further departures from paradigmatic
voluntary slavery. First, in permissible fixed-term slave contracts, the
slave-owner will have a duty not to kill the slave. Since the slave-owner
can have no duties to the slave, the duty not to kill the slave must be a
matter of the criminal law, and it will permit the exceptions that are
generally recognised in that law, such as self-defence, and perhaps
assisted suicide. Second, the slave will be required to give away her
property before the start of the contract term. Although each of these two
conditions moves fixed-term slavery further away from paradigmatic
voluntary slavery, it seems that neither makes an essential difference to
the relationship. After all, a genuine slave-owner wants a live slave. Even
if the slave becomes unduly troublesome the slave-owner has the option
of terminating the contract, and few people would choose to kill a person
when they have an effective alternative means of being rid of her.
Similarly, a genuine slave-owner wants to acquire a slave rather than her
property.
In fixed-term voluntary slavery, recall, two people agree a legally
binding contract in which one person accepts the enforceable duty to live
as directed by the other. To say that it is permissible for there to be such
legally binding contracts is to say that people ought to be permitted to
enter such contracts and enforce them legally, which is to say that such
contracts ought to be legally enforceable. This does not exclude the
imposition of some paternalistic conditions to be met prior to entry into
such a contract. For example, a prospective slave could be required to
receive a critical challenge, provided by a person trained for the purpose,
the point of which would be to ensure that she has considered
alternatives, is familiar with the risks and drawbacks and has strategies
for dealing with them. To avoid any adverse impact on third parties, the
costs of the critical challenge would be charged to the prospective slave.
Further, legal provisions would be needed for a range of special
circumstances, including children born to slaves; but the various options
for dealing with such circumstances cannot be discussed here.
We can summarise (i) - (iii), as follows:
(iv) each person has the moral right to experiment with any kind of
life that may possibly fulfil her.
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
134
DANNY FREDERICK
This right involves the liberty of the person to experiment, expressed
in (i), and the duty of others not to prevent such experimentation by using
a person simply as a means to promote the flourishing of others,
expressed in (ii). Like all rights, it is subject to limitation by the
flourishing of others, as reflected in (iii). Given that one of the permissible
ways in which a person may experiment is with fixed-term voluntaryslavery contracts, another way of expressing (iv) is:
(v) each person has the temporarily alienable moral right to direct
her own life.
That is to say, each person has the moral right to direct her own life,
unless she alienates that right; but she can alienate that right only for a
fixed term. The fact that a person has the moral right to direct her own
life unless she voluntarily gives up that right means that compulsory
slavery is impermissible. The fact that she can give up that right means
that voluntary slavery is permissible. The fact that she can give up that
right only for a fixed term means that only voluntary slavery of a nonparadigmatic form is permissible. The permissible form of nonparadigmatic voluntary slavery is that which is for a fixed period, in which
the slave-owner cannot kill the slave, and in which the slave gives away
all her property before the start of the contract term. These moral rights,
impermissibilities and permissibilities are grounded in the essential
personal task of self-discovery, the possibility of natural slaves and of
natural slave-owners, and the trial-and-error process by which we acquire
our always fallible knowledge.
IV. Conclusion
A free person is one who has the legal right to direct his own life.
Slavery is a relationship between persons in which one person has the
legal right to direct the life of another. In compulsory slavery, a person is
enslaved against his will. In voluntary slavery, a free person contracts
with another free person that from a specified future time the first person
will have the legal duty to fulfil all the legitimate commands of the second.
The first person thereby ceases to be free and becomes a slave; the second
becomes a slave-owner with the legal right to direct the life of the first. In
paradigmatic voluntary slavery, the slave has the duty to obey the slaveowner until the slave-owner releases him from that duty; the slave-owner
is also entitled to kill the slave and to take possession of the slave’s
property.
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
135
Facts about permissibility and impermissibility depend upon facts
about human flourishing and the separateness of persons. Each person
has the essential personal task of discovering for himself, through trial
and error, which kind of life will fulfil him; though the knowledge thus
gained is always fallible. This undergirds the right of each person to
experiment with any kind of life that he conjectures may fulfil him, so
long as it is possible for a person to be fulfilled in that way and so long as
such experimentation does not have substantial adverse impact on other
people’s flourishing or involve using anyone simply as a means. This
makes compulsory slavery and paradigmatic voluntary slavery
impermissible. However, it is possible that there are natural slaves of (at
least) five different types, for whom the most worthwhile life is that lived
as a slave to another person; and there are permissible slave-owners. It is
therefore permissible for people who conjecture that they may be natural
slaves to experiment with a form of non-paradigmatic voluntary slavery,
provided they can find a permissible slave-owner. In this form of voluntary
slavery, the contract is for a fixed period, the slave-owner has a legal duty
not to kill the slave and, before the start of the contract, the slave has to
give away all his property. People who are convinced that they are natural
slaves will be able to renew their contracts at the end of the fixed term
and continue such renewals until such time as they change their mind (if
they ever do). Thus, each person has the temporarily alienable moral right
to direct his own life; that is, fixed-term voluntary slavery contracts ought
to be legally enforceable. If fixed-term voluntary slavery were thus legally
recognised, we might be surprised at how many people experiment with
slavery as part of their process of self-discovery, and even at how many
appear to be natural slaves.
V. Bibliography
Archard, David (1990), “Freedom not to be Free: the Case of the Slavery
Contract in J. S. Mill's On Liberty,” Philosophical Quarterly, 40 (161):
453-65.
Aristotle (2006), Nicomachean Ethics, translated by W. D. Ross, Adelaide,
eBooks@Adelaide. Downloaded on 30 September 2011 from:
http://ebooks.adelaide.edu.au/a/aristotle/nicomachean/
——— (2007), Politics, translated by Benjamin Jowett, Adelaide,
eBooks@Adelaide. Downloaded on 30 September 2011 from:
http://ebooks.adelaide.edu.au/a/aristotle/a8po/
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
136
DANNY FREDERICK
Block, Walter (2003), “Toward a Libertarian Theory of Inalienability,”
Journal of Libertarian Studies, 17 (2): 39-85.
Boehm, Christopher (1999), Hierarchy in the Forest, Cambridge, MA,
Harvard University Press.
Ellerman, David (2010), “Inalienable Rights: a Litmus Test for Liberal
Theories of Justice,” Law and Philosophy, 29 (5): 571–99.
Feinberg, Joel (1986), Harm to Self, Oxford, Oxford University Press.
Frederick, Danny (forthcoming), “Pro-tanto Obligations and Ceterisparibus
Rules,”
Journal
of
Moral
Philosophy,
http://booksandjournals.brillonline.com/content/10.1163/174552
43-4681036;jsessionid=9gsnrqr3a20a.x-brill-live-01.
Gabriel, Jazmine (2012) “Forced to be Free, Not Free to be Slaves,”
Journal of Value Inquiry, 46 (1): 39-50.
Heath, Malcolm (2008), “Aristotle on Natural Slavery,” Phronesis, 53 (3):
243-70.
Henrich, Joseph and Gil-White, Francisco (2001), “The Evolution of
Prestige: Freely Conferred Deference as a Mechanism for Enhancing
the Benefits of Cultural Transmission,” Evolution and Human
Behaviour, 22 (3): 165-96.
Ishiguro, Kazuo (1989), The Remains of the Day, London, Faber & Faber.
Kant, Immanuel (1785), Groundwork of the Metaphysic of Morals, tr. H. J.
Paton in his The Moral Law (53-123), London, Hutchinson (1948).
Kershnar, Stephen (2003), “A Liberal Argument for Slavery,” Journal of
Social Philosophy, 34 (4): 510-36.
Mill, John Stuart (1859), On Liberty, Harmondsworth, Penguin (1983).
Nozick, Robert (1974), Anarchy State and Utopia, Oxford, Blackwell
(1980).
Patterson, Orlando (1982), Slavery and Social Death, Cambridge, MA,
Harvard University Press.
Popper, Karl (1945), The Open Society and its Enemies, fifth edition,
London, Routledge and Kegan Paul (1966).
Popper, Karl and Eccles, John (1977), The Self and its Brain, London,
Routledge and Kegan Paul (1983).
Rousseau, Jean-Jacques (1762), The Social Contract, in The Social
Contract and Discourses, trans. G. D. H. Cole, London, Dent & Sons
(1913).
Steiner, Hillel (1994), An Essay on Rights, Oxford, Blackwell.
Thomson, Judith Jarvis (1990), The Realm of Rights, Cambridge, MA,
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137
Voluntary Slavery
137
Harvard University Press.
Turgenev, Ivan (1859), On the Eve. Harmondsworth, Penguin (1950).
Vallentyne, Peter (2000), “Left Libertarianism: A Primer,” in Left
Libertarianism and its Critics, ed. Peter Vallentyne and Hillel Steiner
(1-20), Basingstoke, Palgrave.
Zwolinski, Matt (2008), “The Separateness of Persons and Liberal Theory,”
Journal of Value Inquiry, 42 (2): 147-65.
Las Torres de Lucca
Nº 4 (enero-junio 2014): 115-137