Ethical Sex
Ethical Sex
Sexual Choices and their
Nature and Meaning
Anthony McCarthy
Fidelity Press
South Bend, IN
2016
Published by Fidelity Press in South Bend, Indiana
www.culturewars.com
Copyright © 2016 by Anthony McCarthy
First Edition
All rights reserved. No part of this book may be reproduced, stored in a
retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without the prior permission of Fidelity Press.
Printed and bound in the United States of America
by St. Martin de Porres Dominican Community
Cataloguing in Publication Data is on file with the Library of Congress
Anthony McCarthy
Ethical Sex: Sexual Choices and their Nature and Meaning
p. cm.
Includes bibliographical references and index
ISBN 0-9289891-17-0
Table of Contents
Preface
Acknowledgements
1
7
Introduction
10
Thick or thin?; Specialness of sex; Rape; The rape of
Lucretia; Rape, shame, guilt; Sex and phenomenology.
1.
Contraception as Contralife
34
Basic goods; Incommensurability; The contraceptive
act; Necessarily contralife; Possible persons; Masek’s defence; Rape case; Conclusion.
2.
Natural Law, Functions, Teleology
66
Natural law and practical rationality; Naturalists;
Anti-Naturalists; Teleology; Functions, flourishing and
the good; The good; Applications; Unity, goods and enjoyment; Reproductive organs; Clarifying perverse use of
faculties; Sterilisation; Conclusion.
101
3.
Marriage and Meaning
Meaning as social construct; Species of act; Marriage
as standard; Sacramental marriage; Natural marriage;
Role of marriage; Quasi-marital arrangements; Designing marriage; Substitutionary sex; Privileged acts; Counterarguments; Role of sex in marriage; Substitutionary sex
and society; Conclusion.
4.
Marital Willing
127
Conditional and preparatory intentions and the good of
marriage; Conclusion.
5.
Sexual Desire
141
Setting the heart; Bad desires; Dynamic of lust; Felt desire,
informed desire – desire, choice and cognition; Willed and unwilled desire; Intentions and evaluating prospective acts; Hate
crimes: Bad motivating desires; Fantasy; Teleology of desire;
Plain sex; Functionality again; Conclusion.
6.
Love, Virtue and Vice
169
Love; Affirmation of beloved’s existence; Unconditional
love; Three aspects of love; Procreative union; One flesh: instrumental to procreation; Objectification; Objectification in
and outside the sexual arena; Sexual arousal and sexual interest; Morally problematic low-level desires; Appreciating erotic
goods; The primacy of the virtuous; What are virtues?; Sexual
virtue and the common good; The common good; Chastity and
roles; Conclusion.
Appendix
Notes
Bibliography
Index of Names
208
220
307
323
Chapter 1
Contraception as Contralife
“If anyone for the sake of fulfilling sexual desire or with premeditated hatred does something to a man or to a woman, or gives
something to drink, so that he cannot generate, or she cannot conceive, or offspring be born, let it be held as homicide.”
Pope Gregory IX1
In their paper “‘Every Marital Act Ought to Be Open to New
Life’: Toward a Clearer Understanding,”2 the influential Catholic
philosophers Germain Grisez, Joseph Boyle, John Finnis, and William May (hereafter GBFM) propose an argument that they believe
will (i) explain what constitutes a contraceptive act and (ii) explain
why such acts are always morally impermissible. Their argument
attempts to explain the wrongness both of contraceptive acts within
marriage, and of contraceptive acts outside of marriage, by reference to the ‘contralife’ character of such acts in the context of what
is known as the New Natural Law Theory.3
It may come as a surprise to readers that this opening chapter
is concerned with somewhat more abstract matters concerning the
general approach to moral reasoning proposed by these authors.
However, the experiences people may have with sexual value/
disvalue and the importance such values may occupy in human
life are not to be understood in a vacuum, but need to be somehow
34
Contraception as Contralife
situated in a general theory of ethics. Beginning with the GBFM
argument is useful in opening the discussion, even if some aspects
of GBFM’s overall theory and its application to sexual ethics may
fail to convince on closer examination.
The framework for GBFM’s argument involves a method of
evaluating individual human acts (which, for these writers, necessarily involve some kind of choice) in terms of certain aspects of
human fulfilment or ‘basic human goods’.4 What makes the GBFM
paper distinctive is GBFM’s separation for purposes of moral evaluation of contraceptive acts from acts of sexual intercourse, and
their refusal to analyse these two acts as one and the same human
act–i.e. an act of contraceptive sex.
Another distinctive feature of GBFM’s argument that contraception is necessarily contralife is that it does not rely on any prior
theory of what constitutes specifically marital goods in the light of
which contraception must be assessed.5
Basic Goods
To see the GBFM paper in its proper context it is necessary
to outline briefly the authors’6 more general theory of the basic
goods–sometimes referred to as the ‘New Natural Law Theory’
or theory of the Grisez-Finnis school. Adopting as a starting point
and directive for action Thomas Aquinas’s7 formulation of the selfevident first principle of practical reasoning that “the good is to be
pursued; the bad is to be avoided” (a principle which plays a role
analogous to that played by the principle of non-contradiction for
theoretical reasoning) the theory proposes that careful reflection on
what motivates human agents, regardless of location or time, reveals a limited variety of real basic intelligible human goods (values) which are realisable for human beings. The list of these basic
35
Ethical Sex
goods is made up of the most general and distinguishable features
of human activities that constitute human flourishing. These ‘basic
goods’ are, in the words of Germain Grisez, “aspects of persons,
not realities apart from persons…the basic goods by which they
[persons] enjoy self-fulfilment must be aspects of persons, not
merely things they have.”8 The practical intellect (as opposed to the
speculative intellect) is said to grasp these basic human goods directly in non-inferential acts of understanding. These basic human
goods correspond to the “inherent complexities of human nature.”9
The basic goods are, as one commentator surmises, “kinds of
goods that we can intelligibly conceive any or every human agent
as acting towards or for the sake of, in and of themselves, and with
no further objective beyond those goods in mind.”10 They are, in
other words, non-instrumental goods (although they may also be
used instrumentally). John Finnis categorises these basic forms of
human goods as Life, Knowledge, Play, Aesthetic Experience, Sociability (friendship), Practical Reasonableness, and ‘Religion.’11
As the basic goods provide ultimate reasons for acting they are,
according to the Grisez-Finnis school, incommensurable. In other
words, instantiations of differing basic goods present in options for
choice cannot be weighed up against one another; for each can be
a distinct and ultimate reason for acting. There is no more fundamental good by which we compare the relative values of the basic
goods.
Practical Reasonableness is one of these basic goods, and from
its requirement (namely the ‘self-evident’ principle that good be
done and pursued and evil be avoided) the Grisez-Finnis school
derives a series of practically applicable moral principles which
Finnis terms the Basic Requirements of Practical Reasonableness.12
36
Contraception as Contralife
Practical Reasonableness, in the words of Finnis, “…both is a
basic aspect of human well-being and concerns one’s participation
in all the (other) basic aspects of human well-being.”13
Given that the purpose of practical reasonableness is to advance human flourishing, to act directly against (i.e. to destroy,
damage or impede) any aspect of human flourishing is of itself
practically irrational. The Grisez-Finnis school therefore holds, as
one of the basic requirements of Practical Reasonableness, that one
can never be morally right in choosing directly against a basic human good. One may pursue a basic human good in action, or one
may at least respect it (i.e. not damage it), but may never choose to
damage such a good. As Finnis observes in Natural Law and Natural Rights:
Reason requires that every basic value be at least respected in each and every action. If one could ever
rightly choose a single act which itself damages and
itself does not promote some basic good, then one could
rightly choose whole programmes and institutions and
enterprises that themselves damage and do not promote
basic aspects of well-being, for the sake of their ‘net
beneficial consequences’.14
GBFM take it that to aim for ‘net beneficial consequences’ is
to adopt an incoherent general objective for action, because the
incommensurability of the basic goods means that consequences of
actions cannot be commensurably evaluated.
Much more can be said about this approach to ethics, but a
number of points should here be noted. The approach, though
attractive in its focus on human flourishing, invites searching
questions as to how we are to understand concepts of ‘damaging,
37
Ethical Sex
destroying, or impeding’ a basic human good, especially, for our
purposes in this chapter, that of life.
We need to ask which acts count as ‘direct’ attacks and whether
all such acts are truly immoral. For example, in the case of capital punishment actions that do seem to be ‘direct’ attacks on the
good of life have been and are approved in certain circumstances
by many, including the religious tradition out of which the GrisezFinnis theory arises (even if that Church has increasingly frowned
on the use of capital punishment in practice).15 Yet on the GrisezFinnis account it is difficult to see how such actions could ever be
morally justified. Similar problems crop up with regard to what are
usually taken to be morally licit actions of armed police or soldiers
in just wars: even where such people are not intending to kill aggressors, they are at least intending to do serious bodily harm.
Furthermore, how is one supposed to characterise actions that
‘impede,’ for example, the basic good of play, and why need such
actions always be necessarily wrong?
Incommensurability
As well as providing answers to such questions, the theory
needs to give a clear account of the practical implications of the
‘incommensurability’ of the basic goods. With regard to the contraception case, GBFM posit that, as there are no goods commensurable with the basic good of human life, one can never make
a rational choice against life. It would only be rational to choose
against the basic good of life for the sake of a greater good, but
given the incommensurability of any other good with this (or any
other) basic good, such a choice could never be indicated.
GBFM apply their theory of basic goods to contraceptive acts,
to which they object on the ground that they are identifiable as
38
Contraception as Contralife
necessarily contralife choices–i.e. choices made directly against
the basic good of life. As contraception is, according to GBFM, a
contralife choice, it is, necessarily, irrational.
The fact that couples do try to calculate whether to bring about
a child at some given time, and appear to compare that to the option of not bringing about a child at that time does not, according
to GBFM, mean that their calculations are in fact rational. In comparing two possible futures, one in which a baby comes to be and
one in which he/she does not, the couple appear to be calculating
on the basis that they “know that the future without the baby will
be rationally better,”16–and, say GBFM, this they cannot know.
However, this approach seems fraught with difficulties. Such
high standards for knowledge claims threaten to call into question
any knowledge claims based on future probabilities for mediumterm decisions.17 All of us make, all the time, exactly these kinds of
decisions, based on reasonable projections of future probabilities.
Consider also a couple that decide not to conceive a child in
January but to conceive one in March instead. They do this for
reasons of the mother’s health. Cannot the mother claim that, in
choosing to have the baby later in the year, she is comparing one
life against one life–and, all other things being equal, doing so for
reasons of health? In which case, cannot she be said to be choosing
the same amount of life (beginning two months later) plus the good
of health? Given this, it certainly appears that the choice to try and
conceive in March overrides the rational appeal of conceiving a
child in January.
If GBFM refuse to concede the above, then it follows that the
sorts of commensurating decisions that GBFM do allow as rational are not possible either. Why is it rational (ceteris paribus
39
Ethical Sex
and with no better options) to crash your aircraft into a field that is
less densely populated than a neighbouring field? Such a decision
is rational according to GBFM because the pilot’s “comparison of
possible futures established the rational preferability, in terms of
saving human lives, of steering towards the less densely populated
area.”18 However, it is unclear how the loss of V’s unique and irreplaceable life in field 1 is being compared to the instantiations
of X, Y, and Z’s lives in field 2. The value of X, Y and Z’s lives in
field 2 may exceed but does not include the value of V’s life.19 Put
another way, proposal (a) “to crash into field 2” does not have all
the beneficial features of proposal (b) “to crash into field 1” and
some more. Moreover, how is one supposed to identify the ‘future
benefits’ of the pilot’s decision, following GBFM’s approach? Yet
GBFM in this case wish to characterise the pilot’s decision as rational. If that decision over apparently ‘incommensurable’ futures
can be rational, then why cannot other such decisions?
In light of these examples we can ask regarding incommensurability: if a chooser is to be a virtuous chooser, i.e. one who aims to
make those choices most fulfilling of human nature, then he must
be choosing with an eye to greater or lesser value (which is not to
say that value can be identified independently of what it is virtuous to prefer). And the virtuous chooser is precisely a chooser who
is good at identifying just how valuable certain choices are. The
same applies to GBFM’s point about side-effects and the Golden
Rule.20 Here the morally praiseworthy application of the Golden
Rule to situations requiring a calculation of the possible benefits/
burdens of side-effects requires an agent to be virtuous–and a virtuous agent is precisely the agent who can identify correctly which
instantiations of goods are most important.
An example might be the following: While playing dominoes
I notice that a person in the same room is drowning in the bath. It
40
Contraception as Contralife
appears that I have a duty to save this person, and in so doing stop
playing dominoes. I have this duty, according to the Grisez-Finnis
school, not because life is a greater good than play (or because this
instantiation of life is a greater good than this instantiation of play),
but because, following the Golden Rule, I should do unto others
as I would have them do unto me (on this see also George (1999)).
However, the proper application of this rule assumes that I am not
a domino-playing fanatic, who thinks that playing dominoes is so
important a good that even if I were dying, I would urge others to
carry on playing dominoes rather than save me. In order for the
rule to be properly applied we implicitly assume a virtuous agent
of the kind I have described. This is not to deny that the virtuous
agent is bound by certain non-negotiable demands concerning
particular virtues and their related goods. It is merely to note that
the GBFM view of incommensurability seems far too strong to account for some actions that seem morally required.
The Contraceptive Act
Having briefly considered incommensurability and the theory
of basic goods it is important now to be clear how GBFM define a
contraceptive act. The authors state that:
...to contracept one must think that (1) some behaviour
in which someone could engage is likely to cause a new
life to begin, and (2) the bringing about of the beginning
of new life might be impeded by some other behaviour
one could perform. One’s choice is to perform that other
behaviour; one’s relevant immediate intention (which
may be sought for some further purpose) is that the prospective new life not begin.21
This definition shows that GBFM see contraception as only
loosely related to intercourse: a sexual act and a contraceptive act
41
Ethical Sex
are, in their view, two separate acts. The definition contrasts with a
definition which takes intentional contraceptive acts of sexual intercourse as the basis for a definition of contraception.22
One of the reasons for setting up the debate this way is that,
by focusing upon the intention of those who contracept, their chosen human act can be properly analysed in terms of intentionality.
GBFM appear to define contraceptive acts in a way that captures
what Pope Paul VI in Humanae Vitae #14 condemns, namely any
act, “which either before, at the moment of, or after sexual intercourse, is specifically intended to prevent procreation – whether as
an end or as a means.”
In order to stress what their definition of contraceptive acts includes, GBFM state that contraception is not defined by a pattern
of outward behaviour. An example they give is of a populationcontrolling dictator who for his own purposes ‘contracepts’ by
adding a fertility-reducing chemical to the public water supply. The
dictator’s behaviour is, on GBFM’s account, contraceptive; that
is, he makes a choice the moral object23 of which is to impede the
beginning of the life of possible persons. The moral object of the
choice is what renders his behaviour contraceptive. Such behaviour, on this account, necessarily exhibits a contralife will. Couples
who drink the water and subsequently have intercourse, insofar as
they did not drink the water for contraceptive reasons, would not
be contracepting, although their intercourse would be infertile.
Whether the dictator’s own actions should be brought under
the term contraceptive is an interesting question. It does seem confusing to say, with Janet Smith,24 that the dictator has turned the
marital acts of the water-drinking victim couples into contracepted
acts (albeit non-culpably contracepted acts)–unless contracepted
acts merely means acts physically incapable of procreation (in
42
Contraception as Contralife
which case it might be better simply to speak of physically sterile
acts). What the dictator cannot do is turn anyone else into a contraceptor by such means, for it is not in his gift to impinge upon
the moral objects of the chosen behaviour of others by means of
acts of which they are unaware. All he may do is render sexual acts
physically infertile. However it seems that we can usefully say of
the dictator example that not only parents but society needs to welcome (a) fertility and (b) new lives. On the assumption that contraceptive and other sterilizing acts are in some sense immoral, the
dictator might corrupt married couples and destroy their marital
acts at the subjective level (promoting contraception and not just
secretly changing the water) or he might fail to respect the goods
that couples’ own actions do respect (i.e. a) and b)).
There appears to be a moral difference between a couple
choosing to perform an act of contraceptive sex, and the dictator
choosing to pollute the water hoping to sterilise others. A common
argument against contraception, based on characterising contraception in terms of contraceptive sex acts, will say that the couple are
intentionally doing something to their marital act which renders it
non-marital and that this rendering differs radically from what the
dictator is doing, as he is simply incapable of rendering anything
non-marital without engaging the couple’s own intentions.25
An example to elucidate this seeming difference might be the
following. Imagine a dictator who, in introducing his fertility-reducing drug into the water, somehow ensures that it will only affect
the fertility of non-married couples. He does this because he thinks
it is unjust for people to have children out of wedlock. He also believes it to be immoral for any couple to contracept their own acts
of intercourse, because, according to him, in so doing the couple,
married or not, render their acts opposed to the good of marriage.
43
Ethical Sex
Now imagine a non-married couple contracepting their sexual
act with a similar motive to the dictator sterilising the water i.e. to
avoid a child being born out of wedlock. The couple have turned
one type of act (their act of merely non-marital intercourse) into
another type of act entirely (a contracepted act of non-marital intercourse). The couple have not, of course, turned an otherwise
marital into a non-marital act, but rather, have further distanced
their non-marital act from a marital act as GBFM and the Catholic
Church to which they belong would understand that act.26 In so doing the couple have affected their characters differently from the
way in which the dictator has affected his.
A common approach to the question of contraception would see
the couple’s act but not the dictator’s as coming under the heading
of specifically sexual immorality. Is the dictator being sexually immoral in some way? We might imagine a case where the dictator
does not even know what marital meanings the sexual act is meant
to express–he may merely have some hazy idea that normal sexual
intercourse is loosely connected with the possibility of conception.
But if this is the case, is the dictator’s wrongdoing to be characterised as merely a case of illicit use of power, such as illicitly
sedating a couple would be? Is it only the illicit use of power over
the couple that would characterise the wrongdoing of the dictator
in a case where the dictator, with perhaps a relatively benign further intention, sought to prevent the couple’s sexual act from being
fertile? Or is his wrongdoing to be characterised more specifically
as a species of sexual wrongdoing? It might be argued that the dictator does not relate properly to certain couples’ reproductive wellbeing and in this sense behaves somewhat analogously to a couple
who contracept their sexual act. However, there are very real differences too.
44
Contraception as Contralife
On the GBFM account, what makes both the couple and the
dictator contraceptors is their possession of a contralife will. Any
further difference in the moral nature of their acts is, for GBFM,
extraneous to the question of whether they are engaging in the type
of wrongdoing known as contracepting.
Such an account of contraception cannot properly exclude the
following type of case. Imagine an aunt expressing to her niece
disapproval of her niece’s intention to have a child. The aunt foresees that some behaviour of her niece is likely to result in a new
birth, and believes rightly or wrongly that her verbal expression
of disapproval will make this eventuality less likely. Is she, on the
GBFM definition, a contraceptor? If she is a contraceptor she is
surely a worse one than the dictator, for she is directly attempting
to influence the intentions of her niece and not merely whether or
not she has a child.
Would it matter if her advice to her niece not to have a child
were grounded in a recognition that the niece, for very good reasons, had a duty not to have a child? GBFM might try and argue
that their use of the word “impede” in relation to new life rules
out cases such as this, but they provide no definition of the word
“impede” or any independent reason for thinking that such cases
should be ruled out by their definition.27 Lawrence Masek imagines
a case where a father, fearing his daughter and son-in-law may not
chart their fertility correctly (they have a serious reason to avoid
pregnancy), adds an anovulant drug to his daughter’s drink. The
couple do chart correctly and the father’s intervention is superfluous. For Masek, the father “does not commit an injustice by frustrating someone’s desire to become pregnant. In this case defenders
of Humanae vitae cannot circumvent the contralife will argument
by appealing to the injustice of using contraception on others. (To
45
Ethical Sex
emphasize that the woman and her husband are not victims of
an injustice, one could add that they would welcome her father’s
intervention if they knew about it).”28 This argument, however,
misses the point that an assault is still an assault even if the victim
or victims would not object to it if they knew. And a virtue ethicist
might say that such an assault is precisely against the virtue of justice. And surely something qualifies as an assault that involves a
non-consensual thwarting of - or attempt to thwart - a function: in
this case the non-trivial matter of a woman’s fertility.
If it can be shown that GBFM’s definition of contraception is
in important ways inadequate in drawing a distinction between
contraceptive acts and the practice of ‘Natural Family Planning’
(NFP), which GBFM follow the Catholic Church in seeing as licit
in principle, there will be no reason to include the above cases under the term of contraception. For, if contraception has such a wide
meaning, inclusive of many activities GBFM and the Catholic
Church regard as licit, then clearly GBFM will no longer be merely
defending the teaching of Humanae Vitae, but actually rejecting it
as too lax.
One reason given by GBFM for treating contraceptive acts
independently from sexual acts is that by doing so, “one cannot
argue that since marital intercourse is good, contraception involved
in it can be acceptable. If the contraceptive act and the marital act
were one and the same human act, that argument might succeed,
since that one act could be analysed with two effects.”29 This,
GBFM believe, would make this approach open to the objection
that the ‘principle of double-effect’ might justify such an act.
However, many of those who object to contraceptive sexual
acts will deny that such acts are marital in nature, precisely because
they have been deliberately rendered contraceptive.30 Such an ac46
Contraception as Contralife
count, which I have called the contraceptive sex act account, allows that ‘taking a pill’ is an act of contraceptive intent (assuming
that is why the pill is taken) and that ‘having sexual intercourse’ after taking the pill is acting with an intention that may well confirm
the original intention (i.e. the intention to render infertile any possibly fertile sexual act). So, while there are two acts–the taking of
the pill and the sexual act–the contraceptive sexual act is one that
confirms the prior intention embodied in the pill-taking act.31
Necessarily Contralife
Given their wide definition of contraceptive acts, GBFM proceed to argue that “every contraceptive act is necessarily contralife.” A couple, according to them, “in choosing contraception as
a means to [a] further good… necessarily reject a new life. They
imagine that a new person will come to be if that is not prevented,
they want that possible person not to be, and they effectively will
that he or she never be. That will is a contralife will. Therefore,
each and every contraceptive act is necessarily contralife.”32 As
GBFM put it elsewhere, “the contraceptive acts seeks to impede
the beginning of the life of a possible person.”33
Obviously an account of the wrongness of contraception in line
with Catholic teaching will need to exclude the actions/omissions
of couples who licitly practice NFP, but we will come to that later.
The central claim that GBFM make about the wrongness of contraceptive acts appears to be that a couple that choose such acts can
be said to be choosing against the life of a new possible person.
GBFM further state that “…in and of itself, a contraceptive act is
nothing but contralife,”34 that “insofar as contraception is contralife, it is similar to deliberate homicide,”35 and that insofar as a will
is contralife it embodies a “practical (though not necessarily emotional) hatred of the possible baby they project and reject, just as
47
Ethical Sex
the will to accept the coming to be of a baby is a practical love of
that possible person.”36
Possession of a contralife will presumably means at very least
something formal, such as ‘having an attitude that does not appropriately value the good of life,’ or ‘having an attitude that relates
to the good of life inappropriately.’ However, GBFM imply that a
contralife will is something more than this. They state that, “an essential condition of the immorality of deliberate homicide is that
it involves a contralife will,”37 and further that, “contraception is
similar to deliberate homicide, despite their important differences,
precisely inasmuch as both involve a contralife will…the contralife
will that contraception involves also is morally evil, although we
do not claim that it is usually as evil as the homicidal will.”38
However, a homicidal will is a will that aims at depriving an
identifiable individual or individuals of the good of life, thereby
harming him/her/them.39 For GBFM the contralife will involved
in contraception is a refusal to help to bestow the benefit of existence on a possible person, but I take it that this does not involve
a ‘harm’ to any such person who will not in fact exist: there is a
major difference between an action that harms by depriving a real
future person, and an action that fails to benefit a possible person.40
True, a choice of failing to benefit can be a wrongful harm in some
circumstances where there will be a real future victim (and even
where there will not be there may be a failure in generosity involved). However, what the distinction shows is that the term contralife operates in far too wide and formal a way to capture what is
an “essential”41 condition of a homicidal will.
Indeed, there is a question as to whether it is correct to characterise the contralife will as an “essential” condition of a homicidal
will if by that we mean a non-trivial condition. We might say that a
48
Contraception as Contralife
homicidal will is a will directed to the end that a particular person
or actual future person or persons be (effectively) harmed by depriving them of life. But, if I will to kill X I am willing against X’s
life (even if I may not be willing to ‘harm’ X as such–I may not
believe death will be harmful). There is no reason to suppose that
there need be a separable ‘contralife’ component of the homicidal
will which acts as a ‘condition’ for the homicidal will, except in
the trivial sense that life is inappropriately related-to (at least if the
killing is unjust). Similarly a theist might argue that an essential
condition of any sin is that it involves a will opposed to God’s law,
such that insofar as one type of sin involves such a will, it has that
in common with any other sin. Here what is constitutionally a sin
need not be something intended to violate God’s law as such, even
if it is known that it will do so (the relationship between one’s intentions and awareness of this kind is discussed further in the Appendix).
Admittedly GBFM’s understanding of ‘contralife’ will is narrower than our imagined ‘contra God’s law’ will; nevertheless,
there are multifarious ways in which one might have what GBFM
term a contralife will. The aunt in our earlier example may have a
contralife will insofar as she takes verbal steps to discourage her
niece from conceiving: if there is no moral reason why her niece
should not conceive, then the aunt relates inappropriately to the
good of potential new life. However, what of a couple who do have
a duty not to have a child, but nonetheless will to conceive one?
Surely their will must also be, in some sense, contralife, in that
they relate inappropriately to the good of life by their very pursuit
of that good (we might say, they value life but fail to respect it).
A couple, on the GBFM account, have a contralife will if they
intend their extra-marital sex not to result in conception. According to GBFM, a couple who practise NFP with the intention that
49
Ethical Sex
no child come to be, and choose the not-coming-to-be of a child
as a means to achieving their further goal of avoiding baby-related
burdens, have a contralife will. A couple who refuse to bestow
the good of life on a possible person, if such a formulation makes
sense, and use contraceptive means to achieve this goal also have a
contralife will.
These examples make it clear that there are many different
ways of possessing a contralife will. They should also make clear
that the attempt to draw a close analogy between a homicidal will
and a merely contralife will is misconceived.
Possible Persons
A number of further objections can be made to the GBFM account. Firstly, the idea that a couple necessarily ‘imagine’ a possible person whom they make sure, by their actions, will not come to
exist, does not seem to be justified. There is no reason to think that
‘contracepting’ couples necessarily have such thoughts.42 What are
the identity conditions for such a possible person? Just how specific does the imagining of the possible person have to be, and does
the specificity of the couples’ visualisation of the possible person
they are choosing against affect the level of moral wrongness of
the contralife choice made against that possible person?43 As concerns the possible baby that results from contracepted sexual intercourse being unwanted/hated, GBFM would need to establish that
this practical hatred is a necessary motivation for the couple to undertake a contraceptive act, not just that it is a likely later reaction
to any actual baby that does in fact come to be.
In translating precise definitions into predicate logic we need
a criterion of identity. As Frege argued, if we are to use the symbol a to signify an object, we must have a criterion for deciding
50
Contraception as Contralife
in all cases whether b is the same as a, even if it is not always in
our power to apply the criterion.44 Not only can we have no object
without an identity; we can have no naming of an object without a
criterion of identity. No such criterion for the ‘imagined’ or ‘foreseen’ possible person is provided by GBFM. This is an important
point, for the wrongness of contraception on the GBFM model runs
along the lines that it is illicit to will against the good of life of soand-so. However, the most one might will against in fact would be
a range of possible persons one of whom will, in coming to exist,
fall under the described range.
Can we make intelligible the notion of practical hatred for a
person who does not yet exist? Imagine two families, the Pascals
and the Molinas, who are involved in a long-running blood-feud.
One day a member of the Pascal family, Rufus, poisons the well
of the Molina family with the intention of killing the future yetto-be-conceived grandchild of the Molina’s. Cannot Rufus be said
to have, at least, a malicious intention towards the future Molina
grandchild? In a sense this seems a reasonable assumption, but
needs to be qualified. Surely the most Rufus can be said to have is
a conditional general attitude to any possible future Molina grandchild. As a general intention this is condemnable but even here,
Rufus cannot be said to have a practical hatred for a person who
may never come into existence. Rufus’ intention cannot be characterised as “there exists x–I intend to make sure that x doesn’t
exist,” precisely because x does not yet exist. Moreover, Rufus can
only will against a range of possible persons. One may always ask
the question, of someone who refers to “the future Molina grandchild,” “why do you suppose that there is only one (non-existent)
future Molina grandchild?”45 So the most that Rufus may have is a
negative general propositional attitude towards a range of possible
51
Ethical Sex
persons.
The analogy with contraception breaks down, however, when
we remember that the contracepting couple is acting precisely to
ensure that the possible person is not actualised, whereas Rufus is
doing nothing to prevent the coming-to-be of the future grandchild,
and may even welcome his/her coming to be so that his evil plan
can be realised. Whereas Rufus wills harm to one of a range of
possible persons he anticipates will be actualised, the contracepting couple wills against a range of possible persons who they will
never to be actualised. Rufus’ will can here be called homicidal in
a way in which the contracepting couple’s can’t be.
Admittedly, Grisez and Boyle46 do not now use the word ‘imagine’, and talk of the couple necessarily ‘foreseeing’ a baby coming to be. This seems more hopeful insofar as it jettisons the idea
of active imagination of a specific possible person and is more
suggestive of the idea of a range. The theory, however, runs into its
biggest difficulties with the NFP/contraception distinction.
GBFM appear to be making the counterfactual claim that if
sperm X were to enter ovum Y at such-and-such a point, and at
such-and-such a time, and at such-and–such a temperature etc., we
would have a specifiable possible person. Given this, GBFM then
seem to be saying that a contracepting couple must be necessarily foreseeing such a specifiable possible person, and acting so as
to prevent his/her coming into being. Of the status of such a being
GBFM say: “the possible person whose life is presented is no mere
abstraction but an absolutely unique and unrepeatable individual
who would exist if he or she were welcomed rather than prevented.”47 According to GBFM, such an intention to prevent is always
wrong because it entails a contralife will. It entails a contralife will
because a couple, in choosing to contracept, is doing no more and
52
Contraception as Contralife
no less than impeding the beginning of the life of that specifiable
possible person (or, at least, a narrowly specifiable range of possible persons).
However, we need to ask here how the intentional relation to
the specific possible person or persons involved in the case of the
contracepting couple can differ from that involved with an NFP
couple. The overarching choice in both these cases is “not to cause
[or “not to help to cause”] the initiation of new life” as the result of
a chosen sexual act. GBFM appear, in characterising contraceptive
acts in terms of an intentional relation to the bringing into being of
a possible person, to cast their net too widely. The argument is that
both a contracepting couple and an NFP couple can have good reason not to cause the initiation of new life, due (for example) to the
burdens that a baby’s coming-to-be would bring about. To contracept in order to achieve this goal is, for GBFM, never morally licit
because contraception necessarily involves intentionally impeding
the beginning of a possible person’s life. NFP, however, on the
GBFM account, need not be contralife, although it can be chosen
with “contraceptive intent.”48
The distinction of importance for GBFM is the following. To
contracept is to choose to prevent the beginning of the life of a
possible person–it is “a choice to do something, with the intent that
the baby not be, as a means to a further end.” However, the noncontraceptive choice of NFP49 differs because “It is a choice not
to do something–namely, not to engage in possibly fertile sexual
intercourse–with the intent that the bad consequences of the baby’s
coming to be will be avoided, and with the acceptance as side effects of both the baby’s not-coming-to-be and the bad consequences of his or her not-coming-to-be. In this choice and in the acceptance of its side-effects, there need be no contralife will. The baby
53
Ethical Sex
who might come into being need not be projected and rejected.”50
The authors further add: “[T]rue, not choosing to realise that good–
and, indeed, choosing to avoid the burdens one anticipates if one
were to realise it–means not willing that the good be realised, but
it does not mean willing that the good not be realised.”51 The claim
is that the NFP couple can be, “choosing not to realise something
they have good reason to choose to realise [a baby], but whose
realisation would conflict with avoiding something else they have
good reason to avoid [the burdensome side-effects of that baby’s
coming-to-be].”52
Against this, it can be objected that the contracepting couple
and the NFP couple stand in the same intentional relation to the
beginning of new life, at least in the sense that both couples intend
that life not be conceived. The contracepting couple, like the NFP
couple, may well characterise their decision to contracept as a decision to choose against the burdens that the coming-to-be of a new
baby will bring about. GBFM say that this necessarily involves for
the contracepting couple a contralife will because it necessarily
involves the intention to impede the coming-to-be of a new child.
The NFP couple are said to be willing only against the bad sideeffects that the coming-to-be of a new baby will bring about. But
how is one to make sense of the idea that one can try to avoid the
effects of having a baby without trying to avoid having a baby?
According to the NFP couple’s plan, their means of avoiding the
effects of having a child is that they not have a child. One of the
propositions a couple must go through en route to reaching the intended outcome of their plan (not having those burdensome baby
side-effects linked with the baby’s coming-to-be53) is the proposal:
not having a baby.54 The deliberate removal of the necessary causal
precondition of avoiding the baby side-effects cannot be character-
54
Contraception as Contralife
ised as a mere side-effect of the couple’s plan.55
Masek’s Defence
Lawrence Masek, in an ingenious defence of the GBFM thesis
argues that a couple using NFP and desiring that human life not
begin do not have to be intending the non-existence of human life
either as an end or as a means. He proposes three examples to defend this claim:
Case 1: A baby has an illness that causes both
earache and constipation. A pediatrician informs
the baby’s parents that the two symptoms cannot be
treated simultaneously. The parents need to choose
which symptom to treat first, but the pediatrician
says that each symptom causes an equal level of
discomfort and that there is no medical basis for
choosing one over the other. The parents toss a coin
to determine which symptom to treat first.
Case 2: The baby’s situation in case 2 is the
same as the baby’s situation in case 1. The parents
are about to toss a coin to determine which symptom to treat first. They then consider that they have
a long car trip planned and that treating the earache
first would make the car trip easier, since it would
reduce the hassle of changing dirty diapers on the
trip. The parents therefore choose to treat the earache first.
Case 3: Parents of a healthy baby are planning
a car trip. They add a chemical to their baby’s food
in order to impede the baby’s digestion, in order to
55
Ethical Sex
cause constipation, in order to reduce the hassle of
changing dirty diapers on the trip.
Masek tell us, “I see no reason that the parents act wrongly in
case 1. If they may toss a coin in case 1, I see no reason that they
may not consider the effects of each option in case 2. In case 3,
however, the parents do act immorally since they intentionally
impede their baby’s health as a means to having fewer hassles on
their trip.”
We can cautiously agree with Masek that these judgements
“seem intuitively plausible.”56
One does not need to reject the idea (strongly argued for by
Frances Kamm57) that an effect can motivate an action without being intended, in order to reject Masek’s arguments with regard to
contraception and NFP. What we need is to examine the examples
Masek gives, and others, in the light of questions about contraception.
It would seem that the difference between Case 2 and Case 3 is
like the difference between (3) contraception (‘deliberate constipation’58) and delaying sexual activity along NFP lines (2).
For Masek it is unreasonable to say that the couples in cases
1 and 2 ‘intend’ the baby to be constipated. For when you choose
when a particular bad side effect happens (constipation), you need
not be intending it.
However, this becomes more difficult to argue the more useful
to you is the aspect that guides your choice (as opposed to e.g. situations where that useful aspect is a mere ‘defeater of a defeater’, as
Kamm would call it). Let us imagine two cases involving fertilityaffecting treatment, which is more directly related to contraception
56
Contraception as Contralife
than Masek’s cases:
Case A: A woman needs treatment that will
make her temporarily infertile (a bad side-effect
of some morally good medical treatment). The infertility is unwelcome (or at least, not particularly
focused-on): it is simply a bad side-effect. She
will choose when to have the treatment and so will
choose the time when she has that bad side-effect.
Case B: A woman doesn’t want a baby and
chooses the same treatment that the woman in
Case A receives, but regards the ‘side effect’ as the
doctors see it as a bonus. She generally objects to
contraception (i.e. rendering otherwise fertile acts
infertile), but wants infertile sex with her husband
when he comes back from his lengthy naval posting. In this case she is intending to make herself
infertile. This is not a mere side-effect. Her actions
differ from those of the NFP couple who use infertile or fertile times for intercourse, because here the
woman is choosing to make a fertile time infertile.
In light of these relevant examples what might be said about
Masek’s points? Certainly it can’t be the case that you intend sideeffects which are still altogether unwelcome but which you shift to
a less bad time, e.g. I will shift agonising pain to Thursday rather
than Wednesday. Surely it is ludicrous to say I am intending agonising pain as an effect of my operation!
However, when the ‘side-effect’ is welcomed and seen as
conferring a benefit overall, it is a different story, e.g. when the
secondary effect of infertility is brought about so as to ‘benefit’
57
Ethical Sex
the woman at a time of her choosing. If you time the ‘side-effect’
to get the perceived benefit (as opposed to minimising perceived
harm) then how can we say this effect is not intended?
In an NFP case similarly, there is certainly an intention that
children not come to be. To claim that the NFP couple are merely
not choosing to do anything to achieve something is to mischaracterize what they are doing. NFP can be adopted by a couple precisely because they intend that a new child not come to be on the
grounds that the burdens this would bring about are serious enough
for them to avoid having a child at that time. By deliberately refraining from intercourse when the woman is thought to be fertile,
the couple intend precisely not to bring about a new child.59 While
the use of contraception actively prevents (or lessens) the possibility of a new child coming to be, the NFP couple in contrast are said
to act licitly in avoiding by omission a new child coming to be.
However, the NFP couple, insofar as they are pursuing that strategy, are not just omitting to actively bring about the coming-to-be
of a new child, they are intending by omission that a new child not
come to be.60 This is precisely the rationale behind the adoption of
NFP.61
The distinction, for GBFM, between the NFP couple and the
contracepting couple in their intentional relation to the existence
of new life is difficult to see. Certainly the authors when writing about euthanasia62 see as morally equivalent deliberate active
euthanasia and euthanasia by deliberate omission (as opposed to
‘letting die’ without intending death). Is the idea then that the NFP
couple must somehow try to intend only to avoid the burdens of
the possible baby’s coming-to-be, without intending that baby’s
not-coming-to-be? But as we have seen, the avoidance of the coming-to-be of the baby is integral to the plan of refraining from in58
Contraception as Contralife
tercourse. If we do accept the GBFM position here, then cannot the
couple using a condom who also will to avoid the bad side-effects
of pregnancy be said to be not necessarily willing against conception itself? If that is the case, then GBFM must characterise such
couples as non-contraceptors, which is highly counterintuitive.63
The only way to distinguish NFP and contraception, on the
GBFM account, appears to be that with contraception the couple
seem to be foreseeing and directly choosing against a relatively
specific possible person. So, with a barrier contraceptive, the couple (assuming full knowledge of sexual reproduction) are said to
be willing against any possible person that would result from any
of the sperm that would have otherwise penetrated the woman’s
ovum. But even here there are, on average, over 40 million sperm
per ejaculate that might fertilise the women’s ovum at any of the
many possible times: as with the NFP couple, it may be claimed,
their refraining from intercourse during a certain period cannot be
said to be a direct willing against a clearly specifiable possible person, given that there is no clearly specifiable possible person.
However, if we accept this distinction, then surely we must
accept that the couple using an ovulation-impeding contraceptive
(e.g. the Pill) are also not directly willing against a clearly specifiable possible person (leaving aside abortifacient effects they may
realise are possible and positively intend should there be ‘breakthrough’ ovulation). In point of fact, the couple who have sex after
taking the pill are even more distanced from a specific possible
person than the NFP couple. For the NFP couple are deliberately
refraining from sexual activity due to the fact that they believe a
particular ovum at a particular time will be present, whereas the
pill-taking couple are having sex when they believe no ovum will
be present at the time of sexual activity. So, the pill-taking couple,
59
Ethical Sex
insofar as they relate to possible persons, are further distanced
from a specifiable possible person than the NFP couple–who stand
in closer relation to the possible persons that would result from
ovum x being fertilised, at time y, by one of over 40 million sperm.
Again, the NFP couple are omitting to act at time y, but it is an
intentional omission adopted on the grounds that they intend that
none of a range of possible persons come to be. GBFM might have
other grounds for rejecting pill-usage (such as that it temporarily
suppresses healthy functioning), but that position would be independent of their contraception/contralife argument.
Moreover, the NFP couple, in refraining from intercourse at
a certain time, with the intention that a new baby not come to be,
could be said to be choosing that a specifiable possible person (i.e.
the person that would have resulted if they had chosen to have
sexual intercourse instead of refraining) not-come-to-be. The NFP
couple intentionally omit to have intercourse because they seek to
avoid any such intercourse causing the beginning of a new person’s
life.
If the claim were that the difference between the NFP couple
and the contracepting couple is that these couples have a different
intentional relation to the sexual activity and its being a possible
cause of new life (the NFP couple has an intentional relation to
abstain from sexual activity–something GBFM themselves note),
then we have a rather different argument from the one proposed by
GBFM which is concerned too broadly with the intentional relation
to the “beginning of new life.”64
Rape Case
On the GBFM account, then, there appear to be no cogent
grounds for excluding the licit practice of NFP from the category
of contraception. On top of this GBFM also have difficulties with
60
Contraception as Contralife
the case of pregnancy prevention after rape. The latter is generally
accepted by Catholic commentators on these issues, and various
Bishops’ Conferences or their Committees have said that a woman
who has recently been raped is quite within her rights to seek to
evacuate the rapist’s sperm from her body.65 GBFM have this to
say on such an action: “…the measures that are taken in this case
are a defence of the woman’s ovum (insofar as it is a part of her
person), against the rapist’s sperm (insofar as they are parts of his
person).”66 The question here is whether a woman in this situation
is, according to GBFM, morally permitted to prevent the conception of a specific possible person.
Imagine a woman who either knows or does not know that she
is fertile. She is much more likely to try and rid herself of the rapist’s sperm if she believes that by not doing so she risks the conception of a possible person. But if that is the case, then she very
definitely is willing against a possible person. She foresees the possible person coming-to-be (sperm X, Y, Z etc. entering her ovum),
and she acts in such a way that that possible person does not come
to be. Such action, on GBFM’s account, exhibits a contralife will.
Likewise, if I were to see eminent and wealthy fertility expert, Dr
W. Inston, at one of London’s glamorous IVF clinics, about to inject a single sperm into an ovum in a petri dish, and I grabbed his
arm to prevent the procedure, would I be exhibiting a contralife
will? I certainly can envisage the IVF (or rather ICSI – introcytoplasmic sperm injection) created baby – I foresee his/her coming
to be, and I will against it. More realistically and less dramatically,
we can imagine a couple who have agreed to try IVF, but, at the
eleventh hour, decide that it is wrong to produce a child in such a
way. Do this couple exhibit a contralife will in grabbing the doctor’s arm to stop conception going forward?
61
Ethical Sex
One of the authors of the paper, Germain Grisez, has written
further on this subject and suggested that conception should be
understood as the completion or continuation of sexual union, and
that the victim of the rape is justified in resisting the continuation
of that union.67 Such a position implies, however, that a rape victim
who does not attempt to have the rapist’s sperm removed from her
may be consenting to (or at least not resisting) the completion of
that sexual union even though she is now in a safe environment. So
the rape victim is not only justified in having the sperm removed,
but would appear on this view to have a duty to have it removed
(insofar as this is practicable).68 However, Grisez does not state this
implication, making only the weaker claim that a woman in this
situation can be justified in having the sperm removed.
It is difficult to make sense of Grisez’s attempted solution to
the problem of rape victims and contraception. He makes the claim
that “…women who are victims (or potential victims) of rape and
those trying to help them are morally justified in trying to prevent
conception insofar as it is the fullness of sexual union” (emphasis
added).69 If conception simply means the completion of sexual union then we appear to have a simple identity statement, such that,
insofar as a couple knows that conception is the fullness of sexual
union, in willing against the fullness of sexual union they are necessarily willing against conception (i.e. the coming to be of a new
human being). If the fertilisation process includes conception we
also appear to have a problem of symmetry. For, if a woman may
licitly will against the fertilisation process qua the end of the process (conception), then cannot a contracepting couple will against
conception qua the beginning of a (gestational) process and as a
means to their end of avoiding pregnancy burdens?
However, what Grisez can say is that there are two separable
events here–1) the fertilisation process, and 2) the moment of con62
Contraception as Contralife
ception; and that the second event is related to the first as an effect
is related to a cause. Up to the moment of conception then, we
have a part of the man (sperm) engaged in the fertilisation process.
The fullness of sexual union is therefore achieved just prior to conception, and conception is an effect of that union. At the moment
of conception, the part of the man (sperm) breaks up and the sexual
union ceases. On Grisez’s account, the rape victim can licitly will
against the completion of sexual union qua sexual union, and not at
the same time necessarily be willing against conception.70
Whether this is psychologically plausible (assuming accurate
knowledge on the part of the woman as to what constitutes fertilisation and conception) is extremely doubtful. Nevertheless, even if
we were to accept this account, there are further problems. Imagine
a couple who plan to have sexual intercourse and then stop midway
so that they do not begin a fertilisation process. On this account,
the couple might be able to say that they had no intention against
conception (they might even welcome conception), but that, for
whatever reason, they were willing against the fertilisation process
only. Perhaps such a couple could be criticised by Grisez for intending incomplete sexual acts–but this would be to criticise them
on grounds other than that they are contraceptors. If one were to
say that this example is hugely psychologically implausible, then
one undermines the plausibility of the rape victim case also.
Conclusion
Given the inadequacies of the GBFM thesis in accounting for
the distinction between contraception and NFP, and the permissibility of pregnancy prevention in cases of rape, it would appear
that a different approach is needed in defining, and putting a case
against, contraception. By attempting to provide an account of
contraception that is not reliant on a prior theory of marital goods,
63
Ethical Sex
or presuppositions about the inherent values and meaning of the
sexual act, GBFM have to adopt an externalist account as to the
wrongness of separating the unitive and procreative meanings of
the sexual act.71 They do account for the ‘inseparability thesis’,
but only by pointing to what they take to be the wrongness of on
the one hand, contraception, and on the other, IVF/AIH (artificial
insemination by husband). Contraception, they claim, is always
wrong, and for the reasons their paper has summarised. But, as I
contend, their account is inadequate as it stands. Moreover, what
the failure of GBFM’s account suggests is that it is difficult to give
an account of the wrongness of contraception that is entirely separable from an account of the inherent goodness of the marital act–
an inherent goodness that contraception is said to pervert. If we
cannot understand the wrongness of contraception without a prior
understanding of the value of the marital act, then this suggests
that the whole GBFM project has started off on the wrong track (or
that the wrongness of contraception and IVF is illusory). With IVF
also, is it so evident that we can understand the alleged wrongness
of IVF without also having some notion of how children should
in fact be conceived–i.e. by a marital act? While it is prima facie
wrong to make a human being a product of technological manufacture, it might be difficult to account for the repugnance felt toward
this procedure without some prior understanding of how children
ought to be brought about–i.e. by marital union.
The GBFM account attempts to explain what they see as the
wrongness of contraception as applied to both marital and nonmarital acts. However, it may be possible to account for this perceived wrongness by giving a prior account of how marital acts
instantiate marital goods, and by demonstrating that sexual acts
outside marriage which are contracepted are somehow further dis-
64
Contraception as Contralife
tanced from marital goods than non-contracepted sexual acts outside of marriage.
An account of the kinds of acts that constitute chaste acts – acts
of full committed union and, perhaps, responsible parenthood –
might be the account we are looking for. Such an account need not
deny that contraception insufficiently respects the life-giving nature of acts of sexual intercourse (i.e. their teleology and meaning),
and is, in this sense, contralife. On the contrary, it may give us a
better idea of just how such contracepted acts are contralife (i.e.
fail to respond appropriately to the value of life/fertility), and in a
way free from the difficulties into which the GBFM account falls.
65
Notes
Chapter 1: pp. 34-65
1
Cited in Grisez, Boyle, Finnis, and May (1988), 366. The quote has appended to it the following footnote (2): ““Decret. Greg IX, lib. V, tit. 12,
cap. v; Corpus iuris canonici, ed. AL Richter and A. Freidberg (Leipzig:
Tauchnitz, 1881), 2, 794: “Si aliquis causa explendae libidinis vel odii
meditatione homini aut mulieri aliquid fecerit, vel ad potandum dederit,
ut non possit generare, aut concipere, vel nasci soboles, ut homicida
teneatur.” Some translate “causa explendae libidinis,” which is broad
225
Ethical Sex
enough to cover all motivation by sexual impulse, “to satisfy lust,” which
unnecessarily limits the motive to habitual vice.”
Such texts are philosophically interesting, although it is worth recalling
here that this book is a work of philosophy and does not rely on the authority of religious texts/teachings for its conclusions.
2
Reprinted in Ford, Grisez, Boyle, Finnis and May (1988), 35-116.
All further references to this text in this book (referred to henceforth as
GBFM (1988)) will be to this source. This paper has subsequently been
further defended/amended by Grisez (1993), 506-516 and by Grisez and
Boyle (1998), 228-232.
3
For more on the arguments focusing on contraception’s contralife character within the Church’s tradition see Noonan (1986) and the relevant
references in GBFM (1988) 37 fn.2 & 38 fn.3.
4
Grisez (1993), 513 fn. 104 states: “An argument against contraception
grounded in its opposition to the good of life can be articulated without
articulating the general theory of basic goods and modes of responsibility”; he refers to the Thomist article (reprinted in GBFM (1988)) as such
an argument. However, to make sense of the idea of willing against the
good of life does require some kind of foundational ethical theory and it
is unclear which account GBFM can be appealing to other than the basic
goods theory. Moreover, the appeal GBFM make to the incommensurability of future goods apparently assumes such a theory (see GBFM
(1988) p.52 footnotes 8 & 9).
5
In addition to the contralife argument, other writings of John Finnis in
particular have developed theories on sexual ethics involving marital
goods which could be applied to contraception – see Finnis (1997). In
this paper Finnis refers to the good of marriage as “one of the basic human goods to which human choice and action are directed by the first
principles of practical reason” p.97. See also Grisez (1993), 567.
I refer here specifically to John Finnis, Germain Grisez and Joseph
Boyle, who have developed their theory in a number of works including:
Grisez (1983), Grisez (1993), Finnis (1980), Finnis (1983), Grisez and
Boyle (1979), Grisez, Boyle, Finnis (1988).
6
226
7
Summa Theologica, I-II, q.94, a.2.
8
Grisez (1983), 121.
9
Finnis (1991), 42.
10
Chappell (2005), 29.
11
Finnis (1980), ch.4.
12
Finnis (1980), 100-133 where Finnis adumbrates 9 basic requirements.
Grisez generally refers to these requirements as ‘Modes of Responsibility’.
13
Finnis (1980), 101-102.
14
Finnis (1980), 120-121. For more on respecting basic goods see Finnis
(1983) 124-127.
15
See for example Catechism of the Catholic Church (1994), #2266. For
criticism of Finnis’s understanding of how to respect the basic good of
life see Lamont (2002).
16
GBFM (1988), 54. Of course, couples may have other motives for contracepting or abstaining from potentially fertile intercourse, such as that
they feel too fragile psychologically to conceive now, and recoil from
the psychological burden without assuming that a possible future with
a child would be a ‘worse’ possible future. In such cases the decision
need not involve the kind of comparison of long-term futures that GBFM
envisage. Moreover, a couple deciding to delay for two months bringing
a new child into the world, so as to avoid the woman being heavily pregnant during a predictably hot summer, may be making their decision not
on the basis of two long-term unknowable futures, but rather on the basis
of a short–term concern. They may not focus at all on comparing the
long-term futures, or they may assume that the good of the lives of the
two possible children would cancel each other out so that only short-term
considerations are relevant.
17
GBFM (1988) appear to adopt two lines of argument. One concerns the
incommensurability of basic goods such that commensurating calculation
between them is practically impossible. The choice to contracept is, on
their grounds, contralife and therefore contrary to a reason (not directly
227
Ethical Sex
to will against the good of life) that cannot be rationally outweighed.
The other line of argument concerns our lack of knowledge of the future,
even when any commensuration might be done in terms of just one basic
good.
Consider a case where a couple make a decision about when to conceive
with reference only to the good of life. A couple that choose, via contraception, not to bring about a child at t1 might be taking into account
all sorts of reasons not to bring about a child at this time. Those reasons
might well relate to ‘respecting’ the good of life, in that bringing about
a child at t1, although instantiating life, is not thought to be an act sufficiently respectful of life (due to, for example, the inappropriateness of
the situation for receiving a child). Contraception may indeed be an inappropriate way by which to avoid bringing a child into the world, but this
argument about commensuration does nothing to prove this. What would
need to be established would be the ‘contralife’ character of contraceptive acts as distinct from the practice of abstinence.
A further question is whether GBFM are committed to the view that,
ceteris paribus, if a couple were to have a choice between conceiving
twins or a single child, they should aim at conceiving two, for example if
they are receiving fertility enhancement. Presumably they would have to
answer yes, particularly given their position regarding the pilot example
(see main text below). But surely in this case, the couple cannot know, at
least on the GBFM view, that two is better than one as concerns the longterm future – in which case why would it be necessarily better to choose
two?
18
GBFM (1988), 55. GBFM might reply that three lives as opposed to
one life is the kind comparison that can rationally be made. However,
imagine if you knew that the three people in field 2 all had one second
to live, regardless of whether you crashed the plane into them or not –
whereas the one person in field 1 had many years to live. Would this not
affect what the pilot should do? If so, we are implicitly acknowledging
that numbers are relevant only other things being equal (such as potential
victims having equal expectations of future life). And if this is the case,
then cannot a couple, in delaying conception for two months for, say,
228
the health of the mother, be recognising that by doing this they achieve
life + health with conception in month 3, which outweighs life + lack
of health with conception in month 1? If the pilot’s decision cannot be
rationally faulted on grounds of incommensurability, then neither can the
contracepting couple’s (of course, this is not to deny that the choice to
delay conception via contraception may be irrational/immoral for other
reasons).
19
GBFM (1988), 56 offer another example of the conditions required
for rational decision-making. Here a house-hunter is deliberating between two houses, one of which has all the value-and-more of another,
in terms of price, size and proximity to the school (the exclusive criteria
for choosing in this example). The house with all the value-and-more
than the other, say GBFM, provides one with an unchallenged reason to
buy it. However, this immediately suggests another example where house
(a) which far outweighs house (b) in terms of two of the set criteria but
underweighs it a tiny bit in the remaining one criterion (e.g. size) cannot be ‘rationally’ chosen over the other house (assuming that there is
no priority amongst the three criteria). If house (a) is chosen over house
(b) here, the size (i.e. specific spatial area) of the house rejected is ‘lost’,
but it might be argued that this is not a very important loss, and does not
change the outcome of commensuration. To anyone who objects that
the fact of a particular choice means that one is, in this case, commensurating by rating the criteria in some way the question remains why, at
the general level, commensuration should be thought to be impossible.
Moreover it is unclear whether GBFM think that one can commensurate
even at the level of particular instantiations of goods.
In the pilot example the loss of V’s life clearly is, at least, an important
loss. But what reason is there to dismiss the idea that the choice of house
(a) over house (b) is simply more rationally fulfilling of the chooser in
that almost nothing will be lost in comparison with what is gained?
20
GBFM (1988), 57-58.
21
GBFM (1988), 42.
22
See Smith (1992), 340-370 and Anscombe (1972).
229
Ethical Sex
23
For a representative statement as to what the Grisez-Finnis school
takes to be the defining feature of an action see Grisez (1983), 233. For
a more recent account of what constitutes the moral object of an act see
Finnis, Grisez, and Boyle (2001).
24
Smith (1992), 360-361.
25
I do not here propose to give a full exposition of the ‘contraceptive sex
act’ or an account of how sexual acts ought to relate to marital goods;
these are subjects for later chapters. GBFM recognise this different approach to the question of contraception, which they see as compatible
with the contralife approach, observing that “recent Church teaching
focuses almost entirely on contraception’s wrongness in relation to other
values, especially chastity, marital love, and the sacred character of virtuous sexual activity in marriage.” GBFM (1988), 38.
26
The Catholic Church holds that use of contraception prevents sex in
marriage from being what it holds it should be: an act of marital intercourse: complete bodily and personal self-giving. For the same reason,
it holds that sex between unmarried people falls short of what it should
be (again: marital intercourse) on yet another ground, hence “further distanced.”
Self-giving, the giving of a gift, would also be violated by a couple who
after sex used a spermicide out of fear of pregnancy. Perhaps rather similarly, for the Catholic Church, attempting to cancel your marriage vows
doesn’t nullify those vows but does radically disrespect them. The effects
of those vows continue over time and are open-ended and have implications for the couple’s life.
Sex and contraception are ‘separable’ as they are in any case of non-barrier contraception (a person forgets they took the pill or had an implant
3 months ago). Nevertheless, for the Catholic Church the distortion of
self-giving is clear if we see the giving as being potentially prolonged in
effect over a period of time even if the nature of the gift may not be fully
known.
27
It is unclear what the term ‘impede’ is meant to encompass in GBFM’s
paper. If they mean by ‘impede’ any action undertaken with the intention
230
of preventing the completion of a causal process which could otherwise
lead to the creation of a new life, an argument needs to be made as to
why the impeding of such a causal process is always wrong. Such an
argument would, presumably, have to take into account the significance
of the teleology of sexual activity (see later chapters in this book) and of
any objective meanings such activity embodied, as well as explore why
it might be morally wrong to have a dual intention of performing and
thwarting an act that could give rise to the causal process leading to conception.
Suffice it to say here that I do not believe GBFM have given any satisfactory account of why ‘impeding’ is wrong in this context, nor have they
sufficiently defined what they mean by ‘impede’ so as to rule out those
actions which they would accept as morally licit. Is the thought that
there is some general principle of the type: It is always impermissible to
impede a process that is under way that might lead to a good, because
to choose to do something to stop that process involves willing that that
good not be?
This cannot be a general principle. For example do I have a duty not to
impede the recovery from illness of someone who is a threat to the common good? Do I have a general duty not to impede someone about to
make a true but dangerous communication to a third party? It seems clear
that these are not moral duties even though my choice that someone not
recover now or learn something now is surely against the good of health
and knowledge respectively.
By using the term ‘impede’ GBFM are indicating that there is more to
their contralife argument than mere ‘willing against life’. Given that
some ‘impeding’ of processes under way that might lead to a good appears to be morally justifiable, the term is used in too wide a sense.
None of this is to deny that ‘impeding’ can indeed be morally significant
and even morally conclusive in view of what is being ‘impeded’ and
how, in the context of chosen sexual activity. But if ‘impeding’ must take
a particular form in this particular area to be morally conclusive, then we
are dealing with a substantively different issue from the mere possession
of a contralife will.
231
Ethical Sex
28
Masek (2011).
29
GBFM (1988), 44.
30
Finnis (1993), 189 states: “…contracepted intercourse has, objectively,
the masturbatory feature that it simulates an aspect of the conjugal good
that has in fact and in intention been excluded from it by an act which affects the reality of sexual activity (qua chosen) itself.”
31
A woman taking a pill and subsequently repenting of, or at least regretting, her decision, who proceeds to have intercourse which has been
rendered sterile by the pill she took, is not, in having that intercourse,
confirming the intention embodied in the former pill-taking act. It is a
more complex question whether a woman who takes a pill and is later
completely indifferent as to whether her intercourse is rendered infertile
by the pill is having intentionally contraceptive sex. It is worth noting
that a woman could take a pill in case she chooses to have intercourse
(which she is hoping to resist as immoral or unwise), not with the intention to have intercourse. See the discussion in Chapter 4 of conditional
intentions.
32
GBFM (1988), 43. This statement has since been amended by Grisez
and Boyle (1998), 231 to “contraceptors necessarily foresee that a baby
might come to be, they want that foreseen baby not to come to be, and
they choose to do something in order to make it less likely that he or she
will come to be.”
33
GBFM (1988), 36.
34
GBFM (1988), 43.
35
GBFM (1988), 44.
36
GBFM (1988), 46. It is not clear whether the ‘practical love’ of a possible person is even possible. Even God surely doesn’t create out of love
as opposed to creating a real person though a willingness to share love
with him/her.
37
GBFM (1988), 45.
38
GBFM (1988), 46-47.
232
39
The moral object of such a will – the deprivation of life – is what defines the nature of such a will.
40
On this point see also Bayles (1976), 298-299. It should be clear that
‘harm’ and ‘benefit’ are not being used in their ordinary senses when relating to non-existent possible persons. A couple may bring about future
benefits, including the benefit of existence, but cannot be said to be benefiting pre-existing persons. Possible persons may be specifiable only in
the sense of exactly the person who is conceived as it happens.
41
By “essential” condition I presume GBFM mean something at least as
strong as “necessary” condition.
Defining contraceptive acts in terms of a will that is necessarily contralife would also have the result that a couple using a condom to prevent
the transmission of an infection was not contracepting in any sense. I
think that GBFM are correct not to characterise such activity as necessarily contraceptive as opposed to wrong for non-contraceptive reasons.
42
However, the GBFM account of the wrongness of contraception would
appear to make the following case, suggested to me by Edward Feser,
wrong and for contraceptive reasons: Suppose someone creates a drug
which guarantees that every act of sexual intercourse will result in
pregnancy, and indeed will result in twins, triplets, or quadruplets. Now
suppose he tells me that he has put it into the water I just drank and that
unless I vomit it up within five minutes it will take permanent effect. Do
I act immorally if I induce vomiting, and if I induce it precisely because I
don’t want 50 or 100 children? Would this mean I have a ‘contralife will’
or ‘hate’ these babies? On the GBFM account the answer would appear
to be ‘yes’. This example at least shows that the answer requires much
thought before ‘yes’ could be accepted as reasonable (one might ask,
though, of Feser whether in a case of ovarian hyperstimulation someone
could destroy excess ova for avoidance of multiple conceptions and not
be said to be contracepting).
An example of activity that many would consider contraceptive but
would not be deemed so on the GBFM model would be the following
(again from Feser): Suppose someone develops a drug/device which en-
233
Ethical Sex
sures that after conception takes place, the fertilized ovum (zygote) is put
into suspended animation for a year, or five years, or whatever, within the
woman’s body, and that after this period the pregnancy will resume as
normal. And suppose further either (a) that after this happens the woman
cannot get pregnant again until the existing child is born, or (b) that other
children may be conceived, but they will go into suspended animation
too so that a ‘backlog’ is built up and the woman will give birth to multiple children later on.
Now, in either case – and whatever moral differences there are between
(a) and (b) – we can imagine that people might use this device precisely
to enjoy sex as long as possible without children. But they are not (especially in case (b)) acting with a ‘contralife will,’ at least if we imagine that they are perfectly willing to have the child(ren) when the time
comes.
Such an action does seem to have many of the earmarks of contraception
many people find objectionable. If GBFM’s account of what contraception consists in is lacking (see the discussion of NFP in this chapter),
then we need to find an independent way of judging the morality of such
acts, as opposed to the type of acts GBFM are concerned with. Whatever
one thinks of Feser’s example (and one might object that a couple could
reasonably delay childbirth in this way for serious reasons, not involving
contraception or a ‘contraceptive mentality’ but relating to the mother’s
or children’s health), the example does raise questions for the GBFM approach.
Moore (1992) held that it is difficult even to make sense of the notion of a specific possible person, not least because only actual human
persons are truly individuated. As he puts it, “babies not yet conceived
are not particulars.” p.167. See also Bayles (1976), 298-300. But this is
too radical a claim, for we can certainly make sense of a proposition that
conditions a, b, c (etc.) for conception z were such that only specifiable
person X, and no-one else, could have resulted. As Hare (1975), 220 put
it, such a potential person is “identifiable in the sense that identifying reference can be made to him,” although clearly he is not identifiable with
some already existing person. On a Kripkean model a possible particu43
234
lar must be individuated by some connection with the real world - and
on Kripke’s scientific essentialist view, this would involve identifying
the gametes which ‘would have’ combined. We can make sense of that
proposition (assuming determinism at the relevant ontological level – i.e.
only one sperm could have ‘won’). However, we’d need to know the intimate details of the course of the sexual act in question (body positions
etc) – and it would have to be a possible alternative course in which
contraception was not used. But, arguably (depending on one’s view of
so-called externalist theories of meaning) even having a theoretical criterion for identifying the possible person doesn’t help much if we are talking about the couple having a pre-existing identifying reference to that
person which is accessible to their minds. For the best they are likely to
be able to manage is ‘the person that could result from this sexual act’ –
which is not a unique description of possible persons – certainly not from
their epistemic position, anyway. For an argument about how we can
harm future people see Carter (2001).
44
We do not have to accept Frege’s description theory of reference. Even
on Saul Kripke’s New Theory of Reference, there would appear to be an
analogous problem with possible persons.
45
GBFM (1988) refer to the couple not wanting “that possible baby to
begin to live” p.46 emphasis added.
46
I know from personal communication that Finnis agrees with this revision. I am unaware of the position May has taken with regard to the revision.
47
GBFM (1988), 66.
48
GBFM (1988), 81.
49
Grisez and Boyle (1998), 232 now deny that a couple could carry out
a contraceptive choice via NFP. As they put it, “The choice to practice
natural family planning is a choice to abstain from acts in which a baby
would become a real possibility; the choice to contracept, presupposing
the intention to engage in such acts, is a choice to try to prevent the baby
who might result from them.”
235
Ethical Sex
50
GBFM (1988), 84-85.
51
Ibid.
52
GBFM (1988), 86.
53
Obviously some baby side-effects might be avoided by other means,
such as employing an army of nannies. But here we are only interested
in the couple’s plan to avoid baby side-effects by preventing the baby’s
coming-to-be.
54
GBFM might make the claim that the baby’s not-coming-to-be is to
function as a causal means to the end of avoiding baby burdens, but is
not a means chosen by the couple. But, insofar as a couple know that the
causal means functions in this way, it is likely they are choosing this very
causal path as part of their plan to achieve their end.
It is of course quite possible to choose an end y without knowing that
means x is a necessary causal precondition of end y. In such cases there
is present a false belief that an imagined particular pathway is all that is
necessary to bring about y. But such is not the case with the NFP couple,
whose plan for end y includes a plan of chosen means x, insofar as the
couple is aware of this causal relation. The incoherence of any other position on their part is logical and not merely psychologically improbable
in adoption (Vacek (1998), 62-63 appears to believe it is only psychologically improbable).
55
Compare GBFM’s argument to GBF’s statement : “an action is intentional if it is part of the plan on which one freely acts. That is to say, what
one tries to bring about, whether it be the goal one seeks to realise or the
means one chooses to realise that goal, is intended.” p.79.
56
Masek (2011), 93.
57
Kamm (2007).
58
I take it that bringing about deliberate constipation, while in this case
morally wrong as unfair on the baby, does not fall under any absolute
prohibition of the kind that applies to contraceptive use in the view of
Masek and the Catholic Church.
59
Not merely “do not intend to.”
236
60
Humanae Vitae (16), to which GBFM are committed, states, with regard to a contracepting couple and a couple practising NFP: “In reality,
these two cases are completely different. In the former married couples
rightly use a faculty provided them by nature. In the latter they obstruct
the natural development of the generative process. It cannot be denied
that in each case married couples, for acceptable reasons, are both perfectly clear in their intention to avoid children and mean to make sure
that none will be born.” (my emphasis). A couple charting their fertility
are ‘avoiding’ rather than ‘preventing’ children coming to be. A couple
don’t chart in order to have knowledge regarding the promotion of other
goods. They promote other goods via the non-conception of a new child.
Here abstaining is something the couple ‘do’ (just as they might choose
to be silent).
61
Patrick Lee has suggested to me in correspondence that the claim that
a couple are intending by omission that a new child not come to be is
groundless, and that a couple making a choice to use NFP is analogous
to a man choosing not to write a book because of the burdens involved in
undertaking such a task. However, the analogy is not exact. The proper
analogy with the NFP case would be a man choosing not to write a book
because of the burdens that would come about if the book were produced
(e.g. getting into trouble with the government). In this case the man does
choose specifically against the proposed book in the same way that the
NFP couple choose against the proposed baby.
62
See for example Grisez and Boyle (1979); Finnis (1995).
63
See endnote 55 above for GBF’s statement on what counts as an intentional action, which is consistent with their views expressed in later
work.
64
Rhonheimer (1989).
65
Joint Committee on Bioethical Issues of Bishops’ Conferences of
Scotland, Ireland, England and Wales (1986); Joint Committee on Bioethical Issues of the Bishops’ Conference of Great Britain and Ireland
(1986); Pennsylvania Catholic Conference, “Guidelines for Catholic
Hospitals Treating Victims of Sexual Assault” (1993). It remains an open
237
Ethical Sex
question in Catholic moral theology whether such evacuation of sperm is
morally permissible (though almost all writers would accept this), as well
as whether other methods such as suppression of ovulation may be used.
The feasibility of suppressing ovulation without unduly risking the life of
any newly conceived human is a subject of intense debate.
66
GBFM (1988), 69.
67
Grisez (1993), 512. May (2000), 140-142 observes that: “…the moral
object specifying the rape victim’s (or potential rape victim’s) human
act is not to prevent the conception of a new human person but rather to
prevent ultimate completion of an unjust act of sexual violence.” He later
states, with regard to acts undertaken to prevent the rapist’s sperm from
making his victim pregnant: “ Such acts are not acts of contraception
because the object freely chosen and morally specifying them is not the
impeding of a new human life that could begin through a freely chosen
genital act (=definition of contraception) but is rather the protecting of
the raped woman from further violence by the rapist.”
68
It might be argued that a rape victim does have a prima facie duty to
remove the rapist’s sperm from her body prior to any possible conception
on the grounds that the sexual act has not been freely chosen and this an
inappropriate way for a child to be brought into the world. This argument
would depend on an idea of parental duty to the future child, and would
not apply in a case where the woman knew that the act was in any case
infertile. It seems that Grisez would have to accept, if the presence of
sperm is truly seen as part of the act even after its apparent completion,
that a woman has a duty to have the sperm removed as soon as feasible
even if she knows the act is already infertile. This seems particularly unreasonable.
69
Grisez (1993), 512. See also Grisez (1997), 251-255.
70
In GBFM (1988), 36 it is stated that the distinction between seeking to
prevent conception and seeking to impede the beginning of the life of a
possible person is a merely conceptual distinction.
71
GBFM (1988), 97: “Since contraception is always wrong and since
producing babies is always wrong, the only morally acceptable way to
238
engage either in lovemaking or life giving is by engaging in sexual intercourse that is open to new life.”
239