© 2005 by Blackwell Publishing, Inc. Philosophy & Public Affairs 33, no. 4
It is a common observation that moral compromise is a pragmatic
necessity in the political life of a democracy. All societies are characterized
by moral disagreement and it is naïve, if not dangerous, to eschew
any form of compromise in the face of such disagreement. Some theorists
have tried to make a virtue of this necessity.They claim that we often
have principled reasons to engage in moral compromise, particularly
when we disagree with sincere and reasonable people. Since the abortion
controversy is a paradigm of reasonable moral disagreement, it has
been claimed that we ought to pursue some form of principled compromise
between the pro-choice and pro-life camps. My primary aim in
this article is to repudiate this claim. Moral disagreement about abortion,
however reasonable it may be, generates no principled reason to
compromise on abortion policy. Because many of the arguments I consider
are not specific to the abortion case, my secondary aim is to cast
doubt on the very idea of a principled compromise. The more general
conjecture of the article is that moral compromise in political life is only
ever warranted for pragmatic reasons. It is quite enough that moral compromise
is accepted as a necessity without its being revered as a virtue.
In Section I, I define principled and pragmatic compromises, and distinguish
them from some related phenomena. In Section II, I set out an
egalitarian pro-choice position as well as two proposals for how it could
be modified in a moral compromise: George Sher’s rejection of public
Principled Compromise and
the Abortion Controversy
SIMON CA˘BULEA MAY
This article is a revised version of the second chapter of my doctoral dissertation, “The
Significance of Moral Disagreement” (Stanford University, 2004). I am grateful to the
Editors of Philosophy & Public Affairs for their generous comments. I am also grateful to
a number of people at various universities for discussion of the ideas contained in the
article, particularly the members of the Stanford Social Ethics and Normative Theory
group, where an earlier version was presented in late 2003. I am especially indebted to
Debra Satz for her supervision of the original doctoral project.
funding for abortion, and Amy Gutmann and Dennis Thompson’s suggestion
that public subsidies for abortion only be drawn from the tax
contributions of pro-choice supporters.1
In Section III, I defend the original
egalitarian pro-choice position against these specific proposals for
moral compromise on the public funding of abortion. In Section IV, I
criticize four arguments in defense of some sort of principled compromise
on abortion: arguments from complexity, respect, accommodation,
and reciprocity. The failure of these arguments means that there is no
obvious reason why any sort of principled compromise on abortion
should be sought. In Section V, I conclude by way of a generalization of
my arguments about the abortion controversy. If there is no principled
reason to pursue a moral compromise on abortion, there are good
grounds to believe that there is no principled reason to pursue a moral
compromise on any political question.
I. Moral Compromise
A
Political compromise occurs when a political agent invokes the fact of
disagreement as a reason to accept an alternative that she perceives to
be worse on its own merits than her initial position. Although prudential
compromises can arise between political agents motivated purely by
their self-interest, my concern is with moral compromise, which occurs
when disagreement is invoked as a reason to accept a political position
otherwise perceived to be morally inferior. Moral compromise can vary
in degree from relatively mild forms of accommodation to the complete
sacrifice of fundamental principles. Moral compromise is quite distinct
from moral correction or rectification, although both involve the modification
of political positions. If an agent is persuaded to correct her
political position, she comes to see it as morally inferior on its own
merits to the new alternative. Moral correction involves the recognition
that one’s earlier commitments were mistaken. But if the agent comes to
accept an alternative as a moral compromise, she still views her initial
position as morally superior (setting aside the relevance of the fact of
318 Philosophy & Public Affairs
1. George Sher, “Subsidized Abortion: Moral Rights and Moral Compromise,” Philosophy
& Public Affairs 10 (1981): 361–72; Amy Gutmann and Dennis Thompson, Democracy
and Disagreement (Cambridge, Mass.: Belknap Press, 1996).
disagreement). Moral compromise involves no recognition of any error.
There is a corresponding distinction between reasons for moral correction
and reasons for moral compromise (which I understand to be
exclusive). The former are reasons to think an initial position morally
worse than an alternative, whereas the latter do not undermine the
position itself. It may be helpful to think of reasons for moral correction
as first-order reasons that concern the merits of a position itself,
and reasons for moral compromise as second-order reasons that
concern how firmly one should hold to a first-order position in the face
of moral disagreement.
It can very often be a good thing for political agents to engage in moral
compromise. First, political parties representing competing interest
groups may grudgingly strike a moral compromise that happens to be a
closer approximation of fairness than any party’s conception of justice.
Second, political parties championing competing values may agree on a
compromise position that happens to be the best balancing of the relevant
values. But neither of these good consequences of moral compromise
translates as a good second-order reason for moral compromise.
Instead, they are both good first-order reasons for moral correction. If
an agent were to accept them as reasons for any sort of modification of
her views, she would be committed to correcting her initial political
positions accordingly. Since the final positions are morally superior to
the initial positions, the agent ought to recognize that her earlier commitments
were in error. The good thing about the final positions has
nothing to do with the fact that they emerged as moral compromises,
and everything to do with the independent fact that they are simply
the best positions.
B
Reasons for compromise are brought into focus if we consider a politician
(call her Jane) who endorses a position that is, in actual fact, the
best balancing of values and interests on an issue (although perhaps only
after a long period of philosophical reflection, practical experience, and
political deliberation). Assume that the issue remains the subject of
intractable moral disagreement in her society. This fact of moral disagreement
does not give rise to any reason for moral correction, since
we are assuming that Jane already endorses the best position on the
319 Principled Compromise and
the Abortion Controversy
issue. But if the distribution of beliefs and power is such that Jane is
unlikely to realize any of her goals without some modification of her
position then she has a pragmatic reason to accept a moral compromise.
Pragmatic compromise involves an extrinsic appeal to disagreement:
moral disagreement gives rise to a reason for compromise, not in itself,
but only insofar as it is contingently connected with a logically independent
consideration, in this case the threat posed to the realization of
Jane’s goals generally.
Three further points about pragmatic compromise and moral disagreement
should be noted. First, Jane has a pragmatic reason to compromise
whether or not the disagreement about the issue is reasonable,
since unreasonable moral disagreement can impede the realization of
important goals just as much as reasonable moral disagreement.
Second, pragmatic reasons for compromise only have normative significance
for those sympathetic to the goals the compromise facilitates.
Sometimes parties to a dispute do share an underlying goal that is incidentally
threatened by continued discord, as when no party has a majority
of votes but all have an interest in reaching a decision expeditiously.
But political parties will usually differ in their pragmatic reasons for
compromise given their different aims. Third, not all extrinsic appeals to
moral disagreement give rise to reasons for moral compromise. Sometimes
the existence of moral disagreement can be invoked as a contingent
indicator of a good reason for moral correction. For instance, one
could point to moral disagreement about sexuality to argue that, despite
the prevailing orthodoxy to the contrary, homosexual relationships can
be forms of the good life. This appeal to disagreement is extrinsic, since
it is not the presence of disagreement that is relevant in itself, but rather
the independent ethical fact that it brings to light. This fact counts as a
first-order reason to correct positions predicated on heterosexist
assumptions, rather than a second-order reason to compromise them.
C
In contrast with pragmatic compromise, principled compromise
involves an intrinsic appeal to disagreement: reasonable moral disagreement
gives rise to a reason for compromise in itself, aside from any
impediment to other goals it may incidentally generate. Thus, even if
Jane were in a position to successfully implement her political position
320 Philosophy & Public Affairs
in its entirety, say after leading her political party to a decisive electoral
victory, some think she may still have principled reason for moral compromise
with the dissenting minority independent of any pragmatic
consideration of future political fallout. Patrick Dobel remarks that
“principled compromise can occur for good reasons separate from a
sheer concern with another’s power [since] individuals can compromise
because they respect the personal autonomy or dignity of other citi-
zens.”2
In a similar vein, Richard Bellamy and Martin Hollis write that
“liberals have principled as well as pragmatic reasons for embracing
compromise [as] it goes together with a democratic politics informed by
duties of mutual respect and a concern for the collective good.”3
In addition,
Joseph Carens suggests that a “commitment to union” can support
the view that “in a democracy people should make concessions to their
opponents even when the concessions are not needed to create a major-
ity.”4
Colin Bird writes that the “imperative to compromise . . . derives
from the moral norm of mutual respect, not from prudential calcula-
tion.”5
Stephen Macedo claims that when faced with intractable moral
disagreement, “the most reasonable thing may well be mutually to moderate
our claims in the face of the reasonable claims of others, to balance,
and split at least some of our differences.”6
And David Wong claims that
“a willingness to bridge differences” forms the basis of a principled inclination
towards compromise and accommodation.7
Three further points correspond to those about pragmatic compromise
above. First, principled reasons for moral compromise are
usually thought to arise only when moral disagreement is reasonable
(or else when it satisfies some similar minimal moral criterion). The
321 Principled Compromise and
the Abortion Controversy
2. Patrick Dobel, Compromise and Political Action: Political Morality in Liberal and
Democratic Life (Savage, Maryland: Rowman and Littlefield, 1990), p. 80.
3. Richard Bellamy and Martin Hollis, “Consensus, Neutrality and Compromise,” in
Pluralism and Liberal Neutrality, ed. Richard Bellamy and Martin Hollis (London: Frank
Cass & Co., 1999), pp. 54–79, at p. 76; Richard Bellamy, Liberalism and Pluralism: Towards
a Politics of Compromise (London: Routledge, 1999), p. 114.
4. Joseph Carens, “Compromise in Politics,” in Nomos XXI: Compromise in Ethics, Law,
and Politics, ed. J. Ronald Pennock and John Chapman (New York: New York University
Press, 1979), pp. 123–42, at p. 135.
5. Colin Bird, “Mutual Respect and Neutral Justification,” Ethics 107 (1996): 62–96, p. 92.
6. Stephen Macedo, Liberal Virtues: Citizenship, Virtue, and Community in Liberal
Constitutionalism (Oxford: Clarendon Press, 1990), p. 71.
7. David Wong, “Coping with Moral Conflict and Ambiguity,” Ethics 102 (1992): 763–84,
p. 779.
unreasonable disagreement of white supremacists and the like with
liberal democratic practices generates no principled reason for moral
compromise. Arthur Kuflik writes that in “seeking a compromise, each
acknowledges the reasonableness of the other side’s view while insisting
on comparable recognition for its own.”8
Second, since the parties to a
reasonable disagreement should be able to recognize their mutual reasonableness,
principled reasons for compromise are thought to have
normative significance for everyone, regardless of their particular pragmatic
interests.The idea of principled compromise can therefore be seen
as offering the possibility of a stable common ground for citizens otherwise
divided by their political views; principled reasons for compromise
can be shared by all when there is no fortuitous convergence of pragmatic
reasons for compromise, and when reasons for correction fail to
persuade. Third, not all intrinsic appeals to moral disagreement are used
as reasons for moral compromise. For instance, one way (but by no
means the only way) to defend state neutrality about questions of the
good life is to invoke the fact of reasonable moral disagreement.9
This
does not count as compromising an otherwise superior liberal pluralist
or perfectionist position since the standard political liberal claim is that
no such position can be consistently defended: state neutrality is a firstbest
position.There are reasons to doubt the cogency of intrinsic appeals
to moral disagreement in general, but my present target is simply the
idea of a principled compromise.
D
The conjecture that there are only pragmatic reasons for moral compromise
is compatible with talk of principled compromise in three additional
and quite different senses. First, a pragmatic compromise can be
said to be principled or morally acceptable because, although it involves
some moral loss, it violates no fundamental moral principle.10
Second, a
322 Philosophy & Public Affairs
8. Arthur Kuflik, “Morality and Compromise,” in Pennock and Chapman, pp. 38–65, at
p. 51.
9. John Rawls, Political Liberalism, paperback ed. (New York: Columbia University
Press, 1996); Charles Larmore, Patterns of Moral Complexity (Cambridge: Cambridge
University Press, 1987).
10. Before the first democratic elections in South Africa in 1994, a compromise between
the major parties led to the inclusion of a Government of National Unity provision in the
interim constitution: all parties registering more than 10 percent of the vote were to be
pragmatic compromise can be principled or morally binding insofar as
parties ought to remain faithful to the pact they have made, even if they
entered into it for pragmatic reasons. Once agreed upon, moral compromises,
like other kinds of deals, may acquire a normative status that
they would not otherwise have.Third, a pragmatic reason for moral compromise
can be said to be principled or morally constrained because it
only arises given a commitment to moral constraints on one’s political
activity, such as a commitment to democratic decision making. Parties
can have a principled commitment to majoritarian ground rules of political
decision making without being committed to any form of moral
compromise beyond that necessary to secure a majority. Accepting only
pragmatic reasons for compromise within these moral constraints does
not entail any Leninist commitment to disenfranchising one’s political
opponents at the earliest convenient opportunity. It should go without
saying that reasons for compromise ought to be morally constrained,
and that compromises ought to be morally acceptable and at least to
some extent morally binding. But these truisms have no bearing on
whether there are ever principled reasons for forging moral compromises
as I have defined them.
The claim that there are only pragmatic reasons for compromise is
compatible with the intuitive idea that those who refuse to compromise
are potentially dangerous political actors. Political conflict is a cause of
much human suffering. Reasonable political actors should avoid ideological
fundamentalism and strive to resolve those political disputes that
threaten such misery as best they can. Indeed, a commitment to improving
the lot of the most vulnerable should be the principle that trumps all
other ideologies. But a willingness to engage in this sort of moral compromise
is best understood as a healthy pragmatism in the pursuit of a
basic humanist commitment. Pragmatic reasons for compromise are
sometimes the weightiest moral reasons a politician has.We do not need
to underscore this point by introducing the idea of a principled reason
for moral compromise.
323 Principled Compromise and
the Abortion Controversy
represented in the cabinet. This compromise was morally acceptable since coalition government
violates no fundamental democratic principle. In contrast, the U.S. slavery compromises
of 1820 and 1850 were morally unacceptable since they permitted the practice of
slavery in Missouri and the New Mexico territory seized in 1848. Perhaps there was no way
for abolitionists to eliminate slavery before the Civil War, but that would make the compromises
unavoidable, not morally acceptable.
II. The Abortion Controversy
A
Jane endorses an egalitarian pro-choice policy on abortion, (A), that
comprises a number of more particular positions:
(r) Women have the right to an abortion for any reason during the
first two trimesters;
(n) The state should be neutral about whether abortion is
unethical;
(f) Abortion ought to be publicly funded (commensurate with similarly
significant healthcare needs) from money accumulated
through the standard involuntary taxation procedures;
(w) Consultations and waiting periods that aim to promote reflection
about the decision to abort should not be required; and
(j) The judiciary may enforce (r), whereas (n), (f), and (w) are
matters for the legislature to decide.
(A) can be expanded to include positions on other abortion-related
matters, for instance whether minors may have abortions without
parental consent, but for the present discussion it is unnecessary to
present a comprehensive version of the policy. It is naturally an open
question whether there are any reasons for correction that imply that (A)
is not the best position on abortion. Pro-life proponents argue that the
fetus’s right to life grounds a reason for correction that defeats (r). Similarly,
a pro-choice majoritarian who rejects judicial review of legislation
would not accept (j), and could instead argue that (r) should be directly
implemented by the legislature. Although I take it that (A), or something
quite like it, really is the best position on abortion, focusing on the
second-order issue of moral compromise requires setting aside the firstorder
issue of moral correction, other than to note that it is a difficult
moral problem about which reasonable people disagree. The claim that
there are no principled reasons for moral compromise on abortion is
compatible with quite different policies, even pro-life positions.11
It is
324 Philosophy & Public Affairs
11. It is worth emphasizing this point. If the pro-life position were to be correct then,
on the argument advanced here, there would be no principled reason to forge a compromise
that permitted (at least elective) abortion. It does not follow from this that intransigent
pro-life proponents are acting justifiably when they refuse to compromise in a
nevertheless useful to assume a specific, reasonably plausible view on
abortion. It is a mistake to think that arguments against principled compromise
on abortion must themselves be scrupulously neutral between
all reasonable positions, as this may be to rule out in advance the best
arguments about what (egalitarian, pro-choice) Jane has reason to do.
An advocate of a more extreme pro-choice position, a proponent of
the permissibility of infanticide for example, could reject (A) as insufficiently
permissive, either because it restricts the right to abortion to the
first two trimesters, or because it does not positively endorse abortion
as ethically unproblematic. (A) is not, therefore, the most extreme prochoice
position available. Defending an uncompromising pro-choice
position, however, does not count as defending an extreme pro-choice
position. That would be to confuse second-order intransigence about
compromise (how one holds to a position) with first-order extremism
about content (the substance of the position held). Sometimes moderate
pro-choice positions on abortion are described as compromises
between the value of autonomy and the sanctity of life. But this mistakes
a balancing of values, warranted by reasons for correction, with a moral
compromise between opposing viewpoints. For instance, Ronald
Dworkin’s position on abortion might resemble a possible moral compromise
between pro-life and pro-choice camps in that it emphasizes
the sanctity of life in addition to women’s autonomy.12
But Dworkin only
argues that these camps have, as a matter of fact, more in common than
might otherwise be thought, not that there is any principled reason to
find common ground in a moral compromise. Dworkin offers his position
as the right answer on its own merits, not as the best compromise
given the existence of intractable moral disagreement.
325 Principled Compromise and
the Abortion Controversy
position of advantage. Those who refuse to compromise unjustifiable positions are acting
improperly, not because they refuse to pursue principled compromises, but because their
positions are unjustifiable from the beginning. The error lies at the first order, not the
second. This means that the moral stakes are high when we come to a conclusion about
the moral permissibility of abortion, the justifiability of a military conflict (see n. 40 below),
or any other profoundly controversial moral question. Some may prefer a less daunting
approach. But why should we accept a philosophical analysis that made our moral dilemmas
seem any easier or less consequential than they actually are?
12. Ronald Dworkin, Life’s Dominion: An Argument about Abortion, Euthanasia, and
Individual Freedom (New York: Alfred A. Knopf, 1993).
B
Sher claims that the fact that there are good considerations on both sides
of the abortion question means that one must in all sincerity admit that
one could be mistaken about the correctness of the view one espouses.
Jane should accept that her endorsement of (A) is fallible in a way in
which the view that children ought not to be tortured for fun is not. “But
once this acknowledgment is made, such considerations as respect for
one’s opponent and the value of mutual accommodation may permit (or
even require) adjustments in behavior that would otherwise be inap-
propriate.”13
Jane has reason to moderate her position on abortion, so
that conflicting duties of conscience do not commit each side to permanent
acrimony and antagonism. Gutmann and Thompson’s reasoning
is similar. They believe that fair terms of social cooperation ought to
be mutually justifiable to citizens. When citizens defend arrangements
and policies, the ideal of deliberative reciprocity requires that they practice
an economy of moral disagreement in which they accommodate, as
far as possible, the reasonable views of others. Although moral accommodation
can occasionally occur if opposing sides find they already
have sufficient points of congruence between their initial positions,
“moral compromise may sometimes be permitted or even required by
principles of accommodation.”14
The abortion controversy is one of the
clearest examples of a deliberative disagreement, that is, a disagreement
“in which citizens continue to differ about basic moral principles even
though they seek a resolution that is mutually justifiable.”15
This means
326 Philosophy & Public Affairs
13. Sher, p. 369.
14. Gutmann and Thompson, p. 88. That moral accommodation often involves moral
compromise is indicated by the language that Gutmann and Thompson use: pro-choice
advocates should “move a small but significant way towards accommodating pro-life concerns”
and should support policies “even if they would otherwise place those policies significantly
lower on their own list of political priorities” (ibid., pp. 87–89). In this way, moral
accommodation usually involves some moral loss. Gutmann and Thompson are clear
though that moral accommodation does not warrant “compromising one’s moral convictions
solely in the interests of agreement,” which I take to be a rejection of morally unacceptable
compromises, not moral compromise per se (ibid., p. 85). See also Amy Gutmann
and Dennis Thompson, “Democratic Disagreement,” in Deliberative Politics: Essays on
Democracy and Disagreement, ed. Stephen Macedo (New York: Oxford University Press,
1999), pp. 243–79, esp. p. 266.
15. Gutmann and Thompson, Democracy and Disagreement, p. 73.
that the best resolution of the abortion problem ought to accommodate,
as far as is morally acceptable, both pro-life and pro-choice convictions.
The arguments advanced by Sher and Gutmann and Thompson are
appeals to disagreement, because if there were no abortion controversy,
Jane would have no opponents to accommodate or respect, and hence
no reason to modify (A). They are not offering first-order reasons for
moral correction, since the values they cite do not ground any argument
that (A) is flawed on its own merits. (A) is not incoherent in the way in
which the political liberal claims that liberal perfectionism is ultimately
unstable on its own terms. So they are offering second-order reasons for
moral compromise. The abortion controversy has obviously led to significant
social strife (including murder) and has the potential for a good
deal more. Although this may ground a separate, pragmatic reason for
moral compromise, it is clearly not the focus of their arguments. The
values they cite would not cease to operate in a situation where Jane’s
party could implement its political agenda without damaging side
effects. So the reasons offered for moral compromise are principled
rather than pragmatic; the appeal to moral disagreement is intrinsic
rather than extrinsic.
C
The abortion controversy is the stock example of proponents of principled
compromise, but few actually suggest what an appropriate compromise
on abortion should look like, beyond vague exhortations to
moderation.16
This is not entirely improper, since whether a compromise
327 Principled Compromise and
the Abortion Controversy
16. Dobel writes that nowhere is the need for moral compromise “more evident than
in the continuing conflict over abortion [but] I would not even attempt to suggest a compromise
. . . [except that any compromise should involve] respecting the need for an
autonomous time of decision . . . as well as recognition of interests in protecting the potential
person in the fetus” (Dobel, pp. 87–88). Bellamy and Hollis write simply that “there are
a host of middling positions between those who oppose abortion of any sort and those
who believe it should be freely available on demand up to 26 weeks or beyond” (Bellamy
and Hollis, p. 75). Macedo does briefly suggest that Jane should revise (r) and drop (w):
“legalized abortion, perhaps, but only up to a certain point in the pregnancy, and only with
consultations that help insure that the decision to abort is carefully considered” (Macedo,
Liberal Virtues, p. 72). But no elaboration is offered. Bird advocates setting viability as the
point after which abortion should no longer be legal (Bird, p. 94). The reason for this is that
neither conception nor birth could serve as points that would enjoy the support of both
pro-life and pro-choice camps. Bird also suggests that the beginning of the second
trimester might be a reasonable compromise criterion (ibid., p. 91).
proposal is a good one depends on its being the realistic object of a reasonably
broad consensus, which can be a difficult empirical question.17
Yet this cannot be the only criterion of a good principled compromise,
since it does not offer one any moral guidance on how compromising
one should be to reach such a consensus. Obviously, no reason for moral
compromise can demand unconditional capitulation, but neither can it
be satisfied with purely cosmetic concessions that fail to incorporate any
of one’s opponents’ substantive views. As such, if principled reasons for
moral compromise are to be something more than otiose rhetoric, they
must be followed up with principled reasons in (at least tentative)
support of specific proposals. The compromises that Sher and Gutmann
and Thompson offer are supported in precisely this way. The proposals
would require Jane to relinquish the public funding policy, (f). In its
place, Sher suggests:
(h) Elective abortion, though legal, should not be publicly funded
(akin to the Hyde Amendment).
Sher proposes (h) in part because he denies that women have any moral
right to be provided with abortion. The state could fulfill basic welfare
and self-determination rights with financial support to struggling
mothers and an effective adoption program for women who simply do
not wish to raise children. Sher believes that it is more plausible to think
that utilitarian rather than rights-based considerations support public
funding. But utilitarian considerations are much easier to override than
rights. Although there may be some utilitarian costs to (h), the values he
cites in support of principled compromise should be able to trump
them. Since public subsidies amount to an implicit state endorsement
of abortion, and since they “place conservatives in a position of actively
supporting abortions rather than reluctantly tolerating their perfor-
328 Philosophy & Public Affairs
17. Martin Benjamin qualifies his endorsement of a restriction of (r) to the first
trimester with the observation that pro-life proponents may simply never be able to agree
in good conscience to any compromise permitting abortion. Given the deeply religious
identity of most pro-life proponents, one cannot “persuade them to compromise without
changing them into different people” (Benjamin, p. 169). Nevertheless, Benjamin is optimistic
that there may be political and philosophical ways to circumvent this obstacle. Benjamin
also seems to suggest that (j) may be weakened or dropped entirely: “it might be
best, if historical conditions were to make it possible, to shift the focus [of the abortion
issue] from the Supreme Court to Congress” (ibid., p. 166).
mance by others,” public funding of abortion is inappropriate, however
utilitarian it may be.18
Sher’s position faces a serious difficulty, since all government-funded
programs place those who dissent in the position of supporting policies
with which they disagree. This cannot mean that all such programs
ought to be compromised, however. Sher acknowledges this difficulty in
a footnote, where he claims that his point “applies only to cases in which
public policy is contested on moral grounds [and where] the principles
underlying the disagreement [are] both supported by reasonable arguments
and serious enough to justify extra-legal activity.”19
Yet, this does
not provide an acceptable criterion to distinguish cases where compromise
on public funding is appropriate from cases where it is not. I take
it that moral and religious pro-life arguments against abortion are no
more reasonable than moral and religious pacifist arguments against
war, at least in any significant way. If pacifists were right that all wars are
unjust forms of mass slaughter, civil disobedience campaigns and other
forms of extra-legal activity would presumably be justified. But this does
not provide a good reason to curtail defense spending, unless the
antecedent of the hypothetical is accepted. Nor is the depth of feeling
about the abortion controversy a good criterion, since many may feel
just as strongly that abortion services should be publicly funded as those
who reject abortion outright. Measuring the strength of emotions is
never particularly fruitful in political theory. It is unclear whether there
is any more plausible criterion to identify moral controversies where
compromise on public funding is appropriate. Even if there is, Sher’s
position is subject to other criticisms.
Gutmann and Thompson reject Sher’s argument that (h) would
not violate women’s rights. In subsidizing childbirth but not abortion
“the government literally makes an offer the indigent women cannot
afford to refuse.”20
Even if abortions are publicly funded, however,
Gutmann and Thompson claim it should be possible to substitute (f)
with a more accommodating policy. Three options are suggested by
their discussion:
329 Principled Compromise and
the Abortion Controversy
18. Sher, p. 371.
19. Ibid., n. 6.
20. Justice Brennan, writing in dissent in Harris v. McRae, 448 U.S. 297 (1980) at 334,
quoted in Gutmann and Thompson, Democracy and Disagreement, p. 99.
(o) Abortion ought to be publicly funded, but only if there is a conscientious
objection clause exempting sincere pro-life proponents
from some portion of their mandatory tax liability
(presumably an amount proportionate to the percentage of
public expenditure directed to abortion services).
(v) Abortion ought to be publicly funded, but only with voluntary
donations from pro-choice taxpayers made in excess of their
mandatory tax liability.
(m) Abortion ought to be publicly funded, but only with that portion
of their mandatory tax liability that pro-choice taxpayers specifically
authorize for the purpose.
Gutmann and Thompson reject (o) and seem to suggest (m), but without
clearly distinguishing it from (v). They use two quite different examples
to illustrate the idea of funding abortion services from pro-choice
contributions: Colorado’s income tax form check-off boxes and the
Presidential Campaign Fund box on the federal 1040 income tax form.21
The Colorado check-off boxes are equivalent to (v): “all state check-off
contributions are voluntary donations from a taxpayer’s refund, not
a redirection of his or her liability.”22
The Presidential Campaign Fund
contribution, however, is the equivalent of (m): checking the box does
not increase one’s tax payments or reduce one’s refund.23
Naturally, some
combined version of these policies could be implemented. Nevertheless,
it is instructive to consider all three possibilities separately.
III. Public Funding of Abortion
There are at least three questions that figure in determining the best
position on the public funding of abortion. First, is it inappropriate for
the state to actively support abortions, and not merely tolerate them,
given a context of reasonable moral disagreement? Second, are there
adequate reasons of justice that support subsidized abortions, assuming
public funding would not be inappropriate? Third, should subsidies for
abortion services only be drawn from pro-choice taxpayers’ contributions,
given some or other system of public funding is warranted?
330 Philosophy & Public Affairs
21. Gutmann and Thompson, Democracy and Disagreement, p. 90.
22. <http://www.checkoffcolorado.org/info.html>
23. <http://www.irs.gov/pub/irs-pdf/i1040.pdf>
A
Sher’s argument against public funding relies on a distinction between
endorsement and toleration. Whereas toleration is acceptable, state
endorsement is not desirable in a situation where citizens reasonably
disagree about ethical matters. Sher’s argument may gain credibility
from an analogy with religious toleration. It is one thing to tolerate
different religious practices, quite another for the state to proclaim the
truth of a particular religion and support it with taxpayers’ money. Subsidizing
abortion would seem to amount to a similar adoption of a partisan
ethical doctrine, especially given the close association of pro-life
views and traditional religious convictions. Thus, although Jane may
have ethical views favorable to abortion, it would be just as wrong of her
to use her political office to divert public funds in support of her personal
convictions as it would be for an evangelical president to divert
public funds to his favorite religious group so that it may spread the
Word of God.
The analogy is not compelling. There are two basic moral questions
associated with the abortion controversy: is abortion ethical and should
it be legal? Sher’s point must be that public funding violates state neutrality
on the first question, since the second question has to be
answered one way or the other. But public funding of abortion has no
more bearing on the ethical question than any other pro-choice position
(including Sher’s). In this it differs from the case of the evangelical president,
whose policies only make sense given a commitment to the truth
of his religion. No pro-choice advocate thinks that the state has any business
dictating the ethics of abortion to women. Proponents of public
funding of abortion argue that it is necessary to ensure women’s effective
ability to choose, whatever their ethical views happen to be. In turn,
this ability to choose is a crucial determinant of the quality of women’s
lives. Public funding is not premised on any desire that women choose
one way rather than another, nor on any belief that abortion should be
no more ethically troubling for a woman than an appendectomy, but
simply on the idea that an inability to control their reproductive destinies
is not in women’s interests.24
There is therefore no reason to think
331 Principled Compromise and
the Abortion Controversy
24. That this claim about women’s ability to choose is compatible with neutrality about
the ultimate rightness or wrongness of abortion can be illustrated by the claim of some
pro-life proponents that women who have elective abortions are selfishly pursuing their
that public funding of abortion violates state neutrality on the ethical
question of abortion; public funding and state neutrality, (f) and (n), are
quite compatible.
B
Gutmann and Thompson claim that a failure to subsidize abortion violates
the basic liberty of indigent women: “the refusal to fund abortions
for poor women, when childbirth is funded, creates an almost irresistible
pressure on indigent women to carry a child to term.”25
This argument
does not, however, adequately support public funding of abortions. One
would not ordinarily think that it violates a person’s liberty to offer her
an almost irresistible incentive to choose one of her two rightful options.
If the state offered large tax incentives to encourage small families, it
would not, presumably, be violating the rights of those who would otherwise
prefer to procreate with abandon (although such a tax policy
might conceivably be excluded on other grounds). Whether or not subsidizing
only childbirth violates a woman’s rights depends on whether
justice entitles her to subsidized abortions in the first place. And this is
precisely the (reasonably controvertible) claim that is in question.26
An entitlement to subsidized abortions is much more directly supported
by straightforward egalitarian considerations than it is by considerations
of liberty, where egalitarianism can be loosely understood in
332 Philosophy & Public Affairs
own interests to the detriment of their maternal responsibilities. So it is quite possible to
think that abortion is profoundly wrong and that the ability to have an abortion is in
women’s interests. That said, there are also many pro-life proponents who think that the
emotional and psychological harms of abortion outweigh the benefits, so my claim is not
uncontroversial.
25. Gutmann and Thompson, Democracy and Disagreement, p. 88.
26. Alan Wertheimer notes the strangeness of Gutmann and Thompson’s argument
here in his “Internal Disagreements,” in Macedo, Deliberative Politics, pp. 170–83, at p. 176.
The latter are attempting to economize on deliberative disagreement, but the question of
a right to subsidized abortion is obviously one about which sincere and reasonable people
seeking mutually acceptable policies could disagree. As such, it is odd that Gutmann and
Thompson seem to insist on it as a fixed point, when they are not so insistent on other
points of deliberative contention. It could be that Gutmann and Thompson are simply
noting that Sher’s rejection of such a right is a reasonably controvertible position itself. This
could not support a rejection of Sher’s (h), however, since that proposal could still be the
best compromise position, all things considered, even though it is not without controversy.
Shifting to public funding would require additional arguments.
terms of John Rawls’s difference principle, i.e., the claim that inequalities
in social and economic primary goods are only just if they are to the
advantage of the worst-off.27
Egalitarian political parties must defend
those policies that will, as a matter of fact, most improve the quality of
life of the worst-off, and oppose those policies that will, as a matter of
fact, aggravate it. Opposition to the public funding of abortion in the
United States has had, on this measure, severe anti-egalitarian conse-
quences.28
The Hyde Amendment prohibits federal funding of abortion
through Medicaid, except in cases of rape, incest, or a threat to the
woman’s life.29
As of May 2005, only seventeen states provide Medicaid
subsidies for abortions beyond the Hyde Amendment’s restrictive limits,
mostly under court order.30
Accordingly, there are a number of difficulties
that face Medicaid recipients who wish to abort a pregnancy. When
public funding is unavailable, many women are simply unable to terminate
unwanted pregnancies.31
Those who do pay for abortions are often
333 Principled Compromise and
the Abortion Controversy
27. Rawls, p. 6.
28. I follow Sher and Gutmann and Thompson in focusing on the United States,
although I do not think that the basic philosophical issues in question depend on the contingencies
of the U.S. case. In particular, I assume that the egalitarian concerns I mention
also arise in other societies. A complete argument for public funding of abortion services
in any one society will need to balance the urgency of abortion services with the urgency
of other medical services, as well as take into account the level of resources available for
welfare provision in general. Naturally, not all societies will have the same priorities or the
same capacities. Moreover, I do not assume that a failure to fund abortion services is particularly
worse than other serious failures to protect the health and welfare of the worstoff.
The ability to control one’s reproductive destiny is an important element in human
well-being, but I do not assume any distinctive right to subsidized abortion that marks it
out as a special case, entirely unlike other aspects of healthcare. Egalitarian considerations
defeasibly support publicly subsidized healthcare across the board.
29. The Hyde Amendment to the Social Security Act came into effect in August 1977. In
1981, the Reagan Administration excised the rape and incest provisions, although they were
later restored by the Clinton Administration. In June 1980, the Supreme Court ruled in
Harris v. McRae that the federal government is under no constitutional obligation to fund
abortion services.
30. The Alan Guttmacher Institute, “State Funding of Abortion under Medicaid,” State
Policies in Brief, May 2005. The states are Hawaii, Maryland, New York, Washington (voluntarily),
Alaska, Arizona, California, Connecticut, Illinois, Massachusetts, Minnesota,
Montana, New Jersey, New Mexico, Oregon, Vermont, and West Virginia (under court
order).
31. A 1999 study of North Carolina’s intermittent provision of abortion funds found that
when funds were unavailable 37 percent of women who would otherwise have obtained
an abortion carried the pregnancy to term instead, as cited in Heather Boonstra and Adam
Sonfield, “Rights Without Access: Revisiting Public Funding of Abortion for Poor Women,”
The Guttmacher Report on Public Policy 3 (2000): 8–11, p. 10.
only able to do so with significant difficulty and delay.32
These delays in
turn increase the cost of abortion and can increase the medical risks and
legal complications involved.33
The lack of public funding does not affect
only women on Medicaid. Private insurance coverage for abortion is low,
with only 13 percent of abortions billed directly to private insurance
providers.34
The number of abortion facilities is declining steadily,
further retarding the capabilities of women to control their reproductive
destinies, especially those who experience considerable difficulty in
traveling to metropolitan areas.35
The desire to keep the U.S. government “out of the abortion business”
has also had some detrimental consequences for U.S. international population
assistance. The Mexico City gag rule prohibits U.S. assistance to
foreign organizations that either provide abortion, provide information
to pregnant women about legal abortion options, or advocate legalizing
abortion in their countries.36
In 1995, Congress cut funding for the U.S.
334 Philosophy & Public Affairs
32. Boonstra and Sonfield cite a 1983 Alan Guttmacher Institute report which found that
60 percent of Medicaid recipients reported serious hardship in paying for abortion (including
diverting money from rent, utilities and food, and even resorting to prostitution),
compared with 27 percent of non-Medicaid women (ibid.). The same study notes that
Medicaid-eligible women wait between two to three weeks longer than other women to
have an abortion because of these difficulties in obtaining funds.
33. In 2001 the mean charge for an abortion at an abortion clinic was $364 at 10 weeks,
$720 at 16 weeks, and $1290 at 20 weeks; see Stanley Henshaw and Lawrence Finer, “The
Accessibility of Abortion Services in the United State, 2001,” in Perspectives on Sexual and
Reproductive Health 35 (2003): 16–24, p. 18.
34. Ibid., p. 20. Four states (Idaho, Kentucky, Montana, and North Dakota) place restrictions
on private medical insurance coverage of abortion.
35. The number of abortion providers dropped 14 percent between 1992 and 1996,
including a 23 percent drop in the number of public hospitals performing abortions; see
Stanley Henshaw,“Abortion Incidence and Services in the United States, 1995–1996,” Family
Planning Perspectives 30 (1998): 263–70, p. 268. There was an 11 percent drop in abortion
providers between 1996 and 2000, and 34 percent of U.S. women live in those counties (87
percent of the total number of counties) with no abortion provider; see Lawrence Finer
and Stanley Henshaw, “Abortion Incidence and Services in the United States in 2000,” Perspectives
on Sexual and Reproductive Health 35 (2003): 6–15, p. 10. In 2001, 24 percent of
women having abortions in non-hospital facilities traveled more than fifty miles to do so,
a difficulty compounded by mandatory waiting periods and women’s own desire for confidential
counseling; see Henshaw and Finer, p. 18.
36. Susan Cohen, “Global Gag Rule: Exporting Antiabortion Ideology at the Expense of
American Values,” The Guttmacher Report on Public Policy 4 (2001): 1–3, p. 2. The name
refers to the 1984 Mexico City conference where the Reagan Administration first announced
the policy. It was rescinded by President Clinton on 22 January 1993 (the twentieth anniversary
of Roe), but restored by President Bush exactly eight years later.
Agency for International Development’s population assistance program
from $542 million to $356 million, in part as an attempt to pressure the
agency to accept the gag rule. It is estimated that restoring funding to
1995 levels would result in 4.3 million fewer unintended pregnancies, 2.2
million fewer abortions, 15,000 fewer deaths of women from pregnancyrelated
causes, and 92,000 fewer deaths of infants worldwide.37
Although
neither the gag rule nor the low funding of usaid are entailed by laws
prohibiting federal funding of abortion, they are political bedfellows.
Thus, President Bush has justified the re-imposition of the gag rule on
the grounds that “taxpayer funds should not be used to pay for abortions
or advocate or directly promote abortion.”38
Even if Gutmann and
Thompson are wrong to think that Sher’s Hyde-like (h) violates a basic
liberty, it still trivializes the urgency of public funding to describe its
absence as a merely utilitarian cost since this suggests that it is the sort
of cost a society may choose to incur without injustice. As an egalitarian,
Jane cannot accept that measures that significantly aggravate the
quality of life of the worst-off are just. Hence, although Sher professes a
willingness to supplement his compromise policy with additional
welfare provisions, he has not provided any reason to think that the antiegalitarian
consequences of (h) can be easily avoided by any realistic
alternative to (f). The state needs to be in the “abortion business” precisely
because it needs to be in the business of aggressively improving
the quality of life of the worst-off.
C
Public funding of abortion is warranted, at least defeasibly, by egalitarian
concerns about the quality of life of the worst-off, and it is compatible
335 Principled Compromise and
the Abortion Controversy
37. Susan Cohen, “Analysis Says Boost in Overseas Family Planning Aid Would Yield
Improved Health, Fewer Abortions,” The Guttmacher Report on Public Policy 3 (2000): 9–11,
p. 9, citing an analysis conducted by researchers from the Alan Guttmacher Institute, the
Futures Group International, Population Action International, and the Population Reference
Bureau.
38. Cohen, “Global Gag Rule,” p. 2. In July 2002, President Bush suspended the entire
$34 million U.S. contribution to the United Nations Population Fund (unfpa) on the uncorroborated
grounds that the unfpa’s program in China constituted a violation of the 1985
Kemp-Kasten amendment that prohibits support for any program of coercive abortion or
involuntary sterilization; see Susan Cohen, “Bush Bars unfpa Funding, Bucking Recommendations
of its Own Investigators,” The Guttmacher Report on Public Policy 5 (2002): 13.
with state neutrality on the ultimate ethics of abortion. The remaining
question concerns whether public funding for abortion should be drawn
only from the contributions of pro-choice supporters. Gutmann and
Thompson reject the exemption for conscientious objectors in (o)
because an unequal mandatory tax burden would be unfair to those who
pay more. If justice requires that a society adopt a certain policy then the
burden for implementing this policy falls on everyone. It does not fall
only on those who believe the policy is a good one. One would need to
discern a compelling reason for exempting people from this burden
before one could maintain that only some of the population ought to
shoulder it. Mere reasonable disagreement with the policy is not a good
enough reason for exemption. For instance, the wealthy are, generally
speaking, less inclined to support egalitarian measures than other
people. Much of this dissent may be quite reasonable since egalitarianism
is not self-evidently true. Nevertheless, it would be self-defeating for
an egalitarian government to exempt them from their fair contributions.
Gutmann and Thompson note one possible basis for exemption: if a
pro-life supporter’s “fellow citizens truly acknowledge the moral seriousness
of her views, they should find some way to reduce her complicity
in acts she regards as murder.”39
They claim that although the
argument for reducing complicity is powerful, it does not warrant an
exemption. But the reason the argument does not warrant an exemption
is because it is not cogent at all. Complicity in an activity is only really a
moral problem if that activity really is unethical. Merely believing it to
be immoral does not in itself ground a claim to special treatment.40
Since
336 Philosophy & Public Affairs
39. Gutmann and Thompson, Democracy and Disagreement, p. 89.
40. Gutmann and Thompson note that “although we permit conscientious refusal in
military service, we do not extend the permission to taxes” (ibid.). My argument does not
rule out conscientious objection to military service, and in fact helps make sense of the
distinction between the cases of conscription and taxation. Conscientious objection can
be justified by the actual injustice of the ends served by military conscription. For instance,
the justification for refusing to serve in the apartheid-era South African Defence Force
stemmed not from the fact that objectors reasonably believed that white male conscription
perpetuated an unjust system, but from the fact that they were right to believe this. It
is not the presence of conscience that is relevant here, but its veracity. In contrast, public
funding for abortion services, by hypothesis, does not serve unjust ends. Moreover, in cases
of justifiable military conscription, those objectors whose pacifist beliefs render them
emotionally and psychologically incapable of using a weapon may be appropriately
redirected to noncombat units even if this does not reduce their complicity in the overall
war effort. In this case, moral beliefs are only of extrinsic relevance; many factors may
(n) requires state neutrality on the ethical question of abortion, however,
the state cannot make the requisite assumption that abortion really is
unethical. Jane could only regard pro-life complicity as the basis for a
conscientious tax exemption if she had compelling reasons of correction
to relinquish (n). In the absence of some such argument for the state to
officially condemn abortion as unethical, there is no basis for Jane to
regard pro-life complicity in supporting abortion as of any more concern
than libertarian complicity in supporting public libraries.
Considerations of fairness also undermine (v), the policy that provides
funding for abortion through voluntary donations. Although the
extra amount paid by pro-choice supporters is given voluntarily, it is still
unfair to expect any one group of people to shoulder the entire burden
of a policy that is required for reasons of justice. A proponent of (v) could
argue that pro-choice supporters should be willing to pay the price of
their convictions. This is presumably true. But it has no bearing on
whether other people who happen not to share those convictions should
also contribute. What it is fair for a person to pay in taxation is not determined
by what it would be appropriate for the person to be willing to
pay, given their political convictions. At most, the argument implies that
pro-choice supporters should make their contributions with a smile and
that pro-life supporters may make theirs with a grumble.
Since (m), the policy in which pro-choice taxpayers direct a portion
of their total mandatory tax burden to abortion services, does not
involve any inequality in the total amount paid by pro-life and prochoice
supporters, it does not involve any substantive unfairness.
However, (m) is inadequately supported. First, Gutmann and Thompson
defend it as a fairer way to reduce pro-life complicity in abortion than
(o), the exemption for pro-life conscientious objectors. But I have argued
that there is no need to reduce pro-life complicity unless state neutrality
on the ethical question of abortion is rescinded. Second, (m) does not,
in any case, reduce pro-life complicity; it simply makes this complicity
indirect. If funds from pro-choice tax returns are used to fund abortion,
337 Principled Compromise and
the Abortion Controversy
incapacitate soldiers, and not every conscript with sincere moral objections to war is incapable
of serving in combat. There is no parallel with taxation, because nobody is psychologically
or emotionally incapable of paying tax (nor, for that matter, is any pharmacist
incapable of providing women with “morning-after” contraception). I develop this position
on the relative insignificance of conscience in a work-in-progress paper.
instead of what they would otherwise fund, then pro-life tax returns
must make up the shortfall. This means that pro-life supporters would
effectively be subsidizing pro-choice supporters so that the latter can
subsidize abortion. As such, (m) is, in essence, a somewhat convoluted
yet entirely cosmetic modification of the original public funding policy
(f). It therefore cannot be the basis of a reasonable moral compromise
on public funding between pro-life and pro-choice protagonists.41
IV. Compromise on Abortion
Reasons of egalitarian justice support some form of public funding of
abortion, and reasons of fairness support placing the burden of this
policy on pro-choice and pro-life proponents equally. Considerations of
state neutrality and moral complicity do not undermine this position.
However, if a strong enough case for some form of principled compromise
on abortion can be developed, Jane may have independent reasons
to modify (f) that override these considerations of justice and fairness.
Alternatively, Jane may have good principled reason to accept a moral
compromise on some other component of (A), if public funding is seen
as non-negotiable. At least four arguments for some sort of principled
compromise on abortion can be distinguished, namely, arguments from
complexity, respect, accommodation, and reciprocity.
A
I assumed above that Jane is correct to endorse (A). This may be a permissible
assumption for the purposes of making an independent philosophical
point about reasons for moral compromise, but it is not an
assumption that Jane herself can be certain of in the day-to-day deliberations
of her political life. It would be dogmatic of Jane not to recognize
the reasonableness of opposing views. Since Jane cannot insist on
the correctness of her own view, she “might rightly compromise because
of the moral complexity” of the controversy.42
Principled compromise
338 Philosophy & Public Affairs
41. The Presidential Campaign Fund check-off box is itself not without controversy.
Opponents of public funding of political campaigns quite correctly reject the idea that (m)type
devices constitute voluntary donations from a subset of taxpayers.
42. Dobel, p. 86.
may be the best way to acknowledge our fallibility and the limitations of
our ability to discern moral truth.
Moral complexity, however, constitutes no reason for moral compromise.
A policy that splits the moral difference between two opposing yet
reasonable viewpoints need be no more self-evident than the viewpoints
themselves. If Jane cannot know that (A) really is, all things considered,
the best policy on abortion, neither can proponents of principled compromise
know that their favored compromise policy really is, all things
considered, the best reaction to the abortion controversy. Fallibility is a
shortcoming that affects all reasonably controvertible positions on abortion,
whether uncompromised initial positions or moderated compromises.
Proponents of a principled compromise on abortion require
additional reasons why Jane, after acknowledging that she might be
wrong about (A), ought to modify it for another policy that is just as
uncertain and reasonably controvertible. A moral compromise involves
some moral loss without any offsetting gain in epistemic reassurance.
The initial plausibility of the argument from complexity may stem
from an ambiguity in the notion of insisting on a position. In the first
sense, to insist on a position is to deny that there may be good first-order
reasons for moral correction against it. In the second sense, to insist on
a position is to refuse to compromise it, given that one has come to the
defeasible conclusion that it is better than any other position. Insistence
in the first, dogmatic sense is unreasonable in the context of reasonable
deliberative disagreements. This does not mean, however, that it is
unreasonable for Jane to be insistent in the second, intransigent sense.
This crucial distinction is often ignored, so it is worth articulating in a
different way. Each participant in public deliberation ought to appreciate
the complexity and variety of considerations that are relevant to an
issue such as abortion policy. Each participant ought to endeavor to
bring these different considerations into reflective equilibrium so that
the position she advances can justifiably be regarded as the best policy,
all things considered. In the context of reasonable moral controversies,
no person will be justified in adhering to her position without some
measure of doubt, since her fellows may always raise relevant considerations—reasons
of correction—that demonstrate that she has not, after
all, attained true reflective equilibrium. But this doubt is normatively
inert. Unless they actually do raise considerations that demonstrate this,
she has no reason to think that she has not, in actual fact, formulated
339 Principled Compromise and
the Abortion Controversy
the best policy. That a political agent must admit the possibility of
being wrong does not give her a reason to modify her position unless
that doubt is accompanied by independent reasons that show she
actually is wrong.
Moreover, moral complexity does not generate a principled reason for
moral compromise even in those cases where the agent is not at all sure
that her position is the best one to hold, but where the exigencies of
political life require her to adopt some or other position. However weakly
the position is supported, if it is the position that is best supported by
all the available first-order considerations, including considerations of
the various risks involved, then the agent has no basis to assume that
a compromise position is a safer bet. Naturally, there is an increased
likelihood of being wrong in cases of greater uncertainty. But that is
a shortcoming endemic to human fallibility for which principled
compromise can be no remedy.
B
The second argument in favor of principled compromise on abortion
supposes that giving one’s reasonable opponents the respect they
deserve requires a willingness to appreciate their point of view and
adjust one’s positions accordingly. A respect-based approach to grappling
with the abortion controversy would move beyond disrespectful
acrimony and hostile incivility towards a satisfactory moral compromise
between pro-life and pro-choice positions. Thus, if Jane is to respect her
political opponents, she should not insist that (A) become official policy
without modification. A satisfactory compromise on abortion would
allow all to feel that they have been given their due regard.
Following Stephen Darwall, we can distinguish between two senses in
which a person can be respected.43
The first sense, recognition respect,
concerns the respect we ought to have for all human beings, and perhaps
some other creatures, as beings with intrinsic moral status. To imprison
a person without trial is to fail to respect her, since it is to fail to recognize
that she is a creature who matters from the moral point of view. Yet
this sort of respect has no bearing on principled compromise. To fail to
respect a person in this sense is to treat her unjustly. If a public policy
340 Philosophy & Public Affairs
43. Stephen Darwall, “Two Kinds of Respect,” Ethics 88 (1977): 36–49, esp. pp. 38–39.
were to be unjust in this way, it would have to be corrected rather than
compromised. For instance, one argument for state neutrality about
conceptions of the good life supposes that a perfectionist state would
fail to respect those who reasonably reject the government’s favored
conception, even if they were left free to pursue their own values
and commitments.44
If this argument is cogent, then state neutrality
will be superior, on its own merits, to the establishment of any one
perfectionist doctrine. If perfectionism unjustly disrespects those who
adhere to dissident conceptions of the good, it must be discarded and
not simply compromised.
The second sense, appraisal respect, is more discerning. To respect a
person in this sense is to have a positive attitude towards her above and
beyond mere respect for her as a person. This sort of respect is more
fragile than the first. One can lose respect for a person, or lose it in one
regard, after finding out more about her ulterior motives in performing
a good deed, her despicable beliefs about racial superiority, or her allround
bad character. This is the sense in play when it is argued that the
value of respect generates principled reasons for moral compromise.
Thus, Gutmann and Thompson write that mutual respect “requires a
favorable attitude toward, and constructive engagement with, the
persons with whom one disagrees [and] consists in an excellence of
character that permits a democracy to flourish in the face of fundamental
moral disagreement.”45
Having a positive appreciation for the reasonableness
of one’s political opponents may not mean that the content
of one’s own positions should be corrected, but it could mean that one
should be willing to forge a moral compromise with them. Whereas
recognition respect for individuals affects what the content of our political
positions should be in the first place, appraisal respect for the reasonableness
of our political opponents could affect how we hold to these
positions in democratic decision-making procedures.
It is certainly true that no side of the abortion controversy enjoys a
monopoly on moral respectability. Jane ought to respect the moral sensitivity
and profound moral conviction of those with whom she disagrees
about (A). Until such time as she is presented with good reasons for
341 Principled Compromise and
the Abortion Controversy
44. Larmore, chap. 3.
45. Gutmann and Thompson, Democracy and Disagreement, p. 79. Gutmann and
Thompson refer to Darwall’s concept of appraisal respect in a note to the quoted text.
moral correction to revise (A) on its own merits, Jane ought to regard
these people (with all due respect) as in error but not at fault. It is one
thing to say that Jane ought to respect those with opposing yet reasonable
views about abortion, however, and quite another to say that on
that basis she ought to compromise with them. There is a lacuna in the
argument here. It can often be unpleasant to disagree with those whom
one esteems and respects, and having a positive regard for others
can often incline one towards agreeing with them. Yet these psychological
tendencies are just facts about people, and provide no normative
reason for moral compromise. There is no reason to think that
Jane’s intransigent opposition to compromise on (A) must represent a
failure to respect her opponents or any desire for the state to officially
denigrate pro-life citizens.
One argument designed to bridge the gap between respect and compromise
claims that it is insufficient to simply respect those with whom
one disagrees. In addition, one needs to make this respect manifest.
Respect, like justice, must be seen to be had. One expresses respect for
others by being willing to find mutually acceptable compromises with
them. A willingness to meet one’s reasonable opponents halfway, or
thereabouts, sends a message that would otherwise be lost. But there are
many ways to express respect without resorting to compromise. Jane can
take the arguments of her pro-life opponents seriously and take time to
respond to them appropriately. She can stick to criticisms of their arguments
and eschew insulting ad hominem attacks. She can even form
working relationships and alliances with regard to other matters. Yet
none of this provides any principled reason for moral compromise.
Jane’s government can refrain from disrespecting pro-life citizens by not
deriding or ridiculing their arguments in official state policy on abortion.
Jane herself can express her respect for reasonable pro-life protagonists
simply by being respectful towards them.
C
The third argument for principled compromise invokes the values of
accommodation and inclusiveness. Since a democratic government
should be representative of all citizens, and not simply its partisan supporters,
Jane’s party should accommodate pro-life citizens by including
some of their views in a workable moral compromise on abortion. (A) is
342 Philosophy & Public Affairs
an insufficiently inclusive state policy since it fails to accommodate
pro-life views. The argument is correct insofar as equal participation
in a democratic process is not sufficient for a thoroughly democratic
public culture, even if the process is entirely legitimate; political legitimacy
is not all of democracy. Something would seem to have gone
wrong with a society’s democratic culture, even if not its institutions, if
a pro-life minority were to be treated dismissively as an evangelical
nuisance, rather than fellow citizens with interests, perspectives, and
ideals of their own.
Nevertheless, as with respect, the argument falls short of providing
principled reasons for moral compromise. There are many ways to
accommodate people in the development of public policy that fall
between formal participation and full-blooded moral compromise.
Efforts can be made to incorporate each party into policy deliberations
and other political processes. Political parties can be open about how
they caucus instead of secretly relegating all effective decision making
to anonymous politburos. In addition, individuals can engage with each
others’ views, try to construct more persuasive arguments, and be willing
to adapt their views when they are not, after all, supported by the best
balance of reasons for correction. This sort of interaction demonstrates
that it is possible for Jane to treat those with whom she disagrees as
fellow citizens engaged in a common deliberative project without resorting
to manufacturing compromises.
Moreover, it is no criterion of a democratic state policy that it be
eclectically representative of the reasonable views of all of its citizens.
Accommodation in the sense concerning the inclusion of citizens in
democratic practices should not be confused with accommodation in
the sense of compromise between their political positions.46
If Jane’s
political party has been democratically elected on the basis of a manifesto
that includes (A), it already has sufficient democratic mandate to
implement (A) in its entirety. An official state policy is not more democratic
because it emerges as a moral compromise between opposing
parties if this also means that it diverges from the fair and just policy
that a majority has voted for after public deliberations in which all
parties were accommodated.
343 Principled Compromise and
the Abortion Controversy
46. Wertheimer, p. 175.
D
My claim is that Jane’s party should only compromise on (A) if there are
pragmatic reasons to do so. Beyond such pragmatism, Jane has no principled
reason to seek to economize on moral disagreement between her
party and its pro-life opponents. As a consequence, we can expect there
to be just as much controversy about abortion (or perhaps even more)
after (A) is implemented as before. Gutmann and Thompson claim that
their view “can deal with moral disagreement” better than this.47
This
assumes, however, that moral disagreement is a problem that must be
dealt with in the first place, independently of the pragmatic difficulties
it creates. Gutmann and Thompson defend this pivotal assumption by
invoking the idea of reciprocity. Because terms of cooperation are
binding on all, they should be acceptable to all. This means that citizens
should practice a reciprocal give-and-take of reasons: “you make your
claims on terms that I can accept in principle [and] I make my claims on
terms you can accept in principle.”48
As such, if a position on abortion
is not morally acceptable to all reasonable people motivated to find
mutually acceptable terms of cooperation, the value of reciprocity provides
a principled reason to seek out an alternative that is. Since (A) is
not morally acceptable to pro-life proponents, Jane has a principled
reason to seek a moral compromise, as long as any such compromise is
morally acceptable to her. If there is no such mutually acceptable compromise,
citizens should endeavor to find the best approximation. A
principled reason for moral compromise could still be a good prima facie
reason, even if it is not ultimately conclusive. Yet my conjecture is that
there are no good principled reasons for moral compromise at all.
The key premise of this argument is that reciprocity requires mutual
acceptability. I have argued that the values of respect and accommodation
can be realized without any form of moral compromise between
opposing reasonable viewpoints, so this premise must be supported by
some other value. Gutmann and Thompson tend to equate mutual
acceptability with fairness: their “deliberative perspective does not
address people who reject the aim of finding fair terms for social cooperation;
it cannot reach those who refuse to press their public claims
344 Philosophy & Public Affairs
47. Gutmann and Thompson, Democracy and Disagreement, p. 54.
48. Ibid., p. 55.
in terms accessible to their fellow citizens.”49
The cooperative values of
fairness and reciprocity are certainly closely connected. For instance,
in Rawls’s justice as fairness, society is conceived of as a fair system of
cooperation. This cooperative conception of society involves an idea of
reciprocity or mutuality: “all who do their part as the recognized rules
require are to benefit as specified by a public and agreed-upon stan-
dard.”50
Rawls uses this notion of reciprocity to ground the egalitarian
difference principle.51
The appropriate standard of mutual benefit is that
which would be agreed to in an original position where parties are ignorant
of their socioeconomic status and ethical beliefs, and motivated
only by their holdings of primary goods.52
This account of fairness and
reciprocity, however, has no bearing on the moral disagreement that
divides pro-life and pro-choice supporters. Mutual benefit, according to
the principles of justice agreed to in the original position, is quite different
from mutual acceptability, according to the reasonable ethical
doctrines that each person is assumed to have. (A) can be the best policy
from the point of view of fairness and reciprocity precisely because it
is implied by, or most in accord with, the principles of justice chosen
in the original position (as seems plausible), even if this standard of
justice proves morally unacceptable to some reasonable people (as
seems the case).
In Political Liberalism, Rawls presents a somewhat different understanding
of fairness and reciprocity, one which involves the idea of
mutual acceptability in addition to the more familiar idea of mutual
benefit. The “criterion of reciprocity” states that for terms of social cooperation
to be fair, “citizens offering them must reasonably think that
those citizens to whom such terms are offered might also reasonably
accept them.”53
Since any just democracy will be characterized by a reasonable
pluralism of comprehensive moral doctrines, Rawls argues that
principles of justice ought to be articulated and defended in terms of
political values, because these are values that all reasonable people can
345 Principled Compromise and
the Abortion Controversy
49. Ibid. See also pp. 52–53 and p. 73.
50. John Rawls, Justice as Fairness: A Restatement (Cambridge, Mass.: Harvard University
Press, 2001), p. 6.
51. Ibid., p. 122ff.
52. Ibid., p. 14ff.
53. Rawls, Political Liberalism, p. xliv; see also pp. 49–50.
accept whatever their broader ethical convictions. Although there are
reasons to doubt the wisdom of this broadened understanding of reciprocity
(and the more general shift to political liberalism), Rawls’s
endorsement of mutual acceptability does not support the case for principled
compromise on abortion. First, as I mentioned above, although
political liberalism involves an intrinsic sensitivity to moral disagreement,
it does not involve any principled reasons for moral compromise.
Reasonable moral pluralism has significance for the first order of political
theorizing, insofar as it generates reasons for moral correction
against any theory of justice articulated in terms of comprehensive
values. Revamping justice as fairness as a political conception does not
involve any sort of compromise of a morally superior Kantian theory of
justice, for either principled or pragmatic reasons. Political liberalism is
supposed to be as good as it gets.
Second, (A) is quite compatible with political liberalism, and its attendant
idea of public reason, because it does not involve any commitment
to any comprehensive moral doctrine. The right protected in (r) and the
equality promoted by (f) are both political values, while (n) requires state
neutrality on the deeper question of the ultimate ethics of abortion. Similarly,
the rejection of waiting periods in (w) and the assertion of judicial
authority in (j) can be defended without any appeal to comprehensive
moral doctrines.54
On the political liberal view, (A) satisfies the criterion
of reciprocity because it can be accepted by all reasonable people
without requiring any alteration of their comprehensive moral doctrines.
As such, even if reciprocity should be cashed out in terms of
mutual acceptability in addition to mutual benefit, we have no reason
to think that this grounds any principled reason to compromise an abortion
policy such as (A).
346 Philosophy & Public Affairs
54. Saying that (A) is compatible with political liberalism should not be read as any
endorsement of political liberalism. Nothing in the way I have presented (A) or defended
it from compromise should be read to suggest that it is in any way a good policy because
of the existence of reasonable moral pluralism. In particular, the requirement of state neutrality
in (n) can be defended without appeal to the fact of reasonable moral disagreement
about the ethics of abortion. The state simply has no interest in making any judgment on
this point, once the right to choose is enshrined. State advocacy of a particular position on
the ethics of abortion may undermine the ability of women to exercise the right to choose
autonomously.
V. Principled Compromise
Principled compromise on abortion is not supported by the fact that the
abortion controversy is an instance of reasonable or deliberative moral
disagreement. An egalitarian pro-choice politician can properly respect
her pro-life opponents and accommodate them in democratic decisionmaking
processes without seeking any form of moral compromise or
accommodation of their convictions. In addition, the cooperative values
of fairness and reciprocity do not obviously require that political positions
be mutually acceptable to all reasonable people, at least in any way
that raises difficulties for an egalitarian pro-choice position such as (A).
I have not defended the substantive content of (A), except to note that
egalitarian considerations provide defeasible reasons of justice for the
public funding of abortion. Naturally, many arguments can be developed
both for and against the provisions of (A). Although I assume that
something like (A) would emerge undefeated from this first-order contestation
of arguments, my only concern in this article is to support the
claim that there are no principled reasons to compromise the policy that
does appear best on its own merits, whatever that might be.
In one respect, the abortion controversy seems a good candidate for
moral compromise because compelling moral arguments can be constructed
on both sides. In another respect, however, it is a poor candidate
for moral compromise because the most crucial point of contention
is a discrete question: either a woman has the right to abort a pregnancy
in her first trimester simply because she does not wish to carry it to term,
or she does not. This question of right must be decided one way or
another; there is no via media here. However, my arguments against the
view that we have principled reasons to compromise on abortion are not
premised on this difficulty.The four arguments from complexity, respect,
accommodation, and reciprocity do not support principled compromise
on any question. The criticism of these arguments in Section IV does not
rest on any point specific to the abortion controversy. If the moral complexity
of the abortion controversy does not give rise to a principled
reason for compromise, then it is quite likely that the complexity of other
moral controversies does not either. Similarly, if it is possible to respect
and accommodate pro-life advocates properly without resorting to
moral compromise, it is unclear why these values would have a different
347 Principled Compromise and
the Abortion Controversy
significance in other questions. Last, if a political position, such as (A),
really is the fairest way to distribute benefits and burdens then it is not
at all obvious why the same values of fairness and reciprocity should
imply that it ought to be moderated, even if only minimally, in a moral
compromise between opposing camps. Presumably, other arguments
in support of principled compromise can be developed. But in their
absence, the simple fact that compromise involves some moral loss,
however small, stands as an undefeated reason against those moral compromises
that are not pragmatically necessary.
To focus on principled rather than pragmatic reasons for compromise,
I have assumed throughout that Jane’s party enjoys a majority in
the legislature. But Jane is a fiction. In this article, I have used abortion
policy in the United States as an example. At present, political power in
the United States is exercised by individuals who are deeply antithetical
to a woman’s right to choose, who promote a backlash against the gains
made in women’s equality in preceding decades, who are hostile to the
notion that justice requires assistance to the worst-off, and who are, generally
speaking, in no mood for compromise. It would be far better if they
thought they had some principled reason to accept a moral compromise
on abortion that preserved Roe, even if it also preserved the Hyde
Amendment or waiting periods. This gives progressives in the here-andnow
of the real world some reason to keep talking about mutual respect,
moral accommodation, and principled compromise, even if no such
reason arises for imaginary pro-choice egalitarian politicians who enjoy
fairy-tale electoral landslides. Nevertheless, the existence of a pragmatic
reason to speak well of principled compromise does not imply the existence
of any principled reason for moral compromise itself. Talk of principled
compromise may be good agitprop, but it would be somewhat
ironic if the best argument for principled compromise did not recognize
any principled moral constraint against dissemblance. If (A) really is the
position on abortion best supported by philosophical arguments, then
there are good moral reasons to correct any public policy that contradicts
it. These are not arguments that can be expected to convince every
reasonable person. But supporting (A) is about as far as philosophical
considerations go. The rest is politics.
348 Philosophy & Public Affairs