88 Yale LJ950
88 Yale LJ950
88 Yale LJ950
Citations:
Please note: citations are provided as a general guideline. Users should consult their preferred
citation format's style manual for proper citation formatting.
-- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and
Conditions of the license agreement available at
https://heinonline.org/HOL/License
-- The search text of this PDF is generated from uncorrected OCR text.
-- To obtain permission to use this article beyond the scope of your license, please use:
Copyright Information
Bargaining in the Shadow of the Law:
The Case of Divorce*
950
Divorce and Dispute Resolution
951
The Yale Law Journal Vol. 88: 950, 1979
In this article we first examine the degree to which the law today
authorizes private ordering at the time of divorce: to what extent
can divorcing spouses create their own legally enforceable commit-
ments? In this context, we will also explain why we think the legal
system should provide divorcing couples broad power to resolve the
various questions that arise. Second, we develop a simple bargaining
model to suggest how the legal system affects negotiations between
spouses and their representatives at the time of divorce. Finally, we
apply this framework to several issues that have dominated much of
the academic discussion concerning family law during recent years:
(1) the advantages and disadvantages of discretion-conferring legal
standards for child custody; (2) Goldstein, Freud, and Solnit's pro-
posed visitation standard; (3) the role of lawyers in the divorce
process; and (4) the role of courts in "undisputed" divorces.
of resources during marriage; and whether and when to divorce. These effects, however,
seem more speculative and remote. Many believe that people decide to marry and raise
children without any consideration of the legal standards governing divorce dispositions.
At any rate, given the present state of knowledge, both theoretical and empirical, con-
cerning the effects of legal rules on behavior, this article does not attempt to trace out
more general, longrun effects. An interesting comparison might be made with the growing
discussion of the effect of economics on family decisions in, for example, ECONOMICS OF
THE FAMILY (T. Schultz ed. 1974).
5. The fact that a continuum exists can best be seen if one considers two extremes. At
one extreme would be a society that prohibited married couples from determining for
themselves the circumstances that permit divorce. The restriction could take the form of
an absolute prohibition on divorce, or it might require an inquiry by a state official to
determine whether narrowly defined "grounds" for divorce had been met. After a divorce,
the state might assert a continuing regulatory power over child care and over transfers of
resources between the spouses. At the opposite extreme would be a society that allowed a
substantial degree of private ordering. Cf. R. NozICK, ANARCHY, STATE, AND UTOPIA (1974)
(discussing minimalist, "night-watchman" state). Marriage could be dissolved at the
request of either spouse. A divorcing couple would then be free to make any mutually
agreeable deal governing the division of their property, their respective child-rearing
responsibilities, and their future economic claims on each other. The state would provide
a court system to enforce this deal and to resolve any subsequent disputes. The state
would also stand ready to impose a division of the couple's resources and responsibilities;
but only if the spouses themselves were unable to reach a mutually acceptable bargain.
Between these two extremes lie a variety of intermediate measures by which the state
can control or influence the process of dissolution. These measures can have varying
degrees of intrusiveness on the power of the parties to strike their own bargain. For
example, certain promises-e.g., never to request child support-might be made unen-
forceable. Certain minimum conditions-e.g., that each spouse receive a certain minimum
952
Divorce and Dispute Resolution
953
The Yale Law Journal Vol. 88: 950, 1979
954
Divorce and Dispute Resolution
18. See, e.g., Leigh v. Aiken, 54 Ala. App. 620, 623, 311 So. 2d 444, 447 (1975); Sheets
v. Sheets, 22 A.D.2d 176, 178, 254 N.Y.S.2d 320, 323 (1964).
19. See, e.g., UNIFORM MARRIAGE AND DIVORCE Aar § 306(f) (enacted in five jurisdictions).
20. See, e.g., Stewart v. Stewart, 130 Cal. App. 2d 186, 193, 278 P.2d 441, 445 (1955)
(parental contract as to custody is "binding upon them," but cannot interfere with "that
wide discretionary power given to courts in the disposition of the custody of children, in
accord with their best interests, or independently of the desire of a parent"); Kritzik v.
Kritzik, 21 Wis. 2d 442, 448, 124 N.V.2d 581, 585 (1963).
21. See H. CLARK, supra note 6, at 498-99, 598-99 (child support and custody); R.
MNOOKIN, CHILD, FAMILY AND STATE 197 (1978) (child support).
22. The study found that, in uncontested divorce cases involving children, the court
adjourned for further information or negotiation in less than 10% of cases; that a
welfare report was available in only 8.2% of uncontested cases; and that the courts
changed the child's residence in only 0.6% of uncontested cases. J. Eekelaar & E. Clive,
supra note 3, at 65-66.
23. See J. DESPERT, CHILDREN OF DIVORCE 189 (1953) (ABA report in 1948 that 85% of
divorced parents reach agreement on custody); Hansen, The Role and Rights of Children
in Divorce Actions, 6 J. FAm. L. 1, 2 (1966) (90%).
955
The Yale Law Journal Vol. 88: 950, 1979
24. See Foster & Freed, Child Custody (pt. 1), 39 N.Y.U. L. REv. 423, 423 (1964);
Mnookin, Child-Custody Adjudication: Judicial Functions in the Face of Indeterminacy,
LAW & CONTEMP. PROB., Summer 1975, at 226, 249-54; Oster, Custody Proceeding: A Study
of Vague and Indeterminate Standards,5 J. FAM. L. 21, 23-25 (1965).
25. See Mnookin, supra note 24, at 229 (in custody, dispute settlement involves
"[choosing] between two or more private individuals, each of whom claims an associational
interest with the child"; child protection "involves the judicial enforcement of standards
of parental behavior believed necessary to protect the child").
26. R. MNOOKIN, supra note 21, at 628. A study of Connecticut courts revealed that the
average amount of time for a court hearing on an uncontested divorce was four minutes.
Project, The Unauthorized Practice of Law and Pro Se Divorce: An Empirical Analysis,
86 YALE L.J. 104, 127 (1976) [hereinafter cited as Pro Se Divorce].
27. See Note, Lawyering for the Child: Principles of Representation in Custody and
Visitation Disputes Arising from Divorce, 87 YALE L.J. 1126, 1131-32 & nn.19-24 (1978)
(citing sources) [hereinafter cited as Lawyering for the Child].
28. In Connecticut and New Jersey, for example, an uncontested divorce case can be
immediately scheduled for hearing as soon as statutory waiting periods have passed and
processing by the court administrators has been completed. Contested cases, however,
generally wait at least a year from the time the parties are ready until a trial date can
be scheduled. Telephone Interview with Hon. Robert Berdon, Judge, Connecticut Superior
Court (Mar. 28, 1979); Telephone Interview with Hon. Harvey R. Sorkow, Presiding
Judge, Matrimonial Division (temporarily assigned), Superior Court of New Jersey, Bergen
County (Mar. 28, 1979) (Bergen County only).
956
Divorce and Dispute Resolution
erences of each spouse, and acceptable over time, than would a result
imposed by a court.
In divorces that involve no minor children, divorcing couples
should have very broad powers to make their own arrangements;
significant limitations are inconsistent with the premises of no-fault
divorce. After all, who can better evaluate the comparative advan-
tages of alternative arrangements than the parties themselves?2 9 Courts
should not, of course, enforce agreements that reflect fraud or over-
reaching. Nor do we wish to minimize the importance of appropriate
standards for alimony and marital property, for, as will be discussed,
these standards very much affect negotiated outcomes. 30 Nonetheless,
against a backdrop of fair standards, parties should be encouraged
to settle these economic issues for themselves. The state should pro-
vide an efficient and fair mechanism for enforcing such agreements
and for settling disputes when the parties are unable to agree.
When there are minor children, the state obviously has broader
interests than simple dispute settlement. The state also has a re-
sponsibility for child protection.31 To acknowledge this responsibility,
however, is not to define its limits. Indeed, the critical questions
concern the proper scope of the child-protection function at the time
of divorce and the mechanisms that best perform this function.
For reasons one of us has spelled out at length elsewhere, the actual
determination of what is in fact in a child's best interest is ordinarily
quite indeterminate. 3 2 It requires predictions beyond the capacity
of the behavioral sciences 33 and involves imposition of values about
which there is little consensus in our society. 34 Thus, the fundamental
question is: who gets to decide on behalf of the child? To what
extent should the child's parents be given the freedom to decide
between themselves how responsibility for their children is to be
allocated following divorce?
We believe divorcing parents should be given considerable freedom
to decide custody matters-subject only to the same minimum standards
for protecting the child from neglect and abuse that the state imposes
on all families. A negotiated resolution is desirable from the child's
29. Each spouse, in the words of John Stuart Mill, "is the person most interested in
his own well-being: . . . with respect to his own feelings and circumstances, the most
ordinary man or woman has means of knowledge immeasurably surpassing those that can
be possessed by any one else." J.S. MILL, On Liberty, in ON LIBERTY AND REPRESENTATIVE
GOVERNMENT 68 (R. McCallum ed. 1947) (1st ed. London 1859).
30. See pp. 968-70 infra.
31. Mnookin, supra note 24, at 229, 232.
32. Id. at 255-62.
33. Id. at 258-60.
34. Id. at 260-61.
957
The Yale Law Journal Vol. 88: 950, 1979
36. Professor Clark's hornbook on domestic relations, for example, devotes seven chap-
ters (over 300 pages) to divorce, alimony and property-division, child-support, and
custody issues. Only one chapter deals with separation agreements, and it emphasizes
problems of judicial enforcement and construction rather than the influence of the rules
on bargains. See H. CLARK, supra note 6, at 280-601. Law review articles in this field
typically focus on questions of the rules and procedures courts do or should employ in
adjudication, and they largely ignore the effect of the rules and standards on negotiations
outside of the courtroom.
37. Commentators typically discuss, for example, each doctrinal strand without ac-
knowledgment of its relationship to the others. See, e.g., Symposium, The Uniform
Marriage and Divorce Act: Strengths, Weaknesses, Alternatives, 18 S.D. L. Rxv. 531 (1973).
But see Watson, The Children of Armageddon: Problems of Custody Following Divorce,
21 SYRAcusE L. Rxv. 55, 59 (1969) (noting link between custody and money issues).
959
The Yale Law Journal Vol. 88: 950, 1979
38. See Registrars Study, supra note 16, at 32 (English study found fathers more willing
to pay child support than alimony).
39. On joint consumption, see J. HEAD, PUBLIC GOODS AND PUBLIC WELFARE 77-80, 167-
69, 176-79 (1974); J. HENDERSON & R. QUANDT, MicRoECoNoMIC THEORY 270-72 (2d ed.
1971); E. MALINVAUD, LEcTURs ON MICROECONO.tiC THEORY 211-18 (1972).
40. An economic loss arises because separation or divorce typically causes the loss of
certain economies of scale that are common within a household. Housing for four people
living together in a single household will often cost less than the same quality of housing
for two separate households composed of a single individual and three persons respectively.
Economies of scale may also arise because of medical and auto insurance costs. Further-
more, some durable goods (washing machines, freezers, etc.) may come in sizes that
are too large for a single user. Parents may easily share one car until dissolution. After
the divorce, two cars may become necessary, or one parent may be forced to use public
transportation.
960
Divorce and Dispute Resolution
Our legal and cultural norms reflect the notion that children should
not bear the economic loss, 41 and that, other things being equal, the
spouses should bear the loss equally. 42 Because joint consumption
implies that the custodial parent and children must essentially share
the same standard of living, a dilemma arises: either the children
must bear some part of the economic loss, or the noncustodial parent
must bear much more of the extra financial burdens imposed by
43
divorce than the custodial spouse.
Financial provision for a spouse terminates automatically on re-
marriage or death, while child-support normally ends when a child
reaches his majority or is emancipated. 44 The parents will know
how many years remain before a child will reach majority; however,
there may be considerable uncertainty about the probability that a
spouse who is receiving alimony will remarry. Therefore, the char-
acterization of the elements of payments will probably affect a party's
calculation of the risks concerning how long a payment will last and
41. The goal has been to set child support at a level allowing the children "'to be
brought up with as nearly as possible the same standard of opportunity as they would
have enjoyed had the marriage not failed."' R. LEVY, UNIFORM MARRIAGE AND DIVORCE
LEGISLATION: A PRELIMINARY ANALYSIS 205 (undated) (quoting PurrING ASUNDER: A
DIVORCE LAW FOR CONTEMPORARY SOCIETY (report of group appointed by Archbishop of
Canterbury to study reform of divorce law, Jan. 1964)); see H. CLARK, supra note 6, at
496 ("Within the limits of the husband's means, the child is entitled to an adequate
provision for his needs, one which reflects the income level and scale of living of the
family before the divorce.'); cf. Commonwealth ex rel. Kaplan v. Kaplan, 236 Pa. Super.
Ct. 26, 28, 344 A.2d 578, 579 (1975) (stating as "blackletter" law principle that "respon-
sibility of the parents, to support the child to the best of their ability, consistent with
their own station in life, is 'well nigh absolute' "); UNIFORM MARRIAGE AND DIVORCE ACT
§ 309(3) (enacted in five jurisdictions) (prescribing as one relevant factor in determination
of appropriate child-support payments "the standard of living the child would have
enjoyed had the marriage not been dissolved").
42. The husband and wife now occupy a position of equal partners. Whether the
marriage continues or is severed, the woman is as fully equipped as the man to earn
a living. The husband and wife share equal rights and obligations in the marriage
relationship and share equal burdens in the event the marriage is dissolved.
Spotts v. Spotts, 355 So. 2d 228, 230 (Fla. Dist. Ct. App.), cert. denied, 361 So. 2d 835 (Fla.
1978); cf. Orr v. Orr, 47 U.S.L.W. 4224 (U.S. Mar. 5, 1979) (statutory scheme imposing
alimony obligation on husbands but not wives violates equal protection clause).
43. Professor Areen suggests that the economic allocation between spouses at divorce
can be seen as governed by one of five conflicting principles: the fault principle, the need
principle, the rehabilitation principle, the status principle, and the contribution principle.
J. AREEN, FAMILY LAxw 634-35 (1978). With the exception of the status principle-which
would require spousal support at a level permitting the custodial parent to maintain his
or her prior economic status-each of the other principles said to underlie alimony is
different from the governing principle for child support. Consequently, to the extent there
is joint consumption, the dilemma would exist.
44. Some statutes provide for automatic termination of alimony upon remarriage
unless the parties provide otherwise; absent a statute, remarriage provides grounds for
modification. See H. CLARK, supra note 6, at 457-59. On child support, see R. MNOORIN,
supra note 21, at 192.
The Yale Law Journal Vol. 88: 950, 1979
962
Divorce and Dispute Resolution
a promise to reflect the perceived risks that the full amount may
never be received. Finally, the tax consequences of alternatives can
be evaluated and compared.
This ability to compare different packages has obvious implica-
tions for private bargaining, at least -when the couple has sufficient
economic resources. Sophisticated parties and their lawyers will at-
tempt to seek out circumstances in which a different characterization,
because of tax effects or differences in risk or time preferences of
the parties, can make both spouses better off. 50
2. Custody
The remaining element of the bargain concerns the custodial duties
and rights of the parents. By varying the time the child spends with
each parent, and by assigning particular child-rearing tasks to one
parent or the other, a divorce settlement may divide prerogatives
in many different ways. At the extreme, one parent may be entirely
responsible for the child all the time, with the other spouse spending
no time with the child. Or, divorcing parents may agree to share
child-rearing responsibilities equally after divorce through joint cus-
tody.5 1 For example, the child may live with each parent one-half of
the time, with the parents together deciding where and how the
child should be educated, who the pediatrician should be, etc. Be-
tween these extremes, many other alternatives are often possible.
963
The Yale Law Journal Vol. 88: 950, 1979
these two elements are inextricably linked for two reasons: over some
range of alternatives, each parent may be willing to exchange cus-
todial rights and obligations for income or wealth, and parents may
tie support duties to custodial prerogatives as a means of enforcing
their rights without resort to court.
Economic analysis suggests that a parent may, over some range,
trade custodial rights for money. Although this notion may offend
some, a contrary assertion would mean that a parent with full cus-
tody would accept no sum of money in exchange for slightly less
custody, even if the parent were extremely poor. Faced with such
alternatives, most parents would prefer to see the child a bit less and
be able to give the child better housing, more food, more education,
better health care, and some luxuries. Suggesting the possibility of
such trade-offs does not mean that the parent would be willing to
relinquish all time with the child for a sufficiently large sum of
money. Indeed, with a minimum level of resources, a parent may
have a parallel minimum of custodial rights for the reduction of
which no additional payment, however large, could be adequate
compensation.
The negotiating process itself provides many opportunities for the
parties to link money and custody issues. The most obvious oppor-
tunity exists in the context of enforcement of support or visitation.
The legal system does not permit these connections in most states:
in a suit brought to collect overdue support payments, a father can-
53
not defend on the ground that his ex-wife did not permit visitation.
Nor have courts permitted a custodial parent to cut off visitation
because of a failure to pay support.5 4 Nevertheless, it is often time-
different commodities. The bargain analysis would still apply because these commodities
are substitutes for each other and represent different forms that money transfers can take.
We have identified four dimensions along which alimony, child support, and marital
property transfers may differ: (I) time flow; (2) tax conseqences; (3) risks of nonpayment;
and (4) enforcement characteristics. Along each of these dimensions, every combination
of property, alimony, and child support might be rated. Our use of money as an index for
all three suggests simply that any combination can be reduced to a single valuation
number.
53. See, e.g., In re Dooley, 30 Or. App. 989, 569 P.2d 627 (1977) (custodial parent's
interference with visitation not in itself change of circumstances warranting reduction or
elimination of child support); H. CLARK, supra note 6, at 504 & n.59, 513 & n.63 (majority
of cases hold that child support must continue despite violation of custody decree as to
visitation). But see Hudson v. Hudson, 412 N.Y.S.2d 242 (Sup. Ct. 1978) (child-support
arrearages of father cancelled by court because of mother's interference Mith his visitation
rights); cf. N.Y. Dom. REL. LAw § 241 (McKinney Supp. 1978) (court may suspend alimony
payments or cancel arrears if custodial parent wrongfully interferes with or withholds
visitation); H. CLARK, supra note 6, at 538 & n.11 (many cases refuse recovery for alimony
if custodial parent violates provisions of separation agreement).
54. See, e.g., Johnson v. Johnson, 52 Ohio App. 2d 180, 368 N.E.2d 1273 (1977).
964
Divorce and Dispute Resolution
The links between support and visitation are nurtured not only
by the parties' power to take self-help measures, but also by impor-
tant cultural values. Many believe that support obligations and visi-
tation rights are inextricably tied together in terms of what it means
to be a parent. A father who fails to support his children, at least
when he has the financial capacity to do so, may in popular perception
no longer be entitled to maintain a relationship with his minor chil-
dren if the custodial mother objects. Similarly, a mother who pur-
posely prevents a father from maintaining his relationship with his
children after a divorce may be viewed as no longer entitled to his
support.
1. Parental Preferences
Parental preferences vary with regard to money and child-rearing
responsibilities. Ordinarily, economists assume that a person's tastes
for most goods and services are insatiable: no matter how much a
person has, he will see himself as better off with more. This is cer-
Divorce and Dispute Resolution
967
The Yale Law Journal Vol. 88: 950, 1979
59. See note 29 supra. A major role for lawyers in the divorce process is to help
clients clarify their own preferences. See p. 985 infra (lawyer's role as counselor); cf.
ABA CODE OF PROFESSIONAL RESPONSIBILITY EC 7-8 (1976) (duty of lawyer to ensure that
client is fully informed of all relevant considerations).
60. Obviously, one can get into substantial linguistic tangles here. See G. HARIAN,
THE NATURE OF MORALITY 137-39 (1977) (problem of differentiating self-interest from
desire to promote welfare of others).
Divorce and Dispute Resolution
61. This analysis rests on the assumption that a spouse will act in his own self-interest.
The conclusion allows for altruism or spite, provided that a spouse's preferences meet
certain requirements of consistency. In particular, each spouse must be able to rank every
possible combination of money/custody divisions in a way that is "transitive." In other
words, if a spouse prefers custody/money division A to division B, and prefers division B
to division C, then the spouse must also prefer division A to division C.
62. See Mnookin, supra note 24, at 236-37 & nn.45-47 (citing statutes and cases).
969
The Yale Law Journal Vol. 88: 950, 1979
970
Divorce and Dispute Resolution
4. Transaction Costs
Costs are involved in resolving the distributional consequences of
separation or divorce, and in securing the divorce itself. The trans-
action costs7 ' that the parties must bear may take many forms, some
financial and some emotional. The most obvious and tangible involve
the expenditure of money. Professional fees-particularly for lawyers-
must be paid by one or both parties.7 2 In addition, there are filing
fees and court costs. 7 3 More difficult to measure, but also important,
are the emotional and psychological costs involved in the dispute-set-
tlement process. Lawsuits generally are emotionally burdensome;7 4
the psychological costs imposed by bargaining (and still more by liti-
75
gation) are particularly acute in divorce.
The magnitude of these transaction costs, both actual and expected,
can influence negotiations and the outcome of bargaining. In the
dissolution process, one spouse, and that spouse's attorney, can sub-
stantially affect the magnitude of the transaction costs that must be
borne by the other spouse. As is generally the case, the party better
able to bear the transaction costs, whether financial or emotional,
will have an advantage in divorce bargaining.
In divorce, transaction costs will generally tend to be (1) higher
if there are minor children involved, because of the additional and
intensely emotional allocational issues to be determined; (2) an in-
creasing function of the amount of property and income the spouses
have, since it is rational to spend more on negotiation when the
possible rewards are higher;7 6 and (3) higher when there is a broad
range of possible outcomes in court.
5. Strategic Behavior
The actual bargain that is struck through negotiations-indeed,
whether a bargain is struck at all-depends on the negotiation process.
During this process, each party transmits information about his or
her own preferences to the other. This information may be accurate
or intentionally inaccurate; each party may promise, threaten, or
bluff.7 7 Parties may intentionally exaggerate their chances of winning
73. Generally, the award of costs to either party is within the discretion of the court.
See note 72 supra (citing sources). The Supreme Court has held that due process prohibits
the state from denying, solely on the basis of inability to pay, access to courts for in-
dividuals seeking dissolution of marriage. Boddie v. Connecticut, 401 U.S. 371, 380-82
(1971).
74. See Address by Learned Hand, Association of the Bar of the City of New York
(Nov. 17, 1921), excerpted in D. LOULSELL & G. HAZARD, PLEADING AND PROCEDURE 1295
(3d ed. 1973) ("[A]s a litigant I should dread a lawsuit beyond almost anything else short
of sickness and death.")
75. See, e.g., R. DEWOLF, THE BONDS OF ACRIMONY 39-55 (1970).
76. Cf. R. POSNER, supra note 71, at 436 ("other things being equal, the higher the
stakes in a case the more likely it is to be litigated"); Cooter & Kornhauser, Can Litiga-
tion Improve the Law Without the Help of Judges? (forthcoming). Cooter and Korn-
hauser suggest that the litigation rate may not be higher in disputes involving larger
sums. Transaction costs may be higher even if the litigation rates are not because transac-
tion costs include costs of settlement and attempted settlement. Id.
77. A father, for example, may pretend that he wants custody and is willing to litigate
the issue simply to intimidate a risk-averse mother into settling for less money. See Law-
Divorce and Dispute Resolution
yering for the Child, supra note 27, at 1131 n.21 (citing interview with Assistant Clerk of
New Haven Superior Court). The prevailing best interests standard exacerbates the dis-
advantages of a risk-averse parent because of its great uncertainty. See p. 979 infra.
78. The concept of strategic behavior is not without ambiguity. We use it to mean
behavior in which the parties misrepresent their own intentions, desires, or chances of
winning in order to obtain a strategic advantage in negotiation.
79. Eisenberg, supra note 1, at 638.
80. Id. at 638-39.
81. Cf. id. at 680-81 (interplay of evocation of norms and use of bargaining power).
973
The Yale Law Journal Vol. 88: 950, 1979
974
Divorce and Dispute Resolution
84. See Aubert, Courts and Conflict Resolution, 11 J. CONFLICT RESOLUTION 40, 44
(1967).
85. See p. 970 supra.
86. Child custody would have limited divisibility in jurisdictions in which joint custody
is forbidden or actively discouraged by courts, see, e.g., Rickard v. Rickard, 7 Wash.
App. 907, 503 P.2d 763 (1972), petition for review denied, 81 Wash. 2d 1012, 503 P.2d 763
(1973) (joint custody to be avoided if possible and ordered only in exceptional circum-
stances); Martin v. Martin, 132 S.W.2d 426 (Tex. Civ. App. 1939" (child cannot grow up
normally with split custody), or if the proposal of Beyond the Best Interests of the Child
were adopted, see pp. 980-83 infra.
87. If the parents live a great distance from each other, it may not be possible to
divide child-time between them in a manner reflecting their preferences.
88. This would be the case, for example, if there were one piece of marital property
so valuable, for sentimental or other reasons, that neither party could be compensated
for giving up his share by getting all the couple's other assets, and if neither party would
consent to its being auctioned off.
The Yale Law Journal Vol. 88: 950, 1979
89. In real cases, the parties do not know the probability distribution of the various
outcomes a court might impose. This injects a further complication in the process and
creates the possibility that one or both parties may overestimate their own chances of
winning. This may result in more cases being litigated than in a world where the out-
come is uncertain but the odds are known.
976
Divorce and Dispute Resolution
be litigated. When one party is a risk preferer and the other is risk-
averse, it is difficult to predict the effect on the rate of litigation. In
any negotiated outcome, a risk preferer will have an advantage over the
party who is risk-averse.
90. See, e.g., K. DAvis, DISCRETIONARY JusTIcE (1969); P. NONET, ADMINISTRATIVE JUsTICr
(1969); R. POSNER, supra note 71, at 424-25; Note, The Void-for-Vagueness Doctrine in the
Supreme Court, 109 U. PA. L. REv. 67 (1960).
91. See, e.g., R. LEVY, supra note 41, at 224-25 (recommending presumption that wife
be entitled to custody); Watson, supra note 37, at 82 (maternal preference for children
under 10, plus same-sex presumption for older children). See generally Mnookin, supra
note 24, at 135-36, 283-84.
92. See Mnookin, supra note 24, at 235-36.
93. See id. at 231-32, 235, 255-56.
94. Id. at 236.
The Yale Law Journal Vol. 88: 950, 1979
95. Traditionally, courts have been unwilling to approve joint, split, or divided custody
arrangements for small children. See, e.g., Utley v. Utley, 364 A.2d 1167 (D.C. 1976);
McLemore v. McLemore, 346 S.V.2d 722 (Ky. 1961); Martin v. Martin, 132 S.W.2d 426
(Tex. Civ. App. 1939). More recently, however, some jurisdictions have been willing to
approve joint-custody arrangements when both parents desire such. See, e.g., Childers v.
O'Neal, 251 Ark. 1097, 476 S.W.2d 799 (1972); Perotti v. Perotti, 78 Misc. 2d 131, 355
N.Y.S.2d 68 (1974) (relitigated with custody awarded to father in unreported decision
dated July 18, 1975; see Dodd v. Dodd, 93 Misc. 2d 641, 647, 403 N.Y.S.2d 401, 405 (1978)).
Some commentators today recommend joint custody as the superior alternative. See note
51 supra.
96. Commentators have suggested, for example, a trend "toward rehabilitative alimony
and away from permanent alimony." White & Stone, supra note 64, at 80. For a chart
summarizing changes in alimony laws, see Freed & Foster, supra note 13, at 309-10.
Divorce and Dispute Resolution
979
The Yale Law Journal Vol. 88: 950, 1979
99. See, e.g., Braiman v. Braiman, 44 N.Y.2d 584, 589-90, 378 N.E.2d 1019, 1021, 407
N.Y.S.2d 449, 451 (1978) ("[Joint custody is encouraged primarily as a voluntary alterna-
tive for relatively stable, amicable parents behaving in mature civilized fashion .... As
a court-ordered arrangement imposed upon already embattled and embittered parents,
accusing one another of serious vices and wrongs, it can only enhance familial chaos.")
(citations omitted).
980
Divorce and Dispute Resolution
Much ink has been spilled over this controversial proposal. 10 2 Many
behavioral scientists have suggested the importance from the child's
perspective of maintaining contact with the noncustodial parent. 1° 1
Lawyers have challenged the proposal on grounds of fairness. 104 But
the critics-like the authors themselves-have largely failed to consider
the possible effects of the proposed standard on private ordering.10
From our perspective it is useful to ask what power the parents
would have to make their own law with respect to custodial arrange-
ments. This question points up a peculiar inconsistency. The book
makes it clear that the parents would have the right and the power
to determine by agreement "who will be the custodial parent."'106
A court would determine the custodial parent only if there were a
982
Divorce and Dispute Resolution
110. This would mean not only an increase in litigated cases, but also an increase in
custody claims that are disingenuous. In the hypothetical example, the father might ask
the court for full custody only so that he could then cede much of the child-time back to
his wife without having to worry about trusting his former spouse.
111. See pp. 963-66 supra (inextricable link between child support and visitation).
The Yale Law Journal Vol. 88: 950, 1979
984
Divorce and Dispute Resolution
is a strong possibility the child will be harmed if he continues to live under the present
arrangement.") The standards will also vary in the time limits, if any, that are imposed.
Compare UNIFORM MARRIAGE AND DIVORCE Acr § 409(a) (enacted in five jurisdictions) (no
motion to modify custody may be made earlier than two years after date of initial decree)
with CONN. GEN. STAT. § 46b-56(a) (1979) (court may modify custody order "at any time").
115. Other transaction costs of litigating are those imposed by the legal system-e.g.,
filing fees, court fees, etc. A lawyer will also be a source of information on the available
strategies for inflicting emotional costs on the other party.
116. See Pro Se Divorce, supra note 26, at 123-29.
117. See R. NVEISS, MARITAL SEPARATION 263-66 (1975); Pro Se Divorce, supra note 26,
at 141-43; cf. Q. JOHNSTONE & D. HOPSON, LAVWYERS AND THEIR WORK 81 (1967) ("[o]ne of
the most common lawyer tasks is negotiation"); Eisenberg, supra note 1, at 661-62 (role
of "affiliates," who may be lawyers, in negotiation).
985
The Yale Law Journal Vol. 88: 950, 1979
118. See, e.g., R. EISLER, DISSOLUTION 40 (1977) ("legal system of battle where two ad-
versaries try to get the 'best possible deal' for their clients only tends to exaggerate, rather
than resolve, the emotional tensions of a divorce"); M. VANTON, MARRIAGE-GROUNDS FOR
DIVORCE 99-100 (1977) (accusing divorce lawyers of inflaming antagonisms, prolonging
negotiations, and doing nothing to attempt to save the marriage-all in order to obtain
large fees); cf. Watson, The Lawyer as Counselor, 5 J. FAMr. L. 7, 9, 11-20 (1965) (stressing
counseling role of lawyer and suggesting that lawyers develop interviewing skills grounded
in awareness of dynamics of situation). But see R. FELDER, DIvoRcE 1-2 (1971) (member
of divorce bar defending hired-gun approach: reason for which lawyer is hired is to "do
anything and everything . . . necessary . . . to gain [client] a divorce in which he will
come out financially, psychologically-in every way-on top"). A more positive appraisal
can be found in M. HUNT, supra note 3, at 219-21, which suggests that, although the
legal process is not designed to be a therapeutic mechanism, it generally does have somc
therapeutic aspects-e.g., some "people purge themselves of crippling feelings by the
haggling process," id. at 220.
119. I. ILLIC, MEDICAL NEMESIS 21-25 (1975).
986
Divorce and Dispute Resolution
berg has suggested that a pair of lawyers-each acting for his client-
may make the process of negotiation very much like adjudication,
in which "rules, precedents, and reasoned elaboration . . . may be
expected to determine outcomes."' 120 When each disputant is repre-
sented, the lawyers
987
The Yale Law Journal Vol. 88: 950, 1979
988
Divorce and Dispute Resolution
989
The Yale Law Journal Vol. 88: 950, 1979
990
Divorce and Dispute Resolution
137. See Llewellyn, Behind the Law of Divorce: II, 33 COLUM. L. REV. 249, 286 (1933);
Pro Se Divorce, supra note 26, at 165-66; Note, Untying the Knot: The Cause and Patterns
of Divorce Reform, 57 CORNELL L. REv. 649, 666-67 (1972).
138. The alternatives include mediation, see Mnookin, supra note 24, at 287-88, and
other less formal means of adjudication, id. at 289; see Coulson, Family Arbitration-An
Exercise in Sensitivity, 3 FAm. L.Q. 22 (1969).
139. See CAL. CIV. CODE §§ 4550-4556 (West Supp. 1979). This provision allows summary
dissolution in uncontested divorce only if (I) there are no minor children, (2) the mar-
riage is of not more than two years' duration when the petition is filed, (3) neither party
owns any real estate, (4) neither party has unpaid debts of more than $2000, excluding any
car loan, (5) there is no community property (excluding a car) of more than $5000, (6)
neither spouse has separate property of more than $5000, and (7) spousal support is
waived. Id. § 4550. At any time after six months from the filing of the joint petition for
summary dissolution, the court may, at the request of either party, enter final judgment
dissolving the marriage. Id. § 4553. Because of these severe limitations, only a tiny pro-
portion of divorcing couples in California will qualify for this new summary procedure.
Indeed, it would appear that the divorce bar was entirely successful in limiting the pro-
cedure to cases in which there would have been no potential in any event for a significant
legal fee.
991
The Yale Law Journal Vol. 88: 950, 1979
clog the family law court system and exact a heavy toll on divorcing
40
spouses in the form of delay.'
The requirement of a judicial proceeding in undisputed divorce
cases could easily be eliminated. Getting married does not require
judicial proceedings, so why should getting a divorce? Some countries
have eliminated the requirement that undisputed divorces go through
court, 141 and it therefore seems appropriate to examine the possible
1 42
justifications for the requirement.
1. Ceremonial Function
A judicial proceeding may serve a ceremonial function that re-
confirms, both for the divorcing parties and the general public, the
seriousness with which the state treats marriage and divorce. Rituals
are important, and the court proceeding can be seen as a socially
imposed divorce ritual. One may, however, ask how well the existing
requirement serves the ceremonial function. The marriage ceremony
is an important social ritual, but it can be extremely simple, and it
does not require lawyers and a judge. Moreover, in most states the
parties to a divorce are not usually required to appear in court, but
may simply appear through their lawyers. If the ritual were for the
benefit of the parties, presumably their presence would be required.
Thus, the requirement is more like a civil fine imposed on a divorcing
43
couple-a fine payable not to the treasury but to the divorce bar.1
140. An analogous problem from a negotiation perspective is the existence of significant
differences in waiting periods according to whether the divorce is consensual or not. These
differences may be created by statute, see, e.g., MD. ANN. CODE art. 16, § 24 (Supp. 1978)
(party abandoned may be granted divorce after 12 months; either party may obtain
divorce after separation for three years), or by problems of court administration and
docketing, see note 28 supra (in some jurisdictions, uncontested matters come to court
swiftly, but contested cases usually must wait over one year).
141. See Pro Se Divorce, suPra note 26, at 166 (registration coupled with waiting period
implemented in uncontested divorces in Denmark, Iceland, Japan, Norway, and Sweden).
142. The analysis that follows criticizes only the requirement that uncontested cases
be processed through courts. It is not meant to deny the importance of divorce as a legal
event in even the simplest and most amicable case. Eliminating the involvement of judges
does not necessarily mean eliminating the involvement of one or more lawyers. Divorce
severs some legal obligations and creates others-appropriate legal instruments must often
be drawn. It would seem advisable for all divorcing couples, even the most amicable, to
consult with someone on the legal consequences of divorce. Nevertheless, this does not
suggest that judicial involvement is also necessary. After all, most legal obligations we
agree to in life do not require a judge's approval.
143. Albert Hirschman has suggested that
[s]pecific institutional barriers to exit can often be justified on the ground that they
serve to stimulate voice in deteriorating, yet recuperable organizations which would
be prematurely destroyed through free exit. This seems the most valid, though often
not directly intended, reason for the complication of divorce procedures and for the
expenditure of time, money, and nerves that they necessitate.
A. HIRSCHMAN, EXIT, VOICE, AND LOYALTY 79 (1970).
The argument that making divorce easier might lead to the termination of salvageable
992
Divorce and Dispute Resolution
marriages was often invoked by those who unsuccessfully opposed the no-fault revolution.
The use of costly state-imposed procedures to inhibit divorce seems inconsistent with that
revolution; it might also offend, at least for the poor, the constitutional requirements of
due process. See Boddie v. Connecticut, 401 U.S. 371 (1971). There is, of course, no
empirical evidence demonstrating that procedural simplification in uncontested cases
would substantially increase the number of "hasty" or "unwise" divorces. If this turned
out to be the case, and it was thought desirable to do something about it, then longer
waiting periods or even mandatory marriage counseling might be made preconditions of
divorce.
The Yale Law Journal Vol. 88: 950, 1979
4. Child Protection
When a divorcing couple has minor children, the state has an ad-
ditional interest in child protection. The requirement of court review
of private agreements relating to custody and child support might
be justified on this ground: it may improve the quality of negoti-
ated agreements from the child's perspective. Some parents might
otherwise engage in divorce bargaining on the basis of preferences
that narrowly reflect selfish interests, rather than concern for the
child. The specter of review might serve as an important reminder
to the parents of the social concern for their children, and might
somehow constrain selfish behavior. Even a selfish spouse may be
more concerned about his reputation as a parent if there is some
sort of public process. Thus, in cases involving children, the judge's
role as "audience" is especially important. Finally, although most
parental agreements are approved after only superficial examination
145. See note 26 supra (study in Connecticut revealed that average of four minutes of
court time spent on uncontested divorces).
146. See p. 960 supra.
147. See R. MANooEKI, supra note 21, at 214-16.
There is, of course, substantial judicial supervision if money or property is left directly
to a child: typically there must be a guardian for the child's estate. The law requires
substantial constraints on the guardian's power to invest and spend the money, and
typically imposes considerable costs. Indeed, it is precisely for these reasons that a primary
goal in estate planning is to avoid guardianships, and instead use trusts if property is
intended for the benefit of minor children. See id. at 217-19.
995
The Yale Law Journal Vol. 88: 950, 1979
Conclusion
996
Divorce and Dispute Resolution