What Do We Owe Parents? - TDX (Tesis Doctorals en Xarxa)
-
Upload
khangminh22 -
Category
Documents
-
view
0 -
download
0
Transcript of What Do We Owe Parents? - TDX (Tesis Doctorals en Xarxa)
WhatDoWeOweParents?DistributingtheCostsofChildren
IsabellaAna-MariaTrifan
TESIDOCTORALUPF/2019
DIRECTORADELATESI
Prof.SerenaOlsaretti
DEPARTAMENTDEDRET
iii
Acknowledgments
Long before the opportunity to pursue a PhD at Pompeu Fabra (or anywhere else)
arose, Serena Olsaretti was the first professor to encourage me to pursue
postgraduate study inpolitical philosophy.As chancewouldhave it, I returned two
yearslatertostartaPhDunderhersupervision,andithasprovedthebestdecisionI
could have possibly made. This thesis would simply not have existed without her
tirelesssupport.Ihavenevermetanyonewhocanseamlesslycombinephilosophical
acuity,asenseofpragmatismandefficiency,andhumanconcernlikeshedoes.Ihope
toliveuptothemoralandintellectualstandardsshesoeffortlesslyinspires.
IamgratefultomycolleaguesintheLawandPhilosophyGroupandtheFamilyJustice
research group for the wonderfully supportive and stimulating environment they
haveprovided,inparticulartoPaulBou-Habib,PaulaCasal,AncaGheaus,andAndrew
Williams. Some of them I am now fortunate to count among my dearest friends:
Andrée-AnneCormier,ErikMagnusson,andRiccardoSpotorno.
Iamgrateful,also,toTomDouglasandJeffMcMahan,whohavesupervisedmeduring
myresearchvisitinOxford,andtoMicheleCohenwho,duringthatvisit,hasnotonly
openedherhomeforme,butofferedmegenuinefriendship.Iamespeciallythankful
to Jeff,who has been an invaluable source of knowledge aswell as a generous and
energeticsupporterofmyprofessionalaspirations.
TothemanyfriendsIhavemetalongtheway,thankyoufortheadvice, inspiration,
and challenging discussions we have shared: Enric Adell, Enric Bea, Christopher
Bennett,NicolasBrando,HuubBrouwer,SibaHarb,HwaYoungKim,FayNiker,Tom
Parr, Mauro Rossi, Pedro Silva, Adam Slavny, Areti Theofilopoulou, Kelly Tuke,
iv
AndrewWalton,MaryamWalton,GustavoZabala.IamespeciallygratefultoRuxandra
Matei and Ana Oțelea for our decade-long friendship, and whose resilience and
achievementsintheirrespectivefieldsinspireme.
IamdeeplygratefultoWillGildea,whosegenerousandinsightfulinputhasimproved
this thesis tremendously. His enthusiastic help and encouragement in the difficult
momentsgotmeoverthefinishline.
My greatest debt of gratitude is to my parents for their unconditional love, their
unflinching faith inmy abilities, and their selfless support of all of my endeavours
regardlessofhowfarawayfromhometheytookme.
Finally,thisresearch,aswellasmyresearchstayinOxford,havebeenfundedbythe
European Research Council (ERC) under the European Union’s Horizon 2020
Research and Innovation Programme (Grant Agreement Number: 648610; Grant
Acronym:FamilyJustice).Iamverygratefultohavehadtheopportunitytobeapart
ofthisresearchproject.
vi
Abstract
Thetopicofthisthesisisthejustdistributionofthecostsofchildren.Inparticular,it
askswhetheranidealliberalegalitariansocietyowesittotheparentstoensurethat
the costs of children are shared between parents and non-parents. This issue has
receivedrelatively littleattention frompoliticalphilosophers,yet it seemspertinent
to a concern that is central to anydistributive justice theory: identifying the agents
whoareresponsibleforgivingthenewmembersofsocietywhattheyareowed.The
thesis provides a comprehensive investigation into the grounds in virtue of which
parents can raise claims to have the costs of children socialized. It argues that the
prospects for successfully establishing such claims are very limited. This is either
because(i)theywouldoffendagainsttheliberalegalitariancommitmenttopersonal
responsibilityforprocreativechoice,oragainst(ii)thecommitmenttostateneutrality
between different conceptions of the good. Finally, I show that a relatively widely
endorsed family of arguments that support parents’ claims (iii) rest on specific
versions of a controversial normative principle, namely the fairness principle, that
uponexaminationturnouttobeimplausible.Thethesisthendevelopsanewaccount
of fairness,whichIcall theSharedPreferenceView.Onthisaccount, typicalparents
donothaveclaimsoffairnesstoparentalsubsidies.TheaccountoffairnessIdevelop
has important implications beyond the debate regarding the fair distribution of the
costs of children, including for environmental justice and for establishing a moral
obligationtoobeythelaw.
vii
Resumen
Eltemadeestatesiseslajustadistribucióndeloscostesdelosniños.Enconcreto,se
plantea la pregunta de si una sociedad liberal igualitaria ideal debe a los padres
asegurarque loscostesde losniñossecompartenentre lospadresy losno-padres.
Esta cuestión ha recibido relativamente poca atención por parte de los filósofos
políticos, pero resulta pertinente a una preocupación fundamental para cualquier
teoría distributiva: identificar a los agentes responsables de asegurar a los nuevos
miembrosde la sociedad lo que se les debe. La tesis proporcionauna investigación
exhaustivasobrelosmotivosenvirtuddeloscualeslospadrespuedenpretenderque
sesocialicenloscostesdelosniños.Latesisalegaquelasposibilidadesdeestablecer
con éxito tales reclamaciones son muy limitadas. Esto se debe a que estas
reclamaciones (i) irían en contra del compromiso igualitario liberal con la
responsabilidadpersonaldelaelecciónprocreativa,o(ii)encontradelcompromiso
con la neutralidad del estado entre diferentes concepciones del bien. Finalmente,
muestra que una família de argumentos ampliamente respaldados que apoyan las
reclamaciones de los padres (iii) están basados en versiones específicas de un
controvertido principio normativo, a saber, el principio de equidad, que bajo
escrutinio resultan inverosímiles. A continuación, la tesis desarrolla un nuevo
significado del concepto de equidad, al que llamo Teoría de las Preferencias
Compartidas.Segúnesta teoría, lospadrestípicosnopuedenextenderdemandasde
equidad con respecto a los subsidios genitoriales. La teoría de la equidad que
desarrollo tiene implicaciones importantes más allá del debate sobre la justa
distribucióndeloscostesdelosniños,incluyendoaspectosdejusticiaambientalyel
establecimientodeunaobligaciónmoraldeobedecerlaley.
ix
TABLEOFCONTENTS
Acknowledgments....................................................................................................................................iii
Abstract.......................................................................................................................................................vi
Resumen.....................................................................................................................................................vii
Chapter1......................................................................................................................................................1
Introduction................................................................................................................................................11.1.Thequestionofparentaljustice......................................................................................................................31.2.Thecostsofchildren.........................................................................................................................................101.3.Theplausiblepro-sharingargument.........................................................................................................131.4.Previewofthethesis.........................................................................................................................................15
Chapter2....................................................................................................................................................18
TheArgumentfromInsurance...........................................................................................................182.1.Dworkin’shypotheticalinsurancedevice................................................................................................212.2.Theinsuranceargumentforparentalsubsidies...................................................................................232.3.Adequateparentingandthecostsatstakebehindtheveil.............................................................262.4.Underemploymentinsurancewouldberathermodest:sowhat?................................................322.5.Areparentalsubsidiesthemorecost-effectiveformofinsurance?.............................................362.6.Conclusion.............................................................................................................................................................45
Chapter3....................................................................................................................................................48
TheArgumentfromAutonomy...........................................................................................................483.1.Alstott’sautonomy-basedchallengetoparentaljustice....................................................................503.2.Theburdensofchildrearing..........................................................................................................................523.2.1.NoExit..................................................................................................................................................................533.2.2.Givingchildrenagoodupbringing..........................................................................................................563.2.3.Beingoncall.......................................................................................................................................................57
3.3.Autonomycosts,responsibilitycosts,andtheissueofstakes........................................................583.3.1.Thequestionofresponsibilitystakes......................................................................................................603.3.2.Thequestionofautonomystakes.............................................................................................................62
3.4.Hierarchicalaccountsofautonomy............................................................................................................653.5.Substantiveviewsofautonomy...................................................................................................................713.5.1.Theconstraintsofself-respect...................................................................................................................723.5.2.Theself-authorshipaccount.......................................................................................................................74
3.6.Relationalautonomy.........................................................................................................................................853.7.Thedemandsofequalautonomyfrombehindaveilofignorance...............................................893.8.Conclusion.............................................................................................................................................................95
Chapter4....................................................................................................................................................97
TheArgumentfromFairness..............................................................................................................974.1.Theempiricalpremiseoffairness-basedarguments:parents’contributiontosociety......98
x
4.2.Thenormativepremiseoffairness-basedarguments:FairPlay.................................................1004.3.TheFairPlaydebate........................................................................................................................................1034.4.Unfairfreeridingasafailureofimpartiality........................................................................................1084.5.TheSharedPreferenceViewofFairPlay...............................................................................................1134.6.TheSharedPreferenceViewandmorallyrequiredgoods.............................................................1194.7.BenefitsrecipientswholackFairPlayobligations............................................................................1254.8.BenefitsproducerswholackFairPlayclaims.....................................................................................1274.9.Conclusion...........................................................................................................................................................139
Chapter5.................................................................................................................................................141
AreNon-ParentsUnfairFreeRiders?............................................................................................1415.1.Childrenaspublicgoods...............................................................................................................................1415.2.Childrenasmorallyrequiredgoods.........................................................................................................1535.2.1.Thechallengefromirrelevantintentions...........................................................................................1575.2.2.Thechallengefromdouble-counting....................................................................................................162
5.3.Childrenassocializedgoods........................................................................................................................1635.4.Conclusion...........................................................................................................................................................171
Chapter6.................................................................................................................................................173
AreKidsUnfairFreeRiders?............................................................................................................1736.1.Tomlin’sKidsPayView..................................................................................................................................1746.1.1.KidsPay:thegroundsforaFairPlayobligation.............................................................................1766.1.2.ThemissingstepintheKidsPayargument.......................................................................................182
6.2.KidsPayandtheSharedPreferenceView............................................................................................1886.2.1.Themorallyrequiredgoodsofanupbringing..................................................................................1926.2.2.Theoptionalgoodsofanupbringing....................................................................................................197
6.3.TheSharedPreferenceViewandfilialobligations............................................................................2026.4.Conclusion...........................................................................................................................................................206
7.Conclusion..........................................................................................................................................208
Bibliography..........................................................................................................................................213
1
Chapter1
Introduction
Asparentsknowalltoowell,raisingchildrenisexpensive.Thelatestfiguresfromthe
USGovernmentshowthatafamilycomposedoftwoparentsandtwochildrenspends
$233,610perchildtoraisethemtotheageof17.1Andchildrenimposecostsnotonly
onparents, but society at large too. Children generate costs for society for twokey
reasons.First, theyadd to thenumberof citizens that can lay claimson thenatural
and socioeconomic resources of a state. Secondly, inmany countries, costs-sharing
schemes are in place which spread out the costs of raising children. For instance,
childrenoftenreceivepubliclyfundedhealthcareandeducation.Taxbreaksareoften
applied in respect of children, and paid parental leave is often made available to
familieswithnewbornbabies.Thefact thatchildrengeneratesubstantialcosts fora
number of different agents raises an important questionof justice:What is the just
distribution of the costs of children? How should the costs of children be shared
betweentheparents,therestofsociety,and,perhaps,thechildrenthemselves?
A complete answer to this questionwould have to balance a number of important
considerationsincludingchildren’srights,theinterestthatwomenhaveinagender-
egalitariandistributionof childrearingcosts, and the interestsof societyat large. In
addition to thesewell-recognizedconsiderationsof justice,which Iwillnotaddress,
thereisanothersetofintereststhatshouldbearonhowthecostsofchildrenshould
bedistributed.The latterhave, regrettably, receivedrelatively littleattention.These
aretheinterestsofparentsquaparents,andtheyformthesubjectofthisthesis.
1Lino,Kuczynski,RodriguezandSchap2017.
2
Thisdissertationprovidesacomprehensiveinvestigationintothegroundsinvirtueof
whichparentscanlayclaimsofjusticetohavethecostsofchildrensocializedbyajust
liberal egalitarian society. It argues that the prospects for successfully establishing
such claims are very limited. The most promising avenue for parents can be
summarizedby the followingclaim.Unless thecostsofcaring for infantsandyoung
children is socialized, those who care for them full-time would suffer deficits of
personalautonomythatareincompatiblewiththeliberalegalitariancommitmentto
personalfreedom.Thisconclusionwouldbeinlinewithwidespreadfeministviewson
thematter,2andwouldsupportpoliciessuchaspaidparental leave,publicly funded
childcare facilities,workplaceaccommodation forparentsofyoungchildrenandthe
like.
Ishow,further,thatotherargumentsthatseem,primafacie,toprovideparentswith
distinctive claims to having the costs of children shared are not successful. This is
eitherbecause(i)theyoffendagainsttheliberalegalitariancommitmenttopersonal
responsibilityforprocreativechoice,oragainst(ii)theliberalegalitariancommitment
to state neutrality betweendifferent conceptions of the good. Finally, I show that a
relativelywidelyendorsedfamilyofargumentsthatsupportparents’claims(iii)rest
onspecificversionsofwhatisalreadyacontroversialnormativeprinciple,namelythe
fairnessprinciple,3thatuponexaminationturnouttobeimplausible.Thethesisthen
develops a new account of fairness, which I call the Shared Preference View.
Accordingtothisaccount, typicalprocreativeparentsdonothave fairnessclaimsto
besubsidized.
In this introduction I will explain in more detail the question that this thesis is
concernedwith,namely,thequestionofparentaljustice.IdothisinSection1.1,where
Ialsolayoutthebackgroundassumptionsandthegeneralmotivationbehindtackling
2Seee.g.Bubeck1995;Anderson1999.3Hart1995;Rawls1999a.
3
thisissue.InSection1.2Iexploreinabitmoredepthhowmuchparents,aswellasthe
rest of society, spend on children, and I clarify which of these costs are up for
socializingunderthesortsofargumentsIdiscuss.InSection1.3Ilayoutthecriteria
thataplausibleargumentestablishingparents’claimsofjusticeoughttomeet.Finally,
Section1.4offersapreviewoftherestofthethesis.
1.1.ThequestionofparentaljusticePhilosophers subscribing to some version of liberal egalitarianism have generally
tended tooverlook the implicationsof theirview forwhatSerenaOlsaretti calls the
centralquestionofparentaljustice:What,ifanything,doesajustsocietyoweparents
invirtueofhavingandraisingchildren?4
Mainstreamliberalegalitarianviewsofjusticearefundamentallycommittedtothree
mainvalues:thevalueofequality,thevalueofpersonalfreedomand,finally,thevalue
of personal responsibility. The overwhelming contemporary interest in this type of
viewcanbe tracedback to the seminalworkof JohnRawls,5which, asNagelnotes,
combines“theverystrongprinciplesofsocialandeconomicequalityassociatedwith
European socialism with the equally strong principles of pluralistic toleration and
personalfreedomassociatedwithAmericanliberalism,andhehasdonesoinatheory
that traces them to a common foundation.”6Liberal egalitarianism dominates the
contemporary political philosophy literature, and it holds, in a nutshell, that the
inequalities traceable to unfavorable circumstances are unjust and should be
4Olsaretti2013;Bou-HabibandOlsaretti2013.5Rawls1999a.6Nagel2002,p.88.
4
redressed,while the inequalities traceable to people’s choices or ambitions are not
unjust.7
A central commitment of contemporary liberal egalitarian views, then, is to hold
people responsible for what they make of their own lives. But what exactly they
choosetomakeoftheirlivestheyarefreetodecideforthemselves,withoutthestate
interfering with their choices and without it favoring one lifestyle over another
throughthepoliciesitadopts.Orsoitismaintainedbythoseliberalegalitarianviews
that are also committed to a principle of political neutrality. Political neutrality
providesaprincipledconstraintonwhatcountsaslegitimatestateaction.Itholdsthat
legitimate state action should be neutral in the face of the plurality of views of the
good life that its citizens may adopt. Rawls’s own view of justice as fairness is
committedtothisprinciple,andheexplainsitasfollows.
[Justiceasfairnessdoesnot]trytoevaluatetherelativemeritsofdifferentconceptions
ofthegood.Instead,itisassumedthatthemembersofsocietyarerationalpersonsable
to adjust their conceptions of the good to their situation. There is no necessity to
comparetheworthoftheconceptionsofdifferentpersonsonceitissupposedtheyare
compatiblewiththeprinciplesofjustice.Everyoneisassuredanequallibertytopursue
whateverplanoflifehepleasessolongasitdoesnotviolatewhatjusticedemands.8
Acriticofpoliticalneutrality,AlasdairMacIntyre,definespoliticalneutralitythisway:
[A] community is simply an arena in which individuals each pursue their own self-
chosenconceptionofthegoodlife,andpoliticalinstitutionsexisttoprovidethatdegree
oforderwhichmakessuchself-determinedactivitypossible.Governmentandlaware,
orought tobe,neutralbetweenrivalconceptionsof thegood life forman,andhence,
7PhilosopherswhohavedevelopedotherversionsofliberalegalitariantheoriesincludeDworkin1981;Arneson1989;Cohen1989;Roemer1993,1996,1998;Fleurbaey1995a,b.8Rawls1999a,pp.80-81.
5
althoughitisthetaskofgovernmenttopromotelaw-abidingness,itis(…)nopartofthe
legitimatefunctionofgovernmenttoinculcateanyonemoraloutlook.9
In this thesis I am interested in the positive answers to the central question of
parental justicethatcouldbeformulatedwithintheframeworkof liberalegalitarian
theories of justice that are committed both to holding people responsible for their
choices,aswellastotheprincipleofpoliticalneutrality.Thisispartlyforpragmatic
reasons:themajorityofcontemporaryliberalegalitarianssubscribetosuchaview.10
Itisworthadoptingthisframework,then,iftheargumentsexploredinthisthesisare
to appeal to aswide an audience as possible. Butmore importantly, the reasons in
favorof theneutralityconstraintonstateactionseemcompellingontheirown.The
justificationforstateneutralitybeginsfromthebasicmoralassumptionthatallofus
are born free and equal. As such, we are not naturally under anyone’s, or any
institution’s,authority.Ifapersonoraninstitutionwishestoexerciseauthorityover
people,theycanonlylegitimatelydosoiftheycanjustifytheuseofauthoritytothe
people overwhich theywish to exercise it. The legitimate use of power, therefore,
requires thata justification isprovidedwhich freeandequalpeoplecanaccept.But
freeandequalpeopleinliberalsocietieshavedeep-seateddisagreementsoverwhatis
valuable and what makes for a good life. Given the fact of reasonable pluralism,11
therefore, if the state is to exercise legitimate power over all of its citizens itmust
refrainfromtakingsidesonwhatmakesforagoodlife.12
Given the political neutrality constraint, then, in this thesis I will leave aside any
argumentsthatmightseektoestablishparents’claimstohavingthecostsofchildren
sharedbysocietybyappealingtothevalueofparentingasaworthwhilewayof life
thatshouldbeprotectedorencouragedbythestate. Iwill limit the inquirytothose
arguments that have the potential, at least prima facie, to establish parental
9MacIntyre1984,p.195.10PerhapsthemostnotableexceptionfromthecommitmenttostateneutralityisRaz1986.11Rawls1993,ch1.12ForarecentandelaboratedefenceofpoliticalneutralityseeQuong2011.
6
entitlementsbyappealingtoreasonsthatfreeandequalcitizenscanacceptregardless
oftheirparticularviewofthegoodlife.
Thatthecostsofchildren,oratleastasubsetofthem,shouldbesharedinsomeform
or another by the entire society seems like a common-sense view to hold. Liberal
egalitarians,too,havetendedtoassumethatthecostsofchildrenoughttobeshared.
Policiessuchaspubliclyfundededucationorhealthcareforchildrenareconsidered
tobecompatiblewith,orevenrequiredby,theseviews.Yetonedirectimplicationof
oneofthebasictenetsofliberalegalitarianism,namelythatpeopleoughttobearthe
costs that they are responsible for creating, seems to point to the conclusion that
parentsought tobear all of the costsof children themselves. Insofar asparents are
responsible for creating the need for someone to meet the costs of children (by
knowinglybringingchildrenintoexistencewhentheycouldhaveavoidedthiscourse
of action), it seems that they should be the ones to meet those costs. Conversely,
having to participate inmeeting the costs thatotherpeople have created by having
children seems like an unjust imposition on thosewho choose to remain childless.
Indeed,thecostsimposedonthechildlessbythosewithchildrenseemrelevantlyakin
toastrokeofbadluckratherthansomethingtheycanjustlybeheldresponsiblefor
paying.13Based on this, several philosophers have concluded that responsibility-
sensitive liberal egalitarian views demand that, at least as amatter of basic justice
betweenparentsandthechildless,parentsshouldbearallthecostsofchildren,justas
those responsible for adopting other sorts of expensive lifestyles should be held
responsibleforthecoststhattheirambitionsgenerate.Wecancallthistheanti-cost-
sharingview,oranti-sharingviewforshort.14
13Thegeneralpointthatthenegativeeffectsofindividuals’choicesonthirdpartiesshouldbetreatedasaformofbadluckforthosethirdpartiescanbetracedbacktoSteiner1997.14IborrowthislabelfromOlsaretti2013,p.229.Variousversionsoftheanti-sharingviewhavebeendefendedbyAckerman1980;Casal1999;CasalandWilliams1995,2004,2008;Clayton2006;Rakowski1991;SteinerandVallentyne2009;Vallentyne2002.
7
Thisthesisprovidesacomprehensiveinvestigationintothoseargumentsthatattempt
toestablishthat,despiteparents’beingresponsibleforcreatingtheneedtomeetthe
costsofchildren,basicjusticebetweenparentsandnon-parentsrequiresthatsomeof
thosecostsbesharedbytherestofsociety,evenonaresponsibility-sensitiveliberal
egalitarian theory. This family of views has been labelled pro-cost-sharing, or pro-
sharingforshort.15
Thequestionofparentaljusticedeservescarefultreatmentattheveryleastbecausea
theoryofjustice,insofarasitisatheoryabouthowtodistributebenefitsandburdens
withinasociety,isincompletewithoutsettlingthematterofhowburdensshouldbe
distributedamongparentsandnon-parents.16Moreimportantly,atheoryofjusticeis
incomplete if it ignoresthequestionofparental justicebecausesomeofthecostsof
childrenarecoststhatneedtobeborneinordertogiveeverypersonhisorherjust
due.17New members of society, both during their childhood, and once they have
reached adulthood, have rights that must be met by someone. The question of
parentaljustice,then,seekstoidentifytheagentswhoshouldbeartheresponsibility
tohelpmeettherightsthesenewpersonshave.Thus,thequestionofparentaljustice
concernsacentralquestionaboutjusticeingeneral:whetherornoteveryoneshould
share the costs ofmeeting everyone else’s claims of justice, orwhether these costs
should only be incurred by the ones responsible for the existence of new citizens,
namelytheirbiologicalparents.
To be sure, even if pro-sharing answers to the question of parental justice fail,
governmentsmaystillstepintoensurethatchildrenindisadvantagedfamilieshave
theirbasicneeds taken careof, or that theyhave their rightsmetwhere thismight
include more than just having their basic needs met, when their parents cannot
discharge theseobligations themselves.Thestatemayalso institutespecificpolicies
to redress other kinds of injustice arising in the family, such as gender-related15Olsaretti2013,p.229.16Olsaretti2017.17Bou-HabibandOlsaretti2013,p.422.
8
injustices. Most current societies do have such policies in place to one extent or
another.Butthequestionthisthesiswillbefocusedoniswhetherparentshaveclaims
to have the costs of children shared by the rest of society in virtue of what these
activitiesthemselvesentitlethemtoasmatterofidealjustice,evenonresponsibility-
sensitiveviewsofjustice.
Myapproachtoansweringthequestionofparentaljustice,then,isanexerciseinideal
political theory. The term “ideal theory” refers to the methodology pioneered by
Rawlsforidentifyingprinciplesofjusticethatapplytobasicsocialinstitutions.18Itis
defined in contrast to non-ideal theorizing, and it involves devising principles of
justice against two fundamental assumptions. The first assumption is that of full
compliance.Principlesofjusticearetobeidentifiedontheassumptionthateveryone,
or nearly everyone, complies with those principles. By contrast, non-ideal theory
dealswith questions of non-compliance. Oneway to see this contrast is to think of
non-ideal theory as askingwhat the appropriate responses shouldbe in the face of
injustice,whichiswhytypicalnon-idealtheorytopicsincludetheoriesofpunishment,
compensatory justiceor the justresponses tounjustpoliticalregimes,amongwhich
civil disobedience is one. 19 Ideal theorizing, on the other hand, sets aside the
possibilitythatpeoplemightnotcomplywiththedemandsofjusticeandattemptsto
establish what a perfectly just society would be like in the first place. The second
assumptionofidealtheoryisthatthereexistfavorablesocioeconomiccircumstances
fortheestablishmentof just institutions.Thisassumptionshouldnotbeunderstood
as referring toconditions thatareso favorable that theywouldbe impossible in the
real world, but rather to conditions that are favorable andwhich are also feasible.
Non-ideal theory, by contrast, is concerned with the socioeconomic limitations to
justice.20
18Rawls1999a.19Rawls1999a,p.8.20Forahelpfuldiscussionofthedistinctionbetweenidealandnon-idealtheoryseeStemplowskaandSwift2012.
9
Inreality,ofcourse,injusticeabounds,includinginjusticessufferedbychildrenwhich
could be redressed better and/or quicker by pooling all of society’s resources as
opposedtoonlytheparents’.Iftheultimateaimofpoliticaltheoryistofindsolutions
to“thepressingandurgentmatters”21thatconfrontsociety,onemightwonderwhat
thepointisofengaginginidealtheory.AccordingtoRawls,establishingthedemands
of ideal justice is worth doing for its own sake, but more importantly, it is the
necessary first step inattempting tosolve theproblemsofeveryday life.Systematic
thinkingabout justice in the realworldmustbeginby firstdetermining “thenature
andaimsofaperfectlyjustsociety”22whichwecanuseasastandardtothenevaluate
thecurrentstateofaffairs.Idealtheorycanfunctionasastandardbothinthesenseof
providinganaimfornon-idealtheorytoworktowards,aswellasawayofevaluating
theurgencyofparticularinjusticesbyreferencetohowlargethegapisbetweenour
real-worldcircumstancesandthecircumstancesofidealjustice.23
IfullysubscribetotheRawlsianideathatsystematictheorizingaboutjusticerequires
ustofirstestablishwhatperfectjusticelookslikeina“realisticutopia”24wherethere
areno socioeconomic limitations to establishing just institutions andwherewe can
counton fullcompliance. Ibelieveweshouldregardthedistinctiveclaimsof justice
thatparentsmighthave invirtueofhavingand raising childrenasonepieceof the
puzzleofwhataperfectlyjustdistributionofthecostsofchildrenlookslike.
In addition to the two standard assumptions of ideal theory, inwhat follows Iwill
work with a number of other simplifying assumptions, which it is important to
mention here. First, by “parents” I will refer to those who are their children’s
procreatorsaswell as their custodians, even though these twocategories can come
apart, andoftendocomeapart in reality, for instance in thecaseofadoption. Iwill
alsoassume,second,thatparentshaveampleopportunitytoavoidbecomingparents
21Rawls1999a,p.8.22Rawls1999a,p.8.23StemplowskaandSwift2012,p.376.24Rawls1999b,pp.11-12.
10
butpositivelychoosetohaveandraisechildren.By“non-parents”,conversely,Irefer
to thosewho choose to remain childless. Imake these assumptions in the spirit of
approaching the central question of parental justice in a careful and systematic
manner that attempts, as a first step, to establish the demands of perfect justice as
theyapplytocasesthataresimpleintherespectsIhighlighted.Asecondstepwould
involvereintroducing thesecomplicationsandchecking towhatextent,and inwhat
ways,doingsowouldaffectourconclusionsaboutparents’justclaims.InthisthesisI
willlimitmyself,however,todealingwiththesimplifiedcases.
Ialsomakeanassumptionaboutthemoralpermissibilityofprocreation,namelythat
bringing children into existence can be morally permissible, but is not morally
required. This position is in line bothwith common-sensemorality aswell aswith
most of the procreative ethics literature. 25 Assuming that procreation can be
permissible at least under certain conditions is also necessary to get the parental
justice debate off the ground, for we might think that if procreation were morally
impermissible, this would undermine parents’ claims to have the costs of children
sharedby the rest of society. If bringingpeople into existencewerealwaysmorally
impermissible, we might think that parents should not only bear the costs of
children’supbringing,butalsoofcompensatingthem,andperhapssocietyatlarge,for
wrongfullyhavingbroughtthemintoexistence.Idonotconsiderthesepossibilitiesin
whatfollows.
1.2.ThecostsofchildrenThe costs of raising childrenvarywidely, not only across countries, but alsowithin
eachcountry,dependingonfactorssuchasgeographicallocation,familycomposition
25Seee.g.Parfit1984,p.358andAppendixG;McMahan1981,pp.104-105;Narveson1978,p.48;Velleman2008a,pp.242-244;Velleman2008b,pp.247-250.
11
and income level. A report published in 2017 by theUSDepartment of Agriculture
estimated that a married-couple, two-children, middle-income family would spend
$233,610raisingachildbornin2015totheageof17.Atwo-parentfamilywithonly
onechildwillspend27percentmore.26IntheUK,atwo-personfamilywouldspend
£75,436toraiseachildtotheageof18,thoughthisfiguredoesnotincludethecosts
of housing, health care, and council tax. Also excluding these costs, a one-parent
householdintheUKwouldhavetospend£102,627onchildrearingexpenses.27
Alloftheabovefigurescaptureonlythedirectexpensesmadebyparentsanddonot
include indirect expenses such as time investments, forgone earnings and forgone
career opportunities. For measuring the financial value of the time that parents
devote to childrearing, the economist Nancy Folbre uses a “replacement costs
approach”. This involves determining how much it would cost society to provide
childcareofacceptablequalityifparentswereunableorunwillingtocareforthem.28
Sheestimates“alowerboundforthereplacementcostofparentalservicesperchild
inatwo-parent,two-childfamilyofabout$13,352”peryear.29
Parental spending does not stop once children reach adulthood. Many parents
contribute to their grown children’s expenses, including higher education costs,
housing,orweddingexpenses.AsurveystudycommissionedbytheBankofAmerica
Merrill Lynch showed that 79 percent of American parents continue to transfer
considerableamountsofresourcestotheiradultchildren(between18and34years
old), spending twice theamountofmoney tomeet theirneeds than theycontribute
towardtheirownprivateretirementfunds.30
Finally, consider the public expenditure on childrearing. This, of course, varies
considerablyfromcountryfromcountry.Manyindustrializedcountriesprovidedirect26Lino,Kuczynski,RodriguezandSchap2017.27Hirsch2017.28Folbre2008,p.130.29Folbre2008,p.130.30MerrillLynchandAgeWave2018.
12
cash transfers and/or tax exemptions to parents, and support for childcare, health
care, andpublic education.31When it comes topublic education, for example,OECD
countries spend on average $10,759 a year per student, from primary to tertiary
education.32In the US, the average amount spent per elementary and secondary
school student in the year 2014-2015 was $13,119, adding up to a total of $668
billion.33
These figures lend support to the idea that, as Folbre puts it, “children are an
expensivecrop.”34Butwhichofthesecostsareactuallyplausiblecontendersforbeing
socializedongroundsofparentaljustice?
In order to answer this, it isworth noting two clarifications aboutwhich costs are
relevant.BothpointsareduetoOlsaretti.35First,thecostsofchildrenreferbothtothe
costsofcareaswellastothecoststhatthesechildrenwillimposeoncetheybecome
adults. The costs of care include the loss of time, opportunities, and financial
resources that parents incur in the process of raising their children to be self-
sufficient adults (conventionally, up until the age of 18 years). Sharing these costs
withparentswouldinvolvethingslikeuniversallyfundedchildcare,parentalleave,or
giving tax breaks to families. The costs of “added adultmembers”, asOlsaretti calls
them,refertothecostsofmeetingwhateverclaimstheirchildrenwillmakeasadults,
which depending on our conception of justice, might include their claims to a fair
shareofresources, toequalcapabilities, toabasicminimum,andsoon.AsOlsaretti
points out, sharing the costs of added adultmembers simply involves contributing
equalshares,throughtheirtaxes,“toanypubliclyfundedschemeforcitizens(aimed
atprotectingeithernegativeorpositiverights),ratherthanbeingliabletoalessertax
31SeeFolbre2008,pp.139-177.32OECD2017.33USDepartmentofEducation,NationalCenterforEducationStatistics2018.34Folbre2008,p.65.35Olsaretti2013,pp.229-232.
13
burden,suchthattheycouldbesaidtohavetheirshareunaffected, insofarasthis is
feasible,byparents’choicestocreatenewpersons.”36
A second clarification about the relevant costs of children is key for understanding
what is at stake in the parental justice debate. The costs that are up for socializing
includeonlya subsetof all the costsof careand thecostsof addedadultmembers,
namelythosecostsofcareandthosecostsofaddedadultmembersthataremorally
required. For the burdens that childrenmight impose on their parents or on third
partiescaninclude,andoftendoinclude,burdensthataremorallyoptionalormorally
unjustified.Forinstance,parentsoftengobeyondmeetingthejustclaimsofchildren
in their attempt to give their children the best upbringing they possibly can. They
mightevenendupshoweringtheirchildrenwithluxurygoods.However,theclaimsof
justice they can raiseagainst the restof society canonly refer to the rangeof costs
thatchildrenhaveaclaimtoreceiving.Andthesamegoesforthenewadults:therest
ofsocietycanonlybeenjoinedtocontributetomeetingthenewmembers’justclaims,
butnot tomeetingtheclaimsthat theymightraisewhicharemorally indifferentor
unjustified.37
1.3.Theplausiblepro-sharingargumentForthepurposeofinvestigatingtheprospectsofsuccessofthepro-sharingparental
justice case, it is useful to set out the criteria forwhat I take tobe a plausiblepro-
sharingargument.BeforeIdothis,letmementiononetypeofpro-sharingargumentI
willnotconsider.
36Olsaretti2013,p.230.37SeeOlsaretti2013,p.230.
14
Onetypeofpro-sharingargumentthatIwillnotspendtimeonistheoneaccordingto
which society ought to share the costs of children so as to incentivize people to
becomeparentsinordertoreachadesiredpopulationsize(withaviewtogrowingor
maintaining a country’s economy, for example). The reason I leave such incentive-
based arguments aside is that, for one thing, they are not very philosophically
challenging. Inprinciple, if societyneeds to incentivize fertility, sharingsomeof the
costs of children, thereby lowering the costs of parenthood for the prospective
parents, seems likeapermissibleway todoso takenon itsown(that is,bracketing
other considerations thatmightmake increasing fertility impermissible, suchas the
worries about the carbon footprint of an increased population). More importantly,
this sort of argument merely establishes the permissibility of sharing the costs of
children, rather thananentitlementon thepartofparents.This thesis is concerned
with thosearguments that takeup themore challenging taskof groundingparental
claimsofjustice.
A plausible pro-sharing argument will exhibit the following features. First, it will
attempttoestablishanentitlementonthepartofparentstohavesomeorallofthe
costs of children shared by society, rather than merely establishing that it is
permissible to do so. Second, it will attempt to do so by identifying a normatively
relevant feature of parenting itself, or its effects, that can ground parental
entitlements.Morespecifically,mydiscussion focusesonarguments that identifyan
interestthatparentshavethatwouldbeunderminediftheyhadtobearallthecosts
ofprocreatingandchildrearing.Thisinterestmustnotbereducibletotheinterestof
children in having their needsmet, nor to the interest thatwomen have in gender
justice.Theplausiblepro-sharingargument,then,mustshowthatthestatewouldbe
failingparentsthemselves(asopposedtochildren,orwomen)ifitdidnotsocializethe
costs of children. Thirdly, the plausible pro-sharing argument must show that the
relevant parental interest it identifies as relevant is one that the state should be
committedtoprotecting.Inotherwords,itmustshowthatthestatewouldbefailing
parents along some important dimension of justice in a way that would ground a
15
complaintonthepartofparents.Anditmustdoso,fourth,compatiblywiththeliberal
egalitarian commitment to personal responsibility, and, fifth, compatibly with the
principle of political neutrality. Finally, the plausible parental entitlement thus
established shouldonly apply to themorally required costsof children, rather than
also the morally optional or the morally impermissible costs that children might
create.
In summary, a plausible pro-sharing argument for parental justice is one that
establishes, compatibly with the commitment to personal responsibility and to
politicalneutrality,anobligationofbasicjusticeonthepartoftheliberalegalitarian
state to share (some of) themorally required costs of children by appealing to an
interestthatparentshavequaparents,whichwouldbeunderminedifthestatefailed
tosocializethosecosts,andwhichparentshaveaclaimtohaveprotected.
1.4.PreviewofthethesisI begin the investigation in Chapter 2 by considering an argument that attempts to
show that one brand of liberal egalitarianism that seems particularly averse to
parents’claims,RonaldDworkin’sequalityofresources,caninfactsupportthem.This
viewisdefendedbyPaulBou-Habiband itholds, inanutshell, that thestateshould
subsidize parents if it is true thatmost peoplewould take out “insurance cover” to
cover some of the costs of children under the hypothetical insurance conditions
specifiedbytheequalityofresourcesview.38
I begin with this argument because, if successful, it would provide a particularly
strongcaseforparents.Thisisbecause,first,itexplicitlysetsouttoshowthatapro-
sharing conclusion can be reached compatibly with the liberal commitments to
38Bou-Habib2012.
16
personal responsibility and to state neutrality. And secondly, it would establish
parents’ claims against non-parents without appealing to any further principles or
reasonsexternal to thebasic tenetsof equalityof resources itself.Despite its initial
plausiblity, Ishowthat thisargumentdoesnotsucceedbecause itultimately fails to
grounddistinctiveclaimsofjusticeonthepartofparents.
In Chapter 3 I consider an argument that holdsmore promise as far as grounding
distinctive parental justice claims, namely the argument according towhich parents
wouldsufferanautonomydeficitiftheywereforcedtobearallthecostsofchildren.39
I show that, while it is themost promising avenue for parental justice, this line of
argumentisincomplete.It isincompletebecause,first,onmostaccountsofpersonal
autonomyitisdifficulttoevenidentifyasetofcoststhatshouldcountasautonomy-
undermininginawaythatis(a)plausibleand(b)compatiblewithpoliticalneutrality.
Second,evenoncewehavemanagedtoidentifyonepossiblesetofsuchcosts,namely
thecoststhat full-timecarersof infantsandyoungchildrenface, it isnotclearwhat
citizens’justclaimsofautonomyamountto.Itisnotclear,thatis,thattheautonomy
deficitsthatfull-timecarerssuffercangroundaclaimofjusticethatthestateshould
restoretheirautonomyforthetimetheyarefull-timecarers.
Chapter4begins the investigation into a familyof views thathas enjoyed themost
popularity in the pro-sharing camp: fairness-based arguments. In a nutshell, they
claimthatparentsproducepublicgoodsbyhavingandraisingchildrenthattheentire
societyenjoys,includingnon-parents.Allowingnon-parentstointernalizethebenefits
of children without enjoining them to help share the costs would allow them to
unfairly freerideonparents’effortsofhavingandraisingchildren. In thischapter I
briefly introduce these views and I focus most of my attention on the normative
principle on which they rely to establish parents’ claims. This is the Hart-Rawls
principle of fairness, or Fair Play. Noting that the Fair Play debate has proceeded
largelythroughpiecemealdiscussions,IsetouttoofferasystematicapproachtoFair39Alstott2004.
17
Play.IdevelopanewaccountofFairPlay,whichIcalltheSharedPreferenceView.On
thisview, freeridingcountsaswrongfulwhenwecanjustifiablyascribetoboththe
benefitsproducersandthefreeridersaqualifiedpreferenceforfreeriding.
InChapter5IdrawouttheimplicationoftheSharedPreferenceViewofFairPlayfor
parental justice. I argue that, on this view, parents do not have claims of fairness
againstnon-parents.
Finally, Chapter 6 considers another version of a parental justice argument that
appeals to a principle of Fair Play for its normative support. This is the Kids Pay
view,40anditclaimsthatgrownchildrenshouldbeviewedasunfairfreeridersifthey
internalize thebenefits of anupbringing that theirparentsprovidewithout sharing
thecosts for themonce theybecomeadultsandareable topay their share. I argue
thatthisargument,too,hasonlyverylimitedsuccess.OntheSharedPreferenceView,
theonlybenefitsofanupbringingforwhichparentsmayhavefairnessclaimsarethe
morally optional goods of a good parent-child relationship. But since these are
morallyoptionalgoods,arguablythereisnoclaimforthesefairness-basedclaimsto
beenforcedbythestatebyextractingtaxesfromgrownchildrenwhowerebenefited.
40Tomlin2015.
18
Chapter2
TheArgumentfromInsurance
Of all the theories of distributive justice on offer, the one on which it seemsmost
straightforwardthatparentslackclaimsofjusticetohavethecostsofchildrenshared
bythechildlessisRonaldDworkin’sequalityofresources.Onthisview,inequalities
inpeople’ssharesofresourcesthataretraceabletopeople’sendowmentsareunjust
whiletheinequalitiestraceabletotheirambitionsarenot.Thegoalistoneutralizethe
effectsofdifferentialluckonpeople’sshares,whichiswhyequalityofresourceshas
since been included under the umbrella term “luck egalitarianism.”41By contrast,
justicedoesnotrequire,on thisview, thatwecompensatepeople fordisadvantages
thatareduetotheirlifeplansortotheirexpensivetastes.42
It iseasytoseewhy ithasbeenthought thatparents lackclaimstoassistance from
non-parentson thisview. Insofaraschildrenare theresultofpeople’sambition (as
wemayassumeasamatterofidealtheory),everythingelsebeingequal,itlookslike
thecosts thatareattachedto thechoiceofbecomingaparentought tobeborneby
those who made that choice, namely the parents themselves. Or so most
commentatorsonthisissuehavemaintained.43Forinstance,JustineBurleywrites,“as
a matter of principle, Dworkin’s theory does not endorse redistribution to people
41Inaseminalcriticalpaper,ElizabethAnderson(1999,p.289)coinsthetermluckegalitarianism.OtherluckegalitariantheoriesapartfromDworkin’sequalityofresourcesincludeArneson1989;Cohen1989;Rakowski1991;Roemer1994;Temkin1993.42SeeDworkin2000,pp.65-68.43Burley2000,p.138;Casal1999,pp.367-368;CasalandWilliams1995,pp.97-98;CasalandWilliams2004,pp.156-159;Clayton2006,p.169;Rakowski1991,p.153.
19
whoseinferioreconomicpositionistraceabletoapreferenceforchildrenwhenthis
preferenceformspartoftheirviewofwhatleadingagoodlifeis.”44
However,PaulBou-Habibmountsacasetothecontrary.45Hearguesthatnon-parents
oughttosharethecostsofchildrenasamatterofbasicegalitarianjusticeevenona
view like equality of resources.46This pro-sharing argument,whichBou-Habib calls
the insuranceargument,47starts from theassumptionthatweshoulddesignour tax
systeminthewayDworkinrecommends,namely,soastoreflectwhichpiecesofbad
luckpeoplewould choose to takeout insurance against in a hypothetical insurance
scenario under conditions of equality. Among other things, individualswould likely
choose, the argument continues, to fund a means tested subsidy earmarked to
partially cover childrearing expenses. For most individuals in this hypothetical
insurance scenario would worry about how the bad luck of lacking the talents to
attractagenerousincomewouldaffecttheirprospectsforbeingabletoraisechildren
adequately. Most people in our society, Bou-Habib argues, would be so uniquely
averse to facing this unfortunate prospect that (i) they would want to insure
themselves against it even if, at the time ofmaking this hypothetical decision, they
werenotcertain theywouldendupwanting tohavechildrenafterall, and(ii) they
wouldnotfeelascompelledtoinsureotherpotentialexpensivepursuitsinasimilar
fashion.
Whethertheargumentfrominsuranceworksmattersforseveralreasons.First,ifitis
right, it would be significant for understanding the implications of equality of
resources,since itwoulduprootwhatseemstobeastraightforward implication for
one kind of ambition that individuals can pursue and which can land them at a
disadvantagerelativetothosewhodonotpursueit:becomingaparent.Secondly,this
44Burley2000,p.138.45Bou-Habib2012.46Infact,Dworkinhimselfdeniesthathisviewimpliesparentsshouldinternalizeallthecostsofchildren,thoughheoffersdifferentreasonsthanBou-Habib.SeeDworkin2004,pp.361-362.47Bou-Habib2012,p.198.
20
wouldbeapowerfulpro-sharingargumentwithintheparentaljusticedebate,sinceit
starts from the most seemingly-unfriendly (to parents) premises of equality of
resourcesbutwouldneverthelessshowthatparentshaveclaimsagainstnon-parents
withoutappealingtoanyfurtherprinciplesorreasonsexternaltothebasictenetsof
equalityofresourcesitself.Moreover,crucially,theargumentfrominsuranceaimsto
showthatapro-sharingconclusioncanbereachedcompatiblywiththecommitmentto
liberalneutrality, which is to say,without appealing to the comprehensive value of
parentingasalifeplan.
InthischapterIaimtodrawattentiontosomeaspectsofthepro-sharingargument
from insurance that are either ambiguous or underexplored, and which, once
examined more closely, cast serious doubts on its success. In particular, they cast
doubtonitssuccessasanargumentforparentaljusticeassuch.Ishowthatthebest
wecanhopefromthisargumentisthatitsupportstheconclusionthatthebenefitsto
which poor people have a claim should be more generous than was previously
thought,butthatthesebenefitsshouldnotberestrictedtoparents(non-parents,too,
couldbeeligibleiftheyearnedbelowtheincomeguaranteedbysociety).Assuch,the
hypotheticalinsurancedeviceestablishesnodistinctiveclaimsforparents,butmerely
offersreasonstothinkthattheimpoverishedshouldreceivebenefitsthatcouldcover
more childrearingexpenses than they currentlydo, if thepooralsohappened tobe
parentsandwishedtospendtheadditionalincomethatway.
Ibeginbyexplaining,inthefirstsection,howtheDworkinianhypotheticalinsurance
exercise works. In Section 2.2 I lay out Bou-Habib’s insurance argument in some
detail. Section 2.3 clarifieswhich costs of children are being considered behind the
veilofignorance,andwhattheambitiontoparentadequatelyamountsto.InSection
2.4Iraisethepossibilitythattheparentalambitionisalready(satisfactorily)captured
byDworkin’stheory.ThisissomethingthatBou-Habibhastorejectinordertoclear
the way for his proposal that people would need to choose a further, more
circumscribedformofinsuranceearmarkedforchildrearingexpenses.InSection2.5I
21
raisedoubtsaboutwhethertheproposedparentalsubsidieswouldindeedbechosen
frombehindtheveilofignorance.Ishowthatitwouldbemoreprudentforpeoplenot
to insure in away that restricts eligibility to parents, and that thismeans that the
argument from insurancedoesnotgrounddistinctiveparentalclaims. I concludeby
highlightingthelimitationsoftheDworkin-inspiredargumentfrominsuranceforthe
purposesofparentaljustice.
2.1.Dworkin’shypotheticalinsurancedeviceDworkin’stheoryofjusticerequiresthateveryoneenjoyequalityofresources,where
this can obtain only if no one envies anyone else’s bundle of resources. A just
distributionofnaturalresourceswouldobtainonlyifeveryonehadequalpurchasing
power and could bid against one another for all of the available resources. The
resulting distribution would not be equal in the sense that people’s bundles of
resourceswould look the same,but itwouldbe equal in the sense thatnooneelse
wouldpreferanyoneelse’sbundle,since,byhypothesis,theycouldhaveboughtit in
thishypotheticalauction.Whicheverbundleofresourcesindividualsendupwith,this
must be because they preferred it overall, and cannot complain that they are
disadvantagedrelativetoothers.48
Theymighthavegroundsforcomplaint if,astimegoeson, theyendupatarelative
disadvantageasaresultofvarious formsofwhatDworkincallsbrutebad luck.Bad
bruteluckiscontrastedwithbadoptionluck,thelatterbeingtheresultofcalculated
gambles that couldhavebeenavoided.49Calculatedgambles that turnoutbadlyare
attributed to people’s choices (of taking on the gambles) and, as such, offer no
grounds for complaint. But various pieces of bad brute luck such as accidents,
48Dworkin2011,p.356.49Dworkin1981,p.293.
22
disabilities, or lack of valuedmarketable talents are beyond a person’s control and
lead to justifiedenvyof thosewho, throughmeregood fortune,commandrelatively
moreresources.Theanswertothisproblem,Dworkinargues,istoconsiderwhatsort
of compensation, if any, individuals themselveswouldchoose if theywereoffereda
fairopportunitytotakeoutinsuranceagainstvariousformsofbadbruteluckbefore
anyofthemmaterialized,andpaythecorrespondingpremiumsforit.
Sohowdoesthehypotheticalinsurancemarketwork?Roughly,Dworkinsuggeststhat
wearetoimagineeveryonehavingthesamebiddingpowerandfacingthesamerisk
of being struck by brute misfortune. Individuals would be making such insurance
decisions from behind a thin veil of ignorancewhere theywould knowwhat their
inclinations and ambitions were. However, no one would know their particular
exposuretotheriskofbeingafflictedbyvariousformsofbadbruteluck(forexample,
noonewouldknowwhethertheywerelikelytohaveageneticconditionthatwould
develop into a disability). No onewould have any reason to assume they are safer
fromsuchhazards thananyoneelse, though theywouldknowthatsuchhazardsdo
happen, and they would also know their incidence within their society. This is
importantbecause this iswhat ensures the fairnessof the insurancedevice, and, in
turn,thefairnessoftheresultingtaxationscheme.Itwouldbeunfairtoletonlythose
ofuswhoactuallyfacehighrisksofdisabilitiesandotherkindsofbrutebadluckbear
the burdens of this fact, and, accordingly, to have to take outmore comprehensive
insurance, or to be charged more by an insurance company, because bad luck is
morallyarbitrary.Instead,weoughttosharetherisksofbadbruteluckequally,and
decidewhatlevelofinsuranceisappropriateinthefaceofit,takingintoaccountthe
opportunitycostsoftheseinsurancechoices.
Considernowthebad luckofnothaving income-earning talent.OnDworkin’sview,
being relatively disadvantaged due to lacking the powers and talents needed to
produce valuable goods or provide services that are valued in one’s economy is
relevantlyakintobeingdisadvantagedduetodisabilityoraccidents.Wecannotknow
23
how each individual person would actually insure against the risk of finding
themselvesuntalentedinthissense,andexactlywhatpremiumtheywouldbewilling
topayforit.Butwecanarguablymakesomeconjecturesofastatisticalnatureasto
howmost peoplewould choose.One such conclusion,Dworkin argues, is thatmost
people would not run the risk of being entirely uncovered in the face of
unemployment or underemployment, if they turned out to lack talent.Most people,
that is,wouldagreetopay intoan insuranceschemeinreturnforbeingguaranteed
some level of income. It would be rational for them to pay into a scheme that
guaranteedthemwhatwemightcallasafetynet,shouldtheyeverneedone.Andthe
lower theguaranteedsafetynet, the stronger theargument thatmost peoplewould
indeedfinditworthwhiletoinsure.50
2.2.TheinsuranceargumentforparentalsubsidiesNowBou-Habib proposes that, in this hypothetical insurance situation, one kind of
considerationthatwouldbeonmostpeople’smindsastheydeliberatedabouthowto
insure against lack of income-earning talent would be how this piece of bad luck
would affect thepossibility of their being able adequately to raise children at some
pointintheirlives.Mostpeopleinoursociety,hesuggests,are“moderates”aboutthe
preferenceforchildren,whichistosaytheyknowtheymightsomedaywanttohave
children,eventhoughtheymightnotbecertainofitnow.Further,theprospectofnot
being able to do so due to insufficient funds would arguably be a particularly
distressing one. This potential ambition is so central to people’s lives, Bou-Habib
suggests,thatindividualswouldbeespeciallyconcernedtokeepthisoptionopenfor
themselvesincasetheydecidedtheywantedtopursueit.
50Idem,pp.97-98.
24
Presumably, most people in our society are either certain they want to become
parents,or theyarewhatBou-Habibcallsmoderatesaboutparenting.Aminorityof
people are certain about never wanting to parent. Given that childrearing is very
expensive, in a hypothetical choice situation where everyone faces equal risk of
having the brute bad luck of not having enough income-earning talent, most
individuals,Bou-Habibcontends,wouldwant tomakesure thatat leastsomeof the
costs of childrearingwould be publicly funded throughwhat he callspartialmeans
testedparentalsubsidies.Thesearesubsidiesthatwould“(1)coversome,butnotallof
thecostsofchildrearing,(2)beadjustedfortheincomelevelofrecipients,and(3)be
fundedthroughaprogressiveincometax”.51
In summary, the argument from insurance aims to show that (i) contrary to what
mosttheoristshavethought,aDworkin-inspiredambition-sensitiveviewofjusticeis
compatiblewithsubsidizingparentalexpenses,(ii)inawaythatwouldnotgeneralize
tootherexpensivepursuitsthatpeoplemighthave,and(iii)whichiscompatiblewith
liberalneutrality.
Ifsuccessful,thisargumentwouldofferapowerfulprincipledreasonofbasic justice
tosubsidizeparenting.Suchargumentsarestrongerthanthemorepopularforward-
lookingreasoningaccordingtowhichthestateispermittedtosubsidizeparentingas
an incentive for the creation of new generations.52Incentive-based arguments are
weaker in two respects. First, they aswas noticed earlier, for the permissibility of
sharing the costs of children rather than establish a requirement of justice. Second,
whether a state adopts costs-sharing policies can depend on contingent empirical
factsregardingthefertilitylevelsofapopulationataparticulartime,andtheaimsthe
statemighthaveforwantingtoincentivizeprocreationornot.Forinstance,ifenough
individuals in a particular society have a robust enough desire to become parents
even without any incentives, and if that state is happy with the resulting fertility
51Idem,p.208.52Folbre1994,2008.
25
levels, itmayseenoincentive-basedreasonstosubsidizeparenting.Bycontrast,the
argumentfrominsurancewouldrequirestatestosubsidizethoseparentsthatdonot
earnenoughtoparentadequatelyregardlessofcontingentfactslikeexistingfertility
levelsandthelike.
Anotherfeatureoftheinsuranceargumentthatcanconstituteanadvantageisthatthe
parental subsidies recommendedbyBou-Habib,while universally funded,wouldbe
sensitivetoparents’levelofincome.Inparticular,veryrichparentswouldnotreceive
them.Sincethesesubsidiesarejustifiedaspaymentsmeanttoredressthebadluckof
nothavinghighincome-earningtalent,therewouldbenoreasonfortheverytalented,
richparents,whohavenofinancial impedimentstoparentingadequately, tocash in
on their insurance. This would be a welcomed feature for those whomight find it
troubling to subsidize parents indiscriminately, the very poor and the very rich
alike.53
ForthepurposesofthischapterIgranttheempiricalassumptionsthattheargument
frominsurancemakesaboutthewidespreadpreferenceforparenting.Itseemstobe
true in our society that most people are at least moderates about the prospect of
parenting. I also grant that this preference is a central one for those who have it,
including for the moderates, and that this can make people particularly sensitive
aboutthepossibilityofnotbeingabletopursueitonaccountofbadluck.WhatIwant
tofocusonishowexactlysuchconsiderationsaboutparentingwouldfeatureintothe
hypotheticalinsuranceexperiment,andwhatsortsofconclusionswecandrawinlight
ofsuchconsiderations,consistentlywithequalityofresources.
53Casal1999,p.374.
26
2.3.AdequateparentingandthecostsatstakebehindtheveilInordertoassesstheargumentfrominsuranceweneedtoclarify,first,whichcostsof
children individuals are supposed to be contemplating behind the veil, and what
exactlytheparentingambitionamountsto.
First,wehavetosettlewhichcostsofchildrenarebeingconsideredbehindtheveil.In
settinguptheproblem,Bou-Habibasks:“Inmostindustrializedcountries,parentspay
thelion’sshareofclothing,shelter,food,andtoys,whilethetaxpayerfundseducation
andhealthcare.Isthatdivisionofresponsibilitycorrectorshouldthebalanceshiftin
onedirectionoranother?”54
Onewayofunderstandinghisproject, then, is to say thatbyusing thehypothetical
insurance device in the way that he suggests we are settling the answer to the
fundamental question of who should bear the costs of raising children, where the
entiretyofthesecostsisatstake.However,itisnotentirelyclearthat,asBou-Habib’s
argumentprogresses, theentire rangeof childrearing costsarebeingkept firmly in
mind.Whenmaking their insurance decisions, are individuals behind the veil only
considering the costs that parents inmost current-day developed societies have to
bearthemselves,oraretheycontemplatingtheriskofhavingtobearallthecostsof
children, including all the costs which are now in most societies already publicly
subsidized,i.e.typicallyeducationandhealthcare?
Whathingesonclarifyingwhichcostsofchildrearingareatstakebehindtheveil?If
individuals are contemplating all the costs of children, on the one hand, this
strengthensthecontentionthatpeoplewouldbeextremelydistraughtbythethought
ofnotearningenoughtobeabletogivechildrenevenminimallyadequateeducation
or health care – mainly because most people would be facing this risk, and, if it
materialized, it would be disastrous for both children and parents. One would
54Idem,p.201.
27
presumablyhavetoearnclosetothehighestlevelsofincomepossibleinone’ssociety
tobeabletocoverallofthecostsofraisingchildrenontheirown.Butthisalsomeans
that(a)alargeproportionofsocietywouldbeeligibletocashin(asmostpeoplewho
parentwillnotearnatsuchhigh levelsof income),andtheamountthat theywould
needtocashintocoverallthosecostswouldbefairlysubstantial.Thiscoulddriveup
thecostofpremiumstothepointofrenderingtheinsuranceschemenon-prudent.
Ofcourse,asBou-Habibiscarefultopointout,individualswouldnotchoosetocover
the full cost of whichever range of childrearing expenses are at stake anyway,
preciselybecause thiswouldraise thecostofpremiumstoomuch.But if,under the
veil,thewholerangeofchildrearingcostsisatstake,itisplausibletocontendthatthe
partialmeanstestedparentalsubsidiesthatwouldbechosenwouldonlyamounttoa
sumlargeenoughtocoverforthecrucial,andveryexpensive,costsofhealthcareand
education, and little more. In this case, the division of costs between society and
parentsthattheinsuranceargumentwouldrecommendwouldnotbemuchdifferent
from the division of costs that already exists in most Western societies, in which
parents are expected to bear all the costs of childrearing apart from, at least, basic
health care and education, which is publicly funded. Cover much more, and the
premiumswouldskyrocket.
Thisresultwouldrender theargument from insuranceconsiderably less interesting
fromapracticalpointofview,thoughitwouldremainofgreattheoreticalinterestthat
subsidizing things like health care and education for childrenwould be amatter of
justice to (non-rich, unlucky) parents, rather than to the children themselves –
arguably controversially. It would also imply, perhaps problematically, that the
childrenofrichparentsarenotentitledtopubliclyfundedhealthcareandeducation.
Weneedtolookatwhetherthecostsofhealthcareandeducationforchildrenwould
be independentlycoveredwithinDworkin’s frameworksuchthat theywouldnotbe
onthetablefortheindividualsbehindtheveilcontemplatingtheriskofnotearninga
28
generousincome.AdirectargumentforthisisprovidedbyDworkinhimself,alsoby
appeal to thehypothetical insurancedevice,but asamatterof rectifyingbrute luck
inequality affecting children themselves, rather than parents. He considers the
possibility of being born to poor or unemployed parents as a distinct piece of bad
brute luck that children themselves are victims of. They could choose to take out
insurance against this, or, rather, we can imagine that prudent guardians of their
interestscould:
Wecanask:Howmuchinsurance,atwhatterms,wouldprudentguardiansbuyontheir
behalf, on the understanding that premiums were to be paid later, on some long-
delayed instalment plan, by the children themselves? It seems plain that a prudent
guardianwould buy insurance to provide enough coverage to allow an infant to live
withitsownparents,andtoreceiveenoughmedicalcareandeducationtosurviveand
qualifyforemploymentwhenappropriate.55
Dworkin’s argument shares the potentially problematic implication that children of
richparentsarenotentitledtopubliclyfundedhealthcareandeducation,butIleave
that to one side. If we take this argument on board (or, really, any argument that
wouldbecompatiblewithDworkin’sframeworkandwhichwouldsecurehealthcare
andeducationforchildrenonindependentgrounds),adultindividualsbehindtheveil
who are least “moderates” about their preference forparentingwouldonlyhave to
worry about being unlucky enough to fail to earn at a level that could cover the
remainingcostsofchildrearing.Thesepresumablyincludethecostsofhousing,food,
clothing,schoolmaterials,toys,extracurricularactivities,andsoon,which,nodoubt,
arestillsubstantial,butconsiderablylessthanthesumsneededtoalsoprovidehealth
careandeducation.
Justhowworrisometheriskofnotearningenoughtoprovide for theseothercosts,
andwhether itwouldmake sense forpeople to insure against it, dependsona few
55Dworkin2002,p.339.
29
otherfactors.Theseinclude,first,whattheparticularstandardofadequateparenting
requiresinthesocietyweareimagining.Ifthemoralstandardforadequateparenting
isverylow,forexample,ifitcoincideswiththelegalstandardofnon-neglectandnon-
abuse,itisplausiblethatmostpeopleinthissocietywouldearnenoughtocoverthe
remainingcoststomeetthisminimalstandard.Payingapremiumto insureforthis,
then,mightnotseemappealing.
Second, italsodependsonhowmanychildrenthey think theywillwant tohave.Of
course, if they plan to havemore than three children, say, suddenly it could seem
difficulttomeeteventheminimalstandardsofadequatechildcare,andsigningupto
aninsuranceschemewouldbecomeattractiveagain.Butsurelyitisnotthecasethat
prudentpeoplebehindtheveil,manyofwhomhaveonlyamoderatepreferencefor
parenting, would want to protect the opportunity to parent several children
adequately, given the corresponding premium costs thatwould be attached to that
choice.Most likely,peoplewouldwant toprotect theopportunity toparent at least
onechildadequatelysincethemostdistressingriskofallisthatonemightmissouton
theopportunitytoparentevenonechildduetoeconomichardship.Butonceagain,if
themoralstandardofparentingadequatelyisaslowasthestandardofnon-abuseand
non-neglect, most people will be able to meet this threshold for at least one child
through theirown income-earning talent.Therewouldnotbeamajoritybehind the
veilwillingtotakeoutinsuranceagainsttheriskofnotbeingabletoparentevenone
child by such aminimal adequacy standard, especiallywhen they know that health
careandeducationarealreadycoveredindependently.
Ofcourse, themoralstandard forparentingadequatelycould(andshould)bemuch
higher than a legal threshold of non-abuse and non-neglect. Perhaps parenting
adequatelyrequiresonetoraisetheirchildinsuchawaythatshewillhaveanopen
future.56Ofcourse,thehigherthestandardofadequateparenting,thehigherthecosts
associated with that. But we are here keeping firmly in mind that health care and56SeeFeinberg1992.
30
education, and perhaps even a bit more than that, are already secured, so this
presumablydoessomeoftheworkofsecuringanopenfutureforchildren.
There are many unknowns, and many empirical complexities here that prevent us
fromdrawinganyfirmconclusionaboutwhetherpeoplewouldinsureornot:itreally
dependsonwhatthestandardforadequateparentingis inaparticularsociety,how
expensivemeetingitwouldbeforparentsthemselvesiftheydidnotinsure,andhow
costly the premiums would be if they did insure. But note, also, that we are here
assumingthat, inmakinghypothetical insurancedecisions,parentsreallywouldcare
aboutmeetingthosestandards,whatevertheyare.Throughoutmostofhistorypeople
havehadchildrenwithoutcareful thoughtabout theirability tomeetchildren’s just
claims(atleastifwehaveademandingviewofwhattheseare),buthereweseemto
beassumingindividualsbehindtheveiltobemodern,enlightenedcitizensthatwould
rathermissoutontheexperienceofparentingaltogetherthanbeinadequateparents
in the sense of not meeting this threshold of adequate parenting. This would be a
welcomepsychological feature of individuals behind the veil, if true, but remember
that behind Dworkin’s veil we have “real” people, not fully rational and morally
impeccableversionsofthem.Wedohaveastylizedversionoftheminthesensethat
theylackinformationabouttheirownexposuretocertainrisks,andweareassuming
themtobeprudent.ButitistheriskofusingtheDworkinianhypotheticalinsurance
device to design a system of just compensation that the choices that actual people
would make behind a thin veil of ignorance might fall short of what we would
otherwisedeemmorallydesirable.
This brings us to another important feature of the argument from insurance that
would need to be unpacked.What exactly is folded into the “opportunity to parent
adequately”thatindividualswouldwanttoprotectisrathervague.Bou-Habibwrites
thatpeoplewouldwanttomakesurethattheywillnotbe“forcedbytheireconomic
difficulties tomissouton thatexperience,or that theydonot facesevereeconomic
difficultiesduringthatexperience,”andfurtheraddsthatsubsidieswouldbegivento
31
peopleonly“so longasparentsneedthemtoenjoyadecentstandardof life.”57This
suggeststhatheisnotonlythinkingofthosewhowouldearnattheverylowestlevels
ofincomesuchthattheywouldnotbeabletoprovideevenaminimallydecentlifefor
a child (andwould, therefore, have to giveuponhaving a child altogether). This is
what I have been assuming so far is the risk theywant to avoid. Bou-habib is also
thinking of those who, while being able to give a child a minimally adequate
upbringing(whateverthatmeans),wouldfacesevereeconomichardship.Buthowis
thiseconomichardshiptobeunderstood?Nothavinganydisposableincomeleftafter
providingminimallyadequatecareforthechild?Nothavingenoughresourcesleftto
enjoy an adequate range of options for flourishing besides parenting? Once again,
where exactly we set the standard herewill influence the insurance decisions that
wouldbemadebecause itwouldaffect(a)whatproportionofsocietyactually faces
(and would be worried about) the risk of not earning enough to meet it; (b) how
distressingitwouldbeifthatriskmaterialized.
Allof this is just to flagthat, intryingtounpack importantempiricalandnormative
issuesthattheargumentfrominsurancerelieson,suchaswhatrangeofchildrearing
costsareatstakebehindtheveil,andwhattheopportunitytoparentatleastonechild
adequately would involve, there are complexities that, depending on how they are
resolved,mightalreadycastdoubtontheargument’sconclusions.
IntherestofthischapterIwillchooseaninterpretationofthesetwoelementsthatI
findmostplausible,thoughIcannothopetofullyclarifythem.Iassumethatwhatisat
stake behind the veil are all the costs of childrearing apart from health care and
education,andthatindividualswanttosecuretheopportunitytoadequatelyraiseat
least one child (where the adequacy standard is higher thanmerely non-abuse and
non-neglect) if they ever decide to become parents, while still enjoying a decent
standardof living themselves (whatever this standard is in theirparticular society).
Thatis,Iassumethatpeoplewouldnotonlybeworriedaboutbeingsopoor,through57Idem,p.201.
32
no fault of their own, that they could not adequately raise even one child, but also
aboutnotbeingwell-off enough toavoid living inpoverty themselvesas a resultof
beingtheuntalentedparentsofevenjustonechild.Thisiswhat,fromnowon,Iwillto
refertoastheparentalambitionthatindividualsmightwanttosecurebehindtheveil.
ThequestionIwanttoraiseinthenextsectioniswhyitisthatsuchparentalconcerns
would not already be captured by Dworkin’s generic level of underemployment
insurance.
2.4.Underemploymentinsurancewouldberathermodest:sowhat?Bou-Habibsuggeststhat,indecidingwhatlevelofinsurancepeoplewouldpurchaseif
theyallfacedthesameriskoflackingmarketabletalentandhadequalopportunityto
insure, theprospectofnotbeingable topursue theparentalambitionwouldplaya
centralroleintheirinsurancedecisions.
Individuals face the risk of low incomedue to badbrute luck andneed to decide the
kindofinsurancetheywanttopurchaseagainstthatrisk.Onetypeofinsurancecover
theywillwishtopurchaseforthemselvesisacashincomesupplementpaidoutincase
theirincomelevelshouldfallbelowastipulatedthreshold.(…)Thequestionweneedto
addressiswhetherontopofthiscashincomesupplement,individualswouldpurchase
insurance cover that is tied to childrearing expenditure (a parental subsidy
guarantee).58
This suggests thatweareassuming that thecash incomesupplement thateveryone
below a certain threshold of income would receive (what Dworkin calls
underemployment insurance) would not be generous enough to cover childrearing
expenditures and allow parents to lead a decent life themselves. Thismaywell be
58Idem,p.207,emphasisinoriginal.
33
true. Dworkin argues that the level of insurance most people would choose in a
hypotheticalscenariowouldbehigherthanwhatiscurrentlyofferedincountrieslike
the United States or Britain by way of transfers for minimum wage or
unemployment.59But he also showswhy itwould not be sensible to insure for the
maximum level of income possible. Thiswould involve, as Dworkin puts it, “a very
highchanceofgainingverylittle.”60Thisisbecausemostpeoplewillendupneeding
to cash in on their insurance (most people will not have the talent to earn at the
maximlevelontheirown).Asaresult,thepremiumthatpeoplewouldhavetopayin
order to support this schemewouldbe sohigh that, over time, itwill approach the
expected gain. The expected return, when compared to the cost of purchasing the
insurance,will not beworth the gamble. This iswhy,Dworkin says, “the lower the
incomelevelchosenasthecoveredrisk,thebettertheargumentbecomesthatmost
people given the chance to buy insurance on equal termswould in fact buy at that
level”.61
ThisispresumablywhyBou-Habibthinksthat,givenjusthowexpensiveadequate
childrearing is (even with health care and education already secured),62 the
underemploymentbenefits recommendedbyDworkinwouldnot suffice to allow
individualstoadequatelyparentandofferthemtheopportunitytoenjoyadecent
standardoflivingwhiledoingso.
However,theunderemploymentinsurancethatDworkinarguespeoplewouldbuy
mayalreadybetakingallrelevantfactorsaboutpeople’sambitionsintoaccount.If
so,theneventhoughtheinsurancelevelagreeduponwillnotbeverygenerous,itis
the optimal level given the attached costs. For there is no reason to assume that
59Idem,p.97.60Idem,p.96.61Idem,p.97.62SeethecostscitedinSection1oftheIntroduction.Mostpertinently,accordingtofiguresfromtheChildPovertyActionGroup,thecostsofraisingachildin2017frombirthto18yearsoldintheUK,excludingthecostsofhousing,healthcare,andcounciltax,is£75,436foratwo-personhousehold,and£102,627foraone-parentfamily(Hirsch2017).
34
people’s potential parenting ambitions are completely overlooked by Dworkin
whenherecommendsarelativelylowlevelofunderemploymentbenefit.Parenting
ambitions,aswellasanyotherpotentiallifeplanspeoplemightbeworriedabout
behind the veil, are presumably already captured by the process of deliberating
aboutunderemploymentinsurance.
Soitmaywellbethattheincomeguaranteedinthefaceofbadbruteluckisnotvery
generouswhencomparedtothecostsofparentingadequatelyinoursociety.Butwhy
shouldthissuggestthatweneedtolookforadditional,discreteinsurancesolutionsas
a response to the same piece of brute bad luck (lack of marketable talent) that the
genericunderemploymentinsuranceclaimstotakecareof?
To my mind, what this more readily suggests is another potential limitation of
appealing to thehypothetical insurancedeviceasabasis forparental justiceclaims:
onceagainthesubstantiveresultsyieldedbythehypotheticalinsurancedeviceseem
tofallshortofwhatmanywouldintuitivelythinkismorallyrequired.
Bou-Habib could reply that what his argument assumes is not necessarily that
parental ambitionsarebeing completelyoverlooked,but that the singularnatureof
theparentingambition,anditsbearingonpeople’shypotheticalinsurancedecisions,
has not been adequately taken into account. Theorists like Dworkin may have
underestimatedthepervasivenessandforceofparentalambitions,whichmayexplain
whytheydidnotargueforalevelofunderemploymentbenefitsthatcouldcoverthe
costs of raising at least one child, or why they did not consider other insurance
possibilities like the partial means tested parental subsidies. If this is true, their
conclusionsaboutwhatthehypotheticalinsurancewouldorwouldnotcoverneedto
berevisedaccordingly.
AssumingBou-Habib is right about that,wemightwonderwhy taking the parental
ambition seriouslywould not lead us to the conclusion that peoplewould insure a
35
higherlevelofincomeinthefirstplace(atleastcomparedtowhatDworkinsuggests
they would). If people care so much about the parental ambition as Bou-Habib
suggests they do, would they not find it worthwhile to pay the higher premiums
associatedwithsecuringalevelofunderemploymentbenefithighenoughtocoverthe
costsoftheparentalambition?CallthistheIncreasedUnderemploymentsolution.
Ibelieve this is an importantquestion, anda case couldbemade thatpeople really
mightinsureahigherlevelofunderemploymentbenefits,dependingonanumberof
empiricalfactorstodowiththedetailsofhowmuchhigherthatlevelofbenefitwould
have to be and so on. But Bou-Habib might reply that this would be taking his
suggestion too far. We may have underestimated the importance of the parental
ambition,buteven ifwecorrected for that, itwouldnot lead to the conclusion that
people would be willing to pay substantially higher premiums to protect it. The
premiums would be substantially higher because not only would the guaranteed
income be higher, but also, more people would be eligible to cash in on their
insurance.Asmorepeoplewould fail toearnat thathigher levelof income through
theirownincome-earningtalents,theywouldbeeligibletocashintomakeupforthe
differencebetweenthatguaranteedincomelevelandwhattheyactuallyearn.
Facedwith thisproblem,wecouldgo inat least twodirections.Perhaps,afterall is
saidanddone,wewouldendupwithaguaranteedlevelofincomethatwouldonlybe
slightly higher than what Dworkin suggested it would be (call this the Slightly
Increased Underemployment solution). For even if we revise people’s hypothetical
insurancechoicesinawaythatreflectstheimportancethattheparentalambitionhas
formostindividuals,intheendpeoplewouldbedeterredfromraisingtheguaranteed
incomeleveltoomuch,giventhehighpremiumstheywouldhavetobeartosustain
thatscheme.I,forone,thinktheSlightlyIncreasedUnderemploymentsolutionwould
alsobeaperfectlyplausiblecontenderastowhatinsuranceoptionindividualswould
choosebehindtheveil.
36
Alternatively,we could go inBou-Habib’s direction,whichpurportedly allowsus to
revise our hypothetical insurance scheme in a way that reflects the importance of
parentingambitions,andtodosoinawaythatismorecost-effective,andhencemore
likelytobechosenfrombehindaveilofignorance.
Let us take stock. I am granting that the importance of the parental ambition on
individuals’ hypothetical insurance decisions may have been underestimated by
theoristslikeDworkin.Nowthequestionis,whatisthemostplausiblewaytorevise
ourconclusionsabouthypotheticalinsuranceinlightofthis?ThisiswhatItakeupin
the next section. In particular, I raise some questions about whether Bou-Habib’s
proposalisindeedsignificantlymorecost-effectivethanthealternatives.
2.5.Areparentalsubsidiesthemorecost-effectiveformofinsurance?Thepurportedadvantageofchoosingmeanstestedparentalsubsidiesoverboththe
IncreasedUnderemploymentandtheSlightly IncreasedUnderemploymentsolutions
isthat it ismorecost-effective. ItproducesbetterresultsthantheSlightlyIncreased
Underemploymentbecause, Iassume, itwouldbemoregenerous, so itwouldcover
morechildrearingexpenses.Ontheotherhand,itismeanttobesignificantlycheaper
thantheIncreasedUnderemploymentbecause,whilebothwouldbeabletocoverthe
same,more extended, range of childrearing costs, fewerpeoplewould be eligible to
cashinthechildrearingsubsidies,namelythosewhofailtoearnatthestipulatedlevel
andareparents.So,presumably,prudentpeople inthehypotheticalchoicesituation
wouldchoosethemeanstestedparentalsubsidiesoverthoseothertwooptions.
Anotherreasonwhyparentalsubsidieswouldallegedlybechosenoversimplyraising
thegenericunderemploymentbenefits ispresumablyduetotheuniquecharacterof
the parental ambition. Bou-Habib argues that the parental ambition displays some
37
characteristics that other expensive life plans do not share (which Iwill turn to in
somedetail later), and this explainswhy individualsmay be contentwith finding a
reasonablyaffordablewayto insuretheparentalambitionratherthantakingonthe
costsofinsuringinawaythatcouldcoverforotherexpensiveambitionsaswell.
It ishelpful toseetheargumentfrominsuranceasaproposalto instituteatwo-tier
underemploymentcompensationschemeinresponsetotheriskoflackofmarketable
talent.Thefirstthresholdissetbythegenericunderemploymentbenefitsthatpeople
areentitledtomerely invirtueofbeingunabletoearnat that incomelevel through
theirowntalents.Theformthiscompensationtakesisacashsupplementtomakeup
for the difference between what one earns and the income level that Dworkin’s
underemployment insurance guarantees. A second threshold is set at the higher
income levelrequiredtoaffordanexpensivepursuit like theparentalambition.Call
thistheParentalAmbitionThreshold.Toclaimthebenefitsofferedatthishigherlevel
ofincomeonemustbebothunluckyenoughnottobeabletoearnthatsortofincome
onone’sown,and,ofcourse,onemustalsobeaparent.Thedifferencebetweenwhat
one earns and the income guaranteed at this threshold is made up of subsidies
earmarkedforchildrearingexpenses.Theargumentforbothformsofcompensationis
that,ineachcase,theywouldhavebeenpartofthecontentoftheinsurancepackage
chosenbyprudentindividualsbehindaveilofignorance.
Iwanttosuggestthat,despitewhatmayfirstappear,itisnotclearthatthistwo-tier
scheme is the more cost-effective insurance option compared to the Increased
Underemployment solution. The latter would involve simply raising Dworkin’s
genericunderemploymentthresholdtotheleveloftheParentalAmbitionThreshold
and giving everyone who falls below it a cash supplement. This cash supplement
wouldnotbeconditionaluponone’sbeingaparent.Anyonewouldbeeligiblemerely
invirtueofearningbelowtheIncreasedUnderemploymentthreshold.
38
Instead,Iargue,theparentingambitionisnotasuniqueasBou-Habibarguesitis,and
peoplewould beworried about the effects of bad luck on their ability to pursue at
leasttwootherambitions.Asaresult,theywouldbemovedtoinsurebyreasonsthat
arenotuniquetoparenting,andtheywouldfinditmoreprudenttoinsureinaway
thatdoesnotrestricteligibilitytoparents.Myconclusionisthat,atbest,theargument
frominsuranceonlygivesusreasonstoraisethe levelofgenericunderemployment
benefitsfromwhathasbeenpreviouslythought,butdoesnotgiveusreasonstofund
parenting specifically through targeted subsidies. If true, this conclusion would
undercut theargument from insuranceasanargument forparental justice. For, as I
explained in the Introduction, a good argument for parental justice must provide
reasons of basic justice to share the costs of children that refer to the features of
parentingspecifically,ortoitseffects.Iarguethattheargumentfrominsurancefails
toprovidesuchreasons.
To begin with, recall that the two-tier schemewould purportedly be cheaper than
IncreasedUnderemploymentbecausethecategoryofpeopleeligibletocashinonthe
formerismorerestricted:the“extra”fundscanonlybeawardedtothosewhoactually
endupparentingandfallbelowtheguaranteedincomelevelforparents.OnIncreased
Underemployment non-parents would be eligible too, provided only that they fall
belowthestipulatedthreshold.
It is clear that the two-tier schemewill necessarily be somewhat cheaper than the
IncreasedUnderemploymentscheme.Anysocietyhasbothparentsandnon-parents,
andapolicywhereonlyparentscancashinwillbecheapertosustainthanonewhere
everyonecan.But if it turnsout thatanoverwhelmingmajorityofadultsdoendup
parenting, it isnot soobvious that thepremiumswillbesignificantly lower thanon
IncreasedUnderemployment.Weknowthatparentstendtomakeupthemajorityof
adults in just about any society. Moreover, any policies that provide parental
subsidies,liketheoneBou-Habibfavors,havethepotentialtoactasanincentivefor
39
peopletobecomeparents,whomayotherwisenothavebeen.63Soitisessentialtosee
howmanypeople, outof all thepeoplewhoarenot talentedenough to earnat the
ParentalAmbitionThreshold,wouldremainnon-parents,therebyallegedlyloosening
the strain on the scheme enough to make it a considerably cheaper option than
Increased Underemployment. It is hard to find a clear answer to this empirical
question,butitisworthhighlightingitsimportance.
It is worth highlighting its importance because the Increased Underemployment
solutionhasitsownadvantages:itsecurestheopportunitytopursueotherexpensive
preferences that are similarlywidespread and central to people’s conception of the
goodaschildrearingis.
Bou-Habibargues thatpeoplewouldbewilling to secure theopportunity topursue
the parental ambition in awaywhat theywould not bewilling towhen it came to
other potential expensive life plans such as, for example, mountain-climbing.64The
former ambition, he maintains, is complex in that it depends on inclinations and
circumstancesthatmaychangewithtime,inawaythatapreferenceforbecominga
mountain-climber does not. Thus, a person behind the veil of ignorance who was
unsureshewantedchildrencouldstillbeaware that thismaywell changeand that
shewouldwant toparentoneday.Furthermore, theparentingambition,Bou-Habib
claims, ispotentiallyof central importance toour lives in away that ambitions like
mountain-climbing are not. The prospect of not having enough resources to parent
adequately, shouldweeverdecidewewanted to,wouldbesodistressing thatmost
individuals would want to insure against this possibility even if they were
“moderates”aboutit,thatis,eveniftheywereunsure,atthetimeofbuyinginsurance,
thattheywouldwanttopursueitatsomepointinthefuture.Bycontrast,writesBou-
Habib:
63Hoem1990,2005.64Idem,pp.209-210.
40
Unlike having children, mountain-climbing is a hobby that we can form fairly
unchangeablepreferencesabout–thoseofwhoaredisinclinedtopursueitarelikelyto
remain so in the future. Furthermore, in the unlikely event that those currently
uninterested in thathobbyshouldwant topursue itoneday, their liveswouldnotbe
shatteredshouldtheyhavetoforgopursuingitforlackoffunds.65
So,Bou-Habibconcludes,mostpeople,both thosewhoarecertain theywillwant to
parent and thosewho are unsure about it, would bewilling to insure the parental
ambition but not other ambitions. This is why they would be inclined to choose
parental subsidies over generic underemployment benefits, since it is the parental
ambitionthatmostwouldbeworriedabout,and,giventhis,itwouldbethemostcost-
effectiveoptionsinceonlyparentswouldbeeligibleforthem.Forpeopletobewilling
to takeon thehigherpremiums thatwouldaccompanyanall-purpose cash subsidy
thatbothparentsandnon-parentswouldbeeligiblefor,andwhichcouldbeusedfor
purposes other than childrearing, it must be the case, Bou-Habibwrites, that “(1)
apart from childrearing, there existed a wide range of other expensive first-order
choiceswemightwishtopursueinthefutureand(2)ourinabilitytopursueanyof
those other choices, were we eventually to settle on one, would cause us great
distress.” He continues, “[i]t is doubtful, however, that both of these conditions are
satisfied.”66
In response, however, it is unclear, first, why Bou-Habib believes we need awide
range of other preferences for it to be sensible for us to opt for the generic cash
supplement (which is the same as saying that they would opt for the Increased
Underemployment solution). Second,wecan thinkofat least twootherpreferences
thataresimilartotheparentingambitionintherelevantrespects:theopportunityto
changeone’sprofession,andtheopportunitytoenjoyaminimallyadequateamount
offreetime.FollowingJulieRose,wecanunderstandfreetimeheretomeantimethat
is not committed to meeting one’s own basic needs or the basic needs of a65Idem,p.210.66Idem,p.211.
41
dependent.67Thesepreferencessharewiththepreferenceforparentingtherelevant
characteristicsthatBou-Habibbelievesjustifytheconjecturethatpeoplewouldwant
toinsuretheopportunitytopursuethemiftheysowishedatsomepointinthefuture.
Toseethis,letusfirstisolatetherelevantcharacteristicsoftheparentalambition.
Arguably,theparentalambitionis:
(i) Sharedbymostindividualsinoursociety:mostpeoplewouldeitherknow
for certain that they want to become parents one day, or are at least
“moderates”aboutit,namely,theyareopentothepossibilitythatifvarious
circumstancesobtain(forinstance,iftheyfindtherightpartner),they,too,
might choose to become parents despite not being settled on this
preferenceatthetimeoftheinsurance-makingdecision.
(ii) Likelytobeaffectedbysomeone’sunluckycircumstances,namelyby lack
ofincome-earningtalent.
(iii) Would cause great distress if the risk of being prevented from parenting
duetoone’sunluckycircumstanceswereactualized.
Given these three features, Bou-Habib concludes that most people behind the veil
wouldwant to take out some insurance to protect themselves against not affording
theirparentalambitioniftheyturnedouttohavebrutebadluck.Andthen,ofcourse,
thequestionis,whatkindof insurancewouldbetakenout,whereBou-Habib,aswe
have seen, argues that an affordable package would involve partial means-tested
parentalsubsidies.
Consider, now, the preference for securing the opportunity of changing one’s
professionandtheopportunitytoenjoyaminimallyadequateamountoffreetime.
67Rose2016,pp.48-49.
42
(i) Itseemsplausible tosaythatmost individuals inoursocietywouldshare
thesepreferences.Considerchangingone’scareerfirst.Certainly,notmany
peoplewouldknowatthemomentofinsurancedecision-makingthatthey
will want to change professions someday. But this preference, like Bou-
Habibsaysaboutthepreferenceforhavingchildren,iscomplex.Inlightof
the changes that their life circumstancesmight undergo, theymightwell
find themselves in the position of wanting to, or being forced to, change
careers, or change jobs. In the case of the preference for an adequate
amountoffreetime,perhapssomepeoplebehindtheveilwouldknowfor
certain that they will always prefer to work long hours and enjoy more
resources than enjoy free time. But they will likely be a minority. Most
individuals want to ensure they have an opportunity to enjoy at least a
minimalamountoffreetimethroughouttheirworkinglives.
(ii) Bothofthesepreferencesarelikelytobeaffectedbythebrutebadluckof
lacking income-earning talent. Changing professions would involve a
substantial investment in termsof timeandresources intoacquiringnew
skills. The time invested into re-skilling would be time taken away from
workingproductively,whichissomethingthatonlythosewhowouldhave
acquired enough resources up to that point could afford. Acquiring new
skillswouldalso involvepaying foreducationandtrainingappropriate to
one’s alternative professional path. Among all those who share the
preference of changing their professions, some will encounter more
obstaclesthanothersindoingso,duetolackofresources.Insofarastheir
lackofresourceswouldbearesultofthebadluckofnotenjoyingenough
income-earning talent, they would be entitled to compensation to the
extentthattheywouldhaveinsuredagainstthisrisk.Forsimilarreasons,
someofthosewhowouldprefertohaveanadequateamountof freetime
will enjoy different prospects for fulfilling that preference. The unlucky
might earn so little that they would be precluded, or severely hindered,
43
from affording either to change their profession, or to enjoy adequate
amountsoffreetime.
(iii) Lacking the resources to afford adequate free time or to change one’s
professionseemsdistressingenoughthatmostindividualswouldbewilling
totakeoutsomeformofinsurancetoprotectthemselvesagainstthat.Given
howmuchof people’s time is spentworking, being stuck in a job or in a
profession that is not conducive to one’s flourishing seems a bad enough
prospect thatpeoplewould insure themselves against that, provided that
thepremiumstheywouldhavetopaywouldbeaffordable.Theywouldalso
beanxious to insure themselves,especially, if itwas true, in theirsociety,
that technologywas rapidly developing in away that wouldmakemany
jobsobsolete–inthatcase,onceagain,mostwouldwanttomakesurethey
couldacquirenewskillsfordesirableandlucrativejobs.Someofthesenew
skillswill involve becoming proficient in using new technologies oneself.
Suchskillsmightbeverytime-andresource-consumingtoacquire.Andas
faras freetime isconcerned,contemplatingtheriskofaworking lifetime
byandlargedeprivedofanadequateamountoffreetimeseems,onitsface,
worrisome enough to promptmost individuals to take some precautions
againstthatpossibilitymaterializing.
Ina footnote,Bou-Habibhimselfacknowledgesthat theopportunitytochangeone’s
careerisapreferencesimilartokeepingone’sparentaloptionopen,buthesuggests
that “the prudent way to cover this optionwould be through an in-kind insurance
scheme, rather than through an indiscriminate cash equivalent of a parental
subsidy.”68Sohebelievesthattheexistenceofthispreferencewould, ifpeoplewere
prudent,merelywarrantthechoiceofsigningupforanin-kindbenefitschemesuch
as,perhaps,vouchersforre-skillingcoursesontopoftheirgenericunderemployment
insuranceandtheparentalsubsidies.68Idem,p.215,fn.13.
44
Butif,asIarguedabove,tothesetwowecanalsoaddthewidespreadpreferencefora
minimallyadequateamountof freetime, it isnotsoclearanymorethatBou-Habib’s
proposal is theprudentone. Itmaywellbe thatacashsubsidythatcanbeused for
eitheroneofthesethreepursuitswouldbetheprudentoptionindeed.
Thisconclusionwouldbeevenmoreplausibleifwecanidentifyevenfurthersimilar
preferences that most would share behind the veil. Think, for example, of the
preference of living in a place one considers conducive to one’s flourishing.Noone
wouldwanttobestuck living inaplacethattheydespise,andtheir income-earning
talent might impede their ability to freely choose a place to live in a way that a
talentedpersoncanafford.Perhapsindividualswouldwanttosecuretheopportunity
tohaveaccesstolivinginabigcity,or,conversely,closetonature,ortoliveinaplace
whereone’sspousewouldhavebettercareerprospects,orwhereone’schildrencould
havebetteropportunities.
Themorepreferencessimilartoparentingwecanidentify, themoreprudentanall-
purposesubsidythatcanbedirectedtoanyofthesepursuitsseems.Thatis,themore
prudenttheIncreasedUnderemploymentsolutionseems.Thisconclusionisbolstered
bymyearlierdoubt regarding thenumberofpeopleeligible tocash in theparental
subsidies. If the number of people eligible on Bou-Habib’s two-tier scheme is not
considerably lowerthanonIncreasedUnderemployment,andgiventhatthereareat
least two other first-order preferences, if not more, that most people would be
anxious to secure when faced with brute bad luck, going for the Improved
Underemployment solution starts to look themore prudent option. This is because
mostpeoplecanbefairlycertainthat,shouldtheyenduppoor,eveniftheydonotend
up being parents theywould very likely still want to pursue one ormore of these
otherpreferences:havingenoughfreetime,changingone’sprofessionorrelocatingto
abetterplaceto live in.Soitwouldbeimprudenttorestricteligibilityonlytothose
whodoendupasparents.
45
All of this was on the assumption that perhaps Dworkin failed to register the
importanceoftheambitionsthatmostpeoplewouldwanttoinsureincasebadluck
struck. This assumption had some plausibility when parenting was the only
preferencewe thoughtDworkin failed to account for. But, itmust be said, oncewe
start to recognize that theremaybe anumberofpreferences thatpeoplewouldbe
willingtohavesecured,thequestionIraisedinSection2.4abovereturnsevenmore
forcefully. Would Dworkin’s recommended underemployment threshold not have
registeredalreadytherelevanceofallofthesepreferences?Thisseemsunlikely.
Butletussaythat,somehow,thatisthecase.Recall,also,thattheSlightlyIncreased
Underemploymentsolutionremainsasortofmiddle-groundcontendertoo.Assuming
thatDworkinfailedtoregisterallofthesepreferencesadequately,andassumingthat
oneisnotconvincedbytheadvantagesofadoptingIncreasedUnderemployment,one
could findSlightly IncreasedUnderemploymentpreferable to the solutionproposed
by Bou-Habib. Slightly Increased Underemployment would allow us to revise the
underemploymentbenefitsinawaythatrecognizedthattherearemorepreferences
apartfromparentingthatpeoplewouldgreatlycareabout,evenasmoderates,behind
theveil.At thesametime, itwouldrecognizethatperhaps individualswouldnotbe
preparedtotakeonthehighpremiumcostsofIncreasedUnderemployment.
2.6.ConclusionIhopetohaveshownthattherearegoodreasonstobewearyoftheargumentfrom
insuranceforthepurposesoftheparentaljusticedebate.Foronething,manyinthe
pro-sharing campmay not be satisfiedwith the substantive results yielded by any
argumentthatsettlespeople’sentitlementsbyreferencetowhatactualpeoplewould
havechoseninahypotheticalinsuranceexercise.Secondly,toevengetofftheground,
theargumentreliesontheassumptionthatthegenericunderemploymentinsurance
46
doesnotadequatelytakeintoaccounttheimportanceoftheparentalambition,which
isitselfaquestionableassumption.Thirdly,evenifthatassumptionwerecorrectand
we sought to revise our views aboutwhat sort of insurancewould be chosen, it is
doubtfulthatthepartialmeanstestedparentalsubsidieswouldbechosenbehindthe
veil.Forwecanidentifyseveralotherpreferencesthatbearsimilarcharacteristicsto
the parental ambition andwhich peoplemight alsoworry about protecting. In that
case,SlightlyIncreasedUnderemploymentandIncreasedUnderemploymentlooklike
verygoodcontendersforthetypeofinsurancethatmightbechosenbehindtheveil.
Thesetwooptionswouldhavetoberejectedbeforeconsideringthemorerestricted
parentalsubsidiesoption.
If it is true,as Ihaveargued, thatSlightlyIncreasedUnderemploymentorIncreased
Underemploymentwouldbepreferredbehindtheveilofignorance,thenanargument
fromhypothetical insurancewouldnotbeofmuch relevance to theparental justice
debateassuch.If thebestconclusionwecanget isthatthepoor,betheyparentsor
not, are entitled to somewhatmore generous underemployment benefits than was
previously thought, there is nothing byway of distinctive parental claims here. For
anyonewhoearnedbelowthestipulatedthresholdwouldbeeligibletocashinandto
spendthatincomehowevertheywanted,allmerelyinvirtueofthefactthattheywere
unluckyonthejobmarket.
True, ina limitedsensesocietywouldendupsubsidizing (poor)parents, insofaras
people earning less than the guaranteed amount would receive universally-funded
benefitswhich, if theyalsohappened tobeparents, theycoulduse for childrearing.
Butthisisnomoreadistinctiveargumentforparentaljusticethanitisanargument
for,say,justicebetweenthosewhocanaffordalotoffreetimeandthosewhocannot.
Poor people receiving underemployment benefits might use that money to buy
themselves some free time, but this does not mean we could pass the insurance
argumentasadistinctiveargumentforfairsharesoffreetime.
47
Finally,ifallittookforanargumenttobeaparentaljusticeweretoshowthatthere
are reasons of basic justice to ineffect universally subsidize parents, any argument
supportingtransfersofwealthfrompeoplewhohappentobenon-parentstopeople
who happen to be parents would qualify, regardless of its aim and underlying
justification.Butthisisnotthecase.Toqualifyasanargumentaboutparentaljustice,
a pro-sharing argument must justify transfers by reference to some normatively
salient feature of parenting itself or its effects,which calls for redistribution. In the
nextchapterIdiscussanargumentthatdoesjustthat,namelytheargumentaccording
towhichparentshaveclaimstouniversalsupportinvirtueofthefactthatbearingall
thecostsofparenthoodwouldunderminetheirautonomy.
48
Chapter3
TheArgumentfromAutonomy
Withinmoralandpoliticaltheoryalotofattentionhasbeenpaidtohowthefamilyin
general, and particular styles of parenting more specifically, would foster the
developmentofchildren’sautonomy.Whatparentsmayormaynotdowhenraising
theirchildren is toavery largedegreedeterminedbywhatwould fosterorprevent
children’sgrowinguptobeautonomousadults,69or indeedbywhatwouldpromote
or compromise children´s autonomyqua children.70Considerably less attention has
beenpaidtowhetherhavingandraisingchildren(orraisingtheminparticularways)
might undermine the parents’ autonomy. A notable exception to this gap in the
literature is provided by Anne Alstott’s book-length treatment of the purportedly
autonomy-underminingburdensofchildrearingandtheobligationthatsocietyowes
toparentstoalleviatethem.71
Thischapterexplorestheprospectsforanautonomy-basedpro-sharingargumentfor
parental justice,takingAlstott’sworkasitsstartingpoint.Thegeneral ideasheputs
forthisthat,unlessitsocializedthecostsofchildren,aliberalegalitarianstatewould
be failing in its commitment to secure the conditionsof autonomy forparents, as it
must do for all its citizens. In this chapter I investigate themost plausibleways of
filling in thedetails of this general argument compatiblywith respecting the liberal
egalitarian commitments to holding people responsible for their choices and to a
principleofpoliticalneutrality.
69See,forinstance,BrighouseandSwift2014.70Clayton2006.71Alstott2004.
49
Iarguethat,inordertoestablishanautonomy-basedcomplaintonthepartofparents,
weneedwhatIwillcallaprincipleofautonomystakes.Thatis,weneedtoknowwhich
actual or potential burdens resulting from individuals’ autonomous choices should
count as autonomy-undermining, which are neutral from the point of view of
autonomy,andwhichareautonomy-promoting.Inwhatfollows,Ifirstintroducethe
intuitivecaseforparents’autonomy-basedcomplaintsproposedbyAlstott,aswellas
someimportantconceptualclarificationsregardingthenotionofautonomyatstakein
thisdebate. In section3.2 I describe the threekindsof costs that couldbe thought,
primafacie,toimpingeonparents’autonomy.Section3.3willdescribethechallenge
to any autonomy-based case, namely the challenge of identifying autonomy stakes.
Sections 3.4 through 3.7 explore how the main accounts of autonomy that the
literature offers could deal with this challenge. I conclude that some accounts of
autonomy are indeterminate with respect to the burdens that ought to count as
autonomy-infringing,whileothersofferunappealinganswers.Theaccountthatdoes
best is Joseph Raz’s self-authorship view,72yet the range of burdens that it can
condemn as autonomy-undermining (compatibly with political neutrality) are very
limited,andeventhereitisnotclearexactlywhatthestateiscommittedtobywayof
protectingpeople fromsuffering thoseautonomydeficits. InSection3.8 I scrutinize
one kind of view of what equal claims to autonomy might entail and offer some
reasons to thinkwe should not accept it. As a result, the autonomy-based case for
parental justice question remains open and depends, first, on defending a plausible
principle of autonomy stakes, and second, on developing a view of how the state
shouldbalancetheclaimstoautonomythatcitizenshave.
72Raz1986.
50
3.1.Alstott’sautonomy-basedchallengetoparentaljusticeAnneAlstott hasput forth thebest-developed autonomy-based case for sharing the
costsofchildren.Shehasarguedthatleavingparentstobearallthecostsofwhatshe
calls“continuityofcare”fortheirchildrenisathreattotheirautonomythatnoliberal
egalitarian state should tolerate. On pain of losing their parental rights altogether,
parents are required by the state to provide continuity of care for their children,
whichistosay,theyarerequiredtoensureminimalmaterialandemotionalstability
forthefirst18yearsoftheirchildren’slife.TheflipsideofthisiswhatAlstottcallsthe
NoExitrequirement.Parentswho“exit”theircare-givingrole,thatis,whointerrupt
carefortoolongaperiod,mayseetheirparentalrightsstrippedaway.Alstottargues
thatprovidingcontinuityofcareforchildreninvolvesseriousautonomy-undermining
costsforparentsastheyarerequiredtoputtheirambitionsandneedsonhold,orto
forgo them entirely, whenever they would conflict with their ability to provide
continuityofcare.Ifso,aliberalegalitariansocietythatwascommittedtoprotecting
citizens’autonomyshouldarguablysharesomeofthecostsofprovidingcontinuityof
careforchildrensoasnottocompromiseparents´personalautonomy.73Thisis,ina
nutshell, the challenge that Alstott brings to those liberal egalitarians who would
arguethatparentsshouldbearthecostsoftheirambitiontohavechildren.74
The notion of autonomy invoked by Alstott is that of personal, or individual,
autonomy. This is a highly contested concept inmoral andpolitical philosophy, the
unifying core of which seems to be the idea of self-government.75That is, to be
autonomous is to livea life that is in somerelevant senseone’sown.Thisnotion is
distinct fromthatofbasicautonomy,whichrefers to theminimalcapacitiesneeded
for someone to count as responsible for their own actions.76Basic autonomy is a
73Alstott2004,pp.58-61.74Aswehaveseeninthepreviouschapter,authorswhosubscribetothisviewincludeBurley2000,p.138;Casal1999;CasalandWilliams1995,pp.97-98;Clayton2006,p.169;Rakowski1991,p.153.75AccordingtoChristman1988,p.109.76Seee.g.Dworkin1988forthisdistinction.
51
moral status-conferring capacity that most adults share unless they suffer from
impairments or find themselves in coercive or restricting situations that would
absolvethemofmoralresponsibilityfortheiractions.
Personal autonomy has been treated, at times, as a local property that applies to
preferencesordesires,atothertimesasapropertyofwholelivesorpersons,andyet
at other times as a “programmatic”77 notion applying to particular domains of
someone’slife,forexampletheirprofessionallife.GivenmyinterestsinthischapterI
willtakeautonomytobea“global”propertyofaperson’slife,asGeraldDworkinputs
it.78However, whether someone lives an autonomous life will undoubtedly partly
depend on whether, and to what extent, their particular desires and choices are
autonomous.
Themainquestion thatmotivates this chapter canbeputas follows.Assuming that
someone autonomously decides to raise children (i.e. assuming theymeet the local
autonomy conditions for that decision), which of the costs that are, or could be,
associated with that decision should be considered a matter of respecting their
autonomyindecidingthecourseof their lives,andhence left forthemtobear?And
which of the actual or potential costs that follow from that decision should be
considered,bycontrast, autonomy-infringing if left for theparents tobear? I take it
that a naturalway of understanding this question is that it asks how the costs of a
particularautonomouschoiceshouldbesetinsuchawayastobecompatiblewiththe
agent’scapacityto leadanautonomouslifesubsequenttomakingthatchoice,hence
myfocusonautonomyasaglobalproperty.
Inthecaseofparents,weareinterestedinidentifyingthosecostsofchildrearingthat
areincompatiblewiththeparents’abilitytoleadautonomouslivesespeciallyforthe
durationoftheiroffspring’schildhood,whenthedemandsofparenthoodareusually
77Forthisdistinctionseeesp.Meyers1989.78Dworkin1988,pp.13-15.
52
attheirmostburdensome.Tothisend,inthenextsectionIteaseoutdifferentaspects
ofthecostsofchildrearingthathavethepotential,primafacie, tobesoburdensome
thattheywouldthreatenparents’autonomyunlessthestateintervened.Afterthis, I
embark on a systematic attempt to establishwhich, if any, of themain accounts of
personal autonomy in the literature have the resources to condemn these various
costs of childrearing (or, rather, their various burdensome aspects) as autonomy-
undermining.
3.2.TheburdensofchildrearingWecanidentifyatleastthreekindsofburdensthatthosewhoraisechildrentypically
bear,andwhichcouldbethoughttoinfringeonparents’autonomy.ThecategoriesI
proposehereshouldnotbereadasidentifyingentirelydistinctsetsofcostsbutrather
as organizing and re-framing those costs along different dimensions of
burdensomeness. Grouping the costs of childrearing in thisway anticipates the fact
thatdifferentaccountsofpersonalautonomy,asweshallseelater,condemndifferent
sorts of burdens as autonomy-infringing. It is therefore worth drawing out the
burdensomeaspectsof childrearingalong lines thatwill becomesalient later in the
chapter, when we review the conditions in which parents might fail to count as
autonomousondifferentviewsofwhatanautonomouslifeis.
First, there are what Alstott terms the costs of No Exit.79These are the costs of
providing what she calls continuity of care, a requirement that is state-enforced.
Parentsmustprovideminimalemotionalandfinancialstabilityforchildren,typically
forthefirst18yearsoftheirlife,onpainoflosingtheirparentalrights.Second,there
arewhatIwillcallthecostsofagoodupbringing.Thisreferstoallthecostsofgiving
childrenthekindofupbringingtheyareowedbyjustice,whereIassumethisincludes
79Alstott2004,pp.52-54.
53
continuity of care, and somuchmore beyond that.Depending on the details of our
favoredaccountofwhatchildrenareowed, thismight includeadequatehealthcare,
goodeducation, clothing, toys, intensiveemotional guidanceand soon.Third, there
arewhatIwillcallthecostsofbeingoncall.Thesearethecoststhatarebornebythe
full-timecarersofchildrenwhentheyareparticularlyneedyandentirelydependent
onsomeone’scare,typicallyininfancyandearlychildhood.Iwillnowsayabitmore
abouteachtypeofcost.
3.2.1.NoExitAccording to Alstott, the autonomy-undermining burdens of parenting stem from
havingtoprovidecontinuityofcare.Parentsareexpectedtostaybytheirchildren’s
side and to provide aminimally financially and emotionally stable environment for
the first 18 years of their life. In effect, society imposes a No Exit rule on parents.
Those who “exit” their care-giving role, that is, who interrupt care for too long a
period,mayseetheirparentalrightsstrippedaway.Alstottarguesthatthis involves
seriousautonomycostsforparentsastheyarerequiredtosacrificeanyambitionor
opportunity thatmight jeopardize theirability toprovidecontinuityof care to their
children.
Herargument,then,isfocusedonthecostsofNoExitandtherolethestateplaysin
enforcingthem.AsPaulBou-HabibandSerenaOlsarettilaterpointout,thecostsofNo
Exitcouldbeunderstoodintwoways.First, therearewhatBou-HabibandOlsaretti
havereferredtoasthecostoffailingtheNoExitobligation.80Itmightbethoughtthat
individuals’autonomyisinfringedbytheveryfactthatsocietyimposesastrictNoExit
obligation on parents. Liberal societies are normally committed to leave people (at
leastthenegative)freedomtofashionandpursue,aswellaschange,theirlifeplansas
theyseefit.However,wedenyparentsthefreedomtoparentinanywaytheyseefit.80Bou-HabibandOlsaretti2013,p.427.
54
Inparticular,only thosewhoparent inawaythatsecurescontinuityofcare for the
childarelegallypermittedtoparent.Parentsarefreetoexittheparentalroleentirely
(byleavingtheirchildinthecustodyoftheotherparentorbygivingtheirchildrenup
for adoption), but they are not legally free to parent in a way that involves
discontinuityofcare.
The second type of costs is what Bou-Habib and Olsaretti have called the costs of
complyingwiththeNoExitobligation.81Broadlyspeaking,thesearethecostsofhaving
to forgo valuable goods and opportunities that would conflict with providing
continuityofcare.EvenifaparentfullyembracestheNoExitrequirementaspartof
their own conception of how to parentwell, providing such continuity of carewill
inevitablybeincompatiblewithsomeothervaluablepursuitsthatparentsmayvalue,
andwhichtheywillbeexpectedtosacrifice.Forinstance,onecannotaffordtotakeup
alucrativeandfulfillingjobonadifferentcontinentthatwouldtakethemawayfrom
theirchildren foranunacceptably longperiodof time.Whenever thepursuitofany
valuable projects or relationships would hamper parents’ ability to ensure a
financiallyandemotionallystableenvironmentfortheiroffspring,societyexpectsthat
parentswillprioritize their children’sneed for continuity,onpainof losingcustody
rightsshouldtheyfailtodoso.
Alstott seems concerned with both of these kinds of No Exit costs. At times, she
stresses that society “so heavily regulates just one social role”82through the very
unique, in her view, imposition of a No Exit rule as such. She suggests that liberal
societies typically balk at imposingNo Exit requirements on any pursuit, since this
amounts to “forcing people to persist in a particular way of life,” just like banning
abortioncouldbeunderstoodtodo.83
81Bou-HabibandOlsaretti2013,p.428.82Alstott2004,p.63.83Alstott2004,p.55.
55
However, as Bou-Habib andOlsaretti argue, parents could have a complaint only if
thisrestrictionwasmorallycontingent.84AsAlstottherselfacknowledges,thisisnot
infactthecase.Asshepointsout,justlikethestateismorallyjustifiedinforcingthose
whoowndogstokeepthemona leashoutofconcernforothers’safety, thestate is
alsojustifiedinimposingthecostofthisparticularstyleofparentingoutofconcern
forprotectingchildren’sfundamentalinterestincontinuityofcare.85
So it isdoubtful thatunderstandingthecostsofNoExit thiswaywouldprovetobe
worrisomefromthepointofviewofparents’autonomy.Thesecondunderstandingof
NoExitcostsseemsmorepromising.Alstottremindsusthatparentsareexpectedto
forgo any opportunities, no matter how valuable, whenever pursuing them would
conflict with parents’ ability to provide continuity of care. Travelling the world,
starting a new relationship,moving to a different country, engaging in fulfilling but
verytime-consumingwork–anyofthesecouldpotentiallyaffecttheparent’sability
toprovideafinanciallyandemotionallystableenvironmentforthechild,and,assuch,
shouldbeforgone.Thisstateofaffairswouldgoonfornolessthan18years,untilthe
childreachesadulthood.
Alstottdoesnotclaimthattheparents’freedomtopursuesuchopportunitiesshould
take precedence at the expense of children’s need for continuity of care.What she
does claim is that it is implausible that a liberal state committed to the value of
autonomyshouldletparentsalonebearthesecosts.
IwillreturntoAlstott’sargumentlaterinthechapter,butfornowitisimportantto
notethathermainconcernwiththecostsofcomplyingwithNoExitseemstobethat
it stuntsparents’ capacity for setting and revising theirpriorities among competing
valuesandprojectsovertimeinawayshethinkstheyshouldbeabletodo.Fromnow
onwhenIspeakofthecostsofNoExit(orthecostsofprovidingcontinuityofcare)I
84Bou-HabibandOlsaretti2013,p.427.85Alstott2004,p.57.
56
willbereferringtothislattersensedistinguishedbyBou-HabibandOlsaretti,namely
thecostsofcomplyingwithNoExit.
Importantly, Iwill understand the costsofNoExit as involvingonly those lossesof
goodsandopportunitiesthatarenecessaryforparentstoremainbytheirchildren,as
theirmaincarers,forthefirst18yearsoftheirlives.TheburdensofNoExitcanvary
fromone context to another, but they include complyingwith a prohibition against
beingphysicallyseparatedfromthechildfortoolongatime(forwhateverreason,no
matterhow important), andwith the requirement toprovideminimal financial and
emotionalstability. Iunderstand thecostsofNoExit, then, tocoincideroughlywith
those required in order to complywith the standard of non-abuse and non-neglect
enforcedbythelaw.
3.2.2.GivingchildrenagoodupbringingThis class of costs refers to the sacrifices that parentsmustmake in order to give
children a good upbringing, which I assume involves more than merely providing
continuityofcare.Continuityofcare,as Iunderstand it, isonlypartofwhatagood
upbringing involves. Arguably, children are also owed a good education, adequate
clothing and nourishment, adequate health care and so on. Furthermore, they also
requireagooddealofcloseparentalguidanceandsupport,andanyothereffortsthat
arenecessaryforraisingchildrentobe,ontheonehand,autonomouscitizenswitha
senseofjusticeand,ontheotherhand,abletobeproductive,competitivecitizensin
theirrespectivesocieties.Justwhattheyrequire,andhowexpensiveitistoprovideit
tothem,dependsonthestandardofagoodupbringingwebelieveistherightone.But
notethatthecoststoparentsofprovidingallofthisontheirown,whicheverstandard
is correct, are considerably higher than the costs to them of merely insuring
continuityofcare.Thoughofcoursejusthowmuchhigherdependsonwhichcostsare
already shared by a particular society, for instance in manyWestern societies the
57
costsofeducation(atleastprimaryandsecondaryeducation)andsomeminimallevel
of health care are publicly funded. This is arguably done for reasons other than to
protect the autonomy of parents qua parents. Rather, the reasons are to do with
ensuringequalityofopportunityforthechildren,orinordertoincentivizepopulation
growth,orasawayofpromotinggenderjustice.
3.2.3.BeingoncallTaking inspiration from the literature on the ethics of care, it is worth teasing out
what seems to be a particularly burdensome subset of the costs of childrearing, a
subset of arguably both the costs of No Exit and the costs of a good upbringing.
Following Diemut Bubeck, wemight term these the costs of being “on call.”86They
refer to the costs of having tomeet the day to day needs of an entirely dependent
child, so they are most obviously tied to caring for infants and young children.87
Babies and young children depend for their survival on others’ continuous care, so
someonemustbeavailableatalltimestorespondtotheirneeds.Theyhaveaneedfor
a fixedeatingandsleepingschedule,aswellasa rangeofotherneeds thatmustbe
metregardlessofwhetherthecarerfeelslikeitornot,andregardlessofwhetherthey
conflictwiththecarer’sownneedsatthatparticulartime.Forinstance,atoddleralso
needs an adequate amount of fresh air and physical exercisewhich the carermust
provideevenifshe,thecarer,feelsexhaustedonthatday.Beyondthecostofhavingto
organize one’s life around the predictable and fixed needs of the child, there is the
addedcostofhavingtobereceptiveandresponsivetoanyfurtherneedthechildhas,
asDiemutBubeckputs it. Ifachildcries, thecarermusttrytounderstandwhatthe
needisandtorespondpromptlyandappropriatelytoit.Bubeckwrites:“womenoften
describetheir livesas full-timemothersofsmallbabiesas ‘dictated’bytheneedsof
86Bubeck1995,pp.143-144.87Though,itbearsmentioning,onecanalsoexperiencesuchcostswhencaringforchildrenwithseveredisabilitiespotentiallyfortheirentirelife,aswellaswhencaringforone’sagedparents.
58
theirbabies,bytheirsleepingandwakingperiods.Being‘oncall’asadoctorornurse
in a hospital is an example of such availability in a more formal setting: it
institutionalizes receptivity and responsiveness and thusmakesdoctors andnurses
availabletopatients.”88
Wecanlooktoananecdotalbutveryvividdescriptionofjusthowburdensomehaving
tomeet theseneeds canbe from the letterof amotherof twoyoung childrenwho,
despite loving her kids, feels overwhelmed. The woman signs her letter ‘The Bad
Mom’.
My body is always being stepped on, squished, pulled, yanked, carrying something,
beingpummelled. I can’t sitdown foronesecondwithoutbothkids fightingoverme,
climbing onme, inadvertently punchingme in the eye, the breast, the gut. There are
momentswhereIhavetopullthemoffmelikeleechesandruntotheotherroomfora
hair’s breadth of freedom. […] Someone is always going through a phase, someone is
always having a developmental leap, someone is always tired, someone is always
hungry,someoneisalwaysnotgettingenoughbooksreadtothem,someoneisalways
notgettingenoughcraftprojects.89
3.3.Autonomycosts,responsibilitycosts,andtheissueofstakesAlstott’swork,feminists’work,90andanecdotalaccountslikeTheBadMom’s,makeit
abundantly clear just how burdensome raising children can be. Many of these are
accountsmeanttocapturethestateofaffairsintherealworld.Nodoubtthatvarious
forms of injustice like economic inequality, enduring patriarchal structures and
structural racial injustice make it the case that raising children is a particularly
burdensomeundertaking(intermsofautonomyandnotonly)forthoseatthebottom88Bubeck1995,p.143.89FromaletterbyalistenerreadonDearSugarpodcast,episode36:Motherswhohatemotherhood.90Okin1989;Bubeck1995;GornickandMeyers2004;Esping-Andersen2009.
59
oftheeconomicladderinoursociety,poorwomen,andmostofall,thosewomenwho
arepoorandwhoaremembersofasalientminoritysuchasblackwomenintheUS.91
Butthequestionthatconcernsushereiswhethertheburdensofchildrearingcangive
parents an autonomy complaint as a matter of ideal theory. For an ideal theory
argumenttosuccessfullygroundanautonomy-basedcomplaintitmustestablishtwo
things:(a)thatunlessconditionsCaremet,anindividualsuffersanautonomydeficit,
eitherrelativeorabsolute;(b)thattheautonomydeficitwouldbeunjusttothosewho
suffer it. The autonomydeficitmight not be unjust, for example, because itmay be
unfairtootherstoenjointhemtogiveupsomeoftheirresourcestomakeupforthis
autonomydeficit(itmightbeaninfringementoftheirautonomy).
Tomake a case for socializing the costs of children, then, onewould need to settle
bothoftheseaspects.Iwillarguethat,asthingsstand,wethereisonlylimitedsuccess
tobehadwithsettlingeventhefirstissue,namelythequestionastowhichburdensof
parenthoodshouldcountasanautonomydeficit.Foranautonomouslyledlifesurely
involves forgoing certain opportunities and bearing some burdens, even large
burdens,intheprocessofcarvingoutalifethatistrulyone’sown.And,asIwillshow
below,themaintheoriesofautonomyonofferintheliteratureeitherdonothavethe
resources (plausibly) to tell us which burdens are incompatible with leading an
autonomouslife,or,whentheydo,theyrelyonsubstantiveviewsaboutwhatleading
anautonomouslifelookslike,viewsthatareincompatiblewithaprincipleofpolitical
neutrality.
Before Ido that, letmesetup theproblemabitmore. I said that the first step ina
successfulautonomy-basedpro-sharingargumentisto(a)establishtheconditionsC
that must obtain such that an agent, despite bearing burdens, does not endure an
autonomydeficit.ConditionsCareoftwotypes,andinclude:(i)theconditionsunder
which thechoices,desires,preferences, commitmentsorvalues thatguideone’s life
91Shelby2018.
60
can properly be attributed to the agent’s authentic self. This has been the focus of
much of the literature on individual autonomy: establishing the conditions under
whichwecansaythatsomeone’svalues,desiresandpreferencesetc.aretrulytheir
own. But crucially, we also need to know (ii) which of the costs and benefits that
might spring from thoseautonomouscommitments,preferences,desiresetc. should
ormaybe left for theagent tobearorenjoyrespectively,asamatterof respect for
theirabilitytodecidetheshapeoftheirlives,andwhichofthemshouldnot.Giventhe
parentaljusticeinterestinansweringthequestionofwhoshouldpayforthecostsof
children, I will focus my attention on the burdens and leave the question about
benefitstooneside.
The issue raised by (ii), namely which of the burdens that might follow from
autonomouscommitmentsshouldormaybeleftfortheagenttobearasamatterof
respect for their autonomy (rather than be understood as a threat to it) bears
structural similarity with a familiar question from the literature on responsibility-
sensitivetheoriesofjustice.Thisiswhathasbeencalledthequestionofresponsibility
stakes.92 It is the question of which disadvantages are a matter of individuals’
substantiveresponsibility for theirchoicesandwhichdisadvantagesareamatterof
injustice that calls for some formof redress. It isworth takingabit of timehere to
explaintheissueofresponsibilitystakessoastobetterunderstandwhatIargueisa
parallelissueplaguingtheautonomydebate.
3.3.1.ThequestionofresponsibilitystakesRecall that responsibility-sensitive theories of justice seek to redress the
disadvantagesthatcomeaboutthroughnochoiceorfaultoftheagent,thatis,through
92Olsaretti2009.
61
bad brute luck, whilemaintaining that the disadvantages that are the result of the
agents’choicesshouldbeleftforthemtobear.
Severalauthors,however,havepointedout that it isnotclear,ontheseviews,what
holdingpeopleresponsiblefortheiractionsactuallyamountsto.93Toseethis,thefirst
step we need to make is to distinguish between what Thomas Scanlon has called
moral responsibility from the notion of substantive responsibility. 94 Moral
responsibilityrefers to theappropriatenessofattributinganactiontosomeoneasa
basis formoral appraisal. Substantive responsibility refers toaperson’s liability for
being treated in certain ways as a result of their action, including denying them
certain rights, exacting compensation from them or letting them bear the material
disadvantages thatmight follow,aswellas,say, letting themkeepthe fruitsof their
effort.
AsOlsarettipointsout,substantiveresponsibility isattributedtopeople invirtueof
somefeatureoftheirsthatisconsideredtoprovidetheappropriategroundsforthat.
Theoristsdisagreeoverwhichthosefeaturesshouldbe.Forinstance,DworkinandG.
A. Cohen famously disagreed over whether someone’s finding themselves with an
expensivetasteforcaviarshouldbeconsideredappropriategroundsforholdingthem
liable for the consequences of that taste.95As Olsaretti emphasizes, these theorists
seem to assume that once we settle the appropriate features that should ground
people’s liability, it is self-evidentwhich are the costs that they are to justifiablybe
heldliablefor.96
However, settling the appropriate grounds for substantive responsibility is not
enough to know which burdens or benefits a particular action should attract. For
illustration, Olsaretti asks us to considerMarc Fleurbaey’s case of themotorcyclist
93ColemanandRipstein1995;Olsaretti2009;Stemplowska2009.94Scanlon1998,ch.6.95Dworkin1981,Cohen1989.96Olsaretti2009,p.169.
62
named Bert, who knowingly and deliberately takes on the risks of driving his
motorcycleathighspeedwithoutahelmetandendsuphavinganaccident.97Byany
account,responsibility-sensitiveviewswouldhaveBertbeartheconsequencesofhis
imprudentchoice.ButwhatexactlyshouldBertbeliablefor?Isheliableforbeingleft
atthesideoftheroad,orforhavinghismotorcycleconfiscated?Isheliableforpaying
thebill of his emergency treatment, and atwhatprice? Shouldhebe chargedmore
comparedtosomeonewhosufferedanaccidentthroughnofaultoftheirown?Ishe
liable topay for anyhealth conditions thathemay suffer later on as a result of his
injuries?Shouldheneedanorgantransplant,isheliableforbeingsenttothebackof
thequeue,giventhathisneedwascausedbyhisownimprudentchoice?Andthelist
couldgoon.ThisindeterminacyregardingwhichcostssomeonelikeBertshouldbear
himself calls for what Olsaretti has termed a principle of stakes,98or what Zofia
Stemplowska has called the structure of pay-offs. 99 Without a principle of
responsibilitystakes,oranestablishedstructureofpay-offs,wecannotsaywhichof
theconsequencesthatBert’schoicebroughtaboutheshouldbejustlyheldto,which
areoneshemaybeheldto,andwhichhecanjustlycomplainabouthavingtobear.We
cannot say, then,what holding someone substantively responsible for their choices
actuallyinvolves.
3.3.2.ThequestionofautonomystakesUsingthesamekindofreasoning,IproposethatweneedwhatIwillcallaprincipleof
autonomy stakes for determining what exactly respect for individual autonomy
requires. In the responsibility debate, it is not enough to settle the conditions of
attributionofresponsibilityinordertoknowwhichdisadvantagesshouldbeborneby
the agent alone as a matter of justice. Analogously, it is not enough to settle the
97Fleurbaey1995citedinOlsaretti2009,p.172.98Olsaretti2009,p.167.99Stemplowska2009,p.246.
63
conditions under which someone’s values, commitments, desires etc. should be
consideredone’sowninordertoknowwhichoftheburdensthatmightfollowought
tobebornebytheagentasamatterofrespectfortheirautonomy.
AsBenColburnputsit,activelyleadinganautonomousliferequiresthat“thereason
that my life goes that way must be that I made it so, and also I must bear the
consequencesofthewayIchoosetoliveit.”100So,forgingandpursuingapersonallife
plannecessarily involves takingon somecosts, at least someopportunity costs, and
thesecostsseemtobeautonomy-enhancing,ratherthanundermining. If Ichooseto
become an emergency surgeon this will mean I cannot, at the same time, enjoy a
globetrottinglifestyle.Itseemsrighttoconsiderthislossofopportunityfortraveland
perhaps for relocation as an expression of my autonomy, or an integral part of it,
ratherthanautonomy-hindering.
Ontheotherhand,someburdensthatmightaccompanyone’sautonomouschoicesdo
seem to be autonomy-violating. The enormous time commitment and the
unpredictablescheduleoftheemergencysurgeonmightcausehertohavedifficulties
in forming andmaintaining personal relationships and theymight even undermine
her ability to have aminimally satisfying family life. Intuitively, this sort of burden
underminestheagent’sautonomyinsteadoffosteringit. Finally,someburdensmay
bealtogetherneutraltotheagent’sabilityto leada lifeoftheirown.Surgeonsoften
have towearblueorgreen-coloreduniforms towork, forexample, insteadofwhite
ones,oranyothercolor.Thisrathertriviallossofcontroloverwhatcoloruniformto
wearseemstomakenodifferencetohercapacitytoliveautonomously.
So,thequestionishowweoughttodecidewhichinstancesoflossofcontrolandloss
ofopportunitiesareamatterofdefiningthescopeofautonomyonceanautonomous
choicehasbeenmade.Recall themotorcyclistBert,whoactsrecklesslyandendsup
havinganaccident.Justlikeweneedaprincipleofresponsibilitystakestobeableto
100Colburn2010,p.32.
64
tell,outofallthenegativeconsequencesthatBertexposeshimselfto,whichoneshe
shouldbeheldsubstantivelyresponsiblefor,whichonesareindifferenttosubstantive
responsibility,andwhichonesheshouldnotbeheldresponsiblefor,so,too,weneed
aprincipleofautonomystakestodiscernwhichoftherestrictionsthattheemergency
surgeonfacesareforhertobearasamatterofrespectforherautonomy,whichones
are neutral from the point of view of autonomy, and which ones constitute an
autonomydeficit.Thesamegoesforparents:thereareclearlymanylossesofcontrol
andofopportunitiescurrentlyattachedtoparenting,butweneedtoknowwhichones
areautonomy-promoting,whichonesareneutral,andwhichonesactuallyconstitute
autonomydeficits.
It isworthemphasizingat thispoint thatnoneof this is todeny that, ifwehavean
autonomy-based view of justice, we could not also allow or require autonomous
people to bear costs for reasons other than those given by concern for their own
autonomy,i.e.reasonsotherthanthosegivenbyaprincipleofautonomystakes.Other
factors could, of course, inform the costs that society could or should impose on
agents,suchasconsiderationsaboutwhatburdensitisjusttootherstoaskagentsto
bear.Thelatter,Iassume,arepartoftheconsiderationsthatwouldbepickedoutbya
properlycircumscribedprincipleofresponsibilitystakes.
OverthefollowingthreesectionsIinvestigatewhetherthemainaccountsofpersonal
autonomy in the literature have the resources to provide a plausible principle of
autonomystakes,andtodosoinawaythatiscompatiblewithpoliticalneutrality.Of
course,theseviewsnevertookthemselvestobeprovidingthisexplicitly,butwecan
attemptto“extract”someprinciplesthatseemtobeimpliedintheseviews,orthatat
least are compatible with them. As I havementioned, a full principle of autonomy
stakeswould tell uswhich burdens count as autonomy-promoting, which ones are
neutral from the point of view of autonomy, and which ones count as autonomy-
undermining.However,tomakesomeheadwayintheparentaljusticedebateitwould
be sufficient to be able to identify at least those kinds of costs that are autonomy-
65
underminingontheseviews.Forwhatweareultimatelyinterestedinisestablishing
whether parents have grounds for complaint for having to bear autonomy-
underminingburdens.
The accounts of autonomy I will consider are representative of the major strands
present inthe literature, thoughmylistwillnotbeexhaustive.Mostnotably, I leave
aside the Kantian view of autonomy, which is strictly concerned with agents’
rationality and their capacity to create themoral law for themselves, and to act in
accordancewith it.101The accounts Iwill address canbe classified into threebroad
categories.These are: themainstreamprocedural, content-neutral views, inwhich I
include the hierarchical accounts of autonomy; substantive accounts, which place
substantive conditions either on the content of agents’ choices for them to count as
autonomous in the firstplaceoron theparticular life situation that individuals find
themselves in as a result of their autonomous choices; and finally, relational views,
which can be either content-neutral or substantive, but whose distinctiveness
concernstheimportancetheyplaceontheroleofsocialandinstitutionalconditions
forindividualautonomy.
3.4.HierarchicalaccountsofautonomyAninfluentialstrandoftheoriesproposesthatthenecessaryandsufficientconditions
for autonomy have to do with the quality of agents’ motivational structure. These
mainstream views have sprung from the seminal work of theorists like Harry
Frankfurt, Gerald Dworkin, and John Christman. All three focus on the fact that
people’smotivationsseemtositinasortofhierarchicalstructurethatonecanreflect
onandmakechangesto.
101ForaninstructiveandconcisediscussionseeColburn2010,pp.5-8.
66
Consider Harry Frankfurt’s account. Frankfurt argues that first-order desires are
autonomouswhentheyareendorsedbysecond-orderdesires.Twopeoplemightbe
acting on the same first-order desires, for instance they might be acting upon the
desire to drink alcohol, but one of them could be considered autonomous and the
other not. One person might regret having a desire for alcohol. They would not
identifywith,orendorse,theirfirst-orderdesire,andsotheirdrinkingwouldnotbe
an act of their ownwill in the relevant sense. By contrast, the other personmight
endorsetheir first-leveldesire fordrinking, inwhichcasetheywouldbeconsidered
autonomous.102Inasimilarvein,GeraldDworkinconceivesofautonomyasasecond-
ordercapacity“toreflectcriticallyupontheirfirst-orderpreferences,desires,wishes,
andsoforthandthecapacitytoacceptorattempttochangetheseinlightofhigher-
orderpreferencesandvalues.”103
The historical twist added by John Christman preserves the importance of the
hierarchical structure of one’s motivations, but makes one important change. He
arguesthatitisthehistoryofhowonecametoacquireavalue,desire,orpreference
that iscrucial forautonomy.Onthisview,anagent isautonomouswithrespecttoa
desireonlyif,havingalltherelevantinformation,onedidnotresistitsdevelopment
orwouldnothaveresistedhadtheyattendedtoitsformationprocess.104
Idonottakeastanceonwhichaccountofautonomyweshouldsubscribeto,andsoI
will not invest time in rehearsing themerits and criticisms to these views.What is
importanttonoteforourpurposesisthattheseviewsseemillequippedtoanswerthe
question of autonomy stakes, as they focus entirely on value-neutral internal
processes for attributing someone’s values, commitments, desires, preferences to
theirauthenticself.Solongasone’svalues,desires,preferenceshavebeenarrivedto,
and/oraremaintained,throughtherightinternalprocesses,theircontent,whateverit
is,canbeconsideredtobelongtoone’sauthenticself.102Frankfurt1971.103Dworkin1988,p.18.104Christman1988.
67
Wecanseethelimitsofthisapproachformypurposeshere,whichconsistinfinding
thebases foranaccountof autonomystakes,by lookingdirectlyathow theymight
helpusmake judgmentsaboutparents’autonomy. In thesortof idealscenario Iam
interestedinItakeitthatwecanassumethattheconditionsthatarenecessaryand
sufficient forautonomyonhierarchicalviewsaremet. I assume that something like
Christman’shistorical criterionobtains:parents (includingwomen)didnotdevelop
theirpreferenceforraisingchildrenthroughsomeformofmanipulationorcoercion
byothers,andsoeitherdidsanctiontheformationofthispreferenceorwouldhave
sanctionedithadtheyattendedtoitsformationprocess.Furthermore,Idonothavein
mindparentswhoregrettheirdesiretoparent.Iassume,asamatterofidealtheory,
thatparentshave thecapacity toreflectonandendorse theirpreference forraising
children(thusmeetingDworkin’sconditions),andthattheydoinfactendorseit(as
perFrankfurt’srequirements).
Withall of this inplace, thequestionof autonomystakes remainsuntouched.What
should follow from the fact that we can ascribe parental commitments and
preferences to individuals’ authentic self? Should it follow that, however costly and
restrictivefortheirself-regardingpursuitsthispreferenceturnsouttobe,theyshould
notberelievedof theseburdensoutof respect for theirautonomouspreference for
parenting?Sincethesehierarchicalviewsarecontent-neutral,theylacktheresources
to non-arbitrarily place a limit on just how restrictive or burdensome someone’s
autonomous choice might turn out to be for them. Hierarchical views themselves,
then, are silent on thequestionof autonomy stakes.This is unsurprising since they
focus only on the first part of the two-part question of autonomy that I identified
above. Asmentioned at the beginning of Section 3.3 above, in order to knowwhat
leading a life that is truly one’s own looks like we need to know (i) whether the
choices, commitments, valuesetc. thatguideone’s life canproperlybeattributed to
the agent’s authentic self, and (ii)whichof the costs andbenefits thatmight spring
fromautonomouschoices, commitments,valuesetc. should,may,ormaynotbe left
for theagent tobearorenjoyrespectively,asamatterofrespect for theirability to
68
decide the shape of their lives. Hierarchical views are geared exclusively towards
settlingthatfirstissue,andstaysilentastowhetherthesortsofburdensthatparents
faceshouldconcernus.
Tobeclear, justicemightrequire,forexample,thateveryone(notjustparents)bear
the costs of children’s education, for reasons other than concern for parents’
autonomy, such as a collective commitment for insuring equality of opportunity for
children. But it is still important to establishwhere there could nevertheless be an
autonomy-based complaint on the part of parents. This can be relevant, first, with
respecttothesameclassofcostsforwhichtherearealsootherjustice-basedreasons
to socialize. Parental autonomy concerns could bolster the case for socializing the
costs of children and it could impact the form that sharing those costsmight take.
Policies justified only by ensuring equality of opportunity for children might look
different from those directed (also) at preserving or boosting parents’ autonomy.
Second,itisimportanttoknowwhetherautonomyconcernscanprovideareasonto
sharesomekindsofcoststhesocializingofwhichisnotsupportedbyotherreasons.
Drawingon the literatureon the stakesof responsibility,105we could try to identify
someprinciplesofstakesthatwouldbecompatiblewithcontent-neutralhierarchical
viewsofautonomy.Onesuchprinciplecouldbethefollowing.
Thecausalistapproachtostakes:Itiscompatiblewithrespectforanagent’sautonomy
thatanyandallcausalupshotsofone’sautonomouschoices106beconsideredtheagent’s
tobear.
Thisprincipleiscompatiblewiththehierarchicalaccountsofautonomybecause,asI
mentioned, they only establish conditions for the attribution of a person’s
commitments,choices,andtheresttotheirauthenticself,butstaysilentonwhether105Olsaretti2009,pp.173-182.106Foreaseofexposition,informulatingtheseprinciplesIwillspeakoftheconsequencesofone’sautonomouslymadechoicesbutIwillmeanforthistoincludetheconsequencesofautonomouschoices,desires,preferences,commitmentsorvalues.
69
there are any external conditions that must be met for an autonomous life. So, in
principle,anyconsequencecausedbyanautonomouscommitmentmaybeleftforthe
agenttobearasamatterofautonomystakes.
Thiswayofdefiningautonomystakeslookshighlyimplausible,anditdoesnotseem
likeitwoulddomuchtohelpparents’case.Onerespectinwhichitisunattractiveis
that it seeks to tie autonomy stakes to the causal upshots of one’s autonomous
commitments.Yetcausation isanotoriouslyhardthingtopindown.For instance, if
weunderstandthenotionofcausationasbut-forcausation,thelistofthingsparents
couldbethoughttohavecausedthroughtheirautonomousdecisiontohavechildren
is seemingly interminable. Any misfortune that might afflict the offspring both in
childhoodaswellas inadulthoodcouldbethought tohavebeencausedbyparents.
But forparents’ having created this child, shewouldnothavebeenmugged, or she
wouldnothavecontractedthisdisease,orshewouldnothavesufferedaheartbreak
andsoon,becauseshewouldnothaveexisted.Thestatecouldforceparentstobear
thecostsfornone,forall,orforaseeminglyarbitrarysubsetofthesemisfortunesthat
theypurportedlycausedbybringingchildrenintoexistence.Soonerespectinwhich
thissolutionwouldbeimplausible,then,isthatitwouldallowthestatetoarbitrarily
pickanyconsequencesinthelong(andperhapsopen-ended)chainofconsequences
thatanyactioncouldbetakentohaveandforcetheagenttobearthecostsforthemas
amatterofautonomystakes.
It should be clear that limiting the principle to those causal upshots that are
reasonably foreseeableor intendedby theagents (here, theparents)wouldprovide
onlyaminorimprovement,butwouldnotrendertheprincipleplausibleoverall.The
improvement is that it could help condemn leaving parents to bear all the costs
associatedwithunintendedcausalupshotsofprocreationsuchas,forinstance,having
a childwho suffers froma rare illnesses that couldnothavebeen foreseen.Butwe
would still be leftwith a greatmany cases inwhich the state could leavepeople to
bearconsequencesthatwewouldthink,intuitively,wouldunderminetheirautonomy
70
and should be condemned by a principle of stakes. Furthermore, therewould be a
questionofthe justifiabilityofdeemingonlytheintendedcostsofone’sautonomous
commitments to be the agent’s to bear. And indeed, the justifiability of such an
accountofautonomystakeswouldbequestionable.Non-parentscouldstillcomplain,
for example, of having to pay for the unintended consequences of other people’s
wanting to have children, since it is still the case that even the unintended
consequencesseemtobelongmoretotheparents,ifanyone,thantothenonparents.
Perhapsamoreplausiblecontenderforaprincipleofautonomystakesthatwouldbe
compatiblewithmainstreamhierarchicalaccountsofautonomywouldbesomething
likewhatOlsarettihascalledthecontextualistapproach tostakes.107Itwouldgolike
this.
Thecontextualistapproach:Itiscompatiblewithrespectforanagent’sautonomythat
the agent bear those consequences of their autonomously made choices which are
publicly known in advance to accompany those particular choices in a particular
community.
Inasociety inwhich it ispubliclyknownthat,say,havingchildren involvesbearing
the costs of being on call, the costs of No Exit, and the costs of a good upbringing
oneself,thenthosewhochoosetoparentwillinglyandknowinglyinsuchconditions
cannotcomplainthattheirautonomyishamperedbyhavingtobearthesecosts.
This certainly seems like a better contender for a principle of autonomy stakes
compared to the first one. For example, this principle is not vulnerable to the
following sort of complaint that causalist approach would face. If I hold the right
motivational structurewith respect to a particular ambition but I have no possible
wayoftellingwhatconsequencespursuingthatlifeplancouldbring(becauseIhave
nowayof foreseeingall the causalupshotsofmyambition forwhich Iwill thenbe
107Olsaretti2009,p.176.
71
liable),thenIcouldplausiblycomplainthatIhadnowayofknowingwhatshapemy
life would take when I decided to pursue that ambition, or any ambition for that
matter. This could cast doubt on whether my decision could be considered
autonomousinthefirstplace.Acontextualistprincipleofautonomystakeswouldbe
moreautonomy-preservinginthisrespect.Theconsequencesforwhichtheagentwill
beheld liable inaparticularsocietywillbeclear fromtheoutset.Parentscouldnot
complain that they did not knowwhat burdens the choice to raise childrenwould
bring. But as Olsaretti points out in the context of responsibility stakes, that a
particularstructureofpay-offsisalreadyinplace(e.g.parentsbearingallthecostsof
children) rather a different one (e.g. costs of childrenbeing shared) doesnotmean
thatthefirstoneismore justifiedthanthe latter. Indeed, this ispreciselywhat is in
question.
3.5.SubstantiveviewsofautonomyIt looks likeautonomyviews that focusexclusivelyon themotivational structureof
theagentdonotseemofmuchhelpinfixingautonomystakes.Sowemightwantto
lookataccountsthatposesomeexternalconstraintsonautonomy.Ireviewtwosuch
views, one that places constraints onwhatmay count as an autonomous choice or
preference in the first place, namely Jean Hampton’s self-respect view. The second
accountIreviewinthissectionisRaz’saccountofautonomyasself-authorship,which
allows us to identify at least some minimal external conditions that should
characterizeindividuals’lifesituationsiftheyaretocountasminimallyautonomous.
72
3.5.1.Theconstraintsofself-respectSubstantiveaccountsplaceconstraintsonthecontentofindividuals’desires,choices
and commitments if these are to count as autonomous. One example of such an
account is that developed by Jean Hampton. She argues, very roughly, that choices
which hamper one’s ability to satisfy their basic needs as human beings aswell as
theirneedstoflourishasuniqueindividualscannotbeconsideredtrulyautonomous.
Suchchoices,are, rather, self-denyingandviolateadutyof self-respect.108Shegives
the example of Terry, amotherwho dedicates herself exclusively to caring for her
familyand,assuch,lacksanyself-regardinggoalsorprojectsandevenjeopardizesher
health.TerryisinthesamesituationasTheBadMom,butshedoesnotseemtomind.
HamptonarguesthatratherthanpraisingTerry’sbehaviourasselflessandvirtuous
weshouldcondemnitasself-exploitativeanddisrespectfultoherself.OnHampton’s
view, self-exploitative and self-harming commitments like Terry’s cannot be truly
autonomous. Thomas Hill is another autonomy theorist that criticizes this sort of
choice. Hill argues that the choice of a “deferential wife” to be subservient to her
husbandbetraysalackofself-respect.109
Hampton’sviewhasbeenthoughtimplausibleinsofarasitamountstosomethinglike
“compulsorymoral egoism,” as Richard Arneson put it.110 He has pointed out that
Hampton’sviewamounts to saying thatwheneverourbasicneedsorourneeds for
flourishing clash with the needs of others, morality requires that we prioritize
ourselves.ThisstrikesArnesonasunmotivated(ifweareallmoralequalswhyshould
webemorallyrequiredtoprioritizeourselves?111),aswellasoverlystringent.
In fact, substantive accounts of autonomy more generally have been criticized for
being too stringent as to what may count as an autonomous choice or life.112We
108Hampton1993.109Hill1991,p.5.110Arneson2004,p.40.111Arneson2004,p.43.112Seee.g.Dworkin1988;MackenzieandStoljar2000.
73
wouldbe right to shareHampton’sworries aboutpeople likeTerryout in thenon-
idealworld, that is, womenwhomay have internalized gender normswhich cause
them to lead self-denying lives inmanyways. However, as amatter of principle it
seems implausible that choices, commitments, anddesires that frustrateone’sbasic
needs,forexample,canneverbeconsideredtrulyautonomous.Someonewholivesa
severely constrained life as a hermit, or someone who freely and voluntarily
undergoes gruelling, extrememilitary training programs,would have to be seen as
non-autonomousoutofprinciple.
Forthesamereasonswemightalsoworrythattheprinciplesofautonomystakesthat
wecanextractfromsuchviewsareimplausible.Hereisonepossibility.
The substantive self-respect approach: It is compatible with respect for an agent’s
autonomy that the agent bear the consequences (either causal or contextual) of their
autonomously made choices unless doing so is inconsistent with the substantive
requirementsofself-respect.
One problem of this principle for our purposes here − which, recall, are that of
identifying awayof singling outwhichburdens of parenting are incompatiblewith
parents´ being or remaining autonomous, thus being the potential bases for an
argument for sharing these costs − is that it leaves us particularly open to the
possibility that the rightway to ensure that people´s self-respect is not threatened
wouldbebypreventingthemfrommakingsomeoftheseself-denyingchoices,rather
than helping them bear the burdens. This is the response that many self-harming
pursuitsusuallyelicit.Furthermore,thisprinciplewouldcallforeitherpreventing,or
sharingthecostsof,virtuousinstancesofsacrificingone’sownimportantinterestsfor
the sakeof others, therebyundermining their value. It could also call for deflecting
someofthecostsoftherestrictedlifeofthehermit,orwouldotherwiserecommend
banningthatwayoflifeentirely.Thejudgmentsyieldedbythesubstantiveself-respect
approach,then,donotseemappealing.
74
3.5.2.Theself-authorshipaccountA now-widely endorsed view of personal autonomy is Joseph Raz’s self-authorship
account.AccordingtoRaz,anautonomous life isa life that isself-createdand freely
chosen. There are three sorts of capacities, or conditions, that must be in place to
enablesomeonetoauthor theirown lives.Theseare, first, that theagentmusthave
the appropriate mental capacities necessary to form and pursue a life plan, for
example theminimal rationality required to choose effectivemeans for one’s ends.
Second, theagentmustbe free frommanipulationor coercionbyothers.And third,
theagentmusthavearangeofsuitableoptionsavailabletochoosefrom.113
The thirdcondition isparticularly important forourpurposeshere.Razargues that
someone’s choices cannot be considered autonomous if the agent did not have a
meaningful, adequate rangeof options to choose from in the firstplace.His famous
examples of TheMan in the Pit and The HoundedWoman are meant to show that
havingmeaningfulandadequateoptionsisalsorequiredinordertofashionalifeof
one’sown.
TheManinthePithasonlyafewtrivialoptionsavailabletohim,suchaswhentoeat
orwhentosleep,butsincehe isstuck inapithe lacksanycontroloveranythingof
significanceinhislife.TheHoundedWomanfindsherselfintheoppositepredicament
assheisstrandedonadesertedislandwithacarnivorousanimalhuntingherdown.
Shehasoptionsavailabletoherthatareofgreatconsequenceforherlife,indeed,they
arelifeordeathchoices,likewhentorunandwheretohide.Yetshealsolackscontrol
overherlifesinceallofhereffortsandpowersmustbedevotedtothisonegoalthat
was imposedonherbyhersituation:stayingalive.Suchexamplessupport theview
that there are external conditions thatmust obtain if one’s choices are to count as
113Raz1986,pp.373-374.
75
autonomous. It is necessary, according to Raz, that one have an adequate range of
optionstochoosefrom,aspectrumrangingfromtrivialdaytodayoptionstomajor
commitmentsanddecisionsaboutthedirectionofone’slife.114Otherwise,justlikethe
HoundedWoman and theMan in thePit, one’s lifewould not be truly autonomous
despitehavingcontroloversomepartofone’slifeoranother.
Moreover,Razstresses that self-authorship isadynamic, life-longprocess: it is “the
visionofpeoplecontrolling,tosomedegree,theirowndestiny,fashioningitthrough
successivedecisionsthroughouttheirlives.”115Thisistosay,theautonomousperson
is not the onewhodecides upon a life plan early on in life and then sticks to it no
matterwhat. The autonomous person is someonewhohas the effective capacity to
change their values and plans as time goes on, and to put them into effect. This
suggeststhatthereneedstobesomeadequaterangeofoptionsavailableforonetobe
abletoreviseone’scommitmentsandplansevenoncetheyhavealreadyembarkedon
aparticularkindoflife.
This dynamic feature of the self-authorship, in particular, seems to bewhat Alstott
picks up on when she worries that parents may suffer autonomy deficits for the
durationofhavingtocomplywiththeNoExitrequirement.Theworryisthatparents
maynothaveadequateopportunitiestochangethecourseoftheirlivesortoreshuffle
their priorities when it comes to self-regarding projects. “The No Exit obligation,”
Alstottwrites, “curtails theexerciseof twocapabilities thatcitizensofa freesociety
ordinarilytakeforgranted:thecapacitytosetone’sownprioritiesamongcompeting
projects or values, and the capacity to revise one’s priorities and projects over
time.”116Recall,parentsarefreetorevisetheirvaluesandprioritiestosuchanextent
that theymay voluntarily relinquish the parental role altogether. If, however, they
wishtopersistintheirparentalrole,theirabilitytobetheauthorsoftheirownlivesis
hamperedbythefactthattheyhavetoconsulttheirchildren’sinterestswhensetting114Raz1986,p.375.115Raz1986,p.369.116Alstott2004,p.54.
76
theirlifepriorities,andtheyareindeedexpectedtosacrificepursuingvaluablegoods
andopportunitiesforthesakeoftheirchildren’sinterestincontinuityofcare.
Alstottdoesnotclaimthattheparents’freedomtopursuesuchvaluableopportunities
shouldtakeprecedenceattheexpenseofchildren’sneedforcontinuityofcare.What
shedoesclaimisthatitisimplausiblethataliberalstatecommittedtoprotectingits
citizens’shouldletparentsalonebearthesecosts.Shewrites:
[I]amnotarguingthattheNoExitruleisinappropriate,thatafairsocietyshouldreject
it. (…) The subtler insight that I am advancing is this: not every regulation with a
legitimatepurpose imposesa fairburdenon those individualswhopursue theregulated
activity.TheNoExitruleimplementsthestate’slegitimateinterestincontinuityofcare
for children but simultaneously imposes an extraordinary restriction on parental
autonomy.117
So, is it true that leavingparents tobearall thecostsofcomplyingwith theNoExit
requirement imposes, like Alstott says, an extraordinary restriction on parental
autonomy? I am assuming that parents autonomously decide to have and raise
children (i.e. they meet the local autonomy conditions for these decisions). The
questioniswhethersomeonewhoautonomouslydecidestohaveandraisechildren,
andwhoislefttobearallthecostsofcomplyingwiththeNoExitrule,canstillenjoya
minimallyautonomouslifeforthedurationoftheircompliancewithNoExit.Iamalso
assuming that it is not the case that complying with No Exit affects their internal
capacitiesformakingautonomouschoices.Ifthereisadimensioninwhichtheylack
autonomyoncetheyenterparenthoodandbegincomplyingwithNoExit, itmustbe
relatedtothelackofexternalconditionsforleadingminimallyautonomouslives.
TheexternalconditionsneededforaminimallyautonomouslifeonRaz’sview,aswe
have seen, involve having an adequate range of options for fashioning and revising
one’s lifeplans.Oneversionofaprincipleofautonomystakesthatwemightextract117Alstott2004,p.57,emphasisinoriginal.
77
fromhisviewandwhichwouldberelevant to theseconditions forautonomy is the
following.
Theself-authorshipapproach:Itiscompatiblewithrespectforanagent’sautonomythat
the agent bear the consequences (either causal or contextual) of their autonomously
made choices unless bearing them would leave one without an adequate range of
optionstochoosefromtoreviseandpursuetheirconceptionofthegood.
Relative to the principles that we considered in connection to the hierarchical
approachestoautonomy,bothoftheseversionshavetheadvantageofprovidingsome
formofpickingoutarangeofburdensthatareconsideredautonomy-infringing.Out
ofthetwo,thelatteroptionseemsmoreplausible.Itcouldbereadasalsoproviding
the contextualist approach with a standard for justifiability. That is, whether the
particularstructureofautonomystakes that is inplace is justified isdeterminedby
whetheritleavesindividualswithanadequaterangeofvaluableoptionstoreviseand
pursuetheirlifeplan.Icalledthisthecontextualistself-authorshipapproach.
The contextualist self-authorship approach only partly answers the question of
autonomystakes.Foritonlygivesusajustifiabilitystandardforthecoststhat,were
theynotsocialized,wouldcurtail theadequaterangeofoptions that theagentmust
always be guaranteed. But we might wonder whether it should not be seen as
autonomy-undermining that the agent may, compatibly with preserving her
autonomy,beaskedtobearanyandallcostswhichdonotjeopardizethatprotected
arrayofoptions.Forinstance,arguablytheoptionofkeepingone’scurrentjobwould
notbeonthelistofprotectedoptionsthateveryoneshouldbeguaranteednomatter
whatelsetheychoosetodo.Itcouldbemadepublicthatthosewhodecidetostayat
hometocarefortheirinfantchildrenareguaranteedtolosetheirjobsfornotshowing
uptowork.Everyonewouldknowinadvancethatcaringfull-timeforone’snewborn
child comes with the cost of losing one’s job. But surely this would seem like an
arbitrary attack on parents’ autonomy. No doubt there would be many reasons of
78
justicenottoattachthiscosttobecomingafull-timecarer,butcrucially,itisakindof
costthatalsoseemstoarbitrarilyrestrictparents’optionsforself-authorshipinaway
that,intuitively,shouldalsobecondemnedbyaprincipleofautonomystakes.
Thissaid,asImentionedearlier,forthepurposeofmakingheadwayontheparental
justicequestion, it is enough thatwehaveawayofdiscerningat least thenegative
answertothequestionofautonomystakes,namely,therangeofburdensthat,if left
for the agent to bear, would count as an autonomy deficit. The contextualist self-
authorship approach tells us that whenever someone’s autonomously made choice
lands them inaposition inwhich they lackanadequate rangeofoptions to choose
from to revise and pursue their conception of the good, they are suffering an
autonomy deficit. Insofar as bearing the costs of childrenwould leave parents in a
restrictedsituationdefinedthisway,then,theyaresufferinganautonomydeficit.
SincethisiswhatAlstotthadbeenarguingallalong,itmightseemthatherargument
is vindicated. But whether it really is vindicated, and to what extent, depends on
facinghead-onthequestionofwhatexactlyconstitutesanadequaterangeofvaluable
options.We need to knowwhether the opportunities that parents have to forgo in
providingcontinuityofcareare included intherangeofoptionsthatshouldremain
availabletoeveryone,includingtoparents.Givingananswertothisquestionisavery
difficulttaskinitself.Moreimportantly,itwouldlikelyinvolveusmakingsubstantive
commitmentsaboutwhatsortsofopportunitiesshouldalwaysbesecuredforpeople.
Forexample,itisnotobviousthatanoptionthatshouldbeincludedintheprotected
range should include theopportunity topursueaproject thatwouldbeessential to
parents’ individual flourishing,butwhichwouldtakethemawayfromtheirchildren
for far too long.Whetherornot thisoptionshouldbeprotectedwill inevitablybea
controversialissue.
Whatever that list of adequate options would contain (assuming arguendo, for the
time being, that a list can be compiled despite the danger of offending against the
79
principleofpoliticalneutrality),itdoesnotseemverylikelythatthecostsofNoExit,
whichAlstottisconcernedwith,wouldbesoburdensomeastojeopardizethemtoa
worrisome degree. This is because the costs of No Exit themselves are not as
restrictiveastheyfirstappear.Fornotethatthestandardofparentingimposedbythe
NoExitruleisaminimalone.Thedigressionsthatparentswouldhavetocommitin
order tohave their parental rights revokedon groundsof breachingNoExitwould
have to be fairly extreme. The parents would have to fail to provide even the
minimumlevelofstabilityrequiredforchildrennottocountasabusedandneglected
intheeyesofthelaw.Providingthesortofcarethatmeetsthisstandardisarguably
compatiblewithawholerangeofself-regardingopportunitiesthatparentsmighttake
advantage of. People do move to other cities, and even other countries, with their
children, thus introducing some degree of instability into their lives. They also
separatefromtheirchildrenfor,perhaps,atleastafewmonthsatatime,withoutthis
providing legitimate grounds for stripping them of their parental rights. And they
oftenseverandstartnewromanticrelationships.Itdoesnot,then,seemplausibleto
saythatcomplyingwiththeNoExitrequirementdeprivesthemofanadequaterange
ofopportunitiesforself-authorshipeveniftheyhavetobearallthecoststhemselves.
Itseemsmorelikelythatparents’adequaterangeofoptionsforself-authorshipwould
be frustrated if they had to bear all the costs of providing a good upbringing
themselves,aswellasall thecostsofbeingoncall.Mostparentsstrivetogivetheir
children a good upbringing, which involves investing huge amounts of time and
resourcesintoraisingtheirchildrentobehealthy,happy,productivecitizensthatcan
be economically competitive in their own societies. As a result, no doubt that the
average parent is leftwith relatively little time and few resources to invest in self-
regarding pursuits, hence with less effective opportunity to revise their plans and
reshuffletheirpriorities.Ontheotherhand,wemustalsokeepinmindthatinmost
just societies somecostsofagoodupbringing, suchasprimaryeducationcostsand
basichealthcare,willbesharedbytherestofsocietyforindependentreasons.
80
So,havingtobearthecostsofagoodupbringingdoeslooklikeitwouldleaveparents
withfeweropportunitiesforself-authorshipthanhavingtobearthecostsofNoExit
would,butjusthowmuchmorerestrictivethisisdependsonwhatcostsarealready
sharedbysocietyforotherreasons.And,onceagain,italsodependsonwhichofthese
(further) lossesofopportunitiesshouldalwaysbeprotectedbythestate,something
thatisdifficulttoestablishwithoutmakingsomesubstantivecommitments.
Consider, now,what I have called the costs of being on call. These are the costs of
beinga full-timecarer foran infantorayoungchild.Thinkbacktothewomanwho
wrotealettersignedTheBadMom.Sheseemstosharealotofsimilaritiesbothwith
Raz’sHoundedWomanaswellashisManinthePit.SheisliketheHoundedWoman
because it seems like all of her efforts, all her physical and mental resources are
directed tomeeting thechildren’s fundamentalneeds.She findsshehasno time for
herself,sheisnotevenincontrolofhowmuchshegetstosleeporwhenshegetsto
eat.Worsestill,sheisnotconsumed,liketheHoundedWoman,bythesingular,urgent
goalofkeepingherselfalive.Sheisfullyabsorbedbythesingulargoalofkeepingher
childrenaliveandwell.Herchildrenarestillsoyoungthattheyareentirelyrelianton
her for survival, and they are at such a crucial moment in their development that
virtuallyallofherdecisions,eventhosethatseemtrivial,havepotentiallyimportant
consequences for the child’s development. The relentless pressure of making high
stakes decisions onbehalf of her children ends upusurping her dominion over her
ownlifetotheextentthatshefeelsunfreetodecideevenontrivialthingsforherself
likewhentosleepandwhentoeat.
Furthermore,italsoseemsunlikelythat,atleastforthetimethatsheis“oncall”,she
couldhavemuchopportunitytomakechoicesofgreatsignificanceforherself.Inthis
respectsheisliketheManinthePit.Ifshebarelyhasanycontrolovertrivialaspects
ofherdaytodaylife,itmustbethecasethatshealsolackstheopportunitytomake
high impact self-regarding decisions such as whether to dedicate herself to other
importantcauses,or,say,whethertochangecareerpaths.Indeed,full-timecarersof
81
infants and young children do not even have the time to pursue their usual
occupation, letaloneexploreothercareeroptions,hence theexistenceofpoliciesof
paid(orunpaid)parentalleaveinmodernsocieties.
Perhapsnotallparentshavesuchahardtimeastheself-describedBadMom(who,by
theway,seemslikeagreatmomtoherchildren,toherdetriment).Someparentshave
amorecasualstyleofparentingthanothers,andsomehavepreferencesthathappen
to fit nicelywith the demands of parenthood, so the degree towhich they have to
sacrificethemselvesmayvary.Butbyandlarge,therecanbenodoubtthatcaringfull-
timeforinfantsandyoungchildrencanbeveryburdensomeindeed.
So,at thevery least,wecanextract fromRaz’sself-authorshipaccount thisminimal
constraintconcerningwhichburdensofparenthoodcountasautonomy-undermining
ifparentswere tobear themalone: it is those costs thebearingofwhich leaveone
withverylittletonocontroloverbothtrivialdaytodayoptionsaswellasovermajor
commitments and decisions about the direction of one’s life. And the costs of
parenthoodthatfitthebillmostobviouslyarethecostsofbeingoncall.Thisisinline
withwhat feminists have been pointing out for decades now, namely that full-time
carersfaceextraordinarilyburdensomelivingconditions,118andthesefull-timecarers
arepredominantlywomen.119
Moreover, and importantly, state-commitment to protecting people from living in
such conditions would also be compatible with the principle of political neutrality.
The range of options that we would be committed to protecting for them is so
minimal,involvingforexampletheopportunitytomeetone’sbasicneedswhileinthe
118E.g.Okin1989;Young1990.119Indeed,inadditiontocaringforchildren,womenmakeupthemajorityofthoseinanykindofcaregivingrole,includingcaringforthesick,theelderly,andthedisabled.SeeAmericanPsychologicalAssociation2009.Theautonomy-basedargumentsapplyingtocarersfortheyoungshouldalsoapplytothosewhocarefortheseothervulnerablecategories.Inthecaseofseverelydisabledpersons,theburdensofbeingoncallmaylastalifetime.
82
caring role, that we need not make any controversial decisions about further
opportunitiesandlifeplansthatmustalwaysbekeptopentothem.
This provides a good first step in a pro-sharing argument for sharing the costs of
children. But the conclusion is still far from secured. For, recall, a successful pro-
sharing argument is one that establishes (a) that unless conditions C are met, an
individual suffers an autonomy deficit, either relative or absolute; and (b) that the
autonomy deficit would be unjust. So, while we have discovered at least a partial
answer to (a), we still need to know what are the just autonomy-based claims of
citizensagainstthestatearemoregenerally.Withoutafulltheoryofthejustclaimsto
sufficient, or perhaps equal, autonomy, we cannot have a conclusion to our pro-
sharingargument.
To illustrate why, consider the fact that a full theory of individuals’ just claims to
autonomymightonly involve thestatecommitment to securingequalopportunities
for autonomyacross a lifetime.Citizensmay thenhave autonomy-based complaints
only if their lifetime share of opportunities for autonomy is smaller than others’
shares.Bearingallthecostsofbeingoncall,whileinvolvingsomeautonomydeficits,
canstillbecompatiblewiththecarers’enjoyingequallifetimesharesofopportunities
forautonomy.Forinmostcasespeopleare“oncall”foronlyafewyearsoftheirlives,
typicallywhentheirchildrenareveryyoung.120
Otherquestions that this theorywouldhave to settle include, for example,whether
peoplecanraisecomplaintsagainstthestatefortheautonomydeficitstheysuffer if
theyhadampleopportunitytoavoidplacingthemselvesinthatsituation.Thereisalso
afurtherquestionhereconcerningthemeansbywhichastatemayorshouldensure
thatpeopleavoidautonomydeficits.Itcouldseektosetthecostsofeachactivitysuch
that none is so burdensome as to leave people below a threshold for sufficient
120Withthenotableexceptionofthosewhocarefordisabledchildren,somethingthatmightendupbeingalifelongcommitment.Parentingingeneralisalifelongcommitment,ofcourse,butarguablythetypicallifeofaparentisnotalwaysasrestrictedaswhentheycareforyoungchildren.
83
autonomyforanystretchoftheirlives.Thiswouldtranslateintosocializingthecosts
of those ambitions that otherwise would leave people below that threshold, and it
couldhelp secureapro-sharing conclusion regarding the costsofbeingon call.But
anothermeansbywhichthestatecanpreventpeoplefromfallingbelowthethreshold
is by banning the sorts of ambitions thatwould put them there, like the state does
withtheoptionofsellingoneselfintoslavery.
To sum up this section, the self-authorship view of personal autonomy, though not
offeringafullprincipleofautonomystakes,offersusaplausiblewaytodiscernwhich
costsofparentingcountasautonomydeficitsifleftforparentstobear.Mostplausibly,
andcompatiblywiththeprincipleofpoliticalneutrality,thesearethecoststhatfull-
time carers of infants and young children typically bear. Anne Alstott’s proposed
autonomy-based case for sharing the costs of children, then, has a lot of initial
plausibilitywithrespecttothesecosts,ratherthanthecostsofcomplyingwiththeNo
Exitrulethatherownaccountfocuseson.
Evenso,theargumentsprovidedhereareonlythefirststepinasuccessfulautonomy-
basedargumentforparentalsupport, for inordertoknowwhethertheseautonomy
deficitsgroundacomplaintonthepartofparents,andwhatexactlytheymayclaim,
remainstobesettledbyaviewofhowthestateoughttobalancepeople’sclaimsto
autonomy,andwhatformthestate’seffortsshouldtakeinpromotingthoseclaims.
Perhapssurprisingly,theliteratureonthevalueofautonomyhasmadebutveryfew
attemptsatdrawingouttheimplicationsofthevalueofautonomyforhowautonomy
shouldbedistributedinajust,liberalegalitariansociety,andthefewthatexistarenot
ofhelptousinconstructingtheparentaljusticecase.121Razhimselfdoesnotoffera
viewastohowautonomyasself-authorshipshouldbepromotedbythestate.122Such
aviewcouldbedeveloped,ofcourse,butattemptingtoprovideonemyselfwouldtake
121Forinstancebecausetheyareperfectionistviewsofjustice,whilewewerelookingforproposalsthatarecompatiblewithpoliticalneutrality.SeeHurka1993,chs11and12.122Raz1986,ch.9.
84
usfarbeyondthescopeofthisproject.Iwillconsider,however,oneaccountofhow
theclaimstoautonomyshouldbedistributedthatwasdevelopedspecificallywitha
viewtoultimatelygroundingparents’claimstosocietalsupport.Thisviewhasbeen
proposedbyPaulBou-HabibandSerenaOlsaretti,anditmaintainsthatpeoplehave
claimstoequalautonomy,andtherequirementsofequalautonomyshouldbesetby
usingDworkin’shypotheticalinsuranceexercise.123
Beforeconsideringthatview,Iwanttodiscussanothertypeofaccountofthevalueof
autonomy, namely the so-called relational views. It is worth doing so, first, for the
sakeofamorecompletelandscapeoftheautonomyviewsonofferandthepotential
principles of autonomy stakes thatmight be extracted from them. Secondly, if it is
possibletoextractfromtheseviewsaprincipleofstakesthatcondemnsawiderrange
of burdens as autonomy deficits as compared to the self-authorship view, this will
bode well for parents’ prospects. For, as I have noted, the self-authorship-based
argument, evenwhen supplemented by a viewof the just distribution of autonomy
claims,onlycondemnsarathernarrowsetofcosts, i.e. thosebornewhilebeing“on
call”.Assuch,itcouldonlysupportpoliciesaimedspecificallyatrelievingtheburdens
of full-time carers, such as publicly funded childcare facilities, andwould therefore
havemore limited public policy implications thanAlstott had hoped.124Itwould be
salutaryforthepro-sharingcause,then,ifwecouldprovidethegroundsformorefar-
reaching family support policies, and relational views may seem promising in this
context,sincetheytypicallyimposemoredemandingconditionsforautonomythando
the views we have already reviewed So let us take a look at one last family of
autonomyviews:therelationalviews.
123Bou-HabibandOlsaretti2013.124Alstott2004,part3.
85
3.6.RelationalautonomyRelationalautonomyaccountspropose that individualsand their relevantcapacities
for autonomy should be conceived relationally. These views theorize individuals as
dependent on others and vulnerable to some degree or another, and thus seek to
maketheiraccountofanautonomouspersoncompatiblewithnotionsofdependency
and vulnerability. Furthermore, individuals’ practical identities cannot be separated
from their social, historical, and cultural context. As a result, relational accounts of
autonomyemphasizetheimportanceofthesocialconditionsthatmustbeinplacefor
anindividualtodevelopthenecessarysetofcapacitiesforleadinganautonomouslife.
For instance, socially caused deficits in self-esteem are considered to prevent
someone from living autonomously. Someone whose environment (through its
institutionsandthroughtherelationshipsinwhichtheagentisengaged,forinstance)
is constantlyundermining their senseof self-worth is likely to fail to relate to their
ownattitudesandchoices inawaythatseemsnecessaryforautonomy, for instance
theymight not see their choices as authoritative.125As CatrionaMackenzie puts it,
relational views aim to acknowledge the “extensive interpersonal, social, and
institutional scaffolding necessary for the development and ongoing exercise of the
complexcognitive,volitional,imaginative,andemotionalskills”126involvedinmaking
autonomousdecisions.
Ihastentonotethatrelationalautonomyisanotiondevelopedontheassumptionof
non-idealbackgroundconditions.Relationaltheoristsaremotivatedbysuchconcerns
as theorizing the effects of oppressive socialization and oppressive institutions on
individuals’capacitiesandopportunitiestoleadanautonomouslife.127Assuch,their
theoretical apparatus is geared towards making sense of autonomy in unjust
circumstances. One might worry, then, about the relevance of non-ideal theorizing
125Seee.g.Benson1994.126Mackenzie2014,p.22.127Seee.g.Meyers1989.
86
aboutautonomyforthepurposesofanidealprojectsuchastheoneIamengagedin.
Feministsdonotworkwithinidealframeworkslikeluckegalitariansdo,forinstance,
astheybelievetheseframeworkstobetoonarrowlyfocusedoninstitutionaldesign
ratherthanalsoonthesocialwebofrelationshipsassuch,andtobeconcernedonly
withinequalitiesofresourcesratherthanalsoinequalitiesinpower,esteem,orsocial
standingwhichmayconstituteformsofdomination.
I am inclined to say that relational views are of interest to this chapter at the very
least insofar as we can extract valuable lessons about the relevance of the
“institutionalscaffolding”neededforautonomy.Indeed,wemightbeconvincedbythe
feminist view about the complex role that institutional factors play for the
constitution of the self and for acquiring and exercising the requisite capacities for
autonomy, andwemight think it important toaskhow these institutions shouldbe
organized in light of this.Weneednot assume conditionsof oppression as such, or
other forms of background injustice, in order to be interested in conceptions of
autonomythatplaceimportanceontheinstitutionalfactorsthatmightberelevantfor
autonomy. In fact, this sort of focus seems particularly suited for the question
exploredinthischapter.
One account that is particularly interesting for the question of autonomy stakes is
Mackenzie’s. She offers a complexmultidimensional view of autonomy that tries to
bring together all the main categories of conditions that have been thought to be
uniquelyrelevantforautonomy.Shearguesthatautonomyisnottheunitaryconcept
thattheliteraturesofarhasthoughtittobe.Thatis, itdoesnothavejustonesetof
necessaryandsufficientconditions.Weneednotargueaboutwhether it isonly the
motivationalstructureoftheagent,oralsoexternalconditions,thatarenecessaryand
sufficient for autonomy. Instead, she argues, autonomy “involves three distinct but
causallyinterdependentdimensionsoraxes:self-determination,self-governance,and
87
self-authorization.” 128 Self-determination involves having the freedom and
opportunitiestodecideuponone’svaluesandtoeffectivelyactuponone’svaluesand
preferences.This axis tracks theexternal conditions for autonomy,namely freedom
and opportunity conditions respectively. The self-governance axis identifies the
internalconditionsforautonomy,namelythecompetenceandauthenticityconditions
that mainstream procedural views have tended to focus on. The self-authorization
dimension “involves regarding oneself as having thenormativeauthority to be self-
determining and self-governing.”129The conditions for autonomypicked out by this
axis include self-evaluative attitudes, accountability for one’s reasons, and social
recognition. She then goes on to develop relational conceptions of these three
dimensions,thatis,conceptionsthataresensitivetofactsabouttheembeddednessof
theindividualwithinherhistorical,social,andinstitutionalenvironment.Mackenzie’s
thought is thatdifferentdimensionsofautonomymightbemoresalient indifferent
contexts.Inmedicalsettings,forinstance,shesuggeststhatself-governanceandself-
authorization seem to be crucial, while self-determination seems more salient for
democraticcitizenship.130
Forthequestionofstakestheself-determinationaxis isofparticularrelevance.Like
Raz,Mackenziebelievesthathavingandadequaterangeofsignificantopportunitiesto
choose from when deciding how to lead one’s life are an important condition for
autonomy.Sheargues,however,thatwhatexactlytheseopportunitiesshouldbeisa
matterofsocialjustice,andthatitisbestsettledbyappealtoaviewlikecapabilities
theory.131We need a list of valuable capabilities that anyone in our society should
haveequalaccessto.Unlesspeoplearemakingtheirchoicesaboutwhattovalueand
whattodoagainstabackgroundofguaranteedcapabilities,theirchoicesdonotcount
asautonomous.Thissuggeststhefollowingprincipleofstakes:
128Mackenzie2014,p.17.129Mackenize2014,p.18.130Mackenzie2014,p.40.131Sen1992;NussbaumandSen1993.
88
The capabilities approach to stakes: It is compatible with respect for an agent’s
autonomy that the agent bear the consequences (either causal or contextual) of their
autonomously made choices unless bearing them would jeopardize one’s valuable
capabilities.
ComparedtoRaz’sopen-endedlistofwhatconstitutesanadequaterangeofoptions
to choose from, the appeal to capabilities theory has the advantage of potentially
giving us a substantive list of opportunities for well-being that should form the
backdropof one’s decisions abouthow to shapeone’s life. This is the case if one is
happy to go for a list of basic capabilities likeMartha Nussbaum’s, for instance, as
Mackenzierecommendswedo.132
However, we now face, once again, the challenge from political neutrality. For in
assemblinganylistofbasiccapabilitiesthateveryonemustbeguaranteed,onehasto
commit to some selection or other of the aspects of individual well-being that
everyone should be able to attain should they choose to do so. Deeming certain
aspectsofwell-beingasprotectedtotheexclusionofothersseemstooffendagainst
theprincipleofpoliticalneutrality.
Perhapsevenmoreimportantly,therelationalviewseemsunabletoyieldadistinctive
principleofautonomystakes.Foritseemsthatwecouldby-passtheentirediscussion
aboutautonomyaltogetherandadoptacapabilities-basedviewofjusticethatwould
supportparents’claims,aswellasanyoneelse’sclaims, tohavingacertainrangeof
opportunitiesprotected,presumablythroughouttheirlives.
132Mackenzie2014,pp.29-30.SeeNussbaum2006.
89
3.7.ThedemandsofequalautonomyfrombehindaveilofignorancePaulBou-HabibandSerenaOlsarettihaverecentlyofferedwhatItaketobeavalue-
neutralattempttosettlethedemandsofequalautonomy.Asopposedtotheaccounts
ofthevalueofautonomydiscussedabove,whichfocusontheconditionsunderwhich
someone’s desires or choices are properly attributable to the agent, their proposal
addresses the issueofhowtheclaimstoautonomyshouldbebalancedbythestate.
Thisistheonlyproposalintheparentaljusticeliteraturethatattemptstoestablisha
wayofmeasuringautonomydeficitsthatgivegroundstoegalitarianclaims.Theydo
this by asking, which are the valuable pursuits that everyone should enjoy equal
opportunitytopursueregardlessoftheircircumstances?Andhowwouldthecostsof
thosevaluablepursuitsbeshared?Theyproposethatweshouldanswerbothofthese
questionbyusingDworkin’shypotheticalinsurancedevice.
The veil of ignorance they propose is Dworkinian with respect to the personal
information it allows individuals to base their decisions on. People behind the veil
wouldhave information about their character and inclinationsbutwouldnot know
whichlifeplanstheywillactuallywanttopursue.TheprojectdepartsfromDworkin,
however,inthatthedecisionsbehindtheveilarenotconnectedtobrutebadluck.The
questionthatindividualsbehindDworkin’sveilarecalledtoansweriswhether,and
towhatextent,theywouldbewillingtoinvestinordertomitigatetheeffectsofbrute
badluckontheirlot.Bycontrast,thequestionthatindividualsareaskedbehindthe
Bou-Habib-Olsarettiveilofignoranceis,‘Whichplansoflifewouldyoubeprepared
toshare thecostsof therebyensuring thatyouwillhaveanequalchance topursue
themonceyousettleononeormoreofthem?’Ofcourse,theyarguethat,whenasked
suchaquestion,individualsbehindtheveilwouldincludeparentingonthelistoflife
planstheywantedtosecureforthemselves,muchliketheargumentfrominsurance
thatIhaveaddressedinthepreviouschapter.
90
Theresultingprincipleofstakescouldtakethefollowingform:
Theveilof ignoranceapproach: It is compatiblewithrespect foranagent’sautonomy
that the agent bear the consequences (either causal or contextual) of their
autonomouslymadechoicesunlesstheirchoicesarethosewhicheveryoneshouldhave
equalopportunitytomake,asdeterminedbythehypothetical insurancedecisionsthat
theaverageindividualwouldhavemadebehindaveilofignorance.
Theadvantageofthisapproachisthatitoffersusawaytodefinethecostsofvarious
activities such as raising children by reference to a familiar independent standard,
namelyhowpeoplebehindaveil of ignorancewould choose to fix them.Moreover,
theveilofignoranceapproachalreadygivesusananswertothedistinctquestionof
whatsociety’sanswershouldbewhenfacedwithparticularautonomycosts.Namely,
thelifeplansthatindividualswouldhaveinsuredinahypotheticalinsurancemarket
aretobecollectivelyfundedbytheentiresociety.
Inresponse, Iwant topointout that thereseems tobesomethingproblematicwith
theapplicationoftheveilofignorance-typeofargumentinthiscontext.Theveilisso
thin, and the question so exclusively focused on establishing the preferences of the
majority,thatithardlyseemsusefultousetheveilofignoranceatall.Thesourceof
thetheoreticalproblemwithhowBou-HabibandOlsarettiusethedeviceisthat, for
theirpurposes,thereisnorealneedforobscuringanyinformationaboutindividuals’
actualpreferencesandambitions.
Theveilofignoranceistypicallyconceivedasausefuldeviceforensuringfairness.In
Dworkin’s case, the veil obscures individuals’ level of exposure to various forms of
brute bad luck.Wedonotwant people’s insurancedecisions to dependonmorally
arbitrary factors such ashow safe they are from the riskof developing aparticular
kindofdiseasecomparedtootherswho, throughnofaultof theirown,carrysucha
risk. It seems only fair to pool together suchmorally arbitrary risks. Dworkin then
91
allows for information about people’s tastes and inclinations to influence insurance
decisionsbehindtheveilbecausehethinksthisisnecessaryifpeopleareexpectedto
make any meaningful insurance decisions. The point of this is to render the
hypothetical insurancedecisions, and the resulting tax systemmodelled after them,
ambition-sensitive.133
TheBou-Habib–Olsarettiveilofignorancedeviceservesnofairnesspurpose.There
isnobrutebadluckthattheindividualsareaskedtotakeresponsibilityforinmaking
insurancedecisions.Andtheveilisotherwiseverythin,allowingpeopletoknowtheir
tastesandinclinations,allowingthemtoknowwhatsortsoflifeplanstheyarelikely
towanttopursue,giventheirparticularities.Noinformationthatmightbeconsidered
unfairtouseindeliberationisobscuredbytheveil.Soitishardtoseewhywewould
needaveilatall,ratherthanjustdirectlyask,“Whichpreferencesandwaysoflifedo
wefindpeopleinoursocietytendtohave,andhowwouldtheydecidetosetthecosts
ofpursuingthem?”
There does not seem to be much difference between using the veil, and directly
appealingtoadifferentinterpretationofsettlingthedemandsofequalautonomythat
Bou-Habib and Olsaretti consider, but reject.134This latter approach would simply
consider the specific life plans that individuals happen to prefer (presumably by
lookingatthepreferencesoftheaverageperson,orthemajority,insociety)andthen
askhowtheywouldsetthecostsofthoselifeplansforthemselves.
Bou-HabibandOlsaretticlaimthattheadvantageoftheirapproachisthatitismore
prudentcomparedtothismoredirectapproach.Consideringwhatplans“realpeople”133Forcomparison,consideralsoRawls’suseoftheveilofignoranceindevisinghisprinciplesofjusticeasfairness.Rawls’sveilofignoranceiseventhickerthanDworkin’sashebelievesitwouldbeunfairforindividuals to choose principles of justice based on their personal inclinations and interests, andperhapsendupwithskewedprinciplesthatreflectamajority’sidiosyncratictastesandinclinations.So,here,too,theveilismeanttorendertheprocessofdeliberationaboutjusticeprinciplesfairinthefaceof morally arbitrary facts, which in this case include the personal interests that individuals in thedeliberativecommunitywouldhappentohaveandwouldwanttopushforattheexpenseofthosewhomightnotsharethem.SeeRawls1999,pp.11,16-17.134Bou-HabibandOlsaretti2013,p.431.
92
arecommittedtowillobscurethefactthatmostpeopledonothavefixedplansand
thattheyoftenchangetheirpreferencesandplans.Behindaveil,peoplewouldnotbe
mistakenlyweddedtotheplanstheyhappentohaveatthattime. Instead, theclaim
goes, they would have to rely on information about their general inclinations and
tastesandwouldlikelyendupsecuringawiderrangeoflifeplansthanactualpeople
intherealworldwould.
It is doubtful, however, that the veil of ignorance really presents a prudential
advantageoverthedirectapproachtofixingautonomycosts.Idonotseeanyreason
why,when asking real people how theywouldwant to set the costs of various life
pursuits,theycouldnotalsohaveinformationabouttheirmoregeneraldispositions
such that they, too,would know thatmore than one life planwas compatiblewith
theircharacter,eventhoughtheyhappenedtosettleonaparticularoneatthatpoint
in time. And more importantly, there is no reason why they could not also have
information regarding the likelihood of changing their minds later on. As such, it
strikes me as an even more prudent approach to simply ask actual people, “How
would you fix the terms of various life plans, given your current preferences and
commitments,aswellasthepotentiallifeplansyoumightchoosetopursueatsome
laterpoint,givenyourcharacteranddispositions?”
Tosumupthesepoints,itseemslikeusingtheveilofignoranceinthiscaseis,atbest,
redundant. It serves no fairness purpose, and it does not seem to be especially
prudentially advantageous. At worst, it is less prudent than just asking how actual
peopleinoursocietywouldchoosetosetthecostsofthevariousactivitiesthatthey
know for a fact theywant to pursue,plus further activities that theymightwant to
pursueatsomelaterpoint.
There is a deeper worry with this approach to setting autonomy stakes. It is the
familiarworry of using the hypothetical insurance device to determine substantive
issuessuchaswhatistheproperscopeofautonomyinpursuingparticularlifeplans.
93
Whetherbehindoroutsidetheveil,wewoulddecidewhetherparticularrestrictions
and burdens areworrisome by reference towhat amajority of peoplewould have
chosen.Thisdoesnotseemlikeanappealingsolution.
Hereisonewayinwhichthissolutionseemstolackappeal.Itimpliesthatlifeplans
with structurally similar burdens and restrictions in terms of autonomy could be
treateddifferently simplybecause theyhappen tobepopular. If, as it is the case in
reality, a majority of people (behind or outside the veil) would be inclined to
becomingparents,butwouldnotbeinclinedtobecomingemergencysurgeons,then
equal autonomy would require that the surgeon bear all the autonomy burdens
herself,whiletheparenthavethemrelieved.Theemergencysurgeon,liketheparent,
mustbeavailable tomeet theurgentneedsof theirpatientsatanyhourof theday.
Theyalsohavetomakeenormoustimeinvestmentsintocaringfortheirpatientsina
way that might undermine their ability to sustain an adequate number of other
relationships. They cannot afford to start traveling theworld, get involved in other
meaningfulprojects,andsoon.Comparedtononparents,parentsenjoyconsiderably
fewerdiscreteopportunitiessuchaspursuingcertainexpensiveandtime-consuming
projects. Analogously, compared to, say, freelance software developers, emergency
surgeons enjoy fewer discrete opportunities likeworking from the comfort of their
ownhome,orfromanyplaceintheworldthathasanInternetconnection.Theparent
and thesurgeonseemto facesimilar restrictions,both in termsof thesortsof time
andcarecommitmentstheymustmaketotheirchildrenorpatients,andinthesense
that theybothhave fewerdiscreteopportunities forself-authorship than thosewho
chosemorecommitment-freelifestyles.Yettheveilofignoranceapproachwouldtreat
themdifferentlybecauseparentinghappenstobeamorepopularlifeplanthanbeing
asurgeon.
That similar autonomy burdens would be judged differently is not even the most
problematicaspectofthissolutionfromthepointofviewourpurposehere,whichis
toestablishwhatpeople’sequalclaimstoautonomyare.Itis,instead,thefactthatthe
94
veilofignoranceapproachtoautonomystakesseemstotakeustoofarawayfromthe
reason why we care about autonomy, and about its equal distribution, in the first
place. The reason why we care about autonomy in the first place is that we value
expressingouragencyinmakingchoicesanddeterminingthecourseofourlives.Yet,
the burdens that should count as autonomy-undermining or autonomy-preserving
under the veil of ignorance approach seem problematically tied to how popular or
unpopularcertain lifestylesare inaparticularcommunity,ratherthantosomething
thatreallybelongstoindividuals’agency.
Certainly, one’s deciding on a life plan rather than another, and choosing to insure
behindaveilofignorance,isanexpressionoftheiragency.Butwhethertheburdens
associatedwithone’slifeplancountasaninfringementofautonomyattheendofthe
day, or the rightful expression of it, depends crucially on how many people also
happen toprefer the sort of life plan that brings about that particular burden.This
mightstrikesomeofusasproblematic.ColemanandRipsteinmakeasimilarpointin
connectiontotheattributionofsubstantiveresponsibilitytopeople:
Ifsomeonetakesariskthatothersinsimilarcircumstancestypicallyavoid–pursuing
the lifeofanartist,ormakingarisky investment–thegoodorbad luckthataccrues
belongstothatpersonalone.Incontrast,ifsomeonedoessomethingthatisnormalfor
those in similar circumstances, the results might be thought of as brute luck. […]
Whether something counts as choice or circumstance will depend on the frequency
with which others make similar choices. […] If there is a wild frenzy of stock
speculation, the losers will need to be indemnified. But if a neighbourhood is so
dangerousthatfewpeopleventureoutatnight,theywillcountasrisk-takersand,so,
not have their losses made good. […] this approach ties responsibility to unusual
activities.Asaresult,itseversthetiebetweenagencyandresponsibility.135
The veil of ignorance approach, then, can be useful for many purposes, but not, it
seems, for establishing the just stakes for substantive responsibility or for equal135ColemanandRipstein,pp.124-125.
95
autonomy. This is because responsibility and autonomy are supposed to be tied as
closely as possible to individuals’ agency in making choices and determining the
course of their lives. It seems inappropriate that the stakes of responsibility or the
stakes of autonomywould depend on contingent facts about the actual inclinations
and preferences of those in a particular society. It also seems unfair that, because
these contingents facts areallowed toplay this crucial role, theemergency surgeon
wouldlackanautonomycomplaintdespitefacingthesamesortofautonomydeficits
astheparent,whowould.
3.8.ConclusionIn this chapter I have reviewed some facets of the burdens that parents bear that
intuitively seem to provide grounds for concern about parents’ being able to lead
autonomouslives(atleastforthedurationoftheiroffspring’schildhood):thecostsof
NoExit,thecostsofagoodupbringing,andthecostsofbeingoncall.Ihavetriedto
showthatakey ingredient forourabilitytodeterminewhichof thesecostsdoturn
outtobeautonomy-underminingforparentsislargelymissingfromtheliteratureon
personal autonomy. This ingredient is what I have called a principle of autonomy
stakes.AlthoughtheautonomyaccountsIhaveconsideredbynomeansexhaustthat
literature, Ihope Ihaveofferedenoughreasons tosuspect thatat least someof the
most prominent views of autonomy do indeed face this challenge. Content-neutral
viewsofautonomythatfocusonthepsychologicalfeaturesoftheagentseemtooffer
usnoguidanceastowhichresultingburdensshouldbeamatterofrespectingone’s
agency in determining the course of their life. Autonomy accounts that place some
externalconditionsonwhatcountsasautonomousdesiresorchoicesdoofferussome
guidance.However,whether they are conditions on the choice situation (and,more
broadly,ontheinstitutionalsituation)thattheagentfindsherselfin,orwhetherthey
areconditionsonwhatkindsofdesireorchoicecontentcancountasautonomous,the
96
principles of stakes they yield either are not very attractive (as in the case of
Hampton’s self-respectview),ordonotyielddistinctiveautonomystakes (as in the
caseofrelationalviews),ortheyyieldverylimitedresults(asinthecaseofRaz’sself-
authorshipview).Thisissoatleastifweremaincommittedtoaprincipleofpolitical
neutrality.Asa consequence, theautonomy-basedcase forparental justicehasonly
verylimitedimplicationsforparents’claimsofjustice,namelythattheburdensoffull-
time carers for infants and young children should be alleviated by the state,
presumablythroughpoliciesofpaidparentalleave,subsidizedchildcarefacilities,and
the like,whichdeveloped countriesby and large alreadyoffer (theU.S. is anotable
exceptionamong liberaldemocracies for failing toprovidepaidparental leave).The
autonomy-basedcaseforparents’claimshasthepotentialtogroundawiderrangeof
claims on the part of parents, but its chances for success depend crucially on the
prospectsforthedifficulttaskoffindingaplausibleprincipleofautonomystakesthat
wouldcondemnawiderrangeofcostsofparenthood.
InthenextchapterIturntoadifferentkindofpro-sharingargument,theonethathas
undoubtedlyenjoyedthemostpopularity.Itsgeneralstructureisthis:itmaybetrue
thatparentsareresponsibleforcreatingcostsbyhavingandraisingchildren,butthis
is only half of the story. By having and raising children, they also create important
publicgoods forsociety,goodsthatcanberecognizedasvaluableby freeandequal
people despite their different conceptions of the good life. Centrally, these goods
involve the economic contributions that children provide once they become adults
andstartparticipatingtothevariousschemesofsocialcooperation.Itwouldbeunfair
ofnon-parentstobenefit fromtheworkofparentsbyenjoying, ineffect, thehuman
capitaltheyproduce,withoutalsosharingthecostsforproducingit.
97
Chapter4
TheArgumentfromFairness
Several political theorists have developed arguments to the effect that non-parents
(peoplewho choose not to have children) have an obligation to support parents in
virtueof thecontribution that the lattermake to society.Theseaccounts relyonan
empiricalpremise thatdescribes thebenefitsparentsproduce for societybyhaving
and rearing children, and on a normative premise that establishes the conditions
underwhich thosewhoproducebenefits have a claimof fairness that thirdparties
whointernalizethemhelpmeetthecostsofproduction.Thesearguments,whichwe
mightcallfairness-based,thenconcludethatparentswhoproducebenefitsbyhaving
and rearing children meet the conditions for having fairness claims against non-
parentswhoenjoythebenefitsofnewgenerations.
After a brisk outline of the purported benefits that parents create, I focus on the
principleprovidingtheseargumentswiththeirnormativebasis:theso-calledfairness
principle (or Fair Play principle). Themajority of this chapterwill be dedicated to
developinganewaccountofthefairnessprinciple,whichIcalltheSharedPreference
View.InthenextchapterIinvestigatewhetherparentsmeettheconditionsforhaving
fairnessclaimsagainstnon-parentsontheSharedPreferenceView.
98
4.1.Theempiricalpremiseoffairness-basedarguments:parents’contributiontosocietyOne fairness-basedpro-sharing argumentholds thatparents, byhaving and rearing
children, produce public goods from which everyone benefits. According to these
arguments, the central benefits are thought to be the children’s future economic
contributions to society. Children grow up to be a society's workforce. Their taxes
helpfundretirement,disability,andunemploymentbenefits,aswellasvariousother
social security schemes and public projects.136Note, further, that apart from these
economicbenefits,childrencanalsobe,orcanalsocontributeto,otherkindsofpublic
goods,namelygoodsofanon-materialnature.Childrencanensurethecontinuationof
our culture, ideas, andprojects broadly speaking. The existence of newgenerations
mightevenbeapreconditionofourcapacitytoenjoyourlivesaswedoandtovalue
the things thatwecurrentlyvalue, ifSamuelScheffler is tobebelieved.137But these
benefitsdonotsimplyfallfromthesky.Theyaretheresultofindividuals’decisionsto
havechildrenandraise them incertainways.Havingandrearingchildren is costly,
and it seems only fair that everyone who enjoys these benefits, parents and non-
parents alike, help pay for their production. This type of argument has come to be
knownintheliteratureasthe‘childrenaspublicgoodsargument.’138
AvariationonthisargumenthasbeenproposedbySerenaOlsaretti.Her“socialized
goodsargument”claimsweshouldseetheimportanteconomicbenefitsthatparents
produce as intentionally socialized goods rather than intrinsic public goods.139This
shift invites us to acknowledge thatwelfare states intentionally extend the positive
outcomes of parents' activities to non-parents when they ensure that the new
generations'taxespayforservicesandgoodsthatreachbothparentsandnonparents
136Seee.g.Folbre2002,p.50andGeorge1987,p.31.137Scheffler2013.138Supportersofthisargumentinclude:George1987;Smilansky1995;Folbre2002;Arneson2014;Gheaus2015;Tomlin2015.139Olsaretti2013.
99
indiscriminately.Ourinstitutionscould(andmaypermissibly)bedesigneddifferently,
such thatnonparentswouldonlybenefit fromgoods that they themselves fund (for
example, they could be denied access to public pension schemes and forced to rely
solely onprivate pension funds), as is not currently the case.Non-parents ought to
recognizetheybenefitfromsuchinstitutionaldesignsthat“enlist”parents’effortsfor
the public good, and pay their share in supporting the latter’s procreative and
parentalactivities.
A third version of the fairness-based, pro-sharing argument is available. Richard
Arneson140and Anca Gheaus141independently argue that parents’ contribution to
society consists in discharging an independent, collective duty, namely the duty to
procreate.Thus,havingchildrenwouldnotbemerelyanambition (oranexpensive
taste),butisinfactadutythatwemustallbearthecostsofdischarging.Onthisview
non-parentswouldbeactingunfairlyiftheydidnotsharetheburdensofensuringthe
continuationofthespecies(atleastuptoacertainnumberofchildren).Thisisnotto
saythatweareallunderanobligationtohavechildrenourselvesorhelpwithraising
childrenhands-on,butthatwemayhaveanobligationtohelpparentsmeetthecosts
ofchildrenthroughredistributivetaxationfromnon-parentstoparents.
Whatexactlyaretherelevantbeneficialactivitiesthatparentsengagein,andwhatare
thecoststhatnonparentsareaskedtohelpshoulder?Olsarettiplausiblyarguesthat
the beneficial activities include both procreating as well as rearing children to be
productive, law-abiding citizens. In order to produce the above-mentioned benefits
forsociety,itisnecessarybothtocreatenewpeople(atleastuptoacertainnumber),
as well as to raise them in such a way that they successfully contribute to the
economicandculturallifeofsociety.Shealsoarguesthatthecostswhichnon-parents
are asked to share involve not only the unavoidable costs of raising autonomous,
productive children (costs of care, as she calls them, which include the costs of
140Arneson2014,pp.7-31.141Gheaus2015,pp.87-106.
100
education,nutrition,andsoon),butalsothecostsofaddedadultmemberstosociety,
namelythecostsinvolvedingivingthenewgenerationofadultswhattheyareowed
byjustice.142
4.2.Thenormativepremiseoffairness-basedarguments:FairPlayRegardlessofhowparents'contributiontosocietyisconstrued,allofthesearguments
appeal to a principle of fairness to ground parents' entitlement to have the costs
socialized.ThisprinciplewasoriginallyproposedbyH.L.A.Hart.Hewrites:
...whenanumberofpersons conduct any joint enterprise according to rules and thus
restrict their liberty, those who have submitted to these restrictions when required
have a right to a similar submission from those who have benefitted from their
submission.143
WhatHartcalls“mutualityofrestrictions,”Rawlscallstheprincipleoffairness(orthe
principleoffairplay,whichIwillcallFairPlayforshort),andformulatesitsmainidea
asfollows.
...when a number of persons engage in a just, mutually advantageous, cooperative
venture according to rules and thus restrain their liberty in ways necessary to yield
advantagesforall,thosewhohavesubmittedtotheserestrictionshavearighttosimilar
acquiescenceonthepartofthosewhohavebenefittedfromtheirsubmission.144
If plausible, this principle establishes a special obligation to pay one’s share for
benefitsproducedbyothersincertainconditions.Thatis,itclaimsitiswrongtofree
ride,atleastundersomecircumstances.Thisprinciplewouldpresumablysupplythe
142FormoredetailsseeOlsaretti,2013,pp.229-232.143Hart1955,p.185.144Rawls1999,p.96.
101
fairness-basedargumentswith thenormativebasis theyneed inorder toshowwhy
nonparents ought to pay their share for the benefits that they enjoy as a result of
parents’work.
Fairness-basedargumentshavehadconsiderableappealbothforfeministthinkersas
well as liberals interested in providing an anti-perfectionist case for subsidizing
parents.Ifsuccessful,afairness-basedcasewouldshowthatatypeofworkwhichis
still, in many cases, disproportionally done by women, namely raising children, is
socially valuablework thatwarrants justice entitlements.These arguments are also
attractive for liberalswhofavoranti-perfectionistviewsofpoliticalmoralityas they
do not construe parenting as a valuable view of the good life to be promoted by
politicalinstitutionsthroughtransfersfromnon-parentstoparents.Instead,fairness-
based accounts ground parents’ claims by pointing to the social goods that parents
produce andwhose utility can be recognized regardless of one’s conception of the
goodlife.
Now,criticsofthefairness-basedargumentsforparentaljusticehavecontestedboth
the empirical premise, aswell as thenormativepremiseonwhich these arguments
stand. Somehavenoted that parents can actually contribute to a public bad, rather
than a public good, in a context of (global) overpopulation. Having children, in
particular indevelopedcountries, is themost environmentally impactful choiceone
can make. In a context in which (a) becoming a parent affects the environment
adversely,145(b) particular economies could be sustained or grownby lettingmore
immigrants in rather than incentivizing procreation,146and (c) there are so many
children available for adoption, the claim that parents produce net public goods is
considerablyweakened.
145SeeCasal1999.146SeeMeijers2016.
102
The normative basis of fairness-based arguments has also been questioned. On the
one hand, Fair Play itself has been contested as a plausible principle of special
obligation,inparticularwithintheliteratureonpoliticalobligation.Ontheotherhand,
even assuming Fair Play is a plausible principle in general, there has been
considerable debate around how its conditions of application should best be
understood. And some of its defenders have argued that the circumstances which
make free riding unfair do not obtain in the case of parents and nonparents. For
instance,PaulaCasalandAndrewWilliamshavearguedthatFairPlayonlyappliesto
“non-excludablegoodswhichareproducebycooperativeactivityinwhichindividuals
bearsomecost,whichtheywouldnototherwisebear,inordertoproducethegood.”147
Parentsdonotseemtomeettheseconditionswhentheyhaveandraisechildren,and
thereforeseemtolackafairness-basedcomplaintagainstnon-parents.
Ineffect, theparental justicedebatehasreopened thequestionofwhat, ifanything,
makesFairPlayaplausibleprinciple.Whatexactlyisthewrongnessoffreeridingand
how does this generate an obligation that the free rider pay her share, which
correlatestoarightthatbenefit-producershaveinvirtueofproducingthosebenefits?
In particular, the fairness-based arguments have put pressure on clarifying which
featuresofproducers’beneficialactivitiesgivethemrightsagainstthosewhoreceive
thosebenefits.
InwhatfollowsIputasidetheempiricalpremiseaccordingtowhichparentsmakea
netpositivecontributiontosocietyissecured,andIfocusonthenormativepremise
instead.Itaketheparentaljusticedebateasanopportunitytoclarifythefoundations
oftheFairPlayprinciple,andintherestofthischapterIaimtodefendFairPlayasa
plausible principle of special obligation. This is worth doing for the purpose of
sheddinglightonthequestionofparentaljustice,butalsoforitsownsake,astheFair
Play principle is at the centre of several other important debates in political
philosophy, including environmental justice and political obligation. In section 3.3 I147CasalandWilliams1995,p.106.Emphasisinoriginal.
103
offeracriticalreviewofthemainaccountsofthefairnessprinciple.Iargue,first,that
theliteratureonunfairfreeridinghas,withoneexception,failedtoofferasystematic
account ofwhatmakes free riding unfairwhen it is unfair. Second, I claim that the
existing accounts render the Fair Play principle’s scope of application either over-
inclusiveorunder-inclusive. In section3.4 I review themost systematicaccountof
FairPlaythatexiststodate,namelyGarrettCullity’sviewofunfairnessasafailureof
appropriate impartiality. I argue thatwhileCullity’s account is on the right track, it
fails to provide a conclusive way of distinguishing between unfair free riding and
unproblematic free riding. In sections 3.5 and 3.6 I develop my view of what
characterizesunfairfreeriding.Insections3.7and3.8Ifurtherdrawoutthefeatures
ofmyaccountbyidentifyingtwotypesofconditionsunderwhichfreeridingdoesnot
countasunfair.
4.3.TheFairPlaydebateAs we have seen, the Hart-Rawls principle of fairness claims that one has a moral
obligation to do one’s share in supporting the benefits-producing schemes that one
benefits from. But according to Robert Nozick, foisting benefits on others in the
absenceoftheirpriorconsentcannevergenerateobligationstopaythebenefactors,
for this would implausibly subject us to other people´s will. 148 It is worth
distinguishing this objection, which I call the voluntarist objection, from a closely
relatedonethatisoftenruntogetherwiththefirst.Thesecondkindofworryisthat,
unless Fair Play´s scope of application is properly restricted, it would generate
implausibly numerous obligations. This concern is distinct from the voluntarist one
becauseonemaystillworrythatproliferationofobligationswillrendertheprinciple
implausibleevenifoneisnotmovedbyvoluntaristconcerns.FairPlaytheoristshave
148Nozick1974,pp.90-95.OthercriticsofFairPlayincludeMcDermott2004;Normore2010;Dworkin2011,Zhu2015.
104
beenmotivatedbytheaimofcarvingoutaplausiblescopeofapplicationforFairPlay
mainly in response to the proliferationworry, and some have also been concerned
withthevoluntaristworry.InthissectionIofferabrisk,criticalsurveyoftheattempts
thathavebeenmadeatdefiningtheprinciple´sscope.WhendiscussingtheseviewsI
focus only on those scope-restricting conditions on the principle which I consider
prima facie plausible and which, further, will play some role onmy own Fair Play
view.
Iwillarguethatalloftheseviewsrendertheprincipleeitherover-inclusiveorunder-
inclusive.Toshowthis,IwillappealtowhatIconsidertobestronglyheldcommon-
senseintuitionsaboutwhenfreeridingiswrong.Iassumethattheseintuitionsgive
us some reason to be skeptical of views that are unable to accommodate them,
although this reason isnotdecisive.Of course,decisive reasons foracceptingaFair
PlayviewshouldbegivenbythesortofsystematicanalysisofunfairnessthatIoffer
inSections3and4.Thatsaid, thecapacity tocapturestrong intuitionsaboutwhich
casesoffreeridingareunfairrepresentsanadvantageforanyview,andconversely,
thefailuretodosorepresentsadisadvantage.
OnethingallFairPlaytheoristsagreeuponisthatFairPlay’sscopeofapplicationis
restricted to the receipt of public goods. Classic examples of public goods include
cleanair,theruleoflaw,andpublicdefence.Themostimportantfeaturesharedbyall
of thesegoods iswhatCullitycalls jointnessinsupply: if thegood isavailable toany
member of a group, then it is available to all the other members at no cost to
themselves.Itisthissenseofreceivingagood“forfree,”ratherthan,say,bystealing,
thatconcernsFairPlaytheorists.Publicgoodsusuallyexhibitotherfeaturesthathave
been considered important for Fair Play, such as non-excludability. Goods are non-
excludableifitisimpossibleorexcessivelycostlyforthecontributorstoexcludethird
parties from enjoying them. Jointness in supply often goes together with non-
excludability due to the nature of the good (e.g. clean air). But a good can exhibit
jointness in supplywhile also being excludable. Organizers of a concert in a public
105
parkcould(andmaypermissibly)excludethosewithouttickets.Iftherearenoticket
checksinplace,anyonepassingbycouldenjoytheconcertatnocosttothemselves.149
Oneimportantmatterofcontroversyconcerningtheprinciple’sscopeistherolethat
the benefit-recipients’ attitudes and beliefs regarding those goods should play in
determining their obligations. Some authors, including Rawls, defend voluntarist
accountsofFairPlay,accordingtowhichfreeriderscanbecomeobligatedonlyifthey
voluntarilyacceptthegoodsinquestion,ratherthanregardthemasfoistedonthem
against their will.150By contrast, those who hold receipt-based views of Fair Play
believethatfairnessobligationsmayarisemerelyasaresultofreceivingcertainkinds
ofpublicgoodsincertainconditions.151
VoluntaristaccountsdirectlyaddressNozick’sworries,foronsuchviewsonlygoods
thatpeoplewilltoreceiveinsomerelevantsensecangenerateFairPlayobligations.
However, this strategy seems to render Fair Play under-inclusive. Anyone who
bizarrely thought that crucial goods like physical security or clean air were
disvaluablewouldbeexemptedfromhavingtopaytheirshareeveniftheycontinued
toreceive them.This isparticularlyworrisomegivenpeople’s tendencies todeceive
themselvesintothinkingtheydonotwantcertaingoodswhenthiswouldletthemoff
thehook.152
On the other side of the divide, George Klosko153argues that it is not people’s
subjectiveviews,butthevalueofthebenefitthatplaysthecrucialroleinrestricting
FairPlay’sapplication.Hedistinguishesbetweenpresumptivelybeneficialgoodsand
discretionary goods. Presumptively beneficial goods are indispensable for any
minimallydecent life,andassuch,canpermissiblybepresumedtobebeneficial for
149SeeCullity1995,pp.3-5,32-34foranexhaustivereviewofthefeaturesofpublicgoods,includingthedistinctionbetweenjointnessinsupplyandnon-excludability.150Lyons1965;Simmons1979;Rawls1999a;Simmons2001;Renzo2014.151Arneson1982;Klosko1987,1992;Cullity1995.152SeeRenzo2013foraningeniousresponsetothisissueofself-deception.153Klosko1987,1992.
106
virtually everyone. By contrast, the beneficial character of discretionary goods
dependsonpeople’spreferences.Themerereceiptofpresumptivelybeneficialgoods
isabletogroundfairnessobligations,arguesKlosko,whilethereceiptofdiscretionary
goodscannotgenerateclearobligationsbecausethedisputablevalueofdiscretionary
benefitsoftencannotoverridetheliberalpresumptioninfavorofthelibertytochoose
one´sownobligations.154
AnalternativetoKlosko´sstrategywithinthereceipt-basedcampistosaythatthere
areFairPlayobligations fairly todistribute the costsofdischarginga sharedmoral
duty. Clean air, clean water, public defence are presumptively beneficial goods
(accordingtoKlosko).But,arguably,theyarealsogoodsthatweallshareamoralduty
to provide everyonewith. Sowemight think that, if nothing else, fairness requires
thatwesharethecostsofprovidingmorallyrequiredgoods.155
Both of these strategies have a lot of initial plausibility. They also go some way
towards limiting Fair Play´s scope by restricting its application to crucial benefits,
eitherbecausewecanpresumeeveryonetobebenefitedbythem,orbecausewehave
a shared duty to produce them. Both strategies, however, face a problemof under-
inclusiveness.ForitseemsthatFairPlayobligationsmayalsoarisewhenonereceives
abenefitthatisvaluabletoherevenifitisnotnecessaryforaminimallydecentlife,
andevenifitisnotmorallyrequired.Imagineasharedflatwhereallthreeflatmates
strongly prefer a level of cleanliness that goes beyond what may be considered
presumptivelybeneficialormorallyrequired forhygienereasons(call thiscaseFlat
Share). If twoof themdo theirshareofupholding this levelofcleanliness, it seems,
intuitively,thatthethirdperson’srefusaltodohershareofthe“extra”cleaningwould
makeheranunfairfreerider.
154Klosko1992,pp.39-44.155Casal1999;Arneson2014;Gheaus2015.
107
Ofcourse,thereceipt-basedstrategiesabovecanbereadasprovidingonlysufficient
conditions forFairPlaytoapply.Sotheymightnotbebotheredbytheir inability to
condemnfreeriding incases likeFlatShare,beingopentothepossibility thatsome
othersufficientsetofFairPlayconditionsmightapplytosuchcases.ButItakeitthat
providing a set of necessary and sufficient conditions for Fair Play, able to capture
cases of morally required/presumptively beneficial goods, as well as cases of
desirablebutnotmorally requiredgoods (like, arguably, public libraries), andeven
trivial but subjectively valued goods (like in Flat Share), would provide a more
complete(andhence,moreattractive)defenceofFairPlay.ThisiswhatIaimtodoin
Sections3and4.
Finally, a different set of conditions thought to help limit Fair Play’s scope of
applicationregards theproductionof thebenefits.AllFairPlaydefendersagree that
contributorsmustmeetsome“burdensomeproduction”condition.Thatis,onlythose
whoproducebenefitsbyincurringasacrificehaveFairPlayclaimsagainstfreeriders.
Most commentators agree, further, on an “intentionality” condition: that claims of
fairness may only arise for benefits that the contributors brought about
intentionally.156
The trouble, as others have noted,157is that none of these conditions have been
adequatelyexplainedordefended. It isunclearwhatconstitutesa “sacrifice” for the
purposesofFairPlay,andwhatmakesitnecessaryforgroundingcontributors’claims.
Similarly,theneedforcontributors’havingcertainintentions,andwhatexactlythese
shouldbe,alsoremainsunderexplored.
OneimportantreasonwhysuchambiguitiesstillplaguetheFairPlayliteratureisthat
hardlyanyattemptshavebeenmadeatapproachingFairPlayinasystematicmanner.
Beforeattemptingtorectifythis,letmefirstturntothe“bigpicture”notionoffairness
156Simmons1979,p.336;Casal&Williams1195,p.106;Casal1999,p.368.157Boran2006;Olsaretti2013;Gheaus2015.
108
thatIwilldrawon,anddiscussinsomedetailtheaccountofFairPlaythathasbeen
thoughttobesupportedbyit.
4.4.UnfairfreeridingasafailureofimpartialityGarrettCullityistheonlyFairPlaysupporterwhohassoughttoofferageneralview
aboutwhatunfairnessconsistsinandhowthefreeridingconditionshepicksoutas
wrong-makingcontributetomakingfreeridingunfair.158
At the most abstract level, I join him in thinking that fairness is a matter of
appropriate impartiality.159Fairness requires thatwe treat peoplewith a particular
formof impartiality, suited to the context at hand. In the context of a 100m sprint,
fairnessasimpartialityrequiresthatcompetitorssharethesamestartingpoint.Inthe
context of a game, it requires that everyoneplay by the same rules. It isunfair if a
contenderstartstheracefromanadvancedposition,andifsomepeopledonotabide
by the same rules. Free riding, when unfair, is also an instance of a failure of
impartiality. Thewrongful formof partiality that theunfair free rider is guilty of is
allowing herself a privilege, or unjustifiablymaking an exception of herself, by not
payinghershareforthebenefitsshereceives.160
Butwhendoesthefailuretopayone’sshareactuallyamounttomakinganexception
ofoneself?InthissectionIbrieflylayoutCullity´sanswer.Idonotsubscribetoallof
hisclaims,butIwillnot,here,engagecriticallywiththem.Instead,mystrategyisto
offeraninternalcritiquetohisaccount,havingacceptedallofhisclaimsarguendo.
158Cullity1995,2008.Sometheoristsdosaymoreaboutthesortofwrongnesstheybelievefreeridinginvolves.Arneson1982arguesthatitcanbeunderstoodasabreachoftheproducers’entitlementtothefruitsoftheirlabourwhentheycannotexcludethirdpartiesfromenjoyingthem.Tosi2018believesthatFairPlayisnormativelysimilartotheprincipleofconsent.However,neitherexplainswhyaviolationoftheproducers’entitlement,orofthetermsofacontract,amountstoactingunfairly.159Cullity2008,pp.2-5.160Cullity1995,pp.22-23.
109
Cullity proposes three conditions under which free riding counts as unjustifiably
makinganexceptionofyourself.Twoofthemarefairlyuncontroversial:first,thatthe
free ridersmust receive a net benefit from the benefits-producing scheme they are
askedtoparticipatein,andsecond,thattheydonotraiselegitimatemoralobjections
againstit161
The third condition, the “fair generalization requirement,” states that the refusal to
payforabenefitonereceivesisunfaironlyif it istruethatthepracticeofhonoring
the sorts of demands for payment beingmade in the case at hand, and in all other
similar cases that might arise, would make virtually everyone worse off.162This
requirement captures the thought that a plausible Fair Play principle should not
generate unreasonable demands. A good hypothetical test for whether a particular
demandforpaymentisunreasonableistoaskwhatwouldhappenifwehadtocomply
withthatkindofdemandnotjustinthisparticularcase,butinallthecasesthatmight
potentiallyarise,wherecontributorsmightmakesimilardemandsonus.Ifhavingto
complywith all the further similar demands thatmight arisewouldmake virtually
everyone worse off, Cullity argues, then the particular demands being tested are
unreasonable,andrefusingtocomplywiththemisnotunfair.
For illustration, Cullity offers the Entreprising Elves case,where a band of elves go
aroundrepairingtheshoesofpeopleunwittinglyleavingthemoutovernight,andthen
attempttochargeareasonableprice.163Cullityarguesthatrefusingtopaytheelvesis
not unfair because their demand for payment does not pass the fair generalization
test. Ifwerecognizedas legitimateandcompliedwiththeirdemand,thenwewould
alsohavetohonor furtherrequeststhatwepay“forallunsolicitedbenefits thatare
worth their cost,” and this would be an inefficient commercial system that would
makevirtuallyeveryoneworseoff.164Variousmechanismswouldmake this so.This
161Cullity1995,pp.18-19.162Cullity1995,pp.14-15.163Cullity1995,p.10.164Cullity1995,p.14.
110
system would presumably force us to pay for benefits that were unintended and
unforeseenpositiveexternalitiesofothers´activities.Itwouldalsoforceustopayfor
benefits thatothers foisteduponuswith the solepurposeof gettingpaid in return.
Even though each individual benefitmight beworth its cost, overall such a system
wouldleadtoaharmfulproliferationofobligations.
Cullityjustifieshisthreeconditionsforunfairfreeridingbyreferencetohisnotionof
appropriateimpartiality.Inthecaseofthefairgeneralizationrequirement,refusingto
comply with unreasonable demands like the elves’ does not amount to making an
exception of oneself, for if their demands fail the fair generalization requirement,
everyone has good reason to refuse to comply, including those who are currently
complyingwiththem.
Cullity’s view is meant to offer a Fair Play principle that is not only theoretically
supported by a general view of fairness, but that also yields a plausible scope of
application. It is intendedtodistinguishcasesof“predatorydemands” liketheElves
from cases of unfair free riding like Cullity’s Recalcitrant Fisherman. In this case,
fishermen band together to reduce pollution in the lake they rely on for their
livelihood.Onefishermanproteststhathehasnotaskedforthisbenefit,andhasno
intention to participate in the effort.165The recalcitrant fisherman is an unfair free
rider, according to Cullity, because the demand made on him by the fishermen
(making“afairlyassessedsacrifice”soas“topreservethelivelihoodofall”),ifallowed
togeneralize,wouldnotmakevirtuallyeveryoneworseoff.166
Cullity’sviewisontherighttrack,especiallyinitsattempttolimitourvulnerabilityto
unreasonableFairPlaydemands.However,thefairgeneralizationrequirementistoo
vaguetobeabletodistinguishcasesofunproblematicfreeridingfromcasesofunfair
free riding. Whether free riding is unfair turns on how exactly we describe the
165Cullity1995,p.11.166Cullity1995,p.15.
111
demandsforpaymentthatshouldberunthroughthefairgeneralizationtest.Yetwe
havenoguidanceastohowtodeterminewhatthesalientfeaturesofthedemandsfor
paymentarethatshouldbetested.Toseetheproblem,considerthiscase.
TheEnterprisingScientists
Everyoneinourtownrunstheriskofcontractingamildchronic illnessbecausethe
watersupplysourceistaintedwithadangerouschemical.Oneday,Iamdelightedto
find that thewater supply is safe todrink thanks to a groupof scientistswhohave
passedthroughourtownandimplementedawater-purifyingmechanismovernight.I
amlessdelightedwhenIreceivethe(reasonablypriced)bill.
Here, it isnotclearwhatthesalient featuresof thescientists’demandsforpayment
are.WemightthinkthattheirdemandsareofthesamesortasthosemadebyElves:
“holding everyone liable to pay for all the unsolicited benefits that areworth their
cost.”Thescientistsseemtobeengaging inthesamesortof“business”astheelves.
They have identified some needs, andmet themwithout the beneficiaries’ consent,
beforeattemptingtochargeareasonableprice.Generalizedcompliancewiththissort
of demand, we could argue, would lead to a commercial system that would be
unreasonablyburdensome forvirtually everyone, and thatwould, therefore, fail the
fairgeneralizationtest.
Wemightthink,instead,thatthedemandsmadebythescientistsaremoresimilarto
those made by the fishermen. The relevant features of the scientists’ demands for
payment could be described as follows: the town’s inhabitants are asked to help
preserveeveryone’shealthbycontributingtopayingthescientistsfortheireffort. If
thisdescriptionisright,theirschemepassesthetestbecauseparticipatinginthis,and
any other potential, future efforts to preserve everyone’s health would not make
virtuallyeveryoneworseoff.
112
Indeed, it might be said on Cullity’s behalf that since the good promoted by the
scientistsiswhatKloskocallspresumptivelybeneficial,thisisclearlyasalientfeature
inEnterprisingScientists,sothelatter’sdemandsforpaymentarereasonableindeed,
as it would make everyone better off to pay their share for securing presumptive
goods.
This is true.However, thepointwasneverthat thescientists’demands forpayment
did not share salient featureswith the fishermen’s demands. The point is that they
also share very important features with the elves’ demands for payment. The
scientistsarenotengaged,likethefishermen,inabeneficialcooperativeeffortwhich
thirdpartiesareaskedto join.Rather, theyare implementinganewtechnology ina
town they arepassing through, seeking to get paid. In this respect,what theydo is
similar to the elves’waiting for people to leave their shoes out. Thus, some salient
features of the scientists’ endeavourmake it appropriate to describe it as a sort of
imposedcommercialtransactionthat,ifallowedtogeneralize,wouldarguablyleadto
aninefficientcommercialsystemthatwouldmakevirtuallyeveryoneworseoff.
Cullity’saccountthusremainsinconclusivewithrespecttooneofhismostimportant
goals:todistinguishbetweenunfairfreeridingand“predatorydemands”forpayment.
Therewillbemanycases liketheEnterprisingScientists inwhich it isunclearwhat
the salient features of the demands are, or in which opposing views of the salient
featuresseemdefensible.
TheFairPlay account I propose in the following section takesonboard the coreof
Cullity’s view, but is better equipped to carve out a clear and reasonable scope of
applicationforFairPlay.
113
4.5.TheSharedPreferenceViewofFairPlayRecall that we are looking for a principle of fairness that condemns free riding as
unfair when it is a breach of impartiality, and in particular, when it involves
unjustifiablymakinganexceptionofoneself.Weneedacriteriontodeterminewhen
free riding amounts tomaking an exception of oneself that applies across different
kindsofgoods,includingdiscretionarygoodslikeinFlatShare,butwithoutleavingus
vulnerabletopredatorydemandsliketheEnterprisingElves’.
As a methodological point, I assume that in order to determine when free riding
amountstomakinganunjustifiedexceptionofoneselfweneedtospellouttheways
inwhich the free riders and the contributors are relevantly similarly situated. The
thought is thatwhen the free riders are similarly situated to the contributorswith
respect to all the relevant facts regarding the receipt and production of a benefit
except for the free riders’ failure to pay their share, their failure to pay cannot be
justified on impartial grounds. I argue that the free riders and the contributors are
similarlysituatedwhenwecanjustifiablyascribetobothaqualifiedpreferenceforfree
riding.IcallthisaccounttheSharedPreferenceView.
There are, I arguebelow, two suchpreferences for free riding thatwe can identify,
that correspond to twodistinct (but structurally similar) opportunities to free ride,
depending onwhether the goods in question are “optional goods” (i.e. notmorally
required)ormorallyrequiredgoods(bywhichImeanthatwehaveamoraldutyto
provide people with). The Shared Preference View distinguishes between the
opportunityto freerideby internalizingabenefit thatothershaveproduced,onthe
onehand,andtheopportunitytofreeridebylettingothersdischargeacollectiveduty
thatappliestoone,ontheotherhand.Thefirstkindofopportunitymayarisewhen
eitherkindofgoodisbeingproduced,whilethelatteronlyarisesincasesofmorally
requiredgoods. InthissectionIdeveloptheSharedPreferenceViewas itappliesto
optionalgoods,leavingmorallyrequiredgoodsforthenextsection.
114
I startbypositing that theonlynecessary requirement relating to thenatureof the
optional goods themselves is that they exhibit jointness in supply (which, recall,
meansthatifsomeoneinthegroupreceivesthem,everyoneelsecandosoatnocost
to themselves). To signal this minimalist requirement, I will call such goods
“collective”fromnowon,ratherthanpublicgoods,todistinguishmyviewfromthose
onwhichother typical featuresofpublicgoodsarenecessary forFairPlay toapply,
suchasnon-excludability.167
I propose thatwhen a groupof people receive a collective goodproducedby some
membersofthatgroup,FairPlayobligationsariseonlybetweenthoseamongthefree
riders,andthoseamongthecontributors,towhomtheFreeRider’sPreferencecanbe
ascribed.
TheFreeRider’sPreference (FRP): I prefer that otherspay for this valuable collective
goodthatIcanenjoyforfreeandforwhichIwouldbepreparedtopay,intheconditions
underwhichitisoffered,ifIhadto.
As is familiar from the extensive literature on Prisoner’s Dilemmas and collective
actionproblems (Olson1965,Parfit1984,Gauthier1986),168in contextswhere free
riding ispossible, eachperson faces the followingpotential scenarios,ordered from
theonethatmakesherbestofftotheonethatmakesherworstoff.
1. Ireceivethebenefitwithoutbearingthebenefits-producingcosts.
2. IreceivethebenefitandIbearthebenefits-producingcosts.
3. IdonotreceivethebenefitandIdonotbearthebenefits-producingcosts.
4. Idonotreceivethebenefitdespitehavingbornethebenefits-producingcosts.
167IherefollowCullity,incontrasttoalltheotherFairPlaydefenders,whotreatnon-excludabilityasnecessary.Seee.g.Arneson1982;Klosko1992.Recallthat,asmentionedabove,jointnessinsupplydoesnotnecessarilyentailnon-excludability.168Olson1965;Parfit1984;Gauthier1986.
115
One counts as sharing the FRP when one can be ascribed the two top-ranking
preferencesinthatorder.169Andwecanascribepeoplethisrankingofpreferenceson
thebasis of twoelements: (a) their having a certainpro-attitude to the receipt and
paymentofthecollectivegood,and(b)anassumptionaboutthekindofbeingsweare
thatwemayapplytoeveryoneforfairnesspurposes,namelythatweareprudentially
rational.
Toelaborate: the first condition, (a), thatmustobtain inorder to justifiably ascribe
theFRPtosomeoneisthattheybelievethegoodtheyarereceivingisworthitscostto
them.Thisconditionobtainswhenthebenefit-recipientssubjectivelyvaluethegood
enoughthattheywouldprefertopayitsattendantcost if theyhadto(that is, iffree
ridingwasnotaneffectiveoption), rather than forgo the goodaltogether.When this
condition obtains, the two scenarios inwhich the good is produced, eitherwith or
withoutone’sowncontribution,rankhigherthanthetwoscenariosinwhichthegood
isnotproducedatall.
Nownote that, once the first condition ismet, thequestion iswhichof the top two
scenarios ranks first.On the assumption that (b) spells out, namely that people are
prudentially rational, the scenario in which one receives the valued good for free,
ratherthanhavingtopayforit,wouldcomefirst.Thenotionofprudentialrationality
atworkhereisthefamiliartechnicalnotionofanindividualwhowouldactsoasto
maximizeherownwell-being.Receivingforfreeagoodthatwouldbeworthitscost
totheindividualisprudentiallybetterthanpayingforit,allelseequal.Foroneisthen
free to direct one’s resources to other desirable ends, without losing out on the
benefit.Therefore,freeridingontheproductionofagoodthatisworthitscostwould
be,inprinciple,thetop-rankingoptionforanyprudentiallyrationalindividual.
169Theascriptionofpreferences1and2inthatorderissufficientformyFairPlaypurposeshere.Imentionpreferences3and4forthesakeofcompleteness,asIamborrowingtheentiresetofoptionsfrompreviousworkoncollectiveactionproblems.
116
For illustration, consider the paradigmatic case of unfair free riding, that of, quite
literally, riding free on public transport. It is possible to ascribe the FRP both to
someonewhobuysaticket,andtosomeonewhodoesnot,providedonlythatbothin
factbelievethebenefit isworthpayingthisprice,andassumingitwouldbetotheir
advantagetogetthisgoodfor freeratherthanpayfor it.Thebestscenario forboth
parties,then,wouldbetheonewheretheyenjoythebenefitforfree.Thesecondbest
scenarioforeachistobenefitfromthepublictransportandpay,sincethebenefitof
publictransportoutweighsthepriceoftheticketbytheirownlights.Andsoondown
therankingofpreferences.
Theimportantpointworthstressinghereisthattherelevantrankingofpreferences
applies to the fare-payers regardless of the fact that theydonot actuallyact in line
with the top-ranking option (i.e., the free riding option), like the free riders do.
SomeonemightwonderhowitispossibletoascribetheFRPtothefare-payers.Ifyou
payyourfare,doesn’tthisshowyoudonotactuallyprefertofreeride?Inresponse,it
isimportanttostressthattheFreeRider’s“preference”itselfisjustatermofart.Its
ascriptionisjustifiedonlypartiallybyaperson’sactuallyheld,subjectivepreference,
namely,thatidentifiedbycondition(a),thatpeopledeemthebenefittobeworththe
cost; and partly by an objective judgment about what is in people’s prudential
interest.
So,somepeopleactonthepreferenceidentifiedby(a),bybuyingaticketevenwhen
they could ride for free (when there are no ticket checks and enough others are
buyingticketsanyway).Others,bycontrast,choosetofreeridewhenthisispossible,
even though theywould pay their fare if ticket controlswere inplace and/or if the
publictransportsystemwasindangeroffailingduetolackofcontributions(sothey
alsoshare(a)but theydonotact in linewith it).Despite theirdifferentactions, the
FRP couldbe ascribed tobothparties because it is trueof each that itwouldbe to
theiradvantage,asprudentiallyrationalpeople,forthem tobetheoneswhogetthe
valuedbenefitforfree.ThesimilaritiesthatmatterfortheSharedPreferencevieware
117
(a)thatbothpartiesactuallyvaluethegood,and(b)thatitistruethatbothwouldbe
betteroffiftheyenjoyedthisgoodforfree.Takentogether,thesetwoelementsallow
ustoascribetheFRP.
Still, onemight insist, if the fare-payers would be better off free riding, why don’t
they? There is a host of reasonswhy fare-payersmight decide to buy their tickets.
Theymightbefollowingthelawsandconventionsthatstipulatethateachshouldpay
for a ticket, or they might be more risk-averse than the fare evaders about the
possibilityofgettingcaughtwithoutaticket.Ortheymightrecognizetheunfairnessof
freeridingandpaytheirfareforthisreason,asmanysurelydo.Buttorepeat:noneof
thesereasonschallengethefactthattheycanbeascribedtheFRP.Ridingthepublic
transportforfreewouldbeineachcontributor´sbestinterest,justlikeitisinthebest
interest of the free riders. In section 6 I discuss cases inwhich some reasonswhy
peoplebeartherelevantcostsdounderminetheascriptiontothemoftheFRP.
Wecannowsingleoutwhatitmeansforfreeriderstomakeanunjustifiedexception
of themselves. Among all those who receive a collective good, those who count as
relevantlysimilarlysituatedforfairnesspurposesarethosetowhomwecanjustifiably
ascribetheFRP.WhensomeofthosewhosharetheFRPneverthelessbearthecostsof
securing the collective good, those who allow themselves to free ride arrogate to
themselves an unjustifiable privilege. They get to act in line with the FRP at the
expenseofthosewho,despitesharingit,actinfrustrationofit.
The Shared Preference View allows us to distinguish cases of unfair free riding on
optional goods like Flat Share from cases of “predatory demands” like Entreprising
Elves.TheflatmatesinFlatSharecanallbeascribedtheFRP:eachofthemplaceshigh
value on enjoying a spotless flat, and eachwould be better off if the others did the
workinsteadofthem.Soifoneofthemrefusedtodohershareoftheextracleaning,
she would be acting unfairly. By contrast, the elves in Entreprising Elves are not
similarly situated to those who benefit from their services because the elves
118
themselvesdonothavean interest inenjoying the samegoodsas thebeneficiaries.
TheycannotbeascribedtheFRP,which, torepeat,states the following.Ipreferthat
others pay for this valuable collective good that I can enjoy for free and for which I
wouldbeprepared topay, in the conditionsunderwhich it is offered, if I had to. The
elves are not interested in the benefit of having their shoes repaired. Unlike their
clients,whattheyseemtohaveaninterestinismakingaprofit,orgettingrecognition,
orsomethingofthatsort.
Someonemightwonderwhyitshouldmatter,forthepurposeofestablishingwhether
theElvesoughttogetpaid,thattheydonothaveapreferenceforenjoyingthisgood
themselves.TheSharedPreferenceViewconceivesof therelevantunfair freeriding
situationasonewhere individualsareata “standstill.”Everyonewants toenjoy the
same collective good, and each would be better off by not contributing to its
production.Whensomepeoplecontribute,therebybreakingthisdeadlock,theShared
PreferenceView condemns thosewho fail to do their share in return. By contrast,
EnterprisingElvesseemstobeasimplecommercial transactionbetweenthosewho
actexclusivelyasproducers,andotherswhooccupytheroleofbeneficiaries.Butsuch
transactions are arguably more appropriately governed by explicit consent to
receivingandpayingforabenefit,notbyFairPlay.FairPlayapplies,Ihaveargued,to
situations that are importantly different from typical transactions: those where
individualsarerelevantlysimilarlysituatedinthattheyshareaninterestinbenefiting
fromacollectivegoodthemselves,andaninterestinnotbearingthecoststhemselves.
Itisinlightofthissymmetrythatfailingtopayone´sshareamountstounjustifiably
makinganexceptionofoneself.
119
4.6.TheSharedPreferenceViewandmorallyrequiredgoodsOn the Shared Preference View just sketched, fairness obligations are generated
between those contributors and those free riderswho share the FRP. Andwehave
seen thatwhether or not a free rider can be attributed the FRP partly depends on
whethershesubjectivelyvaluesthebenefitenoughtoprefertoreceiveandpayforit
rather than do neither, which goes some way to meeting Nozick´s voluntarist
objection.
At thispoint, however, someonemight thinkweare giving toomuchweight to free
riders´ subjectiveviews.Thinkback toCullity´sRecalcitrantFisherman. In thiscase,
fishermenbandtogethertocleanupthepollutedlaketheyrelyonfortheirlivelihood.
Theone fishermanwhoprotests that hehasnot asked for this benefit andwill not
participate in the common effort seems very unreasonable indeed. If people claim
theywouldratherdowithoutcrucialgoodslikeacleanlakeinthiscase,shouldthey
beexemptedfrompayingtheirshareforthesegoods?170
Theanswerisno.Thisisbecausethegoodofanunpollutedlakeisarguablyamorally
requiredoneinthiscase,sinceitisstipulatedthateveryone´slivelihooddependson
beingabletofishthere.Asmentionedintheprevioussection,whensuchgoodsareat
stake,thereexistnotone,buttwokindsofopportunitiestofreerideunfairly.Oneis
found in cases of optional collective goods as well, namely the opportunity to
internalizeavaluedgoodthatothershaveproduced.Thesecondarisesonlyincases
ofmorallyrequiredcollectivegoods.Itistheopportunitytohaveacollectivedutythat
applies to you (alongside others) be discharged only by others. Where morally
requiredgoodsareproduced, then, it ispossible tobeguiltyofboth formsofunfair
freeriding(forexample,iftherecalcitrantfishermansecretlysubjectivelyvaluedthe
benefit of a clean lake he would be both internalizing a valued good for free, and
170Thisisanimportantworryfacinganyvoluntaristview.Seee.g.Arneson1982,p.632directingthisobjectiontoSimmons´sview.
120
allowingothers todischargea collectiveduty thatapplies tohim too).But it is also
possible to be guilty only of the latter sort of unfair free riding, as the recalcitrant
fishermanarguablyisifhetrulybelievesthathavingacleanlakeisdisvaluableornot
worth the cost. And in this case, as we shall soon see, his subjective views do not
exempthimfromhavingFairPlayobligations.IntherestofthissectionIexplainhow
theSharedPreferenceViewcanmakesenseofthissecondkindofopportunitytofree
ride.
Let me start by clarifying what is meant by the collective duty to produce certain
goods.Thinkofgoodslikesafedrinkingwater.Mostofuswouldagreethattheyare
morallyrequiredinthesensethatthereisamoraldutytoseethatpeoplehaveaccess
to them, and that this duty is a general, agent-neutral duty that, in political
communitieslikeours,usuallyfallsonthestateasacollectiveagenttodischarge.No
particularcitizenhasanymoreresponsibility thananyother todischarge it,allelse
equal.Moreover,itisadutyforwhichitisusuallynotpracticallyormorallyrequired
thateverymemberofourcommunityactsoastodischargeit,asitcanbeeffectively
dischargedbyonlysomepeople´sbearingtherelevantcosts.
Iproposethat,outofagroupofpeoplewhoareallunderacollectivedutytoproduce
acertaingood,FairPlayobligationsarisebetweenthosewhoactsoastodischargeit
andthosewhodonot,andwhocanbesaidtosharetheFreeRider’sCollectiveDuty
Preference.
The Free Rider’s Collective Duty Preference (CDP): I prefer that others discharge the
collectivedutythatalsoappliestome.
Think of Recalcitrant Fisherman. The duty collectively to avoid dangerous lake
pollution canbedischargedeven if some fishermendonotparticipate in the effort.
Thisgivesrisetoahierarchyofpreferencesthatthefreeridersandthecontributors
share.
121
1. Thecollectivedutyisdischargedwithoutmybearingthecosts.
2. ThecollectivedutyisdischargedandIbearthecosts.
3. ThecollectivedutyisnotdischargedandIdonotbearthecosts.
4. Thecollectivedutyisnotdischargeddespitemyhavingbornethecosts.
One can be ascribed the CDP when one can be said to share the two top-ranking
preferences in thatorder.Andwecanascribe thisrankingofpreferences toanyone
who falls under the scope of a collective duty, on the basis of the following two
assumptionsaboutthekindsofbeingsweare:(a*)thatweareappropriatesubjectsof
moralevaluation,and(b*)thatweareprudentiallyrational.
Condition (a*) assumes thatwe are appropriate subjects ofmoral evaluation in the
sense thatwehave amoral interest in seeing themoral duties andobligations that
apply to us discharged. Assumingwe can conceive of individuals thisway, the two
scenariosinwhichthecollectivedutyisfulfilled,withorwithoutourcontribution,are
superior,fromthepointofviewofmorality,totheonesinwhichitisnot.
Condition(b*)assumesweareprudentiallyrational.Asbefore,thisisjusttosaythat,
forfairnesspurposes,wecanconceiveofourselvesasthekindsofbeingswhowould
actsoastomaximizeourwell-being.Thismakesisthecasethat,outofthetoptwo
scenarios, the one in which the collective duty is discharged without our own
contribution is superior, from the point of view of prudence, to the one where we
contribute;then,wearefreetoinvestourresourceselsewherewhilestillseeingthe
demandsofmoralitythatapplytoussatisfied.
Whoexactlytheobligation-boundfreeridersaredependsontheparticularcollective
dutyatstake.Theparticularitiesofhowwepickouttheindividualsonwhomtheduty
falls in the first place are important, but this can be left aside for now. Once the
appropriatescopeofthecollectivedutyhasbeenestablished,freeridersfallingunder
itcanbeascribedtheCDPautomatically.FortheCDP,liketheFRP,isnotanactually
122
heldpreference.UnliketheFRP,theascriptionoftheCDPtofreeridersdoesnotrely
on any subjective views held by them, not even in part. Forwe ought not to allow
people’ssubjectivepreferencestoplayanyroleindeterminingthegeneraldutiesthey
areheld to.From thepointofviewofmorality, it isbetter foreachperson that the
dutythatappliestothembedischargedratherthannot.Andfromthepointofviewof
prudence,itisbetterforeachthatothersdischargeitratherthanthem.Forthelatter
judgmentweneednot conceiveof people aswanting to shirk their responsibilities.
The collective duty simply requires that a particular outcome be secured, but says
nothing about who exactly should secure it, so there is no obvious wrongdoing in
failingtocontributeiftheoutcomecouldwellbesecuredbyothers.
Thisview,then,allowsustocondemntherecalcitrantfishermandespitehisnegative
appraisalofthevalueofhavinganunpollutedlake.Hissubjectiveviewsmayprevent
usfromascribingtheFRPtohim,forweoughttoallowpeoplethefreedomtodecide
forthemselveswhichgoodstheyconsiderbeneficial,andwhichbenefitstheywishto
receiveandpayfor.However,weoughtnottoallowpeople’ssubjectiveviewstoaffect
thegeneraldutiestheyareheldto.So,assumingtheCDPcanbeascribedtoboththe
recalcitrantfishermanandtotheotherfishermen,heisboundbyFairPlaytodohis
partindischargingthecollectivedutyofavoidingdangerouslakepollution.
Thus, the Shared Preference View has the advantage of capturing all the cases of
presumptively beneficial goods that Klosko is interested in (as I take it that
presumptively beneficial goods are also morally required) without relying on the
paternalisticclaimthattheyarebeneficialevenforthosewhosincerelybelievethey
arenotandwouldbepreparedtoforgothem.ForontheSharedPreferenceViewwe
canfallbackontheopportunitytounfairlyfreerideconstitutedbythefailuretoshare
thecostsofdischargingasharedduty.Todenouncethissortofunfairfreeridingthere
isnoneedtoestablishthatthefreeridersthemselvesarebenefitedbythegoodsbeing
produced.Allweneedtoestablishisthatthereisacollectivedutytoproduceagood
123
(for the benefit of others, if not also for one’s own benefit), and that the relevant
partiessharetheCDP.
Thisview, then,canalsoserve togroundpoliticalobligation,as theorists likeRawls
and Klosko have argued, since governments normally provide a range of morally
requiredgoodslikepublicdefenceortheruleoflaw.Eventhosewhodenythevalue
(for themselves) of suchbenefitsmaybe enjoined to pay their share if they canbe
attributedtheCDP.Itisworthmentioningthatinrealitytheobligationtopayforthe
morally required goods provided by the state is often over-determined.More often
thannot, thosewho fallundera collectiveduty toparticipate in securing themalso
considerthemselvesbenefitedbythemandcanbesaidtosharetheFRPinadditionto
theCDP.However,ifstatesprovidegoodsthataredesirablebutnotmorallyrequired
(like, perhaps, public libraries), they cannot appeal to Fair Play reasons to extract
taxes from thosewho, for example, sincerely believe themselves to beworse off by
having this benefit available to them and having to pay for it compared to doing
neither.
As further illustrationof theSharedPreferenceView,considerhow itapplies to the
case of Entreprising Scientists I mentioned earlier, in which a group of scientists
passingthroughtownimplementawater-purifyingmechanismthatsavesthetown´s
inhabitants fromriskofamildchronic illness,andthenseektochargeareasonable
price for their service. I argued that Cullity’s view of unfair free riding would be
unable to generate a clear answer in this case because it is not clear, on his view,
whichfeaturesofthissituationshouldbesalientforfairnesspurposes.OntheShared
PreferenceView,onefeature,inparticular,makesallthedifference:thefactthatthe
scientists areprovidingamorally requiredgood. If theyhadnot implemented their
technology, the townspeoplewould have beenmorally required to find some other
way to ensure water safety. Here the question that I bracketed earlier becomes
important,namelywhocanbesaidtofallundertherelevantcollectivedutyinthefirst
place.Providedthatthescientistsfallunderthedutyaswell,theCDPcanbeascribed
124
to both the scientists and the beneficiaries of their technology. And since, by
hypothesis,thescientistschargeafairpricefortheireffort,thetownspeoplehaveFair
Playobligationstocompensate thescientists forhavingdischargedacollectiveduty
forthem.
Notethateventhoughthescientistsdonothaveaninterestinenjoyingthebenefitof
safewaterthemselves(becausetheyareonlypassingthrough),herethisfactdoesnot
undermine their Fair Play claims, like it did for the Entreprising Elves. For what
matters here is whether there is a collective duty that both the producers (the
scientists)andthefreeriders(thetownspeople)canbesaidtofallunder.Assuming
thatthereis,FairPlaykicksin.
To take stock, I have provided a framework for identifying the morally relevant
similaritiesbetweenthosewhodotheirpart insecuringanoptionalcollectivegood,
orindischargingacollectiveduty,andthosewhodonot,suchthatthefailureofthe
freeriderstodotheirpartamountstomakinganunjustifiedexceptionofthemselves.
Out of all the individuals of a group that receives a collective good, or that sees a
collectivedutysatisfied,only those individualswhoshare theFRP,or theCDP,have
fairnessobligationstocontributetotheproductionofthegood.Whatthetwocontexts
haveincommonisthatnoteveryonewhocanbeascribedtherelevantpreferencecan
satisfyitcompatiblywitheveryoneelse’sdoingthesame.Ifsomepeopleactinaway
thatfrustratestheirFRPortheirCDPtherebyenablingothers(thefreeriders)toact
in line with theirs, then free riders ought to similarly frustrate that preference.
Otherwise, they count as allowing themselves an unjustifiable privilege. In the next
twosections,IshowhowtheSharedPreferenceViewalsoenablesustoidentifythose
typesofcaseswhere,thoughthereisfreeriding,itisnotwrong.
125
4.7.BenefitsrecipientswholackFairPlayobligations
Intheprevioussectionwehaveseenthatfreeriders´subjectiveviewsplaynorolein
whetherwe can ascribe to them the CDP. If others discharge a collective duty that
appliestothemtoo,itdoesnotmatterwhethertheybelievetheystandtogainfrom
thegoodsbeingproducedorwhethertheyseethevalueofthedutybeingdischarged.
Theywillbeboundtodotheirshareregardless.
ButnowconsidertheSharedPreferenceViewasitappliestotheprovisionofoptional
collectivegoods.WemaybeunabletoascribetheFRPtosomefreeridersbecause(i)
thegoodisdisvaluabletothem;(ii)thepricetheywouldhavetopayforitexceedsthe
benefit;(iii)theymighthavealternativesourcesforenjoyingthesamebenefit,which
they prefer; (iv) they prefer to forgo this benefit altogether in order to invest their
resourcesintootherprojects;finally,(v)theyraiselegitimatemoralobjectionstohow
the good is produced. Those of whom any of (i) through (iv) is truewould have a
complaintagainstbeingheldtoaFairPlayobligation.
These casesof seemingly innocent free ridinghaveoftenbeenacknowledged in the
literature in some form or another.171The Shared Preference View offers a unified
explanationforwhytheyareallcasesinwhichfreeridingisnotwrong.Fortheyare
all cases inwhich the free riders’hierarchyofpreferences isdifferent from theone
sharedbythecontributorsandtheunfairfreeriders.
Recall the top-ranking options in the hierarchy of preferences shared by the
contributorsandtheunfairfreeriders.
1. Ireceivethebenefitwithoutbearingthebenefits-producingcosts.
2. IreceivethebenefitandIbearthebenefits-producingcosts.
171Arneson1982,pp.620-621;Simmons2001,p.20.
126
If free riders sincerely raise complaint (i) or (v), their top-ranking preference
presumablyis:
1. IdonotreceivethebenefitandIdonotbearthebenefits-producingcosts.
Ifmyflatmatessubjectmeto loud jazzmusic,andIhate jazz, Iammadeworseoff.
Alternatively,ifagoodisprovidedthroughsomefromofinjustice(say,itsproduction
infringes on some people’s rights), I may not want to receive it. In either of these
cases, I cannotbeascribed theFRPbecausemy toppreference isnot to receive the
goodatall.172
If the free riders sincerely raise complaints (ii) through (iv), their top-ranking
preferencespresumablyareasfollows.
1. Ireceivethebenefitwithoutbearingthebenefits-producingcosts.
2. IdonotreceivethebenefitandIdonotbearthebenefits-producingcosts.
Asillustration,thinkofNozick’sfamouspublicaddress(PA)systemexample,173which
Nozick tried to use to discredit the fairness principle altogether. Imagine a
neighbourhood where some residents set up a PA system for entertainment. Each
resident isexpectedtocontributebybeing inchargeof it foradayperyear,during
which they can play records, make announcements, and so on. Many authors have
agreedwithNozickthatthereisnomoralobligationtodoone’sshareinsupporting
this entertainment scheme even if one voluntarily enjoys it, say, by opening their
windowfromtimetotime.
172Notethat,incasesofbenefitsproducedbyunjustmeans,theSharedPreferenceViewcansupporttheviewthatFairPlayobligationsdonotariseinthefirstplaceevenwhenrecipientsprefertoreceivethem.Thiswouldbethecaseiftherewereacollectivedutytoopposeunjustbenefits-producingschemes.Ibelievethistobethecase,andthatitwouldpreventFairPlayobligationsfromarisingfromsuchschemes,butIwillnotargueforthishere.173Nozick1974,p.93.
127
OntheSharedPreferenceViewtherecanbefairnessobligationstoparticipateinthis
entertainmentscheme,butonlyforthefreeriderswhocanbeascribedtheFRP.Asa
matteroffact,whensuchtrivial,discretionarygoodsareatstake,itisoftenthecase
that theFRPcannotbeascribedtomanywhofreeride,becausecondition(a) isnot
trueof them, that is, theydonot think thebenefit isworth itscost, typically for the
reasons given by (ii) through (iv). Since, unlike the contributors, they would be
preparedtoforgothegoodofthePAschemeratherthanpayforit,suchdissentersdo
notsharetherelevantrankingofpreferences,whichmeanstheydonotsharetheFRP.
4.8.BenefitsproducerswholackFairPlayclaimsImentioned inSection3.3 thatFairPlaydefendershavemaintained thatonly those
whoproduceapublicgoodthroughsomesacrifice,ornetburden,haveclaimsagainst
the free riders. Yet it remains unclear what the relevant sacrifice or burden is.
Similarlyuncleariswhethertheintentionsthatcontributorsexhibitmatter,andifso,
whatkindsofintentionstheymustbe.TheSharedPreferenceViewallowsustoshed
lightonboththeseissues,andthereby,toalsoidentifythosecasesinwhich,although
contributors benefit others who do not pay for those benefits, they lack Fair Play
claims.
Consider the Flat Share case again, and a variation on it. In Flat Share, all three
flatmateshappentopreferahighlevelofcleanlinessinthecommonareaenoughthat
theywouldbepreparedtodowhatwasrequiredtomaintainthislevelofcleanlinessif
freeridingwasnotaneffectiveoption(forexample,ifeachlivedaloneandtherewas
nooneelsearoundonwhomtheycouldrelytodothecleaning).Iamassumingthat
for some reason they cannot make a cleaning agreement that all can trust will be
respected.Twoof the flatmatesneverthelessdo their shareofmaintaining thehigh
levelof cleanliness thatallprefer.Onedoesnot.Asavariationon this case,wecan
128
think of Flat Share II: this is the same as before except here the cleanliness is
maintained thanks to the fact that two flatmates enjoy exercising around thehouse
withaduster,astheirpreferredwayofstayinginshape.
IsubmitthatonlytheflatmatesintheoriginalFlatSharehaveFairPlayclaimsagainst
thefreerider,forreasonsrelatedtotheirintentionsinbearingthecostsofcleaning.I
share thisopinionwithauthors likeCasalandWilliamswhoclaimthatcontributors
havefairnessclaimsonlywhenthey“bearcosts,thattheywouldnototherwisebear,
in order to produce the good”.174 They do not spend much time explicating or
justifyingthisclaim,however.TheSharedPreferenceViewenablesustodojustthat.
Thenotionofintentionalitythatisofinteresthereregardstheintentionalproduction
ofcertainbeneficialoutcomes inaway thatgrounds fairnessclaimswithrespect to
those outcomes. As such, I will not explore the intricacies of the literature on
intentional action, for thiswould take us too far afield. Herewe need only identify
whichfeaturesoftheproducers’intentionalityarerelevantforFairPlayandwhy.
Using an intuitive and commonly used definition, I will understand intentions as
referring to the goals, or outcomes, or consequences atwhich our actions aim. The
sense inwhichIusethistermissynonymouswithwhatScanlonhascalled“narrow
intentions”,whichhedefinesasfollows.
Toaskapersonwhatherintentionwasindoingacertainthingistoaskherwhather
aim was in doing it, and what plan guided her action – how she saw the action as
promotingherobjective.Toaskthisisinparttoaskwhatherreasonswereforactingin
such away –which of the various features ofwhat she realized shewas doingwere
featuresshetooktocountinfavorofactinginthisway.175
174CasalandWilliams1995,p.106.175Scanlon2008,p.10.
129
Scanloncontraststhenotionofnarrowintentionswiththatofbroadintentions.One
performsanactionintentionallyinthebroadsensewhenoneisawareofwhatsheis
doing,orisawareofcertainconsequencesofheractions.Abroadlyintendedoutcome
ofone’sactivities,then,isonethatdoesnotcomeasasurprisetotheagent.However,
bycontrasttoanarrowlyintendedoutcome,thebroadlyintendedoutcomeisnotone
that theagent tookasareason forher toact in theway that shedid. It ismerelya
foreseen,butnotproperlyaimedat.
ForFairPlaypurposes,itisnarrowintentionalitythatisrelevant,foranoutcomethat
ismerelybroadly intended(that is, foreseenbutnotaimedat) is, forall intentsand
purposes,amereexternalityoftheagent’sintentionalaction.
Of course,anarrowly intendedoutcomeor stateofaffairs can formpartofa larger
planofaction.Thatnarrowlyintendedoutcomecanthenbedescribedasameansto
ultimately achieving what we might call an ultimate aim. Ultimate aims refer to
outcomesorends that arenarrowly intendedandalsonon-derivative.Theyare the
terminus of our actions or plans for action, and cannot be further explained by
pointingtosomefurthergoalwewishtoachieve.Anultimateintention,then,would
refer to the agent’s ultimate aim in a chain of narrowly intentional actions: for
instance, forsomeonewhodeliberatelyworkstowardsobtainingadegreeachieving
thatdegreeismostlikelynottheirultimateaim.Itisameanstoafurtherend,perhaps
theirultimateendofpracticingtheirchosenprofession.
In a tradition going back to John Stuart Mill, Steven Sverdlik also distinguishes
betweenintentionsandmotives.Expressingthiswidelysharedview,hewrites:
[F]romtheagent’spointofview(and,perhaps,fromothers’)hermotivesspecifywhat
isofvalueaboutheraction.(…)Thetwomaintypesofmotiveseemtobeemotionand
130
desire.Typicalmotivesare,ontheonehand,jealousy,spite,affection,orsympathy,and,
ontheotherhand,greed,ambition,curiosity,orasenseofduty.176
So, for example, the flatmates in Flat Share intentionally produce the outcome of a
clean flat in thenarrowsense:havingaclean flat is theaimof theiraction.But this
may or may not be their ultimate intention. If it is, perhaps their motive is self-
regarding:theymightbemotivatedbythepleasureoflivinginacleanenvironment.
Alternatively,iftheirultimateintentionistoensuretheflatiscleanwhentheirfamily
visits,theymightbemotivatedbyother-regardingmotives.Theymightbeawarethat
theirfamilyenjoyacleanenvironmentandtheymightactoutofaltruisticmotivesto
ensuretheirfamilygettoenjoyacleanflatwhenevertheycomevisit.Buttheirmotive
might be self-regarding in this case also: perhaps theywant to be praised by their
familyformaintainingsuchacleanflat.TheflatmatesinFlatShareIIalsoproducethe
outcomeofacleanflatintentionallyinthebroadsense,inScanlon’sterms.Thisisto
say, it isnotanaccidental,unconscious,or involuntaryconsequenceoftheirbearing
thecostsofexercisingwithdusters.But theirnarrow intentionhas todowith their
fitness aims. And their motives could range from vanity to the desire of enjoying
health.
Ofcourse,thedistinctionbetweenintentionsandmotiveshasbeencontested,andin
particular, the distinction between ultimate intentions and motives can be quite
blurry.W.H.F.Barnes,forinstance,writes:
It is ahighlyplausible view to take that intention is a complex consistingofdifferent
elementsandthatmotiveisthenamewegivetooneelementintheintention,namely
thatelementwhichisofchiefinteresttotheagent.Inthissenseitseemstoberoughly
synonymouswithaim,endandpurpose.177
176Sverdlik1996,p.335.177Barnes1945,p.234,citedinPall1965,p.151.
131
Thephilosophyofactionliteraturenodoubthasalottosayaboutthis,butitiswell
beyondthescopeofthisthesistoengageinthesedebates.Forourpurposeshere,the
often-useddistinctionsbetweenintention(understoodasnarrowintention),ultimate
intention,andmotivewillsufficeforthetimebeing.
ThefirstthingweneedtoestablishforFairPlaypurposesistheobjectoftherequisite
narrowintention.Thequestionis,whichoutcome,orconsequence,mustcontributors
purposefully bring about if they are to have Fair Play claims against free riders?
Contributorsmustintendtobringabouttheoutcomethatfreeridersinternalizeand
for which they would be asked to pay their share.178In the Flat Share cases, the
relevantFairPlayoutcomeisaverycleanflat.Notethattheirultimateintentionneed
notbe tobenefitothers by cleaning.Nordo theyneed towant toact fairly towards
othersbydoingso,ortodeliberatelydischargeadutyassuch.Somecontributorsmay
wellexhibitsuchintentions,butthereisnoneedforthemtoaimtobenefitothersor
to act out of duty. Their ultimate intention may only be to benefit themselves by
deliberately producing an outcome that is (also) beneficial to others or that is
requiredbyacollectiveduty.
Having specified the requisiteobjectof contributors’ intentions,wemust clarify the
place that this intention shouldoccupy in theirmotivational structure. I propose that
contributors must have the production of the relevant Fair Play outcome as their
necessary motivating reason for action. We can use the following counterfactual
intentionalitytestfordeterminingwhethercontributorsaimfortherelevantFairPlay
outcomeinthesensejustdescribed.Wecanimagineaskingcontributors:Ifyoucame
tobelievethatthecostsyouarebearingwerenotcontributingtotheproductionofthe
relevantFairPlayoutcome,wouldyoustillbewillingtobearthem?Iftheanswerisno,
thisshowsthatcontributorsregardtheproductionoftherelevantgoodasanecessary
motivatingreasonforbearingtherelevantcosts.Somethinglikethistestseemstobe178Foreaseofexposition,Iwilloftenrefertothegoodswhichcontributorsmustpurposefullybringabout(eitheracollectivelybeneficialoptionalgood,ortheoutcomedemandedbythecollectiveduty)astherelevantFairPlayoutcome.
132
implied by Casal’s and Williams’s contention that contributors do not have claims
unless theybear costs theywouldnototherwisebearbut for theproductionof the
benefit.
ThecontributorsinFlatSharepassthecounterfactualintentionalitytest.Iftheycame
to believe that the burdens they were bearing would not lead to maintaining a
spotlessflat,theywouldceasetobearthem.Bycontrast,thecontributorsinFlatShare
IIwouldpresumablycontinueexercisingwiththedustersregardless.Thisisanatural
consequenceofthefactthatcleaningtheflatwasnottheirnecessaryreasonforacting
inthefirstplace.Andeventhough,asthingsstand,theyaredeliberately(meaning,not
unexpectedly)causingaspotlessflat,thisoutcomeinitselfplaysnomotivationalrole
forthem.If thethirdflatmateweretoconsistentlycleantheflatallbyherself,such
thatthedusterexerciseswerenotneededtocleantheflatanymore,thetwofitness-
oriented flatmateswould presumably still do their exercise.Or if they came to find
that the duster exercises simply were not effective for maintaining the flat clean
enough,onceagaintheywouldcontinuetodothem.Maintainingaspotlessflat,then,
is not their necessary reason for action in the first place: they have enough
independentreasonstoengageinthisactivity.
Now, why should we think that these are the features of the contributors’
intentionality that are required for Fair Play? The intentionality requirement as
describedisanecessaryconditionfortheascriptionoftheFRP.Andtheascriptionof
the FRP is crucial for ascribing the relevant hierarchy of preferences to the
contributors:
1. Ireceivethebenefitwithoutbearingthebenefits-producingcosts.
2. IreceivethebenefitandIbearthebenefits-producingcosts.
Contributors in Flat Share can be ascribed this hierarchy of preferences. They are
bearingcoststheyarenotpreparedtobearbutforwantingtomaintaintheflatclean.
133
This involves investing,or indeed “redirecting” timeand resources from theirusual
activities towardsproducingtherelevantoutcome.Butsincethedesiredoutcomeis
collective innature, and since thereareothersaroundwhoalsodesire it and could
produceit,eachofthesecontributorscouldenjoythebenefitatnocosttothemselves.
As prudentially rational persons, they can then be ascribed the preference not to
divert resources from theirusualpursuits towardsproducing anoutcome that they
maygetanyway.Thisistosay,theycanbeascribedthepreferencetofreeride.
By contrast, contributors in Flat Share II cannot be ascribed that hierarchy of
preferences.Failingthecounterfactualintentionalitytestmeanstheywouldbearthe
benefits-producing costs anyway, in light of their other ends (here, a fitness aim).
Theirhierarchyofpreferences,then,isthis:
1. IreceivethebenefitandIbearthebenefits-producingcosts.
2. Ireceivethebenefitwithoutbearingthebenefits-producingcosts.
We can now explain what the relevant burden, or sacrifice, is, in virtue of which
contributors have Fair Play claims against free riders. I submit that the relevant
burden is the contributors’ actingagainst their free ridingpreference.FairPlayas I
understanditappliesinsituationswhereitisnotpossibleforeachpersontopursue
theirambitions, in thewaytheywanttopursethem,compatiblywitheveryoneelse
pursuing their ambitions, and without this leading to a collectively suboptimal
outcome. So, if the suboptimal outcome is to be avoided, somepeoplemustmodify
their ambitions, or divert resources away from them, as needed to produce the
beneficialoutcome.Butsincethedesiredoutcomeiscollectiveinnature,eachofthe
contributorscouldenjoyitatnocosttothemselves.EachcanbeascribedtheFRP,and
thosewhoactinfrustrationoftheirFRPbybearingthebenefits-producingcostshave
Fair Play claims that the free riders similarly frustrate their FRP. Failing to do so
wouldamounttoenjoyinganunjustifiableprivilegeattheexpenseofthecontributors.
134
In contexts like Flat Share II, where there are enough people willing to bear the
benefits-producingcosts in lightof theirotherends, it ispossible foreachtopursue
theirambitionscompatiblywitheveryoneelsepursuingtheirown,andstillavoidthe
collectively suboptimal outcome. The contributors in Flat Share II enjoy the same
“privilege”thatfreeridersenjoy,whichistopursuetheirambitionsasnormalwithout
losingoutonthecollectivebenefit.
SofarIhavearguedthattheFRPcanbeascribedtocontributorswhoseonlyintention
is to contribute to the relevant Fair Play outcome, but not to those who are only
aimingtoachievegoalsthatareunrelatedtotherelevantFairPlayoutcome.Nowthe
questioniswhatweshouldthinkofcontributorswithmixedintentions,namelythose
whowanttoproducetherelevantFairPlayoutcomeaswellastoachievesomeother
aim at the same time.We can imagine a casewhere the flatmateswant to kill two
birdswithonestone:theywanttobothmaintainaveryacleanflat,andtogetother
benefits at the same time, like getting some exercise. Can they still be ascribed the
preferencenottobearthecostsofcleaning?
If theypassthecounterfactual intentionalitytest, theanswerisyes. If it istruethat,
even factoring in these other benefits (or incentives), they would not bear the
benefits-producingcostsiftheycametobelievetheywerenotneededtoproducethe
relevant collective good, we can still ascribe the FRP to them. They still count as
making some necessary adjustments or changes to how they use their time and
resources in order to enable themselves to bear the benefits-producing costs. They
mightchangethemeansbywhichtheypursuetheirusualambitions(perhaps,if left
totheirowndevices,theywouldratherjoginsteadofcleaningtheflatforexercise),or
they might acquire new aims and preferences which they consider conducive to
producingthecollectivegood(likeacquiringanewfitnessgoalthattheydidnothave
before andwhich they judge to bewell served by cleaning the house). Once again,
thesearechangesthatotherscouldhaveundertakeninsteadofthem.Andonceagain,
they could enjoy the relevant Fair Play outcome without having to modify their
135
preferences and ambitions in any way, provided that enough others did instead.
Insofarastheywouldnotbewillingtobearthecostsofcleaning(evenwithprivate
incentives available) unless they believed they were conducive to producing the
relevantFairPlayoutcome, theystillpass thecounterfactual intentionality test, and
hencetheycountasactingagainst the freeridingpreferencewhentheydobearthe
benefits-producingcosts.
Now,whatabout thecontributors’motives? Is it thecase thatwecanautomatically
ascribetheFRPtocontributorssolongastheypassthecounterfactualintentionality
test? The answer seems to be no. There are somemotives (granted, a minority of
them)thatmakeitthecasethatthosecontributorswhoactfromthemactinlinewith
their ambitions even when they pass the counterfactual intentionality test. An
examplecouldbeseekingpraiseorglory foroneselfbyproducingacollectivegood.
Someone might have the production of a collective benefit as their necessary
motivating reason for bearing the relevant costs, but they positively value the
opportunitytohelpproducethecollectivebenefitasawaytoattractpraiseorglory.
Thissuggeststhatweneedasecondtesttodiscernbetweenthecontributorswhocan
beascribed theFRPand thosewhocannot.Wemustaskcontributorswhopass the
counterfactual intentionality test this second question: Other things equal, if the
relevant Fair Play outcome could be secured through a costless, problem-free process
such asmanna from heaven,would you rather it be secured through that process, or
would you still prefer to contribute to it yourself as you currently are? Call this the
counterfactual motivational test. The only contributors who would prefer to help
producethecollectivebenefitthemselvesratherthanseethatsameoutcomesecured
throughmanna fromheavenare those forwhom, in reality,producing theFairPlay
outcome is in linewiththeirpreferredactivitiesand lifeplans.Assuch, theycannot
complainthatfreeridersareenjoyingaprivilegethattheythemselveslack.
136
Contributors who pass the counterfactual intentionality test but do not pass the
counterfactualmotivational testwill surelybeonlyaminorityofall thosewhopass
theintentionalitytest.Passingthemotivationaltestiscompatiblewithawiderangeof
motivesthatcontributorsmightactfrom,includingmoralreasons.Someonewhodoes
her share out of a sense of duty, or indeed out of a Fair Play-informed motive
specifically,might seem to have an interest in bearing the benefits-producing costs
herselfandnottorelyonotherstodoitinsteadofher.Butsuchapersonwillalmost
certainlypassthecounterfactualmotivationaltest:shewouldratherthatnoonebear
anycosts,ormakeanysacrifices,toproducethisgood.Ifitwerepossible,shewould
ratherthisgoodbesecuredthroughmannafromheaven,andshewouldbewillingto
forgotheopportunitytobearthesecostsherself(and,byextension,shewouldalsobe
prepared to forgoanyotherprivatebenefitsshemightaccrueasaresultofbearing
these costs). A person motivated by sympathy, beneficence or altruism towards
others,too,willpassboththeintentionalityandthemotivationaltestinamajorityof
cases.
At the same time,moral or altruisticmotives are not necessary for contributors to
haveclaims.Self-interest,forexample,isalsoamotivethatiscompatiblewithpassing
boththeintentionalityandthemotivationaltest.Indeed,inmyFlatSharecaseIhave
beenassumingallalongthattheflatmateswhohaveFairPlayclaimsare,first,aiming
solely at producing an exceptionally clean flat, and secondly, that their motive for
wantingtoproducethisoutcomeisself-interest:theysimplywanttoenjoyacleanflat
forthemselves.
This featureof theSharedPreferenceView isnoteworthy for theaimsof this thesis
becausemanyon thepro-sharing side of theparental justice debate have criticized
anti-sharingtheoristsforworkingwithtoostringentaviewofFairPlay,namelyone
that requires contributors to do their part out of an altruistic motive, or with the
137
ultimate intention of benefitting others.179The Shared Preference View agrees that
altruistic or beneficent ultimatemotives towards others are not necessary for Fair
Play claims to arise. As we shall in the next chapter, the Shared Preference View
nevertheless denies that typical parents have Fair Play claims against non-parents,
sincetheydonottypicallypassthecounterfactualintentionalitytest.
Another noteworthy feature of the Shared Preference View is that the relevant
burden, or sacrifice, in virtue ofwhich contributors have Fair Play claims isnot, as
some have been tempted to think, an ex post imbalance in benefits and burdens
betweenthecontributorsandthefreeriders.Itmaybethoughtthatwhatiswrongin
cases of unfair free riding is that, at the end of the day, free riders are enjoying
comparativelymoreresources,ormorewelfare,relativetothecontributorsinvirtue
ofthefactthatthefreeridersgettobenefitforfreewhilethecontributorsalsopaythe
benefits-producing costs.180Or itmaybe thought that theunfairness lies in the fact
thatthefreeridersenjoyanetbenefitoradvantagewhilethecontributorsenjoyanet
burdenordisadvantageasaresultofdoingtheirshare.
Butthesearenotverygoodwaysofunderstandingthewrongnessoffreeriding.First,
withrespecttolocatingthewrongnessoffreeridinginthefactthatcontributorscan
experienceanetlossbypayingtheirshare,thisseemslikeitwouldrestrictthescope
of Fair Play implausibly. In Flat Share, perhaps the two flatmates would be so
botheredbyalowerstandardofcleanlinessthattheywouldnotregardthemselvesas
incurringanetlossoverallevenifforcedtobearallthecostsofcleaning.Butthis,by
179Mostnotably,Olsaretti2013,p.246.180E.g.ThisseemstobewhatbothersGeorge(1987,pp.1-3)whenhearguesthatfairnessdemandswerectifywhathecallsthe“Imbalance”inresourcesbetweenparents,whoinvestinraisingchildren,andnon-parentswhobenefitfromtheirinvestment.Boran(2006,pp.106-107)reconstructsFairPlayasaprincipledenyingclaimsofcompensationtothosebenefits-producerswho,weretheyrewardedforproducingcollectivegoods,wouldenjoymorewelfarethanthosewhofindproducingthesamekindsofbenefitsburdensome.Shecriticizesthisversionoftheview(pp.107-112).
138
itself, should not settle the question of whether they have claims against the third
flatmatewhoalsoprefersaverycleanflat.181
Second,comparativeexpost inequalities inresourcesorwelfare in themselvesneed
notbeunfaireither.Merelypointingtothefactthatsomepeoplereceivedsomething
forfree,orthatsomepeoplehavemoreresourcesattheendofthedaycomparedto
others,isnotobviouslyunfair,asthiscanalsobethecasewiththeresultsofgambles
orwith doing favors for others.More importantly, there aremany caseswhere the
contributorsare,attheendoftheday,betteroffintermsofresourcesorwelfarethan
some free riders. Some parents may be better off both in terms of resources (for
example, if theirchildrengrowuptobeverywealthyandverygenerouswiththem,
especiallyduringtheiroldage)aswellasintermsofwelfare(overalltheymayenjoy
richer,moreflourishinglivesthankstotheirdecisiontobecomeparentscomparedto
somenon-parents).Andmanyparentscertainlyconsideritanetbenefittoenjoythe
goodsofhavingandraisingchildren,evenwhilehavingtobearallthecostsforthem,
compared to doing neither. If Fair Play were a principle meant to redress certain
formsofexpost inequalityofresourcesorofwelfare(namely, thoseresultingoutof
collectivegoodsproduction),weshouldconcludethatcontributorswhoendupbetter
offthanfreeriders(byvirtueofhavingbornesomeofthebenefits-producingcosts)
donothaveFairPlayclaimsagainstthefreeriders.ButItakeitthatthiswouldnotfit
withtheconclusionsthatthepro-sharingcampwouldliketosecureforparents,nor
withhowweunderstandtheroleofFairPlaymoregenerally.
The Shared Preference View maintains that, in order to have Fair Play claims,
contributorsmustseetheproductionoftherelevantFairPlayoutcomeasanecessary
motivating reason for them to bear the benefits-producing costs. Such contributors
haveclaimsbecausetheyarebearingcoststhatareworthwhileforthemtobearonly
insofar as they contribute to the desired beneficial outcome. But this outcome is181Olsaretti(2013,p.243)makesthispointinthecontextofparentaljustice,whereitisacknowledgedthatparentsoftendonotregardhavingchildrenasanetloss,andsherightlyaskswhythisfactaloneshouldentaildenyingparents’FairPlayclaims.
139
collective in nature, which means that the contributors could get it even without
bearingthesecosts,dependingonwhetherthereareothersaroundwhoalsohavean
interest intheoutcomeandwhocouldbearthecosts instead.There is, then,aclear
sense inwhich the contributors, like the free riders, have anexante interest innot
bearing the benefits-producing costs themselves, for thismeans diverting time and
resourcesawayfromtheirotherendsandambitionsanddirectingthemtowardsthe
productionofanoutcomethattheycouldenjoyforfree.TherelevantsacrificeforFair
Play purposes, then, is not an inequality of resources or welfare between the
contributors and the free riders, nor is it incurring a net loss by bearing the costs
(thoughthiscanalsobe,andis,thecaseinmanyinstancesofunfairfreeriding).The
relevantsacrifice,orburden,ontheSharedPreferenceViewisthewillingnesstoincur
whatisultimatelyan“unnecessary”restriction,sotospeak,ofone’susualambitions.
Itis“unecessary”fromthepointofviewofeachindividual,becauseeachpersoncould
sitbackandallowotherstoproducethegoodthateveryoneenjoys.Butitisnecessary
fromthecollectivity’spointofview,becauseunlesssomepeople restrict theirusual
ambitions,allwillsufferacollectivelysuboptimaloutcome.Thosewhodositbackand
allowthecontributorstoproducethecollectivelydesirableoutcomeenjoyaprivilege
at their expense: theyget thebestofbothworlds,namely topursue their endsand
ambitionsinanunalteredfashionwhileenjoyingthecollectivelybeneficialeffectsof
others’restrictingtheirs.
4.9.ConclusionIn this chapter I have put forth a principle of Fair Play that is, first, justified by a
systematicaccountofwhatmakes freeridingwrongfulwhen it iswrongful; second,
thattakesseriouslythevoluntaristobjectionpressedbyNozick;third,thatcarvesout
a plausible scope of application that does not lead to implausible proliferation of
obligations,andthatis,finally,capableofcondemningtheintuitiveunfairnessoffree
140
riding across a range of benefits, both optional and morally required. In the next
chapterIdrawouttheimplicationsofthisFairPlayviewforthequestionofparental
justice.
141
Chapter5
AreNon-ParentsUnfairFreeRiders?
HavingdevelopedwhatIconsidertobethemostdefensibleaccountoftheFairPlay
principle,inthischapterIevaluatetheprospectsforafairness-basedcaseinfavorof
subisidizing parents. I examine three brands of fairness-based arguments that have
beendefendedintheparentaljusticeliterature.Themaindifferencebetweenthemis
how they conceive of the relevant goods that parents create by having and raising
children, andwhichnon-parents are accusedof unfairly internalizing. Thus, Section
5.1 will consider the argument according to which parents create public goods by
havingandrearingchildren;Section5.2discussesargumentsthattakechildrentobe
morally required goods; and Section 5.3 turns to the ‘children as socialized goods’
argument.IneachcaseIinvestigatewhethertheSharedPreferenceViewcansupport
theclaimthatnon-parentsoughttohelpshoulderthecostsofchildrenasamatterof
fairness. I argue that parents do not typically have Fair Play claims against non-
parentsbecausetheyfailtheintentionalityrequirement.
5.1.ChildrenaspublicgoodsItseemsundeniablethatparentscreatemanypublicgoodsthroughhavingandraising
children.Aworldwithchildreninitisclearlypreferabletoonewithoutthem.Thisis
soforvariousreasons.First,asSamuelSchefflerargues,ourcapacitytogoaboutour
livesaswedonowmightdependonthe implicitassumptionthathumanitywillnot
endwithourgeneration.Ifwehadreasontobelievethathumanitywouldgoextinct
142
afterweourselvesdied,saysScheffler,wewouldabandonmanyoralloftheprojects
that we currently find valuable, such as scientific research, upholding cultural
traditions, or political activism. Arguably, we might even lose our appetite for life
altogether, as our confidence in almost all of our values implicitly depends on the
belief that there will be life after we die.182 Whether or not we ultimately find
Scheffler’sclaimsplausible,attheveryleastwecansaythatifthechainofgenerations
were to endwith us,wewould find our lives negatively affected to some extent or
another.Second,andrelatedly,insofaraswehavespecificprojectsandcommitments
thatstretchbeyondourownlives,webenefitfromtheexistenceoffuturegenerations
which can complete the projects we left unfinished or honor the commitments we
helddearandperhapsevenmadesacrificesfor.183
Then there aremorematerialways inwhich new generations benefit both parents
and non-parents, which pro-sharing arguments have tended to emphasize. All, or
virtuallyall,ofusdependonnewgenerationsofproductivecitizenstocontributeto
maintainingwelfare schemes suchaspensions,publichealthcare systems,disability
benefits,unemploymentbenefits,andsoon.Oureconomies,aswellasourinstitutions
andourwelfaremeasures,cruciallydependontheexistenceandtheparticipationof
future generations. Finally, new generations might develop new cultural and
technologicalgoodsthatenrichallofourlives.NancyFolbre,oneofthebetter-known
defendersofthe ‘childrenaspublicgoods’argument,describestherelevantbenefits
asfollows.
Parents who raise happy, healthy, and successful children create an especially
important public good. Children themselves are not the only beneficiaries.
Employers profit from access to productive workers. The elderly benefit from
SocialSecuritytaxespaidbytheyoungergeneration.(...)Fellowcitizensgainfrom
having productive and law-abiding neighbours. These are all examples of
182Scheffler2013,pp.23-26.183SeeLenman2002,p.262.
143
“positiveexternalities”becausetheyareexternaltotheactualdecisiontoprovide
care.184
The ‘children as public goods’ arguments claim that, since parents produce these
substantial benefits, everyone who internalizes them (or at least everyone who
voluntarilyinternalizesthem)oughttosharesomeofthecostsofproducingthem.Rolf
Georgeputsthecoreissueinthefollowingway:
Childrengrowupandbecome,amongotherthings,providersofpensions,maintainers
ofsociety.(...)Theycostasubstantialsumtoproduce.Nowsincetheyarefreeagents,
escaping thus the controlof their investors, theybecomeresomnia, benefit everyone.
Whoshouldreapthebenefitstheyhavetodispense?185
Asmentionedatthebeginningofthepreviouschapter,Iwillnottakeissuewiththe
empirical contention that parents create benefits for the rest of society. Although
theremaybecontextsinwhich,givencertainconditions,parentscreatepublicharms
(suchasperhapsinconditionsofoverpopulationandextremeresourcescarcity),itis
certainlythecasethatoursocietyreliesonthecreationofsomenumberofchildrenor
another.SoIassumethatthe‘childrenaspublicgoods’argumentsarerelevantatleast
when it comes toanumberof children thatdoes represent anetbenefit to society.
Thisisenoughforthepurposeofthisthesis,whichistoestablishwhatsortsofclaims,
ifany,parentscanraiseagainstnon-parentsasamatterofbasicjustice,andassuming
aresponsibility-sensitiveviewofjustice.Forthistocountasaworthwhilequestionof
justice it isenoughthat thereexistsomeconditionsunderwhichhavingandraising
childrendoescreatethepublicbenefitsoutlinedabove,invirtueofwhichparentscan
layclaimsagainsttherestofsociety.
Granting, then, the empirical premise on which the ‘children as public goods’
argumentsrely,wemustnowturntothenormativequestion:aretheydoingsoina
184Folbre2002,p.50.185George1987,p.31.
144
waythatgivesthemaclaimtohavethecostssharedbysocietyasamatteroffairness?
Thequestioncanbefurtherspecified:Arenon-parentsunfairlyfreeridingonparents’
effortsofhavingandraisingchildren?
Before moving on to answer this question, let us address one potential
misunderstanding.Thecharge thatnon-parentsareunfair free riders shouldnotbe
taken as accusing non-parents of contributing nothing whatsoever to the welfare
schemes they benefit from, or of contributing no value to society. Of course, non-
parents, too, pay their taxes throughout their working lives, and their taxes help
support older generations. So non-parents could point to these contributions and
arguethattheyhavedonetheirpart:theyhavesupportedtheretiredpopulation,and
nowtheyareentitled(invirtueofsomeprincipleofreciprocity)tobesupportedby
thenewgenerations.186Inresponse,itisimportanttounderstandthatthepro-sharing
claim isnot thatparentsproduceall thevalue forsocietywhilenon-parentsmerely
internalize it,or thatnon-parentsoughtnot tobesupported inoldage.Rather, it is
that while both parents and non-parents contribute to society as productive
individualsintheirownright,andtheybothhelpsupportoldergenerations,itisonly
parentswhoprovidea furthercrucial“service”ontopofthat.Theymake it thecase
thatanextgenerationexistsandisabletofurtherthecurrentwelfareschemesaswell
asotherinstitutionsorendeavorsthatmustbesupportedovertime.Andsincehaving
andraisingchildrenisexpensive,thepro-sharingclaimisthatnon-parentsshouldbe
enjoinedtopaytheirshareforthissociallybeneficialserviceaswell.
Non-parents, I submit, are not unfairly free riding on parents’ efforts. For parents
seemtobemoreakintotheflatmatesinmyFlatShareIIexamplefromtheprevious
chapter,ratherthantheflatmatesinFlatShare.LikethecontributorsinFlatShareII,
theydonotpassthecounterfactual intentionalitytest.Recall thatthecounterfactual
intentionality test would ask parents: If you came to believe that the costs you are
bearing were not contributing to the production of the relevant Fair Play outcome,186ForaninsightfuldiscussionofprinciplesofintergenerationalreciprocityseeGosseries2009.
145
would you still bewilling to bear them?By “relevant Fair Play outcome” I mean, of
course, thosegoodswhichnon-parentsareaccusedofunfairly freeridingon.Sothe
question is, if parents came to believe that having and raising children did not
contributetotheproductionofbenefits likethrivingeconomicandwelfaresystems,
culturalandtechnologicaladvancement,andeventhepeaceofmindthattheexistence
ofnewgenerationsgivesus,wouldparentsstillwanttohaveandraisechildrenlike
theydonow? I take it thatmost parentswould answerpositively. This is a natural
consequenceofthefactthattheydonotregardtheproductionoftheseoutcomesas
theirnecessaryreasonforaction.
What, then, are theirnecessarymotivating reasons for action?Wecan lookatwhat
ChristineOverall lists as common responses that people givewhen surveyed about
theirreasonsforhavingchildren.
Whenpeople are asked informally to explainwhy theywant tohave childrenorwhy
theyhadthechildrentheyhad,theyoftenspeakinconsequentialisttermsofpotential
benefitsforthemselves:“Ilovechildren”;“Ilovebeingpregnant”;“Iwillnotbefulfilled
unless I have a child”; “I don’twant tomiss out on the experience of parenting;” “To
relivemyownchildhood”;“SoIwon’tbelonely”;“Ijustwantsomeonetoloveme.”The
Planned Parenthood Association says that people explain their desire for children as
follows:“TogivesomeonetheopportunitiesIneverhad.Tohaveachildtobelikeme.
Tokeepmecompany.Topassonbeliefs,valuesandideasto”(quotedinBergum1997,
29).187
Whateverelsewethinkofthesereasons,itseemsclearthatcontributingtosecuring
some collective good is not what motivates the standard person’s procreative and
parental choices. This much has been acknowledged by pro-sharing advocates as
well. 188 Now, the Shared Preference View is able to explain why this widely
acknowledgedfactshouldmeanthatparentsdonothaveFairPlayclaimsagainstnon-
187Overall2012,pp.77-78.188SeeOlsaretti2013,p.246;Arneson2014,pp.19-20.
146
parents.OnthisviewofFairPlay,thosewhobearbenefits-producingcoststhatthey
would have borne anyway, for their collective good-unrelated reasons, cannot be
ascribed the Free Rider’s Preference. For the costs they are bearing are justified in
light of ambitions they want to pursue regardless of whether pursuing those
ambitions produces the relevant Fair Play outcomes. As such, they have an exante
interestinbearingthosebenefits-producingcosts,insofarastheyhaveaninterestin
fulfillingtheirambitions.Thehierarchyofpreferencesthatwemayascribetoparents,
therefore, with regards to bearing the costs of children and to the receipt of the
collectivegoodstheirchildrenhelpcreatepresumablyis:
1. IreceivethebenefitandIbearthebenefits-producingcosts.
2. Ireceivethebenefitwithoutbearingthebenefits-producingcosts.
IfparentscannotbeascribedtheFRP, theycannotclaimthat,byhavingandraising
children,theyareactingagainsttheirFRPandtherebyincurringtherelevantsortof
“sacrifice” or burden. The relevant sacrifice that paradigmatic contributors incur
stems from the fact that they bear costs that they are not prepared to bearbut for
wantingtoproduceacollectivegood.So,paradigmaticcontributorscanbeattributed
an interest in not directing resources away from their other ambitions in order to
produce a good that theymay get anyway,with orwithout their own contribution,
duetoitscollectivenature.Whensuchcontributorspaytheirshare,theycanbesaid
tohaverestrictedtheirambitionsinordertoproduceacollectivegood,whichallows
freeriderstheprivilegeofenjoyingthatgoodwithoutrestrictingtheirs.Butparents
donotseemtolackthisprivilegeeither.Neithertheparents,northenon-parents,are
diverting time and resources from their independently chosen pursuits in order to
create the relevant Fair Play outcomes. Parents, like the non-parents, are pursuing
ambitionsandendsthattheywoulddoanyway,andtheyarealsobenefittingfromthe
collectivebenefitsthatchildrenrepresentorcreatelater.
147
Now,itispossiblethatsomeparentspassthecounterfactualintentionalitytest.There
maybeparentswhosenecessarymotivatingreasonforprocreatingis,forinstance,to
save their traditionsor community fromdisappearing.Suchparentswouldpass the
counterfactual intentionalitytestandcouldbeattributedtheFRP.But, first, I take it
thatsuchparentsaretheexceptionratherthantherule,atleastinWesternsocieties.
Second, some argue that it would be undesirable if parents did regard producing
collectivegoodsastheirnecessarymotivatingreasonforhavingandraisingchildren.
For if producing those collective benefitswere their necessary reason for doing so,
theywouldhavetoregardhavingchildrenandraisingthemintosuccessfuladultsasa
means to the end of producing the relevant collective goods. Treating children as a
meansforproducingsocietalbenefitsisathoughtthatmanywouldfindrepelling.Ina
differentcontext,EmilyMcTernanwritesthatitwouldbeundesirableifparentsmade
their procreative and childrearing decisions based primarily on what is good for
society.“Aparentwhocaresfortheirchildonlyafterreflectingthatitispermissible
given its contribution to wider society introduces one thought too many,” she
concludes.189She is not referring to parental justice issues here, but she is pointing
out that it would be morally undesirable for parents to base their childrearing
decisions not on what is best for their children but on what is best for the wider
society.If it ismorallysuspect,andperhapsevenimpermissible, forparentstotreat
childrenasameansinthisway,parents’claimsFairPlayclaimsmightbeweakened.
For Fair Play does not generate obligations to participate in unjust or immoral
practicesthatmightproducebenefits.
Somehavepointedout thatparentsdonot typically treat theirchildrenmerelyasa
means forbenefittingsociety,butmanydohavemixed intentions,andperhapsalso
mixedmotives,forhavingandraisingchildreninthewaythattheydo.Thatis,while
their decisions are mainly motivated by their own welfare and the interests and
welfareofthechildrentheycreate,manyparentsarealsomotivated,atthesametime,
189McTernan2013,p.102.
148
by wanting to contribute to society by doing so. And we might consider this both
morallypermissibleandsufficientforgroundingparents’fairnessclaimsagainstnon-
parents.HereiswhatArnesonwritesonthis.
Igrantthatpeoplewhoeitherhavesexwiththeaimofhavingchildrenorwhohavesex
and then make a decision to bring the fetus to term when it is discovered that a
pregnancyhasstartednormallyact intheexpectationthatraisingchildrenwillenrich
theirlivesandmakethedecisionforthisreason.Butthereisnormallyanotherelement
inplay.Peopledecidetohavechildrenforself-fulfilment,butthisisamoralizednotion
of self-fulfilment. Procreators think that their childrearing activities will significantly
enhance the community in which they live, and they are also aware, perhaps in a
somewhat inchoateorvagueway, that there is aduty tobe fruitful andmultiply that
fallsontheircommunityandisonethattheirprocreativechoiceshelptofulfil.People’s
motivesaremixed,butthatdoesnotprecludetheirhavingthemotivationsthatarethe
conditions for [the] Hart-Rawls [Fair Play principle] to apply. After all, many who
volunteer to contribute to national defense, the paradigmof a cooperative scheme to
which Hart-Rawls applies, have mixed motives and aim in part at their own self-
fulfillment,throughmeaningfulworkorgloryseekingorthelike.190
Thisparagraphbringsup,first,thatparents’reasonsmaynotbeasdivorcedfromthe
socialbenefitsthatnon-parentsenjoyasImayhavemadethemseeminthischapter.
Second,itsuggeststhatactingoutofsuchmixedreasonsshouldbeenoughtoground
parents’ Fair Play claims. To support this, Arneson brings up the case of the glory-
seeking soldierwho is presumablymotivated both bywanting to contribute to the
collective benefit of national defence, aswell as to engage inwhat he perceives as
meaningful,glory-awardingwork.
190Arneson2014,pp.19-20.Olsaretti2013,p.246sharesthisthought.Shewrites,«Ithinkitisveryplausiblethatmanyparentshavemixedmotives:theywanttheirchildrentobecomeeconomicallysuccessfulandsociallyproductiveadultsbothbecausethisisgoodfortheirchildren,andbecausethisisgoodforothers»;emphasisinoriginal.
149
Inresponse,Iwanttosuggestthatifwelookcloserattheanalogybetweentheglory-
seekingsoldierandparentswewilldiscover, first,thatnotallglory-seekingsoldiers
haveFairPlayclaims,andthatparentsaremoresimilartothosewholackclaimsthan
to those who do, even if we take their mixed intentions and mixed motives into
account.
First, all glory-seeking soldiers seem to pass the counterfactual intentionality test.
Thisisalreadyacrucialdifferencebetweenthemandmostparents,whomostlikely
donotpassthattesteveniftheirreasonsforactionareasmixedasArnesonsuggests
theyare.Butwhenitcomestosoldiers,ifbearingthecoststhattheyarebearing(by
enrolling in the army and participating inmilitary efforts)would not contribute to
nationaldefence,theywouldnothavereasontobearthem.
Now, whether or not glory-seeking soldiers ultimately have Fair Play complaints
against those who free ride depends on whether they also pass the counterfactual
motivational test.Recall that this second testasks:Otherthingsequal,iftherelevant
Fair Play outcome could be secured through a costless, problem-free process such as
mannafromheaven,wouldyouratheritbesecuredthroughthatprocess,orwouldyou
stillprefertocontributetoityourselfasyoucurrentlyare?It ispossiblethatdifferent
soldierswillanswerthisquestiondifferently.Therearesomepeopleforwhomgoing
towar represents a good opportunity to attain glory or to do something they find
meaningful, but whom might still prefer that no one had to attain glory or find
meaningbygoingtowar.Theymightstillprefer thatwarsdidnotexistand thatno
onehadtobearthesecosts.Butsincewarsdoexist, theyarepreparedtobeamong
the people to bear the costs. Such people, I argue, can raise Fair Play complaints
against thosewhobear no costs at all because there is still an identifiable relevant
sacrifice they aremaking. They are bearing costs theywould rather no one had to
bear. Nevertheless, as things stand, they are prepared to bear these costs, first,
because theyarenecessary toproduce the relevantFairPlayoutcome, and, second,
150
thankstothefactthattheirvaluesallowthemtofindotherthingsof interestinthis
pursuit,suchastheopportunitytoattainglory.
As suggested in Section 3.8 in the previous chapter, however, other glory-seeking
soldiersmight not pass the counterfactualmotivational test. Theywould regret not
having the opportunity to go to war themselves if it turned out that any risk to
nationalsecuritywereavertedthroughsomehappyoccurrence.Thatthesepeopledo
nothaveFairPlayclaimsagainsttheirfellownationalsforhavinggonetowarmight
strike some as counter-intuitive. But it is hard to see what sacrifice, or what
restriction of one’s ambition, they are incurring. Of course, they are engaging in an
objectivelycostlyandriskyactivity,butiftheyarethetypeofpeoplewhorelishthe
opportunitytofightinwars,then,ontheSharedPreferenceView,thereisnowayfor
themtobeascribedtheFRP.Thisisnottosaythatwehavenoreasonwhatsoeverto
pay them a wage or to cover their medical expenses. But these reasons are not
grounded in fairnessbecausetheycannotclaimthat their toppreference is toenjoy
nationalsecuritywithoutbearingthenationalsecurity-producingcosts.Indeed,they
considerthemselvesluckytohavetheopportunitytogotowarinthefirstplace.
Whenitcomestoparents,nodoubtmanysincerelyhope,andwelcometheprospect,
thattheirchildrenmightenrichthelifeofthecommunityinvariousways.However,if
the thought of the social benefits that their children couldproduce are not actually
decisive for theirprocreativeandparentaldecisions, theystill fail tohaveclaimson
theSharedPreferenceView. Inotherwords, if it remainstruethat theseparentsdo
notpassthecounterfactualintentionalitytest,theylacktheseclaims.Allthathasbeen
established,itseems,isthatparentsrecognizeandwelcomethebenefitsthattheirlife
planswill cause for the rest of society: these benefits remainmere externalities to
their choices, as far asFairPlay is concerned.Andeven if someparentspassed the
counterfactualintentionalitytest,theywouldnotpassthecounterfactualmotivational
test. Like the soldiers who relish the opportunity to go to war, parents normally
consider themselves happy to have the chance to become parents.We can see this
151
pointbestifweconsiderwhatprospectiveparentswouldpreferinasituationwhere
thenumberofchildrenintheircommunitywassohighthatalimithadtobeimposed
ontheabsolutenumberofchildrenbeingcreated.Supposethereisnoshortageoftax-
payers, but creating too many more children would eventually cause great public
harmforenvironmentalreasons.Prospectiveparentswouldbehappyforthemtobe
theonestotakeadvantageofthelimitedopportunitiesforprocreationdespitethefact
thatthesocialbenefitsofchildrenareplentifulalready.
Another challenge that some pro-sharing theorists have brought regards the
definitionoftheobjectoftheintentionalityrequirement.Olsarettiacknowledgesthat
parentsdonotnormallyaimtoproducesocialbenefitsbyhavingandraisingchildren,
but she argues that Fair Play need not require this of them in the first place. The
beneficialoutcome,shepointsout,issimplyhavinganewgenerationoflaw-abiding,
productivecitizens.Andmostparentsdeliberatelyundertaketoraisetheirchildrento
be law-abiding, economically and socially successful adults. Their ultimate, non-
derivativemotivation for doing so typically regards the expected future welfare of
their children rather than the public welfare. But this, Olsaretti claims, should not
countagainsttheirfairnessclaims.ForsurelyaFairPlayviewthatonlyacknowledged
the fairness claims of those who contribute out of purely beneficent, or altruistic,
ultimatemotiveswouldbetoostringentaview.191
Aspointedoutbefore, I fully agreewith the idea that altruisticmotivations arenot
necessaryforcontributorstohaveFairPlayclaims.Ialsoacknowledgethepointthat
one of parents’ main aims is to raise their children to be law-abiding, productive
adults. The fact thatmost children in society today growup to be (minimally) law-
abiding and productive adults is certainly no accident but rather the result of
deliberateparentingchoices.Sothisoutcomeisintendedbyparents.Furthermore,if
raisingone’s children tobe law-abidingandproductivewere the requisiteobjectof
contributors’ intentions for Fair Play purposes, then most parents would arguably191Olsaretti2013,p.246.
152
passthecounterfactualintentionalitytestwithrespecttoproducingthisoutcome.We
couldaskparents: Ifyoucametobelievethatthecostsyouarebearingbyhavingand
raising children in theway that you currently are did not contribute to turning your
childrenintolaw-abiding,productive,successfuladults,wouldyoustillbewillingtobear
them?Surelymanyormostparentswouldanswernegatively.Nodoubtturningtheir
children into law-abiding, productive, successful adults is the necessary reason for
whichtheybearmanyormostofthecoststheyarebearing.Assuch,theywouldhave
FairPlayclaimsagainstfreeriderswhobenefitedfromthisoutcome.
However,regarding,first,parents’aimtoraiselaw-abidingadults,Iwouldarguethat
this isanoutcomeforwhichparentsdonothaveFairPlayclaimsbecauseit istheir
obligationtoraisechildreninwaysthatdonotharmorwrongothersinthefirstplace.
Second,asregardsraisingchildrentobesuccessfulandproductiveinthesocietythey
livein,thisisnotyetanoutcomethatbenefitsthirdparties.Assuch,parentscannot
have Fair Play claimswith respect to this outcome because they are not aiming to
produce the benefits that non-parents actually free ride on. I will address each of
thesepointsinturn.
Onenecessaryreasonforwhichparentsraisetheirchildrenthewaytheycurrentlydo
isclearlygearedtowardscreatingasocialbenefit,oratleastavoidingasocialharm:
making sure that their children grow up to be law-abiding and respectful of other
people’smoralrightsratherthanlaw-violatingandinsensitivetoothers’moralclaims.
Butarguably,parentsarenotmorallyfreenottoraisetheirchildreninthisway.Inthe
samewayas it istheresponsibilityofapet-ownertoensure,as faraspossible, that
herpetiswell-socializedanddoesnotendupharmingthirdparties,parentshavethe
responsibilitytotakestepstoensurethattheirchildrendonotgrowuptowrongor
harmothers.Arguably,ifanagentisunderanobligationtoactinawaythatdoesnot
wrongorharmothers,theycannothaveclaimsoffairnessagainstthebeneficiariesfor
153
havingsuccessfullyavoidedharmingorwrongingthem.192
As regards parents’ intentions to raise their children to be successful, productive
adults,itdoesnotseemcorrecttosaythatthisoutcomeisthatwhichbenefitstherest
ofsociety.Itiscertainlyakeyingredientofwhatendsupbenefittingnon-parents,but
it doesnot seem tobe it just yet.The society-widebenefitswhichnon-parents free
rideonseemtolieastepbeyondthosethatarepursuedbyparents.Wecanseethis
by considering the fact that if society were organized differently, for instance if
welfare schemes were not in place, or if non-parents were ineligible for welfare
benefits funded by new generations, non-parents would not benefit from the new
generations of productive citizens like they do now. Grown children can also leave
their home countries as soon as they reach adulthood and fail to contribute to the
collectivebenefitsthatnon-parentsaresaidtofreerideon.Grownchildrencanalso
individually or collectively decline to continue important projects that the previous
generations of non-parents hold dear. It is possible, then, for the outcomes that
parentsaimattobeachievedwithoutalsoachievingtheoutcomesthatnon-parents
free ride on. This shows that raising successful adults is not the relevant Fair Play
outcomeforwhichtheycouldclaimfairness-basedsupport.
5.2.ChildrenasmorallyrequiredgoodsSo far, I have been treating the goods that parents create by having and rearing
childrenascasesofmorallyoptionalgoods,andIhavearguedthatparentscannotbe
ascribedtheFRP,andhencedonothaveFairPlayclaimsagainstnon-parents.Yetwe
mightthinkthatsomeofthegoodsthatchildrenrepresentorproduce,andthatareat
thecenterof ‘childrenaspublicgoods’arguments,aremorallyrequiredrather than192Isaymoreaboutthissortofargumentinthenextchapter,whenIconsidertheFairPlayclaimsthatparentsmightraiseagainsttheirchildrenforhavingbenefitedthembygivingthemthesortofupbringingtheyareowed.
154
optional.Beforepresentingsomeargumentsthathavebeenmadetothiseffect,Iwill
note that themorally required nature of the goodswould not, by itself, change the
verdictregardingparents’claimsontheSharedPreferenceView.Thisisbecausethe
intentionalityconditionisstillinplace.Regardlessofwhetherthegoodsatstakeare
morally optional or morally required, on the Shared Preference View those who
contribute to their production only have Fair Play claims if they pass the
counterfactual intentionality test. As I have noted, passing the counterfactual
intentionalitytestinvolvesseeingtheproductionoftherelevantoutcome(thatis,that
onwhich third parties are said to unfairly free ride) as one’s necessary reason for
bearingtherelevantcosts.If,asIhavearguedintheprevioussection,parentsdonot
pass this test with respect to the relevant outcomes that non-parents are said to
benefit from, it does not matter, on the Shared Preference View, whether we
categorize those outcomes as optional ormorally required. As such, acknowledging
that(someof)thegoodscreatedbyhavingandraisingchildrenaremorallyrequired
ratherthanoptionaldoesnotdomuchtochangetheverdictregardingparents’Fair
PlayclaimsontheSharedPreferenceView.
Nevertheless,itisworthconsidering‘childrenasmorallyrequiredgoods’arguments
closely because theydo seem to put somepressure on the Shared PreferenceView
itself, by putting renewed pressure on the intentionality condition. For, intuitively,
thereseemstobesomethingwrongwhenonlysomepeoplebearthecoststhatresult
in discharging a collective duty that others share. More precisely, there may be a
strongintuitionthatexhibitingtheintentionrequiredbytheSharedPreferenceView
isirrelevantfordeterminingcontributors’ fairnessclaimsagainstthosewhobearno
costsatallforproducingtheoutcomedemandedbythecollectiveduty.Inthissection
Iwillconsiderwhatsupportsthisintuition.Iwilltakethisasanopportunitytodefend
theSharedPreferenceView,andmorespecificallytheintentionalitycondition,inthe
faceofthechallengesitfaces.
155
First,letusconsiderthewaysinwhichparentsmightbeproducingmorallyrequired
goods by having and raising children. This would be the case if we were under a
collectivedutytohavechildren.Thereasonforsuchadutymightbethatsomeofthe
central benefits that children grow up to produce (like a minimally functioning
economy, goods that satisfy basic needs, or upholding just institutions) aremorally
requiredbyjusticeorperhapsbybeneficence.AncaGheausarguesthatageneration
facing extinction would suffer grave material harms due to the breakdown of
economicandpolitical systems,aswell asenduregreatpsychologicalburdens from
knowingthattheywillbeamongthelasthumansevertolive.193Ifso,eachgeneration
mighthaveadutytohavechildreninordertoavoidthesevereharmsofbeingthelast
generation.Alternatively,wecouldthinkthatcreatingnewpersons isof impersonal
value to theworld (so longas theyhavea lifeworth living), or thatpreservingour
species,ourcommunity,orourcultureisofgreatimpersonalmoralvalue.Thiscould
ground a collective duty to have and raise at least enough children to avoid the
disappearance of our species, our community, or our way of life.194We could even
think,withArneson,thatthereisacollectiveduty“tobefruitfulandmultiply”which
parentscanbesaidtodischarge.195
Therolethatthecollectivedutytohavechildrenismeanttoplaywithintheparental
justicedebateistoblocktheconclusionthat,becausechildreninstandardcasesare
theresultoftheirparents´choice,theparentsoughttobearalltheresponsibilityfor
givingthemwhattheyareowed.Thethought is thatwhatevertheconditionsunder
which holding someone responsible for their choices are, surely these conditions
should recognize the purported unfairness of holding someone responsible for the
consequences of discharging a moral duty that others share.196So the ‘children as
morallyrequiredgoods’argumentsseektoshowthatwhilehavingchildrenmaybea
193Gheaus2015,pp.91–95.194SeeSmilanski1995,p.46.195Arneson2014,p.9,p.20.196ForthisgeneralpointseeStemplowska2009.
156
choiceforthestandardparent,itisoneforwhichitwouldbeunfairtoleaveparents
tobearallthecosts.
Iamnotconvincedthatthereexistsacollectivedutytoprocreate,butforthesakeof
argumentIwillassumethatatleastoneoftheargumentsaccordingtowhichitdoesis
true. “Children asmorally required goods” argumentsput renewedpressureon the
intentionalityconditionofFairPlayand,ineffect,forceustoreckonwiththequestion
of whether this condition really is necessary for producers to have fairness claims
whenthecosts theybearhelpdischargeaduty thatothersshare.Tostart, consider
thiscase.
GreenTownv.FitnessTown.
InGreenTown,amajorityofpeopleoptforcyclingastheirmainmodeoftransportin
order toensuretheykeeptheirCO2emissions low.Theysucceed inkeepingtheair
pollution in their town below the dangerous threshold. Call these people Green
Cyclists.InFitnessTown,amajorityofpeople(FitnessCyclists)alsochoosecyclingas
theirmainmodeoftransport,withthesamepositiveeffectonairquality.ButFitness
Cyclistsdoitbecausethisistheirfavoritewayofstayinginshape.
I am assuming that the benefit provided by both kinds of cyclists here is morally
required: they are keeping pollution levels below the threshold where it would
threaten the health of the town’s inhabitants. It seems clear enough that theGreen
Cyclistscouldclaimthatthecardriversintheirtownshouldhelptobearsomeofthe
costsoftheircyclingbecausetheyaredischargingadutythatcardriversfallunderas
well:thatofavoidingdangerousairpollution.CantheFitnessCyclistsclaimthesame?
I take it that, intuitively, many people would be inclined to answer positively, for
Fitness Cyclists, too, are producing a morally required benefit. It is helpful to
disambiguatetwoelementsthatwouldseemtosupportthisintuition.Thefirstisthat,
on someviewsabout thedeontic relevanceof intentions, theagent’s intentiondoes
157
notmatter for establishing themoralpermissibilityof their action.ThomasScanlon
provides a recent defence of such a view.197If correct, this view could support an
argument according to which the Fitness Cyclists are discharging a collective duty
regardless of their intentions, and that, further, discharging this duty should entitle
themtohavethecostsofdoingsosharedbythosewhoalsofallunderit.
The second element that supports the intuition that Fitness Cyclists have fairness
claimsfordischargingacollectivedutycanbeunderstoodasaworryaboutdouble-
counting. IfwedidnotconsidertheFitnessCycliststobeamongthecontributorsto
dischargingtheenvironmentaldutyinsomerelevantsensealready,thenwewouldbe
entitled to ask them to do their part if an environmental scheme needed to be
institutedatsomepoint.AschemecouldbeneededifthenumberofFitnessCyclists
shrunktothepointwherethetownfaceddangerouspollutionifothermeasureswere
not implemented. The Fitness Cyclists could be asked to do their part in this new
collectiveduty-discharging scheme.But they could complain that theywerealready
doingtheirshareforavoidingairpollution,andthataskingthemtodomorewouldbe
askingthemtopaytwice.Iwilladdressthesetwoworriesinturn.
5.2.1.ThechallengefromirrelevantintentionsLetmestartwith the firstconcern.Thechallengehere is that therearesomeviews
according to which an agent counts as acting permissibly, or indeed as fulfilling a
morallyrequiredact,bymerelyperformingtheactthatispermissibleorrequiredon
somedescriptionandregardlessoftheintentionwithwhichitisperformed.Ifso,then
bearingthesortsofcostsrequiredbyacollectivedutyissufficientforonetocountas
doinghersharefordischargingthatcollectiveduty.ItseemsthenthatFitnessCyclists,
asindividualswhoare“actinginaccordancewithaduty”shouldhaveFairPlayclaims
197Scanlon2008.
158
even though they might fail the intentionality condition and the motivational
conditionrequiredbytheSharedPreferenceView.
In response, thereare twopotential lineswecould take in supportof theview that
FitnessCyclistslackFairPlayclaims.OneistosaythatinaplacelikeFitnessTownthe
fact that there are enoughFitnessCyclists aroundwhobear the costs of cycling for
theirown,environment-unrelatedpurposes,thecollectivedutytoprotectorimprove
airqualityneverarises.198Thesecondlinewecantakeistoacknowledgethatthereis
acollectivedutytoprotectairquality in thebackground,but thatcertain intentions
arerequiredinordertocountashavingdischargeditinawaythatisrelevantforFair
Playpurposes.
Letusstartwiththefirstkindofreply.IfthereareenoughFitnessCyclistsaroundthat
airquality isneverunderthreat,wecouldclaimthatthereisnocollectivedutythat
theycanbetakentodischarge.Itisastrokeofluckforthecardriversthattheylivein
atownwheresomanypeopleenjoycyclingforitsownsake.Iftherewerenot,then
presumably air qualitywould suffer and some collective solutionwould have to be
found.Butsincethereareenoughpeoplearoundwhofindcyclingtobeanenjoyable
activity that awards them benefits such as improved fitness, the environmental
problem never arises, so the collective duty to address the environmental problem
never arises either. The cyclists, then, cannot raise Fair Play claims for having
dischargedacollectiveduty.
Today’sparentsareinasimilarposition,forhumanityisnotjeopardizedbyimminent
extinction thanks to the fact that so many people prefer to have children anyway.
Gheaus argues that there is a collective duty to have at least as many children as
wouldbenecessarytostaveoffthebasicneedsdeprivationsassociatedwithbeingthe
lastgenerationtoeverlive.199Oneanswertothisproposal,then,isthatthiscollective
198IthankAndrewWilliamsforsuggestingthispointtome.199Notethatevenifthisargumentwentthrough,itwouldnotgroundFairPlayclaimstoallparents,buttoalimitednumberofthem.Thenumberofchildrennecessarytostaveoffbasicneedsdeprivation
159
dutysimplydoesnotexistinsocietytodaybecausethethreatofextinctionitselfdoes
not exist.Millions of people around theworld prefer to have children for personal
reasonsthatareunrelatedtospeciespreservationortotheavoidanceofbasicneeds
deprivationassociatedwiththeimminentthreatofextinction.
DenyingtheFitnessCyclists´fairnessclaimsonthebasisthatthereisnothreatofair
pollution in the first place, and hence no duty to be discharged, has a lot of initial
plausibility.However,wecanseethelimitsofthisresponseifweconsideravariation
oftheGreenTownv.FitnessTownCase.Inthisversion,therearenotenoughFitness
CyclistsinFitnessTown,norenoughGreenCyclistsinGreenTown,toavoidtheriskof
collective harm entirely. The collective duty to avoid dangerous air pollution thus
kicks in,andourmainquestionre-emerges:oughtwetoregardtheFitnessCyclists,
howeverfewofthemtheremightbe,ascontributorswithFairPlayclaims,justaswe
regardGreenCycliststobe?
Herewemustexplore thesecondreplywhichdenies thatFitnessCyclistshaveFair
Playclaims.Takingthislineinvolvesacknowledgingthatthecollectivedutytoavoid
air pollution exists regardless of whether pollution is an imminent threat, and
showing that in order to count as discharging this duty in a relevant sense certain
intentionsarerequiredonthepartofthecontributors.
I submit that the question of whether Fitness Cyclists have fairness claims for
dischargingadutythatotherssharecannotbesettledbythegeneraldiscussionabout
the deontic relevance of intentions. That is, even if a view like Scanlon’s is correct
aboutthefactthatintentionsarenotrelevantforjudgingtherightnessorwrongness
ofanact, itisafurtherquestionwhethertheyarerelevantforestablishingFairPlay
claims.Scanlonarguesthatwhetherornotanactismorallypermissibleisdetermined
solelybythebalanceofmoralconsiderationsinfavororagainstperformingthatactin
associatedwithimminentextinctionissurelymuchlowerthanthenumberofchildrenbeingcreatedbyeachgenerationofadultstoday.
160
those conditions. Of course, the moral considerations that may make an act
permissiblemightincludeactingoutofcertainintentionsandmotives,asisthecase
foractionsthataremeanttoexpressacertainattitude,buttheyneednot.Theagent’s
intentions become relevant in themselves only if we are looking to evaluate the
agent’s character: did the agent recognize and adequately weigh those moral
considerations at the time of acting? If they did not, itmakes sense to say that the
agentdidtherightthingforthewrongreasons,andthattheirreasonsreflectpoorly
ontheircharacter.200
Iamopentotheviewthatone’sintentionsarenotultimatelyrelevantforestablishing
whether someone has discharged a duty, if by this we only mean that they have
performedanaction thatundersomedescription ismorally required.But, I submit,
whether theyhavedischarged theduty in a sense that is relevant for FairPlay is a
furtherquestionthatcanonlybeansweredfullybygivingafull-blownaccountofthe
conditions under which free riding is unfair, as I have tried to do in the previous
chapter.OntheviewIhaveproposed,whileitmaybepossibletodischargeadutyfor
thewrongreasons,itisnotpossibletohaveFairPlayclaimsforhavingdischargedif
onedidsoforthewrongreasons.
Tobeclear,whattheSharedPreferenceViewofFairPlayrequiresofagentsisnotthat
theyactoutofasenseofdutyassuch.ActingoutofdutyiscompatiblewiththeShared
PreferenceView,butnotrequired.Theintentionthatagentsmustexhibit istobring
about the outcome that is required by the collective duty in away that passes the
counterfactualintentionalitytest,andtheirmotiveforwantingtobringitaboutmust
passthemotivationalcounterfactualtest.Thereasonwhythismotivationalstructure
isrelevantforFairPlayisthatitisrelevantfordeterminingwhethertheagentshave
incurredtherelevantsacrificethatgeneratesafairnesscomplaintoverothers.
200Scanlon2008,pp.22-27.
161
Consider the followingpointmadebyStemplowska,whoargues that theconditions
underwhichsomeoneshouldbeheldresponsibleforbearingalltheconsequencesof
theirchoicesshouldreflectwhetherthechoiceinquestiondischargedasharedduty.
[L]etusassumethatallmembersofagivensocietyhaveadutytokeeppollutionlevels
belowa certain limit (…).And letus assume that, because some ignore their shareof
this collective duty, others have to carry heavier burdens than theywould otherwise
have tocarry inorder tomakesure that theduty isbeing fulfilled. (…) Inperforming
whatothersshoulddobutdonot,thedutifulpeopleareinfactexpandingtheoptionsof
thenon-dutifulones–allowingthemtospendtimeandresourcesonotherthingsthan
thefulfilmentoftheirduties.Butthenevenifoneisresponsibleforfindingoneselfata
disadvantageduetoone’schoicetofulfilashareddutyinthefaceofothersslackingoff,
oneisalsoentitledtohavethisdisadvantageremoved.201
Note that the Shared Preference View describes the unfairness of the situation in
muchthesametermsasStemplowska:itisunfairifsomepeople’soptionsforhowto
spend their time and resources are expanded by the fact that they need notworry
aboutinvestingsomeofthemintothefulfilmentofacollectivedutythankstoothers’
havingfulfilledit.However,thecrucialpointofcontentionisonethatStemplowska’s
remarks are not clear enough about: do people count as fulfilling a collective duty
regardlessoftheirintentions?MysuspicionisthatStemplowska’sargumentmightbe
assuming what I have been arguing for: that the relevant loss of “options”, or the
relevantsacrifice,invirtueofwhichcontributorsacquireafairnesscomplaintagainst
othersdependsontheirintentionsandmotives.Itdependsonwhethertheybearthe
duty-fulfillingcostsinordertobringabouttheoutcomerequiredbytheduty(whether
ornottheydothisoutofasenseofdutyoraself-interestedmotive),asopposedto
bearingthesecostsasamatterofpursuingtheirpreferredwayoflife.Ifthelatteris
the case, then the contributors do not seem to lack an option that only free riders
enjoy.
201Stemplowska2009,p.244.
162
So, going back to the Green Town v. Fitness Town case, the Fitness Cyclists do not
seemtobe“pickinguptheslackinordertofulfiltheduty,”asStemplowskasays,but
ratherareonlypursuingtheirlifeplan,asarethefreeriders.Similarly,typicalparents
do not decide to procreate and raise children in order to ensure that we avoid
extinction, for instance.Assuch, theycannotbesaidtobedischargingthecollective
duty to procreate and raise children, if there is one, in a way that gives them a
fairness-basedcomplaintagainstthosewhodecidenottohavechildren.
5.2.2.Thechallengefromdouble-countingSofar,Ihopetohaveshownthatwehavenoreasontobelievethatsomeonecanhave
Fair Play claims merely in virtue of having borne the sorts of costs required by a
collective duty, without exhibiting the relevant intentions. There is one lingering
worry that this view seems to face, however, and it is the worry about double-
counting.IfwedonotrecognizetheFitnessCyclistsascontributorsdischargingtheir
shareofthecollectiveduty,thenperhapswecouldbejustifiedinaskingthemtomake
somefurthereffortinordertopromotetheenvironmentalgoal.Butsurelythiswould
beimplausible.
IagreethatitwouldbeunfairtoaskpeopleliketheFitnessCycliststodomorethan
theyarealreadydoing.ButdoestheSharedPreferenceViewcommitusto,effectively,
penalizepeople like theFitnessCyclists by asking them topay their sharewith the
right intentions, regardless of the fact that they were already acting in accordance
withthedutyforotherreasons?
IdonotbelievethattheSharedPreferenceViewcommitsustosuchaconclusion.The
Shared Preference View commits us to denying that the Fitness Cyclists are on the
same footingwith the Green Cyclists as regards the Fair Play claims they can raise
againstfreeriderswhocontinuetopolluteasusual.IncontrasttotheGreenCyclists,
163
they cannot claim to have restricted themselves in order to help discharge the
environmental duty, and thus cannot complain that the free riders are wrongfully
enjoyingaprivilegeattheirexpense.ButthisneednotentailthattheFitnessCyclists
canbeforcedtobearfurthercoststhemselves.
IntheprevioussectionIsaidthatifthereareenoughFitnessCyclistsinFitnessTown
that air quality is never at risk of dropping below the critical threshold, thosewho
prefertodriveeverywherearelucky;thenatureoftheambitionsoftheirpeersmeans
thattheycanavoidrestrictinghowmuchtheydrive.Butif,conversely,therearenot
enough Fitness Cyclists in town, and everyone else has to restrict themselves by
bearing the sorts of costs that Fitness Cyclists are already bearing, it is the cyclists
whogettoenjoythegoodluckoftheirambitions’beinginlinewiththerequirements
ofduty.Thefactthattheirambitionsaresuchthattheyarealreadybearingthesorts
of costs thateveryone isnowrequired tobear isa formofgood luck,buta formof
good luck that those who are now forced to restrict their driving do not have a
complaintagainst.
5.3.ChildrenassocializedgoodsIn the Children as Public Goods section of this chapter I claimed that parents lack
fairnessclaimsoverthecollectivegoodstheirchildrenrepresent,orproduce,because
the outcomes they produce intentionally in the sense required by the Shared
PreferenceViewarenotthesameonesthatnon-parentsareaccusedoffreeridingon.
Thebeneficialoutcomesthatnon-parentsfreerideon,Iclaimed,lieastepbeyondthe
outcomesthatparentsintentionallyproduce.
However,itisusuallythecasethatwelfarestatesdeliberatelymakethatextrastepof
turning the outcome that parents aim at into a universally beneficial outcome.
164
Institutions of welfare are designed so as to ensure that the existence of a new
generationofadultsisharnessedforthebenefitofparentsandnon-parentsalike.The
new adults are obligated to pay taxes, and their tax money is then invested into
supporting various public schemes which aim to support both parents and non-
parents.
This is the key insight of Olsaretti’s ‘children as socialized goods’ argument.202She
points out that some of the central benefits at the heart of the ‘children as public
goods’ are, in fact, excludable but that welfare states intentionally extend these
benefits to non-parents as well as parents. Welfare schemes could, and may
permissibly,besetupdifferentlythantheyarecurrently,suchthatnon-parentscould
beineligibletoreceivewelfarebenefitsfundedbythenewgenerations.Non-parents,
then, would be forced to make their own provisions for their pensions, disability
support and so on. But since welfare states currently do harness and socialize the
benefits of having new generations of productive citizens, Olsaretti argues, states
should also socialize the costs of creating this valuable human resource by forcing
non-parentstosharethecostsofchildren.203
A second important insight of the “children as socialized goods argument” is that
manyofthebenefitsthatnon-parentsinternalizearenotonlyexcludableintheway
justexplained,butarealsoavoidable(bythepotentialfreeriders)andrivalinnature,
which means that if some people get to enjoy them this diminishes the amount
available for others to enjoy. Non-parents could choose to avoid receiving welfare
benefitsthatarefundedbynewgenerations:theycouldchoosetoinvestintoaprivate
pension fund and forgo publicly funded pensions, for instance, and they could
purchaseprivate insurance forall themisfortunes theyareworriedaboutsuffering.
The benefits of a welfare system are also rival in nature, given that if non-parents
receivevariouswelfarebenefitsfromthesametaxpoolasparents,thisleavesfewer
202Olsaretti2013.203Olsaretti2013,pp.253-254.
165
funds to go around for parents. Surely, Olsaretti suggests, these features help to
bolsterparents´fairnesscaseagainstnon-parents.Foritdoesnotseemfairtoletnon-
parents internalize avoidable benefits that they did not pay to produce andwhich,
onceinternalizedbythem,wouldleaveparentsworseoffthantheywouldhavebeen
ifnon-parentshadnotinternalizedthem.
The idea that non-parents’ free riding leaves parents worse off than they would
otherwisebe isof importancetoOlsaretti’sargument,sincethis issurelyat leastan
aggravatingfactor,asitwere,onanyplausibleFairPlayview,ifnotatthecoreofwhat
makesfreeridingwrongful.204Shepointsoutanotherwayinwhichnon-parents’free
ridingmakes parentsworse off, beyond the fact that non-parents’ enjoyingwelfare
benefitsleaveslesstaxfundstogoaroundforparents.Asecondwayinwhichparents
aremadeworseoff,accordingtoOlsaretti,isthatthecostsofchildrearingarehigher
as a result of non-parents’ free riding. Raising a child to be successful in today’s
society has becomemore andmore costly because being successfulmeans, in part,
beingproductiveintoday’seconomy,whichistosaycreatingvalueforthebenefitof
othersaswell. Insofarasparents canbe said tobeargreaterburdensasa resultof
raising theirchildren inways thatarebeneficial toothers,Olsaretticlaims, theyare
worseoffrelativetoasituationinwhichothersdidnotbenefitfromtheirchildrearing
activities,andmayhaveFairPlayclaimsagainstthebeneficiaries.
Thus,thequestionthatwemustconfrontinthissectioniswhatrelevance,ifany,does
it have for Fair Play that states deliberately take that “extra step” of socializing the
benefits of having a new generation of productive adults, benefits which are
excludable,rival,andavoidable?
Iarguethatthesefactsdonothelptoestablishparents´FairPlayclaimsbecausethey
refertobenefitsbeyondthoseintentionallycreatedbyparentsinthesenserequired
204OntheSharedPreferenceViewtheunfairnessoffreeridingdoesnotcruciallydependonwhetheritmakescontributorsworseoff.
166
bytheSharedPreferenceView.Sincetherelevantsocializedbenefitsliebeyondthose
properlyintendedbyparents,the“childrenassocializedgoods”argumentcouldonly
succeedingroundingparents’FairPlayclaimsifwereliedonthenotionthatbenefits
producersareentitledtoallthefruitsoftheirlabour,includingthosethatareineffect
a positive externality of their efforts. In the realm of parental justice, this would
requireustoaccepttheideathatgrownchildren,andtheresultsoftheirproductivity,
are their parents’ property in some sense, an idea that Olsaretti herself is explicit
aboutwantingtoavoid.205
Now,evenwhenitcomestobenefitsintentionallyproducedinthesenserelevantfor
theSharedPreferenceView,someofthefeaturesofthesebenefitsthatthe‘childrenas
socializedgoods’argumenthighlightsarenotinthemselvesrelevantforFairPlayon
theSharedPreferenceView,whileothersare.Excludabilityisoneofthefeaturesthat
is not relevant on the Shared Preference View. Suppose some people organize a
concertwheretheypublicizethefactthattheconcert-goersareexpectedtopayafee,
but for some reason there are no ticket checks at the entrance. On the Shared
PreferenceView,whoeverwillinglyandknowinglyenterstheconcerthasaFairPlay
obligationtopayforaticket,providedthattheotherconditionsforFairPlayobtain,
and despite the fact that this is an excludable benefit from which they were not
actuallyexcluded.
Other features highlighted by the ‘children as socialized goods’ view, such as
avoidability,are,indeed,relevantforFairPlay.TheSharedPreferenceViewcaptures
therelevanceofavoidabilityinsofarasvoluntarilyseekingortakingabenefitthatone
couldhaveeasilyavoidedisevidenceofthefactthatthefreeridersubjectivelyvalues
the good in a way that is required for them to be attributed the Free Rider´s
Preference.Thatis,afreeriderwhocouldhaveavoidedabenefitbutdecidedtotake
itanywaycouldnotcomplainthattheydonotvalueitorthattheyregarditasbeing
imposed against their will. In the same concert example, willingly and knowingly205Olsaretti2013,p.241.
167
goingintoseetheconcertwithoutpayingfortheticketgeneratesanobligationtopay
forit.Andthisisregardlessofthefactthatonecouldhavebeenexcludedbutwasnot.
TherivalnatureofthebenefitsatstakeisalsorelevantontheSharedPreferenceView
ofFairPlay.IhavenotmentionedrivalrysofarbecausetheSharedPreferenceView
asdevelopedinthepreviouschapteridentifiesthewrongnessoffreeridinginaway
thatdoesnotdependonfreeridingbeingharmfultothecontributors.Thatis,onthe
SharedPreferenceView,freeridingcanbewrongevenifitdoesnotstrictlyspeaking
maketheproducersworseoff.Thatbeingsaid,whenfreeridingisnotonlyunfair in
the way picked out by the Shared Preference View, but also harmful, Fair Play
obligationsarearguablyevenmorestringent.Theproducerscanbemadeworseoffif
freeridingcausesthemtobeargreaterbenefits-producingcoststhantheyotherwise
would.Forexample,intheFlatSharecase,thefactthatthefreerideralsomakesthe
two flatmatesworse off by forcing them to pick up the slack, and therefore bear a
larger shareof cleaningchores than theyotherwisewould, arguablymakes the free
rider´sobligationsevenmorestringentcomparedtoacasewherefreeridingdidnot
increasethecontributors’burden.Theproducerscanalsobemadeworseoffbyfree
riding in cases where the properly intended goods are rival. Suppose that the
flatmatesorganizedapotluckwhereonly twoof thembrought foodwhile the third
ate a portion of it without contributing anything. Again, even though the Shared
PreferenceView fundamentally condemns free riding for different reasons, the free
rider´s obligations would be even more stringent in virtue of the fact that she
diminished the amount of food available for those who had done their part for
producingit.
So,ingeneral,itisrighttoconsiderfeaturessuchasavoidabilityofthebenefit,andthe
variousways inwhichproducersmightbemadeworseoffbyothers’ freeriding,as
relevant for determining Fair Play claims and obligations. However, as argued
previously,parentsdonotproperlyintendtoproducetherelevantbenefitsthatnon-
parents internalize and that exhibit these features. As I tried to show in the first
168
section of this chapter, the benefits that non-parents are said to internalize, and
especially the economic benefits that the “children as socialized goods” argument
focuseson,liebeyondparents´necessaryaimsforaction.Producinggoodssuchasan
enlarged taxbase,avibranteconomy,or fundingparticularwelfarebenefitsarenot
the things that parents aim at when they make their procreative and childrearing
choices.Assuch,bybearingthecostsofchildrentheycanbesaidtopursuetheirown
ambitions, as opposed to frustrating them in a way that would generate Fair Play
claims.
ThefeatureshighlightedbyOlsaretti,then,areexhibitedbybenefitsthatcomeabout
through the further workings of the state and which lie beyond parents’ properly
intendedoutcomes.Thus,itisnotclearthattheyshouldmakeadifferencetoparents’
Fair Play claims. They certainly do notmake a difference on the SharedPreference
View,whichonlycondemnsfreeridingongoodsthatareproperlyintended.Ifthisis
correct, the avenue left for parents to complain against the state’s socializing the
results of their children’s productivity is to say that the state is socializingwhat is
rightfully theirs, which seems like an implausible view of what rightly belongs to
them.Wewouldhavetorelyontheideathattheyhavearighttoallthefruitsoftheir
labour,includingthosethatmakenodifferencetotheirwillingnesstobearthecosts
they are bearing. This view seems implausible, or at least would call for elaborate
support. In the absence of such support, it seems more intuitive to think that the
positiveexternalitiesofpeople’sactivitiesshouldbeconsideredakintomannafrom
heaven,ratherthanthefruitsofone’slabourtowhichonehasaclaim.
Olsarettianticipatesthisobjectionandsuggeststhatclaiming“thatthesebenefitsare
onaparwithmanna fromheavenoverlookshalfof the truthof thematter,namely,
that parents have borne burdens that were necessary for the production of this
‘manna.’”206Indeed,parentsbearcertaincoststhatarenecessaryfortheproductionof
such benefits as an enlarged tax base, but notbecause they are necessary for it. As206Olsaretti2013,p.257.
169
such, it is a cogent question whether we can treat these benefits as manna from
heaven,eventhoughofcoursetheyareinonerespectdifferentfromactualwindfall:
somepeoplebearsomecoststhat(predictably)resultintheircreation.
It seems that, on one side of the equation, we have benefits that are intentionally
producedinthesenserequiredbytheSharedPreferenceView:producersbearcosts
becausetheyarenecessarytocreatecertainbenefits.Thesebenefitsclearlybelongto
the producers in some relevant sense, and socializing themwithout also socializing
theircostswouldbeunfairtotheproducers.Ontheothersideoftheequation,there
are benefits that are produced unintentionally, so much so that they come as a
surprise to their producers. These may be treated as manna from heaven and
socializing them would not wrong the producers. The question is where, on the
spectrum between these two kinds of benefits-production, we should place those
caseswhere theproducersknowandexpect thatsomecollectivebenefitswill come
aboutasaresultoftheiractivities(eitherspontaneouslyorthroughtheworkingsofa
third party like the state), butwhich do notmotivate their willingness to bear the
coststheyarebearing?Theanswercouldonlybeprovidedbyafullyworkedoutview
of producer’s entitlement. A view that was friendly to parents’ complaints against
non-parentswouldhave toclaim thatparents’entitlementsextendedas faras their
adult children’s contributions to the welfare state. We might consider this an
implausible view of how far our producer’s entitlement reaches. But, moreover, in
parents’ case it would have an added layer of implausibility given the unwelcome
thought that childrenare in some important sense theirparents’property, and that
parents can claim some entitlements in virtue of their adult children’s productive
participationinsociety.
Allthisbeingsaid,thereisaparticularfeaturethatthe“childrenassocializedgoods”
argumenthighlightswhichmayyetseemrelevantforestablishingparents’FairPlay
claims,namelythefact,ifitisafact,thatparentsbeargreaterchildrearingburdensas
a result ofnon-parents’ free riding.Even ifweaccept thatparents cannot complain
170
that theyaremadeworseoffby thestate’s socializing theresultsof theirchildren’s
productivity, theymay have a complaint if they aremadeworse off in this second
sense. If they are forced to bear greater burdens as a result of other people’s free
riding, theycancomplainthat it isnottruethatbyrearingchildrentheyaremerely
pursuingtheirusualambitionsandbearingcoststheywouldhaveborneanyway.For
thecoststheymustnowbearinpursuitoftheirambitionsareinflatedasaresultof
these ambitions’ creating benefits for third parties. Olsaretti makes the following
argument,whichisworthconsideringinsomedetail.
Someofthecostsofparentingaresociallycreated,andtheyarecreatedasaby-product
oftheactivity’sbeingbeneficialforthirdparties.Take,forexample,averyconsiderable
cost of children, the expense to provide themwith adequate education. This cost has
been rising steadily, in line with social, economic, and technological changes that
require children tobe in the education system formanymoreyears inorder tohave
goodprospectsasadultmembersof theeconomy.The increasedcosts forparents (in
termsof education costs andmaintenance costs for childrenwhopostponebecoming
financiallyself-sufficientuntilmuchlaterthantheyusedto)aretothebenefitnotonly
of their children, but of all citizens who benefit from a better-educated and skilled
workforce. If someone’sbenefiting fromothers’ activitiesmakes thoseactivitiesmore
costly for theseothers, thismaymake adifference towhether thebeneficiaries incur
obligatonsoffairnessasaresultofbenefitingfromthoseactivities.”207
Thecaseforparents’FairPlayclaimswouldbestrengthenedif itweretruethatthe
costsofchildrearingwerehigherasaresultofnon-parents’freeriding.Iwouldargue,
however, that the increasedcostsofchildrearingof thesort thatOlsarettihighlights
are not the result of free riding. Rather, they can be attributed to the forces of the
marketeconomyandtotheevolutionofthesortsofskillsthathavebecomenecessary
for one to occupy desirable positions in society. In a society in which valuable
positionsrequireahighlevelofeducation,andoftenevensophisticatedtechnological
skills, itisnowonderthateducationhasbecomeapositionalgoodtosuchanextent207Olsaretti2013,p.244.
171
thatwecanspeakof a so-callededucationalarms race.208The fact thatnon-parents
benefitfromtheproductivityofnewgenerationsdoesnotseemtoplayacausalrolein
theprocessofincreasingthecostsofchildrearing.Wecanseethisbyconsideringthe
fact that these market forces would very likely still produce many of these same
effectsevenifoursocietieshadsplitwelfaresystemsinwhichonlyparentsbenefited
fromtheirgrownchildren’staxes.Wehaveeveryreasontothinkthatdevelopments
suchashistoricallyhigheducationalcostsandeducationalarmsraceswouldstillexist
in societies that were like ours in every other respect except for their having
eliminatednon-parents’freeridingonnewgenerations.Sowhilesuchincreasedcosts
nodoubtindirectlybenefitnon-parents,itdoesnotseemcorrecttocharacterizethose
highercostsastheresultoftheirbenefiting.
5.4.ConclusionInthischapterIhavereviewedallthemainversionsofthe‘childrenaspublicgoods’
arguments and I have argued that, on what I have proposed is the most plausible
principle of FairPlay, they fail to establish an entitlementon thepart of parents to
havingthecostsofchildrensharedbytherestofsociety.Thereasonfor this is that
parentsdonotexhibittherequisiteintentions,which,ontheSharedPreferenceView,
shows that theyarenot incurring therelevantsortof “sacrifice”orcost invirtueof
whichtypicalpublicbenefitsproducerscanlayclaimsoffairnessagainstthirdparties.
In the next chapter I turn to a closely related, yet importantly different, attempt at
groundingparents’claims.ThisistheviewthatparentsmayhaveclaimsofFairPlay
againsttheir(grown)children,sincetheirchildrenarethemainbeneficiariesoftheir
efforts. This sort of argumentbringsup somevery interesting issues inprocreative
ethicsaswellasissuesregardingfilialduties.Butmoreover,itisworthexploringits
merits as a Fair Play argument as such because it seems not to be plagued by the208Halliday2016.
172
intentionality challenge that the arguments reviewed in this sectionhave faced: the
benefitsinvirtueofwhichparentswouldraiseFairPlayclaimsagainsttheirchildren,
namelythegoodsofgivingtheirchildrenaminimallydecentupbringing,are indeed
theonesthattheyproperlyintendedtoproduceinthefirstplace.
173
Chapter6
AreKidsUnfairFreeRiders?
InthischapterIconsiderthepossibilitythattheanswertothequestionofwhoshould
pay for the costs of children might include children themselves. According to one
fairness-basedview,themainbeneficiariesofparents’effortsofraisingchildrenare
thechildrenthemselves,andassuchtheymighthaveFairPlayobligationstohelppay
their share for thebenefitsof anupbringing, once theyare grownandable to.This
kindofargumentclaims thatgrownchildrencountasunfair freeridersunless they
deviseawaytohelppaytheirownway.
The most detailed version of this view to date has been developed by Patrick
Tomlin.209His view, termedKidsPay, aims to establish apro-sharing fairness-based
obligationthatallofsocietyshouldhelppayforthecostsofchildrensincealladults,
whethertheyarethemselvesparentsornon-parents,havebenefited,tosomeextent
oranother,fromtheupbringingthattheirparentsprovidedforthem.
Although the Kids Pay view was developed in the context of the parental justice
debate,itcanbereadasoffering,atthesametime,ananswertothebroaderquestion
of filial obligation: what, if anything, do grown children owe their parents? In this
wider context, the idea according to which grown children ought to repay their
parents for the upbringing they received is very familiar. According to one view of
filialobligation,childrenincuradebtinvirtueofthesacrificesthattheirparentsmade
in order to give them an upbringing.210The debt that children accrue is directly
correlated to the effortsmade by parents. Another, perhapsmore popular, account209Tomlin2015.210Post1989.
174
claims that grown childrenowe their parents gratitude for their efforts, though the
content ofwhat is owed ismuchmore open-ended. Grown children ought to show
appropriate expressions of appreciation and gratitude rather than provide a
repaymentofdebtassuch.211
AFairPlayargumentoftheKidsPayvarietywouldcomplementthispictureof filial
obligationbyofferinganotherclosely related,yetdifferent, justification forwhether
grown children owe something to their parents, and what that might be. Children
could be said to owe it to their parents not to unfairly free ride on their efforts of
givingthemanupbringing.Whatthisobligationwouldrequireinsubstantivetermsis
that grown children ought to share the costs of their own upbringing with their
parents.
InthefirstsectionIlayoutPatrickTomlin’sKidsPayviewandIofferreasonstodoubt
itssuccess.InthesecondsectionIarguethatthereisaverylimitedargumentthatcan
bemadeinsupportofparents’fairnessclaimsagainsttheirchildrenifoneadoptsthe
SharedPreferenceViewofFairPlay,which Idefend. In the third section Ihighlight
some of the implications that the Shared Preference View has for the literature on
filialobligations.
6.1.Tomlin’sKidsPayViewTheKidsPayaccountbeginsbynotingthatraisingchildrentobecomeindependent,
autonomousadultsbenefitssocietyinmorethanoneway.First,asthepreviouspro-
sharingfairness-basedaccountshaveemphasized,non-parentsstandtobenefitfrom
theworkthatparentsdobyhavingandraisingnew,productivemembersofsociety.
211Berger1975,pp.298-301;Blustein1982,partII,ch.3;Callahan1985;Wicclair1990,pp.172-173;Zola2001.Othertheoriesoffilialobligationincludethefriendshipaccount(English1979),andthespecialgoodsaccount(Keller2006).
175
But all adultswere once children themselves, and they have all benefited (to some
degreeoranother) fromthechildrearingworkdonebytheirownparents.Focusing
on this secondkind of benefit, Tomlin argues that current adults,whether they are
parentsornon-parents,couldhaveFairPlayobligationstocontributetothecostsof
raising childrenon thebasis that theyhavebenefited from receiving anupbringing
themselves.212Theformthatthisobligationshouldtakeisamatterofdebate,but,he
suggests,an intuitivelydefensibleschemewouldbetohaveallofsociety“front” the
resourcesthatchildrenneedandthussubsidizechildrearing.Then,whenthechildren
are grown, they can be asked to contribute, not by paying their own parents back
directly,butbysharingthecostsofraisingthenextgeneration.
Toevaluatethisproposalitisusefultomaintainafirmdistinctionbetweenthesetwo
parts of the Kids Pay argument: on the one hand, there is the justification for the
existenceofaFairPlayobligationonthepartofchildrentocontributetothecostsof
their own upbringing. On the other hand, there is the argument concerning one
possibleimplementationofthisobligation.Thisargumentinvolves,first,theclaimthat
itisallofsocietywhoshouldfrontthecostsofchildrentoparents,ratherthanparents
collectively sharing thecostsof childrenamongst themselves.Second, theargument
about implementation involvesacommitment to thedirectionalityof theobligation,
namely that grown children should help pay for the next generation rather than
benefittheirownparentsinoldage,forinstance.Ofcourse,thetwopartsarerelated,
for the content of the Fair Play obligation and the direction in which it should be
dischargedpartlydependontheunderlying justification.Still, thesetwoaspectscan
andshouldbekeptseparate.Formypurposesitisnotnecessarytodiscussthedetails
of Tomlin’s scheme of discharging children’s alleged Fair Play obligations. In what
follows I focus on the justificatory component of his argument and I show that the
KidsPayviewfailsatthisstage.
212Alstottalsomakesthispointbutdoesnotpursueitasanindependentpro-sharingargument,distinctfromtheautonomy-basedcaseshemounts(Alstott2004,p.51).
176
6.1.1.KidsPay:thegroundsforaFairPlayobligationToestablishaFairPlayobligationonthepartofchildren,Tomlinadoptssomething
like Klosko’s Fair Play view according to which receiving presumptively beneficial
goods in certain conditions can generate Fair Play obligations. He argues that
receivingpresumptivegoods(i.e.goodsthatarenecessaryforaminimallydecentlife)
generatesaFairPlayobligationtohelppayfortheprovisionofthosegoodswhenthe
recipientisnotavailabletogiveconsent,andwhereitisnotpossibleorreasonableto
wait for the recipient to become available to give consent. Furthermore, he argues,
whentheseconditionsobtain,FairPlayobligationscanariseevenwhenthegoodsin
questionareexcludable,andwhentheyareprivateratherthanpublicinnature.213To
illustratethisviewofFairPlay,considerthiscase.
Beach-Goers214
Someone is drowning near a crowded beach. Suppose there is a duty to save the
victimthatappliestoallthebeach-goerswhoarenearbyandwhocansavethevictim
atlittlecosttothemselves.Someonedoesso,andsavesthevictim.Intheprocess,the
rescuerincurssomemildinjuriesthatrequiremedicalattention.
AccordingtoKidsPay,wecanconsidertherescuedvictimasapartyliabletosharein
the costs of her own rescue as amatter of fairness to the rescuer. For there is no
reasontothinkthatpresumptivebenefits,suchas,here,thebenefitofrescue,should
comeatnocosttothebeneficiary.Andthisisso,saysTomlin,despitethefactthatthe
benefit at stake is not public in nature, nor is it non-excludable, which are both
featuresthatareconsiderednecessaryforFairPlaytokick inonmostviewsofFair
Play.215The good in Beach-Goers is private (it can only be enjoyed by the intended
213Tomlin2015,p.669.214TheexampleisadaptedfromArneson2014,p.14.215SeeSection4.3above.
177
beneficiary,thevictimherself),anditisexcludable(therescuercouldhavedeniedit
to her). Normally, foisting such private, excludable benefits on someone and then
demandingtobepaidinreturnisimpermissible,saysTomlin.Aprincipleofconsent
toreceivingandpayingforsuchbenefitsshouldtypicallybethenormforestablishing
such obligations. However, if obtaining the beneficiary’s consent is not possible,
waitingforthebeneficiarytogainthecapacitytogiveconsentisalsonotpossible,and
the goods are presumptively beneficial, it may be permissible to foist benefits on
someone.ThisseemstobethecaseinBeach-Goers.Thevictimisnotavailabletogive
consent,norisitpossibletowaitforhertobecomeavailable,asdelayingtherescue
would presumably cause her to drown. In such cases it seems plausible to claim,
according to Tomlin, that foisting presumptive benefits is permissible and that the
beneficiaryhas anobligationof fairness to contribute to the costsofproviding that
benefit.ThevictiminBeach-Goers,then,hasafairness-basedobligationtohelpshare
the costsofherown rescue,which in this casemeanshelpingwith the costsof the
rescuer’smedicalbill.
Similarly,Tomlinargues, it ispermissible to imposepresumptivelybeneficial goods
onchildrenandthen,upontheirreachingadulthood,toaskthemtoshareinthecosts
ofprovidingthem.Hedescribestherangeofpresumptivebenefits thatparentsmay
foist on children as including, first, the goods which are essential for reaching full
agency, andsecond, thegoodswhichare “notessential for reaching full agency,but
which are essential for a minimally decent life and which can only be reasonably
providedpriortofullmoralagency.”216Takentogether,wecancallthesethegoodsof
aminimallydecentupbringing.
Of course, there is an obvious difference between the parental case and cases like
Beach-Goers.Unlike inBeach-Goers, someone is responsible for theexistenceof the
needtoprovidechildrenwithcostly,presumptivebenefits.Sowhyshouldsociety(i.e.
nowgrown-upchildrenwhodidnotasktobeborn)havetohelpshoulderthecostsof216Tomlin2015,p.669.
178
theminimallydecentupbringingtheyneededasaresultoftheirparents’decisionto
bring them into existence? It should be only parents themselves, this familiar
objectiongoes,217whoshouldpayforgivingchildrenwhattheyareowed.
Tomlin calls this the “parental responsibility objection.”218His response consists in
the following two moves. First, he argues that moral responsibility for causing
someonetobeinneeddoesnotautomaticallyentailliabilityformeetingthoseneeds.
Liabilitywill follow insomecases,butnot inothers.Second,heargues thatcausing
children to exist, thereby causing them to be needy, is not one of those cases that
attracts liability,provided that it isnot true thatparentsharmchildrenbybringing
themintoexistenceinawaythatrequirescompensation.
Thefirststepofthisresponseclaimsthatanagent’scausalresponsibilityforthefact
thatsomeone is inneedofcostlybenefitsdoesnot,by itself, imply that theagent is
liabletoprovidethosebenefits(alone).Forthissortofliabilitycouldapplytopeople
likeourrescuerinBeach-Goers.Indeed,toillustratethispoint,Tomlinmakesuseofa
case very similar to Beach-Goers, where themain difference is that, in the case he
uses, the rescue is supererogatory rather than required. This change is useful for
makingthecasemoreanalogoustoprocreation,asmostpeoplebelievethatbringing
someone into existence is not amorally required act (even if onebelieves itwould
representabenefitforthepersonbeingbroughtintoexistence),butratheramatterof
theparents’discretion.
BartsavesLisa
Lisaisdrowning,andnooneisresponsibleforherbeinginthissituation.Savingheris
sodangerousthatnooneisobligatedtodoso.Bartdivesin,nevertheless,tosaveher.
Once on the bank of the river, Lisa requires essential and expensive medical
treatment.219
217Thisnow-familiarobjectionhasbeenraisedingeneraltermsbySteiner1997.218Tomlin2015,pp.671-672.219Tomlin2015,p.671.
179
Lisa needs very expensive life-saving medical treatment once on the bank. This
situationiscausedby,oristheresultof,Bart’sdecisiontopullheroutofthewater.
Hadhenotpulledheroutof thewater, shewouldnotbe inneedof costlymedical
assistance,andnoonewouldbeobligatedtoprovidethistoher.Butthisfactdoesnot
seem to imply that Bart is on the hook for paying all her medical bills. If he does
decide topayhermedical bills, and if theFairPlayprinciple sketchedbyTomlin is
plausible,Barthas claimsof fairness against Lisa forhavingpaid forher life-saving
medicaltreatment.Theimportantpointhereisthattheseclaimsarenotundermined
byBart’sresponsibilityforcausingLisatobeinaneedysituationbybringingherto
shore.
SuchFairPlayclaimswouldbeundermined,Tomlinnotes,ifBarthadputthevictimin
thatsituationbyharmingher.
BartharmsLisa
BartintentionallyandwrongfullyknocksoutLisaandshefallsintoacoma.Asaresult,
sherequiresexpensivemedicalcare.220
Here, it is clear enough thatBart hasno fairness claims against Lisa for payingher
medicalbills,sincehehasanobligationtopayfortheseascompensationfortheharm
hecaused.Sothesecondmoveinreplyingtotheparentalresponsibilityobjectionisto
maketheKidsPayargumentconditionalonitnotbeingthecasethatbringingchildren
intoexistence isaharm to them. Ifbringingchildren intoexistencewereaharm to
them inaway that requiredcompensation(like inBartharmsLisa), thenproviding
for children’sneedswould seem to fallunderparents’ obligation to compensate for
wrongfullycausingthoseneedsinthefirstplace.
220Tomlin2015,p.671.
180
We might believe that we harm children by bringing them into existence if their
circumstances will be so bad that their life will be on balance worse than
nonexistence.221Indeed,somemayeventhinkthatnolifeisgoodenoughtobeworth
living, and so eachofus is harmedbyhavingbeenbrought into existence.222Orwe
might believe that even if our lives are worth living, our parents expose us to
inevitableharmsthatcomewithanylife,theimpositionofwhichcannotbejustified
bytheoverallpositivevalueofourlives,andforwhichtheyoweuscompensation.223
Providedthatnoneoftheseisthecase,Tomlinconcludesthatchildrencanbeasked,
infairness,to“paytheirownway.”
The justification for the existence of children’s Fair Play obligation, then, can be
summarizedasfollows.
1. Fair Play claims can arise for the bestowal of private, excludable,
presumptivelybeneficialgoodswhenthebeneficiariesareunavailabletogive
consent to receiving and paying for them, and when it is not possible or
reasonabletowaitforthebeneficiariestobecomesoavailable.
2. Fair Play claims can be undermined in cases where the benefactor is
responsible for causing the need for someone to receive the presumptively
beneficialgoodsbyharmingtheminawaythatrequirescompensation.
3. Parentsbenefittheirchildrenbyprovidingthemwiththeprivate,excludable,
presumptively beneficial goods of aminimally decent upbringing during the
timewhentheyareunavailable togiveconsent,and it isnotpossible towait
forthemtobecomesoavailable.
221Seee.g.Parfit1984,p.358andAppendixG;McMahan1981,pp.104-105.Somebelievethatbringingpeopleintoexistencecannotbenefitnorharmpeople.Seee.g.Narveson1978,p.48;Velleman2008a,pp.242-244;Velleman2008b,pp.247-250.222Benatar2006.223Shiffrin1999.
181
4. Parentsdonotharmchildrenincreatingthemandtherebyplacingtheminthe
situationwheretheyneedthepresumptivebenefits.
Therefore,
5. Parents have Fair Play claims against their children for having given them a
minimallydecentupbringing.
ThegreatadvantageoftheKidsPayviewisthatitavoidstheintentionalityobjection
thathassofarstoodinthewayofparents’claimstohavethecostsofchildrenshared
asamatteroffairness.Mostpeopletakeonthespecificparentalactivitiesthattheydo
inordertoprovidetheirkidswiththosebenefitsofupbringingthattheKidsPayview
highlights, namely meeting children’s essential needs for the duration of their
childhood,andraisingthemtobecomeautonomousagents.Mostparentswouldpass
thecounterfactual intentionality testwithrespect to (at least some)of thesecrucial
benefits, which I defended in Chapter 3 as a necessary condition for deeming
producers tohaveFairPlayclaims.That is, if theycametobelieve that theirefforts
wouldnotcontributetosecuringthesebenefitsfortheirchildren,theywouldnotbear
thecoststhattheyarecurrentlybearing.Sowhen,inreality,theydobearthesecosts,
theyareaimingattherelevantgoodsforwhichchildren,oncetheygrowintoadults,
wouldbeaskedtosharethecostsof.
Parentsalsopassthecounterfactualmotivationaltestwithrespecttothesegoods:if
theycouldbeprovidedsomeother,costlessway,parentswouldnotdotheworkthat
they are currently doing. This is certainly the case when it comes to the financial
efforts that giving someone a decent upbringing requires. For instance, if children’s
needs for shelter, food, clothing,health care, andeducationwerealready financially
covered,mostparentswouldnotinsistoninvestingtheirownresourcesintopaying
for these themselves. Of course, there are many goods that parents provide their
childrenwiththattheywouldwanttocontinuetoprovideevenifthesebenefitswere
182
guaranteed for their children from some other source. Reading bedtime stories to
themwouldprobablybeanexample.224ButtheKidsPayviewholdsenoughinterest
evenifweonlyfocusonthosebenefitswhichparentswouldindeedbehappytocease
to provide hands-on, or to cease to bear the financial costs for, in a counterfactual
scenarioinwhichthesethingscouldbecoveredinaproblem-freeway.
6.1.2.ThemissingstepintheKidsPayargumentIcontendthattheKidsPayviewisnotsuccessfulbecauseitistooquickinassuming
thatparentshaveFairPlay claimsagainst their childrenprovidedonly that causing
them to exist is not a harm to them. That is, it is too quick to assume that
compensatoryobligationsaretheonlysortsofspecialobligationsthatparentsmight
have towards their children and which would prevent them from having fairness
claims against them. There is at least one other class of special obligations that
parentsmight have towards their children,which Iwill call preventive obligations,
andwhichmight also be incompatiblewith parents’ Fair Play claims. This idea has
beenpresentintheliteratureonprocreativeethicsforawhilenow,and,inanutshell,
it claims thatwhileparentsmaynotharmchildrenbybringing themintoexistence,
theydoexpose themwithout theirconsent to therisksofharmthatcomewithany
existence.225This is morally problematic, perhaps even impermissible, unless some
conditionsaremet.Arguably,oneof theseconditions isthattheprocreatorstakeon
the responsibility to make reasonable efforts to ensure that those risks do not
materialize.226And if whatwemay reasonably demand that parents do in order to
shelter their children from these risks includes providing themwith the goods of a
minimallydecentupbringing,parentscannot,atthesametime,raiseFairPlayclaims
againsttheirchildrenforhavingcompliedwiththatobligation.
224SeeBrighouseandSwift,ch.5.225Seee.g.Shiffrin1999;O’Neill1979.226Shiffrin1999,p.139.
183
The idea of being liable for mitigating the risks that one exposes others to, in the
absenceoftheirconsent,liesatthecoreofawell-establishedstrandoftheoriesaimed
at explaining how parental obligations are acquired, namely causal accounts of
parental obligation. Causal accounts claim that those who are responsible for the
existence of a needy child are also responsible for ensuring that those needs are
adequately taken care of, either by the procreative parents themselves, 227 or
potentiallybyotherwillingparents,e.g.throughadoption.228Importantly,thenature
oftheobligationtocareforthechildneednotbeastrictlycompensatoryobligation.
Causalists need not, and do not, claim that bringing children into existence is a net
harmtothechild,orevenjustthat it involvesimposingsomediscreteharmsonthe
child, which attract liability for compensation. The causalist claim has more of a
preventive flavour. It regards procreation as a morally risky behaviour that could
resultinthecreationofaneedychildthatwouldbeharmedbyhavingbeencreatedif
itdidnotreceivethecareneededtoavoidsufferinganddeath.Onthestrict liability
version of the causal account of parental obligation, someone who engages in a
procreativeact is automaticallyon thehook formaking thenecessaryprovisions to
ensurethattheharmsofnotbeingcaredfordonotbefallthechild,shouldoneresult
outofthis,evenifthepersonresponsibleforitsexistencehadnointentionofcreating
achildorbecomingaparent,andtookstepstoensuretheywouldnot(forinstance,if
the child is the result of contraceptive failure, coupled with the unavailability of
abortion).
Forourpurposes,weneednotappealtoastrictliabilityversionofthecausalaccount,
astheparentaljusticedebateistypicallyconductedonthe(idealtheory)assumption
that parentswillingly and knowingly bring children into existence in order to raise
them.TheKidsPayviewisnoexception.Withrespecttosuchdeliberateprocreative
endeavours, then, the causal claim is evenmore plausible. The thought is that “the
righttobegetorbearisnotunrestricted,butcontingentuponbegettersandbearers
227Nelson1991;Callahan1992.228O’Neill1979;Blustein1997;Archard2010.
184
having or making some feasible plan for their child to be adequately reared by
themselvesorbywillingothers.”229Thereason for this is that it is impermissible to
placesomeoneinasituationwheretheyareatriskofseverepreventableharm,inthe
absenceof their consent,withoutbeingprepared toprevent thatharm fromcoming
about.
Within the parental justice debate specifically, Erik Magnusson has developed an
objectiontotheKidsPayviewalongverysimilarlines.Hearguesthatbecauseparents
“(a) electively put their children into a needy circumstance for the purpose of (b)
satisfying a self-regarding interest in meeting their children’s needs, they lack a
legitimateclaimagainsttheirchildrentoshareinitsassociatedcosts.”230
A similar, though distinct, explanation of why causing someone to exist attracts
liability to provide for their needs is Lindsey Porter’s account of procreation as
“choosing for”.Choosing for someone,Porterargues, cansometimesbepermissible,
but only if one does their best to choose well. So we could also understand the
obligationthatPorterarguesforasbeingofapreventivesort.Choosingforsomeone
isonlypermissible if you try tomake it agoodchoice,which is to say, if you try to
prevent it from turning out to be a bad choice. Porter claims that when we cause
someonetoexistwechooseforthem,intheabsenceoftheirconsent,thattheyexist,
whichispermissibleonlyifwedoourbesttoensurethatcomingintoexistenceturns
outwell for that person,which includes, at the very least,making sure theyhave a
decentupbringing.Shewrites:
On this account of the moral force of causing existence, the obligation is not
rectificatory; it is just what morally permissible choosing for requires. Making it the
case thatsomeoneexists isonlymorallypermissible ifonedoesone’sbest tomake it
229O’Neill1979,p.29.230Magnusson2018,p.965.
185
thecasethatthatsomeoneexistsdecently.So,seeingthatthechildis“contentwithher
condition”isjustpartofprocreatinginamorallypermissibleway.231
Ifanyoftheseunderstandingsof themoralrelevanceofcausingsomeonetoexist is
right,thenevenwhilewegrantthatprocreatorsareunlikeBartinBartharmsLisa,we
need not assume that they are, therefore, like Bart in Bart saves Lisa. For the
conditionsinwhichBart’ssavingLisaispermissibleseemtobedifferentthanthosein
whichparentsmaypermissiblyprocreate.WhatmakessavingLisapermissibleisthat
sheissavedfromanevengreaterharm.Placingherinthatneedystateisnecessaryto
avert theevengreaterharmofdrowning ifnoonesavesher.Thisseemsmorethan
enough to grant that Bart’s action is permissible, and that he carries no liability to
bear thecostsofLisa’smedicalbills(cf.Shiffrin1999).Butprocreatorscannotavail
themselves of the same sort of justification. Placing someone in the needy state of
beingachildisnotnecessarytopreventagreaterharm.Nonexistenceisnotaharm
thatparentssavetheirchildrenfrombybringingthemintoexistence.
Considerwhat isarguablyabetteranalogy thanBart savesLisa towhatparentsdo
whentheybringchildrenintoexistence:
BartteachesLisahowtoswim
BartthrowsLisa,thetoddler,intothewaterinordertoteachherhowtoswim.She
faces the risk of drowning unless someone takes the necessary precautions to
ensurethatshedoesnot.BartgetsintothewaterwithLisaandmakessureshegets
theinstructionandsupportneededtolearnhowtoswim.
Forourpurposes Iamassuminganumberof thingsabout thiscase. Iamassuming,
first, that learning how to swim as a child delivers important benefits that justify
teachingLisahowtoswimwhensheisatoddler,ratherthanwaitingforhertoreach
theageofconsent;second,thatthereisnootherwaytoteachherhowtoswimthat
231Porter2014,p.198.
186
involveslessrisk.Then,IassumeBart’smotivationsarenotproblematic(toavoidany
possiblecomplicationswecangrantthatheismotivatedsolelybywantingtobenefit
her,thoughnotethatinthecaseofprocreationparents’motivationsareoftenmixed
and includeselfishreasons).Finally, inorder tomake thecaseanalogous towhat is
commonlythoughtofbringingchildrenintoexistence,Iamassumingthatthebenefits
oflearninghowtoswimasachildarenotnecessarytoavertsomegreaterharmthat
wouldbefallthechild.
IfteachingLisahowtoswimisnotnecessarytoavertagreaterharmcomingherway,
the justification present in Bart saves Lisa is simply not available to Bart in the
swimmingcase.Intypicalcircumstances,throwingsomeonewhodoesnotknowhow
to swim into the water is impermissible. Whether we conceive of Bart’s act as
exposingsomeonetotheriskofpreventableharm(asShiffrinhasit),orwhetherwe
conceiveitaschoosingforsomeone(inPorter’sterms)thattheylearnhowtoswim,
his act canbemadepermissible only if he takes all the reasonable stepsneeded to
ensure that the harm of severe injury and even death do not actualize, or that the
choicehemakesforherturnsouttobeagoodonerespectively.
Sofar,Tomlinmightnotdisagree.Butatthispointhemight insistthatwhileLisa is
owedthosesafetyprecautions,itdoesnotfollowthatsheisowedthematnocostto
herself.SoBartmightstillbeabletoraisefairnesscomplaintsagainstLisaifshefails
toreimbursehimforsomeofthecostsoftheinstructionandsupporthehasgivenher.
Thatwemay be asked to contribute to the costs ofmeeting our own just claims is
quiterightincertaincircumstances.Weareowedlife-savingmedicalattentionwhen
we get into an accident, but through our taxes we do contribute our share to the
medicalservicesthatwereceiveifandwhenweneedthem.However,thedifference
betweenpayingourshareforthemedicalserviceswemighthappentoneedandBart
Teaches Lisa How to Swim is that in the latter case a particular person ismorally
responsible forcreating thatneed inawaythatwouldbe impermissibleunless they
187
took the necessarymeasures tomake sure that their action did not result in harm.
Thischangesthesituationconsiderably.ItestablishesnotonlythatBartisresponsible
for putting Lisa in a needy situation, but that he does so in a way that grounds a
specialobligationthathemeetthoseneeds,lesthewouldbeactingimpermissibly.If
makingtheprovisionsnecessarytoensurethathisactiondoesnotseverelyharmLisa
isaconditionof thepermissibilityofBart’saction inthe firstplace, it ishardtosee
howhecouldaccuseLisaofunfairlyfreeridingonhiseffortsnottoharmher.
Ifthisisright,parents’FairPlayclaimsagainsttheirchildrencanbeunderminedeven
though they do not cause harm to them by bringing them into existence. So even
though we can grant, at least for the sake of argument, that parents do not have
compensatory obligations that could undermine their Fair Play claims against their
children, theymay still have preventive obligations that are also incompatiblewith
raising Fair Play complaints. If the goods for which parents would raise Fair Play
claims are thosewhich they have a preventive obligation to provide their children
with, they cannot accuse children of unfairly taking advantage of their efforts to
providethem.Similarly,ifwefollowPorter,ifthegoodsforwhichparentswouldraise
FairPlayclaimsarethosewhichtheymustprovidetheirchildrentoensurethatthe
choiceofbringingthemintoexistencewaspermissible,thentheycannotcomplainof
unfairnessiftheirchildrenrefusetosharethecostsforprovidingthem.
So,are thebenefits that theKidsPayview focuses included in the rangeofbenefits
thatparentshave apreventiveduty toprovide?Recall, the goods that theKidsPay
viewappliestoarethosenecessarytobringchildrentofullagency,plusthosewhich
arenotnecessarytobringthemtofullagencybutwhichareessentialforaminimally
decentlifeandwhichmustbeprovidedduringchildhood(thatis,fortheprovisionof
whichparentscannotaffordtowaituntilthechildrencangiveauthoritativeconsent).
I have called these the goods of aminimally decent upbringing. Providing children
with the benefits of a minimally decent upbringing seems to be the minima that
parentsowetheirchildren,asamatterofspecialobligation,iftheirprocreativeactis
188
tobepermissible in the firstplace.For surely someof theworst risksofharm that
childrenareexposedtobybeingbroughtintoexistencearetheriskofsufferingfrom
basicneedsdeprivationduringchildhood,aswellastheriskoffailingtodevelopfull
agencyandthereforefailingtocraftalifeforthemselves.Thelatterwouldpresumably
lead to their being under other people’s domination during parts of, or even their
entire, lives. Ifparentsbringchildrenintoexistencewithoutthe intentionofmaking
allthereasonableeffortstoshieldtheirchildrenfromtheseharms,arguablybringing
children into existence is impermissible. And if the very permissibility of parents’
procreative actions depends on their being prepared to alleviate these severe and
urgentneedsthatcomewithanyexistence,surelytheycannotturnaroundandaccuse
theirchildrenoftreatingthemunfairlyifthey,thechildren,refusetosharethecostsof
meeting those needs. Of course, this is not to say that children should not pay
somethingbacktotheirparents,astheymayhavereasonsofloveorgratitudetodo
so. And indeed,many of themdo. But parents cannot lay a justice-based claim that
theyshould.
6.2.KidsPayandtheSharedPreferenceViewIntheprevioussectionIhaveofferedreasonstobelievethattheKidsPayviewfailson
itsownterms,evengrantingsomeof itsmajorpremises for thesakeofargument. I
havenotchallenged,forinstance,theparticularKlosko-inspiredversionofFairPlayat
work in the argument. Secondly, I have granted that parents do not harm their
childrenbybringingthemintoexistence.Ihavearguedthattheargumentstillfailsto
securetheconclusionthatchildrenoughttopaytheirwayasamatterof fairnessto
theirparentsbecauseanimportantstepismissingfromthisargument.Whileparents
maynotharm their childrenbybringing them into existence, I have suggested that
theyneverthelesshavespecialobligationstoprovidefortheirchildrenthat(a)need
not relyon thenotion that causing themtoexist isaharm,and that (b)undermine
189
theirfairnessclaimsagainsttheirchildrenontheFairPlayviewthatKidsPayrelies
on.
InthissectionIexplorehowtheviewofFairPlaythatIdevelopedinChapter3,the
SharedPreferenceView,appliestoparentsandtheirchildren.Thisisworthdoingfor
tworeasons.First,itoffersindependentreasonstorejectthethesisthatchildrenact
unfairly if they fail to pay for themorally required goods their parents owe them.
Secondly, andmore importantly, the SharedPreferenceView, Iwould like to show,
cangroundFairPlayobligations,butforadifferentrangeofgoodsthanTomlin’sview
supposes.WhiletheversionofFairPlaythatTomlinemploysisonlyapplicabletothe
benefits of a minimally decent upbringing, the Shared Preference View is also
applicabletothemorallyoptionalgoodsofanupbringing.Iwillshowthatonthisview
childrenmayindeedhavefairnessobligationstowardstheirparentswhenitcomesto
averyrestrictedclassofgoods,namelymorallyoptionalbenefitsthatparentsmight
provideandfromwhichboth thechildandtheparentdirectlybenefit.However, the
sorts of obligations established thisway are arguably not enforceable because they
applytoarangeofgoodsthatarenotmorallyrequired.
InordertoexaminewhattheSharedPreferenceViewhastosayaboutthepotential
fairnessclaimsthatparentsmightraiseagainsttheirchildren,itishelpfultoremind
ourselves that theconditions inwhich freeridingcountsaswrongfulon theShared
PreferenceViewdifferaccordingtothetypeofbenefitatstake. Ihavedistinguished
betweenmorallyoptionalgoodsontheonehand,andmorallyrequiredgoodsonthe
otherhand.Whereoptionalgoodsareatstake,Ihavearguedthattherelevantkindof
opportunitytofreeridearisesthankstothecollectivenatureofthegood,anditisthe
opportunitytointernalizeagoodthatonevaluesandwhichothershaveproduced.In
thecaseofmorallyrequiredgoodstherelevantopportunitytofreeridearisesthanks
tothecollectivenatureofthedutytoproducethosegoods,anditistheopportunityto
see a collective duty that applies to one, alongside others, be discharged only by
others. If themorallyrequiredgoodhappenstobe itselfcollective innatureaswell,
190
thenbothkindsofopportunitiestofreeridearise.Thisisthecase,forinstance,with
publicdefence.Thereisacollectivedutytosecurepublicdefenceforallthecitizensof
astate,anditisalsothecasethateachcitizencanbenefitherselfbyinternalizingthis
good.
Withthispictureofwhatcharacterizestheopportunitytofreerideinvarioustypesof
cases, recall, furthermore, that according to the SharedPreferenceView, free riding
counts as unfairwhenboth the free riders and the contributors (or producers) are
relevantly similarly situated, such that the free riders’ failure to pay amounts to
makinganunjustifiedexceptionof themselves.OntheviewIdefend, the freeriders
and the contributors count as relevantly similarly situatedwhenwe can ascribe to
both a qualified preference for free riding. In the case of optional goods that
preference,torepeatit,statesthefollowing.
TheFreeRider’sPreference (FRP): I prefer that otherspay for this valuable collective
goodthatIcanenjoyforfreeandforwhichIwouldbepreparedtopay,intheconditions
underwhichitisoffered,ifIhadto.
Inthecaseofmorallyrequiredgoods,thepreferencethatwemustbeabletoascribe
tobothpartieswasthis.
The Free Rider’s Collective Duty Preference (CDP): I prefer that others discharge the
collectivedutythatalsoappliestome.
InordertoestablishwhetherparentshaveFairPlayclaimsagainsttheirchildrenon
theSharedPreferenceView,thefirststepweneedtotakeistoseparatethebenefits
that parents typically provide to their children into the two categories that are
relevantfortheSharedPreferenceView:morallyrequiredgoods,andoptionalgoods
respectively.Thefirstcategorycomprisesthosegoodsthatwebelievechildrenhavea
justiceclaimto.Wemightbelievethese includeonlytherangeofgoodsthatTomlin
was concerned with, namely (a) the goods of a minimally decent upbringing.
191
Alternatively, we might believe that children have claims to much more than that,
namelytowhatwecouldcall(b)thegoodsofagoodupbringing.232Thelattermight
include,forinstance,alevelofeducation,healthcare,andguidancethatexceedswhat
issufficienttogiveachildaminimallydecentlife.
Thesecondcategoryofbenefitsthatparentsoftenoffertheirchildren,thatofoptional
benefits,referstothosegoodswhichgobeyondwhatchildrenhavejusticeclaimsto
receiving.Parentsoftendotheirbesttoactonthepreferencesthattheirchildrenform
aboutthingsthatmattertothem(ofwheretolive,orwhattoweartoschool,orhow
to spend free time). Parents also invest a lot of time, energy and resources into
discoveringanddeveloping their children’s talents, and intoofferinga greatdealof
guidanceandadvice forwhatever childrenneed.And indeed, someparentsprovide
their children with what we might consider luxury goods such as very expensive
clothes,gadgets,orholidays.Thesesortsof thingsbelongto thecategoryofmorally
optionalgoods.
Thenext stepweneed to take inorder toestablishwhetherparentshaveFairPlay
claims against their children is to verify whether we can ascribe the relevant
preferences for free riding, the CDP or the FRP respectively, to both parties. In the
nextsectionIshowthatparentscannotbeascribedtheCDPifitistruethatproviding
children with the goods of a minimally decent upbringing, or the goods of a good
upbringing, isamatterof specialobligationrather thanacollectivedutyof thesort
that is relevant for Fair Play. After this, I argue that a version of the FRP can be
ascribed to both parents and their children for a very restricted range of morally
optionalgoods.
232SeeClayton2012.
192
6.2.1.ThemorallyrequiredgoodsofanupbringingConsider, first, how the Shared Preference View of Fair Play might apply to the
morally required goods that parents provide their children, whether these include
only the goods of a minimally decent upbringing, or also the benefits of a good
upbringing.Forparentstohaveclaimsagainstchildrenwithrespecttothese,itwould
havetobethecasethatthedutytosecurethesegoodsisacollectiveone.InChapter3
Ihavedefinedcollectivedutiesasagent-neutraldutiesthatfallonagroupcollectively
todischarge.Thedutytosecurepublicdefence,forinstance,arguablyfallsonallthe
citizens of a particular country. For this duty to be discharged, it is notmorally or
practicallynecessary forall the citizensof that state to contribute.That is, theduty
simplydescribesanoutcomethatcan,practicallyspeaking,besecuredonlybysome
citizens as opposed to all of them. And morally speaking, the duty does not itself
requirethatallshouldworktosecureit.Importantlyforourpurposesinthissection,
thecollectivedutyalsodoesnotgivefundamentalreasonsforsomeparticularpeople
tosecure it rather thanothers.All that theduty tellsus is thatanoutcomemustbe
secured, and that the responsibility for this falls on the entire collective. Once the
requiredoutcomeisachieved,thedutycountsasdischargedforthewholecommunity
inthesensethateveryoneisreleasedfromhavingtodoanythingmoreforit.
Giving children aminimally decent upbringing, or perhaps a good upbringing, does
notseemtobeamatterofdischargingacollectivedutysounderstood,oncewekeep
inmindthat,onthepictureproposedbyTomlin,wearetoseesocietyascomposedof
parents and their (grown) children. Providing childrenwith the sort of upbringing
that they are owed by justice does not seem to be a duty that applies in an agent-
neutralfashiontoboththeparentsandtheirchildren.
Ashighlightedintheprevioussection,onsomeprocreativeethicsviewsparentshave
aspecialobligationtomitigatetherisksthattheyexposetheirchildrentobybringing
them into existence. If this risk prevention amounts to giving children aminimally
193
decentupbringing,oragoodupbringing,thenitcannotbethecasethatchildrenand
parents are both under a collective, agent-neutral duty to provide these benefits to
children.
Ofcourse,inonesense,everyonehasageneralreason(andsometimesaduty)totend
tothebasicneedsofanychild,aseveryonehasageneralreasontoalleviatesuffering
and prevent harm from occurringwhen possible. Arguably, everyone could also be
said to have a reason to help ensure that children develop into full, autonomous
agents. But this is compatible with the existence of an agent-relative reason or a
specialobligationforparticularpeopletobeartheprimaryresponsibilityforensuring
that children getwhat they need, in virtue of the special relationship they stand in
with those children, just like acknowledging that everyonehas a reason to alleviate
harmwhen they can is compatible with the existence of agent-relative reasons for
whyparticularpeopleshouldbeartheprimaryresponsibilityfordoingsoasamatter
of special obligation. The clearest such cases are those where someone’s harm is
wrongfullycausedbyanagentwhothenbearsprimaryresponsibilityforalleviatingit.
One does not have to subscribe to the particular view of the source of parental
obligationswhichIdiscussedintheprevioussection(namely,aparticularversionof
the causalist view) in order to deny that giving children a minimally decent
upbringing, or a good upbringing, is not (or not merely) a matter of agent-neutral
collectiveduty.Common-sensemoralityaswellasmostothertheoriesofprocreative
ethicsalsotendtoviewobligationstowardschildrenasspecialobligations.AsDavid
Archardwrites,
Recentphilosophicalworkonparenthoodstartsfromtheassumptionthatitisinvirtue
ofstandinginacertainrelationshiptoachildthatadultshaverightsoverandoweduties
194
tothatchild.Theoriesoftheprovenanceofparentalrightsanddutiesincludethecausal,
thegenetic,thegestational,andtheintentional.233
So,anyofthecommonlysubscribed-toviewsofprocreativeethicswouldcharacterize
the responsibility of giving children the sort of upbringing that they are owed as a
matter of special obligation as opposed to a form of collective duty. On the Shared
Preference View, if an agent has a special obligation to provide certain morally
requiredgoods, theycannot, at thesame time,haveclaimsof fairness forproviding
thosegoods,foritcannotbethecasethattheyarerelevantlysimilarlysituatedwith
thebeneficiariesforFairPlaypurposes.
The Shared Preference View, then, offers a principled reason for endorsing the
intuitivelyplausible conclusion,which I invoked in the introduction to this chapter,
and which several prominent philosophers have already advanced,234that grown
childrendonotoweadebttotheirparentswhenitcomestogoodsthattheirparents
are under an obligation to provide. On the Shared Preference View, the reason for
endorsingthisviewhas todowiththerole that thenatureof thedutyplaysfor the
ascriptionoftherelevantfreeridingpreference,theCDP:Ipreferthatothersdischarge
the collectiveduty thatalsoapplies tome. Recall, this preference can be ascribed to
peoplewhofallunderacollectivedutyinvirtueoftwoassumptionsaboutthesortsof
beings we are. The first was the assumption that we are appropriate subjects for
moralevaluationinthesensethatwehaveamoralinterestinseeingthemoralduties
andobligationsthatapplytousdischarged.Thesecondwastheassumptionthatwe
canconceiveofourselvesasprudentiallyrational,whichistosaythatweareinclined
toactinawaythatmaximizesourwell-being.Together,thesetwoassumptionsallow
ustoascribethepreferencetoseeacollectivedutythatappliestousdischargedonly
byothers,sincethisoptionsatisfiesbothofourinterestsbest.Wegetthebestofboth
233Archard2010,pp.106-107.234Seee.g.Daniels1988,pp.30-31;Simmons1979,p.182.
195
worlds:ourinterestinseeingadutythatappliestousdischargedissatisfied,andsois
ourprudentialinterestinnotdoingtheworkourselves.
Butthisstoryreliesonthedutyinquestionbeingcollectiveintherelevantsense.In
particular,itreliesonitsbeingagent-neutral,suchthatourmoralinterestcanbefully
satisfiedevenwhenthemorallyrequiredoutcomeisproducedwithoutourcontribution.
Ifitturnsout,asitdoesinthecaseIamdiscussing,thatthemorallyrequiredoutcome
cannotbedischargedwithoutourcontribution,becausewehaveaspecialobligation
todischargeitourselves(whetherinvirtueofapastactionorarelationshipweare
part of), then the correct description of ourmoral interest is to discharge the duty
ourselves.Itmatters,now,whoprovidesthemorallyrequiredoutcome,andifitmust
be us in particular (as opposed to any other agent), thenwe cannot claim that our
moralinterestissatisfiedifothersdoit.Ourmoralinterestwillbesatisfiedonlyifwe
dischargethespecialobligationsthatapplytous.Forthisreason,wecannotascribeto
parentstherelevantfreerider’spreference,thatis,apreferencetofreerideonothers’
dischargingthecollectivedutythatalsoappliestothem,bydoingtheworkofgiving
childrenthechildhoodgoodstheyareowed.
Not everyone agrees that parents’ obligations to their children are fundamentally
special obligations.235Robert Goodin, for instance, argues that giving childrenwhat
they are owed is part of the general duty of protecting the vulnerable.We all have
duties to act inways that protect the interests of thosewho are vulnerable to our
actions and choices, claims this vulnerability model. If parents owe more to their
childrenthanotherpeopledo,thisisbecausechildrenareparticularlyvulnerableto
theirparents’actions,andbecauseparentsmaybebestplacedtomeetsomeoftheir
children’s needs, and not because there is some fundamental special obligation at
play. “‘Special responsibilities,’ such as those of particular family members for one
another, are on this model just instances of ‘distributed general responsibilities’,”
235TheseexceptionsincludeSwiftandBrighouse(2014,p.51)aswellasVallentyne(2002).
196
writesGoodin.236Onthisview,inordertoknowwhoshouldmeetchildren’sneeds,we
should look at how we, as a society who recognizes general duties to protect the
vulnerable,couldbestdistributetheresponsibilitiestomeetthoseneedsinawaythat
bestprotectschildren’sinterests.Itseemslikeparentsareindeedbestplacedtobear
responsibilityformeetingneedssuchaschildren’sneedforintimacy,forhavingwell-
informed decisions made on their behalf, for guidance. But if meeting such needs
would be facilitated by external, perhaps financial, support, then the rest of society
has a reason (and perhaps an obligation) to provide such support to parents. And
whenitcomestomeetingchildren’smaterialneeds,onceagainthethoughtisthatif
these needs could be best met by sharing the costs across society, rather than by
holdingparentsaloneresponsibleformeetingthem,thensocietyshoulddoitsbestto
meetthatresponsibilitytotheextentthat’spossible,compatiblywithmeetingother
responsibilitiesthatthestatemighthavetoothervulnerablesocialgroups.237
IfweacceptedGoodin’sview, thearguments Ihave justdeveloped fordenying that
children owe Fair Play obligations to their parents for having received morally
requiredgoodswouldnotapply.IdonothereargueagainstGoodin’sview,butnote
that ifweendorsed it,wewouldnotneed toappeal toaFairPlayargumentof any
kindtogivesupporttotheclaimthatnon-parentsshouldsharethecostsofchildren
with parents. This is because, on Goodin’s view, the most natural justification for
sharingthecostsofchildrenisnottodowiththeobligationsthatnon-parentshaveas
former children who were benefited by their parents, as the Kids Pay view would
claim.Thevulnerabilitymodelgroundsanobligationonthepartofallcontemporary
adults (parents andnon-parents) tohelp share the costs of children at ahorizontal
level,ifyouwill,invirtueofthefactthateveryonehasageneraldutytohelpmeetthe
needsofexistingchildrenaswellaspossible.Soifthevulnerabilitymodelisusefulto
makeapro-sharingargument,itisnotusefultomakeapro-sharingargumentofthe
KidsPayvariety.
236Goodin2005,p.67.237Goodin2005,pp.66-67.
197
6.2.2.TheoptionalgoodsofanupbringingLetusnowturntotherangeofchildhoodbenefitsthatarenotmorallyrequired.Can
parentshavefairnessclaimsagainstchildrenforhavingprovidedthemwithbenefits
that are morally optional, such as trying to satisfy their preferences as much as
possibleorgivingthemluxurygadgetsandclothes?
Inansweringthisquestion,weshouldnotethatwithrespecttomostoptionalbenefits,
theSharedPreferenceViewsimplydoesnotapplyinsofarasthesearecasesofprivate
benefits transfers from a benefactor to the intended beneficiary. Indeed, the
beneficiaryofagiftorafavorgetstoenjoyabenefitatnocosttoherself,whichisalso
what unfair free riders do. But the analogy stops there. Recall, the phenomenon of
unfairfreeridingthatweareinterestedinisinimportantwaysdistinctfromenjoying
a benefit for free by receiving a gift, or by stealing, as I explained in Section 3.3. If
someonewhogetssomethingforfreewastheintendedbeneficiaryinthefirstplace
wemightdoubtthatthereissomethingunfairaboutnotpayingthebenefactorback.
Eitherthebenefitwasintendedasagiftwithnostringsattached,or,ifthebenefactor
expected something in return, they should have sought the beneficiary’s consent. If
therewasnoconsenttopayinganything,orifthebenefitwasintentionallygivenasa
gift, thereseems tobenothingunjust (becauseunfair)aboutenjoying thatgood for
free. So if children enjoy, say, expensive gadgets and toys and clothes at no cost to
themselves, most likely they are the intended beneficiaries of gifts, not unfair free
riders.Theymay incuramoralobligationtoshowgratitude for thesegifts,but they
cannotbeenjoinedtopaytheirshareforthemonFairPlaygrounds.
This being said, there seems to be a special class of optional goods that parents
provide their children with. These are goods that are in some sense collective,
meaningthatboththeparentandthechildbenefitfromthematthesametime.Goods
likespendingqualitytimetogether,readingbedtimestories, learningnewskillsand
playing together can be considered collective in that sense. They contribute to
198
relationshipgoods thatenrichbothparties,238and forwhichonlyoneof theparties
mightbepayingthecosts.
Suchsharedgoods,then,cangiverisetoanopportunitytofreeride.Thenextstepfor
applyingtheSharedPreferenceViewistoseewhetherwecanascribetheFreeRider’s
Preferencetobothparties.TheFreeRider’sPreferencestatesthatindividualswould
prefertoenjoytherelevantcollectivebenefitwithoutcontributinganythingtoit,and
here, like with the case of morally required goods, we seem to face a difficulty in
ascribing the Free Rider’s Preference to either the parents or the children. This is
becausethecollectivegoodsoftheparent-childrelationshipcanonlybebeneficialfor
eitherpartyifbothparentsandtheirchildreninvesttherequisitetimeandeffortinto
enjoyingqualitytimetogether,readingbedtimestories,engaginginvariousactivities
togetherandsoon.
If spending time and energy on these shared activities can count as costs, then
enjoying thesharedgoodsof theparent-child relationshipatno cost tooneself, like
theFreeRider’sPreferencewouldrequireintypicalcases,wouldmeannotenjoying
thesharedgoodsatall.Thatis,itseemsthattherearesomecoststhebearingofwhich
is integral to theverypossibilityofenjoyingthebenefit in the firstplace. Ifso, then
neitherpartycanbeascribedthepreferencetoreceivethesebenefitsentirelyforfree.
However,notingthatsomecollectivebenefitsaresuchthatsomecostsareintegralto
thepossibilityofanyindividualenjoyingthemoughtnotleadustotheconclusionthat
theFreeRider’sPreferencecannotbeascribedtoanyparty,andhencethatnounfair
freeridingcanoccur.FortheSharedPreferenceViewwasdevisedwithparadigmatic
cases in mind, in which a free rider could get a benefit for free without paying
anythingwhatsoever for it. Someone can enjoy clean air for free thanks entirely to
others’effortsofreducingtheircarbonprint.Therearenointegralcoststobeborne
fortheenjoymentofcleanair.Forthisreason,investinganyeffortintocontributingto
238SeeBrighouseandSwift2004,ch.4foradiscussionofthevalueoffamilialrelationshipgoods.
199
the clean air effort would be a waste for the free rider, and this is why the top
preferencewecanascribetoheristopaynothing.Butifthenatureofthecollective
goodissuchthatsomesortofeffortorcostisrequiredofanyonewhohopestoenjoy
itatall,likeitisthecasewithrelationshipgoods,thenitwouldnotbeilluminatingto
saythatthepartieshaveaninterestinpayingsomething,i.e.,thatwhichisnecessary
fortheenjoymentofthegoodinthefirstplace.Forthiswouldbeequivalenttomerely
sayingthatbothpartieshaveaninterestinenjoyingthebenefit.Sowhensuchgoods
areatstake,weshouldinsteaddistinguishbetweenthecoststhatareintegraltothe
possibility of enjoying the benefit in the first place, and non-integral costs, and see
who can be ascribed the Free Rider’s Preference with respect to the non-integral
costs.Thereasonforthisisthattherelevantopportunitytofreeride(toreceivethe
benefit without doing one’s share) seems to only arise at the level of non-integral
costsinthefirstplace.
In the case of the parent-child relationship goods, it seems that the integral costs
includethingslikeinvestingtime,thought,andenergyintovaluablesharedactivities
and into building the sort of trust and intimacy that are thought to be important
relationship goods between parents and children.239But then, of course, there are
material costs that might accompany the realization of this relationship-building.
Enjoying an afternoon of family quality time at Disneyland, for example, can be
financially quite expensive. So, too, can be deciding to forgo lucrative work
opportunitiesinordertostayathomeandtutorone’schild.
It seems, then, thatwemightbeable toascribe theFreeRider’sPreference toboth
parent and childwith respect to the non-integral costs, like thematerial costs just
mentioned,thatmightaccompanytheirsharedrelationshipgoods:
I prefer that others pay the non-integral costs for these valuable shared goods of the
parent-childrelationship,whichIwouldbepreparedtopayifIhadto.
239BrighouseandSwift2014.
200
Since the twoparties here are theparents and their children, itmight seemodd to
ascribetoparentsthepreferencethatothers,here,theirchildren,shouldbearallthe
material burdens of their valuable shared activities, like paying for the Disneyland
trip, or compensating them for staying at home to tutor them. One reason this is
intuitivelyodd is todowith the fact thatchildrencannotpay for themselves.Butof
course,thisproblemcanbeby-passedoncewethinknotofchildrenaschildren,but
asfutureadults,abletopayretroactively.
A secondreasonwhy theascriptionof thispreference toparentsmight strikeusas
oddisbecauseitseemsquitemeanorunusuallyselfishforparentstowishthattheir
childrenboreallthematerialburdensoftheirsharedactivities.Ifanything,weknow
that parents are prepared to sacrifice quite a lot for their children, somost would
actuallybepreparedtobearallthesecoststhemselvesforthesakeoftheirchildren.
In response, recall that the Free Rider’s Preference is not an actual preference, but
something we can ascribe to people on the assumption that they are prudentially
rational inaquitenarrowsense.Tosay thatbothparentsand theirgrownchildren
wouldhaveaninterestinnotpayingthematerialcostsissimplytosaythat,assuming
that both value the benefits enough that bothwouldprefer to pay all thematerials
costs rather than pay nothing and forgo the benefits, nevertheless each would be
prudentially better off if for some reason the other party does, in fact, bear all the
material costs. Once again, this is not about actual judgments that selfish, mean
parentsmightmake,butaboutthefactthat,prudentiallyspeaking,parentswouldin
principlebebetteroffiftheydidnothavetobearallthesematerialcostsalone.And
thesamegoesfortheirgrownchildren.Asithappens,sinceatthetimeofproducing
thesebenefitschildrencannotpayforthemselves,parentsbearallthematerialcosts.
Wecansay,however,thatsincebothparentsandchildrenbenefitfromtheseshared
goodsandactivities,thereisnoreasonwhythechildren,oncegrownandabletopay
theirparentsback,shouldnotbeaskedtopaytheirshareforthem.(Whatformtheir
contributionmighttakeisanopenquestion.)
201
It concluding this section, let me consider an objection: it might seem very
counterintuitivethatchildrencanbeaccusedofunfairlyfreeridingontheirparents’
effortsofgivingthemaminimallydecentupbringing,orevenagoodupbringing,andI
suggestedthatitisavirtueoftheSharedPreferenceViewthatitdeniestheexistence
of such fairness obligations. Someone might wonder whether it is even more
counterintuitivetoconcludethatparentshaveclaimsoffairnessagainsttheirchildren
forthemorallyoptional,sharedbenefitsoftheparent-childrelationship.
Itseemstomethat thereasonwhyaccusingchildrenofunfairly freeridingseemed
implausible in the first place was mainly to do with the fact that this claim was
attachedtobenefitsthatchildrenareowedbyjustice–owed,indeed,bytheirparents
invirtueoftheirprocreativeactions.JustlikeIthoughtitimplausibletosaythatLisa
inthe“BartteachesLisahowtoswim”caseowesanythingtoBartforhisavertingthe
harmsthathehimselfexposedherto,Ialsothoughtthatparentscannothavefairness
claimsforgivingtheirchildrenthekindofupbringingtheyareowedforhavingbeen
put in thatneedysituationby theirparents.But ifBart, inaddition to teachingLisa
howtoswimandtreatingherinthewaysthataremorallyrequiredofhim,alsomade
effortstobenefitherinwaysthatwentoverandabovewhatheowedher,thenitdoes
not seem tome so implausible to say that she could be required to share some of
thoseextracosts,ifsheindeedjudgesBart’sactionstobebeneficialforher.
Asecondthingwemightsayinresponseisthatparentsmaywellopttoreleasetheir
childrenfromtheFairPlayobligationtopaytheirshare.This,ofcourse,isanoption
opentoanybenefitsproducer,butitismorelikelytobetakenupinsuchcaseswhere
reasonsofloveorfriendshipmightcauseproducerstogiveuptheirFairPlayclaims.
Soparentswilloften,asamatteroffact,relinquishtheirclaimsagainsttheirchildren,
butthemainpointisthatsuchclaimscanindeedexistforarestrictedclassofgoods
thatparentsprovidetheirchildrenwith: thosebenefits thatcontributetobuildinga
valuable parent-child relationship that both parties value and which go over and
above what parents owe their children as a matter of special obligation. This
202
conclusion is in linewith those views of filial obligation according towhich grown
children have obligations of gratitude or reciprocity only when their parents go
beyondthecallofduty.240
Finally,becauseparentshavefairnessclaimsonlyforhavingprovidedthosesortsof
shared relationship goods that fall outside the realm of justice (i.e. that go beyond
givingchildrenwhattheyareowed),arguablytheirclaimsarenotenforceablebythe
state.
6.3.TheSharedPreferenceViewandfilialobligationsIt isworth taking amoment to consider the place that the SharedPreferenceView
occupieswithin the filial duty literature, especially given that the twomostwidely-
endorsed views of filial obligation, the debt theory and the gratitude theory, bear
manysimilaritiestothefairnessargumentadvancedabove.
To remind ourselves, gratitude obligations are typically defined as obligations to
communicateordemonstratefeelingsofgratitudeforbenefitsthatonehasreceived,
in recognition of the benefactor’s time and efforts as valuable.241The debt theory
claims that children and parents are in one important respect like debtors and
creditors: your parents “fronted” you a lot of resourceswhen youwere a child, so
whenyoubecomeanadultyouought topayback that investment.242Both theories,
then,seemtorelyonanintuitive,generalnotionofreciprocityorfairreturnstomake
theircase.
240Blustein1982,p.182;Callahan1985,p.35.241Keller2006,pp.257-258.242Keller2006,p.256.
203
Despitethesimilarities,thereareimportantsubstantivedifferencesbetweenFairPlay
and these theories. The conditions under which gratitude obligations or filial debt
maybeincurreddifferfromtheconditionsunderwhichfairnessobligationstopayfor
thecostsofchildrenmaybe incurred,and thecontentofwhat isactuallyowedwill
alsoultimatelydiffer.
Takegratitudeobligations.Itisarguedthatgratitudeobligationsaregeneratedwhen
the benefactors are “acting, at least in part, from themotive of doing good for the
beneficiaryforthelatter’sownsake,”243andthattheamountofwhatthebeneficiaries
owe iscorrelatedwith theeffort that thebenefactorsmade,notwith thesizeof the
benefit thatwasgiven.So, ifawealthyparent findscaringforherchilda fairlyeasy
and enjoyable task, the child’s debt of gratitude to that parent will be smaller
compared to the one incurred by someone whose parents had to sacrifice a lot in
ordertobenefitthem.Moreover,somearguethatincurringsuchdebtsofgratitudeis
compatiblewiththeparents’havingmerelydischargedtheirparentalobligationsby
benefitingtheirchildren.244
OntheSharedPreferenceViewofFairPlay,bycontrast, inorderforachildtoincur
fairnessobligations it isnotnecessary, inprinciple, that theparentbemotivatedby
thechild’sowngood.Theymightbemotivated,instead,bywantingtobeseenasgood
parents. Of course, in reality most parents want to benefit their children for the
children’ssake,but,inprinciple,eventhosewhoarenotmotivatedinthiswaycould
havefairnessclaimsagainsttheirchildreniftheotherconditionsforFairPlayaremet.
Secondly,ontheSharedPreferenceViewtheamountofwhatisowedisproportional
to the amount of benefits (and their associated costs) that the parent provided for
their children, not to how difficult or easy it was for the parent to provide these
benefits. Thirdly, as I have argued in this chapter, benefactors cannot have fairness
claims tobepaid for benefits that thebenefactorswereobligated toprovide to the
243Jeske2019,p.239.244Jeske2019,p.239.
204
beneficiariesinthefirstplace.OntheSharedPreferenceViewoffairness,parentscan
only have claims formorally optional benefits that benefit both the parent and the
child.
Mostimportantly,however,thegratitudeviewandtheSharedPreferenceViewofFair
Playarenotinfactincompetitionwithoneanother.Thisisbecausetheydonotseem
toanswerthesamefundamentalquestion.FairPlayisgearedtowardansweringthe
overarchingquestionofthisproject:whoshouldbearthecostsofchildren?Isit just
parents, or perhaps also children themselves? Gratitude theories of filial obligation
fundamentally address a different question: what do grown children owe their
parents?Of course, these twoquestions, and their respective answers, canpartially
overlap. Inparticular,FairPlaycanofferonekindofanswertothequestionof filial
obligation.Forwemightthinkthatwhatgrownchildrenowetheirparents,perhaps
among other things, is to share some of the burdens of their own upbringing as a
matter of fairness. But theories of filial obligation do not generally offer an answer
aboutwhoshouldpayforthecostsofchildren.Showingthatchildrenhaveobligations
ofgratitude to theirparentsdoesnotamount tosaying,andarguablydoesnot lend
itself to supporting the view, that children should pay their ownway. This is true
regardless of the direction in which one might think this obligation should be
discharged, i.e. by paying back one’s parents directly, or, alternatively, by paying it
forwardtothenextgeneration.
Supposewethoughtthatobligationsofgratitudeforwhatourparentsdidforusought
tobedirectedtowardsthem.Itisdoubtfulthattryingtopayone’sownwayisagood
waytoshowgratitude.For thepointofgratitude is toshowappreciation foractsof
goodwill,nottotrytomakegoodtheinvestmentsthatyourbenefactorshavemade.
Infact,attemptingtodothismightevengoagainstthespiritofshowingappropriate
appreciaton.Ifonemakesanefforttovisitafriendinhospitalandtobringagift,one
wouldbeoffended(rightly,Iwouldthink)ifthatfriendattemptedtopayonebackfor
thateffort,forexamplebyofferingtopayforone’stransportandtoreimbursethegift,
205
orevenbyreturningthevisitandbuyingagiftcarefullypickedouttobeofroughly
thesamevalue.
Now suppose the obligation of gratitude forwhat our parents did for us should be
dischargedintheotherdirection,bybenefitingthenextgeneration.Thisseemslikean
evenmoreoddattemptatshowingappreciation forourparents’goodwill. Itwould
seemveryoddifthefriendIvisitinhospitallatermakesanefforttovisitadifferent
person inhospitalasasignofgratitude tome.This isall tosuggest that theoriesof
gratitudeand theSharedPreferenceViewofFairPlayarenot indirect competition
with each other. A grown child might have both obligations to show appropriate
gratitudetoherparentsaswellasaFairPlayobligationtohelpbearsomeofthecosts
ofherownupbringing.
Consider,now,thedebttheoryoffilialobligation.Oneinterestingdifferencebetween
thedebttheoryandtheSharedPreferenceViewasappliedtochildrenisthatthedebt
theory is vulnerable to a sort of voluntarist objection in away inwhich the Shared
PreferenceViewisnot.Onemightobjectthatchildrendidnotasktobeborn,norwere
theyabletoconsenttothebenefitsthattheirparentsprovidedthemwithduringtheir
childhood, and yet theynow find themselves saddledwith this debt.On the Shared
Preference View, given that fairness obligations are only incurred for the morally
optionalgoodsthatbothparentsandchildrenenjoytogether,thegrownchildrencan
beletoffthehookiftheybelievethattheseoptionalgoodswerenot,infact,beneficial
to them, or that they are notworth the cost, even if they aremistaken about these
judgments. Someone might decide, for instance, that their parents ought not have
taken them to Disneyland every summer. They might think that the benefit of
spending quality time together in that particular way was not worth the cost,
especiallywhenthereexistedmanyalternativewaystospendqualitytimethatwere
notasexpensive.Anotherpersonmightdecide that theirparents’ choosing tobond
overrecreationalhuntingtripswereinfactmoreharmfulthanbeneficial,andthat,in
any case, they have moral objections to having been subjected to that activity. As
206
showninSection3.7,theSharedPreferenceViewoffersawayoutoftheobligationto
paytheirparentsbackfortheseactivities,ifitturnsoutthatgrownchildrencannotbe
ascribedtheFreeRider’sPreferencewithrespecttothem.
More importantly, theSharedPreferenceView isnot indirect competitionwith the
debttheory,thoughthereasonwhyitisnotdiffersfromthereasonwhyitwasnotin
competitionwith the gratitude theory. The Shared Preference View can be seen as
providing thedeepernormative story forwhyandwhenadebt is incurred.For the
debttheoryitselfseemstolackthenormativeexplanationforwhy,exactly,andunder
what conditions, we should hold children to a debt, beyond relying on intuitive
notionsofreciprocity.245
6.4.ConclusionInthischapterIhaveconsideredonelastversionoftheFairPlay-basedpro-sharing
argumentforparentaljustice.Accordingtothisview,everyoneinsociety,parentsand
non-parentsalike,hasanobligationtopayforthecostsofchildreninvirtueofhaving
benefitedfromanupbringingthemselves.Asthebestdevelopedversionofthisview
todate, IhavecriticallyassessedPatrickTomlin’sKidsPayview,andIhavetriedto
showthatitfailstosecurethepro-sharingconclusionitsetsouttoestablish.TheKids
PayviewmaintainsthatchildrenhaveaFairPlayobligationtohelppayforthecosts
ofaminimallydecentupbringing that theirparentssecured for them,provided it is
not true that parents owed them these benefits as compensation for the harm of
bringingthemintoexistence. Inresponse, Ihavearguedthatwhileparentsmaynot
owetheirchildrencompensation,theyneverthelesshaveanobligationtopreventthe
severerisksofharmthatanyexistencecomeswith fromeventuating. I thenargued
thatprovidingthegoodsofaminimallydecentupbringingispartofthesepreventive
245SeePartIofBlustein1982.
207
obligations, and that discharging these obligations is incompatiblewith raising Fair
Playcomplaintsagainstchildren.Ihavethenshownthat,onmyproposedviewofFair
Play,theSharedPreferenceView,theremaybesomeFairPlayclaimsthatparentscan
raiseagainst(grown)children,butnotwithrespecttotherangeofbenefitswebelieve
childrenareowedbytheirparents.Parentscanonlyraiseclaimsoffairnessforavery
restrictedrangeofbenefitsthatboththechildandtheparentdirectlybenefitfrombut
forwhichparentsbeardisproportionateburdens.Thesearethegoodsoftheparent-
child relationship that both parents and children have an interest in enjoying, but
which parents are not under an obligation to provide. Precisely because these are
relationshipgoodsthatfalloutsidethescopeofjustice,however,thesortsofFairPlay
obligations established here are not enforceable. Finally, the conclusions of this
chaptercancomplementthoseaccountsoffilialobligationaccordingtowhichgrown
childrenowegratitudeorreciprocity to theirparentsonlywhen theirparentswent
beyondthecallofduty.
208
7.Conclusion
Thedecisiontobecomeaparentissurelyoneofthemostimportantlifedecisionsany
personcanmake.Creatingandnurturingnewlifeinvariablybringsimportantbenefits
and burdens to parents. Some might even say it transforms them as persons.246
Having and raising children, then, seems to be not only a decision of great
responsibility and, possibly, of high rewards for parents, but also of deeply
transformativepotential.
Yetwhatseemstobeadecidedlyintimatematterhassubstantialpublicimplications
aswell.Forsomeone’sprivatedecisiontocreateandraiseachildalsomeans,atthe
same time, adding a new member to the community of free and equal persons to
which one belongs. Bringing new life into the world, then, creates benefits and
burdens forboth thosewhoaredirectly involved, but also for society at large.This
raises an important question of justice, namely what is the just distribution of the
benefitsandburdensthathavingchildrengenerates?
This project has focused on the second part of that question, namely on the just
distribution of the burdens of having children, and in particular on one important
considerationthatshouldfeedintoanall-things-consideredanswertoit:doesajust
liberal egalitarian society owe it to the parents to share some of the burdens of
parenthoodwiththem?
The thesis explored the most promising positive answers to this question and
concludedthat theprospects forgroundingdistinctiveparentalclaimstohavingthe
246Paul2014,p.16.
209
costsofchildrensharedarevery limited,withthemostpromisingavenuebeingthe
appealtothevalueofpersonalautonomy.Arestrictedversionoftheautonomy-based
argumentthatAnneAlstotthasdevelopedisdefensible:itseemsundeniablethatthe
full-time carers of newborns and young children suffer a restricted life in terms of
opportunities forself-authorshipunless theirburdensarealleviated.However,even
heremoreworkneeds tobedone toestablish that theseautonomydeficitsprovide
full-time carers with a complaint against the state. What is needed is a plausible
account ofwhat the liberal state’s commitment to protecting its citizens’ autonomy
amountsto.
The most important positive contribution this thesis makes is to develop a new
accountoftheprincipleofFairPlay.Thiscontributionisimportantforthreereasons.
First, itprovidesasystematicdefenceofFairPlayasaplausibleprincipleofspecial
obligation in the face of widespread skepticism about its plausibility, which is a
significanttheoreticalgain in itself.Secondly, itweighs inononeofthemostwidely
endorsed family of arguments of parental justice according to which parents as
providersofchildren-as-public-goodscanraiseFairPlay-basedcomplaints that they
arethevictimsofunfairfreeridingbytherestofsociety.OntheviewofFairPlaythat
I defend, however, parents do not meet the necessary conditions for having such
claims.Thirdly,theFairPlayprinciplehaswiderimplications,reachingfarbeyondthe
parental justice debate. Chief among these is its potential to provide a basis for
politicalobligation,asarguedinSection4.6.
One interesting set of questions that the thesis does not explore, but whichwould
deserve attention in future work, concerns the implications of the arguments
developed in this thesis for adoptive parents. Throughout the thesis I haveworked
with a definition of “parents” as being both the procreators of, and the custodial
parentsto,theirchildren.Althoughthisremainsthestandardcaseinreality,thereare
countlesscaseswherethisdoesnothold:procreatorsmaynotendupparentingtheir
biological children for a variety of reasons, and many custodial parents are their
210
children’s adoptive, as opposed to biological, parents. Acknowledging this more
complexpicturecanhavesomeimportantimplicationsforwhatthejustdistribution
of the costs of children should look like. For instance, in Chapter 6 I argued that
parents (those who are both the procreative and custodial parents) lack claims of
fairness against their children for the range of morally required goods that they
provide themwithbecause theyhavea specialobligation toprovide thosegoods in
firstplace, invirtueofhavingbroughtthechildrenintotheworld.Butiftheparents
arenot, in fact, their children’sprocreators, this argumentdoesnothold.Thisdoes
not immediately entail that adoptive parents have claims of fairness against their
childrenforprovidingthemwitharangeofmorallyrequiredbenefits,foritmaystill
be the case that society ought to hold only the procreators liable for bearing those
costs.Still,itmaymakeanimportantdifferencethatitisprocreatorswhohavetopay
to give children the morally required benefits of an upbringing, and not adoptive
parents.
Anotherwayinwhichdistinguishingbetweenprocreativeandadoptiveparentsmight
havenoteworthyimplicationsisthatperhapsacasecouldbemadethatthereexistsa
collective duty to adopt orphaned children that adoptive parents are intentionally
discharging on behalf of the entire community. It would be worth investigating
whethertheSharedPreferenceViewcouldgroundfairness-basedentitlementsonthe
partofadoptiveparentsagainsttherestofsociety.Inanyevent,treatingprocreative
andadoptiveparentsasseparatetypesofagentsthatmightbearliabilityforthecosts
of children raises interesting issues, of both parental justice and procreative ethics,
thatmeritfurtherexploration.
In closing, I wish to acknowledge what I consider to be the twomost problematic
theoretical implicationsoftheviewthata just liberalegalitariansocietyshouldhold
parentsresponsibleforbearingallthecostsofchildren.ThroughoutthethesisIhave
beenassuming that the costsof children includeboth themorally required costsof
care aswell as the costs that children imposeonce theybecomeadults, namely the
211
costsrequiredtogivethemtheirfairshareofwhateverjusticerequirestheyreceive
as equal members of society. As Olsaretti has pointed out,247if parents bear the
primary responsibility formeeting all of these costs as amatter of basic justice, it
followsthatnoonehasbasicclaimsofjusticeagainsttheirfellowcitizens,butonly,in
thefirstinstanceatleast,againsttheirparents.Andthis,Olsarettiargues,goesagainst
the egalitarian commitment to the value of solidarity expressed in sharing in each
other’sfortuneandmeetingeachother’segalitarianclaimsofjustice.248
The second challenge comes from Clare Heyward who argues that holding only
parentsliableforbearingthecostsofchildrenisincompatiblewithtreatingthenew
memberswith the respect they are owed as free and equal citizens.249For insisting
that parents ought to internalize all the costs of children sends the stigmatizing
message that thesenewmembersarean imposition,oraburden,onsociety, that it
would have been in one way better if they did not exist, and that they are not
considered part of a community where people are sensitive to each other’s basic
claims of justice. Heyward argues, then, that treating the new members with
appropriate respect requires bearing the costs of children as a society, parents and
non-parentsalike.
IacknowledgetheforceofbothchallengesandIagreethateffectivelycancellingthe
commitmenttomeetingoneanother’sclaimsofjusticeasequalmembersofapolitical
community, and/or treating new members as the negative externalities of their
parents’ambitionisanundesirableplacetoendupin.Onereplythatdefendersofthe
anti-sharing view could explore is to argue that the costs of children that parents
should be held responsible for include only the costs of care, but not the costs of
added adult members to society. In this way, anti-sharing views could make some
space for acknowledging that, once they reach adulthood, children stop being their
parents’ responsibility and enter the social cooperation as moral equals who have247Olsaretti2017,p.165.248Olsaretti2017,p.165.249Hewyard2012,pp.718-719.
212
claimsofegalitarian justiceagainsttheothermembers.Whetherthisdistinctioncan
plausiblybedrawnisworthexploringinfuturework,bothinvirtueof itsrelevance
for thequestionofparental justice,butalsobecause itwouldhavebearingonsome
insufficiently explored issues of procreative ethics. For instance, it is not clear
whether procreators’ responsibilities toward their children end once the children
reachadulthood,orperhapsonlyasubsetofthemdo,andwhichonesthoseare.
Toconclude,Iwouldremarkthattheall-things-consideredanswertothequestionof
who shouldpay for the costs of children should avoidwhat I agree are implausible
implicationsoftheanti-sharingview.Thatis,weoughttoacknowledgeoneanother’s
claims of justice as fellow (adult) members of the community, and we should also
avoid sending themessage that childrenareaburden imposedon the restofusby
theirparents.ButIwouldalsonotethatthisisnotsomethingthatsocietyowestothe
parents.Itissomethingowedtothenewcitizensthemselves,whohavenotaskedto
be brought into existence butwho, once here, have a right to be treated as equals,
whateverthatentails.So,itmayberightthat,whenthepuzzleofthejustdistribution
of the costs of children is complete, it turns out all of us, parents and non-parents,
must share someof the costs of children.But ifwe, as a society, failed todo so for
somereason,wewouldnotbe failing theparents.Wewouldbe failing the children
themselves.
213
Bibliography
Ackerman, Bruce. 1980. Social Justice and the Liberal State. New Haven, CT: Yale
UniversityPress.
Alstott ,Anne.2004.NoExit:WhatParentsOweTheirChildrenandWhatSocietyOwes
Parents.Oxford:OxfordUniversityPress.
American Psychological Association 2009. Gender equality in caregiving: TheUnited
Nations response. UN Report.Washington D.C: American Psychological
Association https://www.apa.org/international/pi/2009/07/un-gender (last
accessed:01/06/2019).
Anderson,Elizabeth.1999.WhatisthePointofEquality?Ethics,109,287-337.
Archard,David. 2010. TheObligations andResponsibilities of Parenthood. Pp. 103–
127inD.ArchardandD.Benatar(eds),ProcreationandParenthood:TheEthics
ofBearingandRearingChildren.Oxford:OxfordUniversityPress.
Arneson,Richard J.1982.ThePrincipleofFairnessandFree-RiderProblems.Ethics,
92,616–633.
Arneson, Richard J. 1989. Equality andEqualOpportunity forWelfare.Philosophical
Studies,56,77–93.
Arneson,RichardJ.2004.MoralLimitsontheDemandsofBeneficence?Pp.33-58
inD. K. Chatterjee (ed.) The Ethics of Assistance. Cambridge: Cambridge
UniversityPress.
Arneson,RichardJ.2014.WhatdoOwetoPoorFamilies?Law,EthicsandPhilosophy,
2,7–31.
Barnes, Winston Herbert Frederick. 1945. Symposium: Intention, Motive, and
Responsibility. Supplementary Volume XIX of the Aristotelian Society, cited in
214
Ardal, Pall S. 1965. Motives, Intentions and Responsibility. The Philosophical
Quarterly,15,146–154.
Benatar,David.2006.BetterNevertoHaveBeen:TheHarmofComingintoExistence.
Oxford:OxfordUniversityPress.
Benson,Paul.1994.FreeAgencyandSelf-Worth.JournalofPhilosophy,91,650–668.
Berger,Fred.1975.Gratitude.Ethics,85,298–309.
Blustein, Jeffrey. 1982.Parentsandchildren:Theethicsof the family. Oxford: Oxford
UniversityPress.
Boran,Idil.2006.Benefits,Intentions,andthePrincipleofFairness.CanadianJournal
ofPhilosophy,36,95-115.
Bou-Habib, Paul. 2012. Parental subsidies: The argument from insurance. Politics,
Philosophy&Economics,12,197-216.
Bou-Habib, Paul and Olsaretti, Serena. 2013. Equality, Autonomy and the Price of
Parenting.TheJournalofSocialPhilosophy,44,420-438.
Brighouse, Harry, and Swift, Adam. 2014. Family Values: The Ethics of Parent-Child
Relationships.Princeton:PrincetonUniversityPress.
Bubeck, Diemut Elisabeth. 1995.Care,Genderand Justice. Oxford: OxfordUniversity
Press.
Burley, Justine.2000.Thepriceofeggs. InJ.HarrisandS.Holm(eds.),TheFutureof
HumanReproduction:Ethics,Choice,andRegulation.Oxford:OxfordUniversity
Press.
Callahan, Daniel. 1985.What do children owe elderly parents?TheHastings Center
Report,15,32–37.
Casal, Paula. 1999. Environmentalism, Procreation, and the Principle of Fairness.
PublicAffairsQuarterly,13,363-376.
Casal, Paula andWilliams, Andrew. 1995. Rights, equality, and procreation.Analyse
undKritik,17,93-116.
215
Casal, Paula and Williams, Andrew. 2004. Equality of Resources and Procreative
Justice.Pp.150-169inJ.BurleyandR.Dworkin,DworkinandHisCritics:With
RepliesbyDworkin.Malden,MA:Blackwell.
Christman,John.1988.Constructingtheinnercitadel:Recentworkontheconceptof
autonomy.Ethics,99,109–124.
Clayton, Matthew. 2006. Justice and Legitimacy in Upbringing. Oxford: Oxford
UniversityPress.
Clayton,Matthew.2015.HowMuchDoWeOwetoChildren?InS.Hannan,S.Brennan,
and R. Vernon (eds.), Permissible Progeny? The Morality of Procreation and
Parenting.NewYork,NY:OxfordUniversityPress.
Cohen,G.A.1989.OntheCurrencyofEgalitarianJustice.Ethics,99,906–944.
Colburn,Ben.2010.AutonomyandLiberalism.NewYork:Routledge.
Coleman, Jules and Ripstein, Arthur. 1995. Mischief and Misfortune. McGill Law
Journal,41,91-130.
Cullity,Garrett.1995.MoralFreeRiding.Philosophy&PublicAffairs,24,3–34.
Cullity,Garrett.2008.Publicgoodsandfairness.AustralasianJournalofPhilosophy,86,
1–21.
Daniels,Norman.1988.AmImyparents’keeper?Anessayonjusticebetweentheyoung
andtheold.Oxford:OxfordUniversityPress.
Dear Sugar podcast, Episode 36. Mothers who hate motherhood.
http://www.wbur.org/dearsugar/2015/12/11/dear-sugar-episode-thirty-six
(lastaccessed:01/06/2019).
Dworkin,Gerald.1988.TheTheoryandPracticeofAutonomy.Cambdrige:Cambridge
UniversityPress.
Dworkin, Ronald. 1981.What is Equality? Part 2: Equality of Resources.Philosophy
andPublicAffairs,10,283-345.
Dworkin, Ronald. 2000. Sovereign Virtue: the theory and practice of equality.
Cambridge,Mass.;London:HarvardUniversityPress.
216
Dworkin,Ronald.2004.ReplytoPaulaCasalandAndrewWilliams.Pp.361-362inJ.
Burley and R. Dworkin, Dworkin and His Critics: With Replies by Dworkin.
Malden,MA:Blackwell.
Dworkin, Ronald. 2011. Justice for Hedgehogs. Cambridge, MA: Belknap Press of
HarvardUniversityPress.
English, Jane.1979.Whatdogrownchildrenowetheirparents?InO.O’NeillandW.
Ruddick (eds.), Having Children: Philosophical and Legal Reflections on
Parenthood.Oxford:OxfordUniversityPress.
Esping-Andersen,Gosta.2009.TheIncompleteRevolution.Cambridge:PolityPress.
Feinberg, Joel. 1992. The Child’s Right to an Open Future. In his Freedom and
Fulfillment.Princeton,NJ:PrincetonUniversityPress.
Fleurbaey,Marc.1995a.EqualityandResponsibility.EuropeanEconomicReview, 39,
683-689.
Fleurbaey,Marc.1995b.EqualOpportunityorEqualSocialOutcome?Economicsand
Philosophy,11,25–55.
Folbre,Nancy.1994.Childrenaspublicgoods.AmericanEconomicReview,84,86–90.
Folbre,Nancy.2002.TheInvisibleHeart:EconomicsandFamilyValues.NewYork,NY:
TheNewPress.
Folbre,Nancy.2008.ValuingChildren.Cambridge,MA:HarvardUniversityPress.
Frankfurt,HarryG.1971.FreedomoftheWillandtheConceptofaPerson.Journalof
Philosophy,68,5-20.
Gauthier,David.1986.MoralsByAgreement.Oxford:OxfordUniversityPress.
George,Rolf.1987.WhoShouldBeartheCostsofChildren?,PublicAffairsQuarterly,1,
1-42.
Gheaus,Anca.2015.CouldThereEverBeaDuty toHaveChildren? In S.Hannan, S.
BrennanandR.Vernon(eds.),PermissibleProgeny?TheMoralityofProcreation
andParenting.Oxford:OxfordUniversityPress.
Goodin, Robert. 2005. Responsibilities for Children’sWell-Being. InM. Prior and S.
Richardson (eds.), No Time to Lose: The Wellbeing of Australia’s Children.
217
Melbourne:MelbourneUniversityPress.
Gornick, Janet.C.,andMeyersMarcia.K.2004.SupportingaDual-Earner/Dual-Carer
Society:LessonsfromAbroad.InJ.HeymannandC.Beem(eds.),ADemocracy
thatWorks: The PublicDimensions of theWork and FamilyDebate. NewYork,
NY:TheNewPress.
Gosseries,Axel.2009.ThreeModelsof IntergenerationalReciprocity. InA.Gosseries
and L. H. Meyer (eds.), Intergenerational Justice. Oxford: Oxford University
Press.
Halliday, D. 2016. Private education, positional goods, and the arms race problem.
Politics,Philosophy&Economics,15,150-169.
Hampton,Jean.1993.SelflessnessandtheLossofSelf.SocialPhilosophyandPolicy,10,
135–165.
Hart,H.L.A.1955.AreThereAnyNaturalRights?ThePhilosophicalReview64,175–
191.
Heyward, Clare. A Growing Problem? Dealing with Population Increases in Climate
Justice.EthicalPerspectives,19,703-732.
Hill, Thomas E. 1991. Autonomy and Self-respect. Cambridge: Cambridge University
Press.
Hirsch,Donald.2017.TheCostofaChildin2017.London:ChildPovertyActionGroup.
Availableat:
http://www.cpag.org.uk/sites/default/files/TheCostofaChildin2017.pdf(last
accessed:01/06/2019).
Hoem, JanM.1990. SocialPolicyandRecentFertilityChange inSweden.Population
andDevelopmentReview,16,pp.735-748.
Hoem,JanM.2005.WhydoesSwedenhavesuchhighfertility?DemographicResearch,
13,559-572.
Hurka,Thomas.1993.Perfectionism.NewYork:OxfordUniversityPress.
218
Jeske,Diane. 2019. FilialDuties. InA.Gheaus,G. Calder and J.DeWispelaere (eds.),
TheRoutledgeHandbookofthePhilosophyofChildhoodandChildren.NewYork:
Routledge.
Keller,Simon.2006.Fourtheoriesoffilialduty.ThePhilosophicalQuarterly,56,254–
274.
Klosko, George. 1987. The Principle of Fairness and Political Obligation. Ethics, 97,
353–362.
Klosko,George.1992.ThePrincipleofFairnessandPoliticalObligation. Lahnam,MD:
Rowman&Littlefield.
Lenman, James. 2002.OnBecomingExtinct.PacificPhilosophicalQuarterly,83, 253-
269.
Lino, Mark, Kuczynski, Kevin, Rodriguez, Nestor and Schap, Tusa Rebecca. 2017.
Expenditures on Children by Families, 2015. Miscellaneous Publication No.
1528-2015. U.S. Department of Agriculture, Center for Nutrition Policy and
Promotion.
Lyons,David.1965.FormsandLimitsofUtilitarianism.Oxford:ClarendonPress.
Maclntyre,Alasdair.1984.AfterVirtue,2ndedn.NotreDame,IncL:UniversityofNotre
Dame.
Mackenzie,Catriona.2014.ThreeDimensionsofAutonomy:ARelationalAnalysis.In
A. Veltman and M. Piper (eds.), Autonomy, Oppression, and Gender. Oxford:
OxfordUniversityPress.
Mackenzie,CatrionaandStoljar,Natalie.2000.Introduction:AutonomyRefigured.In
C.Mackenzie,andN.Stoljar(eds.), RelationalAutonomy:FeministPerspectives
onAutonomy,Agency,andtheSocialSelf.Oxford:OxfordUniversityPress.
Magnusson, Erik. 2018. Parental Justice and the Kids Pay View. Ethical Theory and
MoralPractice,21,963–977.
McDermott,Daniel.2004.Fair-PlayObligations.PoliticalStudies,52,216–232.
McMahan,Jeff.1981.ProblemsofPopulationTheory,Ethics,92,96–127.
219
McTernan,Emily.2013.Theinegalitarianethos:Incentives,respect,andself-respect.
Politics,Philosophy&Economics,12,93–111.
Meijers, Tim. 2016. Justice inProcreation.FiveEssaysonPopulationSize,Parenthood
andNewArrivals.PhDmanuscript,UniversitéCatholiquedeLouvain.
MeryllLynchandAgeWave2018.TheFinancialJourneyofModernParentingReport.
https://www.ml.com/the-financial-journey-of-modern-parenting.html (last
accessed:01/06/2019).
Meyers, Natalie T. 1989. Self, Society and Personal Choice. New York: Columbia
UniversityPress.
Nagel, Thomas. 2002. Concealment and exposure & other essays. Oxford: Oxford
UniversityPress.
Narveson, Jan. 1978. Future People and Us. In R. I. Sikora and B. Barry (eds.),
ObligationstoFutureGenerations.PhiladelphiaPA:TempleUniversityPress.
Normore,CalvinG.2010.Consentand thePrincipleofFairness. InC.Favor,G.Gaus
andJ.Lamont(eds.),EssaysonPolitics,Philosophy,&Economics:Integration&
CommonResearchProjects.Stanford,CA:StanfordUniversityPress.
Nozick,Robert.1974.Anarchy,State,andUtopia.NewYork:BasicBooks.
Nussbaum, Martha. 2006. Frontiers of Justice. Disability, Nationality, Species
Membership.Cambridge,MA:HarvardUniversityPress.
Nussbaum, Martha and Sen, Amartya. (eds.) 1993. The Quality of Life. Oxford:
ClarendonPress.
OECD 2017. Education at a Glance: OECD Indicators. OECD Publishing, Paris.
http://dx.doi.org/10.1787/eag-2017-en(lastaccessed:01/06/2019).
Olsaretti,Serena.2009.ResponsibilityandtheConsequencesofChoice.Proceedingsof
theAristotelianSociety,Vol.CIX,Part2,165-188.
Olsaretti,Serena.2013.ChildrenasPublicGoods?Philosophy&PublicAffairs41,226–
258.
Olsaretti, Serena. 2017. Children as Negative Externalities. Politics, Philosophy &
Economics,16,152-173.
220
Olson, Mancur. 1965. The Logic of Collective Action: Public Goods and the Theory of
Groups.Cambridge,MA:HarvardUniversityPress.
O’Neill, Onora. 1979. Begetting, Bearing and Rearing. In O. O’Neill andW. Ruddick
(eds.),HavingChildren:PhilosophicalandLegalReflectionsonParenthood.New
York:OxfordUniversityPress.
Parfit,Derek.1984.ReasonsandPersons.Oxford:OxfordUniversityPress.
Okin,Susan.M.1989.Justice,GenderandtheFamily.NewYork,NY:BasicBooks.
Overall,Christine.2012.WhyHaveChildren?TheEthicalDebate,Cambridge,MA:The
MITPress.
Paul,L.A.2014.TransformativeExperience.Oxford:OxfordUniversityPress.
Porter,Lindsey.2014.WhyandHowtoPreferaCausalAccountofParenthood.Journal
ofSocialPhilosophy,25,182–202.
Post, S. 1989. What children owe parents: Ethics in an aging society. Thought:
FordhamUniversityQuarterly,64,315–325.
Quong, Jonathan. 2011. Liberalismwithout Perfection. New York: Oxford University
Press.
Rakowski,Eric.1991.EqualJustice.NewYork:OxfordUniversityPress.
Rawls,John.1964.LegalObligationandtheDutyofFairPlay.InS.Hook(ed.),Lawand
Philosophy.NewYork:NewYorkUniversityPress.
Rawls,John.1993.PoliticalLiberalism.NewYork,NY:ColumbiaUniversityPress.
Rawls,John.1999a.ATheoryofJustice,rev.edn.Cambridge,MA:BelknapPress.
Rawls,John.1999b.TheLawofPeoples,rev.edn.Harvard:HarvardUniversityPress.
Raz,Joseph.1986.TheMoralityofFreedom.Oxford:OxfordUniversityPress.
Renzo,Massimo.2013.Fairness,self-deceptionandpoliticalobligation.Philosophical
Studies169,467–488.
Roemer, John. E. 1993. A Pragmatic Theory of Tesponsibility for the Egalitarian
Planner.Philosophy&PublicAffairs22,146–166.
Roemer, John. E. 1996. Theories of Distributive Justice. Cambridge, MA: Harvard
UniversityPress.
221
Roemer, John. E. 1998.Equality ofOpportunity. Cambridge,MA: Harvard University
Press.
Rose,Julie.2016.FreeTime.Princeton:PrincetonUniversityPress.
Scanlon,ThomasM.1998.Whatweowetoeachother.Cambridge,MA:BelknapPress.
Scanlon, ThomasM. 2008.Moral Dimensions: Permissibility, Meaning, Blame.
Cambridge,MA:BelknapPress.
Scheffler,Samuel.2013.DeathandtheAfterlife.NewYork:OxfordUniversityPress.
Sen,Amartya.1992.InequalityReexamined.Oxford:ClarendonPress.
Shelby, Tommie. 2018.DarkGhettos. Injustice, Dissent, andReform. Cambridge, MA:
BelknapPress.
Shiffrin,Seana.1999.Wrongfullife,procreativeresponsibility,andthesignificanceof
harm.LegalTheory,5,117–148.
Simmons,JohnA.1979.ThePrincipleofFairPlay.Philosophy&PublicAffairs8,307–
337.
Simmons,John.A.2001.JustificationandLegitimacy:EssaysonRightsandObligations.
NewYork:CambridgeUniversityPress.
Smilansky, Saul. 1995. Is There a Moral Obligation to Have Children?, Journal of
AppliedPhilosophy,14,41-53.
Steiner,Hillel.1997.Choiceandcircumstance.Ratio,10,296–312.
Steiner, Hillel and Vallentyne, Peter. 2009. LibertarianTheories of Intergenerational
Justice.InA.GosseriesandL.H.Meyer(eds.),IntergenerationalJustice.Oxford:
OxfordUniversityPress.
Stemplowska,Zofia.2009.MakingJusticeSensitivetoResponsibility.PoliticalStudies,
57,237–259.
Stemplowska,Zofia,andSwift,Adam.2012.IdealandNonidealTheory.Pp.373-390in
D. Estlund (ed.),TheOxfordHandbook of Political Philosophy, Oxford: Oxford
UniversityPress.
Sverdlik,Steven.1996.MotiveandRightness.Ethics,106,327-349.
Temkin,Larry.1993.Inequality.NewYork:OxfordUniversityPress.
222
Tomlin,Patrick.2015.Shouldkidspaytheirownway?PoliticalStudies,63,663–78.
Tosi,Justin.2018.RethinkingthePrincipleofFairPlay.PacificPhilosophicalQuarterly.
Epubaheadofprint12March2018.DOI:10.1111/papq.12219.
U.S. Department of Education, National Center for Education Statistics. 2018.The
Condition of Education 2018(NCES 2018-144), https://nces.ed.gov/
programs/coe/indicator_cmb.asp(lastaccessed:01/06/2019).
Vallentyne,Peter.2002.Equalityand thedutiesofprocreators. InD.ArchardandC.
Macleod (eds.), The moral and political status of children. Oxford: Oxford
UniversityPress.
Velleman,J.David.2008a.PersonsinProspectI:TheIdentityProblem.Philosophyand
PublicAffairs,36,221–244.
Velleman, J. David. 2008b. Persons in Prospect II: The Gift of Life. Philosophy and
PublicAffairs,36,245–266.
Wellman,Cristopher.2005.SamaritanismandtheDutytoObeytheLaw.Pp.1-90in
Wellman, C. and Simmons, J. Is There a Duty to Obey the Law?. Cambridge:
CambridgeUniversityPress.
Wicclair,Maria.1990.Caringforfrailelderlyparents:Pastparentalsacrificesandthe
obligationsofadultchildren.SocialTheoryandPractice,16,163–189.
Young, Iris M. 1990. Justice and the Politics of Difference. Princeton: Princeton
UniversityPress.
Zhu, Jiafeng.2014.Fairness,PoliticalObligation,and the JustificatoryGap. Journalof
MoralPhilosophy12,290–312.
Zola, Charles. 2001. Geriatric filial piety. International JournalofAppliedPhilosophy,
15,185–203.