WhatDoWeOweParents?DistributingtheCostsofChildren

IsabellaAna-MariaTrifan

TESIDOCTORALUPF/2019

DIRECTORADELATESI

Prof.SerenaOlsaretti

DEPARTAMENTDEDRET

ii

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.

v

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.

viii

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

xi

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.