The Arrow of Care Map: Abstract Care in Ideal Theory


 
 
 
Feminist Philosophy Quarterly 
 

 
Volume 3 | Issue 4 Article 5 

 
  
 

2017  
 

The Arrow of Care Map: Abstract Care in  
Ideal Theory 
 
Asha L. Bhandary  
University of Iowa, asha-bhandary@uiowa.edu 

 
 
 

 
 
 
 

 
Recommended Citation 

Bhandary, Asha L.. 2017. "The Arrow of Care Map: Abstract Care in Ideal Theory."Feminist Philosophy Quarterly3, (4). Article 5. 



The Arrow of Care Map: Abstract Care in Ideal Theory1 
Asha Bhandary 

 
 
 

Abstract 
This paper advances a framework to conceptualize societal caregiving 

arrangements abstractly. It is abstract in that it brackets the meaning of our 
particular relationships. This framework, which I call the arrow of care map, is a 
descriptive tracking model that is a necessary component of a theory of justice, but 
it is not a normative prescription in itself. The basic idea of the map is then multiply 
specifiable to track various ascriptive identity categories as well as different 
categories of care labor. In this way, the idea of the arrow of care map serves as a 
conceptual frame with which to identify societies’ caregiving arrangements. I 
characterize it as a component of “ideal theory” insofar as it seeks to clarify our 
values without the immediate aim of formulating principles to govern a just society. 
The resultant partial theory of justice is one that responds to nonideal theory’s 
critiques of the ideological nature of liberal contractarianism’s idealizations and 
abstractions while moving towards new visions of a just society. 
 
 
Keywords: ideal theory, nonideal theory, contractarianism, care, dependency care, 
liberal feminism, feminist liberalism, justice, distributive justice 
 
 
 

Care, compassion, and concern are crucial elements of meaningful intimate 
relationships (Held 2006; Tronto 1993; Ruddick 1989). In addition, the labor of 
hands-on care to meet a person’s basic needs often occurs within our closest 
relationships. However, caregiving also occurs in paid and professional contexts, 
which sometimes cross national boundaries through migrations of nannies and 
                                                 
1 I presented this paper to the Society for Analytical Feminism, Wellesley University, 
and as the 2017 Inaugural Robinson Grover Lecture at the University of Connecticut, 
where Elise Springer and Rik Hine offered interesting analyses of the view that I plan 
to pursue in future work. I thank the participants of each of these events. I owe a 
particular debt of gratitude to Cynthia Stark, Lisa Tessman, and Diana Tietjens 
Meyers for philosophically open-minded discussions about the nature of my project. 
The written version of the paper was improved by comments by Richard Fumerton, 
two anonymous reviewers, and the editors of this journal. 

1

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



other care workers. Moreover, even in the contexts of families and friends, care that 
is given and received can be structured by internal hierarchies and inequities.  

Various cultures and societies have different ways of organizing life around 
care, such as gendered arrangements, extended families, and socialized care. These 
arrangements have particular histories. The history of slavery in the US means that 
black women who tend to white women do so in a context that is racially and 
historically charged.2 In societies with extended family systems, systems of care 
provision are structured by gender oppression and internally structured hierarchies 
and norms. For instance, a daughter-in-law may provide elder care to her parents-in-
law, where that practice is linked to preferences for males in those cultures, or a 
youngest daughter may be designated as the caregiver of her own parents, a 
designation which is paired with a cultural custom that she not marry and have 
children of her own. 

Although there is an extensive feminist literature about the highly gendered 
nature of inequalities in responsibilities for providing care,3 we lack a unitary model 
with which to identify arrangements for care in diverse cultures and social forms—
one that goes beyond gender. This paper fills this gap by advancing a new abstract 
concept for tracking the distribution of caregiving labor in a society that I call the 
arrow of care map. It is an abstract representation of caregiving that brackets the 
meaning of the relationships within which it occurs and thereby enables critical 
distance from particular cultural biases about who ought to care for whom. The 
function of this map is to gain clarity about how caregiving actually occurs in 
societies, because transparency is a necessary step prior to theorizing about what a 
just society would be. Due to the pervasiveness of obfuscating narratives about who 
gives care and who receives care, it will be a substantial achievement to arrive at an 
accurate description of the ways that care is provided and received. 

In Section 1, I defend the need for this new concept. There, I also show how 
the arrow of care map is responsive to the nonideal theoretic criticism that ideal 
theory is ideological in virtue of the content of its idealizations and abstractions. 
Section 2 sets forth the core components of the arrow of care map and identifies the 
normative commitments embedded in the model. It is a relational analysis of a 
society’s caregiving arrangements founded on individual units of human beings. 
There, I also identify the map’s variations as specified by categories of labor and 
identity group, while emphasizing the need for the higher-order, abstract map as a 

                                                 
2 I thank Lindsey Stewart for discussion on this topic. 
3 For feminist liberal work that discusses care, see Abbey (2016), Baehr (2004, 2013, 
2017), Bhandary (2010, 2016a, 2016b), Cudd (2006), Dimock (2008), Hartley and 
Watson (2010, 2012), Lloyd (1998), Okin (1989), Schouten (2015), Stark (2007), 
Varden (2006). 

2

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



way to maintain cross-cultural flexibility to track various identity groups. Section 3 
identifies the limited conclusions that can be drawn about what a fair caregiving 
society will be prior to the descriptive tracking endeavor. Finally, in Section 4, I 
contrast the scope of the arrow of care map with both the ethics of care and forms 
of feminist contractarianism that apply a metric of justice directly to the family and 
personal relationships.  

 
1. The Need for a New Concept 
 
1.1. Oppression via Theoretical Invisibility 

Oppression via theoretical invisibility occurs when a person has a lived 
experience of disadvantage or exploitation that is unintelligible by the terms of the 
theory. In the lexicon of that theory, the person is either not disadvantaged, or the 
source of their disadvantage remains out of view even if they are disadvantaged 
according to the theory’s metric. For instance, an African American woman in the US 
might be economically disadvantaged—where the disadvantage is identifiable with a 
metric of income inequality—but the complex structural reasons that create 
disadvantage will not be wholly explicable in terms of income. As a result, an agenda 
for social change that follows from a theory that only tracks income inequality will 
overlook these structures of disadvantage. Thus, one of the tasks of a theorist of 
justice is to articulate frameworks that will identify these overlooked experiences. 

Kimberlé Crenshaw’s concept of “intersectionality” is a prime example of a 
concept that makes visible a set of issues that perhaps could have been triangulated 
from preexisting concepts, but would remain more difficult to theorize without it.4 
The idea of intersectionality is a way to understand the social position of women of 
color that reveals how the categories of race and gender obscure theorizing about 
their lives. Intersectional analyses point out how assumptions of race and gender 
result in the loss of data capture that is particular to the lives of women of color.5 
The articulation of the concept makes subsequent theorizing possible to further 
explore its various dimensions and significance.6 Concepts like intersectionality are 
abstract, insofar as they highlight one set of concerns to the exclusion of others.7 
Thus, the concepts of gender, race, and class have made many aspects of social 

                                                 
4 Crenshaw’s concept of intersectionality is a way to understand the invisibility of 
black women by naming their location and thereby rejecting the idea of “identity as 
woman or person of color as an either/or proposition” (Crenshaw 1991, 1242). 
5 Crenshaw explains the problem of analyses of racism and sexism as one in which 
all black people are men and all women are white (Crenshaw 2015). 
6 See P. Collins (2000). 
7 Similarly, Mills (1997) argues for the need for the concept of white supremacy. 

3

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



oppression visible, and the idea of intersectionality has both augmented and 
complicated them.  

Every culture or society has some way of organizing life around care, such as 
gendered arrangements, extended families, and socialized care. Despite the 
universality of care, though, caregiving arrangements are among the most 
theoretically and practically significant sources of blindness in liberal 
contractarianism. Therefore, if liberal contractarianism is to be a viable doctrine, it 
needs a way to conceptualize how care is provided. I schematize this understanding 
of a society-wide caregiving arrangement with the arrow of care map. The arrow of 
care map theorizes the labor of care—not the nuances of relations of love and 
affection—and it abstracts away from assumptions about family units or socialized 
care to offer a way to tune our intuitions about caregiving in human life. The 
framework embeds within it a normative claim that society must meet its vital care 
needs.8 In doing so, it eliminates the following form of exploitation from traditional 
forms of liberalism. More precisely, adding the arrow of care to an account of the 
just distribution of the benefits and burdens of social cooperation removes from 
theory the following form of exploitation as it occurs for care labor.  

 
~Persons X are assigned to do work C when work C is not recognized as work. 
Consequently, persons X look like non-contributors. 
~ Persons Y do work A, which is highly valued. They are able to gain 
additional advantages by assiduously avoiding work C. Their avoidance of C is 
not identifiable as shirking because C is not identified as work. 
 
If a person’s contribution to social cooperation is not identified as a 

contribution in a theory of distributive justice, and the benefits that some people 
receive are not identified as benefits, then the resultant theory will have a loophole 
that leads to advantages for the people who avoid that work. This is the pattern that 
many societies have adopted for the social contributions needed for repair and 
maintenance.9 Eva Kittay has argued that liberalism, in particular, results in 
disadvantages to dependency workers due to their invisibility in its basic 
assumptions and concepts (1999). 

 
 

                                                 
8 It proceeds from the premise, unargued for here, that societies need to meet their 
care needs, and that doing so plays a crucial role in systems of social cooperation. 
For that argument, see Bhandary (2016b). See also Reiheld (2015) for an argument 
that society has a duty towards caregivers. 
9 On repair and maintenance, see Tronto (1993), Spelman (2002), Walker (2007). 

4

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



1.2. Non-ideological Ideal Theory 
More broadly, liberal contractarianism has been criticized for idealizations 

and concepts that make the sources of group-based oppression impossible to 
identify or, more strongly, require the subordination of a subset of society (Pateman 
1988; Jaggar 1995; Kittay 1999; Held 1987; Mills 2004; Schwartzman 2006a). I agree 
with these critics that an ideal theory that refuses to scrutinize whether its core 
concepts result in the perpetuation of illicit privilege is a non-starter as a viable 
theory of justice. Despite these problems with the history of the contract tradition, I 
share the growing consensus that one of the main targets of the critique of ideal 
theory—the core liberal contractarian idea of fairness—retains value as a basis for 
identifying a number of contemporary forms of social injustice.10 Fairness, defined in 
a way that requires transparency, can rule out arrangements that exploit or 
systematically disadvantage a subset of society. A necessary, but not sufficient, step 
to rule out these exploitative arrangements is to develop concepts and claims that 
make visible what Lisa Schwartzman calls “the sources and manifestations of group-
based oppression” (2006a). Thus, the project of advancing non-ideological liberalism 
requires new theoretical work to better specify the model of the person, the facts 
known to deliberators, and the domain of the basic structure of society.11 
Idealizations that perpetuate ideology must be criticized, and better concepts for 
mapping the realities of existing societies are needed. The arrow of care map is a 
contribution to the project of advancing non-ideological ideal theory. 

 
2. The Arrow of Care Map: Overview of Functions 

The framework I advance is one that can be used for all social forms. It will 
be further specified by the culturally specific details that structure social inequality 
in that social form—where the assignment of care may track onto a person’s caste 

                                                 
10 Okin (1989), Hampton (1993), Hay (2013), and Mills (2012). I follow a broadly 
Rawlsian form of contractarianism (2001). For the criticism that contractarianism 
does not necessarily rule out oppression, see Schwartzman’s critique of O’Neill 
(2006b, 572). Specifying the precise form of contractarianism will be necessary for a 
complete account of just caregiving, but that is not the task of this paper. 
11 My usage of ideal theory here is compatible with Stemplowska (2008) and 
Valentini (2012). It differs from Mills (2004), who says that ideal theory is necessarily 
ideological theory, so that if a person addresses oppression, they are doing nonideal 
theory. Thus, if one accepts Mills’s definitions of ideal and nonideal theory, the 
arrow of care map will count as nonideal theory because it makes a source of 
oppression visible. See also Tessman (2009). For alternative formulations of the 
distinction, see Rawls (2002), Valentini (2012), Stemplowska and Swift (2015), and 
Kang (2016). 

5

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



heritage rather than race or gender. Thus, the basic model of the arrow of care map 
is universal but multiply specifiable.  

The arrow of care map, as a systems-level depiction of societal caregiving 
arrangements, gives theories of justice something they have been lacking: a way to 
understand care given and received as a fundamental social good that is the 
outcome of social cooperation, and that is prior to—and not dependent on—
contingencies such as the particular economic structure of the society. In virtue of 
bracketing relationships of particularity and intimacy as well as the robust meaning 
of giving and receiving care in human life, it is decisively abstract and moderately 
fact-sensitive—incorporating the most fundamental aspects of human life, while 
leaving aside others.12 Because the need to receive care is fundamental to all human 
lives, it is a fact that must be included on any list of basic facts employed in a theory 
of justice.13 In its most abstract form, it has value for a broad range of social forms 
to track who tends to receive large amounts of care and who tends to provide it. At 
one level less abstracted, it must track historically oppressed identity groups. I do 
not reduce it to these specific maps, though; instead, I insist on the need to maintain 
an abstract version of the map so that societal variations in the exploitation of 
caregivers can be found even when they do not coincide with dominant narratives. 
It may be the case that class has a higher correlation with caregiving than gender, 
for instance. Theorists need critical distance from cultural preconceptions about 
who ought to receive care and who ought to provide it because—as scholars of 
affect study show—our concern for others is shaped by gender, race, and class 
(Greyser 2017).  

Thus, it is crucial that a map that tracks care given and care received remains 
conceptually separable from the range of ascriptive factors that will correspond to 
inequities in care. Maintaining a separate metric may reveal unmapped social 
locations, such as a caregiver who may be from an otherwise advantaged group but 
who becomes disadvantaged as a caregiver.14 Therefore, in addition to serving as a 
new descriptive mapping concept, the arrow of care map makes it possible to 

                                                 
12 I therefore aim to respond to the problem that abstractions have too often 
masked false claims—that humans are maximally rational (O’Neill 1996), or that 
they fail to be explicit about what they bracket, and why (Schwartzman 2006a). My 
account is moderately fact-sensitive in that it includes a set of basic facts but 
brackets others. See also Farrelly’s characterization of the moderate fact-sensitivity 
of Rawls’s theory, which “acknowledges some moderate feasibility constraints (e.g. 
pluralism) but also employs a number of idealizing assumptions (e.g. society is 
closed, full compliance, etc.) when deriving the principles of justice” (2007, 844). 
13 For a lengthier argument for this claim, see Bhandary (2016b). 
14 See also Kittay’s “dependency workers” (1999). 

6

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



identify inequalities that are not already captured by economic analyses or racial 
hierarchies.  

Arrow of care maps have two functions. The first function of the maps is to 
organize fact-finding about societies’ caregiving arrangements. The basic map 
template will then be modified into multiple maps with overlays tracking trends by 
race, gender, geographic origin, and class. Its main purpose is to gain an 
understanding about trends in levels of care received and provided as a relational 
good. 

The second function of the maps is to guide thinking about what a fair 
arrangement will be. It is my view that there are likely many possibly fair 
arrangements. On my preferred view, deliberations should be guided by a broadly 
Rawlsian methodology, which evaluates the arrangement based on the idea that 
you could be anyone in it, variously interpreted as the claim that you could occupy 
any social position in it. However, the arrow of care map idea is compatible with a 
broad range of interpretations of transparency and fairness tests. What it rules out 
are arrangements that are based on, and require, the widespread exploitation of 
groups of people (minorities, women) in virtue of their roles as caregivers.15 After 
greater transparency is achieved, further deliberation will be required to 
imaginatively conceptualize potential caregiving arrangements as candidates for a 
fair arrangement. For instance, just caregiving arrangements may include nuclear 
families, extended families, and non-family-based care.  

 
2.1. Function One: A System-Wide Descriptive Mapping Concept for Caregiving 

Arrangements  
The arrow of care map does not seek to capture the whole truth of the role 

of caring relationships in human life, nor does it seek to prescribe how people 
should think about care in their private lives. It is a systems-level analysis of a society 
in the sense that it holds up caregiving arrangements as objects of deliberation in a 
way that facilitates clearer thinking about the different possible arrangements that 
would be acceptable to people in a society, when the labor of care is made visible.16 

                                                 
15 See Schwartzman’s (2006b) argument that a theory without ideals will not have 
resources to rule out oppression. Because the paper’s aim is to advance the idea of 
the arrow of care map—and not to defend a specific conception of fairness—I do 
not specify an ideal or respond to Schwartzman’s claim about the necessity of ideals 
here.  
16 The project is aptly characterized by Stemplowska’s characterization of the point 
of ideal theory as clarifying our values (2008). See also Tessman (2015) on the need 
for feminist theory that apprehends our normative realities even when there are no 
morally viable paths of action. 

7

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



The arrow of care map is properly classified as a “descriptive mapping concept” 
(Mills 2004, 174), insofar as it creates a concept to use in order to identify what 
occurs in the real world.  

Caregiving norms are deeply culturally entrenched; they are embedded 
within gender norms about who should care for whom—by, for example, bringing 
them tea. They suffuse assumptions about who deserves sympathy for their 
suffering. Therefore, identifying what actually occurs in the domain of care is a 
significant task of its own. The fact-finding task will give us real data with which to 
inform our social understandings about who provides care, who receives it, and at 
what levels. In the descriptive use, it remains somewhat abstract because it brackets 
information about particular relationships in order to gain an understanding of 
caregiving that does not simply legitimate existing assumptions about who should 
care for whom.  

Although the arrow of care map will promote greater clarity about the 
nature of a society’s caregiving arrangements, it does not purport to avoid all 
cultural biases. For instance, leaving nonhuman animals out of the metric is a 
normatively laden starting point. The arrow of care map does not avoid the 
possibility of embedding and legitimating existing biases, but it seeks to be 
transparent about the claims and assumptions it makes, and the reasons for those 
assumptions. It is precisely because of the cultural entrenchment of caregiving 
norms that we need to think abstractly about care. The preponderance of 
philosophical effort directed towards thinking about care counterfactually has been 
directed towards new action-guiding policy proposals. These proposals often embed 
within them assumptions of the immutability of other features of the status quo. 
Theories of justice need a more abstract methodology to think about care, and that 
is what the arrow of care achieves.  

However, because care ethicists and other feminists have roundly criticized 
abstraction, it is incumbent on me to clarify and defend the role of abstraction, and 
also to show how my account does not fall into the pitfalls of ideal theory as it has 
been criticized by nonideal theorists. Schwartzman represents the nonideal criticism 
of ideal theory as fundamentally a problem of misrepresentation: “[Ideal theory] 
misrepresents both agents and structures in ways that deny and obscure the 
realities of oppression” (2016, 7). Thus, I seek to avoid misrepresentation by 
characterizing the project and scope of the arrow of care map. The arrow of care 
map is an abstraction that brackets numerous features of our lived experiences of 
caregiving. It does not seek to characterize the complexity of lived experiences 
within caring relationships. Although it does not accurately represent the complexity 
of human agents, it offers a structural analysis of caregiving that can illuminate 
forms of oppression. It does so in a way that seeks to attend to the ways that 

8

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



abstractions and idealizations can be ideological to mask and perpetuate 
entrenched inequalities. 

 
2.2. Normativity in Abstraction 

The arrow of care map includes normativity at two levels: first, with the 
values and conceptualizations that are incorporated into the arrow of care, and then 
to guide and assess various caregiving arrangements. It includes value claims by 
taking individual human beings as what is fundamental in order to then depict the 
amount of care each person provides, for instance, over a lifetime. As a framework 
for tracking actual distributions, the arrow of care map will unveil the care people 
give and receive in a way that is individualistic but that also recognizes the relational 
nature of caregiving. The individualism of the framework is built into the metric of 
care that is given and received for and by each person. The relational nature of care 
given and received occurs in a set of arrows that specify from whom and towards 
whom care goes. Sometimes the arrows will connect two members of a family, at 
other times they will not connect them. Quantifying care in this way will lay bare 
distributive inequalities that have been masked by cultural narratives. Theorists 
must (a) peel away the norms of femininity that mask the fact that caring is labor, 
(b) debunk ideas of gender essentialism that naturalize the labor of care, and (c) 
contest narrative and social constructs about racial and ethnic identity that 
normalize, for instance, Western presumptions that women of color should tend to 
the care, maintenance, and repair needs of whites.  

Furthermore, assumptions about oppressive gender norms can leave 
undetected exploitative labor distributions that occur within same-sex marriages, or 
the subordination of a man within a heterosexual couple. Consequently, the map 
will require extensive fact-finding by sociologists and anthropologists who attend to 
dominant norms, but are aware that these norms will not always describe what 
actually occurs. 17 To see just how difficult it may be to identify actual distributions 
of care when they conflict with social narratives, consider the following case: An 
affluent, white stay-at-home mom who does not have a paid career is credited as a 
full-time caregiver, but closer analysis reveals that substantial amounts of care are 
provided by babysitters, or unpaid caregivers like a friend, mother, sister, neighbor, 
or older child.  

Mapping amounts of care given and received will also make evident the 
differential entitlements of groups of people occupying different social positions 

                                                 
17 Approximately 1.4 million children between the ages of eight and eighteen serve 
as caregivers for parents, grandparents or siblings with disabilities (New York Times, 
May 23, 2016, “Supporting Children Who Serve as Caregivers,” by Jane E. Brody). I 
thank Amy R. Baehr for bringing this statistic to my attention. 

9

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



marked by race, ethnicity, and histories of slavery and colonization. In these ways, 
the first usage of the arrow of care map satisfies a few of the nonideal theorists’ 
requirements: it is quite centrally about dependency care, it identifies a category of 
labor that has sometimes coincided with slavery, and it unmasks ideologies. In order 
to do this, arrow of care maps must specify the gender, race, ethnicity, citizen 
status, and class of care receivers and providers.18 There will be different arrow of 
care maps to identify these caregiving arrangements—as a short time slice—over a 
year, or over the lifetime of individuals. In addition, it will be a theoretically robust 
project to delineate the boundaries of a society in order to be able to identify the 
“global care chain” and care drains from one country to another (Hochschild 2000; 
Kittay 2008; Gould 2008).  

Even in its descriptive form, the arrow of care map is abstract because it 
brackets facts about our relationships, such as the love parents have for their 
children, and that children have for their parents. Moreover, it does not present 
action-guiding recommendations with respect to caregiving, and so it says nothing 
about how people should approach their caring relationships in their real lives.19  

 
2.3. Mapping Care  

The arrow of care map is fundamentally an idea that admits of increasing 
levels of sophistication. I specify some of its uses here only to make more concrete 
the idea at its core. Below, I offer a schematic depiction of the idea of the arrow of 
care map. This particular depiction does not describe an actual caregiving 
arrangement, nor does it specify an ideal. Instead, the idea of the arrow of care is a 
way of quantifying caregiving labor as it occurs across both public and private 
domains. It offers an idea of tracking caregiving that is analogous to the idea of 
tracking blood pressure. In addition, just as the metric used for blood pressure is 
independent of any particular person’s blood pressure measurement, the idea of 
the arrow of care is an abstract idea that is not equivalent to the sum of what occurs 
in every person’s life. 

The arrow of care map tracks the intensity and duration of care given and 
received over the course of a lifetime. The width of the arrow corresponds to the 
intensity of care given and the direction of the arrow identifies care going from one 
person and towards another person. The length of the arrow corresponds to the 
duration of care.  

 
 

 

                                                 
18 I am indebted to Amy Mullin for pressing me to make this claim abundantly clear. 
19 For further discussion of this claim, see Section 4.1. 

10

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



Figure A. The Arrow of Care Map 
 

 
 
 
 

 
2.4. Categories of Care 

Arrow of care maps will be specified to track different subsets of labor, from 
the narrow category of dependency care to the broader inclusive category of care, 
and even personal services. Dependency care is care that seeks to respond to a 
legitimate need. Dependency care is the hands-on labor without which a person 
would not survive. It includes, for instance, feeding a baby, helping a paraplegic 
person with toileting, and assisting a person with mental disabilities to obtain food 
and live safely (Kittay 1999). I define dependency care, in part, by the relative 
immediacy of the harm that the vulnerable person will experience when care is 
absent. A paradigm case is that of a baby left in a crib without a caregiver; he will 
not survive long because he needs significant care to survive. This definition of 
dependency care does not offer necessary and sufficient conditions. It follows my 
view (Bhandary 2016b) that the nature of caregiving will change as caregiving 
becomes transparent in society, and the definition is consequently provisional.  

The caring work the map tracks maintains its historic connection with the 
category of labor that women have typically performed as mothers, nannies, 
mammies, daughters-in-law, and so forth. However, my account of caring work 
excludes the activities of teaching children to read, write, and perform basic math.20  

                                                 
20 I disagree with Daniel Engster’s account of care (2007, 27), which includes these 
activities and is broader than my view. For Engster, “Caring may be defined as 
everything we do directly to help individuals to meet their vital biological needs, 

11

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



The aim of the activity must be meeting a basic need, but attempts that fail 
should still be tracked. Consequently, I do not build efficacy into the caregiving that 
is tracked because caregiving is performed at varying degrees of skillfulness. 
However, I agree with Daniel Engster that actions that can be successfully 
completed without satisfying biological and developmental needs of a person will 
not count as caregiving labor. Engster offers the example of a cook in a restaurant, 
who can prepare and sell a meal without caring for anyone, whereas a parent 
cannot nourish a child without being sure that he or she actually ingests the food.  

Of course, there is a commonsense threshold for an activity to count as 
caregiving. A person who says they intended to care for their baby but leaves the 
baby crying in the crib for days while enjoying time out in town with friends violently 
harms the child and clearly fails to care for her. Thus, intentions matter. Some cases 
are more difficult to assess. For instance, imagine that Eugene makes his daughter 
soup while she is ill. He has taken off a day from work to do so. But he then spills the 
hot soup on his daughter, requiring that she receive medical care from someone 
else and also that her mother now take off a day from her work. The daughter still 
has not eaten anything and she now has burns that will require additional care to 
treat, resulting in additional caregiving by others. How do we quantify the care? Do 
Eugene’s contributions count as care, despite his ineffectiveness? If effectiveness is 
a necessary condition, then Eugene’s labor will not count, but a distributive 
assessment should include his attempted care, and therefore his earnest efforts 
should be counted in some way, although the additional needs he created must also 
be tracked. 

For these reasons, I reserve “effective” as a modifier to care, instead of 
saying that care must be minimally effective for it to count as care. In contrast, Amy 
Mullin offers a normative account of care for children as an activity that should not 
increase their vulnerability by failing to understand their material, psychological and 
developmental needs (Mullin 2014). Mullin’s account of care tells us what good care 
is, but tracking labor for an abstract representation of a caregiving arrangement 
should also track the care that aimed at meeting the person’s needs despite its 
failure to do so. 

It is well established in medical contexts that the labor of doctors that seeks 
to help a patient counts as work, even if they fail to achieve this aim. Other forms of 
care should follow the same convention. Imagine a case in which Beatrice, a six-
year-old child, sees a physician for a skin rash. The physician, Dr. Lyons, 
misdiagnoses it as a bacterial infection and prescribes an antibiotic, but the 

                                                 
develop or maintain their basic capabilities, and avoid or alleviate unnecessary or 
unwanted pain and suffering, so that they can survive, develop, and function in 
society” (2007, 29). 

12

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



antibiotic results in a larger rash as a side effect. Dr. Lyons has not provided effective 
or good care. Nonetheless, the health care system does not deduct the hours the 
physician spent seeing the child from the overall number of hours of care provided 
by the physician during that week. There is value in separating a quantification of 
labor hours from the quality of the labor—for it allows us to track that a person has 
spent time and effort aiming at an outcome, even if that person has failed to achieve 
the outcome, or if they do it poorly.  

In this example, Dr. Lyons has acted in ways that should be counted in a 
distributive account of labor. He has also offered bad care by failing to meet the 
needs of the recipient of care. Good care is the proper aim of caregiving. The reason 
for permitting bad care to count as care in the arrow of care map is that the map 
seeks to capture the labor people perform. Assessments of labor do not nullify labor 
that fails to meet a metric of high quality—which is what is illustrated with the 
medical case. Of course, it is of the utmost importance that vulnerable and 
dependent persons need good caregivers, not caregivers who are ineffective, and so 
we need educational programs and apprenticeships to train excellence in caregiving.  

 
2.5. Map Typologies 
 
Unpaid dependency care 

There will be several iterations of the abstract but descriptive arrow of care 
maps. The least complex map may be one that tracks unpaid dependency care 
because it will identify labor that is not currently tracked in any systematic way in 
virtue of its exclusion from the market—for the market is among the main sources 
of socially tracking labor.  

 
Dependency care that is poorly paid 

The need to track dependency-care labor that is paid is evident in the case of 
dependency work that is paid poorly so that the person is resultantly made 
vulnerable in virtue of performing the labor.21 If the ability to meet your needs and 
wants is dependent on money, then remunerating care well could relieve the 
disadvantage of it. However, I do not advocate increasing the domain and scope of 

                                                 
21 A host of additional questions arise at this stage—how do we weigh caregiving 
against other social contributions? How do we quantify burdens that result in 
compounded vulnerability? Answering these questions will be difficult, but 
answering them is just as necessary as any other comparative assessment of 
individual and social disadvantage for a theory of distributive justice. However, we 
may legitimately postpone answers to these questions until the present project is 
complete. 

13

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



market norms, for to do so treats all goods as fungible (Anderson 1993). If we live in 
a fully marketized economy, then caregiving should be marketized and the arrow of 
care would identify a type of social productivity that would be included in the gross 
domestic product. But I do not advocate that kind of society, for increasing the 
scope of markets will inevitably result in the creeping of market norms. Care is 
among the most fundamental domains of social cooperation, and it follows from its 
fundamentality that we will need to meet our care needs at least at a minimal level 
irrespective of whether we have a market economy or not, and those caregiving 
arrangements need to be fair. 

 
Personal services 

Another iteration of the arrow of care map must include personal services. I 
adopt Kari Waerness’s term and definition of “personal services” for the category of 
activity that a person does for someone else when that person is capable of meeting 
the basic need him or herself, or the need is not pressing and/or legitimate (1984).22 
Personal services include making tea for others, attending to the emotions of others, 
and serving as an audience to the ego of another. 

Why should theorists of justice bother mapping personal services if the 
needs they meet are not essential to human life? We should map this labor because 
it is work that people—often women—perform. There may be strong social pressure 
for a person to perform the labor even if the care needs to which he or she responds 
are not legitimate care needs. Thus, anti-oppression theory needs to include a 
personal-services map because the recipients of that labor reap benefits from it and 
the people performing it remain disadvantaged. A personal-services map is needed 
to track distributions of effort and time on the part of the person performing that 
labor, and it will include emotional work, ego-tending, and some forms of sex 
work.23 As in the case of dependency care, the first iteration of the personal-services 
map is unpaid personal services. The second iteration, which is no less important, is 
a map of poorly remunerated personal services. 

For distributive purposes, the burdens of caregiving—as both dependency 
care and personal services—are defined by the effort it requires in terms of time 

                                                 
22 Engster includes care for self in the activity of caring, and therefore it need not 
always be other-directed (2007, 32). I do not include care for self because the nature 
of caring for someone else and caring for oneself differ significantly. I seek to 
capture the category of other-directed care that promotes the recipient’s well-being 
and goals. See Bhandary (2016a) for further analysis of the differences between self-
directed and other-directed care. 
23 See MacKinnon (2011). 

14

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



and the demands on the person.24 A full analysis of the burdens of personal services 
must include psychological costs experienced by the caregiver. Personal services 
that radically harm the self of the caregiver are not justified and so a just social form 
should eliminate them. For instance, some forms of sexual labor that destruct the 
self of the sex worker should be eliminated from just social forms. In the meantime, 
the labor should be descriptively tracked. 
 
Unmet care needs  

A supplemental iteration of the dependency-care map requires an index of 
the dependency-care needs of vulnerable persons that no one meets in the current 
arrangement. One can expect that disabled children of parents living in poverty will 
have fewer of their needs met. There will likely be significant correlations between 
relative wealth and the level at which one’s care needs are met, but it should not be 
assumed that inequalities in wealth and income will subsume inequalities in care.  

 
3. Function Two: Fair Caregiving Arrangements  

The second function for the arrow of care map is to serve as an object for 
deliberations about what a fair caregiving arrangement would be. In this use, it plays 
an essential role in a form of revised contractarianism—one that pays attention to 
the need for sensitivity to the facts that are included and to those that are excluded 
in the use of abstractions, where abstraction is defined here as the bracketing of 
various predicates. The normative question for the most abstract iteration is: What 
would a just caregiving arrangement be? Is it one with arrows of equal widths and 
lengths? Must the arrows be exactly reciprocal, modeling something like exchange 
reciprocity? For instance, we can model the idea that children have duties to their 
parents to care for them with a set of equal arrows going to and from parents and 
children over the course of a lifetime. We must also ask: Is a just arrangement one 
where caregiving arrangements are clustered by gender or race? The outcome of 
deliberations behind the veil of ignorance will rule out arrangements that exploit 
any group of people. No group of people will want to be designated as the set who 
must be other-directed caregivers, even if they may choose to do so in their own 
lives. A form of hypothetical contractarianism with an “I could be anyone” device 
that is moderately fact-sensitive, insofar as it includes a set of important facts that 
are needed for it to satisfy its function, will exclude arrangements that designate 
ascriptive identity groups to be other-directed. The important facts that must be 
included are the race, gender, class, and geographic origin of the person, but these 

                                                 
24 The metric requires collaboration among philosophers, anthropologists, 
economists, and sociologists (see, e.g., Folbre and Bittman 2004 for an excellent 
case of time-use studies). 

15

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



facts are mutable, and different categories can become the most significant ones in 
different societies.  

Moving caregiving arrangements to the forefront of the philosophical agenda 
changes contractarianism because of the way that these arrangements have been 
assumed to be matters of the family. Thus, the exploitation that occurs when a 
person does activity C, but C is not considered socially useful or necessary, where C 
is care, will be eliminated. Insofar as performing care that is invisible in the tracking 
of societal benefits and burdens is part of a form of group-based oppression, the 
arrow of care map will eliminate that source of group-based oppression.25 

We can expect that the arrow of care map will show how racial hierarchies 
and disadvantage overlap with social norms about who should receive care, and 
who should provide it. A normatively acceptable caregiving arrangement cannot 
distribute care in ways that differentiate by race and class—nor should any 
inequalities systematically result from these group memberships. But these 
inequalities must be tracked in the descriptive form, and doing so requires 
bracketing details of people’s particular relationships and situations in order to 
document the extent to which inequalities in care given and received overlap with 
other categories of ascriptive identity groups. The arrow of care map is a framework 
within which it is intelligible to ask undertheorized questions, such as: Do African 
American men receive more or less care than white men? And from whom? It also 
offers a way of systematizing inquiry into more established areas of inquiry, such as: 
Do women of color receive less care than white women? On whom does the care 
deficit fall with the greatest effects? Additional data can be added to incorporate 
how immigrant and legal-resident status correspond to care arrangements.  

Different societies may arrive at different conclusions about the outcome of 
hypothetical acceptability because our actual experiences and values influence our 
intuitions about what we would choose in an ideal choice situation.26 Therefore, I 
leave open the principle that meets the criterion of hypothetical acceptability. In 
this way, the account I offer is deliberately less determinate than Rawls’s use of the 
original position. It leaves the principles that would result open to the person 
contemplating the thought experiment. In this usage, the map enables imagining a 
variety of caregiving arrangements and evaluating the acceptability of these 
arrangements—when one does not know where they will be in the arrangement 

                                                 
25 I think eliminating invisible forms of labor will eliminate the corresponding 
varieties of oppression via theoretical invisibility. However, other forms of 
oppression may emerge.  
26 I will not endeavor to defend this claim here. My view, in brief, is that the actual 
cultural context within which we deliberate as philosophers influences our 
estimations of what ideally rational people would decide, given certain inputs. 

16

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



(Stark 2000). This type of theorizing is a valuable mechanism to clarify our thinking 
about care in human societies, for it leads us to think about what an ideally just 
society would look like when it takes caregiving into account alongside all other 
goods. Because care is one good among several fundamental social primary goods, 
inequality in the domain of care may be part of an overall just arrangement. 

 
3.1. Adaptive Preferences and Legitimacy 

A principle of fair distribution of care requires an account of legitimate needs 
for care. Any metric of needs-differences must also take into account the ways in 
which the needs of privileged members of a society are perceived by all members of 
the society as more pressing. Due to the ways that people’s expectations are shaped 
by what is available to them, and by that to which they have become accustomed, 
we cannot assume that we will get a clear view of what legitimate expectations for 
care needs will be. A woman who is expected to be self-abnegating may well 
demand less care from others.27 More broadly, the set of phenomena theorized as 
adaptive preferences suffice to give reasons to not merely rely on existing 
preferences and states of affairs.28  

There are some clear cases of legitimate care needs. For instance, a child 
with a severe cognitive or physical disability will legitimately require more care—of 
greater intensity and duration—than a nondisabled adult man. However, there are 
likely racial and economic differences in the extent to which these needs for care are 
diagnosed and met. The arrow of care will need to track the care people are actually 
receiving, and add facts about race, gender, and disability, so that it can identify 
disparities in the care received by disabled black and white children, for instance. 
Race must also be included to be able to theorize about the significance of white 
disabled children receiving more care than disabled children of color.  

For instance, a principle of equality of capability would recommend that 
everyone receive the amount of care that they need to function (Engster 2007), and 
this requires accounting for differences in needs. Therefore, race-based aggregate 
data about care received and given are needed in addition to data about the 
amounts of care people receive above or below a threshold for basic functioning. If 
the gap in average care received by racial groups is very great, it gives a prima facie 
reason to scrutinize our intuitions about which needs are legitimate, and for whom. 
An asymmetry in care at a group-based level is an injustice even if it does not result 

                                                 
27 On “deformed desires” as a phenomenon of the self-abnegating woman, see 
Superson (2005). A related set of phenomena have been theorized as adaptive 
preferences. See, for example, Elster (1982) and Khader (2011).  
28 See Farrelly (2007) on the connection between maximal fact-sensitivity and the 
perpetuation of adaptive preferences. 

17

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



in disadvantages of income and wealth. A group-based caregiving arrangement 
raises a prima facie problem for justice because it will create problems for everyone 
who shares in the ascriptive identity group of the caregivers. Finally, historically 
informed understandings of oppression are necessary but not sufficient because the 
arrow of care map is a metric that can track unmapped disadvantage and relational 
asymmetries. 

 
4. Comparing the Arrow of Care Map to Care Ethics and Feminist Liberalism 

Applied within the Family 
 
4.1. The Challenge from Care Ethics 

One of the commitments that differentiate care ethics from the liberal 
contractarian tradition is the care-ethical view that emotions are a source of moral 
knowledge, which is also linked to the claim that our emotions and particular ties to 
real people should not be overcome in order to achieve impartiality. Care ethicists 
argue that moral theory must pay attention to particularity, and that theorizing 
about abstract subjects who have no ties to others misses what is important for a 
moral life.29 

The arrow of care map enables thinking about society-wide caregiving 
arrangements in a way that abstracts away from particular intimate relationships 
and from the emotions and meaning of care in our lives so that we may gain an 
understanding of the distributive justice or injustice of that arrangement. Unlike 
care ethics, it does not prescribe how we should respond to the people in our lives. 
Care ethicists’ accounts of caregiving relationships and caring persons offer 
sophisticated analyses of caring practices and the aims of good relationships, but 
they do not include assessments of the underlying distribution of benefits and 
burdens.30 As I have argued, a theory of care will be incomplete and inadequate if it 
does not have an abstract way to depict and evaluate caregiving arrangements, for 
abstraction is necessary to depart from existing conventions and assumptions about 
who should care for whom.  

Care ethics advances a related criticism of thinking in the terms of justice. 
Alison Jaggar characterizes this care ethical criticism of “justice thinking” in the 
following passage:  

 

                                                 
29 Held (2006), S. Collins (2015), and Noddings (1982) advocate a contextual 
approach to moral reasoning. 
30 Engster (2007) theorizes about a just society that provides care. Kittay (1999) 
theorizes about the disadvantage of dependency workers within the terms of her 
dependency theory.  

18

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



Justice thinking is impersonal and general because it regards both 
moral subjects and the objects of their moral concern in terms of 
their moral status as representatives of humanity . . . rather than in 
terms of their concrete specificity; care thinking is personal and 
particularized in that both carers and those cared for regard each 
other as unique, irreplaceable individuals. (1995, 190–191)31  
 
My view is diametrically opposed to care ethics on this matter because some 

level of abstraction is valuable in order to step back from the facts of one’s life. It 
enables the theorist to stand back from his or her family’s assumptions about who 
within the family ought to provide actual caregiving if they are to express love and 
concern. Abstraction in this form is distinct from the kind of abstraction against 
which Bernard Williams (1981) counsels—the application of impartial thinking to our 
loved ones. And I embrace care ethicists’ claims about the moral requirements of 
our actual relationships. The arrow of care does not recommend standing back from 
love for one’s family to evaluate whether the terms of the relationship are fair. As 
actual people, we are embedded within imperfect and unjust social forms, and it is 
likely that this will always be the case. The project of theorizing about the nature of 
the good life within any particular social form is a task for moral theory, philosophy 
of action, and practical reason. A moral theory that rejects all real relationships 
when they are unfair would leave real persons relation-less in any foreseeable 
future. Whereas an account of morality must discern our particularity and identify 
how it is that we can live good lives, a political theory should evaluate the basic 
structure as the matrix within which we live our lives. Consequently, political theory 
must engage in abstractions. My argument for abstraction counsels a departure 
from particularity at the level of social forms—not at the level of a person’s own life.  

Claims for abstraction at the level of justification are sometimes taken to 
imply that we should implement these claims abstractly in some way. Stark (2010) is 
right to argue that we can be sensitive to context when we apply ideals we have 
arrived at through the use of abstraction in justification. Therefore, care ethicists 
need not reject abstract care, for it invites attention to morally salient details at the 
level of an individual’s life. My view is that we must respond to our lives as they are 
in their particularity—to the facts about who our parents are, or to particulars about 
our brothers, sisters, children, neighbors, and so forth. The particularized details 
about who is in our intimate sphere are not defined voluntarily. It is not up to us to 
choose whether or not we have a neighbor who falls down after an illness and is 
alone in the world; or to determine our parents’ abilities, and particular types of 

                                                 
31 Jaggar goes on to criticize care ethics. See also Walker (2007) for a criticism of 
theoretical-juridical approaches to moral theory. 

19

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



physical or mental decline towards the end of their lives; or to decide if we have a 
sibling, and to decide on that sibling’s need for support or help—for example, if he 
or she has a severe disability. In all of these cases, and in our real lives, we will have 
responsibilities that arise from the details of the real world, from contingencies that 
have nothing to do with our choices as well as from actions that we have chosen, 
but the consequences of which we had no way of predicting.  

What I wish to underscore, though, is that these responsibilities and choices 
occur against the backdrop of a social form that is regulated by some account of 
distributive justice. Our sense of our responsibilities is influenced by background 
conditions and culture.32 For instance, is a son responsible for his mother’s needs as 
she ages, or is his wife responsible for them? Is the younger son responsible, or is it 
the older one? In some pockets of Indian cultures, it is the eldest son who is held 
responsible for his parents, but hands-on caregiving responsibilities are assigned to 
his wife. Is the society-at-large responsible for an aging population, or is it the 
individuals themselves who should plan for their future with financial savings that 
will allow them to pay a professional caregiver? Or is it children overall who are 
responsible for their parents’ care? If a community overall is responsible for meeting 
care needs, how does this actually break down by race and ethnicity? If white 
women regularly expect a higher level of needs-satisfaction while women of color 
have adapted to meeting fewer of their needs, then will a distributive assessment 
that relies on subjective assessments of well-being be adequate?33 To answer these 
questions, we will need to evaluate a number of different permutations of 
caregiving arrangements. 

Faulty abstractions, not abstractions per se, are the cause of 
contractarianism’s perpetuation of oppression. Consequently, a criterion of 
adequacy for contemporary forms of liberalism is that they must do the theoretical 
work needed to identify the abstractions the theory should employ. The arrow of 
care map meets these criteria as a conceptual device that identifies caregiving 
arrangements and then holds them up as an object of hypothetical consent. When 
the labor of caregiving is made visible, a new layer of privilege and disadvantage 
becomes evident that can address intersectionality by showing how the expectation 
to receive care and the social expectation to give care can map onto race, ethnicity, 
gender, caste, and social class. The concept of the arrow of care thereby disrupts 
status quo social arrangements in societies where caregiving arrangements are 

                                                 
32 See Walker’s (2007) analysis of moral understandings, which has influenced this 
account. 
33 See Khader (2011) for the argument that “inappropriate adaptive preferences” 
must be identified in virtue of their conflict with an objective account of well-being. 

20

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



enforced through informal mechanisms like widely accepted expectations and 
narratives.34  

 
4.2. No Action-Guidance 

The arrow of care map differs from two influential feminist liberals’ approach 
to inequality and caregiving responsibility.35 Susan Okin argued for the application of 
principles of justice within family—and in doing so she rejected the Rawlsian view 
that representatives be heads of households.36 Although I share Okin’s project of 
feminist Rawlsian contractarianism, the arrow of care map departs from Okin in that 
it maintains a higher level of abstraction to accommodate cross-cultural differences 
in caregiving and familial structures while Okin applies Rawlsian reasoning within 
the family. Moreover, I do not assume the family as a starting point, and in this way, 
operate at a greater level of abstraction. Instead, the starting point for the arrow of 
care is a claim that we all need to receive care and therefore caregiving is one of the 
core functions that a scheme of social cooperation must satisfy. 

Another foil for the arrow of care map is Jean Hampton’s (1993) feminist 
contractarian test for personal relationships that tells participants to assess whether 
they would accept the terms of the relationship as the product of an unforced 
agreement. In contrast with Hampton’s test, the arrow of care is not a method of 
abstraction for individuals to map their relationships in service of demanding 
equality in the care given and received within each one of their intimate 
relationships. 37 Instead, as a framework for mapping society-wide distributions 
which then evaluates their fairness, the arrow of care map does not seek to proffer 
action-guiding recommendations for individual lives. Our lived engagement with 
families and friends should be guided by values that extend far beyond justice. 
 

                                                 
34 I do not recommend that individuals map their relationships within the arrow of 
care. It is not a comprehensive assessment of fairness, nor does it provide an 
account of what it means to be concerned about someone or to have a legitimate 
responsibility towards a particular other.  
35 Ann Cudd’s (2006) account of society-wide oppression offers a systematic analysis 
of the context of choice, and so the arrow of care map is structurally closer to her 
theory than it is to Hampton’s or Okin’s theories. 
36 See Baehr (1996) for a discussion of the way Okin’s view changed over time. See 
also Abbey (2016). 
37 Similarly, I do not offer a solution to directly address cooperative conflicts for 
people here and now (Sen 2011); nor does the arrow of care embed within it ways 
to increase women’s agency in diverse cultural contexts (Meyers 2002; Khader 
2011). 

21

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



Conclusion 
This paper has advanced a concept—the arrow of care map—that gives us a 

way to think about care that is not embedded in a set of assumptions that are 
culturally conservative insofar as they are specific to the social form in which we 
live. It is a concept in the sense that it sets forth a problem. The problem it sets forth 
is that every society needs to advance an arrangement to secure its care needs. 
There will then be multiple conceptions to specify what a good caregiving 
arrangement is—these conceptions are solutions to the problem set forth by the 
concept, and they will take the form of various arrow of care maps.38 The concept of 
the arrow of care map takes a bird’s-eye view of the labor of care—it does not seek 
to identify the nuances of relations of love and affection, and it abstracts away from 
assumptions about family units or socialized care to offer a way to tune our 
intuitions about caregiving in human life.  

The paper has also advanced a methodology for ideal theory that seeks to be 
non-ideological. The insights and intuitions on which it relies are informed by 
existing states of affairs, and it is explicit about its value commitments. By advancing 
this abstract concept, and limiting my theoretical project to this conceptual work, 
my account may appear to conflict with the commitments of a dominant strain of 
feminism and anti-racist philosophy that advocates nonideal theory. Whereas 
nonideal theories locate the core philosophical project in improving the world, and 
seek to do so quite directly, this paper has shown that theory that is “ideal”—in the 
sense that it seeks to clarify our values—must also begin with insights from the 
pressing problems of society, and then advance concepts that can better identify 
these problems. These theories must interrogate and evaluate the normative claims 
that are embedded in these concepts. Finally, by formulating the concept of the 
arrow of care map, I hope to enable further theorizing about the facts of societal 
caregiving arrangements in order to imagine new visions of just societies. 
 
 
Works Cited 
Abbey, Ruth. 2016. “Susan Okin's Justice, Gender, and the Family: Twenty-Five Years 

Later.” Hypatia 31 (3): 636–637.  
Anderson, Elizabeth. 1993. Value in Ethics and Economics. Cambridge, MA: Harvard 

University Press. 
Baehr, Amy R. 1996. “Toward a New Feminist Liberalism: Okin, Rawls, and 

Habermas.” Hypatia 11 (1): 49–66. 

                                                 
38 I thank Julie Walsh for identifying the similarity between Rawls’s concept-
conception distinction and my own use of the arrow of care map. 

22

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



———. 2004. “Feminist Politics and Feminist Pluralism: Can We Do Feminist Political 
Theory Without Theories of Gender?” Journal of Political Philosophy 12 (4): 
411–436. 

———. 2013. “Liberal Feminism: Comprehensive and Political.” In Feminist 
Interpretations of Rawls, edited by Ruth Abbey, 150–166. University Park: 
Pennsylvania State University Press. 

———. 2017. "A Capacious Account of Liberal Feminism." Feminist Philosophy 
Quarterly 3 (1): article 4. doi:10.5206/fpq/2016.3.4.  

Bhandary, Asha. 2010. “Dependency in Justice.” Hypatia 25 (1): 140–156. 
———. 2016a. “A Millian Concept of Care: What Mill’s Defense of the Common 

Arrangement Can Teach Us About Care.” Social Theory and Practice 41 (1): 
155–182. 

———. 2016b. “Liberal Dependency Care.” The Journal of Philosophical Research 
41:43–68. 

Collins, Patricia Hill. 2000. Black Feminist Thought. New York: Routledge. 
Collins, Stephanie. 2015. The Core of Care Ethics. London: Palgrave Macmillan. 
Crenshaw, Kimberlé. 1991. “Mapping the Margins: Intersectionality, Identity Politics, 

and Violence against Women of Color.” Stanford Law Review 43 (6): 1241–
1299. 

———. 2015. "Say Her Name: Politicizing Black Women's Vulnerability to State 
Sanctioned Violence." Keynote lecture at the Association for Feminist Ethics 
and Social Theory (FEAST) Conference, Clearwater, FL, October 2. 

Cudd, Ann. 2006. Analyzing Oppression. Oxford: Oxford University Press. 
Dimock, Susan. 2008. “Why All Feminists Should Be Contractarians.” Dialogue 47 (2): 

273–290. 
Elster, Jon. 1982. “Sour Grapes – Utilitarianism and the Genesis of Wants.” In 

Utilitarianism and Beyond, edited by Amartya Sen and Bernard Williams, 
219–238. Cambridge: Cambridge University Press. 

Engster, Daniel. 2007. The Heart of Justice. New York: Oxford University Press. 
Farrelly, Colin. 2007. “Justice in Ideal Theory: A Refutation.” Political Studies 55 (4): 

844–864. 
Folbre, Nancy, and Michael Bittman, eds. 2004. Family Time. New York: Routledge. 
Gould, Carol. 2008. “Recognition in Redistribution: Care and Diversity in Global 

Justice.” Southern Journal of Philosophy 46 (S1): 91–103. 
Greyser, Naomi. 2017. On Sympathetic Grounds. New York: Oxford University Press. 
Hampton, Jean. 1993. “Feminist Contractarianism.” In A Mind of One’s Own, edited 

by Louise Antony and Charlotte Witt, 227–256. Boulder, CO: Westview Press.  
Hartley, Christie, and Lori Watson. 2010. “Is a Feminist Political Liberalism Possible?” 

Journal of Ethics and Social Philosophy 5 (1): 1–21.  

23

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



———. 2012. “Political Liberalism, Marriage and the Family.” Law and Philosophy 31 
(2): 185–212.  

Hay, Carol. 2013. Kantianism, Liberalism, and Oppression. New York: Palgrave 
Macmillan. 

Held, Virginia. 1987. “Non-Contractual Society: A Feminist View.” In Science, 
Morality, and Feminist Theory, edited by M. P. Hanen and Kai Nielsen, 11–
137. Calgary, AB: University of Calgary Press. 

———. 2006. The Ethics of Care. Oxford: Oxford University Press. 
Hochschild, A. R. 2000. “Global Care Chains and Emotional Surplus Value.” In On the 

Edge: Living with Global Capitalism, edited by W. Hutton and A. Giddens, 
130–146. London: Jonathan Cape. 

Kang, Hye Ryoung. 2016. “Can Rawls’s Nonideal Theory Save His Ideal Theory?” 
Social Theory and Practice 42 (1): 32–56. 

Khader, Serene. 2011. Adaptive Preferences and Women’s Empowerment. New York: 
Oxford University Press. 

Kittay, Eva. 1999. Love’s Labor: Essays on Women, Equality, and Dependency. New 
York: Routledge. 

———. 2008. “The Global Heart Transplant and Caring Across National Boundaries.” 
The Southern Journal of Philosophy 46 (S1): 138–165. 

Jaggar, Alison. 1995. “Care as a Feminist Practice of Moral Reasoning.” In Justice and 
Care: Essential Readings in Feminist Ethics, edited by Virginia Held, 825–833. 
Boulder, CO: Westview Press. 

Lloyd, S. A. 1998. “Toward a Liberal Theory of Sexual Equality.” Journal of 
Contemporary Legal Issues 9:203–224. 

MacKinnon, Catharine. 2011. “Trafficking, Prostitution and Inequality.” Harvard Civil 
Rights–Civil Liberties Law Review 46 (1): 271–309. 

Meyers, Diana Tietjens. 2002. Gender in the Mirror. New York: Oxford University 
Press. 

Mills, Charles. 1997. The Racial Contract. New York: Cornell University Press. 
———. 2004. “‘Ideal Theory’ as Ideology.” In Moral Psychology, edited by Peggy 

DesAutels and Margaret Urban Walker, 163–181. Lanham, MD: Rowman and 
Littlefield. 

———. 2012. “Liberalizing Illiberal Liberalism.” Closing keynote address for the 
Oxford Graduate Conference in Political Theory, Oct. 4. University of Oxford 
Podcasts. MS on file with author. 

Mullin, Amy. 2014. “Children, Vulnerability and Emotional Harm.” In Vulnerability: 
New Essays in Ethics and Feminist Philosophy, edited by Catriona Mackenzie, 
Wendy Rogers, and Susan Dodds, 266–287. New York: Oxford University 
Press. 

Noddings, Nel. 1982. Caring. Berkeley: University of California Press. 

24

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



Okin, Susan. 1989. Justice, Gender and the Family. New York: Basic Books. 
O’Neill, Onora. 1996. Towards Justice and Virtue: A Constructive Account of Practical 

Reasoning. Cambridge: Cambridge University Press. 
Pateman, Carol. 1988. The Sexual Contract. Palo Alto, CA: Stanford University Press. 
Rawls, John. 2001. Justice as Fairness: A Restatement. Edited by Erin Kelly. 

Cambridge, MA: Harvard University Press. 
———. 2002. The Law of Peoples. Boston: Harvard University Press. 
Reiheld, Allison. 2015. “Just Caring for Caregivers: What Society and the State Owe 

to Those Who Render Care.” Feminist Philosophy Quarterly 1 (2): article 1. 
doi:10.5206/fpq/2015.2.1. 

Ruddick, Sara. 1989. Maternal Thinking. Boston: Beacon Press. 
Schouten, Gina. 2015. “Citizenship, Reciprocity, and the Gendered Division of 

Labor.” Politics, Philosophy and Economics. 
doi: 10.1177/1470594X15600830. 

Schwartzman, Lisa. 2006a. Challenging Liberalism: Feminism as Political Critique. 
University Park: Pennsylvania State University Press. 

———. 2006b. “Abstraction, Idealization, and Oppression.” Metaphilosophy 37 (5): 
565–588. 

———. 2016. “Action Guidance, Oppression, and Nonideal Theory.” Feminist 
Philosophy Quarterly 2 (1): article 4. doi:10.5206/fpq/2016.1.4. 

Sen, Amartya. 2011. The Idea of Justice. Cambridge, MA: Harvard University Press. 
Spelman, Elizabeth. 2002. Repair: The Impulse to Restore in a Fragile World. Boston: 

Beacon Press. 
Stark, Cynthia A. 2000. “Hypothetical Consent and Justification.” Journal of 

Philosophy 97 (6): 313–334. 
———. 2007. “How to Include the Severely Disabled in a Contractarian Theory of 

Justice.” Journal of Political Philosophy 15 (2): 127–145. 
———. 2010. “Abstraction and Justification in Moral Theory.” Hypatia 25 (4): 825–

833. 
Stemplowska, Zofia. 2008. “What’s Ideal about Ideal Theory?” in “Social Justice: 

Ideal Theory, Nonideal Circumstances,” edited by Ingrid Robeyns and Adam 
Swift, special issue, Social Theory and Practice 34 (3): 319–340. 

Stemplowska, Zofia, and Adam Swift. 2015. “Ideal and Nonideal Theory.” In The 
Oxford Handbook of Political Philosophy, edited by David Estlund, 373–389. 
New York: Oxford University Press. 

Superson, Anita. 2005. “Deformed Desires and Informed Desire Tests.” Hypatia 20 
(4): 109–126. 

Tessman, Lisa, ed. 2009. “Introduction.” Feminist Ethics and Social and Political 
Philosophy: Theorizing the Non-ideal. New York: Springer Science and 
Business Media. 

25

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



———. 2015. Moral Failure. Oxford: Oxford University Press. 
Tronto, Joan. 1993. Moral Boundaries. New York: Routledge. 
Valentini, Laura. 2012. “Ideal v. Non-ideal Theory: A Conceptual Map.” Philosophy 

Compass 7 (9): 654–664. 
Varden, Helga. 2006. "Kant and Dependency Relations: Kant on the State’s Right to 

Redistribute Resources to Protect the Rights of Dependents." Dialogue: 
Canadian Philosophical Review 45:257–284. 

Waerness, Kari. 1984. “Caring as Women’s Work in the Welfare State.” In Patriarchy 
in a Welfare Society, edited by H. Holter, 67–87. Oslo: Universitetforlaget. 

Walker, Margaret Urban. 2007. Moral Understandings. Second edition. Oxford: 
Oxford University Press.  

Williams, Bernard. 1981. “Persons, Character and Morality,” In Moral Luck, edited by 
James Rachels, 1–9. Cambridge: Cambridge University Press. 

 
 
ASHA BHANDARY is Assistant Professor of Philosophy at the University of Iowa. Her 
areas of specialization include feminist ethics, social and political philosophy, and 
autonomy theory. Her published articles on liberalism, care, autonomy, and 
distributive justice have appeared in Hypatia, Journal of Philosophical Research, 
Social Theory and Practice, and Journal of Political Philosophy. 

26

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



The Arrow of Care Map: Abstract Care in Ideal Theory1 
Asha Bhandary 

 
 
 

Abstract 
This paper advances a framework to conceptualize societal caregiving 

arrangements abstractly. It is abstract in that it brackets the meaning of our 
particular relationships. This framework, which I call the arrow of care map, is a 
descriptive tracking model that is a necessary component of a theory of justice, but 
it is not a normative prescription in itself. The basic idea of the map is then multiply 
specifiable to track various ascriptive identity categories as well as different 
categories of care labor. In this way, the idea of the arrow of care map serves as a 
conceptual frame with which to identify societies’ caregiving arrangements. I 
characterize it as a component of “ideal theory” insofar as it seeks to clarify our 
values without the immediate aim of formulating principles to govern a just society. 
The resultant partial theory of justice is one that responds to nonideal theory’s 
critiques of the ideological nature of liberal contractarianism’s idealizations and 
abstractions while moving towards new visions of a just society. 
 
 
Keywords: ideal theory, nonideal theory, contractarianism, care, dependency care, 
liberal feminism, feminist liberalism, justice, distributive justice 
 
 
 

Care, compassion, and concern are crucial elements of meaningful intimate 
relationships (Held 2006; Tronto 1993; Ruddick 1989). In addition, the labor of 
hands-on care to meet a person’s basic needs often occurs within our closest 
relationships. However, caregiving also occurs in paid and professional contexts, 
which sometimes cross national boundaries through migrations of nannies and 
                                                 
1 I presented this paper to the Society for Analytical Feminism, Wellesley University, 
and as the 2017 Inaugural Robinson Grover Lecture at the University of Connecticut, 
where Elise Springer and Rik Hine offered interesting analyses of the view that I plan 
to pursue in future work. I thank the participants of each of these events. I owe a 
particular debt of gratitude to Cynthia Stark, Lisa Tessman, and Diana Tietjens 
Meyers for philosophically open-minded discussions about the nature of my project. 
The written version of the paper was improved by comments by Richard Fumerton, 
two anonymous reviewers, and the editors of this journal. 

1

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



other care workers. Moreover, even in the contexts of families and friends, care that 
is given and received can be structured by internal hierarchies and inequities.  

Various cultures and societies have different ways of organizing life around 
care, such as gendered arrangements, extended families, and socialized care. These 
arrangements have particular histories. The history of slavery in the US means that 
black women who tend to white women do so in a context that is racially and 
historically charged.2 In societies with extended family systems, systems of care 
provision are structured by gender oppression and internally structured hierarchies 
and norms. For instance, a daughter-in-law may provide elder care to her parents-in-
law, where that practice is linked to preferences for males in those cultures, or a 
youngest daughter may be designated as the caregiver of her own parents, a 
designation which is paired with a cultural custom that she not marry and have 
children of her own. 

Although there is an extensive feminist literature about the highly gendered 
nature of inequalities in responsibilities for providing care,3 we lack a unitary model 
with which to identify arrangements for care in diverse cultures and social forms—
one that goes beyond gender. This paper fills this gap by advancing a new abstract 
concept for tracking the distribution of caregiving labor in a society that I call the 
arrow of care map. It is an abstract representation of caregiving that brackets the 
meaning of the relationships within which it occurs and thereby enables critical 
distance from particular cultural biases about who ought to care for whom. The 
function of this map is to gain clarity about how caregiving actually occurs in 
societies, because transparency is a necessary step prior to theorizing about what a 
just society would be. Due to the pervasiveness of obfuscating narratives about who 
gives care and who receives care, it will be a substantial achievement to arrive at an 
accurate description of the ways that care is provided and received. 

In Section 1, I defend the need for this new concept. There, I also show how 
the arrow of care map is responsive to the nonideal theoretic criticism that ideal 
theory is ideological in virtue of the content of its idealizations and abstractions. 
Section 2 sets forth the core components of the arrow of care map and identifies the 
normative commitments embedded in the model. It is a relational analysis of a 
society’s caregiving arrangements founded on individual units of human beings. 
There, I also identify the map’s variations as specified by categories of labor and 
identity group, while emphasizing the need for the higher-order, abstract map as a 

                                                 
2 I thank Lindsey Stewart for discussion on this topic. 
3 For feminist liberal work that discusses care, see Abbey (2016), Baehr (2004, 2013, 
2017), Bhandary (2010, 2016a, 2016b), Cudd (2006), Dimock (2008), Hartley and 
Watson (2010, 2012), Lloyd (1998), Okin (1989), Schouten (2015), Stark (2007), 
Varden (2006). 

2

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



way to maintain cross-cultural flexibility to track various identity groups. Section 3 
identifies the limited conclusions that can be drawn about what a fair caregiving 
society will be prior to the descriptive tracking endeavor. Finally, in Section 4, I 
contrast the scope of the arrow of care map with both the ethics of care and forms 
of feminist contractarianism that apply a metric of justice directly to the family and 
personal relationships.  

 
1. The Need for a New Concept 
 
1.1. Oppression via Theoretical Invisibility 

Oppression via theoretical invisibility occurs when a person has a lived 
experience of disadvantage or exploitation that is unintelligible by the terms of the 
theory. In the lexicon of that theory, the person is either not disadvantaged, or the 
source of their disadvantage remains out of view even if they are disadvantaged 
according to the theory’s metric. For instance, an African American woman in the US 
might be economically disadvantaged—where the disadvantage is identifiable with a 
metric of income inequality—but the complex structural reasons that create 
disadvantage will not be wholly explicable in terms of income. As a result, an agenda 
for social change that follows from a theory that only tracks income inequality will 
overlook these structures of disadvantage. Thus, one of the tasks of a theorist of 
justice is to articulate frameworks that will identify these overlooked experiences. 

Kimberlé Crenshaw’s concept of “intersectionality” is a prime example of a 
concept that makes visible a set of issues that perhaps could have been triangulated 
from preexisting concepts, but would remain more difficult to theorize without it.4 
The idea of intersectionality is a way to understand the social position of women of 
color that reveals how the categories of race and gender obscure theorizing about 
their lives. Intersectional analyses point out how assumptions of race and gender 
result in the loss of data capture that is particular to the lives of women of color.5 
The articulation of the concept makes subsequent theorizing possible to further 
explore its various dimensions and significance.6 Concepts like intersectionality are 
abstract, insofar as they highlight one set of concerns to the exclusion of others.7 
Thus, the concepts of gender, race, and class have made many aspects of social 

                                                 
4 Crenshaw’s concept of intersectionality is a way to understand the invisibility of 
black women by naming their location and thereby rejecting the idea of “identity as 
woman or person of color as an either/or proposition” (Crenshaw 1991, 1242). 
5 Crenshaw explains the problem of analyses of racism and sexism as one in which 
all black people are men and all women are white (Crenshaw 2015). 
6 See P. Collins (2000). 
7 Similarly, Mills (1997) argues for the need for the concept of white supremacy. 

3

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



oppression visible, and the idea of intersectionality has both augmented and 
complicated them.  

Every culture or society has some way of organizing life around care, such as 
gendered arrangements, extended families, and socialized care. Despite the 
universality of care, though, caregiving arrangements are among the most 
theoretically and practically significant sources of blindness in liberal 
contractarianism. Therefore, if liberal contractarianism is to be a viable doctrine, it 
needs a way to conceptualize how care is provided. I schematize this understanding 
of a society-wide caregiving arrangement with the arrow of care map. The arrow of 
care map theorizes the labor of care—not the nuances of relations of love and 
affection—and it abstracts away from assumptions about family units or socialized 
care to offer a way to tune our intuitions about caregiving in human life. The 
framework embeds within it a normative claim that society must meet its vital care 
needs.8 In doing so, it eliminates the following form of exploitation from traditional 
forms of liberalism. More precisely, adding the arrow of care to an account of the 
just distribution of the benefits and burdens of social cooperation removes from 
theory the following form of exploitation as it occurs for care labor.  

 
~Persons X are assigned to do work C when work C is not recognized as work. 
Consequently, persons X look like non-contributors. 
~ Persons Y do work A, which is highly valued. They are able to gain 
additional advantages by assiduously avoiding work C. Their avoidance of C is 
not identifiable as shirking because C is not identified as work. 
 
If a person’s contribution to social cooperation is not identified as a 

contribution in a theory of distributive justice, and the benefits that some people 
receive are not identified as benefits, then the resultant theory will have a loophole 
that leads to advantages for the people who avoid that work. This is the pattern that 
many societies have adopted for the social contributions needed for repair and 
maintenance.9 Eva Kittay has argued that liberalism, in particular, results in 
disadvantages to dependency workers due to their invisibility in its basic 
assumptions and concepts (1999). 

 
 

                                                 
8 It proceeds from the premise, unargued for here, that societies need to meet their 
care needs, and that doing so plays a crucial role in systems of social cooperation. 
For that argument, see Bhandary (2016b). See also Reiheld (2015) for an argument 
that society has a duty towards caregivers. 
9 On repair and maintenance, see Tronto (1993), Spelman (2002), Walker (2007). 

4

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



1.2. Non-ideological Ideal Theory 
More broadly, liberal contractarianism has been criticized for idealizations 

and concepts that make the sources of group-based oppression impossible to 
identify or, more strongly, require the subordination of a subset of society (Pateman 
1988; Jaggar 1995; Kittay 1999; Held 1987; Mills 2004; Schwartzman 2006a). I agree 
with these critics that an ideal theory that refuses to scrutinize whether its core 
concepts result in the perpetuation of illicit privilege is a non-starter as a viable 
theory of justice. Despite these problems with the history of the contract tradition, I 
share the growing consensus that one of the main targets of the critique of ideal 
theory—the core liberal contractarian idea of fairness—retains value as a basis for 
identifying a number of contemporary forms of social injustice.10 Fairness, defined in 
a way that requires transparency, can rule out arrangements that exploit or 
systematically disadvantage a subset of society. A necessary, but not sufficient, step 
to rule out these exploitative arrangements is to develop concepts and claims that 
make visible what Lisa Schwartzman calls “the sources and manifestations of group-
based oppression” (2006a). Thus, the project of advancing non-ideological liberalism 
requires new theoretical work to better specify the model of the person, the facts 
known to deliberators, and the domain of the basic structure of society.11 
Idealizations that perpetuate ideology must be criticized, and better concepts for 
mapping the realities of existing societies are needed. The arrow of care map is a 
contribution to the project of advancing non-ideological ideal theory. 

 
2. The Arrow of Care Map: Overview of Functions 

The framework I advance is one that can be used for all social forms. It will 
be further specified by the culturally specific details that structure social inequality 
in that social form—where the assignment of care may track onto a person’s caste 

                                                 
10 Okin (1989), Hampton (1993), Hay (2013), and Mills (2012). I follow a broadly 
Rawlsian form of contractarianism (2001). For the criticism that contractarianism 
does not necessarily rule out oppression, see Schwartzman’s critique of O’Neill 
(2006b, 572). Specifying the precise form of contractarianism will be necessary for a 
complete account of just caregiving, but that is not the task of this paper. 
11 My usage of ideal theory here is compatible with Stemplowska (2008) and 
Valentini (2012). It differs from Mills (2004), who says that ideal theory is necessarily 
ideological theory, so that if a person addresses oppression, they are doing nonideal 
theory. Thus, if one accepts Mills’s definitions of ideal and nonideal theory, the 
arrow of care map will count as nonideal theory because it makes a source of 
oppression visible. See also Tessman (2009). For alternative formulations of the 
distinction, see Rawls (2002), Valentini (2012), Stemplowska and Swift (2015), and 
Kang (2016). 

5

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



heritage rather than race or gender. Thus, the basic model of the arrow of care map 
is universal but multiply specifiable.  

The arrow of care map, as a systems-level depiction of societal caregiving 
arrangements, gives theories of justice something they have been lacking: a way to 
understand care given and received as a fundamental social good that is the 
outcome of social cooperation, and that is prior to—and not dependent on—
contingencies such as the particular economic structure of the society. In virtue of 
bracketing relationships of particularity and intimacy as well as the robust meaning 
of giving and receiving care in human life, it is decisively abstract and moderately 
fact-sensitive—incorporating the most fundamental aspects of human life, while 
leaving aside others.12 Because the need to receive care is fundamental to all human 
lives, it is a fact that must be included on any list of basic facts employed in a theory 
of justice.13 In its most abstract form, it has value for a broad range of social forms 
to track who tends to receive large amounts of care and who tends to provide it. At 
one level less abstracted, it must track historically oppressed identity groups. I do 
not reduce it to these specific maps, though; instead, I insist on the need to maintain 
an abstract version of the map so that societal variations in the exploitation of 
caregivers can be found even when they do not coincide with dominant narratives. 
It may be the case that class has a higher correlation with caregiving than gender, 
for instance. Theorists need critical distance from cultural preconceptions about 
who ought to receive care and who ought to provide it because—as scholars of 
affect study show—our concern for others is shaped by gender, race, and class 
(Greyser 2017).  

Thus, it is crucial that a map that tracks care given and care received remains 
conceptually separable from the range of ascriptive factors that will correspond to 
inequities in care. Maintaining a separate metric may reveal unmapped social 
locations, such as a caregiver who may be from an otherwise advantaged group but 
who becomes disadvantaged as a caregiver.14 Therefore, in addition to serving as a 
new descriptive mapping concept, the arrow of care map makes it possible to 

                                                 
12 I therefore aim to respond to the problem that abstractions have too often 
masked false claims—that humans are maximally rational (O’Neill 1996), or that 
they fail to be explicit about what they bracket, and why (Schwartzman 2006a). My 
account is moderately fact-sensitive in that it includes a set of basic facts but 
brackets others. See also Farrelly’s characterization of the moderate fact-sensitivity 
of Rawls’s theory, which “acknowledges some moderate feasibility constraints (e.g. 
pluralism) but also employs a number of idealizing assumptions (e.g. society is 
closed, full compliance, etc.) when deriving the principles of justice” (2007, 844). 
13 For a lengthier argument for this claim, see Bhandary (2016b). 
14 See also Kittay’s “dependency workers” (1999). 

6

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



identify inequalities that are not already captured by economic analyses or racial 
hierarchies.  

Arrow of care maps have two functions. The first function of the maps is to 
organize fact-finding about societies’ caregiving arrangements. The basic map 
template will then be modified into multiple maps with overlays tracking trends by 
race, gender, geographic origin, and class. Its main purpose is to gain an 
understanding about trends in levels of care received and provided as a relational 
good. 

The second function of the maps is to guide thinking about what a fair 
arrangement will be. It is my view that there are likely many possibly fair 
arrangements. On my preferred view, deliberations should be guided by a broadly 
Rawlsian methodology, which evaluates the arrangement based on the idea that 
you could be anyone in it, variously interpreted as the claim that you could occupy 
any social position in it. However, the arrow of care map idea is compatible with a 
broad range of interpretations of transparency and fairness tests. What it rules out 
are arrangements that are based on, and require, the widespread exploitation of 
groups of people (minorities, women) in virtue of their roles as caregivers.15 After 
greater transparency is achieved, further deliberation will be required to 
imaginatively conceptualize potential caregiving arrangements as candidates for a 
fair arrangement. For instance, just caregiving arrangements may include nuclear 
families, extended families, and non-family-based care.  

 
2.1. Function One: A System-Wide Descriptive Mapping Concept for Caregiving 

Arrangements  
The arrow of care map does not seek to capture the whole truth of the role 

of caring relationships in human life, nor does it seek to prescribe how people 
should think about care in their private lives. It is a systems-level analysis of a society 
in the sense that it holds up caregiving arrangements as objects of deliberation in a 
way that facilitates clearer thinking about the different possible arrangements that 
would be acceptable to people in a society, when the labor of care is made visible.16 

                                                 
15 See Schwartzman’s (2006b) argument that a theory without ideals will not have 
resources to rule out oppression. Because the paper’s aim is to advance the idea of 
the arrow of care map—and not to defend a specific conception of fairness—I do 
not specify an ideal or respond to Schwartzman’s claim about the necessity of ideals 
here.  
16 The project is aptly characterized by Stemplowska’s characterization of the point 
of ideal theory as clarifying our values (2008). See also Tessman (2015) on the need 
for feminist theory that apprehends our normative realities even when there are no 
morally viable paths of action. 

7

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



The arrow of care map is properly classified as a “descriptive mapping concept” 
(Mills 2004, 174), insofar as it creates a concept to use in order to identify what 
occurs in the real world.  

Caregiving norms are deeply culturally entrenched; they are embedded 
within gender norms about who should care for whom—by, for example, bringing 
them tea. They suffuse assumptions about who deserves sympathy for their 
suffering. Therefore, identifying what actually occurs in the domain of care is a 
significant task of its own. The fact-finding task will give us real data with which to 
inform our social understandings about who provides care, who receives it, and at 
what levels. In the descriptive use, it remains somewhat abstract because it brackets 
information about particular relationships in order to gain an understanding of 
caregiving that does not simply legitimate existing assumptions about who should 
care for whom.  

Although the arrow of care map will promote greater clarity about the 
nature of a society’s caregiving arrangements, it does not purport to avoid all 
cultural biases. For instance, leaving nonhuman animals out of the metric is a 
normatively laden starting point. The arrow of care map does not avoid the 
possibility of embedding and legitimating existing biases, but it seeks to be 
transparent about the claims and assumptions it makes, and the reasons for those 
assumptions. It is precisely because of the cultural entrenchment of caregiving 
norms that we need to think abstractly about care. The preponderance of 
philosophical effort directed towards thinking about care counterfactually has been 
directed towards new action-guiding policy proposals. These proposals often embed 
within them assumptions of the immutability of other features of the status quo. 
Theories of justice need a more abstract methodology to think about care, and that 
is what the arrow of care achieves.  

However, because care ethicists and other feminists have roundly criticized 
abstraction, it is incumbent on me to clarify and defend the role of abstraction, and 
also to show how my account does not fall into the pitfalls of ideal theory as it has 
been criticized by nonideal theorists. Schwartzman represents the nonideal criticism 
of ideal theory as fundamentally a problem of misrepresentation: “[Ideal theory] 
misrepresents both agents and structures in ways that deny and obscure the 
realities of oppression” (2016, 7). Thus, I seek to avoid misrepresentation by 
characterizing the project and scope of the arrow of care map. The arrow of care 
map is an abstraction that brackets numerous features of our lived experiences of 
caregiving. It does not seek to characterize the complexity of lived experiences 
within caring relationships. Although it does not accurately represent the complexity 
of human agents, it offers a structural analysis of caregiving that can illuminate 
forms of oppression. It does so in a way that seeks to attend to the ways that 

8

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



abstractions and idealizations can be ideological to mask and perpetuate 
entrenched inequalities. 

 
2.2. Normativity in Abstraction 

The arrow of care map includes normativity at two levels: first, with the 
values and conceptualizations that are incorporated into the arrow of care, and then 
to guide and assess various caregiving arrangements. It includes value claims by 
taking individual human beings as what is fundamental in order to then depict the 
amount of care each person provides, for instance, over a lifetime. As a framework 
for tracking actual distributions, the arrow of care map will unveil the care people 
give and receive in a way that is individualistic but that also recognizes the relational 
nature of caregiving. The individualism of the framework is built into the metric of 
care that is given and received for and by each person. The relational nature of care 
given and received occurs in a set of arrows that specify from whom and towards 
whom care goes. Sometimes the arrows will connect two members of a family, at 
other times they will not connect them. Quantifying care in this way will lay bare 
distributive inequalities that have been masked by cultural narratives. Theorists 
must (a) peel away the norms of femininity that mask the fact that caring is labor, 
(b) debunk ideas of gender essentialism that naturalize the labor of care, and (c) 
contest narrative and social constructs about racial and ethnic identity that 
normalize, for instance, Western presumptions that women of color should tend to 
the care, maintenance, and repair needs of whites.  

Furthermore, assumptions about oppressive gender norms can leave 
undetected exploitative labor distributions that occur within same-sex marriages, or 
the subordination of a man within a heterosexual couple. Consequently, the map 
will require extensive fact-finding by sociologists and anthropologists who attend to 
dominant norms, but are aware that these norms will not always describe what 
actually occurs. 17 To see just how difficult it may be to identify actual distributions 
of care when they conflict with social narratives, consider the following case: An 
affluent, white stay-at-home mom who does not have a paid career is credited as a 
full-time caregiver, but closer analysis reveals that substantial amounts of care are 
provided by babysitters, or unpaid caregivers like a friend, mother, sister, neighbor, 
or older child.  

Mapping amounts of care given and received will also make evident the 
differential entitlements of groups of people occupying different social positions 

                                                 
17 Approximately 1.4 million children between the ages of eight and eighteen serve 
as caregivers for parents, grandparents or siblings with disabilities (New York Times, 
May 23, 2016, “Supporting Children Who Serve as Caregivers,” by Jane E. Brody). I 
thank Amy R. Baehr for bringing this statistic to my attention. 

9

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



marked by race, ethnicity, and histories of slavery and colonization. In these ways, 
the first usage of the arrow of care map satisfies a few of the nonideal theorists’ 
requirements: it is quite centrally about dependency care, it identifies a category of 
labor that has sometimes coincided with slavery, and it unmasks ideologies. In order 
to do this, arrow of care maps must specify the gender, race, ethnicity, citizen 
status, and class of care receivers and providers.18 There will be different arrow of 
care maps to identify these caregiving arrangements—as a short time slice—over a 
year, or over the lifetime of individuals. In addition, it will be a theoretically robust 
project to delineate the boundaries of a society in order to be able to identify the 
“global care chain” and care drains from one country to another (Hochschild 2000; 
Kittay 2008; Gould 2008).  

Even in its descriptive form, the arrow of care map is abstract because it 
brackets facts about our relationships, such as the love parents have for their 
children, and that children have for their parents. Moreover, it does not present 
action-guiding recommendations with respect to caregiving, and so it says nothing 
about how people should approach their caring relationships in their real lives.19  

 
2.3. Mapping Care  

The arrow of care map is fundamentally an idea that admits of increasing 
levels of sophistication. I specify some of its uses here only to make more concrete 
the idea at its core. Below, I offer a schematic depiction of the idea of the arrow of 
care map. This particular depiction does not describe an actual caregiving 
arrangement, nor does it specify an ideal. Instead, the idea of the arrow of care is a 
way of quantifying caregiving labor as it occurs across both public and private 
domains. It offers an idea of tracking caregiving that is analogous to the idea of 
tracking blood pressure. In addition, just as the metric used for blood pressure is 
independent of any particular person’s blood pressure measurement, the idea of 
the arrow of care is an abstract idea that is not equivalent to the sum of what occurs 
in every person’s life. 

The arrow of care map tracks the intensity and duration of care given and 
received over the course of a lifetime. The width of the arrow corresponds to the 
intensity of care given and the direction of the arrow identifies care going from one 
person and towards another person. The length of the arrow corresponds to the 
duration of care.  

 
 

 

                                                 
18 I am indebted to Amy Mullin for pressing me to make this claim abundantly clear. 
19 For further discussion of this claim, see Section 4.1. 

10

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



Figure A. The Arrow of Care Map 
 

 
 
 
 

 
2.4. Categories of Care 

Arrow of care maps will be specified to track different subsets of labor, from 
the narrow category of dependency care to the broader inclusive category of care, 
and even personal services. Dependency care is care that seeks to respond to a 
legitimate need. Dependency care is the hands-on labor without which a person 
would not survive. It includes, for instance, feeding a baby, helping a paraplegic 
person with toileting, and assisting a person with mental disabilities to obtain food 
and live safely (Kittay 1999). I define dependency care, in part, by the relative 
immediacy of the harm that the vulnerable person will experience when care is 
absent. A paradigm case is that of a baby left in a crib without a caregiver; he will 
not survive long because he needs significant care to survive. This definition of 
dependency care does not offer necessary and sufficient conditions. It follows my 
view (Bhandary 2016b) that the nature of caregiving will change as caregiving 
becomes transparent in society, and the definition is consequently provisional.  

The caring work the map tracks maintains its historic connection with the 
category of labor that women have typically performed as mothers, nannies, 
mammies, daughters-in-law, and so forth. However, my account of caring work 
excludes the activities of teaching children to read, write, and perform basic math.20  

                                                 
20 I disagree with Daniel Engster’s account of care (2007, 27), which includes these 
activities and is broader than my view. For Engster, “Caring may be defined as 
everything we do directly to help individuals to meet their vital biological needs, 

11

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



The aim of the activity must be meeting a basic need, but attempts that fail 
should still be tracked. Consequently, I do not build efficacy into the caregiving that 
is tracked because caregiving is performed at varying degrees of skillfulness. 
However, I agree with Daniel Engster that actions that can be successfully 
completed without satisfying biological and developmental needs of a person will 
not count as caregiving labor. Engster offers the example of a cook in a restaurant, 
who can prepare and sell a meal without caring for anyone, whereas a parent 
cannot nourish a child without being sure that he or she actually ingests the food.  

Of course, there is a commonsense threshold for an activity to count as 
caregiving. A person who says they intended to care for their baby but leaves the 
baby crying in the crib for days while enjoying time out in town with friends violently 
harms the child and clearly fails to care for her. Thus, intentions matter. Some cases 
are more difficult to assess. For instance, imagine that Eugene makes his daughter 
soup while she is ill. He has taken off a day from work to do so. But he then spills the 
hot soup on his daughter, requiring that she receive medical care from someone 
else and also that her mother now take off a day from her work. The daughter still 
has not eaten anything and she now has burns that will require additional care to 
treat, resulting in additional caregiving by others. How do we quantify the care? Do 
Eugene’s contributions count as care, despite his ineffectiveness? If effectiveness is 
a necessary condition, then Eugene’s labor will not count, but a distributive 
assessment should include his attempted care, and therefore his earnest efforts 
should be counted in some way, although the additional needs he created must also 
be tracked. 

For these reasons, I reserve “effective” as a modifier to care, instead of 
saying that care must be minimally effective for it to count as care. In contrast, Amy 
Mullin offers a normative account of care for children as an activity that should not 
increase their vulnerability by failing to understand their material, psychological and 
developmental needs (Mullin 2014). Mullin’s account of care tells us what good care 
is, but tracking labor for an abstract representation of a caregiving arrangement 
should also track the care that aimed at meeting the person’s needs despite its 
failure to do so. 

It is well established in medical contexts that the labor of doctors that seeks 
to help a patient counts as work, even if they fail to achieve this aim. Other forms of 
care should follow the same convention. Imagine a case in which Beatrice, a six-
year-old child, sees a physician for a skin rash. The physician, Dr. Lyons, 
misdiagnoses it as a bacterial infection and prescribes an antibiotic, but the 

                                                 
develop or maintain their basic capabilities, and avoid or alleviate unnecessary or 
unwanted pain and suffering, so that they can survive, develop, and function in 
society” (2007, 29). 

12

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



antibiotic results in a larger rash as a side effect. Dr. Lyons has not provided effective 
or good care. Nonetheless, the health care system does not deduct the hours the 
physician spent seeing the child from the overall number of hours of care provided 
by the physician during that week. There is value in separating a quantification of 
labor hours from the quality of the labor—for it allows us to track that a person has 
spent time and effort aiming at an outcome, even if that person has failed to achieve 
the outcome, or if they do it poorly.  

In this example, Dr. Lyons has acted in ways that should be counted in a 
distributive account of labor. He has also offered bad care by failing to meet the 
needs of the recipient of care. Good care is the proper aim of caregiving. The reason 
for permitting bad care to count as care in the arrow of care map is that the map 
seeks to capture the labor people perform. Assessments of labor do not nullify labor 
that fails to meet a metric of high quality—which is what is illustrated with the 
medical case. Of course, it is of the utmost importance that vulnerable and 
dependent persons need good caregivers, not caregivers who are ineffective, and so 
we need educational programs and apprenticeships to train excellence in caregiving.  

 
2.5. Map Typologies 
 
Unpaid dependency care 

There will be several iterations of the abstract but descriptive arrow of care 
maps. The least complex map may be one that tracks unpaid dependency care 
because it will identify labor that is not currently tracked in any systematic way in 
virtue of its exclusion from the market—for the market is among the main sources 
of socially tracking labor.  

 
Dependency care that is poorly paid 

The need to track dependency-care labor that is paid is evident in the case of 
dependency work that is paid poorly so that the person is resultantly made 
vulnerable in virtue of performing the labor.21 If the ability to meet your needs and 
wants is dependent on money, then remunerating care well could relieve the 
disadvantage of it. However, I do not advocate increasing the domain and scope of 

                                                 
21 A host of additional questions arise at this stage—how do we weigh caregiving 
against other social contributions? How do we quantify burdens that result in 
compounded vulnerability? Answering these questions will be difficult, but 
answering them is just as necessary as any other comparative assessment of 
individual and social disadvantage for a theory of distributive justice. However, we 
may legitimately postpone answers to these questions until the present project is 
complete. 

13

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



market norms, for to do so treats all goods as fungible (Anderson 1993). If we live in 
a fully marketized economy, then caregiving should be marketized and the arrow of 
care would identify a type of social productivity that would be included in the gross 
domestic product. But I do not advocate that kind of society, for increasing the 
scope of markets will inevitably result in the creeping of market norms. Care is 
among the most fundamental domains of social cooperation, and it follows from its 
fundamentality that we will need to meet our care needs at least at a minimal level 
irrespective of whether we have a market economy or not, and those caregiving 
arrangements need to be fair. 

 
Personal services 

Another iteration of the arrow of care map must include personal services. I 
adopt Kari Waerness’s term and definition of “personal services” for the category of 
activity that a person does for someone else when that person is capable of meeting 
the basic need him or herself, or the need is not pressing and/or legitimate (1984).22 
Personal services include making tea for others, attending to the emotions of others, 
and serving as an audience to the ego of another. 

Why should theorists of justice bother mapping personal services if the 
needs they meet are not essential to human life? We should map this labor because 
it is work that people—often women—perform. There may be strong social pressure 
for a person to perform the labor even if the care needs to which he or she responds 
are not legitimate care needs. Thus, anti-oppression theory needs to include a 
personal-services map because the recipients of that labor reap benefits from it and 
the people performing it remain disadvantaged. A personal-services map is needed 
to track distributions of effort and time on the part of the person performing that 
labor, and it will include emotional work, ego-tending, and some forms of sex 
work.23 As in the case of dependency care, the first iteration of the personal-services 
map is unpaid personal services. The second iteration, which is no less important, is 
a map of poorly remunerated personal services. 

For distributive purposes, the burdens of caregiving—as both dependency 
care and personal services—are defined by the effort it requires in terms of time 

                                                 
22 Engster includes care for self in the activity of caring, and therefore it need not 
always be other-directed (2007, 32). I do not include care for self because the nature 
of caring for someone else and caring for oneself differ significantly. I seek to 
capture the category of other-directed care that promotes the recipient’s well-being 
and goals. See Bhandary (2016a) for further analysis of the differences between self-
directed and other-directed care. 
23 See MacKinnon (2011). 

14

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



and the demands on the person.24 A full analysis of the burdens of personal services 
must include psychological costs experienced by the caregiver. Personal services 
that radically harm the self of the caregiver are not justified and so a just social form 
should eliminate them. For instance, some forms of sexual labor that destruct the 
self of the sex worker should be eliminated from just social forms. In the meantime, 
the labor should be descriptively tracked. 
 
Unmet care needs  

A supplemental iteration of the dependency-care map requires an index of 
the dependency-care needs of vulnerable persons that no one meets in the current 
arrangement. One can expect that disabled children of parents living in poverty will 
have fewer of their needs met. There will likely be significant correlations between 
relative wealth and the level at which one’s care needs are met, but it should not be 
assumed that inequalities in wealth and income will subsume inequalities in care.  

 
3. Function Two: Fair Caregiving Arrangements  

The second function for the arrow of care map is to serve as an object for 
deliberations about what a fair caregiving arrangement would be. In this use, it plays 
an essential role in a form of revised contractarianism—one that pays attention to 
the need for sensitivity to the facts that are included and to those that are excluded 
in the use of abstractions, where abstraction is defined here as the bracketing of 
various predicates. The normative question for the most abstract iteration is: What 
would a just caregiving arrangement be? Is it one with arrows of equal widths and 
lengths? Must the arrows be exactly reciprocal, modeling something like exchange 
reciprocity? For instance, we can model the idea that children have duties to their 
parents to care for them with a set of equal arrows going to and from parents and 
children over the course of a lifetime. We must also ask: Is a just arrangement one 
where caregiving arrangements are clustered by gender or race? The outcome of 
deliberations behind the veil of ignorance will rule out arrangements that exploit 
any group of people. No group of people will want to be designated as the set who 
must be other-directed caregivers, even if they may choose to do so in their own 
lives. A form of hypothetical contractarianism with an “I could be anyone” device 
that is moderately fact-sensitive, insofar as it includes a set of important facts that 
are needed for it to satisfy its function, will exclude arrangements that designate 
ascriptive identity groups to be other-directed. The important facts that must be 
included are the race, gender, class, and geographic origin of the person, but these 

                                                 
24 The metric requires collaboration among philosophers, anthropologists, 
economists, and sociologists (see, e.g., Folbre and Bittman 2004 for an excellent 
case of time-use studies). 

15

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



facts are mutable, and different categories can become the most significant ones in 
different societies.  

Moving caregiving arrangements to the forefront of the philosophical agenda 
changes contractarianism because of the way that these arrangements have been 
assumed to be matters of the family. Thus, the exploitation that occurs when a 
person does activity C, but C is not considered socially useful or necessary, where C 
is care, will be eliminated. Insofar as performing care that is invisible in the tracking 
of societal benefits and burdens is part of a form of group-based oppression, the 
arrow of care map will eliminate that source of group-based oppression.25 

We can expect that the arrow of care map will show how racial hierarchies 
and disadvantage overlap with social norms about who should receive care, and 
who should provide it. A normatively acceptable caregiving arrangement cannot 
distribute care in ways that differentiate by race and class—nor should any 
inequalities systematically result from these group memberships. But these 
inequalities must be tracked in the descriptive form, and doing so requires 
bracketing details of people’s particular relationships and situations in order to 
document the extent to which inequalities in care given and received overlap with 
other categories of ascriptive identity groups. The arrow of care map is a framework 
within which it is intelligible to ask undertheorized questions, such as: Do African 
American men receive more or less care than white men? And from whom? It also 
offers a way of systematizing inquiry into more established areas of inquiry, such as: 
Do women of color receive less care than white women? On whom does the care 
deficit fall with the greatest effects? Additional data can be added to incorporate 
how immigrant and legal-resident status correspond to care arrangements.  

Different societies may arrive at different conclusions about the outcome of 
hypothetical acceptability because our actual experiences and values influence our 
intuitions about what we would choose in an ideal choice situation.26 Therefore, I 
leave open the principle that meets the criterion of hypothetical acceptability. In 
this way, the account I offer is deliberately less determinate than Rawls’s use of the 
original position. It leaves the principles that would result open to the person 
contemplating the thought experiment. In this usage, the map enables imagining a 
variety of caregiving arrangements and evaluating the acceptability of these 
arrangements—when one does not know where they will be in the arrangement 

                                                 
25 I think eliminating invisible forms of labor will eliminate the corresponding 
varieties of oppression via theoretical invisibility. However, other forms of 
oppression may emerge.  
26 I will not endeavor to defend this claim here. My view, in brief, is that the actual 
cultural context within which we deliberate as philosophers influences our 
estimations of what ideally rational people would decide, given certain inputs. 

16

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



(Stark 2000). This type of theorizing is a valuable mechanism to clarify our thinking 
about care in human societies, for it leads us to think about what an ideally just 
society would look like when it takes caregiving into account alongside all other 
goods. Because care is one good among several fundamental social primary goods, 
inequality in the domain of care may be part of an overall just arrangement. 

 
3.1. Adaptive Preferences and Legitimacy 

A principle of fair distribution of care requires an account of legitimate needs 
for care. Any metric of needs-differences must also take into account the ways in 
which the needs of privileged members of a society are perceived by all members of 
the society as more pressing. Due to the ways that people’s expectations are shaped 
by what is available to them, and by that to which they have become accustomed, 
we cannot assume that we will get a clear view of what legitimate expectations for 
care needs will be. A woman who is expected to be self-abnegating may well 
demand less care from others.27 More broadly, the set of phenomena theorized as 
adaptive preferences suffice to give reasons to not merely rely on existing 
preferences and states of affairs.28  

There are some clear cases of legitimate care needs. For instance, a child 
with a severe cognitive or physical disability will legitimately require more care—of 
greater intensity and duration—than a nondisabled adult man. However, there are 
likely racial and economic differences in the extent to which these needs for care are 
diagnosed and met. The arrow of care will need to track the care people are actually 
receiving, and add facts about race, gender, and disability, so that it can identify 
disparities in the care received by disabled black and white children, for instance. 
Race must also be included to be able to theorize about the significance of white 
disabled children receiving more care than disabled children of color.  

For instance, a principle of equality of capability would recommend that 
everyone receive the amount of care that they need to function (Engster 2007), and 
this requires accounting for differences in needs. Therefore, race-based aggregate 
data about care received and given are needed in addition to data about the 
amounts of care people receive above or below a threshold for basic functioning. If 
the gap in average care received by racial groups is very great, it gives a prima facie 
reason to scrutinize our intuitions about which needs are legitimate, and for whom. 
An asymmetry in care at a group-based level is an injustice even if it does not result 

                                                 
27 On “deformed desires” as a phenomenon of the self-abnegating woman, see 
Superson (2005). A related set of phenomena have been theorized as adaptive 
preferences. See, for example, Elster (1982) and Khader (2011).  
28 See Farrelly (2007) on the connection between maximal fact-sensitivity and the 
perpetuation of adaptive preferences. 

17

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



in disadvantages of income and wealth. A group-based caregiving arrangement 
raises a prima facie problem for justice because it will create problems for everyone 
who shares in the ascriptive identity group of the caregivers. Finally, historically 
informed understandings of oppression are necessary but not sufficient because the 
arrow of care map is a metric that can track unmapped disadvantage and relational 
asymmetries. 

 
4. Comparing the Arrow of Care Map to Care Ethics and Feminist Liberalism 

Applied within the Family 
 
4.1. The Challenge from Care Ethics 

One of the commitments that differentiate care ethics from the liberal 
contractarian tradition is the care-ethical view that emotions are a source of moral 
knowledge, which is also linked to the claim that our emotions and particular ties to 
real people should not be overcome in order to achieve impartiality. Care ethicists 
argue that moral theory must pay attention to particularity, and that theorizing 
about abstract subjects who have no ties to others misses what is important for a 
moral life.29 

The arrow of care map enables thinking about society-wide caregiving 
arrangements in a way that abstracts away from particular intimate relationships 
and from the emotions and meaning of care in our lives so that we may gain an 
understanding of the distributive justice or injustice of that arrangement. Unlike 
care ethics, it does not prescribe how we should respond to the people in our lives. 
Care ethicists’ accounts of caregiving relationships and caring persons offer 
sophisticated analyses of caring practices and the aims of good relationships, but 
they do not include assessments of the underlying distribution of benefits and 
burdens.30 As I have argued, a theory of care will be incomplete and inadequate if it 
does not have an abstract way to depict and evaluate caregiving arrangements, for 
abstraction is necessary to depart from existing conventions and assumptions about 
who should care for whom.  

Care ethics advances a related criticism of thinking in the terms of justice. 
Alison Jaggar characterizes this care ethical criticism of “justice thinking” in the 
following passage:  

 

                                                 
29 Held (2006), S. Collins (2015), and Noddings (1982) advocate a contextual 
approach to moral reasoning. 
30 Engster (2007) theorizes about a just society that provides care. Kittay (1999) 
theorizes about the disadvantage of dependency workers within the terms of her 
dependency theory.  

18

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



Justice thinking is impersonal and general because it regards both 
moral subjects and the objects of their moral concern in terms of 
their moral status as representatives of humanity . . . rather than in 
terms of their concrete specificity; care thinking is personal and 
particularized in that both carers and those cared for regard each 
other as unique, irreplaceable individuals. (1995, 190–191)31  
 
My view is diametrically opposed to care ethics on this matter because some 

level of abstraction is valuable in order to step back from the facts of one’s life. It 
enables the theorist to stand back from his or her family’s assumptions about who 
within the family ought to provide actual caregiving if they are to express love and 
concern. Abstraction in this form is distinct from the kind of abstraction against 
which Bernard Williams (1981) counsels—the application of impartial thinking to our 
loved ones. And I embrace care ethicists’ claims about the moral requirements of 
our actual relationships. The arrow of care does not recommend standing back from 
love for one’s family to evaluate whether the terms of the relationship are fair. As 
actual people, we are embedded within imperfect and unjust social forms, and it is 
likely that this will always be the case. The project of theorizing about the nature of 
the good life within any particular social form is a task for moral theory, philosophy 
of action, and practical reason. A moral theory that rejects all real relationships 
when they are unfair would leave real persons relation-less in any foreseeable 
future. Whereas an account of morality must discern our particularity and identify 
how it is that we can live good lives, a political theory should evaluate the basic 
structure as the matrix within which we live our lives. Consequently, political theory 
must engage in abstractions. My argument for abstraction counsels a departure 
from particularity at the level of social forms—not at the level of a person’s own life.  

Claims for abstraction at the level of justification are sometimes taken to 
imply that we should implement these claims abstractly in some way. Stark (2010) is 
right to argue that we can be sensitive to context when we apply ideals we have 
arrived at through the use of abstraction in justification. Therefore, care ethicists 
need not reject abstract care, for it invites attention to morally salient details at the 
level of an individual’s life. My view is that we must respond to our lives as they are 
in their particularity—to the facts about who our parents are, or to particulars about 
our brothers, sisters, children, neighbors, and so forth. The particularized details 
about who is in our intimate sphere are not defined voluntarily. It is not up to us to 
choose whether or not we have a neighbor who falls down after an illness and is 
alone in the world; or to determine our parents’ abilities, and particular types of 

                                                 
31 Jaggar goes on to criticize care ethics. See also Walker (2007) for a criticism of 
theoretical-juridical approaches to moral theory. 

19

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



physical or mental decline towards the end of their lives; or to decide if we have a 
sibling, and to decide on that sibling’s need for support or help—for example, if he 
or she has a severe disability. In all of these cases, and in our real lives, we will have 
responsibilities that arise from the details of the real world, from contingencies that 
have nothing to do with our choices as well as from actions that we have chosen, 
but the consequences of which we had no way of predicting.  

What I wish to underscore, though, is that these responsibilities and choices 
occur against the backdrop of a social form that is regulated by some account of 
distributive justice. Our sense of our responsibilities is influenced by background 
conditions and culture.32 For instance, is a son responsible for his mother’s needs as 
she ages, or is his wife responsible for them? Is the younger son responsible, or is it 
the older one? In some pockets of Indian cultures, it is the eldest son who is held 
responsible for his parents, but hands-on caregiving responsibilities are assigned to 
his wife. Is the society-at-large responsible for an aging population, or is it the 
individuals themselves who should plan for their future with financial savings that 
will allow them to pay a professional caregiver? Or is it children overall who are 
responsible for their parents’ care? If a community overall is responsible for meeting 
care needs, how does this actually break down by race and ethnicity? If white 
women regularly expect a higher level of needs-satisfaction while women of color 
have adapted to meeting fewer of their needs, then will a distributive assessment 
that relies on subjective assessments of well-being be adequate?33 To answer these 
questions, we will need to evaluate a number of different permutations of 
caregiving arrangements. 

Faulty abstractions, not abstractions per se, are the cause of 
contractarianism’s perpetuation of oppression. Consequently, a criterion of 
adequacy for contemporary forms of liberalism is that they must do the theoretical 
work needed to identify the abstractions the theory should employ. The arrow of 
care map meets these criteria as a conceptual device that identifies caregiving 
arrangements and then holds them up as an object of hypothetical consent. When 
the labor of caregiving is made visible, a new layer of privilege and disadvantage 
becomes evident that can address intersectionality by showing how the expectation 
to receive care and the social expectation to give care can map onto race, ethnicity, 
gender, caste, and social class. The concept of the arrow of care thereby disrupts 
status quo social arrangements in societies where caregiving arrangements are 

                                                 
32 See Walker’s (2007) analysis of moral understandings, which has influenced this 
account. 
33 See Khader (2011) for the argument that “inappropriate adaptive preferences” 
must be identified in virtue of their conflict with an objective account of well-being. 

20

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



enforced through informal mechanisms like widely accepted expectations and 
narratives.34  

 
4.2. No Action-Guidance 

The arrow of care map differs from two influential feminist liberals’ approach 
to inequality and caregiving responsibility.35 Susan Okin argued for the application of 
principles of justice within family—and in doing so she rejected the Rawlsian view 
that representatives be heads of households.36 Although I share Okin’s project of 
feminist Rawlsian contractarianism, the arrow of care map departs from Okin in that 
it maintains a higher level of abstraction to accommodate cross-cultural differences 
in caregiving and familial structures while Okin applies Rawlsian reasoning within 
the family. Moreover, I do not assume the family as a starting point, and in this way, 
operate at a greater level of abstraction. Instead, the starting point for the arrow of 
care is a claim that we all need to receive care and therefore caregiving is one of the 
core functions that a scheme of social cooperation must satisfy. 

Another foil for the arrow of care map is Jean Hampton’s (1993) feminist 
contractarian test for personal relationships that tells participants to assess whether 
they would accept the terms of the relationship as the product of an unforced 
agreement. In contrast with Hampton’s test, the arrow of care is not a method of 
abstraction for individuals to map their relationships in service of demanding 
equality in the care given and received within each one of their intimate 
relationships. 37 Instead, as a framework for mapping society-wide distributions 
which then evaluates their fairness, the arrow of care map does not seek to proffer 
action-guiding recommendations for individual lives. Our lived engagement with 
families and friends should be guided by values that extend far beyond justice. 
 

                                                 
34 I do not recommend that individuals map their relationships within the arrow of 
care. It is not a comprehensive assessment of fairness, nor does it provide an 
account of what it means to be concerned about someone or to have a legitimate 
responsibility towards a particular other.  
35 Ann Cudd’s (2006) account of society-wide oppression offers a systematic analysis 
of the context of choice, and so the arrow of care map is structurally closer to her 
theory than it is to Hampton’s or Okin’s theories. 
36 See Baehr (1996) for a discussion of the way Okin’s view changed over time. See 
also Abbey (2016). 
37 Similarly, I do not offer a solution to directly address cooperative conflicts for 
people here and now (Sen 2011); nor does the arrow of care embed within it ways 
to increase women’s agency in diverse cultural contexts (Meyers 2002; Khader 
2011). 

21

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



Conclusion 
This paper has advanced a concept—the arrow of care map—that gives us a 

way to think about care that is not embedded in a set of assumptions that are 
culturally conservative insofar as they are specific to the social form in which we 
live. It is a concept in the sense that it sets forth a problem. The problem it sets forth 
is that every society needs to advance an arrangement to secure its care needs. 
There will then be multiple conceptions to specify what a good caregiving 
arrangement is—these conceptions are solutions to the problem set forth by the 
concept, and they will take the form of various arrow of care maps.38 The concept of 
the arrow of care map takes a bird’s-eye view of the labor of care—it does not seek 
to identify the nuances of relations of love and affection, and it abstracts away from 
assumptions about family units or socialized care to offer a way to tune our 
intuitions about caregiving in human life.  

The paper has also advanced a methodology for ideal theory that seeks to be 
non-ideological. The insights and intuitions on which it relies are informed by 
existing states of affairs, and it is explicit about its value commitments. By advancing 
this abstract concept, and limiting my theoretical project to this conceptual work, 
my account may appear to conflict with the commitments of a dominant strain of 
feminism and anti-racist philosophy that advocates nonideal theory. Whereas 
nonideal theories locate the core philosophical project in improving the world, and 
seek to do so quite directly, this paper has shown that theory that is “ideal”—in the 
sense that it seeks to clarify our values—must also begin with insights from the 
pressing problems of society, and then advance concepts that can better identify 
these problems. These theories must interrogate and evaluate the normative claims 
that are embedded in these concepts. Finally, by formulating the concept of the 
arrow of care map, I hope to enable further theorizing about the facts of societal 
caregiving arrangements in order to imagine new visions of just societies. 
 
 
Works Cited 
Abbey, Ruth. 2016. “Susan Okin's Justice, Gender, and the Family: Twenty-Five Years 

Later.” Hypatia 31 (3): 636–637.  
Anderson, Elizabeth. 1993. Value in Ethics and Economics. Cambridge, MA: Harvard 

University Press. 
Baehr, Amy R. 1996. “Toward a New Feminist Liberalism: Okin, Rawls, and 

Habermas.” Hypatia 11 (1): 49–66. 

                                                 
38 I thank Julie Walsh for identifying the similarity between Rawls’s concept-
conception distinction and my own use of the arrow of care map. 

22

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



———. 2004. “Feminist Politics and Feminist Pluralism: Can We Do Feminist Political 
Theory Without Theories of Gender?” Journal of Political Philosophy 12 (4): 
411–436. 

———. 2013. “Liberal Feminism: Comprehensive and Political.” In Feminist 
Interpretations of Rawls, edited by Ruth Abbey, 150–166. University Park: 
Pennsylvania State University Press. 

———. 2017. "A Capacious Account of Liberal Feminism." Feminist Philosophy 
Quarterly 3 (1): article 4. doi:10.5206/fpq/2016.3.4.  

Bhandary, Asha. 2010. “Dependency in Justice.” Hypatia 25 (1): 140–156. 
———. 2016a. “A Millian Concept of Care: What Mill’s Defense of the Common 

Arrangement Can Teach Us About Care.” Social Theory and Practice 41 (1): 
155–182. 

———. 2016b. “Liberal Dependency Care.” The Journal of Philosophical Research 
41:43–68. 

Collins, Patricia Hill. 2000. Black Feminist Thought. New York: Routledge. 
Collins, Stephanie. 2015. The Core of Care Ethics. London: Palgrave Macmillan. 
Crenshaw, Kimberlé. 1991. “Mapping the Margins: Intersectionality, Identity Politics, 

and Violence against Women of Color.” Stanford Law Review 43 (6): 1241–
1299. 

———. 2015. "Say Her Name: Politicizing Black Women's Vulnerability to State 
Sanctioned Violence." Keynote lecture at the Association for Feminist Ethics 
and Social Theory (FEAST) Conference, Clearwater, FL, October 2. 

Cudd, Ann. 2006. Analyzing Oppression. Oxford: Oxford University Press. 
Dimock, Susan. 2008. “Why All Feminists Should Be Contractarians.” Dialogue 47 (2): 

273–290. 
Elster, Jon. 1982. “Sour Grapes – Utilitarianism and the Genesis of Wants.” In 

Utilitarianism and Beyond, edited by Amartya Sen and Bernard Williams, 
219–238. Cambridge: Cambridge University Press. 

Engster, Daniel. 2007. The Heart of Justice. New York: Oxford University Press. 
Farrelly, Colin. 2007. “Justice in Ideal Theory: A Refutation.” Political Studies 55 (4): 

844–864. 
Folbre, Nancy, and Michael Bittman, eds. 2004. Family Time. New York: Routledge. 
Gould, Carol. 2008. “Recognition in Redistribution: Care and Diversity in Global 

Justice.” Southern Journal of Philosophy 46 (S1): 91–103. 
Greyser, Naomi. 2017. On Sympathetic Grounds. New York: Oxford University Press. 
Hampton, Jean. 1993. “Feminist Contractarianism.” In A Mind of One’s Own, edited 

by Louise Antony and Charlotte Witt, 227–256. Boulder, CO: Westview Press.  
Hartley, Christie, and Lori Watson. 2010. “Is a Feminist Political Liberalism Possible?” 

Journal of Ethics and Social Philosophy 5 (1): 1–21.  

23

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



———. 2012. “Political Liberalism, Marriage and the Family.” Law and Philosophy 31 
(2): 185–212.  

Hay, Carol. 2013. Kantianism, Liberalism, and Oppression. New York: Palgrave 
Macmillan. 

Held, Virginia. 1987. “Non-Contractual Society: A Feminist View.” In Science, 
Morality, and Feminist Theory, edited by M. P. Hanen and Kai Nielsen, 11–
137. Calgary, AB: University of Calgary Press. 

———. 2006. The Ethics of Care. Oxford: Oxford University Press. 
Hochschild, A. R. 2000. “Global Care Chains and Emotional Surplus Value.” In On the 

Edge: Living with Global Capitalism, edited by W. Hutton and A. Giddens, 
130–146. London: Jonathan Cape. 

Kang, Hye Ryoung. 2016. “Can Rawls’s Nonideal Theory Save His Ideal Theory?” 
Social Theory and Practice 42 (1): 32–56. 

Khader, Serene. 2011. Adaptive Preferences and Women’s Empowerment. New York: 
Oxford University Press. 

Kittay, Eva. 1999. Love’s Labor: Essays on Women, Equality, and Dependency. New 
York: Routledge. 

———. 2008. “The Global Heart Transplant and Caring Across National Boundaries.” 
The Southern Journal of Philosophy 46 (S1): 138–165. 

Jaggar, Alison. 1995. “Care as a Feminist Practice of Moral Reasoning.” In Justice and 
Care: Essential Readings in Feminist Ethics, edited by Virginia Held, 825–833. 
Boulder, CO: Westview Press. 

Lloyd, S. A. 1998. “Toward a Liberal Theory of Sexual Equality.” Journal of 
Contemporary Legal Issues 9:203–224. 

MacKinnon, Catharine. 2011. “Trafficking, Prostitution and Inequality.” Harvard Civil 
Rights–Civil Liberties Law Review 46 (1): 271–309. 

Meyers, Diana Tietjens. 2002. Gender in the Mirror. New York: Oxford University 
Press. 

Mills, Charles. 1997. The Racial Contract. New York: Cornell University Press. 
———. 2004. “‘Ideal Theory’ as Ideology.” In Moral Psychology, edited by Peggy 

DesAutels and Margaret Urban Walker, 163–181. Lanham, MD: Rowman and 
Littlefield. 

———. 2012. “Liberalizing Illiberal Liberalism.” Closing keynote address for the 
Oxford Graduate Conference in Political Theory, Oct. 4. University of Oxford 
Podcasts. MS on file with author. 

Mullin, Amy. 2014. “Children, Vulnerability and Emotional Harm.” In Vulnerability: 
New Essays in Ethics and Feminist Philosophy, edited by Catriona Mackenzie, 
Wendy Rogers, and Susan Dodds, 266–287. New York: Oxford University 
Press. 

Noddings, Nel. 1982. Caring. Berkeley: University of California Press. 

24

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5



Okin, Susan. 1989. Justice, Gender and the Family. New York: Basic Books. 
O’Neill, Onora. 1996. Towards Justice and Virtue: A Constructive Account of Practical 

Reasoning. Cambridge: Cambridge University Press. 
Pateman, Carol. 1988. The Sexual Contract. Palo Alto, CA: Stanford University Press. 
Rawls, John. 2001. Justice as Fairness: A Restatement. Edited by Erin Kelly. 

Cambridge, MA: Harvard University Press. 
———. 2002. The Law of Peoples. Boston: Harvard University Press. 
Reiheld, Allison. 2015. “Just Caring for Caregivers: What Society and the State Owe 

to Those Who Render Care.” Feminist Philosophy Quarterly 1 (2): article 1. 
doi:10.5206/fpq/2015.2.1. 

Ruddick, Sara. 1989. Maternal Thinking. Boston: Beacon Press. 
Schouten, Gina. 2015. “Citizenship, Reciprocity, and the Gendered Division of 

Labor.” Politics, Philosophy and Economics. 
doi: 10.1177/1470594X15600830. 

Schwartzman, Lisa. 2006a. Challenging Liberalism: Feminism as Political Critique. 
University Park: Pennsylvania State University Press. 

———. 2006b. “Abstraction, Idealization, and Oppression.” Metaphilosophy 37 (5): 
565–588. 

———. 2016. “Action Guidance, Oppression, and Nonideal Theory.” Feminist 
Philosophy Quarterly 2 (1): article 4. doi:10.5206/fpq/2016.1.4. 

Sen, Amartya. 2011. The Idea of Justice. Cambridge, MA: Harvard University Press. 
Spelman, Elizabeth. 2002. Repair: The Impulse to Restore in a Fragile World. Boston: 

Beacon Press. 
Stark, Cynthia A. 2000. “Hypothetical Consent and Justification.” Journal of 

Philosophy 97 (6): 313–334. 
———. 2007. “How to Include the Severely Disabled in a Contractarian Theory of 

Justice.” Journal of Political Philosophy 15 (2): 127–145. 
———. 2010. “Abstraction and Justification in Moral Theory.” Hypatia 25 (4): 825–

833. 
Stemplowska, Zofia. 2008. “What’s Ideal about Ideal Theory?” in “Social Justice: 

Ideal Theory, Nonideal Circumstances,” edited by Ingrid Robeyns and Adam 
Swift, special issue, Social Theory and Practice 34 (3): 319–340. 

Stemplowska, Zofia, and Adam Swift. 2015. “Ideal and Nonideal Theory.” In The 
Oxford Handbook of Political Philosophy, edited by David Estlund, 373–389. 
New York: Oxford University Press. 

Superson, Anita. 2005. “Deformed Desires and Informed Desire Tests.” Hypatia 20 
(4): 109–126. 

Tessman, Lisa, ed. 2009. “Introduction.” Feminist Ethics and Social and Political 
Philosophy: Theorizing the Non-ideal. New York: Springer Science and 
Business Media. 

25

Bhandary: The Arrow of Care Map

Published by Scholarship@Western, 2017



———. 2015. Moral Failure. Oxford: Oxford University Press. 
Tronto, Joan. 1993. Moral Boundaries. New York: Routledge. 
Valentini, Laura. 2012. “Ideal v. Non-ideal Theory: A Conceptual Map.” Philosophy 

Compass 7 (9): 654–664. 
Varden, Helga. 2006. "Kant and Dependency Relations: Kant on the State’s Right to 

Redistribute Resources to Protect the Rights of Dependents." Dialogue: 
Canadian Philosophical Review 45:257–284. 

Waerness, Kari. 1984. “Caring as Women’s Work in the Welfare State.” In Patriarchy 
in a Welfare Society, edited by H. Holter, 67–87. Oslo: Universitetforlaget. 

Walker, Margaret Urban. 2007. Moral Understandings. Second edition. Oxford: 
Oxford University Press.  

Williams, Bernard. 1981. “Persons, Character and Morality,” In Moral Luck, edited by 
James Rachels, 1–9. Cambridge: Cambridge University Press. 

 
 
ASHA BHANDARY is Assistant Professor of Philosophy at the University of Iowa. Her 
areas of specialization include feminist ethics, social and political philosophy, and 
autonomy theory. Her published articles on liberalism, care, autonomy, and 
distributive justice have appeared in Hypatia, Journal of Philosophical Research, 
Social Theory and Practice, and Journal of Political Philosophy. 

26

Feminist Philosophy Quarterly, Vol. 3 [2017], Iss. 4, Art. 5

https://ir.lib.uwo.ca/fpq/vol3/iss4/5


	Feminist Philosophy Quarterly
	2017

	The Arrow of Care Map: Abstract Care in Ideal Theory
	Asha L. Bhandary
	Recommended Citation


	The Arrow of Care Map: Abstract Care in Ideal Theory