1

Social Capital and the Capability Approach:

are Putnam and Sen incompatible bedfellows?(*)

Flavio Comim and Frank Carey

Dept. Economics/UFRGS and Von Hügel Institute
St Edmund's College, University of Cambridge
fvc1001@cam.ac.uk and fc222@cam.ac.uk

Paper to be presented at the 2001 European Association for Evolutionary Political Economy, Siena, 8-
11 November 2001. Comments are welcome.

Introduction

Few concepts have achieved such prominence in the social sciences in recent years as the concepts of
social capital and capability. Indeed, the concept of social capital –in its distinct origins and versions
such as those of Gleen Loury (1977), Pierre Bourdieu (1979 and 1990) and James Coleman (1988,
1990) to quote just a few- became at Robert Putnam’s (1993a, 1993b, 1995 and 2000) hands a
widespread tool of analysis concerning the consequences of social networks and cooperative forms of
behaviour on economic performance and growth. Dasgupta and Serageldin, for instance, have claimed
that (1999: x), “It is difficult to think of an academic notion that has entered the common vocabulary
of social discourse more quickly than the idea of social capital. Not only do academic journals devote
special issues to discuss the concept, journalists make frequent references to it and politicians pay
regular homage to it.” The importance of Putnam’s work to the development of the social capital
literature cannot be emphasised enough; for instance, Fine (2001: 18) has suggested that “Putnam has
become the crown prince of social capital”.

Similarly, the concept of capability, originally introduced and developed by Amartya Sen (1980,
1985a, 1985b, 1992 and 1999), has been extremely influential at an academic and institutional level.
Some of UNDP’s and World Bank’s most important recent publications, such as the Human
Development Report and the World Development Report have been largely influenced by Sen’s
approach. Pressman and Summerfield, for instance, have argued that (2000: 98) “the capabilities
approach leads to fundamental changes within the field of economic development. It has helped
change the development paradigm from promoting economic growth to promoting human well-
being.”

(*) An earlier version of this paper has been presented at the Workshop on Poverty, Social Capital and
Development, St Edmund's College, Cambridge, 27 January 2000.
It is interesting to note that both concepts seem to have acquired the referred prominence
approximately at the same time and for the same sort of general reasons. Both concepts are relevant to
social and development issues related to poverty1 and inequality problems; they have both been
guided by practical considerations regarding policy implications for development; they put forward
conceptual structures that claim to overcome the narrowness of development analysis based
exclusively on economic considerations; they also appear to be in affinity with contemporary notions
of human development, such as the one put forward by Ul Haq (1998). It seems that there is a prima
facie case for an investigation of the relation between these two concepts, in particular, because given
their influence, there has been so little intersection between them in the literatures that they have
generated.

The main argument of this paper consists of two interrelated but independent claims. The main
justification for these claims comes from an examination of the limitations and potentialities of both
concepts. The first, which could be called the weak claim, suggests that a ‘simultaneous’ reading of
both concepts is possible and might be illuminating. The second, which could be called the strong
claim, proposes that a ‘complementary’ reading is needed if these concepts are to be applied to
development issues. This last claim should be considered with caution on the basis that it could imply
a de-contextualisation of the particular research agendas of Putnam and Sen. Whereas Putnam’s wide
objective is to investigate how social networks and voluntary associations develop and result in
economic prosperity, Sen’s general concern is with the evaluation and assessment of social
arrangements. However, this de-contextualisation is an issue that cannot be settled a priori. Rather, it
is the issue to be examined in this paper.

The paper is organised into three parts. The first part examines the main differences, limitations and
gaps between Putnam’s and Sen’s approaches. It discusses the reasons for avoiding any simultaneous
reading of their contributions. The second part investigates their points of complementarity and

1 Poverty ranks high in the conceptual implications provided by social capital and capability analyses. Putnam,
who sees poverty as part of the poor people’s inability to associate for collective endeavours, has argued that
(1993: 159) “studies of the Italian case have the potential to contribute importantly to our understanding of why
many (but not all) Third World countries remain inextricably and inexplicably mired in poverty”. He also quotes
Esman and Uphoff (1984: 40), who corroborate the view that (p. 221) “A vigorous network of membership
organizations is essential to any serious effort to overcome mass poverty under the conditions that are likely to
prevail in most developing countries for the predictable future. Putnam also mentions that (2000: 318), “Precisely
because poor people (by definition) have little economic capital and face formidable obstacles in acquiring
human capital (that is, education), social capital is disproportionately important to their welfare.” Sen, who
defines poverty as a deprivation of basic capabilities, has suggested that (1992: 151) “the reorientation from an
income-centred to a capability-centred view gives us a better understanding of what is involved in the challenge
of poverty”. He has also noted that (1999: 88) “ ‘real poverty’ (in terms of capability deprivation) may be, in a
significant sense, more intense than what appears in the income space”.. The important point to note here is the
potential relevance of the concepts of social capital and capability for addressing poverty issues. One practical
question of interest for policy making might be raised here: if we are interested in tackling poverty problems,
should we follow the social capital or the capability path?
develops the weak and strong claims mentioned above. Finally, the third part concludes by suggesting
the notion of social capability as a way of describing the institutional and ethical concerns expressed
by the concepts of social capital and capabilities.

Incompatible bedfellows?

The concepts of social capital and capability belong to distinct theoretical lineages. The concept on
social capital, based on the significance of communitarian ties to economic prosperity, is in its
contemporary perspective largely a product of intellectual work made originally in the fields of
sociology and political science. Alternatively, the concept of capability, discussed within the context
of the Rawlsian liberalism, is mainly a product of intellectual work made in the fields of moral
philosophy and economics. There is enough diversity in the characterisation of these concepts
(especially in the social capital literature) to blur the above distinction but it is possible to note how
the conceptual framework (theoretical categories) and tacit postulates used by the main perspectives
on social capital (on that see Woolcock and Narayan, 2000) differ from the ones used by the main
contributors to the capability approach. While the main categories of social capital are built around,
for instance, the notions of social networks, cooperation, social trust, community life and density of
associations, the main categories of the capability approach are elaborated from the notions of human
freedom, individual agency, individual choice, individual responsibility and individual functionings
and capabilities. Thus, taken at their face value (based on the nature of the categories they employ) it
seems that these concepts are founded on completely distinct formal principles and that any attempt to
integrate them would have to face the contradiction of having individuals and communities as the
main unit of analysis2.. In what follows we provide a brief critical discussion of these concepts.

Social capital3 brings into economics an important feature of social and political science; the feature
of actors as socialised beings, whose behaviour is not merely governed by their self-interest but by
norms, institutions and social bonds among individuals. Indeed, this is perhaps the main reason for the
prominence of the concept of social capital in the development literature. This feature of social capital
was emphasised by Coleman (1988, 1990) –the main theoretical source of Putnam’s work on social
capital- who proposed an investigation on the way social organisation affects the functioning of
2 Not everyone would agree on this point. For instance, Fine (2001: 18) strongly criticises “the rational choice
origins of social capital”.. He argues that (2001: 26) “Social Capital is an explicit recognition of adding society to
an otherwise asocial economy”.. Fine would claim that the individual is the unit of analysis of social capital.
3 In face of the many alternative interpretations of social capital, it is important to note that it must be
characterised according to the particular perspective we have in mind (for that see Woolcock, 1998; Portes,
1998, and the contributions in Dasgupta and Serageldin, 1999 and in Baron, Field and Schuller, 2000).
Therefore, we focus here on the Coleman/Putnam characterisation of social capital.
economic activity. As he observed (1988: S95), “The conception of social capital as a resource for
action is one way of introducing social structure into the rational action paradigm”4. In its simplest
terms, the notion of ‘social’ refers, as Coleman puts it (1988: S100-1), to ‘relations among persons’.
The notion of ‘capital’ presupposes that relations among persons constitute a form of resource with
economic consequence. This intrinsic vagueness in the definition has encouraged authors to include
all sorts of relations among persons in the definition of social capital.

Putnam, acknowledging the origin of the term in Coleman’s work, refers to social capital as those
(1993: 167) “features of social organization, such as trust, norms, and networks, that can improve the
efficiency of society by facilitating coordinated actions”.. Ultimately, what seems to matter in his
definition is not merely the role of trust, norms, and networks in explaining economic performance,
but rather, the character of civic life, that is, the patterns of civic involvement and social solidarity,
and its economic implications. Do citizens actively participate in public affairs? Is a community
bound by horizontal relations of reciprocity and cooperation or by vertical relations of authority and
dependency? More recently, Putnam (2000) seems to concentrate his definition of social capital on
the existence of ‘networks of reciprocity’ or ‘networks of social connection’ that produce norms of
reciprocity and trustworthiness among individuals. As he summarises (2000: 171), “Social capital is
about networks”. By developing this concept, Putnam shows how people’s motivations and choices
are not simply triggered by their individual circumstances, but also by their social insertion and
involvement in community life.

According to Coleman (1988), social networks, in order to be qualified as social capital, need not only
to facilitate economic performance but also:

1. be subject to closure, because the closure of the social structure is important to warrant
trustworthiness and effectiveness to norms and expectations;

2. be of the kind of multiplex relations, such that resources of one relationship can be
appropriated for use in others. This would imply, as Ostrom (1999: 174) has observed, that
investments in “social capital may be developed as a by-product of other activities as well as
purposely”;

3. have public goods properties, such that they will benefit all those individuals that are part of
such structure independently of their participation.

From a social capital perspective, what can or cannot be achieved by a society depends on the

4 Putnam has noted in his 1999 Marshall Lectures that social capital has been criticised by economists for
smuggling soft concepts into economics and criticised by sociologists for smuggling ‘rationality’ into sociology.
properties of the social (or collective) structure within which individuals exercise their choices. What
can or cannot be achieved is independent of individual attributes or abilities. Indeed, it is not relevant
to the concept of social capital if a particular individual can or cannot participate in community
activities. What matters for the concept of social capital is the existence of system-level behaviour
(Coleman, 1990) that is, the existence of relations, such as trust, reciprocity, social integration, etc.,
that cannot be achieved at an individual level, neither be derived from a characterisation of
individuals nor reduced to it. It emphasises the factors of interdependence behind collective action
and externality problems. It is important to note that social capital elements should not be seen as
external to the characterisation of the individuals, as if they were exogenous constraints, but rather as
their constitutive elements. For instance, trust is a constitutive property of a relationship that cannot
be reduced to any of the individuals holding it.

Much has been said about social capital not having the characteristics of (physical) capital. For
instance, Solow (1999: 6) has claimed that social capital “is a bad analogy with the concept of
capital”. Capital, he argues (p. 6) “stands for a stock of produced or natural factors of production that
can be expected to yield productive services for some time”, and he cannot see how shared attitudes
and behaviour could provide earnings or a rate of return with economic sense. Solow’s critique of the
concept of social capital is due not only to its measurability problems but mainly to its lack of
correspondence with the intuitive definition of capital in economics. Similarly, Arrow (1999) argues
that a defining characteristic of social networks is that they are constituted for social reasons that have
no direct relation with economic value to their participants. However, he goes further than Solow
when claiming that (1999: 4) “I would urge abandonment of the metaphor of capital and the term
‘social capital’.” According to him, to qualify as ‘capital’, a factor of production must involve a)
extension in time; b) deliberate sacrifice for future benefit and c) alienability. He argues that the
concept of social capital –that supports the view that economic benefits come from unintended
consequences of social networks- fails to fulfil the criterion b) and for this reason cannot qualify as
‘capital’ (for further discussion on the differences between social and physical capital see Ostrom,
1999)5.

Behind these critiques of the notion of ‘capital’, in social capital, there is a sociological dispute –
largely ignored by social capital researchers- between economists and social scientists over the use of
capital as ‘a principle of action’ in social theorising. A very remarkable characteristic of that is the
instrumentalisation of social relations that can be seen in some discussions on social capital. For
instance, Putnam (2000: 19) emphasises that “social contacts affect the productivity of individuals

5 Moreover, it is not unusual to find the ‘social’ dimension of social capital loosely defined such that the links
between the ‘social’ and the ‘capital’ are not clearly established; it appears that any social has the potential to
become capital, per definition.
and groups” and that social capital was created to “call attention to the ways in which our lives are
made more productive by social ties.” Similarly, Putnam (1993), in his discussion on ‘the Italian
regional experiment’, examines how institutional performance (instrumentally) depends on
institutional design, organisational determinants and socio-economic and socio-cultural factors. In
particular, he puts emphasis on the character of civic life, or what he calls ‘the civic community’ and
its influence on institutional performance. The ideal of civic virtues –marked by interest in public
issues and active participation in public affairs- is at the core of Putnam’s idea of civic engagement
and civic community. He takes Tocqueville’s comments on Democracy in America as the building
block of his theory of associations.. He stresses the role of norms and values of the civic community
and the (instrumental) consequences of civic associations on individuals and on the wider polity. As
he puts it (1993: 89-90),

Internally, associations instill in their members habits of cooperation, solidarity, and public-
spiritedness. Tocqueville observed that ‘feelings and ideas are renewed, the heart enlarged,
and the understanding developed only by the reciprocal action of men one upon another’. (…)
Participation in civic organizations inculcates skills of cooperation as well as a sense of
shared responsibility for collective endeavours.

There are two different arguments being made here. The first is that individuals’ acts could be
partially explained by the habits and norms prevalent in the communities they live. The second is that
these habits and norms are also responsible for the creation of values and virtues that individuals are
expected to hold as members of these communities.

In operational terms, Putnam defines ‘the civic community’ by the density of local, cultural and
recreational associations. He clearly gives the impression that ‘associations’ is all that matters. He
seems to over-emphasise the importance of all sorts of associations as a determinant of the nature of
public life at the expense of other factors that facilitate coordination and cooperation in society.
Many contributions, such as those of Levi (1996), Foley and Edwards (1996) and Szreter (2001), have
called attention to remaining “omissions” in Putnam’s work concerning the role of the state, political
organisations, government and culture. These “omissions” can be seen as opportunities of
enlargement of the scope of the social capital theory.

Within this scope, it could be argued that another remarkable “omission” in Putnam’s work appears to
concern the role of justice and rights as foundations of social capital. Law-abidingness appears in
Putnam’s analysis as a direct and straightforward result of civic behaviour. It does not enter in his
analysis as a causal or foundational element. However, it could be argued that issues related to justice
are at the very root of the foundations of society and of any theory of associations among individuals.
To a certain extent, this “omission” is an internal limitation of Putnam’s emphasis on associations as
the main engine behind constitution and development of civic communities. It could be contrasted
against other “omissions” (e.g. the role of state) that are external to the workings of civic
communities. Therefore, it might be suggested that justice is an intrinsic issue to the definition of
Putnam’s theory of associations and thus that it should not be ignored.

Justice, as put Adam Smith in his Theory of Moral Sentiments(1976 [1759]), “is the main pillar that
upholds the whole edifice” [of society]. More recently, Rawls has stated that (1971: 476) “without a
common or overlaping sense of justice civic friendship cannot exist” and that (1971: 6) “In the
absence of a certain measure of agreement on what is just and unjust, it is clearly more difficult for
individuals to coordinate their plans efficiently in order to insure that mutually beneficial
arrangements are maintained. Distrust and resentment corrode the ties of civility”.. Indeed, social
cooperation appears to depend on a division of fundamental rights, duties and advantages in society.
Or, as Rawls puts it, (1971: 142), “a conception of justice is to be the public basis of the terms of
social cooperation”. The morality of association, which seems to be the main principle used by
Putnam in the formulation of his theory of associations, is nothing but a constitutive part of a wider
concept of morality which involves among other elements, what Rawls calls ‘the morality of
principles’, where there is a common allegiance to justice and rights. Therefore, it might be argued
that the foundation of Putnam’s notion of civic community cannot be claimed without a previous
discussion of the conception of justice required for social cooperation ..

But this “omission” in Putnam’s work appears to be just the tip of an iceberg. Putnam’s disregard for
the public sense of justice as a basis for civic friendship seems to be a sign of a narrow normative
approach to the evaluation of social arrangements. Whereas his normative analysis is exclusively
focused on the criteria of responsiveness and efficiency, important normative considerations related to
justice and equality remain ignored by the concept of social capital he puts forward. The concept of
social capital does not seem to provide an adequate context to investigate, for instance, the problem of
the ‘disorganising effects’ of economic inequality on civil society. To a certain extent, the
instrumental nature of social capital given by the use of “capital” as ‘a principle of action’ does not fit
well with the non-instrumental characteristic of a concept of justice based on rights. For instance, how
could we assess a social capital arrangement that is conducive to economic growth but that
undermines the rights of the poorest people in society? Putnam’s normative foundation, based on an
assessment of consequences, does not seem broad enough to include this last kind of evidence.

Should social capital be concerned only with an instrumental view of communities and associations, it
will not be able to address normative or ethical issues that are foundational and constitutive of social
arrangements. Thus, an expansion of the normative basis used by Putnam seems to be an important
step in broadening the concept of social capital. One might wonder to what extent the concept of
“capital” constraints a broader understanding of the social relations that the concept purports to
analyse. Perhaps, for reasons other than those put forward by Arrow and Solow, social capital should
not be seen as a form of “capital” but as something else, broader, that could include not simply the
instrumental aspects of social relations but also the constitutive principles behind the existence of
these relations, such as the principle of justice. It must be noted that the idea of broadening the
normative instrumental views (approximated by utilitarian ethics) held by social scientists is at the
core of Amartya Sen’s approach centred on the notion of capabilities.

Sen’s capability approach (in Sen 1980, 1985, 1992, 1999) is a framework for evaluating and
assessing social arrangements. It is not meant to be a substantive theory of these issues. Its main
contribution consists in broadening the informational space (the information needed to make
evaluative judgements) of other ethical approaches by defining as units of evaluation not utilities or
monetary values but functionings (doings and beings) and capabilities (sets of functionings). The
ethics behind the capability approach requires the identification and weighting of valuables
functionings and capabilities. Many functionings are essential and important for a ‘good life’, whereas
others can be trivial and valueless. The elaboration of a broader informational space for evaluating
social arrangements is not the only important element in Sen’s normative approach. He argues that a
person’s freedom to live the way one would like has intrinsic value and therefore it is constitutive of a
person’s being. This means that not only achieved functionings are valuable but also the individuals’
capability of choosing and discriminating among possible livings. Emphasis on freedom or capability
reflects, according to Sen, the agency aspect of a person. Now, if Sen’s capability approach focuses
on broadening the informational space needed for properly assessing social arrangements, could it be
used to help Putnam’s social capital concept to overcome its limitations, namely, its narrow normative
foundation which emphasises the instrumental nature of social relations? Before discussing this
alternative, we examine an important limitation of the capability approach, which might affect the
way that these theories could be associated.

The normative individualistic ethics of the capability approach has been pointed up by its
sympathisers and critics as one of its main limitations. Carmen (2000: 1023) notes that “capabilities
represent a concept relatively closer to an individualistic mindset”. This means, as Cameron (2000:
1038) puts it, that “the ultimate analytical concern is with the vulnerable atomistic household”.. It has
also been observed by Gore (1997: 243) that “The informational basis of the capability approach thus
requires that judgements about the goodness of states of affairs are based exclusively on properties of
individuals”.. According to him, there are many different senses from which the capability approach
could be considered individualistic (1997: 243):

It does not see individuals atomistically, and it does not rely wholly on individual preferences
to judge states of affairs. But the goodness or badness of social arrangements or states of
affairs is evaluated on the basis of what is good or bad for individual well-being and freedom
and is also reduced to the good of those individuals.

But this should not come as a surprise. Sen, himself, acknowledges at the beginning of his book
Development as Freedom that (1999: 18) “The analysis of development presented in this book treats
the freedoms of individuals as the basic building blocks” [emphasis added] and later that (1999: 117)
“the information focus of this work has been on individual freedoms” [emphasis added]. Concern
with the individual is certainly important and this should not be under-emphasised. Yet, it seems that
a natural extension of Sen’s capability approach would involve a broadening of the informational
space by the inclusion of other valuable elements that transcend individual functionings and
capabilities. Indeed, Gasper (2000:997) suggests that “a broader development ethic would require
more adequate pictures of ‘culture’ and ‘the individual’” than those provided by the capability
approach. Similarly, Alkire and Black (1997: 276) also put forward that “the task is how to move
from principles derived for individual morality to procedures by which complex development
decisions are made”.

Robeyns (2000) suggests that the capability approach’s ethical individualism allows many different
roles for social and environmental factors and that (2000: 17) “a commitment to ethical individualism
is not incompatible with a personal ontology that recognises the connections between people, their
social relations, and their social embedment.” By doing so she denies a constitutive role to social
factors in the set of valuable capabilities. Social structures serve either as constraints or instruments to
the individuals’ evaluative space. They are exogenous to individuals. Indeed, she argues that (2000:
18) it is “impossible to speak of the capability of a community”. Functionings are therefore solipsistic
in the sense that there are no significant interrelations among individuals’ functionings that would
have value in itself. It seems that most social capital elements, such as trust, norms and cooperation –
that cannot be reduced to individuals’ properties- cannot be assessed from a capability approach
perspective. However, this dichotomy between the individual and the social is much less clear in
Sen’s own work. As Douglas and Ney noted (1998: 72) on the capability approach,

the individual is nominally to the fore and the rest of the society to the rear. But only
nominally, for the measures are designed to assess the institutional support for the individual.

It could be argued that there is a tension in Sen’s capability approach between its formalisation as an
ethical liberal theory and its use as a developmental normative framework. Sen makes wide use of
social features as important but instrumental elements6 of value judgements about individuals well-
being. However, it seems that Sen (1999) extends the scope of its evaluative framework to include
constitutive social features that are intrinsically valuable such as those of democracy, trust, “social
attitudes about sharing”, etc. For instance, when discussing the importance of democracy, he
attributes intrinsic value to democracy and political freedom. As he puts it (1999: 152), “We have
reason to value liberty and freedom of expression and action in our lives, and it is not unreasonable
for human beings –the social creatures that we are- to value unrestrained participation in political and
social activities”.. It might be difficult to characterise participation and democracy as properties of
individuals. Now, if they are intrinsically important –independently of their consequences on the
capabilities of individuals- they should be incorporated into the informational space of the capability
approach. In other words, system-level properties seem to matter to Sen from an evaluative
perspective.

Crucial development problems seem to depend on the assessment of unjust and unfair social
structures, such as the international system of trade and transfers, international regulations of property
rights, international financial markets, etc that end up benefiting the richer countries in international
commercial disputes, foreign debt payment arrangements, etc. Should these structures be assessed
only in terms of their impact on individuals’ capabilities? Are they not intrinsically good or bad, fair
or unfair in themselves? Shouldn’t structural, system-level properties, be assessed according to their
intrinsic properties?

It is not clear how the extension of Sen’s individual ethics to a social ethics would be compatible with
the idea of counterfactual choice present in the notion of freedom. What one would have chosen if
one had the choice might not be a social option. Be it as it may, problems of social injustice and
inequality might benefit from an explanation of social change in terms of co-realisable or system-level
capabilities that may be experienced as a result of the evolution of social structures.

Having discussed the main limitations of Putnam’s and Sen’s approaches, it seems that there are many
differences between them that could make harder any simultaneous reading of the concepts of social
capital and capability. They are:

1) Putnam and Sen seem to be working at two different levels of analysis. Sen’s
development ethics is based on moral considerations that follow a well-specified
philosophical discussion on equality, justice and social choice theory. Putnam’s analysis

6 Sen’s emphasis on the instrumental nature of social characteristics can be seen in his claims that (1999: 142)
“Individuals live and operate in a world of institutions. Our opportunities and prospects depend crucially on what
institutions exist and how they function. Not only do institutions contribute to our freedoms, their roles can be
sensibly evaluated in the light of their contributions to our freedom”..
 of institutional performance is shaped by empirical remarks that follow closely a debate
well contextualised within political science. It seems that Sen’s philosophical work and
Putnam’s econometric formulations lie at two separate ends of methodological and
substantive particular inquiries;

2) Putnam restricts his comments on economic factors, emphasising only growth and other
aggregative considerations (or ‘the efficiency aspect’) as the standard of economic
comparison. Sen is openly against this approach. He emphasises that economic growth is
merely a means and not an end in itself;

3) Putnam doesn’t seem to be interested in any concept resembling Sen’s functionings. For
instance, he argues that (1993: 65-6) “In the language of policy analysis, we want to
measure ‘outputs’ rather than ‘outcomes’ –health care rather than mortality;
environmental policy rather than air quality; economic development programs rather than
business profits”.. There is a clear emphasis on means here that is at odds with Sen’s
emphasis on the importance of ends;

4) Putnam’s concept of social capital is presented at a macro level while Sen’s notion of
capability is basically a micro foundation for his ethics;

5) Putnam carries out a historical analysis in order to characterise the concept of social
capital. Sen’s characterisation is logical rather than historical. Capabilities seem to be a-
historical concepts. Path-dependence does not seem to apply to Sen’s development ethics.

Let’s take stock. Social capital’s ethical foundations seem to be incomplete. Capability is an
individual attribute that might be little representative of ‘system-level’ phenomena of social
deprivations influenced by structural elements. Putnam and Sen address two quasi-hermetic academic
debates from different methodological perspectives and haven’t shown in their theories scope for any
exchange of fruitful ideas. So far, they seem quite incompatible bedfellows.

Is there any ‘social capital’ between Putnam and Sen?

Despite their differences, we cannot ignore that there are some areas of intersection between Putnam’s
and Sen’s approaches, such as their emphasis on non-economic variables as crucial elements in their
arguments and their attempts to integrate the demands of efficiency and equality in their claims (as
superficial as it might seem sometimes). But these areas of intersection do not appear to be very
significant. A more promising avenue of investigation consists in examining the possible
‘complementary’ between their approaches. Indeed, if the above characterisation of their approaches
is correct, there is not only room for some fruitful exchange between these two authors but also there
is some compelling reasons for a simultaneous reading of their theories.

As mentioned in the introduction, the first claim developed here, called the weak claim, suggests that
a ‘simultaneous’ reading of the concepts of social capital and capability is possible and might even be
illuminating. By ‘simultaneous’ reading we understand an attempt to integrate both concepts as part
of the same theoretical story. The second claim developed here, which is a second way of
understanding a ‘simultaneous’ or combined reading of the concept of social capital and capability,
can be called the strong claim because it states that this simultaneous reading is ‘needed’.. The
difference between the claims will depend on the cogency of the arguments that follow.

1. perhaps, after all, Putnam is not referring to ‘capital’ but to a sort of capability;

Although Putnam does not seem to be interested in functionings in his analysis, his definition of civic
community involves a description of ‘beings and doings’ of communities, that are, actually,
functionings. There is a tension here between Putnam’s measurement needs, which force him into
dismissing ‘outcomes’ to the benefit of ‘outputs’, and his conceptual basis, which relates institutional
performance on ‘civic engagement’, or the character of the citizens. The existence of social capital, as
discussed by Putnam, does not imply the ‘achievement’ of economic growth. Perhaps, it will be best
seen as a freedom that a certain polity or civic community has to achieve the real opportunity of
growth. Arrow’s and Solow’s objections could then be respected not only by dropping the expression
‘capital’ from social capital but by referring to the potential freedom or capabilities that may arise
from civic associations. Because social networks are formed by social reasons that are not directly
related to any economic benefit to their participants, no one can ‘invest’ in social capital as if one
were investing in ‘physical’ capital. More importantly, there is no reason why social capital
arrangements should be assessed only in terms of its consequences. The instrumental nature of social
capital is actually a limitation to this concept that could be overcome by the introduction of a wider
normative perspective given by the notion of capabilities. Indeed, social capital seems to say more
about the general capability that distinct civic communities have to achieve development (or growth,
if you might prefer) than about a stock of social networks that must deterministically produce
economic benefits to their participants.

In addition, Sen’s critique of the use of resources as the main determinant of capabilities could be
used to give a more dynamic character to the concept of social capital. If the wide diversity of human
beings prevents resources to be an accurate proxy for individuals’ capabilities, it could be argued that
social capital, understood as a ‘social resource’, must also be an imperfect proxy for individuals’
capabilities. However, if social capital could be understood as a capability, it would better be
described as a property of system-level behaviour (e.g. trust, cooperation) and would be assessed
according to its intrinsic valuable characteristics. Alternatively, Sen’s achievement-freedom
distinction might help us to understand the link between the stock of ‘social assets’ and the flow of
benefits that is at the core of the notion of social capital. How diversity affects communities’
conversion of resources into freedoms? How different 'social’ translate into ‘capital’? As observed by
Krishna (1999: 73),

This distinction between stock and flow, between the level of productivity of the resource, is
finessed in much of the available literature. In the conception of social capital made popular
by Putnam, Leonardi, and Nanetti (1993), Putnam (1995, 1996), and Fukuyama (1995), the
extent of flow is related directly to the level of the stock –that is, a high stock of social capital
invariably results in profuse flows of benefits, while low stocks always result in impoverished
flow. The connection between stock and flows is direct, proportionate and invariant.

The conversion problem, as discussed by Sen (1992, 1999), must address not only problems of
interpersonal or group variations within a civic community but also problems of social and
institutional nature. Social command over resources needs to be contextualised within the broader
frame of freedom, or capabilities that communities have in order to command these resources. Or,
conversely, the need for resources could be judged according to the communities’ capability to
function derivable from these resources.

2. Putnam’s ‘theory of associations’ could be nicely complemented by a theory of justice;

There is no reason why Putnam’s ‘morality of associations’ should not be complemented by a

‘morality of principles’.. If justice, as Rawls puts it (1971: 3, 586), “is the first virtue of social
institutions”, there is much to be learned from an investigation (even an empirical investigation) of
justice in the constitution of social capital. If we understand by justice not the formal set of already-
established laws and norms, but a set of moral principles that guide behaviour that people consider
‘just’, we might even say that justice is the main form of social capital in western contemporary
societies. The principles of justice, like those of social capital, are public; justice, like social capital,
presents increasing returns. Effective operation of laws reinforces the case for subsequent laws. And
as Rawls observes (1971: 498), “a more effective sense of justice leads to a more secure intention to
do one’s share, and the recognition of this fact arouses more intense feeling of friendship and trust”.

Empirical research might then be reoriented from a search for density of local, cultural and
recreational associations to a search for density of just arrangements vis-à-vis unjust arrangements.
Because compliance with law depends on people’s perfection of how just particular laws are, we
would unavoidably ended up discussing ‘moral sentiments’ of civic communities. There is no
incompatibility between justice as the public basis of social cooperation and Putnam’s approach.
Comparing the more with the less civic regions in Italy, Putnam comments that (1993: 111),

Honesty, trust, and law-abidingness are prominent in most philosophical accounts of civic
virtue. Citizens in the civic community, it is said, deal fairly with one another and expect fair
dealing in return. They expect their government to follow high standards, and they willingly
obey the rules that they have imposed on themselves. (…) In a less civic community, by
contrast, life is riskier, citizens are warier, and the laws, made by higher-ups, are made to be
broken.

An important point here is that Putnam’s reading of this correlation between civic behaviour and law-
abidingness seems to suggest that people’s morality of principles are somehow influenced by their
morality of associations, when, perhaps, if there are causal elements in these vicious circles they are
to be found in the sense of justice of citizens. However, collective action depends not merely on the
density (the quantity) of norms and networks of civic engagement, as Putnam seems to suggest, but
mainly on their justice (the quality).

However, as mentioned above, Putnam’s disregard for the public sense of justice as a basis for civic
friendship is just the tip of the iceberg. It is just a sign that his narrow normative approach,
exclusively focused on the criteria of responsiveness and efficiency, has provided some empirical
guidelines for investigation of social capital that ignore evaluation of social arrangements. Sen’s
emphasis on a broader evaluative framework seems to provide an important complement of Putnam’s
theory of associations.

3. perhaps, the best ‘use’ of Sen’s capability approach needs a taxonomy on ‘social’
capabilities;

If the capability approach’s informational basis could be extended to include value assessments of
social structures, then it seems that social capital theory could provide a framework for the
identification of social capabilities. There is a certain discontinuity between the formal and applied
aspects of the capability approach. There are no clear signs that the capability approach, defined as an
individualistic development ethics, can address social and structural aspects of development
problems. For instance, there is nothing structurally or historically minded in Sen’s (1999) contrast
between the African Americans living in Harlem and the people living in the Indian state of Kerala.
We would like to suggest here that many capabilities could be better understood as properties of the
communities of societies in which individuals live, such as the social attitudes concerning health care
arrangements, etc. For this reason, an attempt to understand social capabilities must incorporate
elements produced by system-level behaviour.
To emphasise: the informational basis of evaluative judgements of the capability approach could be
expanded to include information on the social structures that individuals live. They could be assessed
in terms of their intrinsic properties and not for their instrumental effects on individuals’ capabilities.
They would allow a direct assessment of social structures in terms of their quality.

Social capital literature can provide a framework for social capabilities. Can ‘horizontal’ relations of
reciprocity and cooperation within a community entitle people to capabilities that ‘vertical’ relations
of authority and dependency cannot? Putnam’s analysis of ‘internal effects’ of associations seems to
be useful here. They appear to shape individuals’ capabilities and their capacity of converting
resources into functionings. Though manifested at individual level, reciprocity and associationist
behaviour depends on the influence of variables that are defined at systemic level. Other important
variables, such as, norms, trust and reliability cannot be defined at an individual level. They are not
‘found’ in any of the participants of the civic community, but rather constitute a ‘third element’ that
results from the association of all parts.

The social nature of capabilities, although not formally included by Sen in his conceptual discussion,
appears quite often in his examples. When criticising Rawls’s characterisation of justice based on
toleration, Sen argues that to limit a political conception to toleration excludes many cases, such as
the case of the Ethiopian famines of 1973 and the Emperor Haile Selassie’s (and his opponents’)
disregard for any principle of toleration. He claims that (1992: 78),

each side pursued their own objectives with no quarter given to the objectives of the others,
and as far as one can judge, had no interest in looking for a political solution based on
toleration with the desire of living together.

Now, isn’t he saying that vertical relations of authority and dependency turned cooperation and
solidarity into scarce commodities in the Ethiopian society? Could this absence of social capital or
social capabilities be an attribute of the individuals of that society or of that society as a whole? As
much as we might agree with Sen that a political conception of justice must not be limited to
toleration, the point here is that the social determinants of the Ethiopian famines have proven to be
decisive. Similarly, when Sen explains the successful history of Kerala (1992: 128), he observes that

There are also other factors involved, including a more favourable position of women in
 property rights and in inheritance among a substantial and influential section of Kerala’s
population, and the greater public activism connected with educational campaign as well as
politics in general.

Again, references to social capabilities, in this case here, to property and inheritance rights and public
activism, are part of the explanatory setting of a specific situation. Thus, if people’s ability to convert
resources into functionings cannot be exclusively determined at individual bases, the concept of
capability would need to be extended to include social capabilities. There is no classification or
taxonomy of capabilities in Sen’s discussion. He only comments that capabilities can go from
elementary physical functionings to more complex social achievements. Based on Putnam’s and
Coleman’s investigation of social capital, social capabilities could be characterised by:
- social closure;
- horizontal relations;
- participation;
- evaluative principles, such as common good;
Other requisites, such as, sense of justice, could also be included. The important issue here is that
Sen’s individualist approach to capabilities, motivated by philosophical considerations, could benefit
from social considerations found in Putnam’s work. The perspective we have in mind here is the ‘use’
of Sen’s contribution to conceptual and empirical investigation of development problems, in
particular, the problem of poverty. Sen’s argument that (1992: 110-1) “Income adequacy to escape
poverty varies parametrically with personal characteristics and circumstances” can be extended
according to the notion of social capabilities. This might be relevant in cases where poverty seems to
be much more a problem of social characteristics than of individual circumstances.

The expression “social capabilities” has been used in a different context. For instance, Abramovitz
(1995: 28-29) defines social capabilities in terms of “a set of national characteristics”, such as those
of social attitudes, ability to exploit technology and the country’s legal regime. Stiglitz (1995: 48-50)
includes under this term the legal-regulatory environment in which firms operate and the public tax
and expenditure system. Other authors, such as Bebbington (1999) and Temple and Johnson (1998)
have also referred to this term. Yet, this use of social capabilities was not directly influenced by Sen’s
use of the term and the instrumental nature of these factors was not related to a normative assessment
of these social structures. Perhaps, the best approximation of our use of the expression social
capabilities is given by Charles Taylor’s concept of irreducibly social goods (in Gore, 1997: 243),
according to which “Such goods are objects of value which cannot be ‘decomposed into individual
occurrences’”, but refer to societies.

It could be argued that social capabilities are those capabilities that can only be achieved socially, that
represent those sets of beings and doings that can only be achieved as a result of social interaction.
Thus, social capabilities are those capabilities that cannot be reduced to properties of individuals;
rather they reflect properties of social structures and systemic level outcomes. They may represent
valuable opportunities that people can achieve as a result of their collective agency or valuable
freedoms that arise from their “social connectedness” and cooperation. Social capabilities are
properties of social structures and processes and their characteristics such as trustworthiness, justice,
reciprocity, etc.

Conclusion: social capability and compatible bedfellows

The core of the argument developed here is that both Sen’s and Putnam’s approaches are subject to
limitations that are to some extent complementary. Putnam’s concept of social capital may better refer
to ‘capability’ than to ‘capital’. Similarly, the sort of ‘moral sentiments’ that he suggests are behind
economic development can be better qualified by a morality of principles than by a morality of
associations. There is no incompatibility between both. Sen’s approach could help identifying social
interactions that have intrinsic rather than merely instrumental value. Sen’s individualistic
development ethics can be complemented by social considerations that can better characterise
capability deprivation along poverty lines. Putnam’s multilevel analysis could provide some social
structure to a capabilities assessment. Our weak claim is that, despite their differences, it is fruitful to
start thinking in terms of these two concepts as part of a unique idea; the idea of complex deprivation-
laden social structures. We are not denying here individuals’ idiosyncratic capabilities but suggesting
that much of the deprivation problems in the word could be better understood from a social capability
perspective. In addition to it, our strong claim is that Putnam’s analysis may not stand to a close
scrutiny of its ethical foundations and that Sen’s work may benefit from an expansion of its
information basis that would allow an examination of the social and structural elements as
constitutive elements of people’s capabilities.

The significance of this proposal can be assessed empirically. While most empirical analyses of social
capital, such as those of Temple (1998) and Knack and Keefer (1997), focus on variables that
measure aggregative concerns based on associationist features of civic societies, important
distributive considerations, based on justice and equality remain unexplored. Empirical remarks, such
as Temple’s (1998: 342) that “Overall, high income inequality, at least to the extent it is reflected in
low middle income shares, does not seem to have been a key source of Africa’s slow growth” could
be assessed from a different perspective. Paradoxically, he argues that “the extent of social capital
appears to be the most important” to define Africa’s growth pattern. He is implicitly assuming that
income inequality, and the consequences it might generate on people’s sense of equality and justice, is
not directly related to social capital. An alternative empirical way of proceeding would consist in
measuring the distributive aspects of social capabilities. Emphasis would be put to measurement of
social arrangements that sustain unjust economic structures that interact with other capabilities at a
more individual level. Different social capabilities would need to be distinguished according to
different contexts and a whole range of evaluative considerations would be needed to classify these
capabilities as intrinsically valuable or not. A proper understanding of system-level behaviour could
be achieved by establishing links between communities’ resources and their rates of conversion of
resources. Variables, such as corruption, could be endogeneised by seen them as a characteristic of
social capabilities that are influenced by institutional arrangements that influence justice and equality
in civic communities. We might find out, at the end, that Putnam and Sen are indeed compatible
bedfellows.

References

COLEMAN, J. (1988) “Social Capital in the Creation of Human Capital”, American Journal of
Sociology, Vol. 94, pp. S95-S120.

DOUGLAS, M. and NEY, S. (1998) Missing Persons: a critique of personhood in the social sciences.
New York: Russell Sage Foundation.

FOLEY, M. and EDWARDS, B. (1996) “The Paradox of Civil Society”, Journal of Democracy, 7(3):
39-52.

GASPER, D. (2000) “Development as Freedom: taking economics beyond commodities –the cautious
boldness of Amartya Sen”. Journal of International Development, 12, pp. 989-1001.

GORE, C. (1997) “Irreducibly Social Goods and the Informational Basis of Amartya Sen’s Capability
Approach”. Journal of International Development, 9(2): 235-250.

LEVI, M. (1996) “Social and Unsocial Capital: a review essay of Robert Putnam’s ‘Making
Democracy Work’”, Politics and Society, vol. 24, n. 1, pp. 45-55.
PORTES, A. (1998) “Social Capital: its origins and applications in contemporary sociology”, Annual
Review of Sociology, Vol. 24, pp. 1-24.

PUTNAM, R. (2000) Bowling Alone: the collapse and revival of American community. New York:
Simon & Schuster.

PUTNAM, R. (1993) Making Democracy Work: civic traditions in modern Italy. Princeton: Princeton
University Press.

ROBEYNS, I. (2000) “An Unworkable Idea or a Promising Alternative? Sen’s Capability Approach
Re-Examined” Discussion Paper Series, Centre for Economic Studies, DPS 00.30, University of
Leuven, mimeo.

SEN, A.K. (1999) Development as Freedom. Oxford: Oxford University Press.

SEN, A.K. (1998) “Mortality as an Indicator of Economic Success and Failure”. Economic Journal,
108: 447, pp. 1-25.

SEN, A.K. (1993) “Capability and Well-Being”. In: NUSSBAUM, M. and SEN, A.K. (1993) The
Quality of Life. Oxford: Oxford University Press.

SEN, A.K. (1992) Inequality Reexamined. Oxford: Oxford University Press.

SEN, A.K. (1987) “The Standard of Living”. In: Sen, A.K., MUELLBAUER, J., KANBUR, R.,
HART, K., and WILLIAMS, B. The Standard of Living: The Tanner Lectures on Human Values.
Cambridge: Cambridge University Press.

SEN, A.K. (1985) Commodities and Capabilities. Oxford: Oxford University Press.

SEN, A.K. (1980) “Equality of What”. In: SEN, A.K. (1982) Choice, Welfare and Measurement.
Cambridge: Harvard University Press.

SMITH, A. (1976 [1759]) The Theory of Moral Sentiments. Glasgow edition. Liberty Fund.

SZRETER, S. (2000) “Social Capital, the Economy and Education in Historical Perspective”. In:
BARON, S., FIELD, J. and SCHULLER, T. (2000) (eds.) Social Capital: Critical Perspectives.
Oxford: OUP.

WOOLCOCK, M. and NARAYAN, D. (2000) “Social Capital: Implications for Development Theory,
Research and Policy”. The World Bank Research Observer, 15(2).