Development as Capability Expansion PDF

DEVELOPMENT AS CAPABILITY EXPANSION Amartya Sen* CONTENTS...

DEVELOPMENT AS CAPABILITY EXPANSION Amartya Sen* CONTENTS Page Introduction........................................................................................................................41 The capability approach: conceptual roots.........................................................................43 Commodities, functionings and capability..........................................................................43 Utilitarian calculus versus objective deprivation................................................................44 Ambiguities, precision and relevance.................................................................................45 Quality of life, Basic needs and capability.........................................................................46 Rawls, primary goods and freedoms...................................................................................47 Freedom, capability and data limitations............................................................................48 Inequality, class and gender................................................................................................51 Conclusion..........................................................................................................................54 ________________ Introduction In his Grundlegung zur Metaphysik de Sitten, ImmanueI Kant argues for the necessity of seeing human beings as ends in themselves, rather than as means to other ends: “So act as to treat humanity, whether in thine own person or in that of any other, in every case as an end withal, never as means only.”1 This principle has importance in many contexts–even in analysing poverty, progress and planning. Human beings are the agents, beneficiaries and adjudicators of progress, but they also happen to be– directly or indirectly–the primary means of all production. This dual role of human beings provides a rich ground for confusion of ends and means in planning and policy-making. Indeed, it can–and frequently does–take the form of focusing on production and prosperity as the essence of progress, treating people as the means through which that productive progress is brought about (rather than seeing the lives of people as the ultimate concern and treating production and prosperity merely as means to those lives). Indeed, the widely prevalent concentration on the expansion of real income and on economic growth as the characteristics of successful development can be precisely an aspect of the mistake against which Kant had warned. This problem is particularly pivotal in the assessment and planning of economic development. The * Lamont University Professor, Harvard University. 41 problem does not, of course, lie in the fact that the pursuit of economic prosperity is typically taken to be a major goal of planning and policy-making. This need not be, in itself, unreasonable. The problem relates to the level at which this aim should be taken as a goal. Is it just an intermediate goal, the importance of which is contingent on what it ultimately contributes to human lives? Or is it the object of the entire exercise? It is in the acceptance–usually implicitly–of the latter view that the ends–means confusion becomes significant–indeed blatant. The problem might have been of no great practical interest if the achievement of economic prosperity were tightly linked–in something like a one-to-one correspondence–with that of enriching the lives of the people. If that were the case, then the pursuit of economic prosperity as an end in itself, while wrong in principle, might have been, in effect, indistinguishable from pursuing it only as a means to the end of enriching human lives. But that tight relation does not obtain. Countries with high GNP per capita can nevertheless have astonishingly low achievements in the quality of life, with the bulk of the population being subject to premature mortality, escapable morbidity, overwhelming illiteracy and so on. Just to illustrate an aspect of the problem, the GNP per capita of six countries is given in table 1, along with each country’s respective level of life expectancy at birth. TABLE 1. ECONOMIC PROSPERITY AND LIFE EXPECTANCY, 1985 Country GNP per capita Life expectancy at birth China..................................................... 310 69 Sri Lanka................................................ 380 70 Brazil..................................................... 1 640 65 South Africa............................................ 2 010 55 Mexico................................................... 2 080 67 Oman..................................................... 6 730 54 Source: World Development Report 1987 (New York, Oxford University Press, 1988), table 1. A country can be very rich in conventional economic terms (i.e., in terms of the value of commodities produced per capita) and still be very poor in the achieved quality of human fife. South Africa, with five or six times the GNP per capita of Sri Lanka or China, has a much lower longevity rate, and the same applies in different ways to Brazil, Mexico, Oman, and indeed to many other countries not included in this table. There are, therefore, really two distinct issues here. First, economic prosperity is no more than one of the means to enriching lives of people. It is a foundational conclusion to give it the status of an end. Secondly, even as a means, merely enhancing average economic opulence can be quite inefficient in the pursuit of the really valuable ends. In making sure that development planning and general policy- making do not suffer from costly confusions of ends and 42 means, we have to face the issue of identification of ends, in terms of which the effectiveness of the means can be systematically assessed. This paper is concerned with discussing the nature and implications of that general task. The capability approach: conceptual roots The particular line of reasoning that will be pursued here is based on evaluating social change in terms of the richness of human life resulting from it. But the quality of human life is itself a matter of great complexity. The approach that will be used here, which is sometimes called the “capability approach”, sees human life as a set of “doings and beings”–we may call them “functionings”–and it relates the evaluation of the quality of life to the assessment of the capability to function. It is an approach that I have tried to explore in some detail, both conceptually and in terms of its empirical implications.2 The roots of the approach go back at least to Adam Smith and Karl Marx, and indeed to Aristotle. In investigating the problem of “political distribution”, Aristotle made extensive use of his analysis of “the good of human beings”, and this he linked with his examination of “the functions of man” and his exploration of “life in the sense of activity”.3 The Aristotelian theory is, of course, highly ambitious and involves elements that go well beyond this particular issue (e.g., it takes a specific view of human nature and relates a notion of objective goodness to it). But the argument for seeing the quality of life in terms of valued activities and the capability to achieve these activities has much broader relevance and application. Among the classical political economists, both Adam Smith and Karl Marx explicitly discussed the importance of functionings and the capability to function as determinants of well-being.4 Marx’s approach to the question was closely related to the Aristotelian analysis (and indeed was apparently directly influenced by it).5 Indeed, an important part of Marx’s programme of reformulation of the foundations of political economy is clearly related to seeing the success of human life in terms of fulfilling the needed human activities. Marx put it thus: “It will be seen how in place of the wealth and poverty of political economy come the rich human being and rich human need. The rich human being is simultaneously the human being in need of a totality of human life-activities–the man in whom his own realization exists as an inner necessity, as need.”6 Commodities, functionings and capability If life is seen as a set of “doings and beings” that are valuable, the exercise of assessing the quality of life takes the form of 43 evaluating these functionings and the capability to function. This valuational exercise cannot be done by focusing simply on commodities or incomes that help those doings and beings, as in commodity-based accounting of the quality of life (involving a confusion of means and ends). “The life of money-making”, as Aristotle put it, “is one undertaken under compulsion, and wealth is evidently not the good we are seeking; for it is merely useful and for the sake of something else.”7 The task is that of evaluating the importance of the various functionings in human life, going beyond what Marx called, in a different but related context, “commodity fetishism”.8 The functionings themselves have to be examined, and the capability of the person to achieve them has to be appropriately valued. In the view that is being pursued here, the constituent elements of life are seen as a combination of various different functionings (a “functioning n-tuple”). This amounts to seeing a person in as it were, an “active” rather than a “passive” form (but neither the various states of being nor even the “doings” need necessarily be “athletic” ones). The included items may vary from such elementary functionings as escaping morbidity and mortality, being adequately nourished, undertaking usual movements etc., to many complex functionings such as achieving self–respect, taking part in the life of the community and appearing in public without shame (the last a functioning that was illuminatingly discussed by Adam Smith9 as an achievement that is valued in all societies, but the precise commodity requirement of which, he pointed out, varies from society to society). The claim is that the functionings are constitutive of a person’s being, and an evaluation of a person’s well-being has to take the form of an assessment of these constituent elements. The primitive notion in the approach is that of functionings–seen as constitutive elements of living. A functioning is an achievement of a person: what he or she manages to do or to be, and any such functioning reflects, as it were, a part of the state of that person. The capability of a person is a derived notion. It reflects the various combinations of functionings (doings and beings) he or she can achieve.10 It takes a certain view of living as a combination of various “doings and beings”. Capability reflects a person’s freedom to choose between different ways of living. The underlying motivation–the focusing on freedom–is well captured by Marx’s claim that what we need is “replacing the domination of circumstances and chance over individuals by the domination of individuals over chance and circumstances”.11 Utilitarian calculus versus objective deprivation The capability approach can he contrasted not merely with commodity-based systems of evaluation, but also with the utility-based assessment. The utilitarian notion of value, which is invoked 44  explicitly or by implication in much of welfare economics, sees value, ultimately, only in individual utility, which is defined in terms of some mental condition, such as pleasure, happiness, desire-fulfilment. This subjectivist perspective has been extensively used, but it can be very misleading, since it may fail to reflect a person’s real deprivation. A thoroughly deprived person leading a very reduced life, might not appear to be badly off in terms of the mental metric of utility, if the hardship is accepted with non- grumbling resignation. In situations of long-standing deprivation, the victims do not go on weeping all the time, and very often make great efforts to take pleasure in small mercies and to cut down personal desires to modest–”realistic”–proportions. The person’s deprivation, then, may not at all show up in the metrics of pleasure, desire- fulfilment etc., even though he or she may be quite unable to be adequately nourished, decently clothed, minimally educated and so on.12 This issue, apart from its foundational relevance, may have some immediate bearing on practical public policy. Smugness about continued deprivation and vulnerability is often made to look justified on grounds of lack of strong public demand and forcefully expressed desire for removing these impediments.13 Ambiguities, precision and relevance There are many ambiguities in the conceptual framework of the capability approach. Indeed, the nature of human life and the content of human freedom are themselves far from unproblematic concepts. It is not my purpose to brush these difficult questions under the carpet, In so far as there are genuine ambiguities in the underlying objects of value, these will be reflected in corresponding ambiguities in the characterization of capability. The need for this relates to a methodological point, which I have tried to defend elsewhere, that if an underlying idea has an essential ambiguity, a precise formulation of that idea must try to capture that ambiguity rather than attempt to lose it.14 Even when precisely capturing an ambiguity proves to be a difficult exercise, that is not an argument for forgetting the complex nature of the concept and seeking a spuriously narrow exactness. In social investigation and measurement, it is undoubtedly more important to be vaguely right than to be precisely wrong.15 It should be noted also that there is always an element of real choice in the description of functionings, since the format of “doings” and “beings” permits additional “achievements” to be defined and included. Frequently, the same doings and beings can be seen from different perspectives, with varying emphases. Also, some functionings may be easy to describe, but of no great interest in the relevant context (e.g., using a particular washing powder in doing the washing).16 There is no escape from the problem of evaluation in 45 selecting a class of functionings as important and others not so. The evaluative exercise cannot be fully addressed without explicitly facing questions concerning what are the valuable achievements and freedoms, and which are not. The chosen focus has to be related to the underlying social concerns and values, in terms of which some definable functionings and capabilities may be important and others quite trivial and negligible. The need for selection and discrimination is neither an embarrassment nor a unique difficulty for the conceptualization of functioning and capability.17 In the context of some types of welfare analysis, for example, in dealing with extreme poverty in developing economies, we may be able to go a long distance in terms of a relatively small number of centrally important functionings and the corresponding capabilities, such as the ability to be well-nourished and well-sheltered, the capability of escaping avoidable morbidity and premature mortality and so forth.18 In other contexts, including more general problems of assessing economic and social development, the list may have to be much longer and much more diverse.19 The task of specification must relate to the underlying motivation of the exercise as well as dealing with the social values involved. Quality of life, basic needs and capability There is an extensive literature in development economics concerned with valuing the quality of life, the fulfilment of Basic needs and related matters.20 That literature has been quite influential in recent years in drawing attention to neglected aspects of economic and social development. It is, however, fair to say that these writings have been typically comprehensively ignored in the theory of welfare economics, which has tended to treat these contributions as essentially ad hoc suggestions. This treatment is partly the result of the concern of welfare theory that proposals should not just appeal to intuitions but also be structured and founded. It also reflects the intellectual standing that such traditional approaches as utilitarian evaluation enjoy in welfare theory, and which serves as a barrier to accepting departures even when they seem attractive. The inability of utility-based evaluations to cope with persistent deprivations was discussed earlier, but in the welfare-economic literature the hold of this tradition has been hard to dislodge. The charge of “ad hoc-ness” against the development literature relates to the different modes of arguing that are used in welfare theory and in development theory. As far as the normative structure is concerned, the latter tends to be rather immediate, appealing to strong intuitions that seem obvious enough. Welfare theory, on the other hand, tends to take a more circuitous route, with greater elaboration and defence of the foundations of the approach in question. To bridge the gap, we have to compare and contrast the 46 foundational features underlying the concern with quality of life, Basic needs etc. with the informational foundations of the more traditional approaches used in welfare economics and moral philosophy, such as utilitarianism. It is precisely in this context that the advantages of the capability approach become perspicuous. The view of human life seen as a combination of various functionings and capabilities, and the analysis of human freedom as a central feature of living, provide a differently grounded foundational route to the evaluative exercise. This informational foundation contrasts with the evaluative bases incorporated in the more traditional foundations used in welfare economics.21 The “basic needs” literature has, in fact, tended to suffer a little from uncertainties about how basic needs should be specified. The original formulations often took the form of defining basic needs in terms of needs for certain minimal amounts of essential commodities such as food, clothing and shelter. If this type of formulation is used, then the literature remains imprisoned in the mould of commodity-centred evaluation, and tan in fact be accused of adopting a form of “commodity fetishism”. The objects of value tan scarcely be the holdings of commodities. Judged even as means, the usefulness of the commodity-perspective is severely compromised by the variability of the conversion of commodities into capabilities. For example, the requirement of food and of nutrients for the capability of being well-nourished may greatly vary from person to person depending on metabolic rates, body size, gender, pregnancy, age, climatic conditions, parasitic ailments and so on.22 The evaluation of commodity-holdings or of incomes (with which to purchase commodities) tan be at best a proxy for the things that really mutter, but unfortunately it does not seem to be a particularly good proxy in most cases.23 Rawls, primary goods and freedoms The concern with commodities and means of achievement, with which the motivation of the capability approach is being contrasted happens to be, in fact, influential in the literature of modern moral philosophy as well. For example, in John Rawls’ outstanding book on justice (arguably the most important contribution to moral philosophy in recent decades), the concentration is on the holdings of “primary goods” of different people in making interpersonal comparisons. His theory of justice, particularly the “difference principle” is dependent on this procedure for interpersonal comparisons. This procedure has the feature of being partly commodity-based, since the list of primary goods includes “income and wealth”, in addition to “the basic liberties”, “powers and prerogatives of offices and positions of responsibility”, “social bases of self-respect” and so on.24 Indeed, the entire list of “primary goods” of Rawls is concerned with means rather than ends; they deal with things that help to 47 achieve what we want to achieve, rather than either with achievement as such or even with the freedom to achieve. Being nourished is not a part of the fist, but having the income to buy food certainly is. Similarly, the social bases of self-respect belong to the list in a way self-respect as such does not. Rawls is much concerned that the fact that different people have different ends must not be lost in the evaluative process, and people should have the freedom to pursue their respective ends. This concern is indeed important, and the capability approach is also much involved with valuing freedom as such. In fact, it can be argued that the capability approach gives a better account of the freedoms actually enjoyed by different people than can be obtained from looking merely at the holdings of primary goods. Primary goods are means to freedoms, whereas capabilities are expressions of freedoms themselves. The motivations underlying the Rawlsian theory and the capability approach are similar, but the accountings are different. The problem with the Rawlsian accounting lies in the fact that, even for the same ends, people’s ability to convert primary goods into achievements differs, so that an interpersonal comparison based on the holdings of primary goods cannot, in general, also reflect the ranking of their respective real freedoms to pursue any given–or variable–ends. The variability in the conversion rates between persona for given ends is a problem that is embedded in the wider problem of variability of primary goods needed for different persona pursuing their respective ends.25 Hence, a similar criticism applies to Rawlsian accounting procedure as applies to parts of the basic needs literature for their concentration on means (such as commodities) as opposed to achievements or the freedom to achieve. Freedom, capability and data limitations The capability set represents a person’s freedom to achieve various functioning combinations. If freedom is intrinsically important, then the alternative combinations available for choice are all relevant for judging a person’s advantage, even though he or she will eventually choose only an alternative. In this view, the choice itself is a valuable feature of a person’s life. On the other hand, if freedom is seen as being only instrumentally important, then the interest in the capability set lies only in the fact that it offers the person opportunities to achieve various valuable states. Only the achieved states are in themselves valuable, not the opportunities, which are valued only as means to the end of reaching valuable states. The contrast between the intrinsic and the instrumental views of freedom is quite a deep one, and I have discussed the importance of the distinction elsewhere.26 Both views can be accommodated within 48 the capability approach. With the instrumental view, the capability set is valued only for the sake of the best alternative available for choice (or the actual alternative chosen). This way of evaluating a capability set by the value of one distinguished element in it can be called “elementary evaluation”.27 If, on the other hand, freedom is intrinsically valued, then elementary evaluation will be inadequate since the opportunity to choose other alternatives is of significance of its own. To bring out the distinction, it may be noted that if all alternatives other than the chosen alternative were to become unavailable, then there would be a real loss in the case of the intrinsic view, but not in the instrumental, since the alternative chosen is still available. In terms of practical application, the intrinsic view is much harder to reflect than the instrumental view, since our direct observations relate to what was chosen and achieved. The estimation of what could have been chosen is, by its very nature, more problematic (involving, in particular, assumptions about the constraints actually faced by the person). The limits of practical calculations are set by data restrictions, and this can be particularly hard on the representation of capability sets in full, as opposed to judging the capability sets by the observed functioning achievements. There is no real loss involved in using the capability approach in this reduced form if the instrumental view of freedom is taken, but there is loss if the intrinsic view is accepted. For the latter, a representation of the capability set as such is important. In fact, neither the instrumental view nor the intrinsic view is likely to be fully adequate. Certainly, freedom is a means to achievement, whether or not it is also intrinsically important, so that the instrumental view must be inter alia present in any use of the capability approach. Also, even if we find in general the instrumental view to be fairly adequate, there would clearly be cases in which it is extremely limited. For example, the person who fasts, that is, starves out of choice, can hardly be seen as being similarly deprived as a person who has no option but to starve because of penury. Even though their observed functionings may be the same, at least in the crude representation of functionings, their predicaments are not the same. In practice, even if in general the capability approach is used in the reduced form of concentrating on the chosen functioning combination, some systematic supplementation would be needed to take care of cases a which the freedom enjoyed is of clear and immediate interest. There may be no great difficulty in doing this supplementation in many cases, once the problem is posed clearly enough and the data search is made purposive and precise. Sometimes it would be useful to redefine the functionings in what is called a “refined” way, to take note of some of the obviously relevant alternatives that were available, but not chosen. Indeed, fasting is an 49 example of a “refined” functioning, and contrasts with the unrefined functioning of “starving”, which does not specify whether or not this was by choice.28 The important issue does not concern the existence or not of some actual word (such as fasting) that reflects the refined functioning (that is largely a matter of linguistic convention), but assessing whether or not such refining would be central to the exercise in question, and if central, deciding how this might be done. As a matter of fact, the informational base of functionings is still a much finer basis of evaluation of the quality of life and economic progress than various alternatives more commonly recommended, such as individual utilities or commodity holdings. The commodity fetishism of the former and the subjectivist metric of the latter make them deeply problematic. Thus, the concentration on achieved functionings has merits over the feasible rivals (even though it may not be based on as much information as would be needed to attach intrinsic importance to freedom). And in terms of data availability, keeping track of functionings (including vital ones such as being well- nourished and avoiding escapable morbidity or premature mortality) is typically no harder–often much easier–than getting data on commodity use (especially divisions within the family), not to mention utilities. The capability approach can, thus, be used at various levels of sophistication, and how far we can go would depend much on the practical consideration of what data we can get and what we cannot. In so far as freedom is seen to be intrinsically important, the observation of the chosen functioning bundle cannot be in itself an adequate guide for the evaluative exercise, even though the freedom to choose a better bundle rather than a worse one can be seen to be, in some accounting, an advantage even from the perspective of freedom.29 The point can be illustrated with a particular example. An expansion of longevity is seen, by common agreement, as an enhancement of the quality of life (though, strictly speaking, I suppose one can think of it as an enhancement of the quantity of life). This is so partly because living longer is an achievement that is valued. It is also partly because other achievements, such as avoiding morbidity, tend to go with longevity (and thus longevity serves also as a proxy for some achievements that too are intrinsically valued). But greater longevity can also be seen as an enhancement of the freedom to live long. We often take this for granted on the solid ground that given the option, people value living longer, and thus the observed achievement of living longer reflects a greater freedom than was enjoyed. The interpretative question arises at this precise point. Why is it evidence of greater freedom as such that a person ends up living longer rather than shorter? Why can it not be just a preferred achievement, but involving no difference in terms of freedom? One 50 answer is to say that one always does have the option of killing, oneself, and thus an expansion of longevity expands one’s options. But there is a further issue here. Consider a case in which, for some reason (either legal or psychological or whatever), one cannot really kill oneself (despite the presence in the world of poisons, knives, tall buildings and other useful objects). Would we then say that the person does not have more freedom by virtue of being free to live longer though not shorter? It can be argued that if the person values, prefers and wishes to choose living longer, then the change in question is in fact an expansion of the person’s freedom, since the evaluation of freedom cannot be dissociated from the assessment of the actual options in terms of the person’s evaluative judgments.30 The idea of Freedom takes us beyond achievements, but that does not entail that the assessment of freedom must be independent of that of achievements. The freedom to live the kind of life one would like to me has importance that the freedom to live the kind of life one would hate to have does not. Thus, the temptation to see more freedom in greater longevity is justifiable from several points of view, including noting the option of ending one’s life and being sensitive to the evaluative structure of achievements which directly affect the metric of freedom. The bottom line of all this is to recognize that the use of the capability approach even in the reduced form of concentrating on the achieved functionings (longevity, absence of morbidity, avoidance of undernourishment etc.) may give more role to the value of freedom than might have been initially apparent. Inequality, class and gender The choice of an approach to the evaluation of well-being and advantage has bearings on many exercises. These include the assessment of efficiency as well as inequality. Efficiency, as it is formally defined, is concerned with noting overall improvements, and in standard economic theory, this takes the form of checking whether someone’s position has improved without anyone’s position having gone down. A situation is efficient if and only if there is no alternative feasible situation in which someone’s position is better and no one’s worse. Obviously, the content of this criterion depends crucially on the way individual advantage is defined. If it is defined in terms of utility, then this criterion of efficiency immediately becomes that of “Pareto optimality” (or “Pareto efficiency”, as it is sometimes–more accurately– called), On the other hand, efficiency can be defined also in term of other metrics, including that of the quality of life based on the evaluation of functionings and capabilities. Similarly, the assessment of inequality too depends on the chosen indicator of individual advantage. The usual inequality measures that can be found in empirical economic literatures tend to 51 concentrate on inequalities of incomes or wealth.31 These are valuable contributions. On the other hand, in so far as income and wealth do not give adequate account of quality of life, there is a case for baling the evaluation of inequality on information more closely related to living standards. Indeed, the two informational bases are not alternatives. Inequality of wealth may tell us things about the generation and persistence of inequalities of other types, even when our ultimate concern may be with inequality of living standard and quality of life. Particularly in the context of the continuation and stubbornness of social divisions, information on inter-class inequalities in wealth and property ownership is especially crucial. But this recognition does not reduce the importance of bringing in indicators of quality of life to assess the actual inter-class inequalities of well-being and freedom. One field in which inequalities are particularly hard to assess is that of gender differences. There is a great deal of general evidence to indicate that women often have a much worse deal than men do, and that girls are often much more deprived than boys. These differences may be reflected in many subtle as well as crude ways, and in various forms they can be observed in different parts of the world–among both rich and poor countries. However, it is not easy to determine what is the best indicator of advantage in terms of which these gender inequalities are to be examined. There is, to be sure, no need to look for one specific metric only, and the need for plurality of indicators is as strong here as in any other field. But there is still an issue of the choice of approach to well-being and advantage in the assessment of inequalities between women and men. The approach of utility-based evaluation is particularly limiting in this context, since the unequal deals that obtain, particularly within the family, are often made “acceptable” by certain social notions of “normal” arrangements, and this may affect the perceptions of women as well as men of the comparative levels of well-being they respectively enjoy. For example, in the context of some developing countries such as India, the point has been made that rural women may have no clear perception of being deprived of things that men have, and may not be in fact any more unhappy than men are. This may or may not be the case, but even if it were so, it can be argued that the mental metric of utility may be particularly inappropriate for judging inequality in this context. The presence of objective deprivation in the form of greater undernourishment, more frequent morbidity, lower literacy etc. cannot be rendered irrelevant just by the quiet and ungrumbling acceptance of women of their deprived conditions.32 In rejecting utility-based evaluations, it may be tempting to go in the direction of actual commodities (enjoyed by women and men, respectively) to check inequalities between them. There is here the problem, already discussed earlier in this paper, that commodity- 52 based evaluations are inadequate because commodities are merely means to well- being and freedom and do not reflect the nature of the lives that the people involved can lead. But, in addition, there is the further problem that it is hard–sometimes impossible–to get information on how the commodities belonging to the family are divided between men and women, and between boys and girls. For example, studies on the division of food within the family tend to be deeply problematic since the observation needed to see who is eating how much is hard to carry out with any degree of reliability. On the other hand, it is possible to compare signs of undernourishment of boys and girls, to check their respective morbidity rates etc., and these functioning differences are both easier to observe and of greater intrinsic relevance.33 There are indeed inequalities between men and women in terms of functionings, and in the context of developing countries the contrast may be sharp even in basic matters of life and death, health and illness, education and illiteracy. For example, despite the fact that when men and women are treated reasonably equally in terms of food and health care (as they tend to be in the richer countries, even though gender biases may remain in other–less elementary–fields), women seem to have a greater ability to survive than men, in the bulk of the developing economies, men outnumber women by large margins. While the ratio of females to males in Europe and North America tends to be about 1.06 or so, that ratio is below 0.95 for the Middle East (including countries in Western Asia and North Africa), South Asia (including India, Pakistan and Bangladesh) and China.34 This crude figure of the ratio of survived females to survived males already tells a story that has much informational value in judging inter–gender inequalities. Sometimes there are sharp contrasts even within a country (e.g., the ratio of females to males varies within India all the way from 1.03 in Kerala to 0.87 or 0.88 in Haryana and Punjab). From the point of view of studying both the actual situations and the causal influences operating in the generation of inter–gender inequalities, these regional contrasts may be particularly important. Being able to survive is of course only one capability (though undoubtedly a very Basic one), and other comparisons can be made with information on health, morbidity etc. The ability to read and write is also another important capability, and here it can be seen that the ratio of female to male literacy rates is often shockingly low in different parts of the world. The combined effects of low literacy rates in general (a deprivation of a basic capability across genders) and gender inequalities in literacy rates (unequal deprivation of this basic capability for women) tend to be quite disastrous denials for women. It appears that even leaving out many countries for which no reliable data exist, in a great many countries in the world, the female literacy rate is still below 50 per cent. In fact, it is below even 30 per 53 cent for as many as 26 countries, below 20 percent for 16 and below 10 percent in at least five.35 In general, the perspective of functionings and capabilities provides a plausible approach to examining inter–gender inequalities. It does not suffer from the type of subjectivism that makes utility-based accounting particularly obtuse in dealing with entrenched inequalities. Nor does it suffer from the over-concentration on means that commodity-based accounting undoubtedly does, and in fact it has better informational sources in studying inequalities within the family than is provided by guesswork on commodity distributions (e.g., who is eating how much?). The case of inter-gender inequality is, of course, only one illustration of the advantages that the capability approach has. But it happens to be an illustration that is particularly important on its own as well, given the pervasive and stubborn nature of inequalities between women and men in different parts of the world. Conclusion The assessment of achievement and advantage of members of the society is a central part of development analysis. In this paper, I have tried to discuss how the capability approach may be used to substantiate the evaluative concerns of human development. The focus on human achievement and freedom, and on the need for reflective–rather than mechanical–evaluation, is an adaptation of an old tradition that can be fruitfully used in providing a conceptual basis for analysing the tasks of human development in the contemporary world. The foundational importance of human capabilities provides a firm basis for evaluating living standards and the quality of life, and also points to a general format in terms of which problems of efficiency and equality can both be discussed. The concentration on distinct capabilities entails, by its very nature, pluralist approach. Indeed, it points to the necessity of seeing development as a combination of distinct processes, rather than as the expansion of some apparently homogeneous magnitude such as real income or utility. The things that people value doing or being can be quite diverse, and the valuable capabilities vary from such elementary freedoms as being free from hunger and undernourishment to such complex abilities as achieving self-respect and social participation. The challenge of human development demands attention being paid to a variety of sectoral concerns and a combination of social and economic processes. In the collection of papers of which this one is a part, there are a number of specific studies dealing with such matters as education, health and nutrition, as well as the processes of agricultural expansion and industrial development. The problems of resource mobilization and participatory development are also addressed. Some of the 54 subjects thus covered deal with variables that are direct determinants of human capability (e.g., education and health), while others relate to instrumental influences that operate through economic or social processes (e.g., the promotion of agricultural and industrial productivity). The uniting feature is the motivating concern with human development and its constitutive characteristics. In the distinction between functionings and capabilities, emphasis was placed on the importance of having the freedom to choose one kind of life rather than another. This is an emphasis that distinguishes the capability approach from any accounting of only realized achievements. However, the ability to exercise freedom may, to a considerable extent, be directly dependent on the education we have received, and thus the development of the educational sector may have a foundational connection with the capability-based approach. In fact, educational expansion has a variety of roles that have to be carefully distinguished. First, more education can help productivity. Secondly, wide sharing of educational advancement can contribute to a better distribution of the aggregate national income among different people. Thirdly, being better educated can help in the conversion of incomes and resources into various functionings and ways of living. Last (and by no means the least), education also helps in the intelligent choice between different types of lives that a person can lead. All these distinct influences can have important bearings on the development of valuable capabilities and thus on the process of human development. There are also other interconnections between the different areas covered in the collection; for example, good health is an achievement in itself and also contributes both to higher productivity and to an enhanced ability to convert incomes and resources into good living. In focusing on human capabilities as the yardstick in terms of which successes and failures of human development are to be judged, attention is particularly invited to addressing these social interconnections. Given clarity regarding the ends (avoiding, in particular, the pitfall of treating human beings as means), the social and economic instrumentalities involved in the ends-means relations can be extensively explored. One of the most important tasks of an evaluative system is to do justice to our deeply held human values. The challenge of “human development in the 1980s and beyond” cannot be fully grasped without consciously facing this issue and paying deliberate attention to the enhancement of those freedoms and capabilities that matter most in the lives that we can lead. To broaden the limited lives into which the majority of human beings are willy-nilly imprisoned by force of circumstances is the major challenge of human development in the contemporary world. Informed and intelligent evaluation both of the lives we are forced to lead and of the lives we would be able to 55 choose to lead through bringing about social changes is the first step in confronting that challenge. It is a task that we must face. ________________ NOTES 1 Grundlegung (1785), sect. II; English translation Fundamental Principles of the Metaphysics of Morals, in Kant’s Critique of Practical Reason and Other Works on the Theory of Ethics, 6th edition, T. K. Abbot, ed. (London, Longmans, 1909), p. 47. 2 Amartya Sen, “Equality of what?”, in Tanner Lectures on Human Values. S. M. McMurring, ed., vol. I (Cambridge, Cambridge University Press, 1980 reprinted in Choice, Welfare and Measurement (Oxford, Blackwell; and Cambridge, Massachusetts, MIT Press, 1982)); Resources, Values and Development (Oxford, Blackwell; and Cambridge, Massachusetts Harvard University Press, 1984); Commodities and Capabilities (Amsterdam, North-Holland, 1985); “Well-being, agency and freedom: the Dewey lectures 1984”, Journal of Philosophy, 82 (April 1985); and “Capability and well-being”, WIDER conference paper, 1988. 3 Aristotle, The Nicomachean Ethics, book I, sect. 7; in the translation by David Ross, World’s Classics (Oxford, Oxford University Press, 1980), pp. 12-14. Note that Aristotle’s term “eudaimonia”, which is often misleadingly translated simply as “happiness”, stands for fulfilment of life in a way that goes well beyond the utilitarian perspective. Though pleasure may well result from fulfilment, that is seen as a consequence rather than the cause of valuing that fulfilment. For an examination of the Aristotelian approach and its relation to recent works on functionings and capabilities, see Martha Nussbaurn “Nature, function and capability: Aristotle on political distribution”, Oxford Studies in Ancient Greek Philosophy, supplementary volume 1988. 4 See Adam Smith An Inquiry into the Nature and Causes of the Wealth of Nations (1776), vol. I, book V sect. II; republished, R. H Campbell and A. S. Skinner, eds. (Oxford, Clarendon Press 1976), pp. 869-872; and Karl Marx, Economic and Philosophic Manuscripts of 1844 (1844); English translation (Moscow Progressive Publishers, 1977). 5 See G. E. M. de Sainte Croix, The Class Struggle in the Ancient Greek World (London, Duckworth, 1981); and Martha Nussbaum, “Nature, function and capability...” 6 Karl Marx, Economic and Philosophic Manuscripts of 1844... 7 Aristotle, op. cit., book I, sect. 5; in the translation by David Ross, p. 7. 8 Karl Marx, Capital, vol. I, English translation by S. Moore and E. Aveling (London, Sonnenschein 1887), chap. 1, sect. 4, pp. 41-55; see also Karl Marx, Economic and Philosophic Manuscripts of 1844... 9 See Adam Smith op. cit., vol. II, book V, chap. II (section entitled “Taxes upon Consumable Commodities”); republished... , pp. 469-471. 10 There are several technical problems in the representation of functioning n-tuples and of capability as a set of alternative functioning n-tuptes, any one n-tuple of which a person can choose. In this paper, I shall not be particularly concerned with these formal matters, for which see Commodities and Capabilities... , especially chaps. 2, 4 and 7. 11 Karl Marx and Friedrich Engels, The German Ideology (1846). The quoted passage is taken from the translation by David McLellan, Karl Marx: Selected Writings (Oxford, Oxford University Press, 1977), p. 190. 12 See Amartya Sen, “Well-being, agency and freedom...”; and Commodities and Capabilities... 13 It is sometimes presumed that to depart from a person’s own actual desires or pleasures as the measuring rod of assessment would be to introduce paternalism into the evaluative exercise. This view overlooks the important fact that having pleasure and desiring are not themselves valuational activities, even though the latter (desire) can often result from valuing something, and the former (pleasure) can often result 56 from getting what one values. A person’s utility must not be confused with his or her own valuations, and thus tying the evaluative exercise to the person’s own utility is quite different from judging a person’s success in terms of the person’s own valuation. The important distinction to note in this context is that a person may not have the courage to desire a big social change weighed down by the circumstances in which he or she lives, and yet given the opportunity to evaluate the situation, which is essentially a political exercise in this context, the person may well value a change. One advantage of valuing as opposed to feeling is that a proper evaluation has to be a reflective exercise–open to critical examination–in a way that feelings need not be (the requirement of critical examination does not apply in the same way to feelings as it does to reflective evaluations). These and related issues are discussed in “Well-being, agency and freedom...” 14 In many contexts, the formal representations will take the form of partial orderings, or of overdetermined rankings, or of “fuzzy” relations. This is, of course, not a special problem with the capability approach, and applies generally to conceptual frameworks in social theory; see Amartya Sen, Collective Choice and Social Welfare (San Francisco, Holden-Day, 1970 republished, Amsterdam, North-Holland, 1979); and On Ethics and Economics (Oxford, Blackwell, 1987); see also “Social choice theory”, in Handbook of Mathematical Economics, K. J. Arrow and M. Intriligator, eds. (Amsterdam, North-Holland, 1985). The formal problems can be dealt with at different levels of precision (i.e., with varying extent of precise representation of ambiguities). The important general point to note here is that it may be, for substantive social theories, both terribly limiting and altogether unnecessary to shun ambiguities. 15 See Amartya Sen, Choice, Welfare and Measurement... , assays 17-20. 16 Bernard Williams raises this issue in his comments on the Tanner Lectures on the standard of living; see The Standard of Living, Tanner Lectures of Amartya Sen, with discussions by John Muellbauer, Ravi Kanbur, Keith Hart and Bemard Williams, edited by Geoffrey Hawthorn (Cambridge, Cambridge University Press, 1987), pp. 98-101 and 108-109. 17 I have tried to discuss some of the general methodological issues involved in description in “Description as choice”, Oxford Economic Press, 32 (1980); reprinted in Choice, Welfare and Measurement... 18 See Amartya Sen, Resources, Values and Development... , chaps. 15, 19 and 20; and “The concept of development”, in Handbook of Development Economics, H. Chenery and T. N. Srinivasan, eds. (Amsterdam, North-Holland, forthcoming). 19 The range of functionings and capabilities that may be of interest for the assessment of a person’s well- being or agency can be very wide indeed; see Amartya Sen, “Well-being, agency and freedom... 20 See, among other contributions, Michael Lepton, Assessing Economic Performance (London, Staples Press, 1968); Paul Streeten, The Frontiers of Development Studies (London, Macmillan, 1972); Irma Adelman and Cynthia Tuft Morris, Economic Growth and Social Equity in Developing Countries (Stanford, Stanford University Press, 1973); Amartya Sen, “On the development of Basic income indicators to supplement GNP measures”, Economic Bulletin for Asia and the Far East (United Nations publication, Sates No. E.74.II.F.4); H. Chenery and others, Redistribution with Growth (London, Oxford University Press, 1974); Irma Adelman, “Development economics: a reassessment of goals”, American Economic Review, Papers and Proceedings, 66 (1975); James P. Grant, Disparity Reduction Rates In Social Indicators (Washington, D.C., Overseas Development Council, 1978); Keith Griffin and Azizur Rahman Khan, “Poverty in the third world; ugly facts and fancy models”, World Development, 6 (1978); Paul Streeten and S. J. Burki, “Basic needs: some issues”, World Development, 6 (1978); Morris D. Morris, Measuring the Conditions of the World’s Poor: The Physical Quality of Life Index (Oxford, Pergamon, 1979); Paul Streeten, Development Perspectives (London, Macmillan, 1981); Paul Streeten and others, First Things First: Meeting Basic Needs in Developing Countries (New York, Oxford University Press, 1981); S. R. Osmani, Economic Inequality and 57 Group Welfare (Oxford, Clarendon Press, 1982); and Frances Stewart, Planning to Meet Basic Needs (London, Macmillan, 1985). 21 This general question of foundations and informational bases is discussed in Amartya Sen, “Informational analysis of moral principles”, in Rational Action, Ross Harrison, ed. (Cambridge, Cambridge University Press, 1979); and “Well-being, agency and freedom...” In the latter analysis, some distinctions are drawn (especially between agency and well-being and between achievement and freedom) that may be worth pursuing in a more elaborate treatment of this matter, but I shall resist the temptation to go into these issues here. 22 On this general question and on the relation between commodities, characteristics and functionings, see Amartya Sen, Commodities and Capabilities... , chap. 2. 23 On this question, see Amartya Sen, Resources, Values and Development... , essays 19 and 20; and Paul Streeten, “Basic needs: some unsettled questions”, World Development, 12 (1984). 24 John Rawls,.9 Theory of Justice (Oxford, Clarendon Press; and Cambridge, Massachusetts, Harvard University Press, 1971), pp. 60-65. 25 See Amartya Sen, “Equality of what?”.... , and Resources, Values and Development... 26 See Amartya Sen, “Freedom of choice: concept and content”, Alfred Marshall Lecture at the European Economic Association, European Economic Review, 1988. 27 See Amartya Sen, Commodities and Capabilities... , pp. 60-67. 28 See Amartya Sen, “Well-being, agency and freedom...”; and “Freedom of choice: concept and content...” 29 On the question of the relation between achieved states and the extent of freedom and liberty, see Amartya Sen, “Liberty and social choice”, Journal of Philosophy, 80 (1983). 30 Indeed, not to take note of the person’s own evaluations of states of affairs in providing a measure of freedom can yield a very peculiar view of freedom, which would be seriously at odds with the tradition of seeing freedom as important. On this, see Amartya Sen, “Liberty as control: an appraisal”, Midwest Studies in Philosophy, 7 (1982); and “Liberty and social choice...” 31 See, for example, A. B, Atkinson, Unequal Shares: Wealth in Britain (London, Penguin, 1972); and The Economics of Inequality (Oxford, Clarendon Press, 1975). 32 I have discussed this question in Commodities and Capabilities... , appendix B, and also in Resources, Values and Development... , essays 15 and 16. The importance of perception biases in the continuation of inter-gender inequalities is discussed in “Gender and cooperative conflicts”, WIDER working paper, in Persistent Inequalities, Irene Tinker, ed. (forthcoming). 33 For an attempt to make such functioning-based comparisons between men and women, see Jocelyn Kynch and Amartya Sen, “Indian women: well-being and survival”, Cambridge Journal of Economics, 7 (l983). 34 See Jocelyn Kynch, “How many women are enough: sex ratios and the right to life”, Third World Affairs 1985 (London, Third World Foundation for Social and Economic Studies, 1985). The ratios of life expectancy seem to have turned in favour of women vis-à-vis men, according to reported statistics in most countries (see United Nations Children’s Fund, The State of the World’s Children 1988 (New York, Oxford University Press, 1988), table 7), but the undoing of past biases against women in the sex composition of the population tends to be a stow process over the years. 35 United Nations Children’s Fund, The State of the World’s Children 1988... , table 4. 58