IN recent years, the Indian city has featured in a wide range of debates and polemics. Whether we are looking at the Indian city historically as a crucible for the emergence of radical publics,¹ as a site of promises, fantasies and fears,² or in terms of changes in the access to spaces in the city with liberalization of the Indian economy,³ it remains both a context and an object of curiosity. The Indian city also features in other discussions and debates, especially on class and mobility. As part of the scholarship that is focused on social stratification and inequality as its primary problematic, there have been a range of studies that have looked at cities as sites of inequality – socio-economic, political and spatial.⁴ On the other hand, the city space can also be visualized as a liberating site for articulating new goals and realizing aspirations.

 

This contribution will approach the debate on inequality⁵ through an engagement with class and social mobility (or immobility). Drawing on the results of a survey of mobility with approximately 600 respondents in Delhi, this article is a preliminary attempt to provide a portrait of Delhi which places objective markers in a dialogue with subjective approaches to categories of class and social mobility. To do this, it highlights the broad macro patterns of social mobility experienced by those in the city, and their divergent experiences and conceptualizations of class and mobility. It begins by tracing the patterns of social mobility over one generation (as indexed by the movement between parents’ and children’s occupational class) in Delhi, disaggregated by gender and length of stay in the city.⁶

The paper then brings out the subjective dimensions of class, i.e. how people articulate their class position and whether they believe that this position has changed over time. Finally, it relates the ‘objective’ occupational measure and the ‘subjective’ articulation of class with opinions on their lives in the city and their role in civic participation. The assumption being that those who are situated at lower or more precarious positions in terms of either objective or subjective class would be less likely to see themselves as having a role to play in the decisions affecting the city.

 

Patterns of intergenerational social mobility (i.e. the movement between socio-economic strata) are an indicator of the ‘openness’ of a society.⁷ The study of mobility is significant as social mobility provides avenues for the redistribution of social and economic opportunities and rewards.⁸ Conversely, the lack of mobility can lead to the reproduction of inequalities over generations.⁹ The implications of the lack of mobility, or indeed the unevenness of mobility opportunities, on persisting social inequalities are hence manifold.

From the time that Pitrim Sorokin¹⁰ wrote the first treatise on social mobility, the area of mobility studies has grown substantially.¹¹ This has also led, at times, to heated debates on the most appropriate way to measure and capture mobility. The debate has been dominated by two key approaches to the study of mobility. First, the status attainment approach¹² which traces the impact of both individuals own as well as their parental socio-economic characteristics on final occupational achievement using path models. And, second, the class mobility approach¹³ that traces the change in class over one or more generations using mobility tables.¹⁴

 

Underlying the debate has been the tension about being able to define and empirically capture ‘status’ and ‘class’. The multiplicity of debates and perspectives on class have led some to claim that ‘there are almost as many class schemas as there are sociologists.’¹⁵ Further, what may or may not constitute class,¹⁶ and how to measure class;¹⁷ and more recently, Savage et al’s attempt to include ‘capital’ (Bourdieu) as a way of capturing class differences, are some of the issues that have been part of the research agenda.

To examine the various constitutive elements of social mobility, especially the intergenerational change in social position, requires the establishment of an appropriate schema that captures social class position across generations. If class is increasingly forming the basis of much of the debate on changing lifestyles and deepening inequalities in India, both academically and in the popular press, then it follows that we need a comprehensive definition of class – one that goes beyond the economic dimension alone. This paper does not attempt to engage with all the debates or the conceptual and operational differences between class and status; rather, it provides an avenue to think about the differences between how one could be objectively ‘classed’ on the basis of occupation¹⁸ and how one subjectively sees their class position, or indeed how class is articulated by respondents in Delhi.

 

A similar engagement is required with the concept of social mobility. Social mobility can include both absolute mobility, i.e., raw numbers of people who move up, down or remain stable in particular strata or classes, and relative mobility, i.e., mobility controlling for structural change that occurs as a matter of course due to economic restructuring. National studies have commented on the slow pace of structural change in India.¹⁹ This lack of rapid structural change has led to a distinctive pattern of intergenerational stability in India over a period of time. Further, there is an intriguing pattern for women’s mobility at the national level, showing that contrary to research on the West, women in India display more stability than men intergenerationally.²⁰ In terms of relative mobility, we find low levels of social fluidity which provide an indicator of the unequal opportunity structures in Indian society. It will be interesting to see how these issues play out in a metropolitan city like Delhi.²¹

 

This contribution draws on the Education and Social Mobility Survey (ESMS 2012), a two-city study on social mobility patterns, mechanisms and aspirations. The survey included the cities of Delhi (n=598) and Patna (n=1118). This paper is based on the Delhi sample. The sample for the ESMS was drawn using Multistage Stratified Random Sampling²² with an achieved sample size of 598 respondents in Delhi (49% men and 51% women). 44% of the sample was born in Delhi, while 64% of the sample stated that they have lived in Delhi their entire life. Of those who have lived in Delhi their entire life, 40% stated that Delhi was also their ancestral home. Of those who have not lived in Delhi their entire life, 46% came to Delhi from a village, 26% from a town and 28% from another city. In terms of states, for those who haven’t lived in Delhi their whole life, when asked which state they came to Delhi from – 29% came from UP, 15% from Bihar, 7% each from Haryana, Rajasthan and Uttarakhand and 6% from Punjab.

‘Objective’ social class in this paper is derived from information on occupation – rather than a detailed occupational schema, the schema in this paper delineates five major ‘class’ groups – from the professionals at the top of the schema to the agrarian class at the bottom (there are of course variations among these larger groups, but for a macro study these are useful to capture a broad measure of class).²³ ‘Subjective’ class is captured by a survey question that asked respondents to place their household (in comparison to other households) on a ladder with the lower class at the bottom and the upper class at the top. This was followed by a question on why they believe that they belong on a particular step of this class ladder.²⁴

 

As mentioned previously, patterns of social mobility are a useful measure to capture intergenerational change and persisting inequalities. Before describing these patterns, it is useful to look at change in the broader occupational structure in Delhi.²⁵

Table I summarizes this change over one generation. It is clear that as compared to fathers’ class, there has been an increase in professional occupations (an increase of roughly 4 percentage points), an increase in the service class (of 12 percentage points), in the business class (a marginal increase of 2 percentage points) and an increase in manual work (6 percentage points). As expected, agriculture shows the largest decline over one generation, as it is fairly non-existent among the current generation in Delhi (this is not surprising as some of the current residents of Delhi grew up outside Delhi, or their father’s had worked outside Delhi. The table further disaggregates these results by gender and by length of stay in Delhi. We find that women in our survey are much more likely than their fathers (and as compared to men) to be in the professional and service classes, and less likely to be in the business and agricultural class (these latter positions are usually inherited by men).

 

In terms of length of stay in Delhi, those who have not lived in Delhi their entire lives are seen to have two trajectories. First, they are more likely to not only be in professional or service jobs themselves (as compared to those who have lived in Delhi their whole lives), but they are also more likely to come from professional and service class origins and less likely to have fathers in manual work. Second, they are also more likely to come from agricultural origins than those who have been in Delhi their whole lives. These patterns seem to suggest not only that the occupational structure has changed over one generation,²⁶ but also that Delhi seems to provide important opportunities for mobility. Further, it underlines the need to capture occupational changes separately by factors such as gender, migrant status, caste and so on.²⁷

 

What does this structural change tell us about patterns of social mobility? Have all those in Delhi experienced similar opportunities of upward mobility? To answer this question, we now turn to patterns of absolute rates of intergenerational occupational mobility (Table II), i.e. the percentages who are mobile (upwardly or downwardly) in the city. We find much higher rates of social mobility in Delhi than authors have found for India overall.²⁸ However, these mobility patterns vary by gender, and by length of stay (in a separate analysis the author finds a caste based difference to these patterns as well).

 

While 54% of the Delhi sample has been intergenerationally mobile (national figures of mobility are around 30%), women rather than men have experienced more overall intergenerational mobility as well as more upward mobility (the net of upward over downward mobility is indicated in the last column of Table II). This is in contrast to national figures, and is similar in some ways to patterns in developed countries (it could perhaps be argued that this is due to the unique occupational structure in a city, or due to the lack of inheritance of agrarian or business occupations by women). However, this needs to be analyzed further as the sample of working women in the survey is fairly small. Further, those who have lived in Delhi their whole life display less mobility than those who have come to Delhi from outside (there is more net upward over downward mobility for the latter category). This latter seems to indicate the opportunities that movement into Delhi provides – which are in line with the results from Table I.

This final point needs to be read with two caveats. First, the category of those who have not lived in Delhi their entire life includes people who have spent varying amounts of time in Delhi and averages the results. Second, the occupational class schema is treated as hierarchical for the purpose of calculating mobility rates, with agriculture at the bottom. This implies that all those who come from agrarian origins would automatically, given that there are few agrarian jobs in the city, experience upward mobility if they are themselves not engaged in agriculture. This could be problematic given that certain forms of manual work themselves may not be considered to be a ‘step-up’ from agrarian jobs in terms of status or life chances. These caveats are applicable to most studies on social mobility.

 

Given these caveats, and as occupation may not capture people’s own articulation of their class position, we now turn to look at subjective expressions of class and mobility in Delhi. Table III traces patterns of social mobility according to subjective class position. Early on in the survey respondents were asked to situate their household in the context of other house-holds when they were growing up (roughly when they were aged 14) and place their household on a ladder with lower class at the bottom rung, to upper class at the top. Later in the survey they were asked to repeat this exercise with regard to the household they currently live in. Any difference between these two positions is treated as a form of subjective social mobility (i.e. an improvement or deterioration in their subjective class position over time).

We find (in Table III), that in comparison to occupational mobility (Table II) there is much less subjective mobility over time. Women in particular are more likely to be stable in terms of subjective class than they are in terms of their own occupations, which underlines the need to specifically study the experiences of women.³⁰ However, similar to Table II, those who have not lived in Delhi their entire lives are more likely to see themselves as mobile (though not necessarily as upwardly mobile as indicated by the last column of the Table) than those who have lived in Delhi their whole lives. This further suggests that subjective position itself may be mediated by how long one has been in one location, and where one comes from.

These patterns of occupational change and mobility beg the question of whether there is a relationship, and of what type, between objective occupational class and subjective class location. That is, how closely are these associated?³¹ We find that there is a weak correlation (p=0.2³²) between the two class positions. Further, on disaggregation we find that those who have professional occupations are more likely to state that their subjective class is upper-middle class (47%) or middle class (44%), while those with white collar service class jobs are more likely to state that they are middle (48%) or lower middle class (22%). Interestingly, the manual work category is more complicated as they are not only most likely to state that they are lower class or middle class (31% each), they are also surprisingly most likely to state that they are upper class (16%). This further highlights the need to delve deeper into a popular understanding of ‘class’.³³

 

To do this, we turn now to an open-ended question in the survey asking respondents to give reasons for why they believe they belong to a particular class.³⁴ The responses were complex and included financial reasons (including income, financial status – arthik sthiti; lack of money – paisa nahin hai), standard of living or lifestyle (rahan sahan), lack of adequate amenities (suvidhayen nahin hain), type of house/dwelling (jhopri me rahte hain). However, class was not simply seen as an economic position – it is seen as tied to other markers of status – for instance, caste was articulated in some instances to be an indicator of class. This could be seen when a respondent claimed a higher class position as their grandfather had been a priest (humare dadaji panda the), while another one claimed a lower class position as they belonged to a lower jati.

 

Further, class is also seen by some as relational. For instance, some articulated their class position in reference to others, i.e. in comparison to those better or worse off than them. Interestingly, even those who claimed to not believe in ‘class’ still situated themselves at a particular step in the class ladder based on how they believe others see them. These various articulations of class further complicate the discussions on operationalizations of class in the city.³⁵

In recent years there has been heated discussion in Delhi with regard to a sense of entitlements and participation in local politics (re: rise of the Aam Aadmi Party is a case in point). The question is whether these sentiments are refracted by class (and other categories). Looking at responses to one question in the survey,³⁶ we find that those who have professional and business occupations, as well as those who have higher subjective class locations, are more likely to agree that they can influence decision making in Delhi. In turn, those at the lower end of the class strata feel that their decisions have little impact on city life. This can help us look at communities that occupy the margins in the city in objective and subjective terms as well as in relation to citizenship rights. Critically, we find that men are more likely than women to believe that they have a role in decision making. In terms of caste we find that the general castes rather than the reserved castes are more likely to have a sense that their opinions make a difference in the life of the city.³⁷

 

The ESMS survey also has a variety of questions on civic rights such as the UID card, ration cards, passport and so on; as well as questions on access to basic facilities such as toilets in the home. Preliminary results seem to display a divergence in these rights and amenities according to gender, class and caste. Further, questions such as ‘what do you think you need to succeed in life’ lead to interesting responses and patterns regarding the role of chance/luck , wealth, effort and education.

 

While this note raises more questions than answers, it does highlight the complications inherent in a study of social class and the need to refract the positions and experiences of those who live in cities such as Delhi with their class, status, caste, locality, identity, income and so on. Further, there is a need to study the intersections between these categories – one is not simply a woman, but a woman from a particular background, with particular education and skills and so on. This would provide a more contextualized view of processes of subjective and objective class identity and social mobility.

Due to space limitations, this article only touches on some of the contours of the debate on social mobility (and the opportunities of movement) and aims really to spark conversations on how to conceptualize and operationalize (measure) class in the city. Further, it highlights the divergences in experiences and expectations of those who may be considered to be at the margins from those who are in relatively more protected positions. We also perhaps need a parallel conversation about the non-urban centres, and the connections between the urban and the non-urban in terms of economic links and migration patterns.³⁸

 

* This survey was funded by a grant from the Indian Council for Social Science Research (ICSSR). I thank the ICSSR for their financial support. I am grateful to Lokniti and the Centre for the Study of Developing Societies, Delhi, where this survey was based, for their support. I thank Anthony Heath, Sanjay Kumar, Suhas Palshikar, Yogendra Yadav, Himanshu Bhattacharya, Kinjal Sampat and Shreyas Sardesai for their advice on the questionnaire and plan of the study. I thank Tania Kahlon for research assistance, Shreya Sarawgi and Maria Mathew for their contribution to the survey, sampling design and data collection, Dhananjay Kumar Singh and Nilanjana Goswami for help with the data entry and data cleaning respectively. I am grateful to the field investigators involved in the survey. Finally, I thank Ankur Datta for immensely helpful comments and suggestions on this paper.

Footnotes:

1. R. Chandavarkar, The Origins of Industrial Capitalism: Business Strategies and the Working Classes in Bombay 1900-1940. Cambridge University Press, Cambridge, 1994.

2. G. Prakash, Mumbai Fables: A History of an Enchanted City. Princeton University Press, Princeton, NJ, 2010.

3. Cressida Jervis Read, ‘Frontier Town: Marking Boundaries in a Delhi Resettlement Colony 30 Years On’, in Ravi Sundaram (ed.), Sarai Reader 07: Frontiers, pp. 516-526. Sarai: The Centre for the Study of Developing Societies, Delhi, 2007; D. Mehra and L. Batra, ‘Neoliberal Delhi: Through the Lens of the Yamuna Pushta Slum Demolitions’, in C. Brosius and R. Ahuja (eds.), Megacities: Approaches to Metropolitan Cities in India. Draupadi Verlag, Heidelberg, 2006.

4. L. Fernandes, India’s New Middle Class: Democratic Politics in an Era of Economic Reform. University of Minnesota Press, 2006. D. Harvey, ‘The Right to the City’, New Left Review 53, 2008, pp. 23-40. T. Caldeira, City of Walls: Crime, Citizenship and Segregation in Sao Paolo. University of California Press, Berekely, 2000. C. Brosius, India’s Middle Class: New Forms of Urban Leisure, Consumption and Prosperity. Routledge, Delhi, London and New York, 2010. H. Donner, Domestic Goddesses: Maternity, Globalisation and the Middle Class Identity in Contemporary India. Aldgate, Aldershot, 2008. A. Krishna, ‘Stuck in Place: Investigating Social Mobility in 14 Bangalore Slums’, The Journal of Development Studies 49(7), 2013, pp. 1010-28.

5. The paper does not claim to provide a comprehensive understanding of all dimensions of inequality and people’s experiences of this inequality.

6. The hypothesis of ‘length of stay’ is a simple attempt to capture the difference between those for whom Delhi has been their only ‘home’ (i.e. those who have lived in Delhi their entire lives) and those who have come to Delhi from the outside (here the issue is not simply ‘migration’, but length of stay). Further to these two disaggregations, there are of course a number of other meaningful categories that require further elaboration – for instance caste, community. These are, however, outside the scope of this paper.

7. A. Heath, Social Mobility. Fontana, London, 1981. R. Erikson and J.H. Goldthorpe, The Constant Flux: A Study of Class Mobility in Industrial Societies. Clarendon Press, Oxford, 1992. R. Breen, Social Mobility in Europe. Oxford University Press, Oxford, 2004. A. Heath and C. Payne. ‘Twentieth Century Trends in Social Mobility in Britain’. Working Paper No. 70 CREST – Centre for Research into Elections and Social Trends, 1999. Available online: www.crest.ox.ac.uk/papers/p70.pdf [accessed September 2014]

8. Ibid.

9. A. Béteille, ‘The Family and the Reproduction of Inequality’, in P. Uberoi (ed.), Family, Kinship and Marriage in India. Oxford University Press, Delhi, 1993.

10. P. A. Sorokin, Social Mobility. Harper & Bros, New York, 1927.

11. Goldthorpe provides an overview of the field of social mobility research since its inception. J.H. Goldthorpe. ‘Progress in Sociology: The Case of Socieal Mobility Research’, in S. Svalfors (ed.), Analyzing I nequality: Life Chances and Social Mobility in Comparative Perspective. Stanford University Press, 2005. pp. 56-82.

12. P.M. Blau, O.D. Duncan and A. Tyree, The American Occupational Structure. Free Press, New York and London, 1967.

13. R. Erikson and J. H. Goldthorpe, op.cit., 1992.

14. Further, studies have looked at social mobility ethnographically – F. Osella and C. Osella, Social Mobility in Kerala: Modernity and Identity in Conflict. Pluto Press, London, 2000, for Kerala; as well as those that have used the life-histories approach to capture mobility – V. Benei, ‘To Fairly Tell: Social Mobility, Life Histories, and the Anthropologist’, Compare: A Journal of Comparative and International Education 40(2), 2010, pp. 199-212. In addition, recent studies have also looked at mobility associated with specific occupations: A. Krishna, ‘Examining the Structures of Opportunity and Social Mobility in India: Who Becomes an Engineer’, Development and Change 45(1), 2014, pp. 1-28. C.J. Fuller and H. Narasimhan, ‘From Landlords to Software Engineers: Migration and Urbanization Among Tamil Brahmans’, Comparative Studies in Society and History 50 (1), 2008, pp. 170-196.

15. S. Kumar, A. Heath and O. Heath, ‘Determinants of Social Mobility in India’, Economic and Political Weekly 37, 2002, p. 2983; E.O. Wright, Approaches to Class Analysis. Cambridge University Press, Cambridge, 2005, brings together various views on class including the neo-Weberian, neo-Marxist, neo-Durkheimian as well as Bourdieu’s class among others; class, of course, is also not unchanged. For more on the ‘death’ of class, see J. Pakulski and M. Waters, The Death of Class. Sage, London, 1996.

16. D. Dorling, ‘Thinking About Class’, Sociology 48(3), 2014, pp. 452-62.

17. R. Erikson and J.H. Goldthorpe, op. cit., 1992; and more recently, M. Savage, F. Devine, N. Cunningham, M. Taylor, Y. Li, J. Hjellbrekke, B. Le Roux, S. Friedman, A. Miles, ‘A New Model of Social Class? Findings from the BBC’s Great British Class Survey Experiment’, Sociology 47(2), 2013, pp. 219-50, attempt to include ‘capital’. P. Bourdieu, ‘The Forms of Capital’, in J. Richardson (ed.), Handbook of Theory and Research for the Sociology of Education. Greenwood, New York, 1986, pp. 241-258.

18. The use of occupation to capture class h as been one way to study mobility (Erikson and Goldthorpe, op. cit., 1992; see Kumar et al., op. cit., 2002; and D. Vaid and A.F. Heath, ‘Unequal Opportunities: Class, Caste and Social Mobility’, in A.F. Heath and R. Jeffery (eds.), Diversity and Change in Modern India, Proceedings of the British Academy 159, OUP, 2010, pp. 129-164, for the use of occupations in the Indian context), though this is not without its critics (see Savage et. al., op. cit., 2013 for a discussion). Further, in addition to information on occupation, economists have used other measures such as income, assets, wealth etc.

19. Kumar et. al., ibid., 2002; also Vaid and Heath, ibid., 2010.

20. D. Vaid, Class Mobility of Women and Men in India. Oxford University DPhil Thesis, 2007. Vaid and Heath, op. cit., 2010.

21. This paper looks at absolute rates of mobility; a study of patterns of relative mobility requires a separate engagement.

22. In the first stage, 20 Assembly Constituencies (ACs) were selected using random circular sampling from all 70 listed constituencies in Delhi. From these 20 ACs, interviews were conducted in 19 ACs, as field investigators were unable to find adequate households in one predominantly industrial AC. In the second stage, the sampling of polling stations (PSs) within each sampled AC was done. In the third stage, the sampling of households within each sampled polling station was done randomly by the investigator based on a sampling constant. The final stage involved the selection of the respondent based on quota sampling. Collectively, age and sex constituted the quota matrix ensuring adequate representation of women and men, as well as people from all age groups. The reason for using a quota sampling in the final stages rather than the list of households in each polling station provided in the electoral register was due mainly to the incompleteness of the electoral lists and the possibility of under-representation of migrants on the list.

23. Kumar et. al., op. cit., 2002. For debates around class see Wright, op. cit., 2005.

24. The ESMS survey question that captures subjective class: ‘Many people talk about class nowadays, and use terms such as lower class, middle class or upper class, while others say they do not belong to these but some other class. In your opinion, compared to other households, the household you live in currently belongs to the: (Probe if middle class – whether upper or lower middle class): 1. Lower class; 2. Lower-Middle class; 3. Middle class; 4. Upper-Middle class; 5. Upper Class; 6. Don’t understand class; 7. No opinion.’

25. For a study on class articulation in Madurai, see S. Dickey, ‘The Pleasures and Anxieties of Being in the Middle: Emerging Middle-Class Identities in Urban South India’, Modern South Asian Studies 46(3), 2012, pp. 559-99.

26. D. Vaid and A.F. Heath, ‘Unequal Opportunities: Class, Caste and Social Mobility’, in A.F. Heath and R Jeffery (eds.), Diversity and Change in Modern India. Proceedings of the British Academy 159, OUP, 2010.

27. The absence of women from social stratification and mobility studies has also been much debated with regards to the participation of women in employment in India and the impact this has on social mobility opportunities (G. Payne and P. Abbott. The Social Mobility of Women: Beyond Male Mobility Models. Falmer, London, 1990; Vaid, op. cit. 2007). The relation between caste and class and the impact of caste on occupational attainment, and in turn social mobility, has been discussed in scholarship (D. Vaid, ‘The Caste-Class Association: An Empirical Analysis’, Asian Survey 52(2), 2012, pp. 395-422) and requires further disaggregation.

28. Kumar et. al., op. cit., 2002; Vaid and Heath, op. cit., 2010.

29. This is derived from respondent’s and father’s occupation information (few respondents reported occupations for their mothers, 89% reported that their mothers were housewives when they were fourteen, making intergenerational calculations of mother to daughter mobility difficult due to small sample sizes). Only 21% of female respondents’ reported an occupation; 58% of the sample did not return an occupation, and from those who did 45% stated that they were housewives.

30. Some research has looked at marriage as an avenue of mobility for Indian women in comparison to their opportunities in the labour market (Vaid, op. cit., 2007).

31. Needless to say, the use of either measure has to be driven by the purpose of the study under question.

32. Pearson’s correlation coefficient (or r) ranges from -1 to +1. An r value of 0 indicates no correlation, and a value close to 0 indicates weak correlation. A value close to +1 indicates a positive correlation between the two variables – i.e an increase in one of these variables is connected to an increase of the other. In this instance, the low correlation indicates that the two measures of class under study appear not to be very closely related, though the correlation is not entirely absent. For more on correlations see A. Agresti and B. Finley. Statistical Methods for the Social Sciences, 3rd edition. Prentice Hall, 1997.

33. There is seen to be a very high positive correlation between subjective idea of class while growing up and current subjective class (r= 0.78). In comparison, the correlation between occupational class of father and class of respondent is also positive and strong (r= 0.57).

34. ESMS survey question: Why do you think you belong to this class? (Open-ended)

35. It would be of interest to study the idea of ‘capital’ in Bourdieu’s (1986) use of the term in relation to these articulations of class (see also Savage et. al., op. cit., 2013). See Dickey, ibid, 2012, for a study of Madurai.

36. ESMS 2012 question: ‘I can influence decisions affecting my city.’

37. It is interesting here to look in detail at the pattern of no responses to this opinion question – Kailash provides an interesting study of the pattern of ‘no opinion’ responses in a national survey and concludes that social position (or marginalized status) may lead to differing social or political opinion formation and expression. K.K. Kailash, ‘The More Things Change, the More They Stay the Same in India: The Bahujan and the Paradox of the Democratic Upsurge’, Asian Survey 52(2), 2012, pp. 321-47.

38. V. Gidwani and K. Sivaramakrishnan, ‘Circular Migration and the Spaces of Cultural Assertion’, Annals of the Association of American Geographers 93(1), 2003, pp.186-213.