defining needs ii: defining community

yeah, so, started the "miniseries" a few months ago when i had an office job in dc and was getting pumped to go to grad school. then didn't go to grad school, got a job in delhi, and have been somewhat busy/distracted since.

but! the job involves a lot of academic research, into the effects of various types of voter education campaigns in slums in india's capital city, which while interesting in its own right, has me thinking a lot about defining community. chris blattman in a (quote in a) recent post touched on what i am going to say baldly: identities, whether individual or group, are created by exclusions. this is true whether you're defining your own (i am this because i am not that) or someone else (they are that because they are not this).

the logical analogy at this point would be to say that identity:individual :: community:group (roughly--i know it's not perfect). but the problem is that looking at even individual identity, one "identity" is actually composed of multiple overlapping identities--loose affiliations with different groups (language, gender, region, occupation, political, etc.) or one's relation to others (mother, loner, leader, etc.). individual identities always, by definition almost, relate a person to other people around him/her.

it would be great if you could extend that to the group/community level, saying everyone sharing one particular identity (language, for example) is a community. two problems with that are that first, it's hard to say who "shares" a specific (in this case linguistic) identity--do you need to speak it from birth, speak it fluently, speak it predominantly, speak it most comfortably, speak it in a certain way (even ignoring the blurry lines between different linguistic varieties); second, this identity is just one of many that the individuals in this group may have--they may be divided by religion, income, nationality, race, gender, education, occupation, etc.

the largest theoretical "problem" (for me) with community, though, is the exclusionary action of this definition. dominant groups (eg, white english-speaking males in the US) usually have the luxury of having their community defined at the most basic level of societal structure, allowing them to exist as the "unmarked" or the "norm", thus discursively becoming the only "neutral" observers, and in many ways becoming the ultimate arbiters of culture (and politics, etc). "ethnic" communities exist often in order to preserve or promote their community in the face of the normalisation of the dominant group, but in this way often cause their group to become, by definition, marked and subordinate (or think of the rite of "coming out" in the gay community, by which one marks oneself as "not-normal", with the assumed sexuality being heterosexual in the absence of such a process. but i digress).

so communities are problematic--but why are they useful, and how are they used? for me, in my work, they're useful because they are a way of segmenting the mass of "everyone in the world" into discrete, manageable groups, setting boundaries to my work or study. by defining one group as the "target population" (ignoring for the moment the way in which that designation is made), you can set boundaries to what you will and will not work on, and with whom you will and will not work. by naming one group as the target, you implicitly place another group or groups out of bounds.

usually, these groups are recognized (by me, the person who has really no business being there) by self-identification--if someone says they're one thing, they're that thing. this is really a great and not-great way to do things. on the plus side, at least you're not putting people in boxes that they wouldn't put themselves in (as long as you're not pressuring them to choose something when the distinction is meaningless to them--see the colonial history of nigeria (or a lot of the rest of africa) for examples of this).

on the negative, it takes a lot of time, and a lot of talking to people, and usually a lot of dead ends. i worked in pune for a couple years, in one slum, and with only two communities. in the end, i got to know pretty much everyone living within a couple of kilometers of there by sight, and got to know the boundaries between different self-identified groups there. but, i don't know the right word for these "groups" in marathi, hindi or even english. the point is, i couldn't go in there and simply ask "to which group do you belong in regards to x dimension?", but rather relied on knowledge passed to me from my colleagues and months of face-to-face discussions in homes, getting to know genealogies, accents, naming patterns and the historical memories of people. on top of that, there was never (ever) a clear and all-encompassing consensus on who was in one group and who was in another. sub-groups, groups that may or may not have been the same but broke off in the past, personal feelings, and the context of the conversation always affected answers, and couldn't really be taken into account.

for my job now, we're working on a "complete" descriptive accounting of the life of the urban poor in delhi (a city of between 12 and 15 million, depending on who and how you ask). this will take a lot of surveys, a lot of pseudo-ethnography, and eventually an identification of group "informants" to answer questions about the "community" as a whole. i've yet to wrap my head around a decent way to decide who is able to speak for a community, and apply that in 100 (at least) different areas of the city. defining a community is difficult enough, especially given that different overlapping communities may form for different purposes--linguistic communities of migrants living near one another, or a community of women who access services at the same community centre, or a community of men working at a cluster of mechanics shops, or a religious community comprised of persons speaking many different languages.

my colleagues have often said that this is going to be "fairly easy, if time consuming", that we can go into an area, generally ask around for the "leader", and once a consensus begins to emerge, interview that person. i'm afraid that that approach is going to leave out a lot of very interesting variation, especially if we prompt people by giving examples of leadership (in political affairs, for example).

at least that's a bridge to cross when we come to it.


related posts:

defining needs i: defining poverty
Burn it to the Ground; or, Defining Needs: The Miniseries