(As a follow up to my first post on electoral surveying, this second, and admittedly long overdue, piece focuses on pre-election reportage and some questions regarding urban elections in Pakistan.)
In the late 90s, Andrew Wilder wrote that the vast majority of electoral coverage, both in the vernacular and the English press, tends to focus on biraderi dominated voting blocs, especially in Punjab. In a typical election-eve report, one would find a brief history of the constituency, the names of all major candidates, and the preferences of local clan heads and their respective followers.
Here’s what one such report would normally look like:
The constituency, which was carved out of NA-92 in 2002, includes the areas which are densely populated like Shahdara, which comprises of some highly populated localities like Shahdara Town, Kot Shahabuddin, Jia Musa, Dher, Match Factory, Qaiser Town and Haji Kot. NA-118 also comprises of areas like Khokhar Road, Khokhar village, entire Badami Bagh, Siddiqia Colony and adjoining areas. Major clans of the constituency are Arain, Kakkay Zai, Rajputs and Kashmiris whereas settlers hailing from the Pakhtoon community are also in good number
And another one:
After the PPP leadership nominated Lalla Asadullah, also an old party worker and brother of former MPA Lalla Shakeel and party’s city president Lalla Idris, the elders of the Arain Biradri decided to field their own candidate for the by-poll. The Biradari elders, including former MPA Chaudhry Shabbir Mehr, Chaudhry Tariq Ali, Chaudhry Babar Mehr, Chaudhry Ehsanullah, Mehr Khalid Mahmood and Chaudhry Azam Iqbal Mehr played a vital role in resolving mutual disputes among various clan members and bringing them on single platform — City Ittehad Group. According to sources, the Arains have launched a vigorous electoral campaign for Sarfraz Mehr and were contacting other politically influential groups in the constituency.
Curiously enough, both of these reports are for wholly urban constituencies.
This utilization of clan politics as a dependent variable in predicting electoral outcomes, especially in urban areas, could be because of two reasons:
1) Clan identity, in so far as it helps to consolidate vote blocs, remains as salient in urban areas as it is in rural parts of the country;
2) Nobody’s really sure what determines an election in urban areas so they go with the laziest explanation available.
Increasingly, I’m beginning to think option number 2 is closer to the truth than option number 1, but also that option number 1 will remain in favor for this and all foreseeable elections. The reason for that is simply because there is really no one theory that can explain electoral competition in Pakistan’s urban landscape, and that nobody is willing to do the legwork required for a holistic understanding of these issues. Sure, we have theories about how violence and ethnic mobilization partially explain voting behavior in Karachi and Quetta, but what about homogenous and less-fragmented cities like Rawalpindi, Lahore, or Faisalabad?
In some of the densely-populated, closely-clustered parts of these cities, clan identities are still quite important in shaping election-day mobilization and patronage networks. At the same time, though, many cities in Pakistan (Lahore, especially) are experiencing rapid suburbanization, and the development of autonomous, inward-looking neighborhood structures where there’s little conception of civic-culture and markedly less interaction between residents in the same area. The interesting thing is that this isn’t just happening in upscale gated communities (like the army-run DHA) but also in squarely low to lower-middle income neighborhoods. With the general absence of ‘baithak’ culture in newer settlements, mosques (and maybe the tea-shop) probably remain the only place of public interaction between household heads in the same area. I haven’t been to a mosque in a while, but I’d be interested to know whether that space provides fertile territory for local level vote-bloc creation.
Another important question – and one which flows directly from the overarching concern of how elections are decided in urban areas – is the impact of campaigning in Pakistan. We’ve seen every single party put up huge billboards in every possible nook and crany, and yet there’s really nothing out there which suggests that mass-advertising does or does not have an impact on electoral outcomes. I discussed this particular issue with Prof. Hassan Javid, a political sociologist at LUMS working on all sorts of interesting stuff, and we both agreed that there’s no way a candidate will ever know the disentangled impact of mass-advertising on his electoral outcome. There’s no evidence to suggest that it goes either way, and I’m not entirely sure what the empirical literature suggests for other similar countries. And yet, despite this complete lack of evidence, we see so much money being spent on text message campaigns, cable tv advertisements, and billboards with huge and largely unpretty faces on them. Surely someone, at some point, will stand up and ask if this is really worth it, right?
Finally, and this is perhaps the biggest and murkiest question of them all, is the issue of patronage – i.e. state employment, public goods provision, mediating between state officials and clients, and reciprocity in terms of votes. Most amateur political scientists and sociologists (me included) are quite hung up on patronage as an overarching, explanatory framework for politics in this country. It’s a bit lazy on our part, yes, but frankly it works so well for rural areas that most people just go along without questioning it in an urban environment. So it was a bit of a reality check for me when while conducting household surveys in a low-income neighborhood of Lahore, I realized the patronage-space in that area had pretty much been saturated. There were two government schools, a well functioning dispensary, freshly carpeted streets, and an underground sanitation system. All households had access to electricity, piped water, and gas, and there was nothing to suggest that residents were completely dependent on state employment, or that the police was overly predatory (hence mandating an intermediary’s existence). What I left with, after spending a few hours there, was that in some (even low-income, densely-populated) urban areas, patronage is non-existent, the electorate either doesn’t vote, or votes autonomously, and that the space for creating hierarchical networks is simply absent. Ancillary to this is where there are such hierarchical networks, enforcement of voting discipline (i.e. ensuring that a voter votes for the anointed candidate) is much more difficult in an urban area, where there’s greater anonymity at the polling station level.
(As a side, the head of a landed family from Sargodha used to ensure his villagers were voting the right way by sitting at the window of a government school that was doubling as a polling station. He would then ask each villager to hold up his or her ballot before putting it in the box.)
Anyway, these were some of the questions that need to be answered before we can start making sense of voting in urban constituencies. As things stand, survey data can show us relative favorability of parties at the macro-level, but that doesn’t give us much in terms of explanations for how they’ll do in a very heterogeneous electoral environment. Finally, reporters and editors really need to get out there and explore electoral dynamics in constituencies that do not conform to their pre-existing view of the universe. Without legitimate anecdotal evidence on these matters, it’ll be even harder to conduct a more serious empirical inquiry on voting behavior in Pakistan.