How do political parties in Pakistan aggregate votes and connect to voters, especially in rural areas? Most explanations for party-voter linkages in rural Pakistan can be grouped into four broad categories: (a) political parties are made up of powerful landlords who use economic power to collect votes from dependent voters; (b) parties are conglomerations of clientelist networks; (c) parties are large aggregations of kinship networks; and (d) parties function by building links with specific types of constituencies across large parts of the country, and voters forge links with political parties based on ideology and party identification. These are, at best, only partial explanations. How is it that they can all co-exist? The answer lies in the level of analysis employed. I argue that when viewed in aggregated form at a macro-level of analysis, such as that of national politics, these varied explanations of party-voter linkages appear to be distinct and sometimes mutually exclusive. However, when disaggregated and examined at a micro-level of analysis, that of village-level politics, it becomes obvious how it is possible for all of these linkages to not only co-exist but often even work together. I adopt this micro-level of analysis through a longitudinal study of one village in Pakistan’s most populous and politically most important province, Punjab. Using the case study of Sahiwal, I argue that the reason for the varied explanations is that the link between political parties and their voters is rarely direct. Instead, it is mediated by different types of local actors. As the national political arena changes, different actors gain precedence, leading to multiple explanations of what is really going on in rural Pakistan between political parties and their voters.