The pesantren archipelago

2026-01-25 - 21:08

On 29 September 2025 a multi-story musala (prayer space) collapsed at Pesantren Al-Khoziny in Sidoarjo, East Java, killing at least 61 people. A similar tragedy occurred again about a month later on 29 October, when the roof of a female student dormitory collapsed at Pesantren Syekh Abdul Qodir Jaelani in Situbondo, East Java, claiming the life of one student. The level of anger and breadth of debate that followed—from demands for criminal investigations and critiques of pesantren governance structures to reflections on how pesantren constitute a distinct subculture within Indonesia’s educational and social systems—illustrate common perceptions about the importance of pesantren in Indonesian society. Observers have argued that pesantren occupy a key role in shaping grassroots political outcomes, whether through the political socialisation of santri (students), the broader authority of pesantren networks in local governance and policy debates, or the general political influence of kyai (religious leaders). Without denying the significance of these findings, we nonetheless lack some basic empirical foundations for systematically measuring the social and political effects of pesantren. Claims about the importance or uniqueness of pesantren often rest on a handful of notable—and likely atypical—cases, such as Pesantren Tebuireng, which produced national leaders like Gus Dur and Ma’ruf Amin, or Pesantren Al-Mukmin in Ngruki, associated with allegations of terrorism. To more systematically understand the place, role, and effects of pesantren in Indonesian society as a whole, we first need answers to fundamental questions: how many pesantren are there? Where are they located? What are their organisational affiliations? Without such baseline information, it remains difficult to ascertain how prevalent pesantren truly are and how far-reaching their influence extends.​ This piece takes a step forward by providing systematic, nationally representative background information on Indonesia’s pesantren landscape. Rather than engaging in an in-depth look at few pesantren, it embraces a bird’s-eye view of pesantren across the archipelago. The analysis addresses key descriptive questions about the number, location, organisational affiliation, and growth of pesantren over time, and offers preliminary evidence on how pesantren presence correlates with political outcomes. Data for this exercise was scraped from the Ministry of Religious Affairs’ (MORA) Education Management Information System (EMIS). First introduced in the early 2000s, EMIS is a MORA’s effort to monitor and supervise religious educational institutions, which include pesantren and madrasah. School administrators enter data into the system following specific forms tapping into different information. Information recorded from each institution includes the number of educators, the number of students, available classrooms and other facilities, founding year, and affiliation, among others. Interested users may access a subset of this dataset on my website. This dataset includes records on all pesantren, but omits several columns such as bank account information and staff contact details. Number, location, and affiliation The simplest analysis to do with the dataset would be to understand how many pesantren are there and where they are located. This question turns out to be more complicated than it looks. The EMIS website indicates that there are 42,391 pesantren. However, 20 of these records are duplicates and, out of the non-duplicates, 23 have broken links, which means their data could not be retrieved. We can thus reasonably assert that there are at least 42,348 pesantren in Indonesia. Figure 1 visualises the distribution of these pesantren at the district (kabupaten/kota) level. Some districts are shaded grey, which means that I was unable to link location information of the pesantren data with the GADM (Global Administrative Areas) map I used to create the choropleth. Figure 1: Number of Pesantren at District (Kabupaten/Kota) Level (click to enlarge) One notable pattern in the figure concerns how districts in West Java and Banten tend to have the most pesantren. Is it simply because districts there are more populated than other districts in the country? Figure 2 offers a more nuanced picture by showing the density of pesantren—that is, how many pesantren there are for every 10,000 Muslim residents in the district. The latter information was based on the 2010 census. The general pattern remains the same. Banten and West Java districts have the highest pesantren densities. However, districts in Aceh have darker shades than they are in Figure 1. In other words, even though there are fewer pesantren in Aceh than in Banten or West Java, once we account for population size, Aceh districts are generally comparable to West Java ones (though perhaps not Banten ones) in how prevalent pesantren are. Figure 2: Pesantren Density at District (Kabupaten/Kota) Level (click to enlarge) The next interesting exercise would be to unpick the affiliations of these pesantren. One notable feature of EMIS is that it asks pesantren to indicate their organisational affiliation. Overall, about 77.5% of pesantren indicate affiliation with Nahdlatul Ulama (NU) and about 15.6% indicate that they are independent or unaffiliated. The remaining 7% indicate various affiliations, such as with Muhammadiyah (1.54%), Nahdlatul Wathan (0.84%), and the Persatuan Tarbiyah Islamiyah (PERTI, 0.42%). Figure 3 presents the proportion of NU pesantren in each district. In the vast majority of districts, the pesantren landscape is dominated by NU. However, some interesting exceptions are evident. In Aceh, NU pesantren tend to be less dominant. Of all pesantren in Aceh, only 23.3% indicate an affiliation with NU. The vast majority (64.9%) indicate being unaffiliated. PERTI also has a quite significance presence in the province, accounting for about 4.4% of the pesantren there. Independent pesantren are also dominant in East Nusa Tenggara (NTT) and Central Sulawesi, where they account for 46.3% and 34.6% of the pesantren, respectively. The most popular affiliations differ in NTT and Central Sulawesi, however. In NTT, after NU at 29.3%, Muhammadiyah (12.20%) and Hidayatullah (4.9%) are the most popular affiliations. In Central Sulawesi, after NU at 27%, the most popular affiliations are Al Khairaat (18.05%) and DDI or Darud Dawah wal Irsyad (6%). Figure 3: Proportion of NU Pesantren at District Level (click to enlarge) Growth Having examined spatial variation, we can also look at temporal variation or how the number of pesantren grew over time. EMIS asked pesantren to indicate their founding year in both international and Islamic calendars. Unfortunately, many of the entries look implausible, and judgment calls have to be made to keep only pesantren with plausibly valid information about their founding year in the dataset. First, I removed 3,010 pesantren that indicated founding years earlier than 1475 AD. This is the founding year of Pesantren Alkahfi Somalanguk, which some regard as the oldest pesantren in Indonesia. Second, I removed 33 pesantren that indicate founding years after 2025—the current year. Lastly, I removed 7,628 pesantren whose founding year in the international calendar does not match its founding year in the Islamic calendar. Such a mismatch makes it impossible for an analyst to decide whether to use information from the international calendar or the Islamic calendar. It might also indicate sloppy data entry. Overall, 29,305 pesantren have information about founding year that I consider valid. Figure 4 draws from this information and presents the cumulative number of pesantren between 1966 and 2024. I treat this growth chart as a disrupted time series and draw two regression lines, one for the New Order era and the other for