The federalism-unitarism debate
In today’s Afghanistan, few debates cut as sharply across ethnic lines as the one over whether the state should be federal or unitary. Many Pashtun elites have long favored a unitary state, arguing that federalism could lead to the partition of Afghanistan, opening the door for neighboring countries to exert dominance or even annex different regions of Afghanistan. By contrast, non‑Pashtun elites, including Hazaras, Uzbeks, and Tajiks, have advocated for a federal system, believing that only with autonomy can they protect and promote their cultures, identities, and political representation. Many in these ethnic groups believe that a unitary system fosters ethno‑authoritarianism and central domination.
These opposing visions are deeply ingrained in the political and intellectual circles of all ethnic groups in Afghanistan. Each side cherry‑picks examples of successful federal or unitary systems from around the world to validate their case, while dismissing other views as insincere or misguided. This entrenched framing leaves little room for compromise. As a result, many have seemingly concluded that reaching consensus on a political system is impossible.
Opportunity for design and dialogue
Yet the deadlock itself may reveal an opportunity to move toward a pluralist process of design and dialogue. A pluralist political framework channels inevitable conflicts into engagement rather than domination and exclusion. Such an approach begins with recognizing that every community’s perspective deserves respect, rejects dismissal, and rests on civic equality and the representation of diversity through innovative institutional designs.
A pluralist approach begins by examining the various possible arrangements that exist between federal and unitary systems. This is essential for challenging mistaken assumptions and widening the range of options available to constitutional designers. The old federal-unitary dichotomy has long lost its descriptive power as countries increasingly experiment with hybrids and crossover models. Today, there are unitary systems that are highly decentralized administratively and politically, such as Spain, the United Kingdom, and Indonesia. Conversely, there are centralized federal systems, such as Malaysia. The federal–unitary binary no longer captures the diversity of constitutional arrangements, and in Afghanistan, the politicization of these terms has made them even less useful. Federalism is often equated with partition and disintegration, while unitarism is linked to authoritarianism. These misconceptions increasingly shape political divides.
This is why a pluralist process should begin with engaging more tangible and operational questions, which are usually not politicized as much, such as:
- Should governors be elected?
- What powers should provincial councils or parliaments have?
- Which decisions should be made locally and which centrally?
When specific design questions like these are discussed, people from diverse communities in Afghanistan express far more agreement than expected. Negotiating the name of the system can wait until after a design is agreed upon. A pluralist design also avoids beginning with judgments about the sincerity or fairness of claims made by different ethnic groups. Simply dismissing the concerns of Tajiks, Pashtuns, Hazaras, or any other group as illogical or self-serving does not make their concerns and interests disappear. Whether those views are fair or unjust, modern or ancient, the fact remains that a majority or significant plurality within each group believes them to be in their interest. A genuine pluralist approach requires engaging with all competing concerns rather than wishing them away. Fortunately, the rich body of literature on constitutional design, based on the experiences of diverse countries, offers ample tools for a thoughtful and informed pluralist to find arrangements that can accommodate most, if not all, the needs and aspirations of each group equally.
Practical steps
Self-governance is key in such an arrangement. Tajiks, Uzbeks, Hazaras, and other groups seek autonomy in regions where they form a majority or plurality. Ignoring these demands amounts to domination, authoritarianism, and exclusion, not a pluralist representative regime. Some level of autonomy should be guaranteed to local governments, including elected governors, councils, and mayors with greater authority over regional issues, and control over matters involving culture, identity, and customs. Interestingly, many Pashtun communities, especially in rural provinces, also resist central government intrusion into their daily lives. Jirgas and Shuras, as indigenous mechanisms of dispute resolution, hold significantly more power than the courts and police in Pashtun-dominated areas compared to other regions.
Many Pashtuns fear that decentralization would split Afghanistan into multiple independent states. This is based on the dominant assumption that decentralization in Afghanistan would stem from a few large, ethnically concentrated territories. But decentralization in Afghanistan does not have to be this way. Instead, this concern can be addressed by creating smaller political units, such as the current provinces or even smaller ones. This approach prevents any single unit from ceding while allowing each group representation and autonomy in multiple units. In addition, a guarantee clause can be added to the Constitution, allowing the central government to take control of a local government temporarily under certain emergency situations.
Meanwhile, concerns over foreign meddling can be addressed by reserving defense, foreign affairs, and trade for the central government while delegating policing locally—a standard feature of federal systems worldwide. Some issues that have long divided ethnic groups, such as higher education law or the design of national identification cards, should be decentralized. During the Republic, key laws on higher education and identification cards stalled in parliament for more than a decade. Localizing such matters reduces political deadlock in the capital, which in turn allows the central government to focus on country-wide strategic priorities.
Both Farsi and Pashto speakers, along with speakers of other local languages, fear that their languages may be eroded through assimilation or the dominance of another language. A pluralist constitutional design should address this by allowing local governments to conduct administration in their own languages, while requiring that all central government communications and interprovincial documentation be bilingual in Farsi and Pashto. When the protection of linguistic rights is explicitly guaranteed in the constitution, these communities are far more likely to feel secure, easing long-standing tensions.
Conclusion
Afghanistan’s political future does not have to be trapped in a binary stand-off between federalism and unitarism. By adopting a design-first pluralist approach, it is possible to address legitimate ethnic grievances, dispel myths, and create a governance system that honors diversity while safeguarding unity. Ultimately, the name of the system matters less than for every community to see itself in it—and feel both protected and empowered within the system.
Download a PDF of this issue »
Bashir Mobasher is a political scientist and legal scholar specializing in constitutional design, governance, and human rights in divided societies, with teaching experience at NYU, American University, and the American University of Afghanistan.
