On January 21, Nepal faced a peculiar rebuke. At least sixteen of the recommendations issued during its Universal Periodic Review named caste discrimination or Dalit rights directly, several calling on the government to enforce its 2011 Caste-Based Discrimination and Untouchability Act—legislation that already exists, already criminalizes untouchability across public and private life, and has been on the books for fifteen years. The call was not for new law. It was a plea for the government to actually use what it claims to have.
This moment, during Nepal’s fourth Universal Periodic Review before the UN Human Rights Council, exposes something structural about how international human rights machinery addresses caste: the system has learned to recognize discrimination. It can issue formal recommendations. It can pressure governments to comply. But it has almost no capacity to ensure that recognition translates into enforcement. The result is ritual masquerading as remedy.
Consider what this means in practice. On January 27, 2026—six days after the UPR concluded—Jaruwa Devi Paswan, a Dalit woman in a Nepali village, was brutally beaten by neighbors. Her transgression: livestock had strayed onto their land. By her own account, she was already apologizing and pulling her goats out of the field when the beating started. Her actual offense: being Dalit in a place where neighbors still believe untouchability requires punishment. She could theoretically file a case under the 2011 Act. Nepal’s constitution explicitly classifies untouchability as a “severe social offence punishable by law.” The statute mandates prison terms up to three years and victim compensation. And yet the odds that Paswan receives justice are remote. Since 2011, Nepal has registered only 30 to 43 caste discrimination cases per year. Of cases that reach trial, the acquittal rate hovers near 63 percent.
The 2011 Act itself was meant to be transformative. When Nepal enacted this legislation—becoming the first South Asian nation to explicitly criminalize caste-based discrimination and untouchability across both public and private spheres—it appeared to represent progress. The law prohibits denial of access to public spaces, temples, water sources; obstruction of inter-caste marriage; employment and education discrimination; refusal of hospitality. On paper, it is among the world’s most comprehensive anti-caste statutes. Yet the chasm between statutory text and lived reality is near-total. In a typical year, fewer than 5 percent of actual untouchability incidents reach law enforcement or court.
This pattern is not unique to Nepal. It reflects a global phenomenon that international human rights bodies have quietly documented but largely failed to confront: caste discrimination can now be named at the highest levels of global governance. But naming almost never produces enforcement.
The CERD adopted General Recommendation XXIX in 2002, explicitly extending the International Convention on the Elimination of All Forms of Racial Discrimination to include “descent-based discrimination”—encompassing caste. The European Parliament issued a resolution on caste discrimination in 2013. The UN Special Rapporteur on Contemporary Forms of Racism documented in 2025 that approximately 250 million people globally face caste-based discrimination. Since 1991, UN treaty bodies have issued 84 Concluding Observations addressing caste discrimination across multiple countries.
Yet this cascade of formal recognition has produced negligible change on the ground. Nepal’s government accepted the UPR’s recommendations, and three months later, in April 2026, the ruling Rastriya Swatantra Party went further still: its chairman, Rabi Lamichhane, told the newly elected parliament that “no Nepali shall ever have to bow down again in the name of caste,” calling historical caste-based discrimination “not a social mistake but an organized crime.” These are not empty words. They represent political consciousness that did not exist a decade ago. But they do not translate into rapid convictions, systemic police reform, or institutional capacity to protect Dalit citizens. Acceptance and apology are not implementation.
The structural weakness lies not in Nepal’s particular failure but in how international human rights law operates. International mechanisms like the UPR, CERD, and UN special rapporteurs lack enforcement power. They issue recommendations that states can simply “note” without legal consequence. They rely on voluntary compliance, what scholars call “naming and shaming,” a euphemism for powerlessness. They abstract caste-based discrimination into a generic category of “descent-based discrimination,” obscuring its nature as a comprehensive hierarchy organizing access to employment, education, political power, and justice itself. They focus on individual recognition harms, untouchability practices, denial of entry to temples, rather than on caste as a structure of graded inequality requiring fundamental redistribution of authority and resources.
India illustrates the problem at scale. The country has had an anti-atrocity law since 1989—older and more established than Nepal’s statute. The Act should have created decades of enforcement precedent. Instead, government data documented an 89 percent acquittal rate in atrocity trials during the late 1990s and early 2000s. More recent NCRB data on SC/ST atrocity convictions shows persistent low conviction rates across states, with high-acquittal patterns even in disaggregated state-level data. The problem is not insufficient legal text. It is institutional refusal, embedded in enforcement mechanisms themselves.
Police demonstrate systematic reluctance to register caste crimes. Judges acquit despite evidence, often favoring reconciliation or downgrading charges. Survivors do not report out of fear of retaliation and because faith in judicial remedies has been systematically eroded by generations of impunity. The criminal justice system exhibits institutional caste bias—not as occasional prejudice but as structural preference built into investigation, prosecution, and judicial practice. International human rights mechanisms have almost no leverage over this. They cannot retrain police forces. They cannot appoint judges. They cannot generate the domestic political will necessary for enforcement.
What would need to change is specifiable, even if international law cannot impose any of it directly. Police forces across South Asia would need to be trained, monitored, and held accountable for investigating caste crimes with the same rigor applied to other serious offenses, not symbolic training but systematic personnel changes and transparent accountability mechanisms; Nepal’s police force remains overwhelmingly non-Dalit, and no caste crime can receive fair investigation when the investigating officer inhabits the same caste hierarchy that motivated the violence. Judges would need the will to convict despite the social pressure caste violence generates, the pressure to reconcile, to protect the accused’s social standing, to treat caste as a private matter amenable to compromise; in India that means treating atrocity law as foundational to dignity rights rather than optional supplement to ordinary criminal law, and in Nepal it means judicial leadership stating plainly that caste crimes are non-negotiable prosecutions, a will that neither international recommendations nor UPR pressure has yet produced. And states would need to face real consequences for non-enforcement: the current system lets governments ratify human rights conventions, enact domestic law, apologize publicly at the UN, and then fail to prosecute without suffering meaningful cost, where a genuine enforcement system would require trade conditionality, conditional aid, or some other mechanism that makes caste justice a condition of international standing. The International Dalit Solidarity Network and other advocates have pushed toward exactly this. The global human rights system resists it, perhaps because enforcement could implicate dozens of states at once, perhaps because the machinery was built to preserve state sovereignty above all else.
The January 21 UPR outcome was not nothing. Formal recognition matters. When sixteen separate recommendations call for enforcement of anti-caste law, they legitimize the claim that caste justice is a global human rights priority, not merely a cultural or regional preference. This creates space for domestic advocates to amplify their demands, to cite international precedent, to argue that Dalit dignity is a universal standard rather than a local option.
But without enforcement capacity, recognition becomes diplomatic theater. Jaruwa Devi Paswan’s neighbors beat her because they believe they will face no consequence. Until that belief changes—until perpetrators see others convicted, until police register their crimes, until judges sentence them proportionally—laws remain script. The gap between what Nepal’s statute says and what happens to Dalit citizens will persist, sustained by an international system that has mastered the language of human rights without developing the power, or the will, to enforce them.
