4.1 Million Votes, Nothing in the Plan: China's Cruelty Law

In October 2025 China's Ministry of Justice did something it does every year: it opened a call for proposals for the next year's legislative programme. Members of the public could submit suggestions by email or post until 26 November.

During that window, an unofficial online vote circulated among animal welfare groups and their supporters. By the time it closed, it had drawn more than 4.1 million votes in favour of putting an anti-cruelty to animals law on the legislative agenda, a reported 96.45 percent of those who took part.

Nine months later, in May 2026, the State Council published its legislative work plan for the year. It sets out the bills the government intends to prepare or submit to the standing committee: fourteen to be put before the NPC Standing Committee, plus a long list of reserve items. There is no anti-cruelty law in either place.

That is not a scandal and it is not a mistake. It is how the process works, and the distance between the two events is worth understanding before anyone decides what the vote meant.

What the vote actually was

The distinction that most coverage loses is this: the Ministry of Justice asked for proposals. It did not hold a referendum, and it did not publish a tally of 4.1 million opinions.

The vote was organised by supporters, not by the ministry. Its 4.1 million figure measures something real, a very large number of people willing to spend a moment registering support, but it is not a count of submissions to the ministry and it carries no formal weight in the process. No official figure appears to have been published for how many proposals the ministry actually received on this subject.

This matters because the number has been repeated in reporting, including in coverage of court cases, in a way that collapses the two things into one. Four point one million people supported it, and four point one million people submitted it to the Justice Ministry, are not the same claim, and only the first is supported by what is publicly available.

None of this makes the number meaningless. A legislature that received four million supportive submissions through an official channel would be remarkable. Four million people voting in an unofficial poll organised by campaigners is a different kind of signal, of public sentiment rather than of procedural record, and it is still a signal worth having. It is just not a vote that any institution is required to count.

Where a proposal goes

The annual call for proposals is a standing part of how Chinese legislation begins, not a special track created for this issue. What arrives is considered alongside everything else the government is asked to legislate on that year.

What happens next is visible in the May 2026 work plan. It lists fourteen bills to be submitted to the NPC Standing Committee: on finance, national defence mobilisation, public tendering, the central bank, water, road traffic safety, tax collection, teachers, earthquake prevention, government offices, exit and entry, pricing, lawyers and customs. The reserve list runs to nearly thirty more, covering telecommunications, metrology, consumption tax, commercial banking, insurance, highways, railways, electricity, trade associations, territorial planning, transport, police, detention centres, space, broadcasting, intangible cultural heritage, food safety, product quality, elderly care, blood donation, pharmacists, property registration, flood control, mine safety, disaster prevention, state reserves, cybercrime and judicial assistance.

An anti-cruelty law appears in neither.

The work plan is a one-year schedule, so its silence is not a permanent refusal. It does tell you what the government has committed its drafting capacity to for 2026, and that list is already long.

How long this has taken

The vote was the latest event in a process older than most of its participants.

Legal scholars prepared a draft in 2009, initially called the Animal Protection Law and later renamed to narrow its scope to cruelty. It went to the NPC Standing Committee in 2010. In 2020 the Ministry of Agriculture and Rural Affairs responded to a proposal by saying that existing laws could be improved to address abuse rather than requiring new legislation. In 2023 a delegation from Hainan proposed bringing cruelty within the Public Security Administration Punishments Law, which is where something did eventually land, though not in the form its proposers wanted.

Sixteen years on from its submission to the Standing Committee, the subject is still at the stage of being proposed.

What has moved instead

Two things have changed on the ground, and neither is the law four million people voted for.

The first is local. Several cities, including Changsha and Nanjing, have passed local regulations prohibiting abuse of animals. The NPC's Agriculture and Rural Affairs Committee has encouraged localities to go first and to strengthen enforcement. Local rules are narrower than a national law and do not create offences enforceable everywhere, but they exist and they can be applied.

The second is the revised Public Security Administration Punishments Law, in force since 1 January 2026, whose Article 89 brings animal keeping into administrative penalties for the first time. It has been described as a breakthrough, and we have written about what it actually does: it penalises the keeper's conduct where that conduct affects people, and an unleashed dog is not itself a violation. It is not a cruelty provision, and it does not reach a person who harms an animal belonging to nobody.

Both are real. Neither is what was voted for.

Why the distance

For a subject to become law in China, a body has to draft it, it has to appear in the State Council's work plan, and then it goes through three readings before the NPC Standing Committee. Each stage is a filter, and each has its own reasons to say not this year.

For this subject specifically, the hard questions are not about whether cruelty is wrong. They are definitional and practical. What counts as cruelty, and who decides in a given case. Which animals are covered, given that the treatment of farmed animals is governed by a different and largely settled set of expectations. Which body enforces it, and with what resources. How a new offence would sit alongside the property and public safety provisions prosecutors already use, which is the route taken in every case we have covered, from Chongqing to Beijing.

Those are answerable questions, and other jurisdictions have answered them. They take drafting time and agreement between ministries, and the 2026 plan shows what was competing for that capacity.

What a law would and would not change

This is worth being precise about, because the frustration behind the votes is real and the answer is not simply that a law would fix everything.

A cruelty offence would change the charging problem directly. In the Chongqing case we wrote about, a man who obtained dogs by pretending to adopt them could be dealt with only through administrative detention, because there was no offence to charge. In the Beijing case, the poisoning of nine dogs was prosecuted only because the poison also endangered people; to the court the dogs were worth 8,000 yuan, and to the criminal charge they were worth nothing. A properly drafted cruelty offence would give both cases somewhere to start.

It would not, on its own, change what courts are willing to award. The Beijing court refused damages for mental suffering even after convicting the man who killed the dogs, on grounds that go to the scope of compensation in criminal proceedings rather than to the status of animals. Legislation about animals would not necessarily reach that question.

The honest position is that a law is a precondition for a great deal and a guarantee of very little, and that people arguing for one are usually right about the first part.

The vote counted something. Four million people, whether or not the figure entered any official channel, is a large number of people who want this and said so. That is evidence about public sentiment, which is one input into the process and not the whole of it. What turns sentiment into a bill happens afterwards, in drafting rooms and committee meetings that generate no votes and no headlines.

The 2026 plan says where that work currently stands. It is not where four million people would like it to be, and it is not nothing.