——从西藏吉隆泥石流看中共的灾难治理逻辑

作者：冯仍

2026年8月26日，一场突如其来的巨大泥石流冲入西藏日喀则吉隆口岸。冰块、岩石、泥沙和洪水从高山峡谷奔涌而下，车辆、道路和建筑转眼之间被吞没，现场流出的视频让人震惊。那些画面没有什么政治色彩，人们看到的只是人在自然灾害面前的渺小和无助。

后来，中国自然资源部门确认，这场灾难源于尼泊尔境内的高位冰川崩塌。冰川崩塌之后形成高速碎屑流，沿河谷一路向下，最后演变成破坏力巨大的泥石流。从目前公开的专业资料来看，这首先是一场严重的自然灾害。对于这一点，并没有什么需要回避。

真正让我无法理解的，是灾难发生以后再次出现的那一套熟悉做法。

截至8月29日18时，中国官方公布吉隆一侧已有16人遇难、546人失联，其中还包括大量外籍人员。数千名解放军、武警、消防、公安和工程救援人员进入灾区参与搜救。那些冒着生命危险寻找幸存者的救援人员，当然值得尊敬。任何一个正常人都希望，他们能够从废墟和泥沙中多救出一个人。

但救援人员值得尊敬，并不意味着政府可以不受监督。恰恰是在这种涉及大量生命的重大灾难面前，公众更有权知道完整的事实。

《西藏日报》发布的对比影像显示，星期三（8月26日）泥石流发生后，西藏吉隆口岸建筑已夷为平地。 

我看到中国数字时代整理这场灾难的相关资料后，最深的感受不是灾害本身，而是一种多年没有改变的熟悉感。灾难刚刚发生，现场视频在网络上传播，人们开始询问伤亡情况和灾害原因，很快便出现视频无法转发、相关文章被删除、网络讨论受到限制的情况。随后，警方又开始通报所谓“造谣”案例，有网民因为讨论死亡数字和灾害原因而受到行政处罚或者批评教育。

我不赞成任何人故意编造伤亡数字。灾难中的谣言会给家属造成新的伤害，也可能干扰救援。但是，一个政府如果真的希望制止谣言，最有效的办法应该是及时、充分、透明地公布事实，而不是一边限制现场信息，一边要求所有人只能接受官方版本。

当完整视频不能正常传播，当记者无法自由进入现场，当一些讨论灾情和风险的文章被删除，当普通人对死亡数字提出疑问也可能受到警方处理，那么所谓“辟谣”就不再只是维护事实，它实际上已经变成了政府对灾难信息解释权的垄断。

这也是我这些年越来越深的一种体会。

我过去长期生活在中国。每当发生重大事故或者公共事件，我们普通人获取真相的方式往往都非常奇怪。官方媒体说一套，微信群里传一套，微博上又出现另一套。有人上传现场视频，大家赶紧保存，因为谁也不知道半个小时以后还能不能看到；有人写了一篇调查文章，朋友圈里刚刚传开，第二天链接就已经打不开了。

久而久之，很多中国人甚至形成了一种非常悲哀的经验：越是被迅速删除的东西，大家反而越想知道里面究竟写了什么。

这不是因为中国人特别喜欢相信谣言，而是长期的信息控制已经消耗了政府自己的公信力。

一个政府如果习惯于只公布对自己有利的内容，那么即使有一天它公布的完全是真话，人们也会本能地怀疑。这种结果并不是境外势力造成的，也不是几个网民造谣造成的，而是一次又一次的信息封锁造成的。

吉隆这场灾难尤其值得追问，因为这里并不是第一次发生严重灾害。

2025年7月，吉隆地区已经发生过严重山洪泥石流，中尼友谊桥被洪水冲毁，中国一侧也有人失联。相关地质研究此前也已经指出，吉隆地区存在高位崩塌、冰湖、泥石流等多种地质风险。如今，仅仅一年多以后，更大的灾难再次发生，公众当然有理由追问：上一场灾害之后，当地到底做过什么风险评估？预警体系有没有得到加强？口岸重新开放以后，有没有重新评估周边人员密度和基础设施安全？如此巨大的冰崩和泥石流发生之前，是否存在可以被监测到的异常信号？

这些问题都很正常，遗憾的是，在中国，很多正常问题一旦涉及政府责任，很快就会变成所谓“敏感问题”。

我这些年观察中国的重大灾难，越来越感觉到，中共真正关心的往往不仅是如何救灾，还包括如何“管理灾难”。所谓“管理灾难”，不只是调动救援人员、恢复道路和安排物资，更重要的一部分是管理舆论：哪些数字什么时候公布，哪些画面可以出现，哪些记者可以采访，哪些家属可以发声，哪些问题不能继续追问。

从汶川地震、温州动车事故，到郑州“7·20”特大暴雨，再到疫情期间的武汉，我们已经见过太多类似场景。灾难发生以后，最先出现的是“领导高度重视”、“作出重要指示”、“全力组织救援”，接下来是救援队伍、感人故事和所谓社会正能量。至于灾难为什么发生，有没有人失职，真正死亡多少人，家属有什么质疑，往往要么被推到后面，要么最终不了了之。

这套模式最大的危险，是它慢慢让灾难本身也成为政治宣传的一部分。

死了多少人，不再单纯是一个生命问题，而可能被视为一个“影响社会稳定”的数字；记者调查，不再只是新闻工作，而可能被认为是在“制造负面舆情”；家属要求解释，也有可能被当成影响救灾秩序；网民提出质疑，则很容易被扣上“造谣”“传谣”的帽子。

一个政府如果连自然灾害都必须按照政治需要进行包装，那么它真正维护的究竟是什么？

显然不是死者。

真正需要被维护的，是政权永远正确、政府永远有力、社会永远稳定的形象。

我现在生活在美国，再回头看中国这些事情，最大的感受之一，就是一个政府到底有没有自信，并不是看它每天宣传自己多么强大，而是看它敢不敢面对坏消息。

一个真正有自信的政府，不会害怕公布死亡人数。死亡数字再难看，也不是政府的敌人；一个真正负责任的政府，不会害怕记者进入灾区，因为记者发现问题，本身就是公共监督的一部分；一个真正把人民生命放在第一位的政府，更不应该害怕家属追问亲人的下落。

如果一场泥石流发生以后，政府第一时间除了救人，还必须担心哪些视频传播出去会“影响形象”，哪些文章需要删除，哪些声音必须压下去，那么所谓“人民至上、生命至上”，就很难不让人怀疑究竟有多少真实成分。

自然灾害没有政治立场。

冰川不会因为一个国家实行什么制度而停止崩塌，泥石流也不会因为政府宣传“社会稳定”就改变方向。自然灾害发生以后，一个政府真正能够做的，是尽可能减少伤亡，把已经发生的事情完整告诉社会，并且认真查清楚哪些损失原本可以避免。

如果灾前存在预警不足，就应该调查预警体系；如果基础设施规划存在问题，就应该追究规划责任；如果地方官员在风险评估中失职，就应该让责任人承担后果。这才是一次灾难之后真正应该留下来的东西。

否则，道路重新修好了，桥梁重新建起来了，领导慰问结束了，媒体也停止报道了，但是造成大量伤亡的制度漏洞却什么也没有改变。几年以后，另一个地方再发生一场灾难，同样的故事又会重新上演。

这才是最可怕的事情。

在我看来，中国这些年的许多灾难都有一个共同问题：天灾结束以后，人为制造的“信息灾难”才刚刚开始。真正的洪水会退去，泥石流会停止，但删除的信息、无法发声的家属、不能追问的责任，却可能永远被埋在地下。

所以我认为，吉隆泥石流最应该留下来的，不只是几张救援照片，也不只是几句“众志成城”的口号。

应该留下死者的名字。

应该留下失踪者真实、完整的数字。

应该允许家属说出他们经历了什么。

应该让记者自由调查。

也应该把2025年那场灾害之后当地究竟做了哪些预防措施，一项一项向社会交代清楚。

如果最终证明所有能够采取的措施都已经采取，那么社会自然能够理解这是一场人力无法阻挡的天灾。可如果其中存在失职，就应该有人承担责任。

这原本就是一个现代社会最基本的常识。

遗憾的是，在中国，公众往往连知道真相本身，都必须付出代价。

泥石流是天灾，删帖不是天灾；冰川崩塌是自然现象，处罚质疑灾情的网民不是自然现象；洪水冲毁建筑无法完全避免，但阻止社会了解灾难的完整情况，却完全是权力主动做出的选择。

所以我始终想问一句：

既然这是一场天灾，究竟有什么不能让中国人知道？

一个政权如果连自然灾害的真实情况都如此紧张，它所害怕的显然不是泥石流，而是真相本身。

天灾可以毁掉房屋和道路，但真正能够摧毁一个社会信任的，是灾难之后仍然有人试图把真相一起埋掉。

编辑：黄吉洲 校对：熊辩 翻译：戈冰

After a Natural Disaster, Why Still Cover Up the Truth?

— Viewing the CCP’s Disaster Governance Logic Through the Lens of the Gyirong Mudslide in Tibet

Author: Feng Reng

On August 26, 2026, a sudden, massive mudslide swept into the Gyirong Port in Shigatse, Tibet. Ice, rocks, mud, and torrents surged down from the high mountain gorges, engulfing vehicles, roads, and buildings in the blink of an eye. The footage emerging from the scene was shocking. Those images bore no political color; people saw only the insignificance and helplessness of human beings in the face of natural disasters.

Later, China’s natural resources authorities confirmed that this disaster originated from a high-altitude glacier collapse within Nepal’s borders. After the glacier collapsed, it formed a high-speed debris flow that swept down along the river valley, ultimately evolving into a highly destructive mudslide. From the currently available public professional data, this was first and foremost a severe natural disaster. On this point, there is nothing that needs to be avoided.

What I truly find incomprehensible is that familiar set of practices that reemerged after the disaster occurred.

As of 18:00 on August 29, Chinese official sources announced that on the Gyirong side, 16 people had been killed and 546 people were missing, including a large number of foreign nationals. Thousands of personnel from the People’s Liberation Army, Armed Police, Fire and Rescue Services, Public Security, and engineering rescue teams entered the disaster zone to participate in search and rescue operations. Those rescuers who risked their lives to search for survivors are, of course, worthy of respect. Any normal human being would hope that they could save one more person from the rubble and mud.

However, that rescuers are worthy of respect does not mean the government can operate without oversight. It is precisely in the face of such major disasters involving a vast number of lives that the public has even more of a right to know the full truth.

Comparative imagery published by Tibet Daily shows that after the mudslide struck on Wednesday (August 26), buildings at Tibet’s Gyirong Port had been razed to the ground.

When reading the materials compiled by China Digital Times regarding this disaster, my deepest feeling was not about the disaster itself, but a sense of familiarity that has remained unchanged for years. The disaster had just occurred, footage from the scene was circulating on the internet, and people began inquiring about casualties and the causes of the disaster; soon enough, situations arose where videos could not be forwarded, related articles were deleted, and online discussions were restricted. Subsequently, the police began issuing circulars on so-called “rumor-mongering” cases, with netizens subjected to administrative penalties or “educational reprimands” simply for discussing death tolls and disaster causes.

I do not support anyone intentionally fabricating casualty figures. Rumors during a disaster inflict fresh pain on family members and may also interfere with rescue efforts. However, if a government truly wishes to curb rumors, the most effective method should be to publish the facts in a timely, full, and transparent manner, rather than suppressing on-the-scene information while demanding that everyone accept only the official version.

When full videos cannot be normally circulated, when journalists cannot freely enter the scene, when articles discussing the disaster situation and risks are deleted, and when ordinary people raising doubts about death tolls may also be targeted by the police, then so-called “refuting rumors” is no longer merely about safeguarding facts—it has actually become a government monopoly over the right to interpret disaster information.

This is also a realization that has grown deeper in me over these years.

I lived in China for a long time in the past. Whenever a major accident or public event occurred, the ways in which we ordinary people obtained the truth were often extremely strange. State media said one thing, WeChat groups circulated another, and yet another version appeared on Weibo. When someone uploaded a video from the scene, everyone hurried to save it, because no one knew if it would still be visible half an hour later; when someone wrote an investigative article and it just started spreading in Moments, the link would no longer open by the next day.

Over time, many Chinese people even developed a deeply sorrowful piece of experience: the more rapidly something is deleted, the more everyone actually wants to know what was written inside.

This is not because Chinese people are particularly fond of believing rumors, but because long-term information control has already eroded the government’s own credibility.

If a government is accustomed to releasing only content favorable to itself, then even if it publishes the absolute truth one day, people will instinctively harbor doubts. This outcome was not brought about by foreign hostile forces, nor was it caused by a few netizens spreading rumors; it was caused by one instance of information blockade after another.

This disaster in Gyirong particularly warrants further questioning, because this is by no means the first time a severe disaster has struck here.

In July 2025, a severe flash flood and mudslide had already occurred in the Gyirong region, washing away the Sino-Nepal Friendship Bridge and leaving people missing on the Chinese side as well. Relevant geological research had also previously pointed out that the Gyirong region faces multiple geological risks, including high-altitude collapses, glacial lakes, and mudslides. Now, just over a year later, an even larger disaster has struck again. The public naturally has every reason to ask: After the last disaster, what risk assessments were actually conducted locally? Was the early warning system strengthened? Following the reopening of the port, was there a re-evaluation of the surrounding population density and infrastructure safety? Before such a massive ice collapse and mudslide took place, were there any detectable abnormal signals?

These questions are all very normal. What is regrettable is that in China, many normal questions quickly become so-called “sensitive issues” once they involve government responsibility.

Observing major disasters in China over the years, I have increasingly felt that what the CCP truly cares about is often not just how to conduct disaster relief, but also how to “manage the disaster.” So-called “disaster management” is not merely about mobilizing rescue personnel, restoring roads, and arranging supplies; an even more crucial component is managing public opinion: which numbers are released at what time, which images can appear, which journalists are allowed to interview, which family members can speak out, and which questions cannot be pursued further.

From the Wenchuan earthquake and the Wenzhou high-speed train collision, to the “July 20” torrential rainstorm in Zhengzhou, and to Wuhan during the pandemic, we have already witnessed all too many similar scenes. After a disaster strikes, what appears first are slogans like “leaders attach high importance,” “important instructions issued,” and “all-out efforts organized for rescue,” followed by rescue teams, touching stories, and so-called positive societal energy. As for why the disaster happened, whether anyone derelicted their duty, how many people actually died, and what doubts the family members harbor, these matters are often either pushed to the back or ultimately left unresolved.

The greatest danger of this model is that it gradually allows the disaster itself to become part of political propaganda.

How many people died is no longer purely a question of human life, but may be viewed as a figure that “affects social stability”; journalistic investigation is no longer just reporting, but may be regarded as “generating negative public opinion”; families demanding explanations may also be treated as disrupting disaster relief order; and netizens raising doubts are easily slapped with labels like “spreading rumors” or “passing on rumors.”

If a government must package even natural disasters according to political needs, then what is it truly maintaining?

Obvious not the deceased.

What truly needs to be maintained is the image of a regime that is eternally correct, a government that is eternally forceful, and a society that is eternally stable.

Living in the United States now and looking back at these matters in China, one of my deepest feelings is that whether a government actually possesses confidence is seen not in how powerfully it proclaims itself every day, but in whether it dares to face bad news.

A truly confident government does not fear publishing death tolls. No matter how grim the casualty figures are, they are not the government’s enemy; a truly responsible government does not fear journalists entering disaster areas, because journalists discovering problems is intrinsically part of public oversight; and a government that truly puts the lives of the people first should even less fear family members asking after the whereabouts of their loved ones.

If, after a mudslide occurs, the government—in addition to saving lives as an immediate priority—must also worry about which videos being circulated will “damage its image,” which articles need to be deleted, and which voices must be suppressed, then it becomes hard not to question how much truth actually lies behind the so-called “people first, life first.”

Natural disasters have no political stance.

Glaciers will not cease collapsing because of what kind of system a country practices, nor will mudslides alter their direction because a government promotes “social stability.” What a government can genuinely do after a natural disaster occurs is to minimize casualties as much as possible, fully disclose what has happened to society, and conscientiously investigate which losses could have originally been avoided.

If there was inadequate early warning before the disaster, the early warning system should be investigated; if problems existed in infrastructure planning, accountability for that planning should be pursued; if local officials were derelict in their duty regarding risk assessment, those responsible should bear the consequences. Only these are the things that truly ought to remain after a disaster.

Otherwise, once roads are repaired, bridges are rebuilt, leaders’ condolences have concluded, and the media has ceased reporting, the institutional loopholes that caused massive casualties will remain completely unchanged. A few years later, when another disaster strikes in a different place, the exact same story will play out all over again.

This is what is truly terrifying.

In my view, many disasters in China over the years share a common problem: after the natural disaster ends, a man-made “information disaster” is only just beginning. The actual floodwaters will recede and the mudslides will stop, but the deleted information, the silenced family members, and the unpursued responsibilities may remain buried underground forever.

Therefore, I believe that what should be left behind most from the Gyirong mudslide is not merely a few photos of rescue efforts, nor just a few slogans about “uniting as one.”

What should be left behind are the names of the deceased.

What should be left behind are the true and complete numbers of the missing.

Family members should be allowed to speak about what they went through.

Journalists should be allowed to investigate freely.

And a clear accounting should be given to society, item by item, of what preventive measures were actually taken locally following that disaster in 2025.

If it is ultimately proven that all measures that could be taken had indeed been taken, then society would naturally understand that this was a natural disaster beyond human capacity to prevent. But if dereliction of duty was involved, someone should bear responsibility.

This should originally be the most basic common sense in a modern society.

Regrettably, in China, the public often must pay a price even just to know the truth itself.

Mudslides are natural disasters, but deleting posts is not a natural disaster; glacier collapses are natural phenomena, but punishing netizens who question the disaster situation is not a natural phenomenon; floodwaters washing away buildings cannot be entirely avoided, but preventing society from understanding the full picture of a disaster is entirely an active choice made by power.

Therefore, I always want to ask one question:

Since this is a natural disaster, what on earth cannot be let known to the Chinese people?

If a regime is so anxious even about the true circumstances of a natural disaster, what it fears is obviously not mudslides, but the truth itself.

Natural disasters can destroy homes and roads, but what can truly destroy a society’s trust is that after a disaster, there are still people trying to bury the truth along with it.

Editor: Huang Jizhou Proofreader: Xiong Bian Translator: Ge Bing