2026年8月8日｜当制度不再默认补件，判断必须发生在递交之前

作者： Xufen Tu 记录类型： Public Observation Record 主题： Human Judgment · Responsibility Boundaries · Verification

---

今天看到美国移民局关于 RFE 和 NOID 的政策更新，心里有一种很强的现实感。过去很多人会以为，材料如果不够完整，系统至少会给一次补件机会；证据如果没有说清楚，审理人员可能会发通知，让申请人再解释、再补充、再把缺口填上。可是现在这个政策提醒我，制度不一定会一直保留这种缓冲。当一个申请在递交时就缺少关键证据，或者证据本身无法支撑申请人的主张，审理人员可能不再先把问题退回来让人修正，而是直接作出拒绝。

这件事表面上是移民程序的变化，但我看到的是一个更深的结构问题：当一个系统不再默认给人第二次解释机会，判断就必须提前发生。不能等到结果出来以后，才发现材料之间没有对应；不能等到收到拒绝以后，才意识到某些证据只是存在，却没有真正支撑主张；也不能把“以后可以补”当成一种安全感。制度一旦把责任前移，人的判断也必须前移。

我以前会觉得，只要一个人的经历是真实的，长期努力是真实的，公开记录也是真实的，后面总有机会慢慢说明清楚。可是今天我更谨慎地理解这件事。真实本身很重要，但真实如果没有被清楚地整理、对应和表达出来，在审查系统里仍然可能变成“不足”。一个人做过很多事，不等于审理人员会自动理解这些事之间的关系；一个人有长期研究，不等于材料会自动证明研究的连续性；一个人有公开记录，不等于这些记录已经足够说明资格、贡献、责任边界和实际意义。

这不是说制度一定不公平，也不是说人应该害怕所有审查。真正需要看见的是，系统里的“记录”并不会自动替人完成判断。文件存在，不代表证据已经成立；链接存在，不代表关系已经清楚；经历存在，不代表价值已经被理解；材料很多，也不代表主张已经被证明。很多时候，问题不在于一个人完全没有事实，而在于事实没有被放在一个能够被审查者理解的位置上。

这也让我想到，为什么“递交前”的确认变得越来越重要。一个申请、一个论文记录、一个公开档案、一个项目说明，进入公共系统之前，都需要人先停下来判断：这是不是当前最清楚的版本？它有没有把核心事实放对位置？有没有把作者、时间、标题、来源和责任关系说明清楚？有没有把真实经历和所提出的主张连接起来？如果这些事情没有在前面完成，后面的系统不一定会替人慢慢修正。

我看到这条政策时，也想到很多普通人面对制度时的处境。很多人可能不是没有真实经历，也不是没有努力，而是不知道什么样的材料在制度里算“足够”。他们以为自己把东西交上去了，就已经说明了事实；以为自己说了真实经历，别人就会理解；以为系统会像人一样，在不清楚的时候主动追问。可是制度未必这样运行。制度更看重的是递交时已经呈现出来的事实、证据和对应关系。它可能不会自动理解沉默的部分，也不会自动替一个人补全没有表达清楚的地方。

这让我更清楚地意识到，人的判断不是只发生在重大选择时，也发生在每一次正式提交之前。一个人要判断什么可以进入系统，什么还不能进入；什么已经清楚，什么还只是自己心里明白；什么是事实，什么只是解释；什么已经有证据支撑，什么还只是未来计划。判断越晚发生，风险越容易被系统放大。判断越早发生，人才更可能在进入审查之前发现缺口。

这和我的研究主线也有关系。一直以来，我关注的不是让系统替人做更多决定，而是人在复杂系统面前如何保留判断、确认和责任边界。今天这件事让我看到一个很现实的例子：当制度变得更少依赖补救程序，人的前置判断就不再是形式，而是决定结果的一部分。不是所有记录都会被重新解释，不是所有缺口都会被提醒，不是所有不清楚都会被退回来修改。系统可以处理文件，但系统不会自动替人承担材料不完整的后果。

对我来说，这也是一个提醒。未来无论是研究发布、公开记录，还是任何需要被外部审查的材料，都不能只追求“有”。有文章、有链接、有文件、有版本，都只是第一步。更重要的是，这些东西之间是否清楚，是否能对应，是否能让外部的人理解它们为什么重要。真正的责任不是等别人指出问题后才补，而是在提交之前尽量完成自己的判断。

我现在更谨慎地理解“机会”。有时候机会不是被拒绝在门外，而是在进入门口之前就因为材料不清楚、判断不完整、证据没有对应而失去了被认真理解的可能。制度不再默认补件以后，很多事情都会变得更严肃。一个人不能把未来的解释当成今天的证据，也不能把真实经历本身当成已经被系统理解。

今天留下的问题是：当越来越多制度把判断和责任前移到提交之前，普通人和独立研究者怎样才能在不夸大、不包装、不暴露不该公开内容的前提下，让真实经历、公开记录和研究连续性被清楚看见？

---

August 8, 2026 | When a System No Longer Defaults to Supplementation, Judgment Must Happen Before Submission

Today I saw the USCIS policy update concerning RFEs and NOIDs, and it gave me a strong sense of reality. In the past, many people may have assumed that if materials were incomplete, the system would at least provide a chance to supplement them. If evidence was not clear enough, an officer might send a notice and allow the applicant to explain, add documents, and fill the gaps. But this policy reminded me that a system may not always preserve that buffer. When an application lacks key evidence at the time of filing, or when the evidence does not support the applicant’s claim, the officer may not first return the problem for correction. The case may instead be denied directly.

On the surface, this is a change in immigration procedure. But what I see is a deeper structural issue: when a system no longer defaults to giving people a second chance to explain, judgment must happen earlier. One cannot wait until the result appears to discover that the materials did not correspond to one another. One cannot wait until a denial arrives to realize that certain evidence merely existed but did not actually support the claim. One cannot treat “I can supplement it later” as a form of safety. Once a system moves responsibility forward, human judgment must also move forward.

I used to think that as long as a person’s experience was real, their long-term effort was real, and their public records were real, there would eventually be a chance to explain everything clearly. But today, I understand this more carefully. Reality itself matters, but if reality is not clearly organized, connected, and expressed, it may still appear insufficient inside a review system. A person may have done many things, but that does not mean an officer will automatically understand how those things relate to one another. A person may have long-term research, but that does not mean the materials automatically prove continuity. A person may have public records, but that does not mean those records already explain eligibility, contribution, responsibility boundaries, and practical meaning.

This does not mean the system is necessarily unfair, nor does it mean people should fear every review process. What needs to be seen is that records inside a system do not automatically complete judgment on behalf of human beings. A file existing does not mean evidence has been established. A link existing does not mean the relationship is clear. An experience existing does not mean its value has been understood. Having many materials does not mean that a claim has been proven. Very often, the problem is not that a person has no facts at all, but that the facts have not been placed where a reviewer can understand them.

This also makes me think about why confirmation before submission becomes increasingly important. Before an application, a paper record, a public archive, or a project statement enters a public or institutional system, someone still needs to pause and judge: is this the clearest current version? Are the core facts placed correctly? Are the author, time, title, source, and responsibility relationship clear? Are real experiences connected to the claims being made? If these things are not completed at the beginning, the later system may not slowly correct them for the person.

When I saw this policy, I also thought about how many ordinary people face institutions. Many people may not lack real experience or effort. They may simply not know what kind of material counts as “enough” inside a formal system. They may believe that once they submit something, the facts have already been shown. They may believe that once they tell a true experience, others will understand. They may believe that a system will ask follow-up questions like a human being when something is unclear. But institutions may not operate this way. They often focus on what has already been presented at the time of submission: facts, evidence, and their correspondence. They may not automatically understand the silent parts, nor automatically complete what has not been clearly expressed.

This made me see more clearly that human judgment does not occur only during major life decisions. It also occurs before every formal submission. A person needs to judge what can enter the system and what cannot yet enter; what is already clear and what is only clear inside one’s own mind; what is fact and what is interpretation; what is already supported by evidence and what remains only a future plan. The later judgment happens, the more easily risk can be amplified by the system. The earlier judgment happens, the more likely a person is to notice gaps before entering review.

This also connects to my ongoing research line. I have not been concerned with making systems decide more on behalf of human beings. I have been concerned with how human beings preserve judgment, confirmation, and responsibility boundaries when facing complex systems. Today’s policy offered a very real example: when institutions rely less on remedial procedures, prior human judgment is no longer a formality. It becomes part of the outcome. Not every record will be reinterpreted. Not every gap will be pointed out. Not every unclear point will be returned for revision. A system can process documents, but it does not automatically bear the consequences of incomplete materials for the person who submitted them.

For me, this is also a reminder. Whether it is research publication, public recordkeeping, or any material that may be reviewed externally, it is not enough to simply have things. Having articles, links, files, and versions is only the first step. What matters more is whether these things are clear, whether they correspond to one another, and whether an outside reader can understand why they matter. Responsibility does not begin only after someone else points out the problem. It begins when a person tries to complete judgment before submission.

I now understand opportunity more cautiously. Sometimes opportunity is not refused only at the door. Sometimes it loses the chance to be properly understood before it even enters, because the materials are unclear, the judgment is incomplete, or the evidence does not correspond to the claim. When institutions no longer default to supplementation, many things become more serious. A person cannot treat future explanation as today’s evidence, nor treat real experience itself as something already understood by the system.

The question I am left with today is this: when more institutions move judgment and responsibility to the moment before submission, how can ordinary people and independent researchers make real experience, public records, and research continuity clearly visible without exaggerating, packaging themselves, or exposing what should remain private?