✦   ✦   ✦

breaking news

News, read through The Primal Race
← 全部评论 · all commentary

一张假名单,一次真试探A Fake List, a Real Probe

国际 结构层 · 元暴力 Vox ↗ 2026-11-02 § 链接
新闻摘要2026年中期选举投票日前一天,特朗普政府的司法部要求最高法院下令加州将多达2万张选票单独隔离、暂不计数。加州洛杉矶县抽取名单中200人初审,发现近40%是在美国出生者,逾70%从未在加州投票;此前内华达类似名单上的185人最终全部被认定是公民。
The newsOn the eve of the 2026 midterms, Trump's Justice Department asked the Supreme Court to order California to segregate up to 20,000 ballots without counting them. A preliminary review of 200 names found nearly 40 percent were US-born and over 70 percent had never voted in California; in Nevada, all 185 people on a similar list turned out to be citizens.
选举争议不是事实之争,是谁有权定义事实的争夺。
Election fights aren't about facts. They're about who gets to define them.

这份诉讼最诚实的地方,是它连胜利都不想要。

Vox 把数字摆得很清楚:加州 2022 年中期选举有接近 1100 万人投票,司法部要隔离的是 2 万人。就算这 2 万人全部非法投票——洛杉矶县抽了 200 个名字初审,近 40% 是在美国出生的人,56% 的登记已注销或不再有效,超过 70% 从未在加州投过票——也翻不动任何一场选举。所以这不是在争结果,这是在争别的东西。

争的是「谁有权宣布什么是事实」。

第三章讲表达的武器化,核心是夺取话语权、夺取解释权、夺取「什么是事实」的制造权。这份名单就是一次制造事实的示范:先在 10 月 13 日向加州扔出一份 2 万人的名单,声称这些人「很可能」不是公民,然后拿着自己造出来的名单去法院,要求法院承认这份名单有资格决定哪些票要进特殊的一堆。名单的内容是假的不要紧,名单被法院受理这件事本身就是真的。

内华达已经演过一遍:先说 16000 名非公民在选民册上,真交名单时只剩 185 个,最后查明 185 个全是公民。误差率接近百分之百,但这不妨碍下一轮再扔一份 2 万人的名单。造假的成本是零,辟谣的成本是一个县一个县去抽 200 个名字核对。

这套机制不靠赢,靠的是让「隔离选票」这个动作在法院里被讨论过一次。今天它被驳回,明天它就是一个先例的种子:2028 年,同一套流程可以再走一遍,而州选举官员要在跑选举的同时应付一份又一份需要核对的假名单——报道里加州自己说,现在建隔离系统「会是一场后勤和行政的噩梦」,甚至可能让登记系统停摆。让它停摆,本身就是收益。

再看这一层下面是什么。Vox 指出,联邦法律要求州在选举前 90 天内不得系统性清除选民名单,而这份名单是选举前不到一个月送到的;最高法院不到两周前才在 Scanlan 案里说过,法院「通常不应在选举前夕改变选举规则」,而司法部的文件是在选举前夜递上去的。连特朗普自己任命的法官马克·贝内特都写了反对意见。这些不是疏忽,是故意把时间压到最紧——因为程序上站不住脚的一方,唯一能赢的方式就是让程序来不及。

到这里,共谋者的位置就出来了。递状纸的是司法部,签字的是一群律师,他们不是被骗的,他们清楚这份名单的成色。他们的回报不是这场选举,是向一个要求忠诚的权力中心证明自己还在牌桌上。第四章说共谋不分性别,看的是利益站队——这里的站队是职业生存,代价由洛杉矶县的登记员、由那 2 万个被点名的人、由每一个以后要证明自己「是公民」的选民来付。报道没有写这 2 万人里有多少人会在投票日被拦下、被要求出示证件、被邻居多看两眼。这个数字缺席,本身就是材料。

所以别把这件事读成一次失败的法律冒险。它是一次成功的叙事投放:从今天起,「2 万张可疑选票」会作为一个说法留在公共记忆里,哪怕它对应的 200 个抽样里没有一个确认的非公民。木已成舟,之后被证伪也没用——第三章写的就是这个。

真正的问题不是最高法院会不会驳回。是驳回之后,这份名单会不会被收进抽屉,等 2028 年再拿出来,换一个州,换一个数字,换一批签字的人。

The most honest thing about this lawsuit is that it doesn't even want to win.

Vox lays out the numbers plainly: nearly 11 million people voted in California's 2022 midterms. The Justice Department wants 20,000 ballots set aside. Even if all 20,000 had voted illegally — and LA County's preliminary review of 200 names found nearly 40 percent were US-born, 56 percent had canceled or inactive registrations, over 70 percent had never voted in California — it wouldn't flip a single race. So this isn't a fight over outcomes. It's a fight over something else.

It's a fight over who gets to declare what a fact is.

Chapter 3 describes the weaponization of expression: seize the discourse, seize the interpretation, seize the power to manufacture facts. This list is a demonstration. Send California 20,000 names on October 13, claim they are "likely" noncitizens, then take your own list to court and ask the court to treat it as grounds for deciding which ballots go into a special pile. The contents of the list can be false. The fact that a court entertains the list is real.

Nevada already ran the rehearsal: first 16,000 alleged noncitizens on the rolls, then a list of 185, then all 185 confirmed as citizens. An error rate near one hundred percent, and it did not stop the next list of 20,000. Fabricating costs nothing. Debunking costs a county sampling 200 names at a time.

The mechanism doesn't need to win. It needs "segregating ballots" to have been discussed in court once. Dismissed today, it becomes a precedent seed tomorrow: in 2028 the same pipeline runs again, and state election officials have to process fabricated lists while also running an election. California says building a segregation system now would be "a logistical and administrative nightmare" that risks shutting down the very system it needs to run the whole election. Shutting it down is itself the payoff.

Look at the layer beneath. Federal law bars states from systematically purging voter rolls within 90 days of a federal election; the list arrived less than a month out. Less than two weeks ago the Supreme Court said in Scanlan that courts "should ordinarily not alter the election rules on the eve of an election"; the brief was filed on the literal eve. Even Mark Bennett, a Trump appointee, wrote that Trump must lose. None of this is sloppiness. It is time compression as strategy — the side that can't win on procedure can only win by making procedure run out of time.

Here is where the co-conspirators sit. The signatories are lawyers who know exactly what this list is worth. Their payoff isn't this election; it's proving to a power center that demands loyalty that they are still at the table. Chapter 4 says complicity isn't gendered — it's interest alignment. Here the alignment is career survival. The cost falls on LA County's registrar, on the 20,000 named individuals, on every voter who will later have to prove they are a citizen. The report never says how many of those 20,000 get stopped at a polling place, asked for documents, looked at twice by a neighbor. That absence is itself the material.

So don't read this as a failed legal adventure. Read it as a successful narrative deployment: from today, "20,000 suspicious ballots" lives in public memory, even though the 200-name sample confirmed not one noncitizen. The wood is already carved; debunking it later changes nothing. That is Chapter 3 exactly.

The question isn't whether the Court dismisses it. It's whether the list goes into a drawer, to be pulled out in 2028 — a different state, a different number, a different set of signatures.