This is a direct continuation of my long analysis “Mukherji v. Miller three months later”. That post contains all the background: the five grounds on which Judge Bataillon struck down the Final Merits, a portrait of attorney Brian Green and his “$11k per suit” model, and the AAO vs. federal court breakdown. I won’t retell any of that here. This post contains only what happened afterward, the court documents themselves (I publish both in full) and an analysis of “what’s next.”
I ended that post with a checkpoint: May 11, the deadline for the government’s first written brief in the Eighth Circuit. The idea was that the brief would reveal how USCIS intended to fight.
So. USCIS won’t be fighting at all.
Yesterday, June 10, 2026, the government withdrew its own appeal. It didn’t lose it, didn’t take it to argument, didn’t file a single brief. It just showed up in court and said: we changed our mind. By the evening the Eighth Circuit closed the case, issued its mandate, and Judge Bataillon’s decision — the one that found the Final Merits Determination unlawful — became final and nonappealable.
In that earlier post, in the section “A Scare for USCIS,” I wrote that in the corridors people were saying USCIS might not want to continue because a loss in the Eighth Circuit would open the road to the Supreme Court, where the result could be “Final Merits unlawful nationwide, period.” I’ll be honest: when I wrote that, it looked like a nice theory. Yesterday it stopped being a theory.
I spent the day in PACER and court databases, purchased documents, and below is the full picture: what exactly happened, a minute-by-minute timeline, why the government gave up, and — most importantly — what this changes for those with RFEs, denials, or pending petitions. Spoiler: it changes less than people want, but more than it looks.
“USCIS withdrew the appeal — so is the Final Merits gone? Will all cases now be approved under three criteria?”
No. And that “no” is the most important part of this post. The government gave up one case to save the whole system. This was not capitulation but a retreat to prearranged positions. Let’s unpack it.
Contents
- One day that ended everything: June 10 timeline by the minute
- The brief that never was: chronology of a quiet surrender
- I publish both documents: the 115-word surrender and the court order
- Why the government folded. The arithmetic that explains it
- “They’ll be swamped with lawsuits now?” They will. And that suits them
- What USCIS is doing right now (checked last night)
- Brian Green is no longer solo: a team of former DOJ lawyers
- Green’s post and the comments: the bar reads the move the same way
- The $11k test: the EAJA window opened
- What this changes for your petition, RFE, and denial
- Questions from the chat: class suits and other “can we”s
- How to read this case yourself (and why the district will block you)
- What to watch next
One day that ended everything: June 10 timeline by the minute
Federal courts rarely let you tell a story by the minute. Here they do, and it’s worth it.
9:56 a.m. Government files withdrawal
DOJ attorney Brian V. Schaeffer (Office of Immigration Litigation) filed in the Eighth Circuit a document titled “WITHDRAWAL OF APPEAL (ASSENTED TO).” Four pages, which in substance amount to one paragraph.
3:02 p.m. Court grants
The Eighth Circuit clerk the same day issues a judgment: the appeal is terminated under FRAP 42(b), a voluntary dismissal at the appellant’s request. Signed: Susan E. Bindler, Clerk. No judges were needed for this: the clerk issues such orders under Rule 27A(a).
3:06 p.m. Mandate
Four minutes later the mandate issues, the formal act that finally returns the case to the district court. For an ordinary merits decision, the mandate is held at least seven days. Here four minutes. That’s it: case 26-1578 ceases to exist.
By evening the district court docket in Nebraska had two final entries: a copy of the appellate court order and the mandate. The procedural history of Mukherji v. Miller is over. Completely. The period was placed not by a judge but by the government itself.
The brief that never was: chronology of a quiet surrender
Now rewind, because the most interesting part is not the withdrawal itself but its trajectory. It reads like a detective story where the criminal spends the whole novel steeling himself and in the final chapter turns himself in.
March 27. Notice of appeal on day 58 of 60
The government files a notice of appeal two days before the deadline. This is the classic “protective” filing: a notice costs nothing and doesn’t bind anyone, it just preserves the option. The decision to fight or not can be made later.
March 30. Schedule: brief due May 11
The Eighth Circuit sets the schedule. That same date I mentioned in my previous post.
April 13. A show of battle-readiness
The government files a statement of issues — a list of questions it intends to challenge — and agrees with Mukherji’s team on joint document appendices. Outwardly it looks like preparation for war.
May 4. One week before the deadline, a request for extension
Instead of a brief DOJ asks for another 30 days, until June 10. The clerk grants it. Standard story, no one noticed anything unusual.
May 5–June 10. Silence
Thirty-six days with nothing filed in the docket. No brief, no appendices, no procedural letters. And on June 10, the last day of the extended deadline, instead of a brief comes a withdrawal.
This detail (the withdrawal filed precisely on the brief deadline) is the most telling in the whole story. This does not look like a spontaneous decision. This looks like the climax of an internal struggle decided at the last hour.
A short primer on how this is structured internally. A rank-and-file DOJ attorney does not decide whether the United States will defend an agency policy in an appellate court. He can file a notice of appeal almost automatically to preserve options. But the substantive appellate brief is the government’s official position, and sanction for it is given by the Office of the Solicitor General. The May 4 extension meant “higher-ups are still thinking.” The June 10 withdrawal means “higher-ups said no.” Someone in DOJ and DHS wanted to fight — otherwise there would have been no April statement of issues. And that someone lost the internal dispute.
What that means in bureaucratic terms
The government kept the appeal as a scare for three months — for Mukherji, whose green card hung in the balance, and for everyone thinking “should I sue over the same scheme?” And when it came time to put pen to paper, under signature, to defend the Final Merits before three appellate judges, it refused to do so. There is no brief anywhere defending the procedure. That fact alone has a long shelf life.
I publish both documents: the 115-word surrender and the court order
There are exactly two documents that ended this story. I obtained both (the motion sits in the open RECAP archive, the judgment I purchased in PACER) and publish them in full, original and translation. The PDFs are attached here: motion to withdraw the appeal and Eighth Circuit judgment.
And first, a detail you cannot invent. The motion included the required certificate of compliance in which the DOJ attorney certified the document’s word count. Quote: “The motion contains 115 words.”
Document one. “Withdrawal of Appeal (Assented To),” June 10, 2026
The entire substantive part, without the heading and signatures:
“Pursuant to Fed. R. App. P. 42(b), Respondents-Appellants respectfully move this Court to enter an order dismissing the above-captioned appeal.
In support of this motion, the government states that it has elected not to pursue this appeal, and that this decision has come before either side filed a brief in the case.
On June 9, 2026, counsel for the government conferred with counsel for Plaintiff-Appellee, and they assent to this motion.
Accordingly, the Government moves to withdraw and dismiss its appeal from the district court's order pursuant to Fed. R. App. P. 42(b), with prejudice, with each party bearing its own costs and fees of the appeal”.
“Pursuant to Rule 42(b) of the Federal Rules of Appellate Procedure, Respondents-Appellants respectfully request that the Court enter an order dismissing the above-captioned appeal.
In support of this motion, the Government states that it has elected not to pursue this appeal, and that this decision was made before either party filed a brief in the case.
On June 9, 2026, counsel for the Government conferred with counsel for the Plaintiff-Appellee, and they assented to this motion.
Accordingly, the Government moves to withdraw and dismiss its appeal from the district court’s order pursuant to Rule 42(b) of the Federal Rules of Appellate Procedure, with prejudice, with each party bearing its own costs and fees of the appeal”.
The signatures on the document are telling: not only the rank-and-file trial attorney Brian V. Schaeffer, but also Brett A. Shumate, Assistant Attorney General, head of the entire Civil Division of DOJ, and Samuel P. Go, Assistant Director. A decision at that level is not made in a junior attorney’s office; it was issued from the top of the Department of Justice.
Note what is not in the document. No reason is given. Nothing like “the parties reached agreement,” nor “the agency is reconsidering the policy,” nor “the petition was approved and the dispute is resolved.” Simply “elected not to pursue.” The government does not explain its motives, and that is deliberate: any written explanation could be quoted against it later.
Three phrases that look technical but are actually substantive:
- “Assented to”: Mukherji’s team assented to the withdrawal on June 9. Of course they did: it’s a complete win for them without a single extra move.
- “With prejudice”: withdrawal with prejudice. The government cannot change its mind and refile the appeal. Never. The door is closed and welded shut.
- Each party bears its own costs “of the appeal”: the phrasing is carefully limited to appellate costs. The question of district-court fees (the EAJA discussed below) remains open.
Document two. Eighth Circuit Judgment, June 10, 2026
First, how this entry appears in the Eighth Circuit’s official electronic system (CM/ECF, Orders/Judgments), because one screen tells the whole denouement:
It reads: Case 26-1578, entry type Judgment, filed and entered June 10, 2026, 15:02:26. In the Description column the story in one line: “Appellant’s motion to dismiss the appeal [5649499-2]; appeal dismissed under Fed. R. App. P. 42(b); mandate to issue forthwith.” The bracketed numbers are internal document numbers. [5649499-2] is that 115-word government motion, [5649760] is the judgment itself. Those document numbers will stay in the docket forever: the bureaucratic machine, unlike agency policy, forgets nothing.
And a particular delight: the PACER receipt at the bottom of the screen shows the search was billed as one billable page — ten cents. The finale of the government’s three-month appeal in a case that could have overturned EB-1A sold for the price of gum.
And here is the document itself, one page. The entire text in substance:
“JUDGMENT
Appellant's motion to dismiss the appeal is granted. The appeal is hereby dismissed in accordance with the Federal Rules of Appellate Procedure 42(b).
The Court's mandate shall issue forthwith.
June 10, 2026
Order Entered Under Rule 27A(a): Clerk, U.S. Court of Appeals, Eighth Circuit. /s/ Susan E. Bindler”.
“JUDGMENT
Appellant’s motion to dismiss the appeal is granted. The appeal is hereby dismissed pursuant to Rule 42(b) of the Federal Rules of Appellate Procedure.
The Court’s mandate shall issue forthwith.
June 10, 2026.
Order Entered Under Rule 27A(a): Clerk, U.S. Court of Appeals, Eighth Circuit. /s/ Susan E. Bindler”.
Note: the court act that buried the government’s three-month appeal bears no judge’s signature — only the clerk’s. Rule 27A(a) allows the clerk to grant uncontested procedural motions where there is no dispute, and there was none: both sides wanted the same outcome. This is how cases end when one side simply stands up and leaves.
Document three. The mandate: the “key” that returned the case downwards
Four minutes after the judgment the Eighth Circuit issued the mandate. In the district docket it appeared as entry No. 30:
“MANDATE from USCA - 8th Circuit (26-1578). In accordance with the judgment of June 10, 2026, and pursuant to the provisions of Federal Rule of Appellate Procedure 41(a), the formal mandate is hereby issued in the above-styled matter”.
Simply put: while a case is on appeal it is legally “upstairs,” and the district court does not have full control over its decision. The mandate is the transfer by which the appellate court returns the case downward. Usually it’s held for at least seven days (to allow motions for rehearing). Here it issued in four minutes because there was nothing to rehear and no one to rehear it with. From that moment Judge Bataillon’s decision operates in full force, and it is backed not just by the district court but by the whole procedural vertical.
Why the government gave up. The arithmetic that explains it
The most common question I was asked yesterday: how does this make any sense? The government had the Fifth Circuit in its pocket (Amin v. Mayorkas, 2022), the argument “our colleagues agree,” and practically unlimited pockets. Why not fight?
Now look at the board through DOJ eyes. It’s a simple two-column table.
Price of capitulation
One approved I-140. Possible EAJA bill: tens of thousands of dollars. One unpleasant, but district decision remains in the books as persuasive authority. That’s it. The damage is localized to one journalist’s case.
Price of losing in the Eighth Circuit
An authoritative precedent for seven states: Nebraska, Iowa, Missouri, Minnesota, both Dakotas, Arkansas. Final Merits would automatically be dead for every applicant in the circuit, without any suits. Plus a direct conflict with the Fifth Circuit, meaning an almost guaranteed Supreme Court fight where agencies are in a weak position post-Loper Bright. Potential conclusion: “the procedure is unlawful nationwide.”
And the key question: what were the odds of winning? The honest answer: poor. Bataillon provided five independent grounds: the government would have had to overturn all five; a failure on any one meant losing. And the core of the decision is not a technical legal theory but a historical fact: Final Merits was introduced in 2010 by memorandum without public notice. That fact cannot be refuted with a brief — it simply happened. The only shield, Amin, was forged in 2022 under the still-living Chevron deference, and by 2026 even those who like it wave it cautiously.
So the choice: almost certain loss with catastrophic systemic consequences, or a certain “loss” with consequences confined to a single case. DOJ chose the latter. I would too.
This is called non-acquiescence; it’s a strategy, not panic
U.S. agencies have a long, refined practice: lose individual cases and not admit a systemic loss. Implement the decision for a specific person and continue applying the same policy to everyone else, because a district court decision binds the agency only in that single case. USCIS yesterday executed a textbook move from this playbook: sacrifice a pawn (one petition) to not expose the queen (Final Merits across seven states and eventually nationwide through the Supreme Court).
I’ll single out something that warms me personally. In the previous post I listed three routes the government could take in a brief: “Loper Bright does not apply,” “this is guidance not a rule,” “this is a unique case.” The answer turned out to be a fourth that I hadn’t listed: do not write a brief at all. Because any of the three arguments, written down and lost, would become precedent at the appellate level. Robinson Immigration warned exactly that back in February: the road is dangerous, a loss creates a stronger precedent than a district decision. It seems DOJ read the same warnings.
“They’ll be swamped with lawsuits now?” They will. And that suits them
Counterintuitive part. You’d think: the decision stood, the path is blazed, Green sends invites, now USCIS will be flooded with suits over FMD — so why give up?
I dug into court databases and counted the “flood.” Method: full-text search of the RECAP archive, which indexes federal-court documents. The phrase “Mukherji v. Miller” appears in documents in just three other cases. Here’s the entire “flood,” with dockets:
- Muthineni v. Seng (Maryland, 8:26-cv-00518, filed Feb. 8), a suit by Brian Green himself, Mukherji’s decision attached as an exhibit. No decision yet; case pending.
- Cannabrava de Sousa v. Selby (Nebraska, 4:26-cv-03056, Feb. 18, attorney Marcelo Gondim), filed in the same circuit to catch the same precedent field. Government asked for an extension to answer in May. No decision.
- Kleinova v. USCIS (Alaska, 4:26-cv-00021, Apr. 3), turns up in a search for “final merits determination” + EB-1A and looks to be on the same theme. No decision.
Why so few? Because the entry filter is brutal: $11k to an attorney, 12–24 months waiting, and a judge looks only at the administrative record that was before USCIS; a weak RFE response won’t be saved in court. Few pass through that sieve.
And now why even a real flood doesn’t frighten USCIS:
Each suit is individual
No class actions (Green deliberately avoids them — I quoted his logic in the first part). One hundred suits are one hundred separate cases, each which the government controls individually.
The government always has an emergency exit
Any case that starts to look dangerous (a picky judge, a looming loud decision) can be diffused in a week: approve the petition before a decision, and the case becomes moot. Plaintiff is happy, no precedent. This lever works in each of the hundred cases.
Individual losses are cheaper than a precedent
Lose twenty district cases = twenty approved petitions plus attorneys’ fees. Unpleasant but operational expense. One loss at the appellate-circuit level = thousands of automatic approvals without lawsuits. The math is obvious.
So USCIS’s strategy is now clear and almost elegant: the system keeps working for those who don’t sue. Those who sue are bought off one by one. And a precedent that could break the whole system never arises, because the government no longer takes such cases to appellate courts. Mukherji taught them this lesson: they took one to appeal and nearly lost everything.
For us there’s a silver lining, discussed further in the practical section below. But first, what’s happening inside the fortress itself.
What USCIS is doing right now (checked June 11)
In the previous post I pictured “the house is burning and the tenants sip tea.” Report: the tea-drinking continues. I checked all official channels twice: the evening of the 10th (the day of the withdrawal) and again today before publishing.
- Policy Manual untouched. Volume 6, Part F, Chapter 2, the one with the two-step analysis, is marked “Current as of May 08, 2026.” The two-step procedure remains (“Officers should use a two-step analysis to evaluate the evidence”), the phrase “final merits” appears five times in the chapter, and there are zero mentions of Mukherji. The last overall update to the manual, May 8, concerned deferred action and is unrelated to EB-1A: the only fresh use of the word “extraordinary” in updates refers to prosecutorial discretion, not ability. Manual update feed: for all of 2026 only two entries, both unrelated to EB-1.
- Newsroom is silent. Recent releases in the USCIS Newsroom: asylum office news in San Antonio (June 5), TPS extension for Lebanon (May 28), denaturalization of twelve people (May 14). Nothing about EB-1A, Kazarian, or Mukherji in May or June.
- AAO continues not to notice. In the AAO non-precedent decisions database a search for Mukherji finds nothing. Caveat: non-precedent AAO decisions are published with delay, so “not citing” here means “no visible signs yet.”
- Service centers keep stamping. From fresh RFEs and denials discussed in the community, Final Merits is being applied as if nothing happened. I won’t link to these: they are live community cases, not public documents, so treat this as observation, not proof.
And the most interesting: RIN 1615-AC85, the proposed rulemaking “Petition for Immigrant Worker Reforms” that would codify the two-step procedure through normal public rulemaking. In my previous post I recommended watching the Federal Register. I checked: the NPRM has not appeared. According to the official schedule, the NPRM was due January 2026 — it’s five months late.
Why a delayed NPRM and the withdrawn appeal are the same story
Combine two facts. The government refused to defend the old method of implementing Final Merits (the 2010 memo issued without notice). And simultaneously is preparing a new method, the same thing but via proper procedure. Why spend the Solicitor General’s political capital defending a memo you plan to replace by rule? Once the NPRM is published and a rule is adopted, Mukherji’s main procedural argument evaporates for all future cases. What remains is a merits dispute (e.g., “the law doesn’t require perpetual top-of-field status”), which the government will prefer to litigate on a fresh, procedurally clean foundation.
So don’t be fooled by the silence. The withdrawal is not “USCIS surrendered.” It is “USCIS moved the fight to a field where it has better odds.”
One more shade on the silence: there is no formal confirmation yet that Mukherji’s petition has already been approved. The order is final; failing to comply would be direct contempt of court, so approval either already happened quietly or will happen within days. But publicly USCIS has not said so.
Brian Green is no longer solo: a team of former DOJ lawyers
Remember my portrait from the first post: a solo practitioner from Denver, a thousand quiet suits, one loud win, Sunday webinars, $11k per suit? That portrait aged in a month in every way except price.
First. On appeal Mukherji already had three attorneys: Green himself, Jessica Ariela Dawgert (Ariela Lake Law & Consulting) and Sarah L. Vuong. That’s visible in the Eighth Circuit docket.
Second. On the day of the withdrawal Green and Ariela Lake issued a joint press release: “USCIS Withdraws Appeal of the Mukherji EB-1A Decision.” And in it, beyond the headline, sits an expansion of the funnel worth noting:
“If you received a denial on an EB-1A or EB-1B I-140 petition between the adoption of the Kazarian policy on December 22, 2010 and the present, you may challenge that denial in federal district court”.
Third, and most telling: who Ariela Lake are. I checked LinkedIn profiles of both partners, and the story brightened. The firm was founded in January 2026 — the same month the Bataillon decision came out. It was founded by two lawyers who together gave the Department of Justice more than thirty years:
- Jess Dawgert: nearly 17 years at DOJ. Initially Trial Attorney and Senior Litigation Counsel in the Office of Immigration Litigation (OIL) — that’s exactly the office that defends USCIS in federal courts. Toward the end of her career she served as Associate Deputy Attorney General, handling the immigration portfolio in the Deputy Attorney General’s office. She left DOJ in summer 2025.
- Sarah Vuong: around 15 years at OIL: Trial Attorney, Senior Litigation Counsel, and since December 2023 Assistant Director. Also nearly two years as counsel inside USCIS (Office of the Chief Counsel). She left in August 2025.
Now open the government’s withdrawal motion (above) and look at the block of signatures: Brett A. Shumate, Assistant Attorney General; Samuel P. Go, Assistant Director; Brian V. Schaeffer, Trial Attorney. The Assistant Director, Office of Immigration Litigation, is the title Sarah Vuong left ten months ago. The government’s capitulation is signed by her former office, by her former title. People who for years decided which appeals the government would defend and which it would quietly let go now stand on the other side; they don’t need to guess what the other side thinks — they were it.
And a detail for the future that seems strategically most telling. Three weeks ago Ariela Lake posted on LinkedIn that they had obtained six recent admissions to federal district courts: Northern California, Maryland, Nebraska, Eastern NY, Southern NY, Northern Texas. Map that: Muthineni is already in Maryland, Cannabrava in Nebraska, and Cyrus Mehta forecasted a wave of suits in New York and Northern California back in February. This is not a random list of jurisdictions. It’s a map of the future offensive, posted three weeks before the government’s surrender.
The irony is that yesterday’s day made that marketing simultaneously stronger and more honest. Previously the formula was “the only attorney who beat Kazarian Step 2.” Now it reads “the attorney who beat Kazarian Step 2, and the government refused to contest it.” From a sales perspective this is bulletproof. From the law perspective, it still isn’t an appellate-level precedent, and now it never will be. What the bar thinks about this is in the next section — a treat.
Green’s post and the comments: the bar reads the move the same way
Green announced the news on LinkedIn, and that post plus the comments are worth separate analysis because the comments turned into an impromptu poll of the U.S. immigration bar. In the first 16 hours: 501 reactions, 37 comments, 33 shares. Here’s the post:
“Big News! USCIS withdrew its appeal of the Nebraska district court’s decision in #Mukherji, which found that the #Kazarian policy was adopted unlawfully in 2010. Congratulations to my appellate co-counsel Jess Dawgert and Sarah Vuong! Details in our #press-release! #SueUSCIS #EB-1A #LoperBright”
Who showed up in the comments: and here’s the interesting part: not applicants but almost exclusively practicing immigration attorneys — firm partners, a former USCIS appeals officer, EB-1A specialists. Reading them in sequence gives a rare picture: the whole bar, without coordination, reads the government’s move the same way. Here’s the gist, my translations.
Sharon L., an EB-1A attorney (NY/NJ), phrases exactly what I analyzed above:
“Looks like the government dropped the appeal to avoid a loss: a Mukherji win at the court of appeals would have made Kazarian unlawful at the circuit level, i.e., precedent. The consequences for EB-1 applicants would have been monumental!”
Toni Xu, immigration attorney, writes with barely hidden disappointment:
“Interesting move. The government is avoiding creating an undesirable precedent that would undo Kazarian? I was hoping they’d lose on appeal and you’d knock Kazarian entirely :)”
Mohammad Shair replies in the thread to those who concluded Final Merits is dead — note, that phrase is the literal takeaway of my analysis:
“But the final merits stage is not dead. A district court ruling binds only that particular case. The government acted strategically: by withdrawing the appeal USCIS avoided an Eighth Circuit decision that would have created binding precedent across multiple states.”
Ron Matten: the thread’s best joke, though it’s not much of a joke:
“Looks like I’ll have to rewrite the blurbs in my EB-1A cover letters: ‘According to Mukherji, the Nebraska district court found application of Kazarian unlawful, and because USCIS is too scared to appeal, the officer should treat Mukherji as controlling precedent… otherwise Brian Green will sue you again’ 😉”
Evan J. Law: and this is not a joke at all. He’s a former USCIS appeals officer — someone who used to review appeals inside the agency:
“EB-1A regulations already embed concepts of recognition and notoriety in the evidentiary structure of the criteria. <…> If the introductory paragraph and the provision on comparable evidence require demonstrating eligibility, what role should a separate subjective ‘final merits’ review play? It will be fascinating to watch Mukherji’s influence on future decisions.”
Olga Prygoda shows how fast the decision becomes a working tool — the same day:
“This made my day — a very, very long Wednesday full of EB-1A argument breakdowns! Happy to discuss how to carefully weave this into our responses to NOIDs.”
And the thread’s central question was posed by Timothy D’Arduini in six words: ‘Can they make it precedent, Brian Green?’ The answer, as we’ve already parsed: no. That’s exactly why the government withdrew the appeal.
What the winning team itself says
Most telling is Jess Dawgert’s phrasing in her repost: “Although we will not get an Eighth Circuit decision, the withdrawal is important. The district court’s ruling remains a useful tool in the ongoing fight against unlawful determinations.” Note the carefulness: “tool,” not “precedent.” Even the winners don’t sell this as cancelling Final Merits, unlike how this news will likely be retold in Telegram channels.
In sum: not a single lawyer in the thread read this as “USCIS admitted error and surrendered.” Everyone reads it as a strategic retreat to preserve the system. When fifty practitioners, a former USCIS officer, and ex-DOJ lawyers converge on one reading, consider that reading established.
The $11k test: the EAJA window opened
In the previous post I discussed Green’s promise to clients: $11k per suit, which if victorious would be recovered under the Equal Access to Justice Act. I warned: recovery is not automatic; the court must find the government’s position not to have been “substantially justified,” and damages are limited.
Well, that promise just got its public test — on its flagship case.
Under EAJA a fee application is filed within 30 days after a decision becomes “final and nonappealable.” While the appeal hung the window was closed. On June 10, with the mandate, it opened. And recall the careful phrasing in the withdrawal: each party bears its own costs “of the appeal”? That does not close district-court EAJA.
His argument is now excellent: it’s hard to claim the government’s position was “substantially justified” if the government itself abandoned defending it and wrote no brief. Watch the district docket; any movement will be visible.
What this changes for your petition, RFE, and denial
Now practicalities. An honest layout: what changed yesterday and what didn’t.
What DID NOT change. Final Merits is alive. The officer who reviews your petition next week uses the same Policy Manual as last month. Mukherji remains a single district-court decision: it has no binding force outside that case, and now it never will, because the appellate court will not weigh in. If you got a denial based on FMD today, it is not automatically invalidated.
What changed. Three things, all significant.
The decision became invulnerable
A month ago any RFE response quoting Bataillon carried an embedded risk: “what if the Eighth Circuit reverses?” That risk died on June 10 at 15:06. The decision is final, unappealable, and will remain in the books forever. The statement “the court finds nothing in the law supporting such a requirement” now comes from a decision the government could have erased but consciously chose not to.
A new argument appeared that wasn’t there a month ago
“The government was given the chance to defend the Final Merits procedure in the appellate court and withdrew its own appeal without filing a single brief.” This is not interpretation; it is a procedural fact with docket numbers. In an RFE response, an AAO brief, or a federal lawsuit it works as a quiet, very uncomfortable pressure: the agency is applying a procedure it refused to defend in court.
The math of a federal suit improved
We now know how the government behaves under pressure in this area: it seeks extensions, stalls, and then yields when a precedent smells imminent. For a high-quality case that means the scenario “USCIS approves a petition rather than allow a decision” is not a theory anymore but an observed behavior. Federal litigation is now a usable tool for quality cases. A weak profile and a weak RFE response still won’t be saved; the judge looks only at what the officer saw.
And a standing rule that, after yesterday, became only more important: write every RFE reply as if a federal judge will read it. The probability of that happening has increased, not decreased, over the past six months.
Questions from the chat: what can and can’t be done
The news spread faster than I finished this post and the community chat is full of questions. I’ll answer the most common ones here so I can reference them later.
“Maybe we should pool and file a class action? And contact these attorneys? 😄”
The temptation is understandable: many of us, similar denials, chip in, swarm them. But a class action will not happen, for three reasons.
First: Green’s own position. I quoted his communications earlier: he purposefully does not bring class actions. His logic: class suits are too complex and costly, and if you hit the wrong judge you lose for everyone. Individual suits cause maximum trouble for the government and give the best chances to plaintiffs. After June 10 this logic only strengthened; we now know the government surrenders to single strong cases.
Second: legal. An APA suit challenges a specific administrative decision and the administrative record behind it: your criteria, your RFE, your denial. Each record is unique. For a class you need common questions of fact and law; here the only commonality is procedure, and even then each plaintiff must individually litigate the specific denial under the administrative record. Class aggregation simply doesn’t fit well here.
Third: tactical. One big suit is one point of failure: wrong judge, wrong circuit, and everyone loses at once, gifting the government a precedent. One hundred small suits in different districts are a hundred separate headaches for DOJ, each of which they must address (and, as we see, they opt to buy them off individually). The swarm strategy beats the battering-ram. That’s the main practical lesson from Mukherji.
Contacting the attorneys is possible and easy: Green’s Sunday webinars, firms’ websites are public (greenusimmigration.com, arielalake.com). Go with your eyes open: $11k, 12–24 months, and the filters discussed next.
“I already have a denial based on Final Merits. Will it be vacated automatically now?”
No. The decision binds USCIS only in Mukherji’s case. Your denial stands and will stand unless you challenge it yourself: motion to reopen, AAO appeal, or a federal suit. There is no automatic reconsideration of past denials; the agency hasn’t even touched the Policy Manual.
“Can I cite Mukherji in an RFE response or petition?”
Yes — and you should, but with realistic expectations. The combination is now triple: quote Bataillon (“the court finds nothing in the law supporting such a requirement”) + Loper Bright + the new fact: the government was given a chance to defend Final Merits in the appellate court and withdrew its appeal without filing a single brief. That’s strong pressure on an officer. But it’s an argument, not a silver bullet: formally an officer is not obliged to follow a district decision from another circuit. The point of citing it is not to “win the RFE with one quote,” but to lay groundwork for a possible suit and force the officer to write a careful denial (a careful denial is harder to defend than a template one).
“Should I wait to file until Final Merits is finally beaten?”
Quite the opposite. “Finally beaten” may never happen: there will be no appellate precedent now, and USCIS is preparing RIN 1615-AC85 to codify the two-step procedure through proper rulemaking, which will neutralize Mukherji’s main procedural argument for future cases. Right now you’re in a window: the procedure is legally wounded, the government showed it doesn’t want to defend it, and the new rule hasn’t been issued. Waiting risks waiting for tightening, not loosening.
“What are the real chances if I go to court?”
The honest answer: almost everything depends on what’s already in your administrative record, and you can’t add to it. The profile where Mukherji’s scheme works: the officer found three or more criteria and the denial was based on the second-stage reasoning — “not enough sustained acclaim,” “not at the top of the field.” That’s exactly the denial type the government now avoids litigating and tends to buy off by approving. If your underlying criteria are not met, federal court will not save you: a judge reviews the officer’s decision, not reassembles your case wholesale. In that scenario $11k is an expensive way to get another denial — now federal.
How to read this case yourself (and why the district will block you)
A small practical bonus because I walked this path yesterday and will save you a couple of hours.
If you dive into PACER’s Nebraska district docket to download documents, most of them will return “You do not have permission to view this document.” This isn’t a bug or your account. It’s FRCP 5.2(c): in immigration and Social Security cases remote access to filings is limited to parties. Outsiders can see only the docket and court orders. The complaint, briefs, and administrative record: only from a courthouse terminal.
Appreciate the irony: a case about an agency that hid its procedure for sixteen years is itself half-hidden by a procedural rule.
How the finale appears in the district docket: the last two entries
For the detail collectors: here’s how the end of the appeal looks in the official district docket report (I purchased the full report, 40 cents for 4 pages). After entry No. 28 on March 30 three months of silence, then two final entries in one day:
- Entry No. 29 (06.10.2026): “COPY of Order (Judgment) from USCA”: the district received a copy of the Eighth Circuit’s order dismissing the appeal. Plainly: the district was formally notified that upstairs everything is over.
- Entry No. 30 (06.10.2026): “MANDATE from USCA”: the mandate. Plainly: the case returned home, the January 28 decision is again fully in the district’s hands, now and forever.
There will be no more entries in this docket except perhaps one: Green’s EAJA fee application (see above). The case status header still reads: CLOSED as of January 28, 2026. A case that shifted the conversation on Final Merits fit entirely into thirty docket entries.
What is freely and publicly available without restrictions:
- Judge Bataillon’s January 28 opinion: in the RECAP district docket, Download PDF.
- The motion to withdraw the appeal: a direct RECAP PDF link (full text and translation above, PDF attached to this post).
- The Eighth Circuit judgment of June 10: full text and translation above, PDF attached. If you want to see it yourself: PACER Appellate (ecf.ca8.uscourts.gov), Orders/Judgments, case 26-1578 — costs a dime. Appellate court orders are open to all, unlike district filings.
- The appellate docket in full: in RECAP.
Install the RECAP extension
If you use PACER, install the RECAP extension. Everything you buy is automatically added to the free public archive, and the next person gets the document for free. Half the links in this post exist because someone with that extension purchased the docs before us.
What to watch next
The watchlist from the previous post has mostly closed (the May 11 brief never appeared; amicus briefs were not needed — the case died before briefing). The new list is:
- By ~July 10: Green’s EAJA fee application in the district docket (RECAP). Will he file? How much will he request? What will be awarded? This is the first public test of the promise “$11k will be returned.”
- Mukherji’s actual I-140 approval. We’ll probably learn this via Mukherji’s LinkedIn or Green’s announcements. If approval doesn’t happen within a month or so, that becomes a separate and loud contempt story.
- RIN 1615-AC85 in the Federal Register. This is now the main front. Publication of the NPRM will mean the start of legalizing Final Merits via proper rulemaking and the opening of the public-comment window, where anyone can participate. Not joking: NPRM comments are read, and the combination “Mukherji + Loper Bright” will be apt there.
- The three live suits: Muthineni (Maryland), Cannabrava de Sousa (Nebraska), Kleinova (Alaska). The first decision among them will show whether Mukherji’s pattern is repeatable or was a single precise shot. Also watch whether the government extinguishes these cases with early approvals — that would be the best confirmation of the model described in this post.
- Policy Manual: uscis.gov/policy-manual/updates. Any movement in Volume 6, Part F is a signal.
Main takeaways in five points
June 10, 2026: 9:56 a.m. withdrawal “with prejudice,” 3:02 p.m. order, 3:06 p.m. mandate. Bataillon’s decision on the unlawfulness of Final Merits is final and nonappealable. USCIS is obliged to approve Mukherji’s petition.
A loss in the Eighth Circuit would have created binding precedent across seven states and almost guaranteed Supreme Court involvement. The government surrendered one case to prevent that. The prediction in my prior post — that USCIS might not want to continue — came true literally.
The Policy Manual was not changed (checked June 11), service centers continue as before, and the overdue NPRM RIN 1615-AC85 is the genuine plan: to adopt the two-step procedure via proper rulemaking and close the procedural hole Mukherji exploited.
Three copycat suits in 4.5 months, zero decisions. But a new argument exists: the government refused to defend the procedure it continues to apply to you. Observed behavior: under pressure USCIS buys off cases rather than fights. That makes federal suit a working tool for quality cases.
His co-counsel are the founders of Ariela Lake: Jess Dawgert (17 years at DOJ, ex-Associate Deputy Attorney General) and Sarah Vuong (15 years in OIL, ex-Assistant Director — the title on the government’s withdrawal). The funnel now invites all EB-1A and EB-1B denials from December 2010 to the present. And until ~July 10 Green has a window to file EAJA in Mukherji: the first public test of the “$11k returns” promise. Watch.
I am not a lawyer and this post is not legal advice. The information is based on open sources: court documents from PACER and the RECAP archive (Eighth Cir. No. 26-1578 and D. Neb. No. 4:24-cv-03170), official USCIS and Federal Register pages, publications by immigration attorneys, and the press release of the firms involved. All documents were verified against primary sources as of June 11, 2026; the situation may change. If you have a denial, RFE, NOID, or questions about choosing among AAO, motion, or federal court, consult a licensed immigration attorney.
