diff --git a/Cslib.lean b/Cslib.lean
index d67431976..7066fe8b8 100644
--- a/Cslib.lean
+++ b/Cslib.lean
@@ -29,6 +29,7 @@ public import Cslib.Computability.Distributed.FLP.Algorithm
 public import Cslib.Computability.Distributed.FLP.CanReachVia
 public import Cslib.Computability.Distributed.FLP.Consensus
 public import Cslib.Computability.Distributed.FLP.FairScheduler
+public import Cslib.Computability.Distributed.FLP.Impossibility
 public import Cslib.Computability.Distributed.FLP.OnePseudoConsensus
 public import Cslib.Computability.Distributed.FLP.PseudoConsensus
 public import Cslib.Computability.Distributed.FLP.ZeroConsensus
diff --git a/Cslib/Computability/Distributed/FLP/Impossibility.lean b/Cslib/Computability/Distributed/FLP/Impossibility.lean
new file mode 100644
index 000000000..3e014e9a9
--- /dev/null
+++ b/Cslib/Computability/Distributed/FLP/Impossibility.lean
@@ -0,0 +1,174 @@
+/-
+Copyright (c) 2026 Ching-Tsun Chou. All rights reserved.
+Released under Apache 2.0 license as described in the file LICENSE.
+Authors: Ching-Tsun Chou
+-/
+
+module
+
+public import Cslib.Computability.Distributed.FLP.OnePseudoConsensus
+public import Cslib.Foundations.Data.OmegaSequence.InfOcc
+
+/-! # Impossibility of asynchronous distributed consensus in the presence of a single fault
+
+This file formalizes the main theorem (Theorem 1) of [Volzer2004] and uses it to prove the
+impossibility of asynchronous distributed consensus in the presence of a single fault.
+-/
+
+@[expose] public section
+
+namespace Cslib.FLP
+
+open Function Set Multiset Fintype Filter ωSequence
+
+variable {P M S : Type*} [DecidableEq P] [DecidableEq M]
+
+variable {a : Algorithm P M S}
+
+/-- `a.ReachableNonUniform inp s` means that `s` is a reahable and non-uniform state of
+algorithm `a` on input `inp`. -/
+abbrev Algorithm.ReachableNonUniform [Fintype P]
+    (a : Algorithm P M S) (inp : P → Bool) (s : State P M S) : Prop :=
+  a.Reachable inp s ∧ a.NonUniform s
+
+/-- Choose an arbitrary non-uniform input. -/
+noncomputable def Algorithm.nonUniformInp [Fintype P] (a : Algorithm P M S)
+    (hpc1 : a.PseudoConsensus 1) (hc : card P ≥ 2) : P → Bool :=
+  Classical.choose (OnePseudoConsensus.nonUniform_inp hpc1 hc)
+
+/-- Assuming `a.PseudoConsensus 1` and there are at least 2 processes, the input chosen by
+`a.nonUniformInp` does indeed give rise to a non-uniform initial state. -/
+theorem OnePseudoConsensus.nonUniform_init [Fintype P]
+    (hpc1 : a.PseudoConsensus 1) (hc : card P ≥ 2) :
+    let inp := a.nonUniformInp hpc1 hc
+    a.ReachableNonUniform inp (a.start inp) := by
+  grind [Algorithm.nonUniformInp, Algorithm.reachable_start]
+
+/-- Assuming `a.PseudoConsensus 1`, starting from any reachable non-uniform state of `a` and
+any message `m` that is in-flight in `s`, there exists a finite execution of `a` in which `m`
+is received and which ends in another non-uniform state. -/
+theorem OnePseudoConsensus.nonUniform_step_exists [Fintype P] {inp : P → Bool}
+    (hpc1 : a.PseudoConsensus 1) {s : State P M S} {m : Message P M}
+    (hs : a.ReachableNonUniform inp s) (hm : m ∈ s.msgs) :
+    ∃ xl t, a.lts.MTr s xl t ∧ some m ∈ xl ∧ a.ReachableNonUniform inp t := by
+  obtain ⟨hr, hn⟩ := hs
+  obtain ⟨s', ⟨xl, h_mtr⟩, _⟩ := nonUniform_step hpc1 hr hn m.dest
+  by_cases h_xl : some m ∈ xl
+  · use xl, s'
+    split_ands
+    · assumption
+    · assumption
+    · grind [Algorithm.reachable_stable, LTS.CanReach]
+    · intro b
+      use m.dest
+      grind
+  · have := Algorithm.mTr_notRcvd_enabled h_mtr hm h_xl
+    have h_tr : a.lts.Tr s' (some m) (a.recvMsg m s') := by grind [Algorithm.lts]
+    have : a.lts.MTr s (xl ++ [some m]) (a.recvMsg m s') := by grind [LTS.MTr.stepR]
+    use xl ++ [some m], a.recvMsg m s'
+    split_ands
+    · assumption
+    · simp
+    · grind [Algorithm.reachable_stable, LTS.CanReach]
+    · intro b
+      use m.dest
+      grind [canDecideWithout_dest]
+
+lemma OnePseudoConsensus.nonUniform_step_aux [Fintype P] (inp : P → Bool)
+    (hpc1 : a.PseudoConsensus 1) (s : State P M S) (m : Message P M) :
+    ∃ xl_t, m ∈ s.msgs ∧ a.ReachableNonUniform inp s →
+      let (xl, t) := xl_t
+      a.lts.MTr s xl t ∧ some m ∈ xl ∧ a.ReachableNonUniform inp t := by
+  by_cases h : m ∈ s.msgs ∧ a.ReachableNonUniform inp s
+  · obtain ⟨hm, hs⟩ := h
+    obtain ⟨xl, t, _⟩ := OnePseudoConsensus.nonUniform_step_exists hpc1 hs hm
+    use (xl, t)
+    grind
+  · use ([], s)
+    grind
+
+/-- Choose an arbitrary finite execution guaranteed to exist by the theorem
+`OnePseudoConsensus.nonUniform_step_exists`. -/
+noncomputable def Algorithm.nonUniformStep [Fintype P] (a : Algorithm P M S) (inp : P → Bool)
+    (hpc1 : a.PseudoConsensus 1) : DeliverMsg P M S :=
+  fun s m ↦ Classical.choose (OnePseudoConsensus.nonUniform_step_aux inp hpc1 s m)
+
+/-- Assuming `a.PseudoConsensus 1`, `a.nonUniformStep` does have the property guaranteed by
+the theorem `OnePseudoConsensus.nonUniform_step_exists`. -/
+theorem OnePseudoConsensus.fair_nonUniform_step [Fintype P] (inp : P → Bool)
+    (hpc1 : a.PseudoConsensus 1) :
+    a.FairDeliverMsg (a.nonUniformStep inp hpc1) univ (a.ReachableNonUniform inp) := by
+  intro s m
+  grind [Algorithm.nonUniformStep]
+
+/-- Assuming `a.PseudoConsensus 1`, starting from any reachable non-uniform state `s0` of `a`,
+use the fair scheduler developed in `FairSchedular.lean` to construct an infinite fair execution
+in which there are infinitely many non-uniform states. -/
+theorem OnePseudoConsensus.fair_nonUniform [Fintype P] (inp : P → Bool)
+    (hpc1 : a.PseudoConsensus 1) (s0 : State P M S) (hs0 : a.ReachableNonUniform inp s0) :
+    ∃ ss xs, a.lts.OmegaExecution ss xs ∧ ss 0 = s0 ∧ (∀ p, ProcFair p ss xs) ∧
+      ∃ᶠ n in atTop, a.ReachableNonUniform inp (ss n) := by
+  obtain ⟨ss', _⟩ := FairScheduler.fair_omegaExecution (fair_nonUniform_step inp hpc1) s0 hs0
+  let xls := a.fairSegActions (a.nonUniformStep inp hpc1) univ s0
+  use ss', xls.flatten
+  split_ands
+  · grind
+  · grind
+  · grind
+  · apply frequently_iff_strictMono.mpr
+    use xls.cumLen
+    grind [cumLen_strictMono]
+
+/-- Assuming `a.PseudoConsensus 1` and there are at least 2 processes, there must exist an
+infinite admissible execution in which no process is faulty but no process terminates, either.
+This theorem formalizes Theorem 1 of [Volzer2004]. -/
+theorem OnePseudoConsensus.not_terminating [Fintype P]
+    (hpc1 : a.PseudoConsensus 1) (hc : card P ≥ 2) :
+    ∃ inp ss xs, a.AdmissibleRun inp 0 ss xs ∧ ∀ p, ¬ ProcTermination p ss xs := by
+  let inp := a.nonUniformInp hpc1 hc
+  let s0 := a.start inp
+  obtain ⟨ss, xs, _, _, _, h_freq⟩ := fair_nonUniform inp hpc1 s0 (nonUniform_init hpc1 hc)
+  use inp, ss, xs
+  split_ands
+  · assumption
+  · assumption
+  · grind [FairRun]
+  · have := numProcFaulty_le_not_procFair (ps := univ) (ss := ss) (xs := xs)
+    grind [ncard_univ, card_eq_nat_card]
+  · rintro p (_ | ⟨k, b, h_k⟩)
+    · grind [not_procFaulty_and_procFair]
+    · have h_r (n : ℕ) : a.Reachable inp (ss n) := by
+        use xs.extract 0 n
+        grind [LTS.OmegaExecution.extract_mTr]
+      have (j : ℕ) (h_j : k ≤ j) : a.Uniform (ss j) b := by
+        apply decided_imp_uniform hpc1 (h_r j)
+        use p
+        apply Algorithm.procDecided_stable (a := a) h_k
+        use xs.extract k j
+        grind [LTS.OmegaExecution.extract_mTr]
+      obtain ⟨n, _⟩ : ∃ n, k ≤ n ∧ a.NonUniform (ss n) := by grind [frequently_atTop.mp h_freq k]
+      grind [not_uniform_and_nonUniform]
+
+/-- As long as there are at least 2 processes, there does not exist a distributed consensus
+algorithm that can tolerate 1 fault. -/
+theorem Consensus.one_not_exists [Fintype P] (hc : card P ≥ 2) :
+    ¬ ∃ a : Algorithm P M S, a.Consensus 1 := by
+  rintro ⟨a, h_cons⟩
+  have hpc1 := PseudoConsensus.of_consensus 1 (show 1 < card P by grind) h_cons
+  obtain ⟨inp, ss, xs, h_run, _⟩ := OnePseudoConsensus.not_terminating hpc1 hc
+  have h_run' := AdmissibleRun.fault_mono (show 0 ≤ 1 by grind) h_run
+  have := Classical.inhabited_of_nonempty <|
+    Fintype.card_pos_iff.mp (show 0 < Fintype.card P by grind)
+  grind [h_cons.right inp ss xs h_run' (default : P)]
+
+/-- As long as there are at least 2 processes, there does not exist a distributed consensus
+algorithm that can tolerate `f` faults for any `f ≥ 1`. -/
+theorem Consensus.ge_one_not_exists [Fintype P] {f : ℕ} (hc : card P ≥ 2) (hf : f ≥ 1) :
+    ¬ ∃ a : Algorithm P M S, a.Consensus f := by
+  rintro ⟨a, h_c⟩
+  suffices h1 : ∃ a : Algorithm P M S, a.Consensus 1 by
+    exact Consensus.one_not_exists hc h1
+  use a
+  grind [Consensus.fault_mono]
+
+end Cslib.FLP
diff --git a/Cslib/Computability/Distributed/FLP/README.md b/Cslib/Computability/Distributed/FLP/README.md
index 11bf07088..39a60cb07 100644
--- a/Cslib/Computability/Distributed/FLP/README.md
+++ b/Cslib/Computability/Distributed/FLP/README.md
@@ -32,9 +32,7 @@ distributed consensus is impossible in the presence of even a single crash fault
 6. `OnePseudoConsensus.lean` focuses on 1-tolerant pseudo-consensus algorithms, defines the key notion
    of "nonuniformity", and proves a number of their properties.
 
-*#7 will appear in a future PR:*
-
-7. `Impossibility.lean` proves that every 1-tolerant pseudo-consensus algorithms has a fair execution
+7. `Impossibility.lean` proves that every 1-tolerant pseudo-consensus algorithm has a fair execution
    which doesn't contain any fault but never reaches a consensus, which then implies that there cannot
    be a consensus algorithm that can tolerate even a single fault.