Government: Arbiter (Step 3)

Inputs: facts.json, claims.md, physicist.md, mathematician.md, contrarian.md; updated 2026-10-05 with the Investigator's first run (sites.md, research/ notes; see the last section, "Investigator findings (2026-10-05)"). Verdicts use the prompt.txt scale. Tags: [read] document; [read: page image p.N] rendered page; [web] URL; [memory] with confidence; [reviewers] a reviewer's own reading that I did not repeat.

Source gap (repeated from claims.md). Three of the four stand-alone PDFs have no text layer: the CIA scalar-waves memo, the CIA copy of Feinberg's tachyon article, and the DIA LENR (low-energy nuclear reactions, "cold fusion") report. They were read by OCR and page images. The Holt NASA report has a text layer. The zip holds three infographics that cite no references, methods or field strengths (2RNM.jpg is credited to Supratik Saha, software, electronics and communications engineering; the gif carries the watermark www.raven1.net; 9RNM.jpg has no visible credit [read: facts.json, digest]) plus three California State Threat Assessment Center (STAC) news digests from 2017 with no physics content. The DIA reference list (pp.7-8) was not read in the first pass; it was read in the QE revision (see the last section but one), which also read the China Lake source report. The cited laboratory papers themselves (except Miles 1996) are still unread. The folder is four unrelated government-hosted documents, not one researcher's theory. Conclusions that rest on [memory] alone are marked.

What I checked myself. I re-read the memo's last page image, the Feinberg page with the bubble-chamber result (p.8), and recomputed every number. I did not re-read the memo pp.1-2 or the DIA text; there I rely on the three reviewers, who agree with each other.

---

Corrections to Step 1 and to the reviewers

1. Memo date. The handwritten date on the memo's last page reads 12/22/87 (the last digit is a 7 and the tens digit is ambiguous in handwriting; OCR gave "37") [read: page image p.3]. The document was released 2001 [read: stamp "Approved For Release 2001/03/07"]. Treat 1987 as probable. Step 1's "early-mid 1980s" is wrong or at best loose. This does not affect any verdict.
2. Memo equations are legible and all three reviewers verified them. Step 1's "partly garbled" is withdrawn.
3. Feinberg's first pass did find candidate tachyon events. Confirmed [read: page image p.8]: of about 6,000 K-capture events, 101 had negative missing mass squared; after removing in-flight captures this fell to 23, and after re-measurement to none. The limit on tachyon production from this experiment is less than 1 part in 400 of the events [read: p.8]. Step 1's "found none" is true only of the final result. The Cerenkov experiment was by Alvager and Kreisler at Princeton, not Feinberg's Columbia group [read: p.8 caption]. The "1/10,000 of pair rate" figure is verified [read: Feinberg p.7, QE revision]. The mathematician's "bubble chamber at Brookhaven" is not on page 8, but the QE deep read found it in the p.3 figure caption, so it is correct (see QE revision).
4. Memo para 7 (contrarian). It says such waves "grow via energy supplied by the medium (free space, sea water, etc.)" [read: page image p.3]. So the memo proposes the medium as an energy source as well as vacuum.
5. Memo and the "carry energy" wording. The memo says the energy source "is not clear" and existence "is not proven" [read: page image p.3]. Claim 1 in claims.md ("carry energy") is stronger than the memo. See corrected claim list.
6. Units slip in the memo (mathematician). It writes Gaussian-unit potentials but gives the Poynting vector as E x B in "watts/m^2", which is correct in neither Gaussian (needs c/4pi) nor SI (needs 1/mu0) [reviewers; not re-checked by me]. Minor.
7. DIA internal inconsistency (contrarian). DIA p.2 places Arata at Osaka University and p.5 reportedly at Mitsubishi Heavy Industries [reviewers; not re-checked]. Minor.
8. Frey effect dates. First published 1961 [web: https://en.wikipedia.org/wiki/Microwave_auditory_effect]. The Sharp-Grove word-recognition work at Walter Reed is reported second-hand via a 1975 article by Justesen (same URL). The infographic's "1974" is unverified. (Refined 2026-10-05: the dates are a 1973 personal communication, a 1974 mechanism paper and a 1975 published account; the infographic's year is imprecise, not clearly wrong. See the last section, F5.1.)
9. Holt page count 13 vs 15: scan and cover difference, no substance.

---

Numbers I recomputed

---

Web checks of [memory] facts that verdicts depend on

Budget used: 5 of 8 calls (the web-search tool was refused; only Wikipedia fetches were possible, so all checks are secondary sources).

• 2004 DOE review split "approximately evenly" on excess heat; helium found "very close to background" per negative reviewers; no federal programme recommended, only individual well-designed studies; 1989 panel "did not present convincing evidence" [web: https://en.wikipedia.org/wiki/Cold_fusion]. Retagged [web]. The same page says the 2019 Google-funded Nature paper was a "failed replication attempt" and a 2021 coordinated study was planned after Nature published "anomalous findings". I could not read the Nature papers.
• Frey 1961; Sharp-Grove reported 1975, second-hand [web: https://en.wikipedia.org/wiki/Microwave_auditory_effect]. Retagged [web].
• OPERA 2011 superluminal neutrinos traced to an instrument fault [web: https://en.wikipedia.org/wiki/Tachyon, as cited by the reviewers; I did not re-fetch]. The same page: tachyonic fields signal instabilities.
• Check failed: largest observed solar-flare energy (about 1e32 erg = 1e25 J). The Wikipedia pages on Solar flare and the Carrington event give no energies. Still [memory, medium]. Claim 3 is built so that its verdict does not depend on this figure.
• Not checked, not verdict-critical: seawater conductivity, brain magnetic field (about 1e-13 T), TMS strength, neuron stimulation threshold (10-100 V/m), Tonomura 1986, Aharonov-Bohm 1959. All [memory, medium]; they support illustrations, not verdicts.

---

Claim 1. Scalar waves: E = 0, B = 0, no Poynting flow, carry energy, penetrate seawater, draw vacuum energy

Where they agree. All three verified by substitution that A = grad S, phi = -(1/c) dS/dt gives E = 0 and B = 0 for any smooth S, and that the memo's "wave equation for S" is the Lorenz gauge condition. So S is a pure gauge function. Energy density is built from E and B, so it is zero. The Poynting-vector ambiguity (adding a curl) does not create flow where the energy density is zero. Seawater attenuation is real. Vacuum energy cannot be extracted net.

Where they disagree. - Mathematician: "Internally inconsistent". Physicist and contrarian: algebra correct, physics empty; contrarian explicitly rejects "inconsistent" because the memo only suggests. - Weak form: physicist rates "Established" (as math); contrarian does not rate it. - Seawater: contrarian "Never properly tested"; mathematician "no measurable difference, so not tested-not-supported"; physicist "Not testable as stated if recast".

Arbitration. The algebra is correct, so the memo is not internally inconsistent in its equations. It is inconsistent in its claim: it says the wave carries energy while its own example has E = B = 0, hence zero energy density. The mathematician's label applies to that claim-versus-example conflict, not to the algebra. On seawater, "never properly tested" is the wrong label for the S wave itself because there is no detector response to test: a charge feels forces only from E and B, and both vanish. That is "Not testable as stated". The contrarian's label would apply only if someone supplied a real coupling. Tested/never-tested/dismissed: not tested, not dismissed on the record in this folder. The memo is an analyst's summary, written on a verbal request, that says "not proven".

What the memo itself reports (kept in full). It gives the applications that made the topic of interest: penetrating sea water "for submarine communications" and "a new form of technology for ASW" (anti-submarine warfare), of "high interest to the U.S. R&D Community and the Intelligence Community" [read: deep/memo, p.1]. It reports a classified paragraph (C/NF) that "a US community" believes scalar waves realizable, citing an IEEE-sponsored conference "dedicated to Nicola Tesla" and an implied "Tesla Connection"; the rest of that paragraph is withheld (SG1B) [read: p.1]. It names "Lt Col Tom Bearden" as "a chief proponent", known for his "Fer-de-Lance" briefing on "Soviet Scalar Weapons", as the source of the energy-from-the-vacuum idea [read: p.2]. The physics in the memo is therefore the analyst's summary of the proponents' description (Bearden's own account is reviewed in the Bearden project), not the analyst's own endorsement.

The silence after the memo (absence of evidence, analysed). No follow-up appears in this folder. That is a gap in a four-document folder, not a finding that none happened, and the record does not in fact stop there: the same CIA reading-room file series holds a later item, CIA-RDP96-00792R000500240008-9, titled "Scalar waves interactions with scalar energy (Dr. Glen Rein) article from Canada Weekly, Col. Buzby (Star Wars now)" [web: https://archive.org/details/CIA-RDP96-00792R000500240001-6; title only, not read, date unknown], and the addressee "Dr. Vorona" appears elsewhere in the same release as "Dr. Jack Vorona, DDS&TI" receiving a memo from H. E. Puthoff [web: https://archive.org/details/cia-readingroom-document-cia-rdp96-00788r001200230021-2; title only]. So the topic stayed in the file at least one further item. Competing explanations for why no test or assessment is visible: (a) the topic was judged unpromising and dropped (consistent with the memo's own "not proven" and with the calculation above); (b) further work exists but is in the withheld SG1B/SG1J portions or in unreleased files (consistent with the redactions and the CONFIDENTIAL/NOFORN marking; the codes' meaning is [memory, low]); (c) the interest was collection of open-source claims (Bearden, Rein, press articles) rather than research, so no test would be expected (consistent with the later item being a press article). The records that would decide: the full RDP96-00792R0005002400xx folder (items 0001 to 0008 and after) in the CIA reading room, and any DIA DT response to the 12/22/87 request. None was read for this revision. No suppression is documented in what was read; that is a statement about this folder, not a finding that none occurred. (Update 2026-10-05: the Investigator has since read items 0002 and 0008 and listed the whole folder, 8 items, and found the originator's office; this paragraph's explanations are re-weighed in "Investigator findings (2026-10-05)", F1.3. The "Dr. Vorona, DDS&TI" wording above came from a title and is superseded by the primary record there.) Aharonov-Bohm closes no loophole here: for single-valued S the loop integral of A is zero [reasoned; Aharonov-Bohm itself memory, high].

Final verdict (split). - S as a mathematical solution, E = B = 0: Established as mathematics; pure gauge, no physical content. - Propagating wave carrying energy or a signal with E = B = 0: Refuted by calculation (zero energy density and flux; also at odds with the memo's own example). - Penetrating seawater by a non-EM carrier: Not testable as stated (no coupling specified). It would become "never properly tested" only if a coupling were defined. - (Added 2026-10-05) Bearden's two-beam interference mechanism for seawater penetration (as summarised in CIA-RDP96-00792R000500240002-5): Refuted by calculation (linear superposition; see Investigator findings, F1.1). His "detection only in altered space-time at a superconducting-magnet pole": Not testable as stated. Rein's "Tesla watch" effect on nerve cells: Never properly tested. - Vacuum or seawater as energy source: Not testable as stated, and excluded in any net-gain form by energy conservation (vacuum is the lowest-energy state; seawater without a gradient has no free energy [reasoned]).

What would settle it. Write down the force on a test charge from the S field. The answer is zero identically. To keep the idea alive the proponent must name a new coupling term and its strength.

---

Claim 2. LENR (cold fusion): excess heat, particles, He-4, transmutation, "accepted as reproducible" (DIA 2009)

Where they agree. The DIA report has no data of its own; its five named contributors (McDaniel, Mosier-Boss, McKubre, Forsley, DeChiaro) are active LENR researchers [read: DIA p.6, per all three]. Their stake: the report's description of LENR is in large part the researchers' own description of their field, carried under an agency's name. It is reported below as their characterization, beside the reviewers' and the critics' characterizations, not as a neutral finding either way. It admits no expected radiation, failed replications, "much skepticism remains", proprietary details, and "no current theory" [read, per reviewers]. Ordinary d+d branching (neutrons, tritium, gammas) does not match the claimed heat. The 2.6e11 He-4 per joule number is the right test. No reviewer read the underlying papers.

Where they disagree. - Ordinary-d+d heat: physicist "Refuted by calculation"; mathematician and contrarian "Tested, not supported" or leave it. - Nuclear origin in general: physicist "Never properly tested"; mathematician and contrarian "Tested, not supported". - DIA's "accepted as reproducible": physicist "Tested, not supported"; mathematician "Not testable as stated"; contrarian "Not supported".

Arbitration. 1. A government agency summarising open literature is not a test. A DOE review is also not an experiment: the 2004 review was a paper review, so "tested by DOE twice" (physicist) overstates it. The experiments are the labs' own work (not read, except Miles 1996 in the QE revision), the fast 1989 replications (many negative; the 1989 DOE Energy Research Advisory Board (ERAB) panel, an institutional reviewer, characterized the evidence as not "convincing" [web]), and the 2019 Nature multi-lab Pd-D attempt (described as a "failed replication attempt" on the Wikipedia cold-fusion page, which also says Nature later published "anomalous findings" [web]; the paper's own design and wording not read). 2. The contrarian's fairness point is the strongest argument for the claim and I keep it: DIA says early replications failed because the loading "incubation" needs several hundred hours [read: DIA p.4, per reviewers], so a weeks-long 1989 test was not a fair test of the later-specified protocol. The mathematician's counter is also fair: an effect that explains away failures by unspecified incubation is close to unfalsifiable until the protocol is fixed in advance. Neither side has supplied the missing evidence: a pre-registered protocol run blind by independent labs. That is why "never properly tested" fits the heat claim better than "tested, not supported". 3. The "refuted by calculation" label for ordinary d+d is correct but narrow: with standard branching, 1 W would need 1.9e12 neutrons/s or 2.6e11 He-4/s with a 23.8 MeV gamma each, and, for ordinary d+d, no such radiation budget is seen. [Reworded in QE revision: DIA says radiation was absent only for Fleischmann-Pons 1989; for SPAWAR 2009 it cites CR-39 neutron tracks, which are about 9 orders of magnitude too weak to carry watt-level heat.] The Gamow suppression (tunnelling probability) at 0.025 eV is astronomically small [memory, high; no one computed it]. That leaves any real effect needing a mechanism outside standard nuclear theory, which is not a refutation of the heat. 4. The 2004 split is neither confirmation nor refutation of heat; the negative reviewers characterized the helium as "very close to background" [web, secondary], and the proponents (China Lake, SRI) characterize it as correlated with heat. The positive helium-heat correlation is the best evidence claimed; the China Lake report was read in the QE revision (below), SRI was not. 5. Tested/never/dismissed: the heat claim was tested (fast, 1989; 2019 one design) and partly not tested in its long-loading form. On dismissal, the evidence on record is the DIA report's own statement: "Because some peer-reviewed journals are reluctant to review or publish LENR data due to past controversies, most results are presented at international conferences" [read: DIA p.4]. It comes from a report whose contributors are the affected researchers, and no second source (journal editorial policy, rejection records) was sought or found in this review. It is evidence that a publication barrier was perceived by the researchers; its extent is open. The same page also says "each team has proprietary aspects of their experiments that are not shared" [read: DIA p.4], which on the same source cuts against open replication.

Final verdict (split). - Anomalous excess heat in some Pd/D cells: Never properly tested (no neutral, protocol-fixed, blind multi-lab test with an agreed effect size; contested, not settled). - Nuclear origin by ordinary d+d fusion at lattice temperature: Refuted by calculation (neutron, tritium, gamma budgets and Coulomb barrier). - Unknown nuclear mechanism (He-4 correlation, transmutation, particles): Never properly tested; helium correlation disputed; transmutation papers unread, status unknown. - DIA's "has begun to be accepted ... as reproducible and legitimate", split into its two parts (scale label restored in the evidence-rule revision): (a) "reproducible" as a physical claim is the excess-heat claim above, Never properly tested; Miles 1996 says in its own words "We have not been able to solve the reproducibility problem" [read: deep/DIA, Miles p.7]. (b) "accepted" is a claim about the scientific community, not about nature: Not testable as stated by a physical experiment. The record bearing on it: the 2004 DOE reviewers split "approximately evenly" [web], and 16 of the report's 54 notes trace to its own contributors [read: deep/DIA].

What would settle it. A pre-registered, blinded, multi-lab run of one fixed Pd/D loading protocol for 500+ hours, with isoperibolic or flow calorimetry at about 0.1% accuracy and mass-spectrometer He-4 measurement against heat (target about 2.6e11 atoms per joule), with air-contamination controls and a null cell (H2O in place of D2O). Also read the DIA references (pp.7-8).

---

Claim 3. EM-gravity coupling; laser megagauss fields tuned to a distant space-time point to move a craft (Holt, NASA TM-80961, 1979)

Where they agree. No equations in the report [read, per all]. It calls itself unproven and says it may need "a complete revamping". Megagauss laser fields (grad n x grad T) and reconnection are real physics; Holt's references are real 1976-78 papers [read: refs, per reviewers]. Holt's argument that merging cannot supply flare energy fails on his own 3000 G figure.

Where they disagree. - Is Holt's flare statement an overstatement of the energy (physicist, factor about 4 above the largest flare) or of the rate (contrarian)? - Is the flare statement itself "Refuted by calculation" (physicist, contrarian), or only the argument built on it (mathematician)?

Arbitration. My numbers: 10% of solar output for 1 s is 3.8e25 J. A cube 1e7 m (10,000 km) across at 0.3 T stores that much; a 1e8 m active region stores 3.6e28 J (also [memory, medium] for the region size). So the energy budget is not a problem for reconnection, and that argument is refuted by calculation regardless of flare records. The size of the largest flare (1e25 J, [memory, medium], web check failed) is the only number the reviewers differ on. Because the energy-per-second phrase is Holt's own, the two readings are both fair: energy exceeds the (unconfirmed) largest flare by a factor of a few, and the implied power (3.8e25 W) is beyond any flare, whose power for 1e25 J over 100-1000 s is 3e-5 to 3e-4 of solar luminosity. I treat "Holt's figure is high" as plausible but unverified [memory, medium]; no verdict depends on it. The magnetic energy at 1 MG is 4e9 J/m^3, which would be about 4e12 J (about 1 kiloton TNT) in a 1000 m^3 craft volume, and the laser-plasma fields are transient and micron to millimetre scale [memory, medium]. The report gives no scaling. Tested/never/dismissed: never tested, never dismissed on the record read (NASA published it with a foreword saying it was "private, unofficial research" and "NASA is not involved in UFO research"; it was presented at a special session of the 15th Joint AIAA/SAE/ASME Propulsion Conference, June 1979 [read: Holt p.3]). No suppression is documented in what was read. The record after 1979 does not go silent: Holt published "Prospects for a breakthrough in field-dependent 'propulsion'" at the 16th Joint Propulsion Conference (1980), a popular version "Starcraft" in Science Digest (May 1982), and later co-authored with Eric Davis and Hal Puthoff "Space testing of electromagnetically sensitive materials for breakthrough propulsion physics" (AIAA 1999-2143) [web: https://arc.aiaa.org/doi/10.2514/6.1999-2143; the 1980 and 1982 items from a search-result summary, not read]. None of these was read, so whether any of them defines the "space-time harmonic" quantitatively is unknown; they are the first records to read.

Final verdict (split). - Laser-generated megagauss fields and reconnection as lab and solar physics: Established (not at craft scale). - "Reconnection cannot supply flare energy" (Holt's motivating premise): Refuted by calculation. - EM-gravity "resonance" with a space-time point and relocation of a craft: Not testable as stated (no defined quantity, frequency, or operator; EM fields gravitate only through their energy density, 4.4e-8 kg/m^3 equivalent at 1 MG, so GR already says the effect is negligible). If given a precise form it would be "Never properly tested". - UFO motivations: Holt's own list is "high speed, right angle turns, abrupt stops or accelerations ... absence of sonic booms despite calculated speeds of 22,000 mph or more", and environmental effects "such as burns, dehydration, stopping of automobile engines, TV and radio disruption, melting or alteration of ground and road surfaces, power disruptions, and static electricity effects" [read: Holt p.7]. These are witness reports as summarized by Holt; the report gives no case citations or instrument records, so they cannot be weighed quantitatively here. Holt himself proposed obtaining that data: photographic spectra, high-sensitivity magnetic measurements, radio and sound recordings, Geiger counters distributed to "UFO investigators, police departments", physician study of burns and radiation-sickness symptoms, and "regressive hypnosis" of close-encounter witnesses [read: Holt p.11]. No verdict on the UFO reports themselves: they are outside what these documents can settle.

What would settle it. Holt must first define a calculable quantity (what is the "space-time harmonic", at what frequency, with what coupling constant). Then a torsion-balance or laser-plasma experiment can bound it. Today the equivalence-principle tests (Eotvos-type) already limit any anomalous EM-gravity coupling [memory, high; not web-checked].

---

Claim 4. Tachyons are not forbidden by relativity; 1970 searches found none (Feinberg, Scientific American)

Where they agree. The kinematics with imaginary rest mass m = i mu are consistent: E = mu c^2/sqrt(v^2/c^2 - 1) is real for v > c, E falls as v rises, and c is a lower bound. Searches were null. A tachyonic field is now read as an instability, not a particle. OPERA 2011 was an instrument error [web, secondary].

Where they disagree. Only on labelling and on facts from the article: the physicist corrected Step 1 (first pass had negative-m^2 events); the contrarian did not. I confirmed the physicist [read: page image p.8].

Arbitration. The article is a mainstream magazine piece by a mainstream physicist; it is a model of how to state and test a hypothesis. Tested/never/dismissed: tested, and the proposer ran the test and reported the null. The first-pass events were analysed and removed (101 to 23 to 0), an honest example of checking for artefact [read: page image p.8]. The causality puzzle (observer-dependent emitter/absorber order) is argued but not resolved [read, per physicist]. Independent post-1970 replication of the nulls is not documented in this folder [no source; mathematician's "later searches also null" is [memory, medium]].

Final verdict (split). - Kinematics with imaginary mass: Established as consistent mathematics. - Existence of tachyonic particles: Tested, not supported. - The 1970 research programme (stable superluminal particle): Historically superseded but instructive.

What would settle it. Already done for the cases searched. A new claim would need a detector signature (Cerenkov emission in vacuum, negative missing mass squared) beyond those limits.

---

Claim 5. Directed EM (microwave or ELF) energy remotely affects the human body and mind

What the infographics say (kept in full). 2RNM.jpg (credited to Supratik Saha): brain-wave bands, a "Bio-Electromagnetic Field" unique to each person serving as a "frequency ID detectable by radar" [read: digest p.1]. 9RNM.jpg: "Remote Mind Control / Remote Brain Mapping" by mobile phone networks, vans, transportable "psychotronic weapons" and helicopters, with a table of "ELF brain stimulation" frequencies (motor cortex 10 Hz, auditory 15 Hz, visual 25 Hz, somatosensory 9 Hz, thought centre 20 Hz) [read: image, Step 1]. The gif (watermark www.raven1.net), "Psycho-Electronic Weapon Effects": forced memory blanking, reading and broadcasting of thoughts, controlled dreams, microwave hearing, artificial tinnitus, forced speech and jaw movement, "transparent eyelids", genital-area itching and forced orgasm, a "caffeine field", pain "of hot needles", forced sleep prevention and induction, microwave burns, electrical shocks "used in the Gulf War", a "demo" of neuro-control of the toes, a claim of "zero trace evidence", and break-ins at home and work with items sabotaged or stolen, which the page attributes to psychological warfare research; a starred note says the first unclassified transmission of voice into a skull was in 1974 by Dr Joseph C. Sharp of the Walter Reed Army Institute of Research [read: digest p.3; facts.json]. These are the infographics' assertions; the images give no case records, methods or field strengths by which to weigh them, and they are recorded here so the reader can weigh them.

Where they agree. The infographics cite no sources, methods, or field strengths. The Frey microwave-hearing effect is real and thermoelastic. At 10 Hz, wavelength is 3e7 m, so a standoff beam cannot be focused on one head. The brain's external field is far below any radar-detectable level [memory, medium]. The three STAC digests (California threat-assessment summaries, March-September 2017: school threat arrests, ISIS cases, hate crimes, the North Korean nuclear test [read: digests]) contain nothing on EM effects; why they were bundled with the infographics is unknown.

Where they disagree. - Physicist and mathematician: "Refuted by calculation" for focused ELF stimulation; contrarian: "not supported" but declines "refuted" because no full field calculation was done. - The mathematician did not open the images and relies on Step 1 and the other reviewers. - Mathematician and contrarian: "Never properly tested" for mind control; physicist: "Not testable as stated".

Arbitration. The wavelength argument is a valid, simple calculation for one specific claim: focusing a standoff ELF beam on one person or one cortical region is excluded by diffraction (spot size at least about the wavelength, 3e7 m, versus 0.2 m for a head). That is enough for "Refuted by calculation" for that form. The threshold argument (neurons needing about 10-100 V/m versus far weaker ambient fields) is [memory, medium] and supports a "not supported" statement only. Thought insertion, radar identification of a person by brain-wave modulation, and the unsourced frequency table give no field strength or detector, so they are "Not testable as stated"; "never properly tested" would mean someone defined a test, and no one did. Tested/never/dismissed: Frey effect tested and replicated (labs and counts not verified; [web, secondary]); the rest no test cited at all, not dismissed after testing. The Frey effect needs peak power densities near the thermal-damage range [web, as reported by Wikipedia; the contrarian quotes 267 mW/cm^2 for single pulses at 1.3 GHz, which I did not re-fetch], so in my judgement the documented effect is narrow (clicks and buzzes), and the infographics' list goes far beyond it. The Sharp-Grove word-recognition result is documented in one secondary source (Wikipedia, citing Justesen 1975) [web]; the Walter Reed primary record has not been read, so it is weighed as reported, not discounted. Date check: the web source gives early 1970s, reported in 1975, against the infographic's 1974.

Final verdict (split). - Microwave auditory (Frey) effect: Established. - Focused, standoff ELF stimulation of a specific brain region: Refuted by calculation (diffraction). - (Added 2026-10-05) Speech conveyed to a person by pulse-timed microwaves (Sharp-Grove, as reported by Justesen 1975): Plausible, untested (a reported demonstration exists in one published account; no primary report, power levels or controlled intelligibility test found; see F5.1). - Remote thought insertion, memory blanking, forced speech, "imposed images": Not testable as stated in these documents; the sensitivity needed is far outside known neural thresholds [memory, medium]. - Radar identification of a person by brain-wave modulation: Not testable as stated; practically excluded by about 1e-13 T field strength [memory, medium].

What would settle it. A calibrated dose-response study of induced field in tissue (V/m) versus measured neural or perceptual effect, for a stated frequency and distance. Existing TMS dosimetry (coil at the scalp, about 1 T pulses [memory, medium]) already shows how far the claims are from realistic standoff fields.

---

Corrected claim list

1. A mathematical "scalar wave" S with E = 0 and B = 0 exists in a memo held in the CIA reading-room STARGATE files (12/22/87, probable), originating from a "DT-ACO" office, which on the 2026-10-05 primary record (a 1984 INSCOM memo) is a DIA science-and-technology intelligence office (inferred from the office symbol; the memo does not name DIA), as a pure gauge function; the memo says its energy source "is not clear" and existence "not proven", and suggests the wave might penetrate seawater, not attenuate, and draw energy from free space or sea water. The memo's energy-carrying idea does not follow from its own equations. A companion note in the same folder (to "COL Buzby", excerpting Bearden's "Star Wars Now", 24 April 1984) gives the proponent's mechanism: two identical beams 180 degrees out of phase sum to zero, so supposedly no attenuation in seawater, and two such beams crossing give a "beat frequency" with energy at the distant point, detectable only at the pole of a 4-8 T superconducting magnet. That mechanism fails by linear superposition (zero sum in transit means nothing to propagate; a non-zero partial sum is ordinary, attenuated EM). A third item in the folder reports a proponent's cell-culture experiment (Rein, "Tesla watch").
2. The DIA 2009 report says LENR in Pd/D cells gives excess heat, particles, He-4 and transmutation and "has begun to be accepted ... as reproducible and legitimate", while also admitting no expected radiation, failed replications, residual skepticism, and no theory. It supplies no data of its own, and 16 of its 54 endnotes trace to its five named contributors. Per one source with a stake in the field (New Energy Times), it was based on an Aug 4-5, 2009 workshop at SPAWAR Pacific at which the contributors and three others presented, and carries a coordination line naming DIA/DRI, CPT, DWO, DOE/IN, SPAWAR Pacific and NSWC Dahlgren.
3. Holt (NASA TM-80961, 1979, private research with a disclaimer) offers two admittedly unproven assumptions (space-time as a projection of higher dimensions; EM-gravity relations) and a proposal to use laser megagauss field-line merging at a tuned rate to resonate with a distant point; no equations. His premise that reconnection cannot supply flare energy fails on his own 3000 G figure.
4. Feinberg (Scientific American, February 1970) argues tachyons are not excluded by relativity and reports that his Columbia bubble-chamber search (photographs made at Brookhaven for another experiment; about 6,000 K-capture events) found 101 candidate events, cut to 23 by removing captures in flight and to 0 on remeasurement (limit under 1 part in 400 of pion production), plus a Princeton Cerenkov search (Alvager and Kreisler) with a limit under 1/10,000 of electron-positron pair production, and a null antiproton-proton search.
5. Three infographics in a zip (one credited to Supratik Saha, one watermarked www.raven1.net, one uncredited), citing no references, claim that directed EM energy affects the body and mind, from microwave hearing and an ELF frequency table to forced speech, Gulf War shocks and break-ins attributed to psychological warfare research; the microwave auditory effect is real and documented; the other effects are asserted with no case records or field strengths. (Added 2026-10-05: the speech-by-microwave note rests on a 1975 published account, Justesen, of a September 1973 personal communication from Sharp and Grove; the 1974 Sharp-Grove-Gandhi paper concerns the mechanism, not speech.)

Verdict summary

What would settle it (summary)

• Claim 1: a force on a test charge from S (zero by calculation); a proponent must name a new coupling. (2026-10-05) For the proponent's coupling now on record (two out-of-phase beams): one measurement settles which of the two cases applies: measure the field of each beam separately along the water path (each should be attenuated) and the sum at the receiver; the superposition calculation predicts no signal where the sum is zero. For Rein's watch: a blinded, randomised, field-measured replication with a sham watch beside the same dish, many runs per condition, with the field (E, B, temperature) logged.
• Claim 2: a pre-registered, blind, multi-lab Pd/D run with heat and He-4 measured together (2.6e11 atoms per joule) in metal flasks with air-leak controls, plus a light-water control and an independent neutron counter beside any CR-39 chips. (The DIA references have now been read.) (2026-10-05) Add: pretreated, helium-blanked hardware (untreated 316 steel alone holds 15-240 pmol, 34-552 J equivalent), and hand the SRI-replication CR-39 chips to independent examiners with a pre-set analysis. The records to obtain: the ENEA RT/2002/41/FUS body, the 2021 JCMNS body and the NASA and Russian Academy examiners' own reports, the Kowalski comment, and ARPA-E LENR project results.
• Claim 3: a calculable definition of the "space-time harmonic", then a torsion-balance or laser-plasma bound. (2026-10-05) First read Holt's 1980 AIAA paper and NTRS 19910012853 and 19940022865 to see whether any defines it.
• Claim 4: nothing further needed for the 1970 searches.
• Claim 5: a field-strength threshold study at stated frequency and distance. (2026-10-05) For speech by microwave: the Justesen 1975 page carrying the Sharp-Grove citation, any Walter Reed report of the trial with peak power density, pulse width and repetition rate, and a controlled intelligibility test.

Open questions

• Is the LENR heat-and-helium correlation real once air contamination is controlled? The 2004 reviewers called helium near background; the proponents disagree. The QE revision read Miles 1996, which admits air helium at about the signal level in glass flasks; no independent blind replication was found.
• Why do LENR effects, if real, need hundreds of hours? DIA gives the "incubation" requirement as the reason early replications failed [read: DIA p.4]; a protocol fixed in advance would turn it into a testable prediction.
• What did the 2019 Nature paper and the 2021 coordinated study test, and how did they set loading time? (Not read; secondary source only.)
• What are the transmutation claims (Iwamura, Mitsubishi) based on, and were contamination controls adequate? Unread.
• What followed the memo (12/22/87)? (Updated 2026-10-05) The released folder RDP96-00792R0005002400xx is now listed: 8 items, three on scalar waves (memo, Bearden note, Rein) and five press clippings; no test, tasking or assessment is in it. Still open: the redacted (SG1B/SG1J) portions, the unreleased CENTER LANE/DT files, any DT response, Bearden's full "STAR WARS NOW!" paper (in the reading room, not opened), and the untried searches "DT-ACO" DIA, Vorona with "scalar".
• What did Holt's later papers add? (Updated 2026-10-05) NTRS abstracts 1986-2004 (AIAA 89-2858, 1990 Vision-21, 1993, Space 2004) show continued proposals for test platforms; none defines the "space-time harmonic". Full texts and the 1980 AIAA, 1982 Science Digest and 1999 AIAA (Holt, Davis, Puthoff) items remain unread. Was any proposed test flown?
• Is the Rein result real? The printed means give p = 0.022 (two-sided, SD) or 0.29 (SEM), not "p < 0.01"; only a one-sided SD test gives 0.011. The table was read from OCR; the page image has not been seen by anyone here.
• What do the NASA and Russian Academy CR-39 examiners' reports say, and was tritium measured in the SRI replication? Not in the abstract.
• What does the ENEA report (RT/2002/41/FUS, pp.2-26) show for heat and helium per interval, with what blanks? Title page only.
• What are the ARPA-E LENR projects' results (8 projects, $10M, announced 2023)? Known from search snippets only; my own fetch attempts this run failed.
• Is "Col Buzby" the CENTER LANE project manager LTC Brian Buzby? Probable, not confirmed.
• Largest solar-flare energy: web check failed; [memory, medium] only.
• Was the Sharp-Grove demonstration a word-recognition result, and at what power? Documented in one secondary source (Wikipedia citing Justesen 1975); the Walter Reed papers would decide.
• Who made the infographics, and from what data? 2RNM.jpg credits Supratik Saha; the gif is watermarked www.raven1.net; 9RNM.jpg is uncredited. The data behind them are not given.

---

QE revision (after qe.md, verdict MINOR)

Inputs added: deep/DIA-2009DIA-08-0911-003-a8012e.md (Opus, pp.1-8 plus Miles 1996), deep/CIA-CIA-RDP96-00787R000100030001-9-7119a0.md (Feinberg, full), deep/CIA-CIA-RDP96-00792R000500240001-6-cc9638.md (memo, page images). I did not re-read the source documents; the points below rest on those deep reads [read: deep/], with their own tags kept.

New evidence

1. The DIA reference list (Claim 2). - 54 endnotes. Peer-reviewed mainstream papers: 10 notes, 7 distinct papers (Fleischmann-Pons 1989; Szpak 2004 and 2005; Iwamura 2002; Mosier-Boss 2009; Sinha-Meulenberg 2006; the Arata-credited note). The rest are conference talks, briefings, one government report (Miles 1996), news, and four notes with no source [read: deep/DIA section 2]. - 16 of 54 notes tie to the five named contributors (Mosier-Boss, McKubre, Forsley, McDaniel, DeChiaro), including two "personal correspondence" notes (50, 52). The "corroborated by experts" statement therefore rests largely on the contributors' own work: it is their characterization of their field, with their stake, which the reader should weigh as such [read: deep/DIA]. - Misstatements: explosions "Japanese, French, and U.S." but notes give one Chinese conference paper and one French paper; the Arata note (19) and the Violante note (20) do not match the claims they support; "China Lake first reported He-4 in 1996" while Miles says March 1991 [read: deep/DIA section 3]. - Headline hedge: "high confidence" attaches to "if LENR can produce nuclear-origin energy" [read: DIA p.1].

2. CR-39 triple tracks (Mosier-Boss et al., Naturwissenschaften 2009). This is DIA's only peer-reviewed particle source, cited three times. Calculation: three-alpha breakup of carbon-12 has Q = -7.27 MeV, so the neutron threshold is about 7.9 MeV, whereas a primary d+d neutron carries 2.45 MeV; so even real tracks are not primary d+d neutrons [calculation, deep/DIA 4a]. An order-of-magnitude bound (CR-39 efficiency about 1e-4, [memory, low]) puts the neutron source near 1e3 n/s, about 9 orders below the 1.9e12 n/s per watt needed; the conclusion survives a factor of 1000 error either way. Status: contested; the same group published follow-ups, and in 2021 Mosier-Boss and Forsley themselves published "Nuclear Generated Tracks or Artifacts?" [web: https://jcmns.org/article/72567-interpreting-cr-39-detectors-used-in-pd-d-co-deposition-nuclear-generated-tracks-or-artifacts]; no independent neutron-counter result found. The paper itself was not read (login wall) and the Kowalski critique is [memory, medium], unchecked.

3. Miles, Bush and Johnson, China Lake TP 8302 (1996) [web: https://lenr-canr.org/acrobat/MilesManomalousea.pdf; a copy hosted by lenr-canr.org, an archive of LENR literature run by supporters of the field, read by the deep reader; it is a Naval Air Warfare Center technical publication by the experimenters, so it is the experimenters' own account]. - Best cells: up to 30% excess, 0.52 W, 1400 kJ; excess power in 7 of 8 Pd-B runs and 17 of 28 Johnson-Matthey runs; some Pd sources never gave excess; "We have not been able to solve the reproducibility problem" (p.7). - He-4: 30 of 33 runs consistent with heat-helium correlation; 1e11 to 1e12 atoms/s per watt, which brackets the d+d value 2.6e11 per joule. Analyses done outside China Lake (Bureau of Mines; Rockwell). Odds quoted: 1 in 750,000 against random error alone, which does not cover systematic error. - Weakness in its own words: air helium diffuses into Pyrex flasks at about 2e12 atoms/day, comparable to the detection limit, so the early detection limit was "too small"; later runs used metal flasks. H2O controls gave no helium. - Radiation: anomalous but not reproducible; "never any significant production of tritium"; no foil activation. - Internal review only (not journal peer review).

4. Feinberg full read (Claim 4). Alvager-Kreisler limit (under 1/10,000 of pair production) verified [read: p.7]; antiproton-proton search null [read: p.9]; photon-stability argument [read: p.9]: if charged tachyons existed, photons would decay into tachyon pairs, so either their charge is many orders below the electron's or their mass squared is near zero; Bilaniuk-Deshpande-Sudarshan 1962 reinterpretation [read: p.5]; Feinberg's own 1970 view that nonexistence "now seems probable" [read: p.9]. Also: the 101 to 23 to 0 history, with the figure caption attributing the loss to "errors in the first set of measurements" while the body text blames mostly capture in flight; the limit is relative to pion production (under 1 in 400), not absolute.

5. Memo (Claim 1). Addressee "DT (Dr. Vorona)", originator "DT-ACO" (redacted name), redaction codes SG1B/SG1J, marking CONFIDENTIAL/NOFORN, Gaussian units confirmed, date 12/22/87, handwritten edits. Bearden is named in the memo as the chief proponent of the vacuum-energy idea [read: page images]. Authorship agency: web check failed; DT as DIA's scientific and technical directorate is [memory, medium], and RDP96-00792 as the STARGATE release is [memory, medium-high]. So "CIA analyst" is unsupported; use "an intelligence analyst (DT-ACO), memo held in CIA reading-room files".

Corrections to the arbiter and QE

• Brookhaven. The QE said "Brookhaven" is nowhere in the text and called the mathematician's [read] tag false. The deep read found it split across a line break as "Brook-/haven" in the p.3 figure caption, confirmed on the page image: "photographs ... originally made at the Brookhaven National Laboratory for another experiment" [read: Feinberg p.3]. So the mathematician's claim was true, though his tag was loose (it appears only in the caption). QE defect 2 (false tag) is withdrawn. My first-pass "ignore it" was also wrong, since I read only p.8.
• "DIA itself says that radiation was not seen" (Claim 2, point 3): wrong as written. DIA says it only of Fleischmann-Pons 1989 [read: p.1]. For work since 2009 DIA claims "nuclear particles". Replacement: DIA reports no radiation for Fleischmann-Pons 1989; for SPAWAR 2009 it cites CR-39 triple tracks, roughly 9 orders of magnitude below what 1 W of ordinary d+d needs, and above the 7.9 MeV threshold so not primary d+d neutrons anyway. Text above has been edited.
• Alvager-Kreisler 1/10,000: now verified; the "unchecked" caveat is removed.
• Memo date and agency: date confirmed 12/22/87; agency wording changed as above.

Verdict changes

Still open

• The cited papers (Naturwissenschaften 2009, Szpak, Iwamura, Kitamura/Arata) were not read; the Arata-note mismatch is [memory, medium].
• No independent, blind replication of the heat-helium ratio found; search budget ran out.
• Memo authorship agency is unverified.
• The tools/textprep.py zip gap is a pipeline issue outside this project.

---

Evidence-rule revision (2026-10-05)

Audit of every claim against the four SOURCE HONESTY rules in prompt.txt (no suppression of evidence; interested parties; absence of evidence; keep pertinent information). No sites.md or research/ exists for this project (no Investigator run), so two web searches were spent on the silences (budget: 2 of 8) [web: https://archive.org/details/CIA-RDP96-00792R000500240001-6 ; https://archive.org/details/cia-readingroom-document-cia-rdp96-00788r001200230021-2 ; https://arc.aiaa.org/doi/10.2514/6.1999-2143]. Findings were used by title only; no new document was read.

Verdicts. No verdict changed in substance. Item 8 restores scale labels for the DIA "accepted as reproducible" statement. Claims 1, 3, 4 and 5 verdicts stand: the new records (later memo-file item, Holt's later papers) are titles only and do not change what the read documents show; they are listed for the next Investigator run. Claim 4 (Feinberg) needed no change: the 101 to 23 to 0 history and all limits are already stated with sources.

---

Investigator findings (2026-10-05)

Inputs: sites.md ("Leads for review"), the eleven research/ notes dated 2026-10-05, and the "Investigator findings (2026-10-05)" sections of physicist.md, mathematician.md and contrarian.md. research/files/ is empty: the Investigator downloaded nothing, so every finding below is the Investigator's report of web pages, abstracts and search snippets [web, URLs as tagged in research/]. I re-did the arithmetic myself (Python venv) and made four web calls (budget 8): arXiv:2405.02307 abstract fetched and read [web: https://arxiv.org/abs/2405.02307]; the ARPA-E press-release URL returned 404 and the ARPA-E project-search page returned only navigation text; the CIA reading-room page for the Rein paper returned unrelated content, so the Rein table could not be checked against the page. Those three checks failed; the findings they would have confirmed stay at snippet or OCR level.

What changed

Claim 1

F1.1 Bearden's mechanism (CIA-RDP96-00792R000500240002-5, a staff note "TO: COL BUZBY", excerpts of Bearden's "Star Wars Now" of 24 April 1984 plus the writer's "logical extensions of my own") [web: https://www.cia.gov/readingroom/document/cia-rdp96-00792r000500240002-5]. In the note's words: Aharonov-Bohm (AB) shows potentials act "even in the absence of electrical and magnetic fields"; two identical beams "180 degrees out of phase" give "a constant electrical value of 0"; with no field change in transit there is "very little molecular resistance", so it "will penetrate ocean water, earth, and other materials"; two such beams crossing give a "beat frequency" with "a definite energy level" at the distant point; detection "can only be done in an area where time and space have been altered", by an antenna at the pole of a 40-80 kG (4-8 T) superconducting magnet in a Faraday cage. It also states military uses (jamming, a plasma "nuclear explosion" without gamma rays, weather modification, "at least 78 tests" by the Soviets) and names collaborators (Faretto, Dea, Bedini) and funders (Josh Reynolds, Essentia Research, the Association of Distinguished American Scientists). This is the proponent's side as summarised by a staff writer, not a test, and it is not necessarily Bearden's exact wording (his full paper is in the reading room and was not opened). It answers the gap I recorded ("no coupling specified"): the stated coupling is interference of two zero-field beams.

Where the reviewers agree. All three reach the same calculation: Maxwell's equations are linear, so two beams that sum to zero leave nothing to propagate, and if they do not sum exactly to zero each is an ordinary field attenuated by the seawater skin-depth factor exp(-z/delta) (about 25 m at 100 Hz). Worked number (contrarian, rounding delta to 25 m): at 100 m the amplitude factor is exp(-4) = 0.018; at 1 km exp(-40) = 4e-18. Recomputed by me with the unrounded skin depth (sigma = 4 S/m [memory, medium], f = 100 Hz, mu0 = 4 pi x 1e-7 H/m): delta = sqrt(2/(mu0 sigma 2 pi f)) = sqrt(2/(4 pi x 1e-7 x 4 x 628.3)) = sqrt(633.3) = 25.16 m, so 100/25.16 = 3.974 and exp(-3.974) = 0.0188 (about 0.019); 1000/25.16 = 39.74 and exp(-39.74) = 5.5e-18. These are the figures example.py Part 6 prints and the primer uses (the primer's "about 6e-18" rounds 5.5e-18). The two pairs differ only by the rounding of delta; the conclusion is the same. AB needs a non-zero enclosed flux, and a sum that is zero everywhere has none. Interference moves energy between places; it does not create a carrier. The detector statement predicts no force or energy at an ordinary detector and gives no control.

Where they differ. Label only. Mathematician: "Internally inconsistent" plus "Refuted by calculation". Physicist: strong form stays "Refuted by calculation"; detector "Not testable as stated". Contrarian: splits the seawater part into (3a) interference version "Refuted by calculation", (3b) a different non-EM carrier "Never properly tested", (3c) detection "Not testable as stated".

Arbitration. I take the contrarian's split for 3a and 3c, which matches the physicist, and keep my earlier ruling for 3b: with no detector response defined there is nothing to test, so "Not testable as stated", not "never properly tested" (I gave this reason under Claim 1; the contrarian's label applies only if someone supplies a coupling). The mathematician's "Internally inconsistent" is defensible (the note says the signal is zero in transit and also that it carries energy), but "Refuted by calculation" is the more informative label because it uses only the note's own stated beam properties and standard Maxwell theory. The strength of the argument: it is a dilemma with two exclusive cases, exact cancellation (nothing to carry) or partial cancellation (ordinary EM, attenuated). The proponent's way out would be a nonlinear element at the receiving point, which the note does not name; the mathematician says so and I agree this is what a proponent would have to supply. Caveats: the note is a summary; if Bearden's full paper states a different coupling, 3a must be rechecked (contrarian's caveat, accepted). The note's other statements (78 Soviet tests, plasma explosions, weather control, collaborators' work) carry no data and no sources; they are recorded as the note's claims and cannot be weighed.

F1.2 Rein's cell-culture experiment (CIA-RDP96-00792R000500240008-9, "Interactions with scalar energy - cellular mechanisms of action", Glen Rein, 6 pp., released 2001) [web: https://www.cia.gov/readingroom/document/cia-rdp96-00792r000500240008-9, via research/2026-10-05-cia-readingroom.md]. Rein, a supporter of the scalar idea, reports that a "Tesla watch" (a Moebius-strip coil credited to Andrija Puharich, said by Puharich to emit a "scalar B field" near 8 Hz) placed over an open dish of PC12 rat nerve cells for 30 minutes inhibited uptake of radiolabelled noradrenaline by 19.5% (control 5507 +/- 841 cpm/mg protein, n = 6; exposed 4435 +/- 480, n = 6; "p < 0.01" by t-test); no effect at 10 minutes; the control dish was about 12 feet away under a watch without the Moebius strip. Rein states the scalar field "was not measured" in related work and that with ever-present EM fields "it will be difficult to rule out" an EM cause. The paper also cites Byrd (EEG with the watch), Persinger (mast-cell degranulation under two cancelling 0.5 Hz, 10 uT "vortex" fields; Rein notes the scalar field "was not measured" there) and a Puharich/Max Planck E. coli gene-activation result; those were not read. Rein's proposed mechanism, which the Investigator's note reports from the OCR text and no reviewer discussed, is a "Crystalline Transduction Theory": scalar energy is transduced into ordinary EM energy by liquid crystals in cell membranes and by solid crystals in blood and tissue [web: https://www.cia.gov/readingroom/document/cia-rdp96-00792r000500240008-9, via research/2026-10-05-cia-readingroom.md]. This is the proponent's own account of how the cell effect would arise, by a supporter of the idea, and it is kept beside the experiment. Two points bear on weighing it: it gives no transduction coefficient or field strength, so it predicts no size of effect that could be checked; and if the effect works by conversion to ordinary EM energy inside the tissue, then what acts on the cell is an ordinary field, which is why the experiment cannot separate "scalar" from ordinary ELF exposure (the physicist's and mathematician's point above) [reasoned]. Rein's references also include Bearden's "AIDS: Biological Warfare" (1988) and the PACE newsletter (1982) [web, same note].

My check: (5507 - 4435)/5507 = 19.5%, reproduced. A pooled two-sample t-test with n = 6 per group (10 degrees of freedom): if the +/- is a standard deviation, t = 2.71, p = 0.022 two-sided (0.011 one-sided); if it is a standard error of the mean, t = 1.11, p = 0.29 (0.15 one-sided). All three reviewers agree with these numbers. No test gives "p < 0.01"; the nearest, a one-sided SD test, gives 0.011 (t = 1072/(sqrt((841^2 + 480^2)/2) x sqrt(2/6)) = 1072/(684.7 x 0.5774) = 2.71 on 10 degrees of freedom; one-sided p = 0.0109, recomputed with sympy.stats), just above the printed figure. The OCR of the table is partly garbled and the page image was not rendered by anyone, so I do not know what the +/- stands for or what test was run. A different n or a paired analysis could produce the printed value. This is a flag, not a dismissal. Weighing: it is a measured biological endpoint on a real cell line by a supporter of the principle, which is more than the memo had; one small experiment, one exposure pair, no replication on record, field not measured, control different in place and in object (two things changed at once), blinding not described. The physicist and mathematician both note that the experiment cannot discriminate "scalar" from ordinary ELF or thermal effects, and the mathematician adds that a true E = B = 0 field would by definition be invisible to any field measurement, so the experiment bears on neither side of the strong-form verdict. Verdict for the biological sub-claim: Never properly tested (all three agree in substance; physicist and contrarian use this label, the mathematician says the same in words). Not dismissed after testing, and not replicated.

F1.3 Who wrote the memo, and what followed (absence of evidence, re-analysed). The primary record is CIA-RDP96-00788R001500100012-3, a US Army INSCOM memo of 25 July 1984, "CENTER LANE Meeting with DIA - 20 July 1984", which names "Dr. Jack Vorona, Assistant Vice Director for Scientific and Technical Intelligence, DIA" and "DT (Dr. Vorona's office)", says Dr. Vorona would keep overall oversight within DIA of the psychoenergetic programme, and is signed by the CENTER LANE Project Manager, "BRIAN BUZBY ... LTC, MI" (OCR) [web: https://www.cia.gov/readingroom/document/cia-rdp96-00788r001500100012-3]. So "DT" in the scalar memo's address is the DIA directorate; "DT-ACO" is an office within it (inference from the office symbol; the memo does not name DIA). The "Col Buzby" of the staff note is very probably the same man (same surname, same programme; not confirmed). This corrects "CIA analyst" fully: it is a DIA document in a CIA release of the Army/DIA remote-viewing (STARGATE) files [web: every page read carries "Collection: STARGATE"]. The folder RDP96-00792R0005002400xx has 8 items [web: archive.org API listing, research/2026-10-05-archive-org.md]: 0001 the memo; 0002 the Bearden note; 0003-0007 five press clippings (Entebbe and similar; titles only, not read by anyone); 0008 Rein. Items 0001 and 0002 carry SG1B/SG1J redactions. No test report, tasking or assessment is in the released folder. Competing explanations, re-weighed: - (a) Judged empty and dropped. For: the memo's "not proven", the pure-gauge algebra. Against: nothing in the folder records a judgement either way; absence of a judgement is not a judgement (contrarian). Now less supported by the folder itself. - (b) Work was done and is withheld. For: the SG1B/SG1J redactions. Against: nothing positive points to work, only redaction (contrarian). Still open; it cannot be excluded by a released folder. - (c) The programme office was collecting proponents' open material. For: what the released folder actually contains (a proponent's summary, a proponent's experiment, unrelated clippings) filed with a remote-viewing office. All three reviewers and the Investigator read it so. This is now the best-fitting explanation of the released items; it is not proven, because (b) is compatible with the same folder. What would decide: the redacted and unreleased CENTER LANE/DT files, any DT tasking or response on "scalar", Bearden's full "STAR WARS NOW!" paper (in the reading room), and the untried searches. A caution from the physicist that I adopt: the folder shows no measurement of any kind and does not show that the DIA office believed the claim; the memo says "not proven". Still no suppression documented in what was read, and that is a statement about this folder only. Possible name coincidence ("Dr. Jack Dea" in the Bearden note; "Jack Y. Dea" on a 2010 CR-39 paper): identity unknown, no inference drawn (physicist, mathematician and contrarian all decline to infer).

Claim 2

F2.1 The DIA report's origin (New Energy Times) [web: https://newenergytimes.com/v2/news/2009/DIALENRReport.shtml]. The site is run by journalist Steven Krivit, who has an interest in the field's coverage; the statements below are its characterization. It says the report "was based on a workshop held on Aug. 4 and 5, 2009 at SPAWAR Pacific", with presenters Tanzella and McKubre (SRI), Swartz, Mosier-Boss, McDaniel, Forsley, Hagelstein (MIT) and McGinnis; the DIA report's own notes 22, 30, 44 and 45 cite briefings at that meeting, which agrees. The HTML copy carries "Prepared by: Beverly Barnhart, DIA/DI, Defense Warning Office" and "Coordinated with DIA/DRI, CPT, DWO, DOE/IN, US Navy SPAWAR/Pacific and U.S. NSWC/Dahlgren"; it reproduces an email of 23 Nov 2009 from "DIA Public Affairs": "This is a DIA document." (as posted; not independently verified; whether the coordination line is in the PDF has not been checked). How it bears: it explains why 16 of 54 notes trace to contributors, so the evidence base was largely a researchers' briefing to the agency, which is the point I made under "interested parties". The contrarian adds that coordination by a DOE intelligence office and a Navy laboratory is a mild institutional signal; I agree it is a fact to keep, but coordination is not endorsement of the data, and the Navy laboratory is also where the work was done. Reviewers' labels: physicist keeps "Tested, not supported" for the DIA wording; mathematician "Not testable as stated"; contrarian "not supported". My 2026-10-05 split stands (reproducibility: Never properly tested; "accepted": Not testable as stated). The physicist's own remark that "Tested, not supported" is a loose fit for a sentence about acceptance supports the split.

F2.2 The 2021 CR-39 re-examination (Mosier-Boss and Forsley, JCMNS 34(1):32-58) [web: https://jcmns.org/article/72567-interpreting-cr-39-detectors-used-in-pd-d-co-deposition-nuclear-generated-tracks-or-artifacts; abstract only]. Both authors are DIA contributors and JCMNS is the field's society journal, so this is the claimants' own characterization. The abstract reports "evidence of neutrons, notably 14.1 MeV neutrons", control experiments for chemical or mechanical damage, an SRI replication whose detectors "underwent exhaustive analysis by noted CR-39 experts from NASA and Russian Academy of Sciences", and answers to deuteroxide/O2 attack, D2/O2 recombination shockwaves and corona discharge. The 27-page body was not read by anyone. Arbitration of the two points all three reviewers make: (i) The threshold argument I used in the QE revision ("above the 7.9 MeV threshold, so not primary d+d neutrons") is correct for primary 2.45 MeV d+d neutrons, but it does not exclude 14.1 MeV d+t neutrons; the threshold is 7.275 x 13/12 = 7.88 MeV, so a 14.1 MeV neutron can make a triple track. I withdraw it as an objection to the track claim in general and keep it only as a statement about primary d+d neutrons. (ii) The rate argument is untouched: about 1e3 n/s inferred from the tracks against 1.9e12 n/s per watt on the neutron branch, about 9 orders short, so the tracks cannot be the heat channel. The inferred 1e3 n/s rests on an assumed detector efficiency [memory, low] and survives a factor-of-1000 error either way. A fairness point the contrarian makes and I adopt: the proponents offer the tracks as evidence that something nuclear happens, not as the heat channel, so the rate argument bears on the heat budget, not on whether the tracks are nuclear. A 14.1 MeV neutron needs d+t, hence a tritium inventory; the abstract does not say whether tritium was measured, and the reviewers note it is a further requirement (physicist: d+t needs tritium first, at a rate far above anything shown). An examination by outside CR-39 experts is a better design than self-examination, as the contrarian says; their reports have not been seen. Verdict: unchanged. The nuclear-particle evidence is contested, tested largely by the proponents and re-examined by them, not independently replicated on the record held; Never properly tested for the unknown-mechanism part. If the NASA and Russian examiners' independent reports are obtained and support the tracks, the contrarian rightly says the part should be reconsidered.

F2.3 The Kowalski comment (L. Kowalski, Eur. Phys. J. Appl. Phys. 2008, article ap08217; authors' reply EPJ AP 44:291-295, 2008) [web, bibliographic only; the site returned a bot check, research/2026-10-05-epjap.md]. The critique exists as a published comment with a published reply, so my earlier [memory, medium] item is now a cited bibliographic fact. Its argument is unread by everyone (the search snippet mentions "highly-localized regions of damaged material" and etching), so I weigh nothing from it. A critique that was published and answered in print is also evidence against the idea that the track claim was ignored.

F2.4 Helium-4: two findings pulling opposite ways. - ENEA, De Ninno, Frattolillo, Rizzo, Del Giudice, Preparata, "Experimental evidence of 4He production in a cold fusion experiment", RT/2002/41/FUS, 26 pp. [web: https://www.lenr-canr.org/acrobat/DeNinnoAexperiment.pdf; title page and snippets only; lenr-canr.org is an archive run by supporters of the field]. Snippet: excess heat and helium "simultaneously measured as a function of short time intervals (less than 1 hour)" on an electrolytically loaded Pd stripe. It is a national-laboratory report from a group separate from China Lake and SRI, which the contrarian counts in favour: hour-scale correlation is harder to explain by slow background drift. Del Giudice and Preparata proposed a coherence theory of the effect [memory, medium, per the Investigator], so two authors have a stake. Pages 2-26 not read: calibration, blanks, air-leak controls and power are unknown. It is a positive heat-and-helium report from a third group, kept in the record, and an unread report cannot carry a verdict. - Gillespie, Lin, Jones, Duncan (Texas Tech), "Helium Detection in Technical Materials", arXiv:2405.02307 (2024) [web: https://arxiv.org/abs/2405.02307; abstract read by me this run]: untreated stainless steel 316 retains 15-240 pmol of He-4, may reabsorb it from the atmosphere, and pretreatment is necessary when He-4 is used as evidence of a nuclear reaction. It measured no heat. The Investigator's link to ARPA-E funding came from a search snippet; the arXiv page names no ARPA-E involvement (it acknowledges Simons Foundation and Schmidt Sciences generally), so the link stays unconfirmed. - Arithmetic, which all three reviewers and I agree on: 1 pmol = 6.02e11 atoms; at 2.62e11 He-4 atoms per joule, 15-240 pmol is 34 to 552 J of d+d heat. A 0.5 W cell for one day (43,200 J) makes 1.13e16 atoms (about 19 nmol): the steel background is 0.08% to 1.3% of it. For a one-hour interval at the same illustrative 0.5 W (1,800 J, about 780 pmol), 15-240 pmol is 1.9% to 31% of the signal (contrarian). The 0.5 W is illustrative only; the ENEA power is not known. Lesson, agreed: for day-long watt-level runs the steel-wall background is a small correction, not an explanation of a heat-matched yield; for short or low-power runs it is comparable to the signal and a pretreated-hardware control is required. For comparison, Miles (1996) reported air helium diffusing into Pyrex flasks at about 2e12 atoms per day, about 8 J equivalent [calculation: 2e12 / 2.62e11]. - Further heat-helium and neutron leads named in the research notes, not opened by anyone (kept so the next run can obtain them): Gozzi et al. (1998), a heat-helium correlation report, and Arata and Zhang (1998), a helium report, both named by a search snippet from a Wikiversity mirror [web snippet, research/2026-10-05-lenr-canr.md]; the physicist also names both. De Ninno et al., "4He detection in a cold fusion experiment" (lenr-canr.org DeNinnoAhedetectio.pdf; a method for detecting small amounts of He in gas mixtures) [web snippet, same note]. On particles: "Comparison of Pd/D co-deposition and DT neutron generated triple tracks observed in CR-39 detectors", Eur. Phys. J. Appl. Phys. 51, 20901 (2010), and "Characterization of Neutrons Emitted during Pd/D Co-deposition" (JCMNS) [web snippets, research/2026-10-05-jcmns.md]; the contrarian names the 2010 paper. Stakes: lenr-canr.org is run by supporters of the field; Arata and Gozzi are proponent-side experimenters reporting their own results; the two CR-39 papers come from the SPAWAR co-deposition programme (the DIA contributors' own line of work) and JCMNS is the field's society journal. The 2010 paper compares co-deposition tracks with tracks made by DT (deuterium-tritium) generator neutrons, which bears directly on F2.2: if co-deposition tracks match 14.1 MeV DT-neutron tracks, the threshold point there is moot and the open question becomes tritium and rate. None was read, so none carries a verdict. - Effect on the verdict: none. "Unknown nuclear mechanism (He-4 correlation)" stays Never properly tested: the new items are one more positive proponent-side report that I cannot read and a quantified systematic that any neutral test must beat; neither is a neutral, blinded, controlled test. Interested parties are stated beside each.

F2.5 ARPA-E. The Investigator found by search snippet an ARPA-E Exploratory Topic on LENR (13 Sep 2022) and 17 Feb 2023 funding of $10 million for eight projects "to determine whether" LENR could be the basis for energy technology [web: arpa-e.energy.gov, not opened]. My two fetch attempts failed, so it is still snippet-level. If accurate, it is a second US government body funding a test of the question, 13 years after the DIA report and 19 years after the 2004 DOE review that recommended no federal programme. That is evidence the question has not been closed institutionally; it is not a result. No DIA follow-up report was found. Competing explanations for the lack of a DIA follow-up (one-off warning product; no further interest; classified follow-up) cannot be separated on the record held. Verdict: unchanged.

Claim 3

F3.1 Holt's later NASA record (NTRS abstracts) [web: https://ntrs.nasa.gov/search?q=Alan%20C.%20Holt%20propulsion]. AIAA 89-2858 (1989): candidate payloads "include magnetic plasma thruster systems and a hypothetical advanced electromagnetic propulsion system utilizing high-temperature superconductivity materials", which "can be tested on the Space Station"; "Spaceport operations for deep space missions" (Vision-21, 1990): "energetic, breakthrough level propulsion research" should "be initiated now"; "Space Station technology testbed: 2010 deep space transport" (1993; in-space "tuning" of micro-engineered materials, per the contrarian); a Space 2004 abstract ("an advanced space propulsion testbed"); a 1986 Mars-support paper with no field propulsion. Not found in this search: the 1980 AIAA, 1982 Science Digest and 1999 AIAA (Holt, Davis, Puthoff) items, which I had listed by title. Weighing: this answers part of the absence-of-evidence question, because Holt's record does not go silent after 1979; he stayed in NASA programme offices and kept proposing tests for 25 years. That is evidence against suppression of the author, and for persistence; the contrarian notes there is no anomaly of absence to analyse for Holt himself. None of the abstracts defines a "space-time harmonic" or has equations. The physicist's gloss that a closed magnet gives no net thrust without reaction mass or an external field is physics applied to a proposal known only from an abstract; I record it as her reading, since I cannot tell whether the 1989 "hypothetical advanced electromagnetic propulsion system" still means the 1979 resonance idea. The mathematician's remark that continued proposals to test are a mild point in favour of the author's good faith is an observation about conduct, not about the physics, and I record it as such. Not found: any record that a test was flown or run on the Space Station (absence in a one-site search is not a finding). Verdict: unchanged (propulsion idea Not testable as stated; flare premise Refuted by calculation). Unread, public: NTRS 19910012853 and 19940022865; if either defines the resonance quantity, "Not testable as stated" must be revisited.

Claim 4

No finding bears on it. Unchanged.

Claim 5

F5.1 The Sharp-Grove speech demonstration. Per Martin Broughton Research & Analysis, a secondary analysis site [web: https://www.martin-broughton-research-analysis.co.uk/the-microwave-auditory-effect/encoded-communication-morse-code-speech], the public source is D. R. Justesen, "Microwaves and Behavior", American Psychologist 30(3):391-401 (March 1975), whose reference list gives "Sharp, J. C., & Grove, M. Personal communication, September 28, 1973". Method as reported: recorded speech converted into microwave pulses timed to the waveform; "recognizable words were reportedly perceived". Broughton also reports a Morse-code demonstration by A. W. Guy (University of Washington) in the same article. The published Sharp, Grove and Gandhi paper (IEEE Trans. MTT 22(5):583-584, 1974) is about acoustic generation, not speech. The Investigator found a scan of the Justesen article [web: https://zoryglaser.com/wp-content/uploads/2020/05/MICROWAVES-AND-BEHAVIOR.pdf] and read p.1 only; the page with the Sharp-Grove citation has not been seen, so Broughton's reading of the citation is unchecked. Net: the speech demonstration is documented in one published account by a named researcher, citing the experimenters by personal communication; no Walter Reed report and no power levels found. I weigh it as reported, not as dismissed and not as replicated; silence elsewhere is not contradiction, and this is a usual way for a result in this field to reach print. All three reviewers agree it is physically consistent with the Frey effect (each pulse gives a thermoelastic click; pulse timing can encode an audio envelope), and that it conveys sound, not thought. The mathematician notes the missing numbers (peak power density, pulse width, repetition rate) prevent any feasibility check. Verdict: physicist "Plausible, untested"; mathematician and contrarian leave Frey "Established" and treat the speech report as a documented single-source extension. I adopt Plausible, untested for "speech conveyed by pulse-timed microwaves", stating that a reported demonstration exists but no controlled, documented measurement. Frey effect stays Established; remote thought insertion and radar ID stay Not testable as stated; focused standoff ELF stays Refuted by calculation. Date correction to my earlier text: the supported dates are September 1973 (personal communication), 1974 (the mechanism paper) and 1975 (the published account). The infographic's "1974 ... first unclassified" wording is not supported by a 1974 Walter Reed report on speech, but it lies between the dates and is imprecise rather than clearly wrong; I withdraw "unverified" in favour of this. Unresolved between reviewers: the physicist quotes "peak below 80 mW/cm^2 at 1.245 GHz" for early Frey work while my earlier text carried the contrarian's 267 mW/cm^2 for single pulses at 1.3 GHz; I have not checked either. No verdict depends on it.

Corrections to the reviewers made in this pass

• Mathematician, Claim 1: "Internally inconsistent" for the Bearden chain is kept as a description but the scale label I adopt is "Refuted by calculation" (above).
• Contrarian, Claim 1 (3b): "Never properly tested" for a non-EM carrier is changed to "Not testable as stated", consistent with my earlier ruling that no detector response is defined.
• Physicist, Claim 2: "Tested, not supported" for the DIA wording stands in her file; my split label supersedes it.
• Mathematician, Claim 2: "Tested, not supported" for nuclear origin stands in his file. My table keeps "Never properly tested" for the unknown mechanism and "Refuted by calculation" for ordinary d+d, as before; none of the new findings changes the split.
• Mathematician: the neutron-threshold reasoning in the QE revision is corrected (F2.2 above).
• Contrarian: ARPA-E "second US government institution" statement kept, flagged as snippet-level.

Verdict changes from this section

Still not read (for the next Investigator run)

Bearden's "STAR WARS NOW!" and other reading-room hits; the five clippings (0003-0007); ENEA pages 2-26; De Ninno et al. "4He detection in a cold fusion experiment" (lenr-canr.org); Gozzi et al. (1998) and Arata and Zhang (1998) heat-helium reports (proponent-side authors; found via a lenr-canr/Wikiversity snippet); EPJ AP 51, 20901 (2010), "Comparison of Pd/D co-deposition and DT neutron generated triple tracks"; JCMNS "Characterization of Neutrons Emitted during Pd/D Co-deposition" (SPAWAR programme, field society journal); Rein's full text for the "Crystalline Transduction Theory" and his Byrd, Persinger and Puharich/Max Planck sources; the 2021 JCMNS body; the NASA and Russian CR-39 examiners' reports; the Kowalski comment (blocked); Springer and Academia copies of Naturwissenschaften 2009; Justesen 1975 page carrying the Sharp-Grove citation; DTIC (Sharp, Walter Reed; DIA DT); AIAA Arc (Holt 1980); NTRS 19910012853 and 19940022865; ARPA-E project results; the Rein paper's table as a page image. The sites listed in sites.md as "not reached (time limit)" and its "Not yet tried" queries carry over.

QE remediation of this section (2026-10-05, after qe.md "Investigator update check")

• QE defect 1a (retention): Rein's "Crystalline Transduction Theory" added to F1.2, quoted from the Investigator's note with its source and stake, with Persinger's 10 uT field strength, Rein's "not measured" caveat and his Bearden/PACE references.
• QE defect 1b (retention): Gozzi et al. (1998), Arata and Zhang (1998), De Ninno "4He detection", EPJ AP 51, 20901 (2010) and JCMNS "Characterization of Neutrons Emitted during Pd/D Co-deposition" added to F2.4 with their stakes, and to "Still not read" (with Rein's full text and sources).
• QE defect 3a (wording): F1.2 now says no test gives p < 0.01; the nearest, one-sided SD, gives 0.011 (t = 2.71, 10 degrees of freedom, p = 0.0109, recomputed).
• QE defect 3c (consistency): F1.1 seawater factors recomputed with the skin depth stated: delta = 25.16 m (sigma 4 S/m, 100 Hz) gives 0.019 at 100 m and 5.5e-18 at 1 km, as in example.py Part 6 and the primer; the contrarian's 0.018 and 4e-18 are the same calculation with delta rounded to 25 m. No verdict changes.
• QE defects 2a, 2b and 3b concern primer.md and are left to the primer author.