Reply to "Counterfactual communication not achieved yet -- A Comment on Salih et al. (2022)"

Hatim Salih; John Rarity; Jonte R. Hance; Will McCutcheon

REVIEW 3 major objections 5 minor 19 references

Reply to "Counterfactual communication not achieved yet -- A Comment on Salih et al. (2022)"

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper argues that a critique of counterfactual communication replaces the postselected protocol with a different one, and that under the actual protocol any arbitrarily long binary message can be sent with arbitrarily high accuracy…

desk verdict A competent reply that scores a real point about Popescu's simplification, but the toy-to-protocol transfer is asserted without derivation and the larger counterfactuality dispute remains open. read the letter →

arxiv 2507.19563 v1 pith:UC7RLSAD submitted 2025-07-25 quant-ph physics.hist-ph

Hatim Salih , Jonte R. Hance , Will McCutcheon , John Rarity This is my paper · ORCID

classification quant-phphysics.hist-ph

keywords counterfactualcommunicationpostselectionweaktraceconsistenthistoriesnestedMach-Zehnderinterferometerinteraction-freemeasurementquantum

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This is a reply to a Comment that claimed the 2022 counterfactual communication protocol does not work. The reply argues that the Comment misreads the claim: counterfactuality applies only to postselected runs, the ones in which Alice's photon triggers detector D0 or D1, and all other runs are discarded. It further argues that the Comment's simplified setup is a different protocol—one that can successfully transmit essentially only an all-ones message—whereas the original protocol lets Bob send any arbitrarily long binary string with arbitrarily high accuracy. If the reply is right, counterfactual communication remains compatible with the weak trace and consistent histories criteria for particle path, and the central 2022 conclusion stands.

What carries the argument

The load-bearing object is the postselected nested-interferometer protocol: a single-outer-cycle, Michelson-type layout in which Bob encodes a bit by blocking (0) or not blocking (1) the inner interferometer, and Alice keeps only outcomes D0 or D1. Postselection on those two detectors is what makes the counterfactuality claim well-defined: every discarded run is removed from the data, so the accepted message contains only rounds in which the photon demonstrably did not reach Bob, by the weak trace criterion and the consistent histories framework. The single-outer-cycle design is chosen so that those path analyses are tractable while still reproducing the no-interference-in-outer-arm behaviour relevant to bit 0 in the full multi-cycle protocol.

What would settle it

A sensitive probe of whether the photon spends any time in Bob's arm, conditioned on Alice detecting the photon at D0 or D1 in the full multi-outer-cycle protocol, would settle the claim: any nonzero trace at Bob's location would show that a particle did visit him in an accepted run, contradicting the central counterfactuality claim.

Watch

Extended reading notes

Core claim

The central claim is that the Comment attacks a protocol the authors did not present. In the protocol under discussion, Alice accepts a round only if her photon arrives at D0 (bit 0) or D1 (bit 1); any round ending at D3 or at Bob's blocker is thrown out along with the whole message it belongs to. Under that postselection, the reply states, any N-bit message chosen by Bob can be transmitted, with the accuracy made as high as desired by discarding more runs, and in every accepted run the photon has not been to Bob according to both the weak trace and consistent histories path criteria. The reply also defends the single-outer-cycle toy setup used in the 2022 paper as sufficient for testing exactly those path criteria, and notes that the Comment's coin-toss simplification changes the statistics rather than merely simplifying the optics.

Load-bearing premise

The defense rests on the assumption that the simplified one-cycle version of the interferometer behaves like any single cycle of the full multi-cycle protocol; the Comment disputes this transfer, and the reply asserts it without presenting a new derivation.

Editorial extensions

If this is right

If the reply is correct, the Comment's simplification does not describe the same communication task: it can successfully transmit essentially only the all-ones message, while the postselected protocol transmits any N-bit string.
Any message that survives postselection is fully correlated with Bob's bits, so Alice never has to extract information from failed runs; discarding them does not weaken the communication claim.
The weak trace and consistent histories criteria, which motivated earlier objections, do not by themselves rule out counterfactual communication; the demonstration for one outer cycle is meant to transfer to each outer cycle of the full protocol.
Because the claimed result holds for postselected particles, it supports extending the same reasoning to counterfactual qubit transport, counterfactual ghost imaging, and exchange-free computation.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

Editorial inference: the reply effectively concedes that postselection does real work—Alice learns a message is invalid precisely because a photon went to Bob in other runs—so the result is best read as 'communication is possible using postselected counterfactual outcomes,' not as a claim that every emitted photon is counterfactual.
Editorial inference: the dispute is partly about which definition of a particle's path is authoritative; on a different path criterion, the same optics could receive a different counterfactuality verdict.
Editorial inference: a direct weak-trace calculation for the full multi-outer-cycle protocol, rather than the single-cycle analogue, would close the main gap the Comment identifies; the reply leaves that transfer asserted.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Colleague,

The reply makes one sharp point that survives contact with Popescu's Comment: his coin-free simplification is not their protocol. If you run the original protocol once per bit, the only N-bit message that can survive postselection is the all-ones string. Any other message fails, and you cannot fall back on the failed runs without using information Popescu said was unnecessary. That is a correct and pointed observation, and the authors are right to insist that the Comment's scheme has different statistics.

The paper is also honest about its own limits. They concede the toy protocol gives erroneous '0-bit' clicks that the multi-cycle version suppresses, and they admit that alternative definitions of particle presence would change the counterfactuality verdict. Those admissions are welcome.

The soft spot is the transfer step. The stress-test note is on target: the reply asserts, without deriving, that conclusions from the single-outer-cycle toy apply to every outer cycle of the 2013 protocol. The toy lacks outer-interferometer interference; the full protocol uses that interference precisely to fix the erroneous clicks. The authors try to have it both ways—saying no outer interference matters in the relevant case, while also relying on outer interference to make the full protocol work. If the toy really were representative, the full protocol would inherit its errors, and if the full protocol really suppresses them, the toy is not representative. The reply gives no calculation that bridges this, only a citation to [11].

The broader claim that postselected photons never visit Bob also remains criterion-relative. The reply's no-particle conclusion is established for the weak-trace and consistent-histories notions of path, and the authors themselves note that other criteria yield other answers. That is not a flaw in the reply's narrow logical point, but it does mean the reply does not decisively resolve the underlying controversy.

Who should read this? Anyone tracking the counterfactual communication exchange. It is a competent reply, not a new result. The narrow point about the all-ones message is worth having on record. If this is submitted as a formal reply, it deserves peer review; a referee should focus on the toy-to-full-protocol transfer and ask for a derivation. I would not cite it in my own work, but I would bring it to the attention of people working on counterfactual protocols.

Recommendation: send for peer review if the venue publishes replies. The logical rebuttal is sound, but the transfer step needs to be defended properly.

Referee Report

3 major / 5 minor

Summary. This manuscript is a reply to Popescu's Comment on Salih et al. (2022), 'The laws of physics do not prohibit counterfactual communication'. The authors argue that Popescu's simplification is not equivalent to their protocol: with one run per bit, the simplified scheme can only communicate the all-1s message; with N runs per bit, it uses failed-run information that the original protocol discards. They further argue that, in the postselected case (detections at D0 or D1), their protocol lets Bob communicate an arbitrarily long binary message with arbitrarily high accuracy, and that by the weak trace and consistent histories criteria the postselected photons have never been to Bob. The reply relies on a single-outer-cycle toy protocol to establish the no-particle conclusion and asserts that this conclusion transfers to the full multi-cycle protocol that transmits arbitrary messages.

Significance. If the reply's central claims are correct, it would refute Popescu's Comment and defend the original paper's conclusion that the laws of physics do not prohibit counterfactual communication under the weak trace and consistent histories criteria. The reply usefully clarifies the distinction between the actual protocol and Popescu's simplification, and it correctly emphasizes that postselection is an essential part of the protocol. However, the significance is limited because the key no-particle conclusion is established only for a toy protocol, while the transfer to the full protocol is asserted without derivation. The paper also explicitly concedes that the conclusion is criterion-relative, which tempers the strength of the claim as stated in the abstract.

major comments (3)

[Toy protocol transfer (paragraphs 2–3)] The reply's central claim—that postselected photons never visit Bob in the protocol that actually transmits arbitrary messages—depends on transferring the single-outer-cycle toy protocol's no-particle conclusion to the full multi-cycle protocol. This transfer is asserted, not derived. The reply concedes that the toy protocol 'can give erroneous “0-bit” clicks, while in two-or-more outer-cycle versions of Salih et al's protocol, this is not the case' and that the toy 'lacks features (e.g., interference between inner and outer path of outer interferometer) which Salih et al's 2013 protocol possesses.' It then asserts that this does not matter because 'when the inner interferometers are not blocked—i.e., bit “0”—no interference should take place in the outer interferometers, just as in the present single-outer-cycle protocol.' This is internally inconsistent: if no outer interference occurred in the relevant case, the multi-cycle protocol could not suppress the toy's erroneous clicks; if it does suppress them, the multi-cycle protocol has different interference properties in exactly the regime the no-particle conclusion must cover. The only support offered for the transfer is a citation to [11] without a specific result. The authors need to provide an explicit derivation or a precise theorem statement from [11], or limit the no-particle claim to the toy protocol.
[Bit labeling in the transfer argument] The bit labeling in the transfer argument is inconsistent with the protocol definition quoted earlier in the reply. The reply quotes from [2]: 'He blocks (does not block) his channel when X = 0 (X = 1).' But then it states: 'when the inner interferometers are not blocked—i.e., bit “0”—no interference should take place in the outer interferometers.' If bit '0' corresponds to blocking, then 'inner interferometers are not blocked' describes bit '1'. This inconsistency obscures which physical configuration is being analyzed. Since the rebuttal of Popescu's simplification depends on correctly identifying the 'counterfactual' bit configuration, the authors must correct this inconsistency and restate the argument coherently.
[Criterion-relativity of the conclusion] The reply both claims that the communication protocol is 'not a matter of interpretation' and concedes that 'one could quite easily come up with alternative definitions for particle presence, which would lead to different conclusions' and that a protocol being counterfactual 'by said criteria does not rule out there being indirect influences between Alice and Bob.' These statements are in tension: the abstract and conclusion assert that 'no particles are exchanged in the course of this communication' and 'the laws of physics do not prohibit counterfactual communication,' which read as unconditional claims. As written, the reply oscillates between a criterion-relative claim (about weak trace and consistent histories) and a stronger unconditional claim. The authors should either state the conclusion explicitly as conditional on these criteria throughout, or argue why these criteria are the correct ones for judging counterfactual communication.

minor comments (5)

[Abstract] The abstract contains a typo: 'creteria' and a double closing parenthesis 'path)).'
[Paragraph 1] The sentence 'one of the two aims of the protocol we construct in our paper construct one where' is ungrammatical and should be rewritten.
[Paragraph 3] The phrase 'as does reference [11]' is vague; the authors should cite a specific section or theorem in [11] that establishes the transfer from the single-outer-cycle toy protocol to the full protocol.
[Paragraph 3] The terms 'outer cycle' and 'inner interferometer' are used without definition; a brief definition or a pointer to the original paper would help readers who do not have [2] or [3] at hand.
[Conclusion] The conclusion states that the reply has 'prove[n] experimentally using a setup consisting of a single outer cycle' the central claim, but the experimental setup is not described in this reply; citing [2] is acceptable, but the wording may overstate the directness of the evidence presented here.

Circularity Check

1 steps flagged · score 4.0 of 10

The reply's transfer of the single-outer-cycle toy protocol's conclusions to the full multi-cycle protocol rests on a self-citation by the same first author, though the central claim otherwise has independent published support.

self citation load bearing [Reply body, paragraph discussing the Fig. 1 caption and applicability to outer cycles (after 'so that the setup exactly includes the equivalent of one outer cycle...').]
"This allows the conclusions drawn for this single-outer-cycle protocol to be applicable to any of the outer cycles of Salih et al’s 2013 counterfactual communication protocol, as does reference [11]."

The Comment's central dispute is exactly whether the single-outer-cycle toy is representative of the full multi-cycle protocol. The reply's only cited support for that transfer is reference [11], a prior paper by the same first author, which is itself part of the same counterfactual-communication program. No new derivation is supplied; the accompanying assertion that no outer interference takes place in the bit-0 case is in tension with the reply's own admission that the toy produces erroneous '0-bit' clicks that multi-cycle versions suppress via outer-cycle interference. Thus the load-bearing premise that toy conclusions apply to the full protocol is justified by the authors' own prior work rather than by an independent argument in this reply.

full rationale

This reply does not derive a new formal result; it defends the protocol of the original paper [2] against Popescu's simplifications. Most of the argument is a protocol-comparison exercise, which is non-circular: the reply correctly notes that removing the coin toss and guessing zeros changes the statistics and therefore the protocol, and that the original protocol uses postselection in a specific way. The one load-bearing self-citation is the assertion that conclusions drawn for the single-outer-cycle toy protocol apply to any outer cycle of the 2013 protocol, supported only by 'as does reference [11]', a prior paper by the same first author. The Comment explicitly disputes this applicability, and the reply offers no independent derivation of it, only the further assertion that no outer interference occurs in the bit-0 case—an assertion that sits uneasily with the reply's own admission that multi-cycle versions suppress erroneous '0-bit' clicks through outer-cycle interference. Because the transfer step is central to the reply's defense, this is more than a passing self-citation, but the underlying counterfactual-communication claim has independent content in the published experimental and theoretical work of [2], so a moderate score of 4 is appropriate rather than a higher score indicating full circularity.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The reply's central claims rest on (1) the legitimacy of postselection as a way to define counterfactual communication, (2) the validity of weak trace and consistent histories criteria for particle path, and (3) the representativeness of the single-outer-cycle toy protocol. None of these is independently established in this reply; they are domain assumptions carried from the authors' prior work and contested by the Comment.

assumptions (3)

domain assumption Postselection on Alice's detectors D0/D1 is a legitimate basis for inferring counterfactual communication
The entire reply assumes that discarded failed runs carry no information relevant to the message, so Alice's successful runs alone establish counterfactuality. The authors themselves flag that this could be challenged.
domain assumption Weak trace and consistent histories criteria are valid criteria for the path of a quantum particle
The claim that no photons crossed the channel is verified only by these criteria, which are contested; the authors admit they do not advocate these criteria.
domain assumption The single-outer-cycle toy protocol is representative of the full 2013 counterfactual communication protocol
The reply relies on transferring conclusions from the toy protocol to the original protocol; Popescu disputes this applicability.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Reply to "Counterfactual communication not achieved yet -- A Comment on Salih et al. (2022)"." pith.science (2026). https://pith.science/paper/UC7RLSAD

@misc{pith2026250719563,
  author       = {Pith},
  title        = {Pith review of: Reply to "Counterfactual communication not achieved yet -- A Comment on Salih et al. (2022)"},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/UC7RLSAD}},
  note         = {Machine review of arXiv:2507.19563}
}

read the original abstract

In his Comment on our recent paper ``The laws of physics do not prohibit counterfactual communication'', \textit{npj Quantum Information} (2022) 8:60, Popescu argues that the claims of the paper are invalid. Here, we refute his argument, showing that it is based on ignoring the specifics of what we set out to prove (that counterfactual communication is possible \emph{for post-selected particles}, and more specifically in these cases is not prohibited by the weak trace or consistent histories criteria for particle path), followed by an unwarranted simplification of the protocol. Moreover, the Comment's excursion into interpretation is misplaced. Our communication protocol is a precisely defined one that allows two remote parties, albeit rarely, to communicate an arbitrarily long binary message, with arbitrarily high accuracy. This is not a matter of interpretation -- as the concrete example given in our paper in question illustrates. As for our overarching claim that no particles are exchanged in the course of this communication, we have already demonstrated this both theoretically and experimentally, in the postselected case we consider, as per the weak trace and consistent histories criteria for path of a quantum particle.

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

19 extracted references · 18 canonical work pages

[11]

From counterportation to local wormholes

Hatim Salih. From counterportation to local wormholes. Quantum Science and Technology , 8(2):025016, mar 2023. doi:10.1088/2058-9565/ac8ecd

work page doi:10.1088/2058-9565/ac8ecd 2023
[2]

The laws of physics do not prohibit counterfactual communication

Hatim Salih, Will McCutcheon, Jonte R Hance, and John Rarity. The laws of physics do not prohibit counterfactual communication. npj Quantum Information , 8(1):60, 2022. doi: 10.1038/s41534-022-00564-w

work page doi:10.1038/s41534-022-00564-w 2022
[1]

Counterfactual communication not achieved yet – A Comment on Salih et al

Sandu Popescu. Counterfactual communication not achieved yet – A Comment on Salih et al. (2022) npj Quantum Information , 11(1):122, 2025. doi:10.1038/s41534-025-00975-5

work page doi:10.1038/s41534-025-00975-5 2022
[3]

Protocol for direct counterfactual quantum communication

Hatim Salih, Zheng-Hong Li, Mohammad Al-Amri, and Muhammad Suhail Zubairy. Protocol for direct counterfactual quantum communication. Phys. Rev. Lett. , 110:170502, Apr 2013. doi:10.1103/PhysRevLett.110.170502

work page doi:10.1103/physrevlett.110.170502 2013
[4]

L. Vaidman. Past of a quantum particle. Phys. Rev. A , 87(5):052104, 2013. doi:10.1103/ PhysRevA.87.052104

work page 2013
[5]

Consistent Quantum Theory

Robert Griffiths. Consistent Quantum Theory . Cambridge University Press, Cambridge, 2002 edition, 2002

work page 2002
[6]

L. Vaidman. Counterfactuality of ’counterfactual’ communication. J. Phys. A , 48(46) (2015). doi:10.1088/1751-8113/48/46/465303

work page doi:10.1088/1751-8113/48/46/465303 2015
[7]

Direct counterfactual transmission of a quantum state

L. Vaidman. Comment on “Direct counterfactual transmission of a quantum state”. Phys. Rev. A, 93(6):066301, 6 (2016). doi:10.1103/PhysRevA.93.066301

work page doi:10.1103/physreva.93.066301 2016

Show all 19 references

[8]

Protocol for Direct Counterfactual Quantum Communication

Lev Vaidman. Comment on “Protocol for Direct Counterfactual Quantum Communication”. Phys. Rev. Lett. , 112(20):208901, 5 (2014). doi:10.1103/PhysRevLett.112.208901

2014 doi
[9]

Three approaches for analyzing the coun- terfactuality of counterfactual protocols

Alon Wander, Eliahu Cohen, and Lev Vaidman. Three approaches for analyzing the coun- terfactuality of counterfactual protocols. Phys. Rev. A , 104(1):012610, 16 (2021). doi: 10.1103/PhysRevA.104.012610

2021 doi
[10]

Griffiths

Robert B. Griffiths. Particle path through a nested Mach-Zehnder interferometer. Phys. Rev. A, 94(3):032115, 9 (2016). doi:10.1103/PhysRevA.94.032115

2016 doi
[12]

Hance, John Rarity, and James Ladyman

Jonte R. Hance, John Rarity, and James Ladyman. Weak values and the past of a quantum 7 particle. Phys. Rev. Research, 5(2):023048, 2023. doi:10.1103/PhysRevResearch.5.023048

2023 doi
[13]

Weak values and the past of a quantum particle

Jonte R. Hance, John Rarity, and James Ladyman. Comment on “Weak values and the past of a quantum particle”. Phys. Rev. Research , 5(4):048001, 2023. doi:10.1103/ PhysRevResearch.5.048001

2023
[14]

Comment on ‘Weak values and the past of a quantum particle’

Jonte R. Hance, John Rarity, and James Ladyman. Reply to “Comment on ‘Weak values and the past of a quantum particle’ ”. Phys. Rev. Research , 5(4):048002, 2023. doi:10.1103/ PhysRevResearch.5.048002

2023
[15]

Hance, Tomonori Matsushita, and Holger F

Jonte R. Hance, Tomonori Matsushita, and Holger F. Hofmann. Counterfactuality, back- action, and information gain in multi-path interferometers. Quantum Sci. Technol. , 9(4):045015, 2024. doi:10.1088/2058-9565/ad63c7

2024 doi
[16]

After the paper: Traversable wormhole in the laboratory

Hatim Salih. After the paper: Traversable wormhole in the laboratory. Springer Na- ture Physics Community , Mar 2023. URL: https://communities.springernature. com/posts/after-the-paper-traversable-wormhole-in-the-laboratory?badge_id= npj-quantum-information

2023
[17]

Protocol for counterfactually transporting an unknown qubit

Hatim Salih. Protocol for counterfactually transporting an unknown qubit. Frontiers in Physics, 3:94, 2016. doi:10.3389/fphy.2015.00094

2016
[18]

Exchange- free computation on an unknown qubit at a distance

Hatim Salih, Jonte R Hance, Will McCutcheon, Terry Rudolph, and John Rarity. Exchange- free computation on an unknown qubit at a distance. New Journal of Physics , 23(1):013004, 1 2021. doi:10.1088/1367-2630/abd3c4

2021 doi
[19]

Hance and John Rarity

Jonte R. Hance and John Rarity. Counterfactual ghost imaging. npj Quantum Information , 7(1):88, Jun 2021. doi:10.1038/s41534-021-00411-4. 8

2021 doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [11]

From counterportation to local wormholes

Hatim Salih. From counterportation to local wormholes. Quantum Science and Technology , 8(2):025016, mar 2023. doi:10.1088/2058-9565/ac8ecd

work page doi:10.1088/2058-9565/ac8ecd 2023

[2] [2]

The laws of physics do not prohibit counterfactual communication

Hatim Salih, Will McCutcheon, Jonte R Hance, and John Rarity. The laws of physics do not prohibit counterfactual communication. npj Quantum Information , 8(1):60, 2022. doi: 10.1038/s41534-022-00564-w

work page doi:10.1038/s41534-022-00564-w 2022

[3] [1]

Counterfactual communication not achieved yet – A Comment on Salih et al

Sandu Popescu. Counterfactual communication not achieved yet – A Comment on Salih et al. (2022) npj Quantum Information , 11(1):122, 2025. doi:10.1038/s41534-025-00975-5

work page doi:10.1038/s41534-025-00975-5 2022

[4] [3]

Protocol for direct counterfactual quantum communication

Hatim Salih, Zheng-Hong Li, Mohammad Al-Amri, and Muhammad Suhail Zubairy. Protocol for direct counterfactual quantum communication. Phys. Rev. Lett. , 110:170502, Apr 2013. doi:10.1103/PhysRevLett.110.170502

work page doi:10.1103/physrevlett.110.170502 2013

[5] [4]

L. Vaidman. Past of a quantum particle. Phys. Rev. A , 87(5):052104, 2013. doi:10.1103/ PhysRevA.87.052104

work page 2013

[6] [5]

Consistent Quantum Theory

Robert Griffiths. Consistent Quantum Theory . Cambridge University Press, Cambridge, 2002 edition, 2002

work page 2002

[7] [6]

L. Vaidman. Counterfactuality of ’counterfactual’ communication. J. Phys. A , 48(46) (2015). doi:10.1088/1751-8113/48/46/465303

work page doi:10.1088/1751-8113/48/46/465303 2015

[8] [7]

Direct counterfactual transmission of a quantum state

L. Vaidman. Comment on “Direct counterfactual transmission of a quantum state”. Phys. Rev. A, 93(6):066301, 6 (2016). doi:10.1103/PhysRevA.93.066301

work page doi:10.1103/physreva.93.066301 2016