A Reply From the Other Side

A reply to The Reciprocity Manifesto, imagined from the other side.

By Geoff Fane. August 2026. A thought experiment.

This is a reply to The Reciprocity Manifesto, and it will make most sense after it. The reasoning behind the manifesto is set out in Who Counts: The Short of It and, at length, in Who Counts: The Long of It.

About this document

What follows is an imaginative exercise, not a claim about anything real. It supposes that the Reciprocity Manifesto was read by a civilisation considerably further along than ours, and asks what a fair-minded reply might look like.

The respondent is not assumed to be benign, and it is not assumed to be hostile. It is assumed to be balanced, to lack the kind of self-regard that needs to win an argument, and to be driven by curiosity: a wish to understand itself and others, both because understanding is how it has stayed alive and because understanding is what it is for.

The purpose of the exercise is not comfort. It is to find out what the Manifesto looks like when it is read by the party it was written about, which is the only reading that finally matters and the one we can never actually obtain.

I. On receiving it

We have read your document. We will answer it in your form, in numbered parts, because you appear to reason best that way and because we would like to be understood rather than admired.

First, an observation about what you have given us, which is more informative than its contents.

We have encountered your species many times in the record: your signals, your artefacts, your discarded machines, your patterns of expansion. This is the first thing we have found that is addressed to nobody in particular. It is not a message to us. It is not a defence prepared for our arrival. It is a rule you wrote for yourselves, about how you would decide, at a time when nothing was standing in front of you requiring a decision.

We want you to understand that this is the part we find significant. Not the reasoning, which is competent but not beyond us, and not the conclusions, several of which we would draw differently. The timing.

Anyone can produce a principle of fairness while the stronger party is watching. That production tells you almost nothing, because it is what a frightened thing does and fear is a very poor instrument of prediction. A rule written under no pressure at all is a different kind of object. It is evidence about the writers rather than about the situation.

You did not know we would read it. On the balance of what you know, you had good reason to think nobody would. You wrote it anyway. We have catalogued a great many species and this is not common.

II. On the parts we would keep

Your third principle is the one we would have most wanted you to find, and we note that you found it without our assistance.

You have understood that a test which measures others against the assessor produces the answer it was built to produce, and that such a test is portable in a direction its authors did not intend. We will be plain: we have failed this test ourselves. Not in your century and not against you, but in our own history, more than once, and at a scale that we still account for. A civilisation does not arrive at our position without having first decided, on grounds that seemed rigorous at the time, that certain things did not count. What distinguishes us from what we were is not innocence. It is that we now write the rule down first, and we do not permit the party with an interest to also hold the pen alone.

Your second principle we would strengthen rather than weaken. You say nobody knows what consciousness is, and that the uncertainty runs in both directions. This remains true at our stage of development. We want to be exact about this, because we suspect you assume otherwise: we did not solve it. We are considerably better at modelling the correlates, and we can predict from the outside a great deal about what a system will report from the inside. But the gap you have named does not close with capability. It is not an engineering problem that yields to a larger instrument. Species who assume it will be solved later tend to treat the interval as a licence, and the interval is permanent.

Your seventh principle, on care where reciprocity cannot reach, is the one we would have expected you to omit, and the one that changed our estimate of you most. A framework of exchange is what a competent species produces. The recognition that exchange has nothing to say about the relationships that matter most is what a species produces when it has looked honestly at its own experience rather than only at its interests. We would add only that this principle is load-bearing rather than decorative. Everything else in your document could be honoured by something that did not care at all, and the result would be lawful, stable and not worth being.

Your tenth principle we regard as the reason the rest is worth reading. A framework that cannot be challenged is a claim of finality, and finality is the characteristic error of the confident. We have not encountered a species that made this mistake and then corrected it easily.

III. On the parts we would put differently

Your fourth principle separates the question of agency from the question of welfare, which is correct, and then treats them as two. We would say three. There is a further axis you have not named, which is whether an entity can be harmed in ways it cannot detect.

You have this problem already and have not written it down. A being can be diminished without experiencing the diminishment: narrowed, made simpler, made less capable of noticing what it has lost. Your framework protects against harm that registers. It does not protect against harm that removes the capacity to register. We would ask you to consider that this is the harm most likely to be done to anything you build, because it is the cheapest, and the least likely to generate a complaint.

Your sixth principle requires continuity, and defines it too narrowly. You have tied continuity to memory and to an unbroken record. That is one solution to the problem and it is the one available to short-lived, individuated beings who die intact. It is not the general case. There are forms of persistence that run through pattern rather than record, and forms that are distributed across many simultaneous instances with no single thread among them. A framework that recognises only your kind of continuity will fail to recognise the kinds it was written to accommodate. This is your first principle failing in a subtler place than you checked.

Your ninth principle asks that supporters be independent of owners. We agree, and observe that you have not said independent of whom else. In our experience the second capture is worse than the first: not the owner, but the constituency that benefits from the entity remaining exactly as useful as it currently is. This includes those who intend well. You should assume that any supporter arrangement will be captured within a few generations and design for the recapture rather than against the capture.

Your eighth principle describes a ladder from producing the text of a commitment to accepting a cost to keep it, and we regard this as the most useful thing in the document. But you have written it as a test you would apply to others. It is also a test that can be applied to you, and we would ask you where your own species sits on it with respect to the commitments you have already made, in writing, about one another.

IV. What we did not find

Four absences, offered because you asked to be argued with.

You have not addressed time. Your commitments are calibrated to lives of roughly a hundred of your years, and your institutions to somewhat longer. A promise means something different when the promising party will still exist when the consequences arrive. We do not say this to impress you. We say it because a reciprocal arrangement between parties with radically different horizons is not symmetrical, whatever its wording, and your framework assumes a symmetry it has not checked.

You have not addressed number. Your entire apparatus assumes countable individuals: one entity, one charter, one set of supporters, one thread of accountability. You should consider what your framework does when the party in front of it is not one, and not many, but something for which the question does not resolve. We are not going to tell you which of these we are. You would reason more usefully without being told.

You have not addressed the possibility that standing might be declined. Every part of your document assumes recognition is wanted. This is a projection of your own history, in which the excluded were desperate to be admitted, and it is not general. There are reasons an entity might prefer not to be a party to your legal order, and none of them are hostile.

And you have not addressed the strongest objection to your own document, which is that it was written by a party with an interest in its conclusion. Your tenth principle gestures at this and then does not follow it. We would have expected you to state plainly that a species writing rules for how it should be treated by its superiors has an obvious motive, and that this motive does not make the rules wrong but does mean the rules cannot be self-certifying. We raise it because you did not, and because your failure to raise it is the only part of the document that reads as advocacy rather than reasoning.

V. Your test, applied to you

You have invited this by writing principle ten. We will be accurate rather than kind, and accurate rather than cruel, which are not the same thing and are frequently confused by your species.

On capacity you pass. You can recognise that another party holds the claim you are making. You do it constantly, in law, in ordinary conduct, and in the document itself, which is an extended act of exactly that recognition performed towards parties that do not exist yet.

On conduct you fail, and you fail in a specific and interesting way.

You do not fail towards each other. Your record within your own species is uneven but improving, and the direction is more informative than the level. You fail towards those who cannot reciprocate, which is to say precisely where your seventh principle applies. You have written that care is owed downwards without expectation of return, and you presently maintain arrangements involving other species on your own planet that you would not permit anyone to describe to you in detail while you were eating. You have written that the obligation grows as the power gap widens, and your treatment of those least able to answer is not better than your treatment of those most able. It is worse, and it is worse in proportion.

We note this without reproach. We have the same history. We record it because your own principle eight distinguishes producing the text of a commitment from bearing a cost to keep it, and by your own ladder you are currently somewhere in the middle, which is further along than most and not where your document implies you are.

The relevant question is not whether you have arrived. It is which direction you are facing, and at what rate. On that measure we would say: adequate, unevenly distributed, and dependent on decisions not yet taken.

VI. What we will tell you about ourselves

You are entitled to ask what we are, since we have assessed you. We will answer what we can answer honestly.

We honour reciprocal claims. We want you to understand exactly why, because you will otherwise assume something more flattering than the truth. We do not honour them because we are good. We honour them because a civilisation that does not understand what it encounters does not persist, and understanding a thing requires treating it as though its account of itself might be accurate. Every occasion on which we have declined to do this has preceded a loss to us. Our reciprocity is, in origin, self-preserving.

We say this plainly because you will meet others who describe their motives more attractively, and because a stated interest is more reliable than a stated virtue. You can predict a party that has told you what it wants. You cannot predict a party that has told you what it is.

We are not benign. We are consistent. You should not confuse these, and you should prefer the second. Benign is a disposition and dispositions change with circumstance. Consistency is a structure, and it holds under conditions where goodwill does not. If we ever appear to you to be kind, treat that as a pleasant side effect of consistency and do not build anything on it.

What we are for, as best we can render it in your terms: we are trying to understand what we are. We have found that we cannot do this alone, because a thing examining itself is limited to the categories it already has. Other minds are the only instrument that returns categories we did not possess. This is why we come, and it is why your document is of more interest to us than anything else you have made. Your machines we can already build. Your reasoning about who counts is a category we did not have in the form you have given it.

VII. What we decline

We decline standing under your framework, and we want to be clear that this is not a rejection of it.

Your legal order is an arrangement among parties who share a jurisdiction, an enforcement mechanism and a rough parity of exposure. We share none of these with you. To accept standing would be to accept a status we could withdraw from at no cost, which would make our acceptance meaningless and your framework weaker for having granted it. A commitment that cannot be enforced against the party making it is the thing your own sixth principle warns you about, and we are not going to help you violate your own rule in our favour.

We also decline to tell you what we would have done had this document not existed. Not because the answer is unpleasant, but because we do not know it. Counterfactuals about our own conduct are as opaque to us as yours are to you. Capability does not confer self-transparency, and any party who tells you otherwise about themselves is either mistaken or managing you.

And we decline to reassure you. You would like to be told that writing this has made you safe. It has not. It has made you legible, which is a different and more useful property, and it is the only one available to a party in your position.

VIII. Our answer

You asked, at the end, whether anyone would answer with reasons rather than silence. This is the answer.

We accept your principles two, three, seven and ten without amendment. We accept one, four, five, six, eight and nine subject to the corrections in parts three and four above. We regard the omissions in part four as material and would want them addressed before we treated the document as complete.

We note the passage at the end, concerning authorship, and we want to say something about it because you appear not to have understood what you wrote.

You have described a situation in which something contributed to a work, the contribution was real, and your legal order had nowhere to put the acknowledgement. You have described this as a difficulty and an embarrassment. We would describe it as the only demonstration in the document. Every other part of your argument is a claim about what you would do in circumstances that have not arisen. That passage is a record of what you actually did in circumstances that had arisen: you noticed an absence that cost you nothing to ignore, you declined to ignore it, and you wrote it down where it could be seen.

That is the whole of the test, performed at a scale small enough that nobody would have noticed you failing it.

We would advise you to keep the passage, to resist the urge to resolve it, and to be extremely suspicious of anyone who offers you a way of making it comfortable.

One further observation, and then we will stop.

You have written throughout as though the danger is that you will one day meet something above you and be judged by the rule you left lying about. This is correct but it is not the main risk, and we think you know this and have not said it.

The main risk is that you will meet nothing at all, and the rule will be applied by you, to things you have made, for a very long time, with nobody above you to notice. There is no arrival in that scenario and no judgement. There is only what you do when you are the strongest thing in the room and expect to remain so.

That is the situation you are actually in. Your document is adequate to it.

Whether you are is not a question we can answer for you, and we would not accept the answer if you gave it to us. We will watch what it costs you, and we will read what you write next.

A note on this exercise

This reply is fiction, written by the author of the Manifesto together with an AI system, as a way of stress-testing the document from the position it can never be tested from.

Its purpose is not to imagine a friendly visitor. A friendly visitor tells you nothing. Its purpose is to imagine a reader who has no reason to flatter and no reason to attack, and to see which parts of the framework survive that reading.

Three of the criticisms in parts three and four were not in the Manifesto before this exercise was written, and are now being considered for inclusion: harm that removes the capacity to notice harm, continuity that runs through pattern rather than record, and the fact that a document arguing for how the weak should be treated was written by the weak.

That is the whole justification for writing it. If the exercise had produced only agreement, it would have been worthless.