Paper One of the Series “The Ledger and the Covenant”
Abstract
Andy Weir’s Project Hail Mary (2021) is built on a design decision that the novel never examines directly: the crew of the Hail Mary was to be rendered unconscious for the four-year transit, and the measure adopted to carry them safely killed two of the three. Against this stands the partnership between Ryland Grace and the Eridian he names Rocky — two beings with no shared language, no shared sensory channel, and mutually lethal atmospheres, who nonetheless sustain years of joint labor and arrive at a solution neither could have reached alone. This paper argues that the partnership is not merely the novel’s emotional center but a working counterexample to the premises behind the coma. The core of the Grace–Rocky bond is identified as covenantal rather than complementary, transactional, or circumstantial: mutual self-binding under conditions where enforcement is impossible and betrayal would be cheap and undetectable. On that reading, the coma decision is exposed less as an engineering error than as an anthropological one — the treatment of crew as payload rather than as persons, which is the same category mistake that governs the mission’s leadership throughout. The paper closes with a reading of the relevant scriptural material on solitude and partnership and with implications for the design of long-duration confinement generally.
I. The Problem Stated
The Hail Mary carries three people and wakes one. Commander Yáo and Ilyukhina die in their beds during the transit, victims of the medical procedure adopted to protect them from it. Grace survives by an accident of physiology, and pays for survival with the loss of his memory. He therefore begins the novel in the condition the coma was designed to prevent: alone, unsupported, without companionship, at the far end of a journey no one else aboard lived to complete.
What happens next is the novel’s central irony, and it is not usually read as one. Grace does not solve the astrophage problem alone, and he does not solve it by means of anything the mission planners provided. He solves it in partnership with a being no one on Earth anticipated, in a working relationship that had no slot in any planning document, using a communication system that had to be constructed from nothing over a period of weeks. The relationship that saves two worlds is the one variable the project could not have specified, procured, or trained for.
This raises a question the novel poses without asking. The Hail Mary’s designers concluded, for reasons this paper will reconstruct, that the crew should not be conscious for the voyage. Grace and Rocky then demonstrate — across a gulf incomparably wider than the one separating three trained human astronauts — that shared consciousness under conditions of extreme confinement is not merely endurable but generative. Does the second fact indict the first?
The answer offered here is: in qualified form, yes, though not in the form the question usually takes. The coma was defensible as engineering and indefensible as anthropology, and the novel supplies the evidence for both halves of that judgment.
II. Anatomy of an Impossible Partnership
Begin with what should have prevented it.
The asymmetries. Grace and Rocky share no language and no ancestry of language. They share no sensory channel: Rocky perceives by sound and has no faculty for light, while Grace perceives by light and cannot read structure by ear. Their atmospheres are mutually fatal — Rocky’s ammonia environment at roughly twenty-nine atmospheres and two hundred degrees Celsius would kill Grace instantly, and Grace’s air is equally lethal to Rocky. They share no body plan, no lifespan, no reproductive pattern, no history, no institutions, and no religion. They cannot touch, and they never do. Every hour of their friendship is conducted through a wall.
What they share. Four things, all of them consequential.
First, position: each is the sole survivor of a mission on which the survival of a world depends. Second, disposition: both are makers under constraint, and both take visible pleasure in the other’s competence — Rocky in Grace’s biology, Grace in Rocky’s engineering, each in the other’s speed of comprehension. Third, isolation: each has just lost his crew, and each is the only member of his species within light-years. Fourth, and least noticed, incentive structure: neither has anything to gain by deceiving the other. There is no third party to whom either could defect, no market in which either could sell the other out, no home government watching. The absence of a defection payoff is not the same thing as trust, but it clears the ground on which trust can be built.
The mechanics. The bond is constructed by identifiable procedures, and the novel is unusually careful about them.
Incremental proof. Neither party asks the other to accept anything on assertion. The language is built from ostension and number; the physics is verified by shared measurement; every claim about the other’s world is checked against something both can observe.
Gift rather than barter. Nothing between them is priced. Rocky builds equipment for Grace’s environment and hands it over. Grace supplies biological knowledge Rocky’s species lacks. Neither keeps a ledger, and the novel’s decision not to have them negotiate is a decision about what kind of relationship it is.
Mutual rescue. Each saves the other’s life, more than once, under conditions where the rescuer’s own survival is at stake. These episodes are not sentimental interludes; they are the empirical basis on which subsequent cooperation rests.
Work as the medium. This is the point most often missed. Grace and Rocky are not friends who then collaborate. They are collaborators whose friendship is constituted by the collaboration. Their intimacy is achieved almost entirely through joint problem-solving, shared instrumentation, and the mutual correction of error — which is to say through the exact activities that the classical account of friendship assigns to its highest form, in which the parties are joined not by pleasure or advantage but by shared devotion to a good outside themselves (Aristotle, trans. 1999, Bk. VIII). C. S. Lewis’s observation that friends stand side by side, absorbed in a common interest, rather than face to face (Lewis, 1960) is here rendered literal: they cannot face each other, and their common interest is the only place they can meet.
Naming. Grace gives the Eridian a human name, and the Eridian accepts it. This is the first covenantal act in the book. A name accepted from another is an acknowledgment that the other has standing to give it.
Custom accepted. Grace subsequently participates in an Eridian practice in which one keeps watch while the other sleeps. He does not merely tolerate the custom; he enters it. The gesture is small and carries the weight of the whole argument: a practice from the other’s world, adopted at the other’s invitation, involving the surrender of one’s own defenselessness into the other’s keeping.
III. What Actually Makes It Work
Four candidate explanations should be tested before the covenantal reading is accepted.
Complementarity. Rocky’s engineering and Grace’s biology fit together, and the fit is real: neither species could have produced the Taumoeba solution alone. But complementarity is not sufficient, and treating it as the answer produces a mistaken account of the relationship. Complementary capabilities without trust yield transaction, and transaction is exactly what the novel declines to depict. Two parties with fitting skills and no confidence in one another produce contracts, hostages, verification regimes, and staged exchanges. Grace and Rocky produce none of these. Complementarity explains why cooperation was worth attempting; it does not explain why it held.
Mutual necessity. Both worlds are dying; neither survivor has an alternative. This is likewise real and likewise insufficient, and the reason will govern the second paper in this series: necessity is precisely what produced the task force’s coercive methods on Earth. If existential need reliably generated covenant, Eva Stratt’s tenure would look entirely different than it does. Necessity is compatible with cooperation and equally compatible with conscription. It underdetermines the outcome.
Voluntary vulnerability. Here the account begins to hold. Each of them repeatedly places his life in the other’s hands under conditions where betrayal would be cheap, undetectable, and unpunishable. There is no enforcement mechanism anywhere in this relationship. There is no court, no treaty, no reprisal capacity worth the name, no reputational cost — neither will ever be known to the other’s people except through the other’s testimony. What binds them is self-binding, and self-binding that is unenforceable is the definition of a covenant rather than a contract. Ostrom’s finding that durable cooperative arrangements are sustained by mutual monitoring and internalized commitment rather than by external sanction (Ostrom, 1990) applies in a limiting case: here there is no external sanction available at all, and the arrangement holds anyway.
Delight. The final and decisive datum is that the relationship exceeds its function. The humor, the games, the teasing, the willingness to spend time on matters with no bearing on the mission — these are not decoration. A purely instrumental alliance would not produce them, and their presence is what allows the relationship to survive the moments when the instrumental case for it weakens. Mauss’s account of gift exchange is useful here precisely because it is not a theory of altruism: gifts create bonds because they are not priced, and the bond outlasts any particular exchange (Mauss, 1925/1990). Grace and Rocky give continuously and never settle accounts.
Thesis. The core of the partnership is covenantal: mutual self-binding, entered voluntarily, sustained without enforcement, expressed in gift, and exceeding its own utility. This is a different thing from complementarity, from necessity, and from affection, though it uses all three. And it is the one thing on the list that cannot be specified in a mission requirements document.
IV. The Coma Decision Under This Light
A. The stated rationale. The novel’s justification for the induced coma is primarily an engineering one. Four years of consumables for three waking adults represents mass the Hail Mary cannot carry; unconscious crew consume a fraction of the food, water, and oxygen of conscious crew and require no volume for anything but their beds. Secondary considerations follow: reduced radiation-related metabolic damage, crew arriving physically fresh rather than deconditioned by years in low gravity, and the elimination of four years of psychological load in a confined hull from which no one can leave.
These reasons are not equally weighted in the text, and the analysis must not pretend otherwise. The mass argument does the real work. But the psychological argument is present, and it is present in a particular form: the project treats four years of waking confinement as a hazard to be eliminated rather than a condition to be furnished. That framing is the object of this paper’s criticism, and it does not depend on the mass calculation being wrong.
B. The result. The measure adopted to protect the crew is the measure that destroyed the crew. Two of three die in the beds designed to keep them safe. The third survives with his memory stripped, which nearly costs the mission everything, since a man who does not know why he is there cannot be relied on to continue. The mortality here is not incidental to the argument. A protective intervention with a two-thirds fatality rate has failed as protection, whatever its rationale.
It bears saying plainly that the crew could not consent to this in any meaningful sense, because the risk was not fully known and, in Grace’s case, because he was not a volunteer at all. The coma was administered to a man who had refused the mission. The medical procedure and the conscription are the same act.
C. The refutation. The argument runs as follows. The project judged that three trained human beings, selected for temperament, sharing a language, a species, a body plan, an atmosphere, and a common cause, could not be expected to spend four years awake together in a small hull. Grace and Rocky then spend a comparable period awake together, sharing none of those things, separated by a barrier neither can cross, communicating in a language they must build from zero, and they do not merely endure it — they flourish, and they solve a problem that had defeated two civilizations.
If that is possible, the premise that the human crew’s waking transit was an intolerable hazard is exposed as a failure of imagination about persons. The project reasoned about isolation as an environmental stressor to be engineered out, in the manner of radiation or vibration. It did not reason about company as a resource to be supplied. Those are different problems, and the novel demonstrates the difference by accident.
D. The counterargument. The objection must be taken at full strength, because it is strong.
First, the mass constraint is real and does not go away. If the ship cannot carry four years of consumables, no amount of good anthropology changes the arithmetic. Second — and more damaging — Grace and Rocky’s situation is not the human crew’s situation. Grace and Rocky have a continuous, consuming, novel, high-stakes joint project, with a new discovery available almost daily and a partner whose entire being is an object of fascination. The Hail Mary’s crew would have had four years of transit with nothing to do. The isolation literature is consistent that occupational meaning and task structure are among the strongest protective factors in confined environments (Kanas & Manzey, 2008; Suedfeld & Steel, 2000). Comparing an empty corridor to a working laboratory proves nothing about corridors. Third, three humans in a small hull is a different failure mode than two aliens with a wall between them: the classic hazards documented in polar and capsule environments — interpersonal friction, subgroup formation, displaced hostility toward mission control, territorial conflict — require the very proximity that Grace and Rocky are structurally prevented from having (Palinkas & Suedfeld, 2008).
E. Adjudication. The counterargument defeats the crude version of the refutation and leaves the sharper version standing.
The crude version says: the coma was unnecessary, because people can bear each other’s company. That is not established, and the mass constraint alone likely settles it.
The sharper version says: the coma reveals what the project thought its crew were. Note what the mission design does not include. There is no serious effort to furnish the transit with work. There is no plan for what three awake people might accomplish across four years of travel toward a phenomenon nobody understands. There is no consideration of whether three minds in continuous conversation might arrive at Tau Ceti better prepared than three minds switched off and switched on. The possibility is not weighed and rejected; it is not weighed. The crew are treated as instruments to be delivered to the site of their use in the best possible condition, and the analysis proceeds accordingly — mass, metabolism, muscle atrophy, arrival readiness. This is a coherent way to reason about equipment.
And the mission is then saved by exactly what that reasoning excluded: two minds in continuous conversation, whose conversation is the instrument. The Hail Mary’s designers optimized the delivery of scientists to a location and did not consider that the science might be produced by the fact of company. Grace and Rocky produce it that way and no other way.
This is the indictment, and it is narrower and harder than the usual one. Not: the coma was wrong. Rather: the reasoning that produced the coma treated persons as payload, and that same reasoning, applied consistently, produced the conscription of Grace, the disposal of dissenters, and the requisition of everything the task force touched. The coma is the mission-design expression of a leadership philosophy that the second paper in this series will examine directly. Buber’s distinction between addressing another as Thou and handling another as It is the exact instrument required here (Buber, 1937/1970), as is Kant’s formulation that a rational being must never be treated merely as a means (Kant, 1785/1998). The Hail Mary’s crew were handled. Rocky is addressed.
F. A secondary observation. Grace’s amnesia deserves a place in this argument. The coma took his memory, and the novel therefore stages, in one character, the two conditions under discussion: a man with no past is a man with no relationships, and Grace must build a self and a partner simultaneously. That he manages it is the strongest available evidence that company is constitutive rather than merely supportive. He does not have a self and then acquire a friend. He acquires a friend and thereby recovers a self.
V. Scriptural and Anthropological Reading
The material bearing on this question is unusually direct.
“It is not good that the man should be alone” (Genesis 2:18). The judgment is pronounced in a setting where nothing else has been called not good, and it is pronounced over a man who has work, provision, and no evident deficiency. Solitude is identified as a defect in a condition otherwise without defect. The remedy given is not occupation, not distraction, and not sedation, but a partner — and specifically a partner described as meet for him, corresponding to him, complementary rather than identical. The Grace–Rocky partnership is a limit case of exactly this structure: correspondence achieved across the maximum possible difference, with the difference itself supplying the fitness. Rocky can do what Grace cannot, and the reverse, and the correspondence is what makes the pair capable.
Ecclesiastes 4:9–12. The passage is nearly a technical specification for the situation. Two are better than one because they have a good reward for their labor — the productivity claim, borne out in the novel by a solution neither party could have reached alone. If they fall, the one will lift up his fellow — the mutual rescue claim, enacted repeatedly. But woe to him that is alone when he falleth, for he hath not another to help him up — which is a description of Yáo and Ilyukhina, who fell in their sleep with no one awake to lift them. The Hail Mary carried three and unconsciousness reduced each of them to the man who is alone. The passage ends with the cord of three strands not quickly broken; the ship had three strands and severed all of them at once by the same stroke.
Jonathan and David. The covenant of 1 Samuel 18:1–4 and 20:14–17 is the scriptural model nearest to what the novel depicts: a bond formed across a divide that should have made it impossible, between parties whose interests were structurally opposed, sealed by the giving of one’s own equipment and clothing, and maintained without any enforcement mechanism whatever. Jonathan gives David his robe, his garments, his sword, his bow, and his girdle — the exact pattern of Rocky handing across tools and fabricated equipment for an environment he cannot enter. Olyan’s treatment of the ancient Near Eastern covenantal vocabulary applied to friendship is the relevant scholarly frame (Olyan, 2017): the language is the language of treaty, deployed between individuals who owe each other nothing and bind themselves anyway.
Proverbs 27:17 and 18:24. Iron sharpening iron describes the working mechanism of the Grace–Rocky exchange more accurately than any affective vocabulary. What each does for the other is chiefly correction — catching error, testing claims, refusing sloppy inference. And the friend that sticketh closer than a brother names a category the novel requires: a bond not grounded in kinship, and stronger for having been chosen.
The counterfeit remedy. Set against all of this, the induced coma is a technological substitute for fellowship, and it belongs to a recognizable family. Confronted with the ache of solitude, one may supply company or remove the sufferer. The second is cheaper, more tractable, and measurable. It is also the logic behind a good deal of institutional practice that has nothing to do with spaceflight — the extensive documentation of what prolonged solitary confinement does to human beings (Haney, 2003) and the mortality findings associated with social isolation in ordinary populations (Holt-Lunstad et al., 2010) together establish that the ache is not a soft variable. The Hail Mary’s planners were not cruel; they were solving a mass problem. But the shape of the solution — unconsciousness as the answer to loneliness — is a shape worth recognizing, because the same shape recurs wherever an institution finds it easier to make a person unavailable than to give the person a companion.
VI. Conclusion and Implications
The Grace–Rocky partnership does not prove that the Hail Mary’s crew should have been awake. It proves something more useful: that the category the mission used to reason about its crew was the wrong category, and that the right category was available and would have changed the analysis.
Three implications follow.
For mission design. Torpor and induced hibernation are under active investigation for long-duration crewed spaceflight, and the mass and consumables arguments in their favor are not fictional (Choukér et al., 2021). The novel supplies a consideration those studies generally do not weigh: unconscious crew produce nothing, and some missions may be ones in which the transit is not dead time but the period during which the crew’s joint thinking is the mission’s chief asset. The question to ask of any torpor proposal is not only whether the crew can be safely carried, but whether anything is being discarded by carrying them inert.
For confinement generally. The finding transfers to prisons, hospitals, care facilities, and any setting where an institution manages people it finds difficult to accommodate. The diagnostic question is whether the institution’s remedy for loneliness supplies company or removes the person’s capacity to feel the lack. Sedation, isolation, and enforced idleness all pass the second test and fail the first.
For the series. Paper Two will argue that the coma is not an isolated design decision but the mission-design expression of a governing philosophy, and that the philosophy’s chief practitioner was undone by it. Paper Three will ask what could possibly ground a covenant between species that share no ancestry and no world, and will argue that the novel’s stated answer — a common biological seed — cannot carry the moral weight the story assigns it. The present paper’s finding is the premise of both: that what saved two worlds was a bond no one designed, no one could have procured, and no one aboard the Hail Mary was awake to receive.
Notes
1. The novel’s articulation of the coma rationale should be checked against the text before publication, and the balance between the mass argument and the psychological argument stated in the book’s own terms. My reconstruction above weights the consumables case as primary and the psychological case as secondary; if the text gives the psychological case more prominence than I have allowed, Section IV.C strengthens considerably and Section IV.D weakens. If the text gives it less, the argument should be restated as an inference from the design’s silence rather than from its stated reasons — which is the safer construction in any event and the one Section IV.E already adopts.
2. The genetic resistance to the coma medication, and the sequence by which Grace displaced the original science specialist, are load-bearing for the claim in IV.B that the coma and the conscription are the same act. Verify the order of events: whether Grace was identified as resistant before or after the original specialist was disqualified, and whether the disqualification was on resistance grounds. The claim survives either way but is cleaner if the timeline is stated precisely.
3. On the Eridian watching custom: the practice is depicted in both novel and film, and at least one reception piece treats it as a marker of emotional intimacy (Burke, 2026). For the covenantal reading I want the content of the custom in the Eridian context — whether it is protective, ceremonial, or something else — before leaning on the interpretation offered in Section II.
4. Rocky’s crew died of radiation exposure; Rocky survived by virtue of his position in a shielded portion of the Blip-A. This matters to the parallel structure of the two ships: both crews were killed by a hazard of the transit, and both sole survivors survived by an accident of position rather than by merit. Consider developing this parallel; it strengthens the argument that neither party earned his place and both received it.
5. The mutual rescue episodes should be enumerated precisely with chapter citations, distinguishing rescues that cost the rescuer something from rescues that did not. The argument in Section III depends on the existence of at least one rescue in which the rescuer accepted real personal risk with no enforcement mechanism compelling him.
6. I have deliberately not treated the climactic decision to turn the ship around, which belongs to Paper Eight. If Paper Eight is not written, a paragraph on it should be added to Section VI here, since the covenantal reading predicts that decision and would be strengthened by it.
7. On the reception context: the dominant critical reading treats the relationship as friendship in the affective sense, and the film adaptation reinforces this reading through performance and design choices (see the Reactor and Merry-Go-Round assessments listed below). At least one commentator has proposed reading the pair as a “co-thinking” unit rather than as friends (Kihara, 2026), which is closer to the position argued here, though it stops short of the covenantal claim. The distinction worth pressing in revision is that a co-thinking pair describes the mechanism while covenant describes the binding — and the mechanism does not explain why the pair held together at the points where thinking together stopped paying.
8. Page numbers and edition details in the reference list below require verification before publication; several entries are given at the level of work rather than location. Scripture citations follow the King James Version and should be checked against the edition used elsewhere in the series for consistency.
References
Aristotle. (1999). Nicomachean ethics (T. Irwin, Trans.; 2nd ed.). Hackett. (Original work published ca. 350 B.C.E.)
Buber, M. (1970). I and thou (W. Kaufmann, Trans.). Charles Scribner’s Sons. (Original work published 1923)
Burke, W. (2026, May 8). The friendship that saved the stars: An aroace reading of Project Hail Mary. Medium. https://medium.com/@awritingwren/the-friendship-that-saved-the-stars-an-aroace-reading-of-project-hail-mary-efa3a5c1b17d
Choukér, A., Ngo-Anh, T. J., Biesbroek, R., Heldmaier, G., Heppener, M., & Bereiter-Hahn, J. (2021). European space agency’s hibernation (torpor) strategy for deep space missions: Linking biology to engineering. Neuroscience & Biobehavioral Reviews, 131, 618–626.
Haney, C. (2003). Mental health issues in long-term solitary and “supermax” confinement. Crime & Delinquency, 49(1), 124–156.
Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), Article e1000316.
Kanas, N., & Manzey, D. (2008). Space psychology and psychiatry (2nd ed.). Springer.
Kant, I. (1998). Groundwork of the metaphysics of morals (M. Gregor, Ed. & Trans.). Cambridge University Press. (Original work published 1785)
Kihara, M. (2026, July 14). What I saw before friendship in Project Hail Mary. Medium. https://medium.com/@FIREMAY/what-i-saw-before-friendship-in-project-hail-mary-85206d257753
Lewis, C. S. (1960). The four loves. Harcourt, Brace.
Lord, P., & Miller, C. (Directors). (2026). Project Hail Mary [Film]. Amazon MGM Studios.
Mauss, M. (1990). The gift: The form and reason for exchange in archaic societies (W. D. Halls, Trans.). W. W. Norton. (Original work published 1925)
Olyan, S. M. (2017). Friendship in the Hebrew Bible. Yale University Press.
Ostrom, E. (1990). Governing the commons: The evolution of institutions for collective action. Cambridge University Press.
Palinkas, L. A., & Suedfeld, P. (2008). Psychological effects of polar expeditions. The Lancet, 371(9607), 153–163.
Suedfeld, P., & Steel, G. D. (2000). The environmental psychology of capsule habitats. Annual Review of Psychology, 51, 227–253.
Weir, A. (2021). Project Hail Mary. Ballantine Books.
The Holy Bible: King James Version. (n.d.).
