The Bubble Does Not Choose the Light

The flash

Hold a small gas bubble at the pressure node of a standing acoustic wave. During the low-pressure phase, the bubble expands; when the pressure reverses, the liquid rushes inward and the bubble collapses. Under stable single-bubble sonoluminescence, the sequence repeats tens of thousands of times a second: expansion, collapse, flash, rebound. In one representative apparatus, the acoustic drive operated at 26.5 kilohertz, so a full cycle lasted about thirty-eight microseconds. The light occupied only a sliver of it. Gompf and his colleagues measured pulses from roughly sixty to more than 250 picoseconds as gas concentration and acoustic drive changed. In another regime altogether—a bubble in concentrated sulfuric acid under stronger forcing—spectral measurements indicated plasma temperatures reaching about 15,200 kelvin. The numbers belong to their liquids, gases, pressures, temperatures, and instruments. They are not a universal biography of a bubble.^{1, 2, 3}

Nor is the flash made by pressure in the abstract. Stable emission depends on gas chemistry, dissolved-gas concentration, liquid temperature, acoustic amplitude, and a narrow region in which the bubble remains both shape-stable and diffusionally stable. Alter those conditions and the pulse may dim, the bubble may fragment, or the recurrent flash may disappear. Accounts of the emission mechanism have also varied across experimental regimes. The defensible scientific sentence is relational: a bubble, inside a liquid, inside a vessel, driven by a wave, can focus energy into light when a demanding set of boundary conditions holds.

Then darkness returns.

The darkness is not evidence of failure. The bubble rebounds; the liquid relaxes; expansion becomes possible again. What recurs is not collapse alone but a cycle whose recovery makes another event possible. If the scene can illuminate a human problem, it begins there—not with the glamour of the flash, but with the structure that permits recurrence. There is a bubble. There is also a liquid, a vessel, and a wave.

The metaphor indicts itself

“Pressure makes light” arrives almost automatically. It sounds like a sermon, a keynote, an epitaph for whatever someone has survived. It is also bad physics and worse ethics.

The bubble does not decide to shine. It has no history, no dignity, no dependents, no memory of earlier collapses. Its emission is not courage. Turn the event into a moral law and a category mistake becomes a demand: suffering should yield radiance, and the injured person should prove the value of the injury by becoming impressive afterward.

The metaphor becomes useful only when it is made to betray that demand.

Sonoluminescence cannot explain human conduct, much less legislate it. It exposes something narrower: the absurdity of explaining a relational event by praising the isolated object. No physicist would account for the flash by admiring the bubble’s character. Yet institutions routinely describe human outcomes in exactly that grammar. The worker burns out and is called insufficiently resilient. The patient cannot turn injury into insight and is called stuck. The survivor reports damage without gratitude and is treated as unfinished. A system disappears from its own story while the person is made to carry both the wound and its interpretation.

“Light” therefore needs restraint. It will not name every admirable human response. It names the metaphor’s temptation: visible transformation offered as proof that pressure worked. Survival is not light. Refusal need not glow. Damage may remain damage; grief may remain without revelation; an ordinary life resumed without a lesson requires no aesthetic defense. Once light becomes a synonym for everything we approve, the metaphor quietly restores the moral ranking it claimed to refuse.

The governing question is not whether pressure can precede transformation. It can. The question is who controls the pressure, who sets the duration, who builds the vessel, who may stop the wave, who can leave, and who is given time to expand again. Human transformation under pressure is neither adversity’s automatic product nor will’s sovereign achievement. It is an event within conditions. Responsibility is shared, but the power to shape those conditions is not.

Agency under conditions

The fantasy of sovereign will imagines a person outside the forces acting upon her, choosing from an intact menu of responses. Stress physiology describes a more entangled creature. Through allostasis, brain and body maintain stability by changing in response to demand. In the short term, that flexibility is adaptive. When demands persist or recovery repeatedly fails, the same mediating systems can accumulate allostatic load: wear produced not by one feeling but by sustained or dysregulated adaptation. Sleep, cardiovascular and immune activity, memory, appraisal, material conditions, and social experience enter the same loop.⁴

Defensive responses do not wait politely for reflective judgment. Clinical accounts of the defense cascade describe patterned shifts through arousal, flight, fight, freezing, tonic immobility, and collapse, with substantial variation across persons and situations. Popular nervous-system language often outruns this evidence, allowing an observer to diagnose a colleague after a meeting or reduce a life to a fashionable label. The research warrants less and matters more: under perceived threat, consequential responses can precede deliberation, and the range of action available in the moment can contract.⁵

Agency is therefore not an on-off possession. It is a capacity exercised within relations. Relational autonomy emerged in feminist philosophy partly to correct the picture of self-direction as independence from others. Socialization, recognition, material provision, and oppressive relations can support or impair the capacities through which a person reflects and acts. Dependence does not cancel authorship; it belongs to authorship’s conditions. A choice can be genuinely mine while the costs attached to each option have been arranged by others.⁶

Context does not dissolve responsibility. It distributes it more accurately. Iris Marion Young’s social connection model distinguishes backward-looking liability from a forward-looking responsibility to change structural processes in which many people participate. Responsibility is shared, but obligations differ with power, privilege, interest, and collective capacity. The employee who sends a cruel message under pressure remains answerable for the message. The executive who maintains an understaffed system of recurring emergency is answerable not only for a discrete act but for the field in which such acts become likelier and harder to contest.⁷ Explanation is not exoneration. It is a refusal to let the least powerful person’s final response stand in for the whole moral event.

This is the beginning of boundary-condition ethics. The framework asks how those who govern exposure justify its purpose, intensity, duration, distribution, and reversibility; what resources and authority accompany the demand; whether dissent can alter the decision; and who bears the work of recovery and repair. It differs from the loose claim that “context matters.” Context can always be invoked. Boundary-condition ethics assigns a burden: the greater an actor’s control over exposure and the less meaningful another person’s refusal, the stronger the actor’s prospective duty to prevent avoidable harm, make uncertainty legible, and revise the conditions when evidence changes.

That allocation turns on more than causal proximity. Role, knowledge, control, benefit, and capacity each matter. A supervisor may not have created an economic shock, yet still control whose workload absorbs it. A board may know less about a technical hazard than an engineer, yet possess the authority to require an independent review and the capacity to fund it. A worker may contribute causally to a harmful outcome while lacking the power to redesign the process that makes recurrence predictable. None of these facts erases individual answerability. Together they explain why institutional responsibility can be primary without becoming total: it attaches to the power to govern the future, not to a fantasy that one office caused everything.

That duty does not turn people into passive products of environment. Better conditions enlarge authorship. Agency may take the form of endurance, but it may also appear as naming the pressure, refusing its assigned meaning, joining with others, or leaving. Sometimes the most lucid transformation is not learning to shine under compression. It is recognizing that the compression has been organized.

Growth without debt

Any critique of compulsory transformation risks becoming unable to hear people who say that suffering changed them for the better. Posttraumatic growth is not a fiction invented by employers or preachers. Richard Tedeschi and Lawrence Calhoun developed the construct to describe positive psychological change arising through struggle with highly challenging circumstances. Their inventory asks about relations with others, new possibilities, personal strength, spiritual change, and appreciation of life. Many people recognize themselves in that account. A person may abandon a life built around approval, discover a friendship capable of candor, or acquire a seriousness she did not possess before. An external theory that dismisses her testimony would seize the authority it accuses redemptive narratives of seizing.⁸

Testimony and demonstrated change, however, are not the same object. A retrospective inventory can establish that someone reports growth. Without a pre-event measure, it cannot show how present capacity compares with baseline. In a 2009 prospective study, Patricia Frazier and colleagues examined students who experienced a traumatic event between assessments. Perceived growth was generally unrelated to measured pre-to-post change in corresponding domains; perceived growth was associated with greater distress, while measured growth was associated with less. The study followed a small, predominantly young, female, and White subsample for only eight weeks. It challenges a measurement inference, not the possibility of transformation.⁹

A later study followed a representative midlife sample across six months and gathered more than 32,000 experience-sampling assessments. Perceived growth after major negative events did not predict subsequent improvement in current standing or aggregated daily states. That design was stronger in ecological detail and still bounded: events varied in kind and severity, attrition mattered, the sample was predominantly female, and there was no clean pre-event baseline for every claim.¹⁰ The result again disciplines a causal conclusion. It does not invalidate a life story.

Narrative authorship is not a consolation prize awarded when measurement fails. To determine what an event will mean in one’s life is an exercise of agency. It can reorganize identity even when a trait score remains stable. The caution runs in both directions: a scale should not overrule a person’s account, and an account should not settle a causal question it was never designed to answer. Most important, authorship belongs to the person whose life is being narrated. An employer, family, or nation may admire what she became; none may use her account to acquit itself.

The grammar changes with the speaker. “What happened made me stronger” may be a survivor’s hard-won description of causation, identity, or both. “What we put you through made you stronger” is an institutional claim to the same life. The first can be respected without cross-examination because its immediate work is self-interpretation. The second requires evidence and standing: Was the exposure necessary? Who authorized it? What alternatives existed? Who bore the losses that the success story leaves outside the frame? Equal sentences can perform opposite acts. One restores authorship; the other appropriates it.

A 2026 multidisciplinary proposal sharpens the problem by requiring credible growth to be lasting, conceptually distinct, meaningfully positive, normatively significant, causally linked to adversity, and empirically verifiable. Its authors then define growth through the cultivation of culturally situated virtues, acknowledging that the content of “positive” character varies across contexts.¹¹ This is a real advance over treating self-reported benefit as a universal metric. It also makes the political question impossible to avoid. Once growth is defined as virtue, who is authorized to name the virtue? Endurance may express courage in one setting and trained submission in another. Harmony may preserve relationship or suppress justified anger. Cultural location cannot be removed from judgment, but neither can “culture” be allowed to conceal who has power to reward a trait and who benefits when it is called growth.

Social support belongs to the same caution. A meta-analysis of 217 samples found a substantial association between support and reported posttraumatic growth, but most included studies were cross-sectional. The finding cannot establish whether support causes growth, growth changes the perception or cultivation of support, or other factors shape both.¹² It supports no universal tale of rescue. It does undermine the romance of transformation as a sealed interior achievement.

Growth can be real without becoming a debt. A good that emerges after suffering does not prove that suffering was necessary, wisely designed, or owed. The achievement belongs to the person and to those who sustained her; it is not a dividend payable to the author of the wound. This is why “What did you learn?” can be intimate or violent. Asked by someone willing to hear “nothing,” “not yet,” or “I learned to leave,” it may open reflection. Asked by an institution seeking evidence that its damage was developmental, it becomes an audit.

After triumph

The pressure to redeem suffering did not begin with organizational psychology. Religious language can place cosmic authority behind it. Genesis gives the imagination one of its deepest images: darkness, divine speech, light. Read carelessly beside sonoluminescence, creation seems to authorize the desired allegory—pressure descends and illumination answers. Yet Genesis does not depict a wounded person earning radiance. Light comes through divine address and the ordering of a world, not through the moral improvement of matter by injury.¹³

Job is more hostile to explanatory appetite. His friends first sit with him in silence; when they begin to defend an intelligible moral order, suffering becomes evidence of fault, discipline, or instruction. At the end, God rebukes them for not speaking rightly, while Job’s accusation and protest remain inside the book’s authorized speech. Carol Newsom reads Job as a contest among moral imaginations rather than a puzzle resolving into one lesson. The friends’ failure is not that they seek meaning. It is that their meaning requires Job’s pain to confirm the order they are committed to preserving.^{14, 15}

Shelly Rambo finds a related danger in the Christian passage from death to new life. A triumphant rush from cross to resurrection can erase Holy Saturday—the interval in which death is not past, the body retains what happened, and life returns as fragment, witness, and remainder rather than clean reversal. Rambo does not discard resurrection. She refuses the timetable by which hope is used to hurry the wounded out of aftermath.¹⁶

Delores S. Williams makes the critique historically harder. In Sisters in the Wilderness, her reading of Hagar confronts the coerced surrogacy imposed on Black women: bodies and labor made to bear another’s need, then sanctified through accounts of sacrifice. That history cannot be converted into a portable metaphor for office strain. Its specificity is the source of its force. Williams shows how redemptive suffering can serve institutions already organized to take Black women’s endurance for granted, and she relocates salvation toward survival, defiance, relation, and quality of life rather than glorified victimization.¹⁷

Theology is not exhausted by these voices, and Christianity has no single account of suffering. A defender of resurrection hope can rightly insist that injury must not have the final word. The ethical limit lies elsewhere: resurrection does not authorize the crucifier, and hope does not transfer interpretive ownership from the wounded to the observer. Bruce Lincoln’s religious-studies method presses the worldly questions that reverent interpretation often resists: who speaks, to whom, through what mediation, with what interests, and with what consequences for winners and losers?¹⁸ A meaning made by the person who suffered may restore authorship. The same words imposed by a pastor, clinician, employer, or state may confiscate it.

Once redemptive language is understood as a distribution of authority, its secular institutional form becomes visible. “This will make you stronger” is not only a belief about the future. Spoken downward, it can be a claim to govern another person’s pain.

The burden of proof

On the evening of January 27, 1986, engineers and managers in Utah, Alabama, and Florida joined a teleconference about whether Challenger should launch the next morning. The forecast was unusually cold. The rubber O-rings sealing the field joints of the solid rocket boosters had shown erosion and blow-by on earlier flights, with the worst blow-by occurring on the coldest previous flight. Morton Thiokol’s engineers recommended against launching below an O-ring temperature of 53 degrees Fahrenheit, the lower boundary of their flight experience.

The recommendation did not arrive in a vacuum. NASA had a formal, multi-tiered Flight Readiness Review built to move problems upward and allow independent judgment. Managers were responsible for a public program of enormous technical difficulty and cost. The data were sparse and not perfectly monotonic; one warmer flight had also shown blow-by, while a static test at a lower temperature had not. NASA officials argued that the evidence for a new temperature limit was inconclusive and that imposing a launch criterion on the eve of flight required more than unease. These were not absurd questions. Institutions sometimes must decide before uncertainty disappears.¹⁹

What followed reveals the difference between questioning evidence and arranging who must bear uncertainty. Thiokol’s initial position was no launch. NASA managers challenged it. During a private caucus, a senior Thiokol executive asked the vice president of engineering to “take off his engineering hat and put on his management hat.” The recommendation reversed. In later testimony, that vice president described the thought process: the contractor found itself trying to prove the motor would not work and could not prove failure absolutely. The burden had changed direction. Those controlling exposure no longer had to establish that departure from the experience base was acceptably safe; engineers had to establish that catastrophe would occur.¹⁹

The Presidential Commission did not conclude that everyone on the call knew the launch would fail. It found a flawed decision process: incomplete and sometimes misleading information, conflict between engineering data and management judgment, and a structure that allowed flight-safety problems to bypass senior Shuttle officials. The officials with final launch authority did not know that the contractor had initially recommended against launch or that engineers continued to object. Formal review existed. Dissent existed. Neither was decision-effective.

The case should be judged from that evening, not only from the wreckage. A successful launch would not have made the process sound; catastrophe did not by itself prove that every advocate of launch had reasoned culpably. Outcomes are evidence, not verdicts on the quality of every prior choice. Had the launch been postponed, no one could have demonstrated which disaster the delay prevented. An ethics that recognizes only visible harm will therefore reward unsafe systems during every interval in which they are lucky. Ex ante judgment asks instead whether uncertainty was represented honestly, whether the people assuming risk had access to what was known, whether contrary evidence reached accountable authority, and whether the institution preserved a real capacity to stop.

The institutional archive also shows how warning can be grammatically domesticated before a crisis. After the coldest prior flight, an eighteen-page analysis connected low temperature with an increased probability of blow-by, then called the condition acceptable. By the next Level I review, the analysis had become a one-page chart resolving the matter as “acceptable risk because of limited exposure and redundancy.” Temperature disappeared from the summary. “Redundancy” remained even though NASA had already classified the seal as a nonredundant single point of failure. Each word sounded technical; together they converted an unresolved mechanism into administrative closure.¹⁹

This is not a parable in which managers represent evil and engineers innocence. Engineers also worked within the institution’s assumptions; evidence could be interpreted in more than one way; delay carried real program costs; and a recommendation at the edge of available data can itself be wrong. The case matters because it resists moral simplification. It shows what boundary-condition ethics adds under uncertainty: the institution imposing risk must identify whose safety is at stake, preserve the path by which adverse evidence reaches decision authority, and assign the burden of proof according to control and consequence. When harm would be grave and refusal belongs to others, uncertainty cannot function as a free asset for the party pressing forward.

An office deadline is not a shuttle launch. The comparison is mechanism, not magnitude. Ordinary organizations rarely govern risks of catastrophic physical death; their duties should scale to the stakes. Yet the same inversion appears whenever a subordinate must prove in advance that a schedule will cause injury, that an accommodation will be necessary, or that a workload cannot be sustained, while the authority imposing the condition treats the absence of certainty as permission. The lesson is not “never proceed.” It is that the party controlling exposure must carry the justificatory burden it has the power to answer.

The institution as boundary condition

Work can be difficult without being wrongful. High standards, urgent projects, and chosen periods of intense effort can generate mastery, solidarity, public value, and joy. The Job Demands–Resources model is useful precisely because it does not collapse demand into harm. Demands require sustained effort and are associated especially with exhaustion; resources such as autonomy, support, feedback, staffing, and adequate tools can support engagement and buffer costs. Consequences arise from a relation between demand and resource, not from intensity alone.²⁰

Voice is part of that relation. Psychological safety, in Amy Edmondson’s formulation, concerns the interpersonal risk of speaking, questioning, reporting error, and admitting uncertainty. It is not low standards or freedom from discomfort. A surgical team may be exacting and psychologically safe; a pleasant office may be unsafe if disagreement threatens status or employment. The relevant measure is not whether leaders invite candor but whether candor can change the information set and survive contact with authority.²¹ A listening session after the decision is not voice. It is ceremony.

Recovery is another boundary condition. Research organized by the stressor-detachment model draws on cross-sectional, longitudinal, and daily-diary studies to show that workload predicts difficulty detaching from work and that impaired detachment is associated with strain and poorer well-being. The evidence does not itself dictate a labor policy. It defeats the idea that recovery is only a private skill practiced after the organization has finished acting.²² A person cannot detach while unofficially on call. A meditation exercise may help her relax; it cannot provide coverage, return control over an evening, or remove the penalty attached to silence.

Boundary-condition ethics turns these findings into a public criterion. First comes necessity: what defensible good requires the pressure, and what less harmful alternatives were rejected? Then proportionality and distribution: are intensity and duration commensurate with the end, and who repeatedly absorbs volatility for whose benefit? Participation follows: can affected people shape the terms, refuse, appeal, or introduce contrary evidence before commitment becomes irreversible? Resources and recovery test whether time, staffing, authority, information, accommodation, and a protected interval match the demand. Finally, revisability and repair ask what evidence will stop or alter the exposure, when the exception expires, how injuries will be investigated, and what the institution will restore. These are normative requirements, not conclusions smuggled out of a burnout scale.

Consent matters within this criterion but cannot carry it alone. Employees may volunteer for punishing assignments because prestige, benefits, immigration status, promotion, or simple economic survival makes refusal costly. A caregiver may have no slack for a schedule others experience as exhilarating. A disabled worker may rely on the same manager for accommodation and advancement. Formal choice does not vanish, but its price must enter the moral accounting. The less credible the exit, the heavier the institution’s burden to govern the condition justly.

A legitimate surge should therefore have a form before it has a slogan. The institution names the condition that triggers exceptional effort, the authority that may declare it, the work that will stop when new work begins, and the date on which the exception expires or must be renewed under fresh evidence. Coverage and recovery are funded as part of the response. Dissent travels to someone empowered to change course. An after-action review examines not only whether the goal was met but who paid for it, which warnings were lost, and what must change before another surge. Without these limits, “emergency” becomes a managerial solvent: it dissolves prior commitments while leaving the power to declare necessity untouched.

Nor is a completed checklist enough. NASA had review architecture; the problem was that warnings could be compressed, reclassified, contained, and prevented from reaching those whose decision they should have changed. The managerial counterfeit of boundary-condition ethics is documentation without consequence: a risk register that cannot slow the schedule, an anonymous survey that cannot alter allocation, a wellness program offered beside unchanged demand. A boundary is real only if crossing it changes conduct.

Repair has three objects: the person, the record, and the rule. The person may need compensation, care, restored time, accommodation, reinstatement, or an exit that does not convert injury into financial punishment. The record may need correction so that refusal is not preserved as disloyalty, lowered output as incompetence, or retaliation as ordinary management. The rule must change if the same allocation, escalation path, or incentive would reproduce the harm for someone else. An apology without material restoration asks the injured person to supply closure. Compensation without institutional learning prices recurrence. Policy reform without attention to the person treats a life as evidence collected for future improvement. Repair is credible when each object constrains the others: testimony shapes redesign, redesign reduces the chance of repetition, and the institution bears costs it cannot quietly return to those already harmed.

No institution can promise a life without injury, and no serious ethics can demand omniscience. Institutions can be judged ex ante by whether they made uncertainty visible, matched authority to consequence, protected adverse information, bounded exceptions, and retained the capacity to reverse course. They can be judged afterward by whether they repaired harm and changed the system rather than recruiting survivors to certify it. No institution has the moral right to manufacture preventable injury as a method of formation and then call the injury development.

Agency after sovereignty

Structural responsibility can end in paralysis if everyone is implicated and no one is answerable. Boundary-condition ethics points in the opposite direction. The question is not whether one person caused the whole system. It is what each person can alter from the position actually occupied. A worker can document a pattern, protect a colleague’s dissent, organize, appeal, refuse, or leave. A manager can change allocation, make escalation consequential, defend an interval of recovery, and absorb the cost of telling the truth upward. A board can ask whether incentives reward manufactured urgency. A clinician or pastor can decline to force a redemptive account. Agency becomes political without ceasing to be personal.

For the person under pressure, authorship may include making meaning. It may also include withholding it. A life can continue without converting pain into instruction for an audience. The right to say “this harmed me” does not depend on a plan for transcendence; the right to say “I changed” does not absolve whoever caused the harm. Growth remains possible. Debt does not follow.

Sonoluminescence is beautiful because a disciplined physical system concentrates acoustic energy into an event that seems impossible at its scale. Its beauty carries no command. The bubble does not choose the light, and the flash cannot tell us what a person should become. It can direct attention away from the luminous instant and toward the liquid, the vessel, the wave, the boundary, and the return.

Sound disciplined into light is an image worth keeping only if discipline belongs first to the structure: pressure bounded, dissent effective, recovery protected, uncertainty borne by those who govern it. The moral achievement is not to make damaged people shine. It is to build conditions in which no one must be broken for illumination, growth can be received without being owed, and after every demand a life is allowed to expand again.

---

Endnotes

1. Michael P. Brenner, Sascha Hilgenfeldt, and Detlef Lohse, “Single-Bubble Sonoluminescence,” Reviews of Modern Physics 74, no. 2 (2002): 425–484, https://doi.org/10.1103/RevModPhys.74.425.

2. B. Gompf, R. Günther, G. Nick, R. Pecha, and W. Eisenmenger, “Resolving Sonoluminescence Pulse Width with Time-Correlated Single Photon Counting,” Physical Review Letters 79, no. 7 (1997): 1405–1408, https://doi.org/10.1103/PhysRevLett.79.1405.

3. David J. Flannigan and Kenneth S. Suslick, “Plasma Formation and Temperature Measurement during Single-Bubble Cavitation,” Nature 434 (2005): 52–55, https://doi.org/10.1038/nature03361.

4. Bruce S. McEwen and Peter J. Gianaros, “Stress- and Allostasis-Induced Brain Plasticity,” Annual Review of Medicine 62 (2011): 431–445, https://doi.org/10.1146/annurev-med-052209-100430.

5. Kasia Kozlowska et al., “Fear and the Defense Cascade: Clinical Implications and Management,” Harvard Review of Psychiatry 23, no. 4 (2015): 263–287, https://doi.org/10.1097/HRP.0000000000000065.

6. Catriona Mackenzie and Natalie Stoljar, “Introduction: Autonomy Refigured,” in Relational Autonomy: Feminist Perspectives on Autonomy, Agency, and the Social Self, ed. Catriona Mackenzie and Natalie Stoljar (New York: Oxford University Press, 2000), 3–31, https://academic.oup.com/book/49675/chapter/422021111.

7. Iris Marion Young, “A Social Connection Model,” in Responsibility for Justice (New York: Oxford University Press, 2011), 95–122, https://doi.org/10.1093/acprof:oso/9780195392388.003.0004.

8. Richard G. Tedeschi and Lawrence G. Calhoun, “Posttraumatic Growth: Conceptual Foundations and Empirical Evidence,” Psychological Inquiry 15, no. 1 (2004): 1–18, https://doi.org/10.1207/s15327965pli1501_01; Richard G. Tedeschi and Lawrence G. Calhoun, “The Posttraumatic Growth Inventory: Measuring the Positive Legacy of Trauma,” Journal of Traumatic Stress 9, no. 3 (1996): 455–471, https://doi.org/10.1007/BF02103658.

9. Patricia Frazier et al., “Does Self-Reported Posttraumatic Growth Reflect Genuine Positive Change?” Psychological Science 20, no. 7 (2009): 912–919, https://doi.org/10.1111/j.1467-9280.2009.02381.x.

10. Meghan J. Gangel et al., “Does Perceived Posttraumatic Growth Predict Observed Changes in Current-Standing and State Posttraumatic Growth?” Clinical Psychological Science 12, no. 3 (2024): 505–516, https://doi.org/10.1177/21677026231182329.

11. Eranda Jayawickreme, Michael Brady, and Nuwan Jayawickreme, “Multidisciplinary Recommendations for Improving the Credibility of Research on Posttraumatic Growth,” Personality Science 7 (2026), first published July 15, 2026, https://doi.org/10.1177/27000710261469023.

12. Jing Ning et al., “Social Support and Posttraumatic Growth: A Meta-Analysis,” Journal of Affective Disorders 320 (2023): 117–132, https://doi.org/10.1016/j.jad.2022.09.114.

13. Genesis 1:1–3, New Revised Standard Version Updated Edition.

14. Job 42:7–8, New Revised Standard Version Updated Edition.

15. Carol A. Newsom, The Book of Job: A Contest of Moral Imaginations (New York: Oxford University Press, 2003), 90–168, 259–264.

16. Shelly Rambo, Spirit and Trauma: A Theology of Remaining (Louisville, KY: Westminster John Knox Press, 2010), 7, 129, 159, 168.

17. Delores S. Williams, Sisters in the Wilderness: The Challenge of Womanist God-Talk (Maryknoll, NY: Orbis Books, 1993), 127, 161, 164–169.

18. Bruce Lincoln, “Theses on Method,” Method & Theory in the Study of Religion 8, no. 3 (1996): 225–227, https://doi.org/10.1163/157006896X00323.

19. Presidential Commission on the Space Shuttle Challenger Accident, Report to the President, vol. 1 (Washington, DC, 1986), chap. 5, “The Contributing Cause of the Accident,” 82–108, https://www.nasa.gov/history/rogersrep/v1ch5.htm; chap. 6, “An Accident Rooted in History,” 120–148, https://www.nasa.gov/history/rogersrep/v1ch6.htm.

20. Evangelia Demerouti et al., “The Job Demands–Resources Model of Burnout,” Journal of Applied Psychology 86, no. 3 (2001): 499–512, https://doi.org/10.1037/0021-9010.86.3.499; Arnold B. Bakker and Evangelia Demerouti, “Job Demands–Resources Theory: Taking Stock and Looking Forward,” Journal of Occupational Health Psychology 22, no. 3 (2017): 273–285, https://doi.org/10.1037/ocp0000056.

21. Amy C. Edmondson, “Psychological Safety and Learning Behavior in Work Teams,” Administrative Science Quarterly 44, no. 2 (1999): 350–383, https://doi.org/10.2307/2666999; Amy C. Edmondson and Derrick P. Bransby, “Psychological Safety Comes of Age: Observed Themes in an Established Literature,” Annual Review of Organizational Psychology and Organizational Behavior 10 (2023): 55–78, https://doi.org/10.1146/annurev-orgpsych-120920-055217.

22. Sabine Sonnentag and Charlotte Fritz, “Recovery from Job Stress: The Stressor-Detachment Model as an Integrative Framework,” Journal of Organizational Behavior 36, suppl. 1 (2015): S72–S103, https://doi.org/10.1002/job.1924.