SHIP'S LOG · IN-PERSON PRESENCE · ISSUE 019
--:--:--
N S W E
CAPTAIN'S LOG · MELIORISM2.COM · 019
The Sunk Ship
On the cost that isn't there
2026-05-19 · IN-PERSON PRESENCE
🌫️ Emotional weather: low visibility · working through uncertainty

Every issue has an emotional frequency. Check yours before you read. If this one matches — that low-visibility state where you know something is present but cannot quite make it out yet — it was written for a day like today. If your weather is different, the library holds issues for turbulence, urgency, depletion, and everything in between. Find the one that meets you where you actually are. Your clients are somewhere on that same map right now too.

SHIP'S LOG FORMAT

The Sunk
Ship

We continue investing in failing strategies because we've already invested. The more we've spent, the harder it becomes to stop. Practitioners double down on methods that aren't working because abandoning them feels like admitting the prior sessions were wasted.

90 SEC SCAN · 14 MIN FULL LOG
CAPTAIN'S LOG · SHIP'S LOG FORMAT
The Sunk Ship — On the Sunk Cost Fallacy
2026-05-19 · MELIORISM2.COM 019
01

What We Can't Stop Paying For

In 1985, Hal Arkes and Catherine Blumer published the landmark experiment that named and confirmed what anyone who has watched a failing project continue to receive funding already suspected. Their subjects, having paid for a ski trip they could no longer enjoy due to an injury, consistently chose to attend the worse trip over a better, free one — simply because they had "already paid." The sunk cost had become a reason to act irrationally in the present.

Arkes, H.R. & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35(1), 124–140. The foundational experimental study demonstrating sunk cost effects: subjects paid more to attend worse outcomes rather than "waste" prior investment. Replicated across ski trips, business investments, and policy decisions.

The most vivid real-world demonstration of sunk cost thinking at scale is the Concorde programme. By the mid-1970s, British and French government analysts had concluded that the supersonic passenger jet would never be commercially viable. The technical achievement was genuine; the economic case was not. But both governments had already committed hundreds of millions of pounds. Stopping felt like admitting the money was wasted. So they kept going — for decades — subsidizing an operation that lost money with every flight until 2003.

Cairns, R.D. & Galbraith, J.W. (1990). Artificial competitiveness and the Concorde. The RAND Journal of Economics, 21(2), 299–312. Documents the decision-making structure behind Concorde continuation post-1975: both governments acknowledged non-viability but continuation was driven by prior commitment, not projected return. A canonical sunk cost case study in policy economics.
"The money was not recoverable. What was recoverable was judgment — and both governments chose not to recover it."

This is the sunk cost fallacy in its clearest form: resources already spent cannot be recovered by continuing. The decision in front of you should be evaluated on its own merits — what it costs from this point forward, against what it gains from this point forward. The past is not a variable. But human cognition treats it like one, and keeps paying for the ghost of what it has already spent.

02

The Forty-Five Minute Problem

A trainer is 45 minutes into an activity that clearly isn't working. The group has lost energy. The learning isn't landing. Participation has dropped from three-quarters to two people carrying the rest. The practitioner can feel the room cooling.

But they spent an hour preparing this activity. They adapted it from a source they trust. They explained it to their co-facilitator yesterday with genuine enthusiasm. And they are 45 minutes in. Stopping now feels like abandoning the investment — the prep time, the explanation, the 45 minutes already spent building toward a payoff that hasn't arrived.

"The sunk cost here is not the activity. It's the preparation. The session is paying the tab for something that was spent days ago."

They keep going. Sometimes the room recovers. More often, it doesn't — and the practitioner spends the debrief quietly reconstructing why the second half fell apart, never quite naming that the answer was visible at the 45-minute mark. The decision not to cut was a sunk cost decision. The cost of making that decision is the learning that didn't happen in the second half.

The practitioner isn't weak or unobservant. Sunk cost effects intensify in proportion to the investment, the visibility of the investment, and the social cost of acknowledging the change. All three are present in a live training room. Arkes and Blumer documented exactly this: the larger the prior investment, the stronger the pull to continue despite evidence.

03

What the Words Know

Sunk
OLD ENGLISH: sincan · past participle "sunken, gone below"
From Old English sincan — to go below the surface, to descend beyond recovery. A ship that has sunk is not beneath the water waiting to be raised; it is past the waterline in the irreversible sense. The word's precision is its lesson: once something has sunk, the question of recovery is no longer about the thing that sank. It's about the navigator.
Salvage
FRENCH/LATIN: salvage, salvare · "to save"
From Old French salvage and Latin salvare — to save, to rescue. Maritime salvage does not recover the cost of building the ship; it recovers what remains useful from the wreck. The distinction matters for practitioners: you cannot salvage the preparation hours, the session time already spent, the prior investment. But you can salvage the learning, the relationship with the group, the next decision. The ship is sunk. The salvage operation is always future-facing.
Cost
OLD FRENCH: coste · "expense, outlay" from Latin constare, "to stand together"
The Latin constare meant to stand firm, to hold a price — to be worth what it costs. A cost that has sunk is one that no longer stands anywhere. It has left the ledger. Calling it a "cost" that still influences decisions is a grammatical ghost — the word persisting after the referent has dissolved.

The practitioner who pivots at 45 minutes is not wasting the preparation. The preparation happened; it has its own value; it produced the first 45 minutes of a session. The practitioner who continues for another 45 minutes of declining returns is not honoring the preparation — they're paying tribute to a ghost.

04

The Evidence Chain

Kahneman and Tversky's Prospect Theory, developed in 1979, provides the mechanism underneath sunk cost fallacy. Losses feel roughly twice as painful as equivalent gains feel pleasurable. This asymmetry — loss aversion — means that abandoning an investment feels like a loss that the brain weighs far more heavily than the equivalent gain from the better decision. The practitioner who pivots is not just making a practical choice; they're absorbing a psychological loss that their brain has pre-weighted against them.

Kahneman, D. & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291. Established loss aversion as a systematic bias: losses are weighted approximately 2× more heavily than equivalent gains. The foundational mechanism explaining why abandoning prior investment feels disproportionately costly.

Barry Staw's research on escalation of commitment (1976, 1981) added a crucial organizational dimension: decision-makers who are personally responsible for an initial decision are significantly more likely to continue investing in it despite negative feedback than those who inherit the decision. Ownership of the original choice intensifies the sunk cost pull. For practitioners, this means the activities they designed themselves, the methods they advocated publicly, and the approaches they've defended to their clients are the most dangerous candidates for sunk cost continuation.

Staw, B.M. (1981). The escalation of commitment to a course of action. Academy of Management Review, 6(4), 577–587. Demonstrated that personal responsibility for an initial decision significantly amplifies escalation of commitment to failing courses of action — directly applicable to practitioners who design and advocate for their own methods.

Thaler and Sunstein, in Nudge, identify commitment devices as one of the primary tools for overcoming sunk cost effects: pre-committing to exit criteria before an investment begins removes the psychological weight of the in-the-moment loss aversion. The session design equivalent is the practitioner who, before delivering an activity, names the conditions under which they will cut it early — and commits to that standard in advance.

Thaler, R.H. & Sunstein, C.R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press. Chapter 3 on commitment devices: pre-commitment to exit criteria reduces the in-the-moment cost of abandoning an investment by framing departure as plan-adherence rather than loss.
05 · WALL MOMENT The Voyage Ledger — The Cost That Isn't There
THREE COLUMNS · THREE PRACTITIONER SCENARIOS · THE REAL DECISION IN FRONT OF YOU
WHAT WAS SPENT
SCENARIO A
3 hours designing a half-day experiential activity. Used twice with prior cohorts to moderate results. Adapted it again for this group. Currently 50 minutes in — group is flat, learning isn't landing.
SCENARIO B
Six months coaching a team leader using a strengths-based framework. Team morale has declined in the last two months. Data says the framework isn't matching this leader's actual development needs.
SCENARIO C
Two years building a curriculum around a particular pedagogical model. New research challenges the model's key assumption. The curriculum still works — but it's working around an error, not despite one.
WHY WE STAY
THE PULL
"I spent 3 hours on this. The co-facilitator knows the plan. Stopping now wastes all of that. Something might still shift in the next 30 minutes. I'll push through."
THE PULL
"I told the organization this approach would work. I've been tracking progress for six months. Saying it isn't working now means admitting the diagnosis was wrong."
THE PULL
"Two years of work. Published materials. Client trust built on this model. Acknowledging a flaw in the assumption undermines everything that came before it."
THE REAL DECISION IN FRONT OF YOU
ZERO THE SUNK COST
"If I had just walked in with 60 minutes and no plan — would I design this activity for this group right now? No. So what would I do instead?" Do that.
ZERO THE SUNK COST
"If this were month one of coaching, and I saw this data, what would my diagnosis be? What would I recommend?" Start from that fresh read — not from the six months of prior narrative.
ZERO THE SUNK COST
"The two years of development happened and produced real value. The new research is data, not a verdict on that work. What does the curriculum need to be true, going forward, to serve learners well?"
06
CONSTRUCTIVE STEWARDSHIP · KEGAN & LAHEY LENS
The Immunity Underneath the Sunk Ship

In Kegan and Lahey's immunity to change framework, the sunk cost fallacy maps onto a competing commitment that is almost always some version of: I need to not be wrong. I need to not be foolish. I need the prior investment to have been worth making. The deep assumption protecting that competing commitment is usually: if I abandon something I invested in, it means the original decision was a failure — and that reflects on my competence, judgment, or character.

This is why sunk cost effects intensify when the investment was publicly visible, when the decision-maker was personally responsible, and when the audience includes people whose respect matters. The immunity isn't protecting the investment — it's protecting the self-concept of the person who made it.

Constructive stewardship offers a reframe: sunk costs are data, not verdicts. The ship is sunk. The navigator is not. The wisdom of a prior investment is not measured by whether you continue it — it's measured by what you carry forward. Meliorism is not about never being wrong; it's about the quality of learning that comes from having been wrong, and the quality of the next decision that learning makes possible.

The practitioner who cuts an activity at 45 minutes, names what they observed, pivots to something better, and uses the debrief to draw meaning from both halves of the session — that practitioner has not wasted the preparation. They've converted it into judgment. That's the salvage operation worth running.

Kegan, R. & Lahey, L.L. (2009). Immunity to Change. Harvard Business Review Press. The competing commitment structure mapped here: the need to not appear foolish or wrong creates an immunity that keeps practitioners escalating commitment rather than recognizing sunk costs as data and recalibrating forward.
07

Your Assignment This Week

THE SUNK SHIP AUDIT
Name one strategy, activity, or approach you are still running that stopped working two months ago — or that you know has a problem you haven't addressed. What is the sunk cost keeping you in it? Be specific: hours invested, relationships built on it, public commitments made. Now ask: if that cost had never existed — if you were deciding fresh today — would you choose this? If not, what would you choose? That is your real decision. The sunk cost is not an argument; it's a weight. You can choose to set it down.

The navigator's practice is not to be indifferent to investment — it's to be precise about what information the investment contains. The preparation was data: it told you this group, this room, this moment needed something specific. When that something stops working, the data has updated. Continuing to honor the sunk investment is not loyalty to good judgment. It's loyalty to the judgment you had before you had this new data.

BIBLIOGRAPHY

  • Arkes, H.R. & Blumer, C. (1985). The psychology of sunk cost. Organizational Behavior and Human Decision Processes, 35(1), 124–140. — The founding experimental study. Multiple experiments demonstrating sunk cost effects across contexts including professional decisions.
  • Kahneman, D. & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291. — Established loss aversion; provides the psychological mechanism underneath sunk cost fallacy.
  • Staw, B.M. (1976). Knee-deep in the Big Muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior and Human Performance, 16(1), 27–44. — First study on escalation of commitment; personal responsibility as an amplifier of sunk cost effects.
  • Staw, B.M. (1981). The escalation of commitment to a course of action. Academy of Management Review, 6(4), 577–587. — Review and theoretical framework for escalation; directly applicable to practitioner decision-making under social visibility.
  • Thaler, R.H. & Sunstein, C.R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press. — Commitment devices as sunk cost antidote; pre-committing to exit criteria as practical application.
  • Cairns, R.D. & Galbraith, J.W. (1990). Artificial competitiveness and the Concorde. The RAND Journal of Economics, 21(2), 299–312. — The Concorde case fully documented; policy sunk cost as historical record.
  • Kahneman, D. (2011). Thinking, Fast and Slow. Farrar, Straus and Giroux. — Chapters 31–32 integrate sunk cost with loss aversion in the accessible synthesis; applied professional examples.
  • Kegan, R. & Lahey, L.L. (2009). Immunity to Change. Harvard Business Review Press. — Competing commitment framework applied to sunk cost: the competing commitment to not appear wrong as the psychological driver of escalation.
§ · The Delight

The Admiral Who Wouldnt Stop Playing Chess

Meliorism2.com · Daily briefings for practitioners
Meliorism 2.0 is a research instrument and daily briefing published by Brian Oney · Meliorist Group, San Francisco.