A tour of the remaining ways your mind deceives you — now with bonus guest appearances from everyone around you
Everything covered in this series so far has been, broadly speaking, an inside job. Cognitive distortions, defense mechanisms, maladaptive schemas, cognitive biases — all of these are errors generated by your own individual mind, working alone, doing its best and producing systematically imperfect results.
This paper is about what happens when other people get involved.
It turns out that the presence, opinions, perceived opinions, imagined opinions, and vague ambient social pressure of other people introduces an entirely new category of thinking errors — errors that cannot be fully explained by individual psychology because they require a group to produce them. You can have confirmation bias entirely by yourself in an empty room. You cannot have groupthink without a committee. You cannot have pluralistic ignorance without a crowd of people all privately believing something while publicly performing the opposite. These are social errors — collective misfires that emerge from the interaction between human minds rather than from any single one of them.
To this we’ll add several concepts from behavioral economics — the field that documented, with increasing delight, that humans make money decisions in ways that are not merely imperfect but reliably, predictably, almost comically irrational. And from epistemology — the branch of philosophy concerned with how we know what we know, and what goes wrong when we convince ourselves we know things we don’t. And from rhetoric — the ancient art of persuasion, updated for the modern era of weasel words, coded language, and phrases specifically designed to make you stop thinking.
This is, in other words, the paper about errors that your social environment, your wallet, your epistemology, and other people’s words contribute to the considerable cognitive chaos already in progress. There is quite a bit to cover. Some of it is alarming. Some of it is, in the specific way that recognition humor works, very funny.
Let’s begin.
Part One: Social Psychology — When Groups Make Everyone Stupider
1. Groupthink
What it is: The phenomenon where the desire for harmony and cohesion in a group overrides realistic, critical appraisal of alternatives — where members suppress dissenting opinions, avoid raising uncomfortable information, and collectively arrive at conclusions that no individual member, thinking alone, would endorse.
Groupthink is not just “a group making a bad decision.” Groups make bad decisions all the time for ordinary reasons — incomplete information, poor leadership, bad luck, the usual. Groupthink is specifically the phenomenon where the group’s social dynamics actively interfere with the group’s ability to think. The desire to not rock the boat, to not be the dissenting voice, to maintain the warm experience of consensus becomes stronger than the desire to be accurate. The group does not think badly because it lacks information. It thinks badly because it has a structural incentive to not use the information it has.
The term was coined by social psychologist Irving Janis in 1972, following his analysis of catastrophic American foreign policy failures — the Bay of Pigs invasion, the escalation in Vietnam — in which groups of genuinely intelligent, experienced people produced decisions that seem, in retrospect, obviously, almost inexplicably bad. Janis’s answer: not stupidity, not malice, but the specific dynamics of cohesive groups under pressure, where the social cost of dissent became higher than the analytical cost of conformity.
The classic symptoms are recognizable: the illusion of unanimity (silence is interpreted as agreement, so the group believes everyone agrees when really everyone is just quiet); collective rationalization (the group generates post-hoc justifications for the chosen direction, which get shared and reinforced until they feel like reasons rather than rationalizations); self-censorship (members who have doubts don’t raise them, because nobody else seems to be raising them); and the emergence of self-appointed “mindguards” — members who actively protect the group from information that might challenge the consensus, without being asked to.
In addiction and recovery, groupthink operates in several distinct and occasionally ironic ways:
During active addiction, the most consequential version is the using group — the social circle organized around shared substance use, where the implicit norms of the group suppress any individual’s private doubts. Nobody wants to be the person who says the thing everybody is privately thinking. The person who is starting to wonder whether this is getting out of hand looks around, sees apparently unworried faces, and concludes: I must be the only one who’s concerned. It must be fine. Meanwhile, three other people in the group are thinking the exact same thought and reaching the exact same conclusion.
This is groupthink and pluralistic ignorance operating simultaneously, which is a particularly efficient combination for maintaining a status quo that nobody fully endorses.
In recovery communities, groupthink can appear in subtler forms that deserve acknowledgment. Recovery communities are, at their best, sources of genuine support, practical wisdom, and the corrective social experience that addiction’s isolation produced. They can also, at their worst, develop the same social dynamics as any other cohesive group: pressure toward doctrinal conformity, discomfort with members who express doubt or heterodox views, the suppression of individual complexity in favor of narrative alignment with community norms.
This is not an argument against recovery communities — it is an argument for the specific kind of critical engagement within them that distinguishes genuine reflection from social performance. The person who raises a hand and says “I’m not sure I agree with that” in a recovery context is not disrupting the community. They are providing the analytical dissent that any group needs to avoid collective stupidity. They are also, incidentally, demonstrating exactly the kind of independent thinking that recovery requires and active addiction suppressed.
In treatment settings, groupthink can produce the phenomenon of everyone presenting as more aligned with treatment goals than they actually are — the group culture of visible compliance that conceals private uncertainty, doubt, or incomplete buy-in. This is understandable. It is also a problem, because treatment cannot address doubts it doesn’t know are present.
A fuller example:
A residential treatment group has been together for three weeks. The group has developed strong cohesion — genuine warmth, real connection, shared humor. One member, David, has been privately struggling with a specific recommendation from the treatment team: the suggestion that he consider ending his relationship with his girlfriend, whose own substance use makes her, the team believes, a significant relapse risk.
David disagrees with this recommendation but is aware that several other group members have made similar relationship changes and spoken about it positively. In group sessions, he has been presenting as aligned with the recommendation — nodding, acknowledging its wisdom, not raising the complications.
What David doesn’t know: two other group members have made similar relationship changes that they are also privately ambivalent about, and have similarly not raised their ambivalence, because the group culture has developed an implicit consensus position on the subject that nobody wants to be the first to question.
In a particularly honest session, another member — someone with nothing to lose because she’s leaving the next day — says: “I’ve been wondering if we’ve all been agreeing with the therapists on this to avoid making things awkward. Because I actually think my situation is more complicated than the way we’ve been talking about it.”
The room goes very quiet and then, in about thirty seconds, becomes very honest. Three people have been silently agreeing with things they privately doubted. The conversation that follows is the most productive session the group has had.
The dissenting voice was the therapeutic gift. It just needed someone to be willing to give it.
The corrective practice: In any group setting — treatment, recovery community, family discussion — deliberately ask: Is there something I genuinely think that I’m not saying because I don’t want to disrupt the consensus? If yes, consider saying it. The value of the group depends on the quality of information it has access to, and the quality of information depends on people actually contributing what they actually think.
2. Pluralistic Ignorance
What it is: A situation where a majority of group members privately reject a norm or belief, but each assumes that most others accept it — and therefore goes along with it publicly, despite privately disagreeing. The result is a group in which nearly everyone is performing a consensus that nearly nobody actually holds.
This is one of the most conceptually elegant errors in social psychology, partly because it requires a very specific and counterintuitive mechanism to produce: a majority of people being wrong about what the majority believes. It sounds like it shouldn’t be possible. It happens all the time.
The classroom version is the canonical example: the professor finishes explaining a complex concept and asks if there are any questions. Nobody raises their hand. The professor concludes everyone understood. Every student concludes everyone else understood. In fact, most students are confused but don’t want to be the only person who didn’t get it. The result: universal confusion, collectively invisible.
The mechanism is the false interpretation of others’ silence. Silence and apparent compliance are read as genuine agreement, when they are actually the product of exactly the same reasoning that’s producing your own silence and apparent compliance.
In addiction and recovery:
Pluralistic ignorance is a powerful force in the maintenance of active addiction, particularly in professional, social, or cultural contexts where heavy use is normalized. “I drink the same amount as everyone at this firm/on this team/in this friend group.” This observation may be accurate. But the comparative context has itself been shaped by pluralistic ignorance: everyone is drinking more than they’re comfortable with, and nobody is saying so, so everyone looks around and sees apparent comfort with the level of consumption and concludes the level must be fine.
In recovery settings, pluralistic ignorance produces the specific dynamic where people assume others in the room have it more together than they do. “Everyone else here seems to have really worked through their stuff. I’m the only one still struggling with X.” The person next to you is thinking: “Everyone else here seems to have really worked through their stuff. I’m the only one still struggling with Y.” The person next to them: “Everyone else here seems to have really worked through their stuff. I’m the only one still struggling with Z.”
This is one of the reasons that honest sharing in recovery communities produces such disproportionate relief — not just for the sharer but for the entire room. The person who says the true difficult thing breaks the pluralistic ignorance and reveals, suddenly, that the private struggle was never as private as it appeared. The almost universal response is some version of: “Oh thank God, I thought I was the only one.”
Nobody is the only one. Almost everyone thinks they’re the only one. This is pluralistic ignorance, and it is defeated by the simple, courageous act of going first.
The corrective practice: When you find yourself assuming that others have resolved something you haven’t, ask: How do I know what others have resolved? What are they actually showing me, versus what am I inferring from their silence? Then, when possible, go first. Say the thing you’re assuming nobody else is struggling with. You will be wrong about being alone with it. The wrongness will be relieving for everyone.
3. Social Loafing (and Its Rationalization)
What it is: The tendency to exert less effort on a task when working in a group than when working alone — the diffusion of individual effort that occurs when the contribution is shared and individual accountability is reduced.
Social loafing is primarily a behavioral phenomenon rather than a thinking error, but it generates a characteristic set of rationalizations that are worth examining: “My contribution won’t make a difference anyway.” “Someone else will handle it.” “The group is taking care of it.” These internal justifications are not just post-hoc explanations for laziness — they are genuinely believed in the moment, and they represent a systematic undervaluation of individual contribution and a systematic overestimation of collective coverage.
In recovery, social loafing manifests in a specific and important way: the group dynamic of recovery can, paradoxically, provide a context in which individual engagement is reduced because the group appears to be doing the work. The person who attends every meeting but never actually engages is social loafing in the therapeutic context. The person who has a sponsor but doesn’t actually call them because “I have a sponsor” is treating the structure as coverage for the behavior the structure is supposed to produce.
Recovery is not a spectator sport. The group does not do the work on your behalf; it creates the conditions under which the work is possible. The individual work — the honest self-examination, the actual application of new communication patterns, the behavioral changes in daily life — remains irreducibly individual. The group provides the container. You still have to bring the contents.
A fuller example:
Thomas has been attending the same home group for five months. He arrives on time, makes coffee, knows everyone’s name, and is genuinely liked. He has not shared anything of substance in a meeting in three months. When his therapist asks how the meetings are going, he says: “Really well — great group.”
What Thomas is describing is the experience of being in a good group. What he is not describing is recovery work. The group’s warmth has become, without his noticing, a substitute for the engagement that was supposed to happen inside it. He is present. He is comfortable. He is, in the clinical sense, loafing — letting the collective energy of the group register in his mind as individual progress, when the individual work has quietly stopped.
His therapist asks a diagnostic question: “When did you last say something in a meeting that cost you something to say?”
Thomas is quiet for a moment. “A while.”
“That’s the work,” his therapist says. “The meeting is where you do it, not where it gets done for you.”
This is the distinction between the structure of recovery and the substance of it. The structure — meetings, sponsor, therapy — is necessary but not sufficient. It is the scaffold. What gets built on the scaffold is individual, effortful, and irreplaceable. Social loafing in recovery is what happens when the scaffold starts to feel like the building.
The corrective question: Am I participating in recovery, or am I attending it? These are different activities with very different outcomes.
Part Two: Behavioral Economics — How You Ruin Your Own Finances and Decisions
4. The Sunk Cost Fallacy (Revisited Through the Addiction Lens)
We touched on the sunk cost fallacy in the previous paper on cognitive biases, but it deserves expanded treatment here because behavioral economics offers a richer framework for understanding exactly why it is so hard to escape — and because in the context of addiction, it operates with a specific cruelty that makes it worth examining at length.
What it is: The tendency to continue investing in a failing course of action because of the resources already spent — time, money, energy, identity — rather than making decisions based purely on the prospective value of future investment. The sunk costs are, by definition, gone regardless of what you decide next. Rational decision theory says they are therefore irrelevant to the decision. Human psychology says they feel overwhelmingly relevant, in ways that frequently produce the opposite of rational decisions.
The sunk cost fallacy is why people finish bad books, stay in worse restaurants, remain in failing projects long past the point where the data clearly indicated it was time to leave, and — most relevantly here — continue patterns of use long past the point where the balance of evidence indicated it was causing more damage than relief.
It is not irrationality exactly. It is a specific kind of loss aversion applied to time and effort rather than money — the psychological experience of “I cannot let all of that go to waste” that makes the cost of cutting your losses feel higher than it mathematically is.
In addiction, the sunk cost fallacy operates at several levels simultaneously:
At the identity level: “I’ve been a drinker for twenty years. Quitting now would mean those twenty years were wasted.” This is the most fundamental form — the investment in a self-concept, an identity, a way of being in the world, that has accumulated enough history to feel like a sunk cost too large to abandon. The drinking is not just a behavior; it is twenty years of experience, relationship, self-understanding. Changing it means letting all of that be the wrong thing. The sunk cost of the identity is the most powerful.
At the relationship level: “My using friends are the people who’ve known me longest. If I get sober I lose them.” The investment in these relationships — real relationships, often genuinely valuable in other respects — creates a sunk cost that keeps people in social contexts that maintain the addiction. The cost of leaving feels higher than the cost of staying, because staying preserves an investment already made.
At the treatment level: The sunk cost fallacy produces the characteristic resistance to changing approaches that aren’t working — staying with a sponsor who isn’t helping because of how much has been disclosed, staying with a treatment modality that’s plateaued because of how much has been invested in it, not seeking higher levels of care because of how much has already been spent on lower ones. Each of these involves mistaking the investment already made for a reason to continue, rather than evaluating whether continuing produces the best future outcome.
The particularly cruel version: In the moment of relapse, the sunk cost fallacy can function as a cognitive accelerant. “I’ve already slipped. The sober time is gone. I might as well continue.” This is the abstinence violation effect plus the sunk cost fallacy: the investment in sobriety has been “lost” by the slip, so the cost of continuing to use has been, in this distorted accounting, already paid. The relapse thus generates its own momentum through the logic of the sunk cost.
The corrective for this specific version is something the series has already covered from the CBT angle: the investment in sobriety is not erased by a slip. The neural changes, the coping strategies, the self-knowledge — these are not stored in the sober-day counter. They are stored in you. They don’t reset.
The corrective practice: For any decision about continuing versus changing a course of action in recovery, write down the future costs and benefits separately from the past investment. Then make the decision as though the past investment belongs to someone else — because in the sense that matters for this decision, it does. The past is spent. The question is only what happens next.
5. The Endowment Effect
What it is: The tendency to assign higher value to things simply because you own them — to demand more to give something up than you would have been willing to pay to acquire it in the first place.
This has been demonstrated experimentally with remarkable consistency: give someone a coffee mug, and they’ll demand significantly more money to sell it than they would have been willing to pay to buy it five minutes earlier. The mug hasn’t changed. Its value hasn’t changed. Their relationship to it has. Ownership, it turns out, creates a psychological attachment that inflates perceived value independent of any objective assessment.
In addiction and recovery, the endowment effect applies to some surprising possessions:
The addiction itself. This sounds counterintuitive until you consider that people in active addiction often describe their relationship with substances in terms that have ownership quality — it’s their way of coping, their release, their one thing. The addiction has been “owned” long enough that its subjective value has been inflated by the endowment effect, independent of its actual contribution to wellbeing. The demand price to give it up — what it would take to be willing to relinquish it — has been inflated by ownership beyond any objective assessment of what it is actually providing.
The coping strategies developed during addiction. Familiar coping strategies — including dysfunctional ones — are “owned” and therefore overvalued. Proposing to replace them with new strategies triggers the endowment effect: the familiar strategy is perceived as more valuable than the unfamiliar replacement, even when the evidence clearly favors the replacement. The resistance to new coping mechanisms in early recovery is not always stubbornness. It is frequently the endowment effect treating the current strategy as more valuable than it is because it is already owned.
The identity narrative. “This is who I am” applied to patterns of behavior and self-concept that happen to organize around addiction is the endowment effect operating on personal identity. The narrative is owned, and owned things are overvalued, and therefore the cost of giving it up feels higher than the benefit of an alternative would — regardless of how compelling the alternative actually is.
A fuller example:
Elena has been in early recovery for four months. Her therapist has been recommending a specific breathing and grounding practice for managing anxiety — a structured, evidence-based technique that takes about ten minutes and has a solid research record for people in her situation.
Elena has tried it twice and found it awkward and artificial. She continues managing anxiety primarily by going very quiet and withdrawn — a strategy she developed during her using years that keeps other people at a distance but provides a kind of internal pressure-release. It doesn’t work well. She knows it doesn’t work well. She continues to prefer it.
Her therapist asks her to describe the withdrawal strategy in terms of what it’s actually providing. Elena lists: familiar, automatic, doesn’t require anything from her, has been with her for years.
Her therapist asks: “If you’d never learned to withdraw and I offered you these two options — grounding practice or going silent and distant — which would you choose for managing anxiety?”
Elena pauses. “The breathing thing, probably. It actually works.”
“So what you’re valuing about withdrawal,” her therapist says, “is that you already own it.”
The endowment effect doesn’t just apply to the substance. It applies to every piece of the addiction’s infrastructure — every coping strategy, every identity narrative, every relational pattern that got built during the using years and is now being carried into recovery because it was acquired during a period when it seemed to help. Ownership inflates value. The unfamiliar replacement, whatever its actual merits, starts at a disadvantage simply because it hasn’t been owned yet.
The corrective question: If I didn’t already have this — this coping strategy, this identity, this way of being — would I pay what I’m currently paying to acquire it? The gap between your answer to that question and your actual behavior is the endowment effect at work.
6. Mental Accounting
What it is: The tendency to treat money (or other resources) differently depending on its origin, intended use, or mental “account” — rather than treating all money as equivalent regardless of where it came from or what bucket it’s been assigned to.
The classic example: people spend money from a tax refund more freely than money from their paycheck, even though both are equally theirs, because the tax refund has been mentally categorized as “windfall” rather than “earned income.” The same $200, depending on which mental account it was deposited in, produces different spending behavior.
People also maintain separate mental accounts for “food money,” “entertainment money,” “savings” — and will refuse to transfer between them in ways that make no mathematical sense. Someone who would not spend $50 of savings on dinner will happily spend a $50 gift card on the same dinner, because the gift card came from a different mental account that was already earmarked as expendable.
In addiction and recovery, mental accounting is less about money and more about effort, cost, and permission:
Effort mental accounting: The person in recovery who works very hard on certain aspects of their recovery and uses that effort to create a mental credit toward slacking on others. “I’ve done so much work this week, I can skip the meeting.” The work done in one account does not transfer to another, but the mental accounting feels like it does.
The windfall permission structure: “I’ve had a terrible week — I deserve a break from being rigorous.” Suffering generates a mental credit toward reduced recovery engagement, as though the pain paid for some amount of reduced effort. It didn’t, mathematically. It feels like it did. And this is precisely the mechanism — a hard week generating a subjective credit toward loosened standards — that relapse research consistently identifies as one of the highest-risk cognitive patterns. Acute stress plus mental accounting plus a credit balance is one of the most common configurations preceding relapse.
Recovery “money” vs. “regular money”: The tendency to treat recovery as a separate domain with its own accounting, sealed off from regular life — to work very hard “in recovery mode” during meetings and therapy sessions and then behave in ways that undermine recovery in daily life, as though the two accounts don’t communicate. They do. The boundaries of the mental account are not the actual boundaries of the effect.
A fuller example:
Victor has had an unusually productive month. He completed a step with his sponsor, attended every meeting he committed to, had a particularly honest session with his therapist, and had a genuinely difficult conversation with his brother that he’d been avoiding for months. By any honest assessment, he has done substantial recovery work.
On Friday, his work colleagues invite him to a bar after work. Victor’s internal accounting runs quickly: I’ve been so on-point this month. I’ve earned a normal night. I can handle this. He goes. The environment is higher-risk than he has been in for several months. He manages it, this time. But the decision was made not by evaluating the risk of the situation but by consulting a mental ledger of effort expended — as though hard work in recovery is a currency that can be spent on reduced vigilance.
His sponsor, when Victor describes this later, asks: “Would you say that a week of perfect eating earns you permission to eat badly? Does the good week protect you during the bad one?”
Victor sees it immediately. “The bank account doesn’t work that way.”
“No,” his sponsor says. “And neither does this one. What you built this month is real. It doesn’t spend.”
This is the corrective insight at the center of mental accounting in recovery: the work done is not a credit balance available for withdrawal against risk. It is more like fitness — maintained through continued effort, not a reserve that accumulates and can be spent on reduced effort. The month of excellent recovery work doesn’t make next Friday’s bar lower-risk. It makes the next month of recovery possible.
The corrective practice: Treat recovery resources — time, energy, engagement — as fungible. What you spend in one place comes from the same pool as every other place. A hard week is not a credit toward a holiday from recovery. A productive therapy session is not permission to skip the meeting. There is one account, and it has the same balance regardless of which sub-account you think you’re drawing from.
Part Three: Epistemology — What We Think We Know, and Why We’re Often Wrong About Both
7. Wishful Thinking
What it is: The formation of beliefs based on what would be pleasant or convenient to believe, rather than based on evidence, logical reasoning, or reality. The mind’s tendency to construct a version of events that matches desire rather than fact, and then to treat that version as accurate.
Wishful thinking is the epistemological equivalent of grading your own exam, giving yourself partial credit for every answer, rounding all the partial credits up, and concluding with satisfaction that you’ve passed. The test was real. The grading was not.
This is the root of a remarkable number of the errors this series has covered. Confirmation bias is, at bottom, wishful thinking applied to information-gathering. Optimism bias is wishful thinking applied to probability. Self-serving bias is wishful thinking applied to attribution. The endowment effect is wishful thinking applied to valuation. Denial is wishful thinking applied to inconvenient reality. The sunk cost fallacy is wishful thinking applied to the narrative that past investment must have been worthwhile. Wishful thinking is not one error among many. It is the substrate from which many of them grow.
In addiction and recovery, wishful thinking is such a pervasive feature that it has been essentially catalogued as a symptom:
“I can moderate.” (Wishful thinking about current capability.)
“This time will be different.” (Wishful thinking about future behavior, without changed conditions.)
“I think I’ve basically beaten this.” (Wishful thinking about the trajectory of recovery.)
“My family is going to forgive me quickly once they see I’ve changed.” (Wishful thinking about others’ healing timelines.)
“I can go to this high-risk event and it will be fine.” (Wishful thinking about resilience in untested conditions.)
Each of these thoughts might, occasionally, be accurate. The problem is that they are generated by desire rather than by evidence, which means they are as likely to be wrong as the desire is strong. And in addiction, the desires producing the wishful thoughts are extremely strong — which means the wishful thinking is extremely convincing and frequently extremely inaccurate.
The corrective practice: When you notice yourself believing something that it would be convenient or pleasant to believe, apply what philosophers call the principle of epistemic charity to your own thinking — the inverse of the usual application. Normally epistemic charity means giving others the benefit of the doubt. Applied to your own wishful thinking, it means asking: Am I believing this because the evidence supports it, or because I want it to be true? This is not a question that eliminates wishful thinking. It is a question that introduces a pause before you act on it.
8. Motivated Reasoning
What it is: The tendency to use cognitive processes — reasoning, evaluation, memory, interpretation — in the service of reaching a desired conclusion rather than an accurate one. Motivated reasoning is not the same as wishful thinking: wishful thinking simply believes what it wants; motivated reasoning works to justify what it wants, deploying genuine intellectual effort in service of a predetermined destination.
This distinction is important. Motivated reasoning is not lazy. It can be extraordinarily sophisticated. The person engaging in motivated reasoning is often not aware they are doing it — they experience themselves as reasoning carefully and rigorously toward a conclusion. What they are actually doing is using their reasoning capacity to construct the most defensible case for the conclusion they already hold, while applying much greater scrutiny to evidence that challenges it.
This is why intelligent people are not necessarily better reasoners than less intelligent people — and in some documented cases are actually worse. A higher intelligence provides a more powerful tool for motivated reasoning: better at finding counterarguments to inconvenient evidence, better at generating plausible alternative explanations for uncomfortable data, better at constructing internally consistent justifications for conclusions that are not actually justified.
The philosopher coined the term “galaxy-brained” for the specific phenomenon of very intelligent people reasoning their way to absurd conclusions through chains of logic that are individually plausible and collectively catastrophic. Recovery communities have their own informal name for this: “stinking thinking,” which lacks the elegance of the philosophical term but captures the same thing with considerably more texture.
In addiction, motivated reasoning is the engine of sustained rationalization:
The person with a substance problem is not typically reasoning badly. They are often reasoning with considerable sophistication — toward conclusions that protect the addiction. Every counterargument to the mounting evidence of a problem is evaluated carefully and found wanting. Every alternative explanation for concerning symptoms is developed with genuine creativity. The result is a case for the defense that is structurally sound and fundamentally dishonest — not dishonest in the sense of deliberate lying, but dishonest in the sense of being motivated by the conclusion rather than by the evidence.
In recovery, motivated reasoning does not disappear. It redirects. The question becomes: what is the current motivated conclusion, and what evidence is the reasoning being deployed to protect?
“I don’t need to go to as many meetings anymore.” Is this an evidence-based conclusion, or is it the conclusion that motivated reasoning is building a case for?
“My sponsor doesn’t really understand my specific situation.” Evidence-based assessment, or the motivated reasoning of someone whose sponsor has been asking inconvenient questions?
“I’m not like the people who relapse.” The conclusion first, the reasoning second.
The corrective practice: When you find yourself building an elaborate case for something, ask: Did I start with the evidence and reach this conclusion, or did I start with the conclusion and find the evidence? The former is reasoning. The latter is motivated reasoning dressed as reasoning. The distinction matters because the former is updatable when better evidence arrives, and the latter is not.
9. The Dunning-Kruger Effect
What it is: The cognitive phenomenon where people with limited knowledge or skill in a domain significantly overestimate their ability in that domain — not because they are arrogant, but because the skills required to accurately assess competence are the same skills required to have competence, so the gap in one produces a gap in the other.
This was documented in a famous 1999 study by David Dunning and Justin Kruger, who found that students who scored in the bottom quartile on tests of logical reasoning, grammar, and humor consistently overestimated their performance — placing themselves near the 70th percentile when they were actually at the 12th. The students who scored highest, meanwhile, slightly underestimated their performance.
The mechanism: competence in an area provides the metacognitive tools to accurately assess performance in that area. Without the competence, you also lack the tools to notice its absence. The unskilled person doesn’t know what good looks like, so they cannot recognize that their work doesn’t look like it.
This is not stupidity. It is a structural feature of learning. Everyone who has ever become genuinely skilled at something has passed through the Dunning-Kruger zone — the early phase where you don’t know enough to know what you don’t know. The problem arises when this phase is not recognized as a phase, when the early confidence of low competence is mistaken for the earned confidence of high competence, and when decisions are made accordingly.
Dunning himself described it evocatively: the first peak of the learning curve is a peak of confidence, not of skill. True competence, when it arrives, is usually accompanied by a more accurate — and therefore more humble — assessment of how much there is still to learn.
In addiction and recovery, the Dunning-Kruger effect operates with notable precision in several areas:
On the addiction itself: “I know my limits.” “I know when I’ve had enough.” “I understand my own relationship with alcohol better than anyone else.” These are claims to expertise in the domain of one’s own substance use — exactly the domain that the addiction has most comprehensively disrupted. The person whose judgment about their own use has been most thoroughly compromised by years of motivated reasoning, denial, and neurological disruption is, by the Dunning-Kruger mechanism, least equipped to accurately assess the extent of that compromise. This is not a character failing. It is the structure of the problem.
On recovery itself: At roughly three to six months of sobriety, many people experience a specific phenomenon that treatment professionals have observed often enough to have given it names: the “pink cloud” or “honeymoon phase,” in which the early relief of sobriety, the neurochemical stabilization, and the genuine early progress create a sense of having essentially solved the problem. The person at three months often has more confidence in their recovery than the person at three years — because at three months, they don’t yet know what they don’t know. The person at three years has encountered enough of the actual territory to have a more accurate map of it.
This is the Dunning-Kruger arc of recovery: early confidence, followed by the humbling encounter with genuine complexity, followed by earned confidence if the work continues. The dangerous moment is when the early confidence produces decisions — reduced treatment, distanced support, declared victory — that are appropriate to earned competence but are being made from the peak of the Dunning-Kruger curve.
On interpersonal skills and communication: Many people in recovery spent years in relationships where emotional intelligence and communication were severely compromised by the addiction’s demands. The person who enters recovery with genuinely limited interpersonal skills is, by the Dunning-Kruger mechanism, also limited in their ability to accurately assess those skills. They may believe themselves to be better communicators, more emotionally attuned, more effective in relationships than the evidence would support — because the evaluative tools are underdeveloped alongside the skills being evaluated.
A fuller example:
Marcus has four months of sobriety and has begun to feel, genuinely and with some justification, significantly better than he did during active addiction. He attends meetings, works with a therapist, and has a sponsor. He also has a growing sense that he has substantially figured this out — that the major insights have been had, the fundamental work has been done, and the remaining task is primarily maintenance.
He mentions to his sponsor that he’s been thinking about reducing his therapy sessions to monthly.
His sponsor, who has fifteen years of sobriety and has watched many people navigate the four-month window, asks a single question: “Can you describe the specific skills you still need to develop for long-term sobriety?”
Marcus begins to answer. The answer reveals, as he speaks it, that the list is considerably shorter than a more experienced assessment would produce — not because he has worked through everything on the list, but because he doesn’t yet know what’s on it. His metacognitive assessment of his own recovery is operating at four months. His actual recovery situation has considerably more territory than he currently has the map to see.
This is not dishonesty. It is the Dunning-Kruger effect in its purest form: not knowing what you don’t know because knowing what you don’t know requires the very knowledge you don’t have.
His sponsor says: “In my experience, the people who most need to keep their therapy are the ones who feel most like they might not need it anymore.”
Marcus keeps the therapy. Six months later, he understands the conversation in a way he didn’t when it happened.
The corrective practice: Seek external assessment of your competence in domains where your own assessment is least trustworthy — and your own assessment is least trustworthy precisely where the stakes are highest and the skill gaps are largest. Ask your sponsor, your therapist, and the people who know you best to tell you what they see — not as a substitute for your own judgment, but as a calibration of it. The person at twelve months who has sought and genuinely absorbed external feedback has a more accurate map of their recovery than the person at twelve months who has primarily relied on their own assessment.
Part Four: Rhetoric and Propaganda — When the Error Is Someone Else’s Gift to You
10. Weasel Words
What it is: Language that is strategically vague, ambiguous, or imprecise — used to create an impression or make a claim while avoiding the commitment that a clear, specific statement would require. The weasel word gives you the feeling of having been told something while actually telling you nothing that can be verified, challenged, or held accountable.
The term allegedly comes from the belief that weasels can suck the contents from an egg while leaving the shell intact — the word does the same to a sentence, removing the meaning while preserving the form.
Classic weasel constructions: “Studies suggest…” (which studies? suggesting what, exactly, with what degree of confidence?); “Experts say…” (which experts? in what context? with what dissent from other experts?); “It is believed…” (by whom? on what basis?); “Up to 50% better results” (up to — a claim that is technically satisfied if the result is 0%); “Some people say…” (some people say the moon is cheese, which makes this phrase capable of introducing literally any claim).
Weasel words are the rhetorical equivalent of the endowment effect applied to information: they give you the feeling of having received something of value — a fact, a statistic, an expert opinion — without actually providing the substance that would make the claim meaningful.
In addiction and recovery, weasel words appear in several contexts that deserve attention:
In the internal monologue: “I’ve been doing pretty well lately.” “Things are kind of improving.” “I more or less have this handled.” These self-addressed weasel words — the hedging, the softening qualifiers, the carefully imprecise language — function to make statements feel like progress reports while evacuating the specific content that would make them accurate or testable. Pretty well by what standard? Kind of improving compared to what? More or less handled in what sense, with what remaining gaps?
Precision is threatening when the precise version of a statement would require a difficult response. Weasel words in the internal monologue are frequently the linguistic vehicle of mild self-deception — close enough to an honest report to feel honest, imprecise enough to avoid the implications of precision.
In early recovery conversations: “I’m working on it.” “I’m doing the best I can.” “I’ve been thinking about this.” These phrases are not false. They are technically true in almost any circumstance. They are also, frequently, used to close conversations that need to remain open — to produce the appearance of progress and engagement while the actual assessment of progress and engagement has been actively avoided. A sponsor who accepts “I’m working on it” as a sufficient answer has been handed a weasel word and let it pass. A sponsor who asks “Working on it how, specifically, and what happened when you tried?” has turned a thought-terminating construction back into a conversation.
In relapse narratives: Weasel words are the syntax of the softened confession. “I kind of had a situation last weekend.” “Things got a little off track.” “I wasn’t doing great for a few days.” Each of these sentences contains, somewhere inside it, information that matters — and has been packaged in language specifically imprecise enough to prevent that information from landing with its full weight. This is not always conscious. It is often the shape that shame takes when it needs to speak but isn’t ready to be fully heard.
A fuller example:
Ana has been in recovery for eight months and has recently had a relapse that lasted four days. She tells her home group about it during a share. Her language: “I went through a rough patch last month — kind of slipped back into some old patterns for a bit, but I’m back on track now.”
Every word of this is technically true. None of it is specific. The group responds with warmth and support. Ana sits down feeling, briefly, better.
Her sponsor pulls her aside afterward. “‘Old patterns for a bit’ — what does that mean, specifically?”
Ana is quiet for a moment. Then she says what actually happened: four days, the specific circumstances that preceded it, the specific moment she decided to call her sponsor, the specific fears she has about what it means for her recovery.
The sponsor listens through all of it. “That’s the share,” she says. “The one you just gave me. The group got the weasel word version. You got to keep the real version to yourself, which means you also kept the weight of it. That’s not easier. It just looks easier.”
Weasel words in the recovery context are not primarily about deceiving others. They are about managing the internal experience of a difficult truth by keeping it imprecise enough that it cannot demand a full response. The cost is that an imprecise truth also cannot receive full support. You can only be helped as specifically as you can be known.
From outside: Recovery-adjacent pseudoscience, wellness products, and some treatment modalities deploy weasel words as their primary epistemic currency. “Holistic approaches can support wellness.” “This method has helped many people.” “Results may vary.” “May promote healing.” When evaluating any claim made about your recovery or treatment, apply a simple test: could I, in principle, verify this claim? Is there a specific, falsifiable assertion here, or is there only the impression of one? Evidence-based treatment makes specific claims that can be checked. Weasel-word treatment makes impressions that cannot.
The corrective practice: Replace weasel words with specific language, especially in your own internal reporting and in conversations with people who are positioned to help. “I’m doing pretty well” → “I attended four meetings this week, called my sponsor twice, had one craving that I managed using my coping plan, and missed one check-in I had committed to.” The specificity is uncomfortable. It is also the only version of a progress report that can actually be used — by you, or by anyone trying to help you.
11. Thought-Terminating Clichés
What it is: Commonly used phrases — simple, catchy, culturally familiar — that function to end inquiry, quell cognitive dissonance, and shut down further critical thinking. The phrase does not resolve the underlying tension; it terminates the engagement with it. The thinking stops not because a conclusion has been reached, but because a sufficiently familiar form of words has been produced.
The term was coined by psychiatrist Robert Jay Lifton in his 1961 study of thought reform — the psychological techniques used by totalitarian regimes and cult systems to prevent members from thinking critically about the group’s claims and demands. Thought-terminating clichés were identified as a key mechanism: simple, emotionally resonant phrases that, when deployed, produce a feeling of resolution without the content of resolution.
“It is what it is.” (Invites acceptance without specifying what the thing is, whether acceptance is appropriate, or what action might be warranted.)
“Everything happens for a reason.” (Closes the question of whether the current situation requires a response, whether something could have been done differently, and what lessons might be drawn — all of which are potentially useful forms of thinking.)
“Let go and let God.” (Enormously meaningful in the right context. Also deployable as a thought-terminator that converts appropriate human agency into passive waiting.)
“One day at a time.” (Genuinely useful as a reminder not to catastrophize the future. Also deployable as a reason not to engage with future planning, structural change, or long-term consequences that deserve attention.)
“It’s all part of the process.” (Sometimes accurate. Also available as a reason to not examine whether what’s happening is actually part of a healthy process or is a problem that needs addressing.)
The recovery community has, alongside its genuine wisdom, developed a particular rich vocabulary of clichés — and this is worth naming not as a criticism but as an observation that requires thoughtful application. The clichés are often compressed wisdom from long experience. “One day at a time” reflects genuine insight about the management of anxiety and the proper scale of daily decision-making. “Fake it till you make it” reflects genuine insight about behavioral practice preceding motivational shift. “This too shall pass” reflects genuine insight about the impermanence of acute distress.
The error is not in the clichés themselves. The error is in deploying them as thought-terminators — using the familiar form of the phrase to produce the feeling of an answer without doing the work of actually thinking about whether the answer applies here, in this specific situation, for this specific person, at this specific moment.
In recovery, thought-terminating clichés function most dangerously when they prevent the processing of legitimate concern:
“I was powerless over it — that’s just step one.” (Legitimate insight about the nature of addiction. Also deployable as a reason not to examine the specific decisions and patterns that preceded the relapse.)
“Progress, not perfection.” (Genuinely important corrective to unrelenting standards. Also deployable as a reason to not examine whether a specific behavior actually constitutes progress rather than stagnation.)
“You’re exactly where you’re supposed to be.” (Sometimes deeply comforting and accurate. Also deployable to a person who is not, in any useful sense, where they’re supposed to be, and who needs to move rather than be comforted.)
The test is whether the cliché is being used to arrive at a conclusion or to avoid examining one. Wisdom that invites further thinking is wisdom. Wisdom that terminates thinking is a thought-terminating cliché, regardless of how much wisdom it contains in other contexts.
A fuller example:
Diane has been sober for six months and has just had a difficult week — a near-relapse that she has not fully processed. She brings it to her home group. Several members respond with warmth and genuine care. One member says: “This too shall pass.” Another: “You’re exactly where you’re supposed to be.” A third: “One day at a time — just today.”
Diane nods. She feels, for a few minutes, better. She has not been asked what happened, what the conditions were, what the near-relapse was telling her about the current state of her recovery, or what she needs to adjust. The clichés arrived with love, and they closed the conversation.
Her sponsor calls her that evening and asks a different kind of question: “Walk me through the day. What happened, in order, from the beginning?”
This question is not warm in the same way. It is useful in a way the clichés were not. It produces a forty-minute conversation that surfaces three specific patterns that contributed to the near-relapse, two of which Diane hadn’t consciously identified before the conversation.
The clichés said: it’s okay. The question asked: what actually happened? Both are expressions of care. Only one keeps the thinking going.
The corrective practice: When you find yourself producing a cliché — internally or externally — pause and ask: Is this phrase answering the question, or am I using it to stop asking the question? If the latter, what is the question you were about to stop asking? That question probably deserves a real answer.
12. Dog-Whistle Communication
What it is: Language that carries two simultaneous meanings — a surface meaning that appears acceptable or neutral to a general audience, and a second, coded meaning that communicates something specific to a targeted subgroup. The phrase is legible at two levels: the innocuous public level, and the meaningful private level. The gap between the two is where the manipulation lives.
The term originated in political analysis, describing language used by politicians to signal positions to specific constituencies without openly endorsing those positions to broader audiences. But the mechanism is not exclusively political, and recognizing it is useful in any context where communication has layers that are not announced.
In addiction and recovery, dog-whistle communication appears in patterns worth recognizing:
The internal dog-whistle: Thoughts that sound like recovery language while communicating something else to the part of the mind that knows what they mean. “I’m just going to assess the situation.” (To the outside observer: a reasonable, measured approach. To the part of the mind that knows: going to a place where using is likely and giving myself permission to decide in the moment.) “I’ll have just one and see how I feel.” (Surface meaning: moderation. Internal meaning: the relapse has already been decided and this is the permission structure for it.) “I deserve to relax.” (Surface: legitimate recovery from stress. Internal: the drink has been authorized.)
The internal dog-whistle is the language of the relapse voice from the communication paper — the internal narrative that gradually shifts toward using language that sounds like recovery while carrying the encoded content of approaching relapse. Recognizing your own private codes — the phrases you use to yourself that sound acceptable and mean something else — is advanced self-knowledge and one of the most protective things available in long-term recovery.
From others: Learning to recognize when communication is operating at two levels simultaneously — when the surface content and the actual function are different — is a broader skill that recovery specifically develops. The person who spent years managing the perceptions of people around them developed considerable fluency in coded communication. That fluency, reversed and applied to incoming communication, becomes the ability to hear what is actually being said beneath what is being said. This is not paranoia. It is appropriate discernment.
The corrective practice: Audit your internal vocabulary. What phrases do you use to yourself that could mean one thing to an observer and something else to you? What are your personal coded constructions — the internal language that sounds like one thing and functions as another? The most useful version of this practice is to bring the private meanings into explicit language, either in journaling or with a trusted person, and examine whether the surface version and the actual version are carrying the same message.
How These Concepts Connect: The Full Ecology of Flawed Thinking
Stepping back from the individual concepts, what emerges is something more comprehensive than a list of errors. It is an ecology — a system of mutually reinforcing mechanisms, each exploiting and amplifying the others, operating across the full range of human cognitive and social experience.
Wishful thinking provides the raw material. Motivated reasoning provides the engine that processes it into defensible belief. Confirmation bias provides the information-gathering strategy that feeds the engine. The availability heuristic provides the fast-access library of confirming examples. The representativeness heuristic classifies new situations as instances of familiar categories before they can be examined. The affect heuristic converts the emotional outputs of the schema into probability assessments. The Dunning-Kruger effect ensures that the domains most in need of accurate self-assessment are the ones where self-assessment is least reliable. The sunk cost fallacy locks the behavior in once it is established. The endowment effect inflates the value of what is already owned. Mental accounting creates the illusion of separate domains where there are none. Groupthink and pluralistic ignorance provide the social reinforcement that makes individual error feel like collective consensus. Weasel words and thought-terminating clichés provide the linguistic infrastructure for avoiding the thinking that would challenge all of the above.
This is the ecology within which addiction develops, within which it is maintained, and within which recovery has to do its work. It is formidable. It is also not invincible, because every element of this ecology can be interrupted — not eliminated, but interrupted — by the specific practices that this series has been describing across its papers.
The cognitive distortions paper: interrupt the surface errors.
The defense mechanisms paper: interrupt the protective structures.
The cognitive biases paper: interrupt the information-processing errors.
The schemas paper: address the source code.
This paper: recognize the social, economic, epistemic, and rhetorical dimensions.
None of these papers, individually, solves the problem. Together, they provide a map of a territory that you are navigating already, with or without the map. The map does not change the territory. It changes your ability to navigate it — to recognize where you are, to see what’s coming, to make choices with more information than you had before.
That is, in the end, what this entire series has been offering: not a cure, not a guarantee, not the elimination of error. The modest but genuinely useful proposition that a person who knows what their mind tends to do, and why, is better positioned to work with it — and around it, and through it — than a person who doesn’t.
Your mind is not your enemy. It is a complicated ally with some well-documented tendencies, most of which make more sense once you know where they came from. Understanding those tendencies is not the same as transcending them. It is, however, the prerequisite for doing so.
That’s enough to be getting on with.
This post is intended as a complement to professional treatment and support — not a substitute for either. If you or someone you love is struggling with addiction, SAMHSA’s National Helpline is available 24/7, free and confidential, at 1-800-662-4357.