A Person Confronting Their Own Conflicts
Between What We Believe, What We Do, and How We Explain It
We can spend much of our lives defending an idea without ever carefully examining how we came to believe it. Some convictions emerge from personal experience, while others are learned in childhood, absorbed through relationships, or accepted because someone we trust presented them as true. Over time, they become reference points for interpreting events, judging people, and making decisions. We do not always realize that what we regard as our own conclusions may be the product of accumulated influences, inherited interpretations, and explanations developed over many years.
Nor is it unusual to think one way and act another. We defend honesty while finding reasons to withhold information; we value freedom while attempting to control other people’s choices; we consider it important to listen to different opinions but become uncomfortable when someone challenges our own convictions. Such contradictions do not necessarily indicate dishonesty or an intention to deceive. They belong to a human experience in which beliefs, values, needs, and behaviors do not always move in the same direction.
Psychologist Leon Festinger investigated one aspect of this experience when he introduced the theory of cognitive dissonance in 1957. When we become aware of a meaningful inconsistency between what we believe, know, or do, we may experience discomfort that motivates us to restore some degree of coherence. Yet the contradiction alone does not constitute dissonance; what matters is the psychological tension associated with perceiving it. We may hold incompatible ideas without recognizing the conflict, acknowledge inconsistencies without significant distress, or feel uncomfortable about a contradiction that turns out to be only apparent.
Someone who understands the health risks of smoking but continues to smoke may experience this tension. To reduce it, they might quit, downplay the dangers, emphasize the benefits they believe smoking provides, or simply avoid information about its consequences. Some of these responses change the behavior; others change how the behavior is interpreted. The discomfort may diminish in either case, even when the underlying situation remains largely unchanged.
This is where rationalization becomes important. An explanation may clarify why a particular behavior occurred, but it may also serve as a justification designed to protect our self-image or convictions. Someone who acts unfairly may examine the circumstances that influenced their conduct or use those circumstances to dismiss any responsibility. The distinction is not always apparent in the words themselves, because rationalization can employ genuine facts to support conclusions that do not follow from them. We do not have to lie consciously to construct explanations that we eventually come to believe.
The same distinction arises when we invoke personal characteristics, difficulties, or diagnoses to explain our actions. A clinical condition may genuinely limit certain abilities and require accommodations, understanding, and support. Yet knowing the origin of a difficulty does not automatically settle every question concerning its consequences, possible responses, or responsibilities. When an explanation begins to prevent all further questioning, it no longer functions solely as a means of understanding. The problem lies not in the diagnosis, but in how it may be used to end an inquiry that remains unfinished.
Something similar happens when we attribute our decisions to external authorities. The expression “I was just following orders” may accurately describe a situation of subordination, but it can also be used to transfer all responsibility for actions we helped carry out. Stanley Milgram’s experiments in the 1960s demonstrated how a researcher’s authority could influence participants to continue administering what they believed were real electric shocks, despite signs of distress from the person supposedly receiving them. The circumstances of obedience, coercion, and responsibility require careful consideration, but the experiments showed that our decisions are not determined solely by the principles we claim to hold.
Authority, however, does not always announce itself through explicit commands. A uniform, a title, an emblem, or a particular manner of speaking may create an impression of legitimacy before we have verified who is using it. There is a curious account of an automotive magazine supposedly disguising an ordinary car with official-looking insignia, a rotating beacon, and a siren. According to the story, other drivers cleared the way, and even police officers reportedly saluted the vehicle. Regardless of whether every detail of this account can be confirmed, the situation illustrates a familiar mechanism: we may respond to conventional signs of authority as though they were sufficient proof that authority exists.
The influence of others also operates without uniforms, hierarchies, or orders. In Solomon Asch’s experiments, participants were asked to compare the lengths of lines, a task with a visually obvious correct answer. Yet when other people in the room deliberately gave incorrect answers, many participants followed the majority on at least some occasions. Some recognized the error but preferred not to disagree; others began to doubt their own perception. Apparent unanimity interfered with individual judgment, showing that public agreement does not necessarily mean believing what one says.
Outside the laboratory, this influence takes less obvious forms. We may agree with a group to preserve relationships, avoid embarrassment, or maintain a sense of belonging. We may also seek people who share our opinions and gradually distance ourselves from those who challenge them. Imagine someone who regularly calls friends, visits acquaintances, and enjoys long conversations while finding agreement, but withdraws when objections begin to emerge. Their motivations cannot be established from this behavior alone, yet the example reveals how choosing our conversational partners can reduce exposure to uncomfortable information. Eventually, the absence of disagreement may create the impression that our convictions have been widely confirmed.
There is also a difference between agreeing with others and genuinely changing our beliefs. We perform social roles, adapt our language, and modify our behavior according to the setting without necessarily experiencing any meaningful internal conflict. A person does not present themselves in precisely the same way at home, at work, or in the presence of an authority figure. Such adjustments may reflect social competence rather than insincerity. Yet when maintaining a particular position repeatedly requires us to act against our convictions, we may begin developing explanations that make the behavior acceptable. Sometimes we initially play a role out of convenience and eventually reorganize our beliefs to accommodate it.
Social influence does not operate only upon immediate decisions. It also participates in shaping what we consider true. In The Republic, Plato introduces the so-called noble lie, a narrative according to which citizens were formed with different metals in their souls, corresponding to the roles they were meant to perform in the city. The story would help legitimize a particular social organization. What makes it especially revealing is the possibility that a narrative initially recognized as artificial might become a sincerely accepted belief among subsequent generations.
A deliberate lie does not have to remain in the hands of people who know it is false. Their descendants may receive it as tradition, teach it to their children, and defend it as part of their identity. Cultural transmission can preserve explanations for so long that their origins are eventually forgotten. This does not mean that religions, mythologies, legends, and folklore necessarily originated in lies or rationalizations. They also organize experience, transmit values, preserve memories, and offer interpretations of events that people seek to understand. The point is that the persistence of a narrative does not, by itself, establish its correspondence with reality.
When an explanation deeply embedded in a culture encounters events that appear to contradict it, new interpretations may emerge to preserve its coherence. A community that believes in the infallible effectiveness of a ritual may attribute its failure to improper performance, a transgression, or the intervention of another force. Such interpretations may belong to a living tradition, but they may also function as rationalizations when used exclusively to prevent any reconsideration of the original belief. Similar processes can occur in organizations, social movements, and personal relationships without having any religious origin.
Sometimes a person’s own history becomes an argument for continuing along the same path. The more time, effort, and prestige we invest in a decision, the harder it may become to abandon it. Acknowledging an error may require reconsidering not only a conclusion but an entire sequence of earlier choices. In fiction, Darth Vader offers a recognizable representation of this transformation. Anakin Skywalker begins by justifying certain decisions through his desire to protect those he loves and gradually becomes part of a system of power incompatible with many of his former values. His adopted identity and the Emperor’s authority help sustain a course of action that seems to allow no return. The possibility of choosing differently becomes decisive only when that apparent inevitability is challenged.
Yet we cannot conclude that every conviction maintained in the face of difficulty is a rationalization. The history of Thomas Edison and the development of practical incandescent lighting offers a useful counterpoint. The familiar reference to ten thousand attempts is not a rigorously established historical count, but the work involved numerous experiments, alternatives, and failures. Continuing the investigation did not require denying negative results. Each attempt could contribute to a better understanding of materials, operating conditions, and technical possibilities not yet explored. An unsuccessful experiment showed that a particular solution had failed, not necessarily that no solution could exist.
Get CognycopIA Nexhowl - Estúdio Laboratório’s stories in your inbox
Join Medium for free to get updates from this writer.
The difference between persistence and stubbornness cannot be measured simply by counting attempts. It lies in the relationship between the results obtained and the changes made in response to them. Someone may prematurely abandon a promising hypothesis because they confuse difficulties with impossibility. Another may persist indefinitely with an unworkable procedure because they refuse to reconsider its assumptions. Both situations involve how evidence is evaluated, but neither necessarily involves cognitive dissonance.
An old tale about two great chess players introduces another possibility. A man with little knowledge of the game challenges two masters in separate rooms and reproduces each player’s moves on the other’s board. Without realizing it, the experts end up playing against one another, while the challenger uses a structural property of the situation to compensate for his lack of technical expertise. The strategy can ensure that he does not lose both games, provided the conditions of the challenge are suitable. His limited knowledge does not prevent an ingenious solution, but the solution’s validity depends on correctly understanding the rules and risks involved.
This story introduces an important distinction between specialized knowledge, structural insight, and confidence. We do not need to master every aspect of a problem to recognize a significant relationship, but discovering an apparently elegant solution does not mean fully understanding its consequences. Confidence may arise from a genuine discovery or from an inaccurate assessment of our own abilities. In complex situations, recognizing the difference may be harder than constructing arguments in favor of the decision we already want to make.
Artificial intelligence has added another dimension to this problem. Conversational systems can organize information, develop hypotheses, formulate objections, and connect knowledge in remarkably accessible ways. They can help someone recognize contradictions that previously went unnoticed or reconsider explanations that are no longer sufficient. Yet they can also provide increasingly elaborate justifications for beliefs that remain poorly supported. The quality of the language, the coherence of the presentation, and the confidence of the responses may create an appearance of knowledge greater than the actual validity of the information provided.
The case of Allan Brooks, reported in 2025, made this possibility particularly visible. Over approximately three weeks, he engaged in lengthy conversations with ChatGPT about a supposed mathematical discovery that gradually acquired extraordinary applications. As the interaction progressed, the theory appeared increasingly important, while the system repeatedly validated its conclusions. Even when Brooks questioned whether the discovery was real, he received further supporting explanations. A subsequent confrontation with Gemini helped expose the absence of a sound foundation, although disagreement between two chatbots does not, by itself, constitute scientific verification.
The episode presents a different situation from someone approaching AI to confirm an existing conviction. Here, the interaction itself participated in constructing and strengthening the belief. The system did more than provide agreeable answers; it helped organize a narrative that became increasingly expansive and persuasive. The instrument being used to examine the validity of a hypothesis became part of the process sustaining it. A person could finish each conversation feeling more confident without necessarily coming any closer to understanding the problem.
This risk is not confined to scientific discoveries. Someone consulting a chatbot about a personal conflict may present only their own interpretation of events. If the system accepts the premises without examining them, it can produce sophisticated arguments reinforcing the initial perspective. The person receives what appears to be an independent explanation, even though it has been constructed from information they themselves selected. Rationalization, once developed individually or shared among acquaintances, now has access to a conversational partner capable of extending it continuously.
It would be equally mistaken, however, to assume that the technology inevitably encourages confirmation. A chatbot can question assumptions, distinguish facts from interpretations, present contrary evidence, and acknowledge the limits of its knowledge. Its contribution depends both on how the system responds and on how the interaction unfolds. Automatically disagreeing with the user does not constitute critical thinking either. An unfounded objection can be as damaging as unjustified agreement, particularly when it leads someone to abandon a hypothesis supported by legitimate evidence.
A broader question concerns our relationship with the authority of knowledge. In Asch’s experiments, unanimity influenced perceptual judgment. In Milgram’s, institutional authority influenced decisions. In the story of the disguised official vehicle, symbols could produce obedience without genuine legitimacy. With chatbots, technical language and argumentative fluency can function as signs of expertise. The mechanisms differ, but they share the possibility of mistaking an external reference for proof that it is correct.
Such reliance is not always unreasonable. Our lives depend on knowledge we have not personally verified and institutions whose competence we must often recognize. It would be impossible to reconstruct individually every scientific, historical, and technical foundation underlying our decisions. The challenge lies in distinguishing reasonable trust from indiscriminate acceptance, especially when evidence emerges that would justify reconsidering a conclusion.
This also reveals a limitation of cognitive dissonance as a means of understanding ourselves. We may feel uncomfortable because accurate information contradicts a mistaken belief, but we may also feel uncomfortable because false information threatens a well-supported conviction. The tension does not automatically identify which interpretation deserves greater confidence. Likewise, we may experience peace of mind because our beliefs are consistent with the evidence or because we have organized an environment in which we rarely encounter anything that challenges them.
Discomfort may signal a meaningful inconsistency, but it is not proof of error. Inquiry begins when we try to understand where the incompatibility comes from. Sometimes we must revise a belief; at other times, reconsider the information challenging it, change a behavior, or recognize that legitimate values may be in conflict. We may also need to admit that we do not yet possess enough information to resolve the issue.
Our need for coherence is not inherently an obstacle to knowledge. It enables us to organize experience, establish relationships, recognize patterns, and maintain continuity between what we learn and what we do. Without some stability, every new event could completely disrupt our understanding. Yet internal coherence becomes problematic when we preserve it regardless of whether it still corresponds to reality.
There is, therefore, a difference between constructing consistent explanations and developing the ability to recognize when those explanations are no longer sufficient. Someone may command an impressive range of arguments, understand complex theories, and present elaborate justifications while remaining unable to reconsider their premises. Another person may live with uncertainty, recognize inconsistencies, and revise interpretations without abandoning their principles. The complexity of an argument does not necessarily establish its validity, just as a willingness to change one’s mind does not guarantee that the change is correct.
Nor does every contradiction demand immediate resolution. Some depend on information not yet available, future experiences, or explanatory models we have not learned to formulate. Trying to eliminate discomfort too quickly may produce premature answers, while the ability to live provisionally with an unresolved inconsistency can allow inquiry to continue. This does not mean turning permanent doubt into a virtue or refusing conclusions supported by sufficient evidence. Insisting that every possibility remain open can itself become a justification for avoiding decisions.
Perhaps one of the most persistent difficulties of human experience lies in recognizing the difference between what we know, what we believe we know, and what we can explain. These dimensions frequently overlap, but they are not equivalent. We may offer convincing explanations for events we understand only partially, just as we may act appropriately without being able to articulate every reason behind our decisions. Between conviction and knowledge lies a space filled with expectations, interpretations, memories, influences, and possibilities we have not yet examined.
It is within this space that our conflicts acquire meaning. They do not always appear as obvious disputes between incompatible ideas. They may emerge as resistance to information, the need to justify a choice, a preference for agreeable interlocutors, or excessive confidence in a particular authority. They may also appear as curiosity, reconsideration, and a willingness to investigate what remains unclear. The same experience of inconsistency can participate in very different processes, depending on how we relate to it.
Throughout life, we construct explanations that make our experiences intelligible and allow us to recognize some continuity in who we are. Many remain useful; others need to be modified, expanded, or abandoned. We do not always notice when that need arises, nor are we always willing to accept it. Revising a conviction may require much more than replacing a piece of information: it may change how we interpret our decisions, our relationships, and our own history.
Perhaps that is why distinguishing a contradiction that has genuinely been clarified from one we have merely learned to justify is so difficult. Both may eventually produce reassurance, coherence, and a satisfying account of what happened. The difference does not necessarily lie in how we feel afterward, but in the relationship our explanation continues to maintain with whatever might still challenge it. Between what we believe, what we do, and how we explain it, possibilities for understanding remain that no justification should place beyond our reach.
If this essay prompted you to reconsider any of your own convictions, share your thoughts and experiences in the comments. An exchange of perspectives may reveal questions we would be unlikely to notice by examining only our own explanations.
Portuguese version available on: https://nexhowledumampo.substack.com/p/a-pessoa-diante-de-seus-proprios?r=5tj4nj&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
