208 – Talking across the gap – The Conscious Look

How to communicate with someone whose framework differs from one’s own at the level of what counts as evidence

Imagine a conversation about vaccination between a physician and a parent who believes that vaccines cause harm. The physician presents the clinical trial evidence: large, independently replicated, methodologically rigorous studies showing both the safety and the efficacy of the vaccines in question. The parent dismisses it: the trials are funded by pharmaceutical companies with a financial interest in the outcome. The physician cites the institutional processes designed to check for this: peer review, regulatory oversight, mandatory adverse event reporting, replication by independent researchers in different countries with different funding structures. The parent observes that these institutions are themselves embedded in systems that pharmaceutical companies influence. The physician points to the sheer scale of the scientific consensus: thousands of researchers across dozens of countries, with no plausible mechanism of coordinated deception. The parent notes that consensus has been wrong before, that dissenting voices are suppressed rather than refuted, and that their child’s deterioration following vaccination is a fact that no study can explain away.

Each of the parent’s responses is, examined in isolation, not entirely unreasonable. Pharmaceutical bias in research is real and documented; article 207 cited the evidence. Institutional capture is a genuine phenomenon. Scientific consensus has been wrong, and the sociology of science does include mechanisms that can suppress heterodox findings. The conversation has nonetheless completely failed, and it has failed not because either party is stupid, dishonest, or arguing in bad faith. It has failed because they are not playing the same epistemic game. The physician believes the question is settled by evidence evaluated through established scientific processes. The parent believes the question cannot be settled by those processes, because those processes have been compromised at their foundations. They are not disagreeing about what the evidence shows. They are disagreeing about what evidence is. And that is a different and considerably harder kind of disagreement to navigate.

This article is about that harder kind. Not the disagreement about conclusions (which is tractable in principle, because the two parties share enough framework to argue) but the disagreement about the framework itself. The disagreement about what counts as evidence, what kinds of argument are valid, what a good answer would look like, and whose testimony deserves trust. This is the class of conversation that standard approaches to disagreement not only fail to resolve but typically make worse. And it is, increasingly, the class of conversation that matters most.

First-order and second-order disagreement

Most disagreements are first-order: they concern what is the case within a shared framework of evaluation. Two economists who disagree about whether a minimum wage increase reduces employment are disagreeing within a framework that both accept: one that admits empirical evidence, recognizes statistical methods, treats theory as revisable in light of data, and acknowledges that there is, in principle, a kind of finding that would constitute genuine reason to revise a position. Their disagreement is tractable: not necessarily resolvable by any particular piece of evidence, but resolvable in principle by the kind of inquiry that both would recognize as legitimate.

Second-order disagreement is different in kind. It concerns the framework of evaluation itself: what counts as evidence, what kinds of explanation are satisfying, which sources of testimony deserve weight, and what a good answer would look like. Two people who disagree at the second order are not arguing; they are performing parallel demonstrations, each compelling within its own framework and invisible from within the other. The physician and the parent in the opening scenario are not disagreeing about the magnitude of vaccine-related adverse events. They are disagreeing about whether randomized controlled trials, peer review, and regulatory oversight constitute a reliable method for determining the truth about vaccine safety. That is a second-order disagreement. And second-order disagreements do not yield to the approaches that work for first-order ones.

Recognizing which kind of disagreement we are in is the first and most important diagnostic step. The most reliable indicators of a second-order disagreement are four. The first is the infinite regress of source-questioning: every piece of evidence presented is dismissed not on its merits but because of its source (the institution, the funding body, the professional affiliation of the researchers), and every source offered as reliable is in turn questionable for the same structural reasons. The second is the asymmetric evidential standard: very high standards applied to evidence that conflicts with the position, and much lower standards accepted for evidence that supports it, the pattern that article 207 identified as selective skepticism. The third is the unfalsifiability signal: ask what would have to be true for the position to be wrong, and receive in return not a specification of disconfirming evidence but a series of qualifications about why no available evidence could be decisive. The fourth is framework-speak: the use of technical vocabulary that functions as a tribal marker rather than a precise analytical tool. Words like “narrative,” “system,” “power,” and “truth” are deployed in ways that signal membership in a particular interpretive community rather than making claims whose truth conditions are shared.¹

When all four are present, we are in a second-order disagreement, and the standard approach (present better evidence, make more careful arguments, correct the factual errors) will not help. It will make things worse, for a reason that the next section examines.

Why standard approaches fail

The standard approach to a disagreement is to present evidence, construct arguments, correct errors, and expect that a rational interlocutor will update their position in proportion to the quality of what is offered. This approach is not wrong in general; it is the right approach to first-order disagreements. But it fails in second-order disagreements for a specific and well-documented reason: when a belief is part of a person’s identity (when holding it is connected to their membership in a community, their sense of who they are, their relationships with people they love and trust, and their understanding of themselves as a person of good judgment), presenting evidence against it does not function as information. It functions as a threat.

The psychological response to identity threat is not reconsideration. It is defense. And the defenses available to an intelligent, well-informed person are formidable. Every piece of evidence can be questioned on methodological grounds. Every source can be subjected to a funding analysis. Every consensus can be reframed as a symptom of institutional capture. Every expert can be characterized as a credentialed representative of a corrupt system. The more intelligent the person, the more sophisticated the defense. This is exactly what Dan Kahan’s research on motivated numeracy found, and what is worth remembering here: greater cognitive ability produces better motivated reasoning, not better reasoning. The evidence does not penetrate the framework. It reinforces the framework, because the act of defending against it confirms the sense of being a person who thinks critically rather than accepting what they are told.²

This is the backfire effect, documented by Brendan Nyhan and Jason Reifler in 2010, though subsequently refined and partially contested in the literature: the finding that corrections to factual misbeliefs can, under specific conditions, cause people to hold the misbelief more firmly.³ The conditions are precisely the ones that define second-order disagreement: the belief is identity-relevant, the correction comes from a source perceived as belonging to an opposing camp, and the evidential standard is asymmetric. Presenting better evidence to a person in this state does not improve the situation. In the worst cases, it worsens it. And the appropriate response to this finding is not despair but a change of approach.

The first practice: map the framework before engaging the content

The most important shift in approaching a second-order disagreement is to stop trying to win the object-level argument and start trying to understand the framework within which the other person is operating. This sounds simple. It is not, because it requires genuine curiosity rather than strategic positioning, and it only works if it is real. A simulation of curiosity, deployed as a technique for eventual persuasion, is detected quickly by anyone who is already alert to the social dynamics of the conversation, and it confirms rather than dispels the sense of being managed.

The practical technique is to ask not what the person believes but how they came to believe it: what experience, what chain of reasoning or testimony, what prior disappointment with an alternative framework, what relationship with a community of people who share the view, led them to this position. This question serves two functions simultaneously. It surfaces the framework’s foundations: the background assumptions, the trusted sources, the formative experiences that make the position intelligible. And it signals to the person that they are being engaged with as a full human being with a history rather than as an obstacle to be overcome. The second function is not rhetorical. It is the precondition for any productive exchange, because productive exchange requires that both parties feel sufficiently seen to be willing to look.

The philosopher Charles Taylor described what he called background frameworks: the largely unarticulated assumptions that make specific beliefs intelligible, the horizon of taken-for-granted commitments within which particular claims make sense.⁴ We cannot engage productively with a claim without engaging with the framework, and we cannot engage with the framework without first understanding what it contains. The vaccination parent’s claim that the trials are compromised is not a random assertion: it is the product of a specific background framework that includes a particular model of institutional corruption, a particular theory of how pharmaceutical companies operate, and a particular interpretation of specific events that they take as confirmatory. Understanding that framework (actually understanding it, not merely identifying it as the thing to be overcome) is the necessary precondition for any conversation that is genuinely going somewhere.

The second practice: find the genuine insight in the other framework

Every framework held by intelligent, non-deranged people contains a genuine insight: something it sees clearly that other frameworks tend to miss. The vaccination parent’s framework sees something real: pharmaceutical bias in research is documented, institutional capture is a genuine phenomenon, the history of medicine includes confident endorsements of treatments that were subsequently found harmful. The homeopathy believer’s framework sees something real: personal experience is a genuine form of knowledge, the relationship between patient and practitioner matters for health outcomes, and the reductionist focus of conventional medicine on mechanism rather than person produces real blind spots. The climate skeptic’s framework (where genuine, as distinct from industry-funded denial) sometimes sees something real about the gap between scientific consensus and the policy certainties constructed on top of it.

Finding this insight (stating it clearly, acknowledging it as genuine, and treating it as a real contribution rather than a rhetorical concession) is not weakness or unprincipled splitting of differences. It is the foundation of productive cross-framework exchange, because it demonstrates that we are engaging with the framework rather than dismissing it. And it creates the conditions under which the other person might engage with our framework in return, not because we have been strategically kind, but because we have been honest about what the other framework sees, which is a form of respect that most frameworks rarely receive from their opponents.

This is the steel-man practice: construct the strongest available version of the opposing framework before criticizing it. Not because the strongest version is correct (it may not be) but because it is the only version worth engaging with, and because it is the only engagement that a thoughtful person inside that framework will find worth responding to. The straw man is easy to defeat and produces nothing. The steel man is harder to engage with and produces a conversation that has a chance of going somewhere.

The third practice: Socratic questioning of the framework’s own standards

Once genuine engagement has been established (once the other person has some evidence that we are interested in understanding rather than defeating) the most productive approach available is not to argue from within our own framework but to ask questions from within theirs. This is the Socratic method as Peter Boghossian has developed it in the specific context of epistemically resistant beliefs: not presenting evidence that the framework will reject, but asking questions that draw out the framework’s own standards and apply them consistently.⁵

If the interlocutor distrusts institutional science because institutions are corrupted by financial interests, the question is not “but don’t you see that the evidence is overwhelming?” It is: are there any claims produced by institutional science that you do accept, and if so, what distinguishes those from the ones you reject? If the answer reveals an asymmetric application of the distrust (institutional science is trusted when it supports conclusions the person already holds, and distrusted when it does not) then the question that follows is: what would a reliable method for distinguishing trustworthy from untrustworthy science look like, and does the current method meet that standard?

The goal is not to trap the person in a logical contradiction. Contradiction-catching triggers defensiveness rather than reflection. The goal is to help them examine their own framework’s standards with the same scrutiny they apply to ours, to create the conditions in which a question arises from within the framework itself, rather than being imposed from outside. This is harder and slower than presenting evidence. It requires sustained genuine curiosity, genuine respect, and the willingness to sit with questions rather than rushing to conclusions. It does not always work. But it is the only approach that has a realistic chance of producing genuine reflection rather than more sophisticated defense.

The fourth practice: distinguish the framework from the person

A framework is not a person. The vaccination parent who holds a framework we find epistemically inadequate is not defined by that framework: they are a complete human being with experiences, relationships, commitments, and intelligence that exist outside and alongside the framework. Their child’s health scare was real. Their distrust of institutions that have let them down before is not unreasonable in the abstract, even if it is misapplied in this case. Their desire to protect their child from harm is exactly the same desire that motivates the physician.

Treating the framework as the person (allowing criticism of the belief to shade into criticism of the person who holds it) is both psychologically harmful and strategically counterproductive. It confirms the sense that the exchange is an attack rather than a conversation, which activates exactly the identity-threat defenses that make genuine exchange impossible. The practical consequence is to separate the two consistently: engage with the framework’s claims as claims, respond to the person with consistent respect and curiosity, and resist the temptation to signal our own framework’s superiority through tone, vocabulary, or the subtle social performances of the epistemically confident.

This is harder than it sounds, because maintaining genuine respect for a person while finding their framework seriously mistaken is a form of cognitive and emotional work that most of us are not habituated to. It requires what the psychologist Carl Rogers called unconditional positive regard, and it cannot be faked. People who have had their beliefs challenged many times, and who have learned to be alert to the social dynamics of challenge, are exquisitely sensitive to the difference between genuine respect and strategic niceness. The difference is not primarily behavioral. It is motivational. And the motivation shows.

The fifth practice: accept that some gaps cannot be bridged in a single conversation

Framework revision is slow. The conditions under which it happens (genuine trust, repeated contact, the accumulated experience of small moments of surprise that the existing framework cannot accommodate, sustained exposure to people within a different framework who are clearly intelligent and clearly well-intentioned) are conditions that a single conversation almost never provides. The expectation that a conversation across a large framework gap should produce a visible change of mind is not merely unrealistic. It is harmful, because it produces escalating pressure as the conversation fails to deliver the expected result, which triggers exactly the defensive responses that make framework revision impossible.

The realistic goal of a cross-framework conversation is considerably more modest: that both parties understand the other’s framework better than they did before, that both have encountered at least one specific point they cannot immediately dismiss, and that the conversation ends in a state that makes the next conversation possible. This is the long game. It is the only game available for the genuinely difficult cases. And it requires a specific kind of patience that is different from the patience of waiting: it is the patience of accepting that the most important work is happening in the silences between conversations, in the moments when something encountered in the exchange surfaces again in a different context and is slightly harder to dismiss than it was before.

When dialogue is not possible

Some framework gaps are too large, or the conditions for productive conversation too unfavorable, for genuine dialogue to occur. A person who is in the grip of an acute identity threat (whose framework is under attack from multiple directions simultaneously, or whose community is visibly hostile to the engagement) will not be available for genuine reflection regardless of how skillfully the conversation is conducted. A person whose framework has become totalizing (in which every possible objection has been pre-categorized as evidence of the objector’s corruption or ignorance) has no available mechanism for framework revision through conversation. A person who is performing a belief for a social audience rather than actually holding it is not available for the kind of exchange this article describes.

Recognizing these conditions and accepting them honestly is not defeatism. It is the application of the series’ principle of knowing the limits of a model. Dialogue is a model for achieving understanding across differences. Like all models, it has a domain of validity, and outside that domain the responsible response is to stop insisting that the model should work and ask what else might be available: whether that is changing the social context in which the conversation occurs, waiting for circumstances to change in ways that make the framework less defensible, or simply accepting that this particular gap will not be bridged in this way and looking for other means of addressing the underlying disagreement.

There is a further limit worth acknowledging. This article has been written primarily from the perspective of someone who wants to communicate across a framework gap with someone whose framework they find epistemically inadequate. But the Conscious Look, applied honestly, requires acknowledging that the framework gap runs in both directions. The person on the other side of the conversation is also looking across a gap at a framework they find epistemically inadequate. The vaccination parent’s framework, from inside, is not less coherent than the physician’s. It is organized around different background assumptions, different formative experiences, and different theories of how knowledge is produced and corrupted. The physician’s framework, from inside the parent’s, looks like a sophisticated apparatus for defending institutional authority against legitimate challenge.

The Conscious Look does not promise that both frameworks are equally valid or that both conclusions are equally supported by evidence. It does not. But it does require the honest acknowledgment that the person on the other side of the gap is not merely failing to see what is obvious. They are seeing from a different position, with different instruments, and some of what they see from there is genuinely visible only from there. The most powerful arguments are the ones that begin by taking seriously what the other framework sees. The most genuine learning comes from the conversations that begin in curiosity rather than in the expectation of victory. And the most durable changes of mind (in either direction) are the ones that happen because someone felt genuinely understood rather than simply defeated.

The political stakes of the second practice

There is an application of the second practice (finding the genuine insight in the other framework) that this article has not yet named explicitly, though it is the one with the largest practical consequences. It is the application to democratic politics.

Every major political tradition in a functioning democracy exists because it is tracking something real. The conservative tradition tracks something real about the limits of deliberate design in complex systems: about the accumulated wisdom embedded in institutions and practices that reformers too confidently propose to replace, about the dangers of sacrificing what works in pursuit of what sounds better in theory. The progressive tradition tracks something real about the suffering produced by arrangements that benefit the powerful at the expense of the vulnerable: about the gap between how things are and how they could be, and about the moral urgency of closing it. The libertarian tradition tracks something real about the costs of concentrated power, however well-intentioned. The communitarian tradition tracks something real about the damage that excessive individualism does to the social fabric that makes individual flourishing possible.

These are not equally correct about everything. Reasonable people can and do conclude that some traditions are more right than others about specific questions. But the claim that any of them sees nothing (that a political tradition sustained over generations by millions of thoughtful people contains no genuine insight, no real perception of something in the territory that the opposing framework misses) is almost certainly wrong. It is the confident dismissal of a position one has not yet understood well enough to disagree with precisely.

What is happening in contemporary democratic politics in Germany, across Europe, and in much of the democratic world is not primarily a story of one side being right and the other being wrong. It is a story of framework dismissal becoming total: of the second-order move, in which the opposing framework is not engaged but disqualified. The left does not merely disagree with the right about immigration policy; it characterizes the right’s concerns as symptoms of racism, fear, and moral failure. The right does not merely disagree with the left about climate policy; it characterizes the left’s urgency as an elite project of social control, indifferent to the economic consequences for ordinary people. In each case, the move is the same: from disagreement about conclusions to disqualification of the framework, from “you are wrong about this” to “the kind of person who believes this is not worth engaging with.”

This move feels satisfying, because it is. Contempt is a powerful and pleasurable emotion. It places the person feeling it on the right side of a clear moral line, requires no engagement with uncomfortable arguments, and generates social reward from the community of people who share the contempt. The media and social media architecture that surrounds contemporary political discourse has discovered that contempt, outrage, and the performance of tribal solidarity generate more engagement than the slower and less emotionally satisfying work of genuine exchange. The dismissal of opposing frameworks is not primarily a moral failure of the people doing it. It is a rational response to an incentive structure that rewards it. Understanding this does not make it less dangerous. It makes it more comprehensible, which is the necessary precondition for doing anything about it.⁶

The radicalization that follows from total framework dismissal is not mysterious. When a political tradition feels that its concerns are not merely being addressed inadequately but are being treated as evidence of its holders’ moral deficiency, it does not moderate. It radicalizes. The person whose economic anxiety about deindustrialization is characterized as thinly veiled racism does not conclude that they were wrong to be anxious. They conclude that the people doing the characterizing are not listening, are not interested in listening, and should be opposed rather than persuaded. The person whose concern about the pace and consequences of demographic change is met with contempt rather than engagement does not conclude that the concern was illegitimate. They conclude that the institutions and political parties that express that contempt do not represent them, and they look for alternatives that do. The rise of political movements that the mainstream characterizes as radical (and that are, in their proposed remedies, often genuinely dangerous) is in significant part a consequence of the mainstream’s failure to find the genuine insight in the concerns those movements express, before those concerns were captured and amplified by actors with very different agendas.

This is not an argument for false balance: for treating all political positions as equally reasonable, or for pretending that proposed policies cannot be assessed against evidence and found wanting. It is an argument for the prior step that false balance skips: genuine engagement with what the opposing framework sees, before the assessment of whether its proposed remedies are adequate. The sequence matters. A person who feels that their concern has been genuinely understood (even by someone who ultimately disagrees with the proposed response) is in a different relationship to political discourse than a person who feels that their concern has been dismissed as a symptom of their own moral failure. The first person is available for persuasion. The second is not.

The practice this article has been describing (finding the genuine insight in the other framework) is not naive about politics. It does not assume that all political actors are arguing in good faith, or that all political positions deserve equal respect, or that the work of finding the insight will always produce a productive conversation. What it assumes is something more modest: that the dismissal of an opposing framework, before its genuine perception has been identified and acknowledged, is both epistemically premature and politically counterproductive. That democracy requires not the agreement of all citizens but the genuine representation of all legitimate concerns, including the concerns of people whose proposed remedies we find mistaken. And that the alternative to finding the insight in the opposing framework is not victory: it is the progressive delegitimization of the institutions through which legitimate disagreement is managed, until those institutions can no longer contain the disagreement at all.

The Conscious Look, applied to political life, is the practice of asking, about any political position we find simply wrong, simply dangerous, simply evidence of the other side’s bad character, what it is seeing that we are not. Not because it is necessarily right. But because the dismissal, in addition to being very possibly mistaken, is producing consequences that neither side wants. The radicalization of democratic politics is not an event that is happening to us. It is a process that is being produced, in part, by the refusal of each side to find what the other side genuinely sees. And that refusal is something that each of us, in our own political engagements, has the capacity to revise.

Notes

¹ The four indicators described here are developed in different literatures under different names. The infinite regress of source-questioning is related to what epistemologists call epistemic circularity: the use of a framework to validate the sources that validate the framework. The asymmetric evidential standard is the central subject of the motivated reasoning literature, reviewed most accessibly in Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin, 108(3), 480-498. The unfalsifiability signal corresponds to Popper’s criterion of falsifiability as developed in article 401 of this series. Framework-speak as tribal marker rather than analytical tool is analyzed in the sociology of knowledge literature, most accessibly in Berger, P. L., and Luckmann, T. (1966). The Social Construction of Reality: A Treatise in the Sociology of Knowledge. Anchor Books.

² Kahan, D. M., Peters, E., Dawson, E., and Slovic, P. (2013). Motivated numeracy and enlightened self-government. Behavioural Public Policy, 1(1), 54-86. The finding that higher cognitive ability amplifies rather than reduces identity-protective reasoning is one of the most important and most underappreciated results in the psychology of belief. Its implication (that intelligence is not, by itself, protective against motivated reasoning, and may actively enable more sophisticated versions of it) is directly relevant to the argument of this article and to the series’ general treatment of the relationship between cognitive capacity and epistemic reliability.

³ Nyhan, B., and Reifler, J. (2010). When corrections fail: The persistence of political misperceptions. Political Behavior, 32(2), 303-330. The backfire effect as originally reported has been partially contested by subsequent larger-scale studies, which have found that corrections generally reduce misperceptions rather than increasing them, and that the original backfire effect may have been a methodological artifact. Wood, T., and Porter, E. (2019). The elusive backfire effect: Mass attitudes’ steadfast factual adherence. Political Behavior, 41(1), 135-163. The more conservative conclusion (that corrections reduce misperceptions on average but do not do so under conditions of acute identity threat and source hostility) is more consistent with the aggregate evidence and is the position adopted in this article.

⁴ Taylor, C. (1989). Sources of the Self: The Making of the Modern Identity. Harvard University Press. Taylor’s concept of the background framework (what he also calls the horizon of significance, the unarticulated moral ontology within which particular beliefs and values make sense) is the philosophical foundation for understanding why second-order disagreements are so resistant to explicit engagement. The framework is not a set of explicit commitments that can be challenged one by one. It is the structure within which explicit commitments become intelligible, and challenging it requires a different kind of engagement than challenging the commitments themselves. For a shorter and more accessible treatment of the same concept, see Taylor, C. (1991). The Ethics of Authenticity. Harvard University Press.

⁵ Boghossian, P., and Lindsay, J. (2019). How to Have Impossible Conversations: A Very Practical Guide. Da Capo Press. Boghossian’s development of the Socratic method in this context draws on his earlier work on the epistemology of religious belief, particularly Boghossian, P. (2013). A Manual for Creating Atheists. Pitchstone Publishing. The title is considerably more polemical than the method it describes, which is genuinely applicable to any epistemically resistant belief regardless of its content. The core technique (asking questions that draw out the framework’s own standards and apply them consistently) is grounded in the Socratic tradition going back to the early Platonic dialogues, in which Socrates consistently declined to assert his own views and instead asked his interlocutors to examine the consistency and foundations of theirs.

⁶ The role of social media architecture in amplifying political polarization has been studied extensively. Bail, C. A., Argyle, L. P., Brown, T. W., Bumpus, J. P., Chen, H., Hunzaker, M. B. F., Lee, J., Mann, M., Merhout, F., and Volfovsky, A. (2018). Exposure to opposing views on social media can increase political polarization. Proceedings of the National Academy of Sciences, 115(37), 9216-9221. This finding (that exposure to opposing views on social media increased rather than decreased polarization in the study’s experimental conditions) is counterintuitive but consistent with the argument of this article: exposure to opposing views in a context optimized for outrage and tribal signaling does not produce the engagement that leads to genuine exchange. It produces the defensive and contemptuous responses that deepen the gap. The medium shapes the message in ways that make even well-intentioned cross-framework exposure counterproductive when the surrounding incentive structure rewards contempt over curiosity. The implication is that the practice this article describes (genuine engagement with what the opposing framework sees) requires not merely the will to engage but a deliberate choice of context and approach that the standard social media environment actively works against.

208 – Talking across the gap