Hard Problem of Sociology

The core tension of agency vs. structure. Sociology seeks to explain human behavior through social structures (class, institutions, norms). But individuals have free will, make choices, and can change structures. So which is the real driver? The hard problem is that any explanation emphasizing one makes the other a mere illusion. If structure determines everything, we're puppets. If agency is paramount, society is just a backdrop and sociology is pointless. The field is stuck trying to describe a dance where it can't tell if the dancers are leading the music or the music is forcing the steps.

Example: Why did you go to college? Agency explanation: You chose to, for your future. Structural explanation: You're a middle-class person in a society where that's the normative, expected path, heavily influenced by family, school counselors, and economic necessity. The hard problem: Both are true simultaneously in a messy way. Sociology can describe the pattern (most middle-class kids go to college), but it struggles to explain any single person's decision without reducing them to a statistic or pretending structures don't push them. It's the science of the forest that keeps getting distracted by unique trees. Hard Problem of Sociology.

by Nammugal January 24, 2026

Hard Problem of the Social Sciences

The prediction problem. Unlike in physics, where you can isolate variables and predict an eclipse to the second, social sciences (economics, political science, sociology) deal with complex, reflexive systems. Humans react to predictions, changing the outcome (the "Lucas Critique"). The hard problem is: Can you have a real science of human society if its core subjects alter their behavior upon hearing your findings? True scientific laws are supposed to be invariant. Social "laws" are more like trends that expire once people know about them, making the field perpetually one step behind a moving target.

Example: An economist develops a perfect model predicting stock market crashes. Once published, investors see it and adjust their behavior to avoid the predicted conditions, thereby preventing the very crash the model forecasted. The model is now wrong. The hard problem: The act of studying the system changes it. This makes falsification—the bedrock of science—incredibly tricky. Social science thus often ends up explaining the past very well (postdiction) but failing at predicting the future, which is what we usually want from a science. Hard Problem of the Social Sciences.

by Nammugal January 24, 2026

Hard Problem of Past Lives & Afterlife

The unsolvable chain of evidence. For a past life memory to be verified, you'd need a documented fact from a deceased person's life that the current person could not possibly know through normal means, and you'd have to rule out fraud, cryptomnesia (hidden memory), and genetic or collective unconscious transmission. For the afterlife, you'd need a verifiable, two-way communication with a specific, identifiable deceased consciousness. The hard problem is that any piece of evidence (e.g., a child knowing a dead person's secret) can be explained by lesser hypotheses (telepathy between living minds, chance, subconscious inference). The signal can never be isolated from the noise of unknown psychic phenomena or pure coincidence.

*Example: A child recalls being a pilot named James who died in a WWII crash, giving specific coordinates. Investigators find wreckage there of a plane piloted by a James. The hard problem: This is astonishing, but is it proof of reincarnation? Alternative explanations include: 1) The child psychically tapped into the collective memory/historical record of the event (clairvoyance, not past life). 2) Extreme coincidence plus confabulation. To prove a past life, you must first disprove all forms of present-life psychic ability, which is itself unproven. The conclusion is always one unproven assumption stacked on another.* Hard Problem of Past Lives & Afterlife.

by Nammugal January 24, 2026

Hard Problem of the Demarcation Problem

The meta-problem: There is no agreed-upon rule to distinguish science from non-science, and the very search for such a rule may be unscientific. Falsifiability (Popper) fails—string theory isn't easily falsifiable but is considered science. Astrology makes falsifiable claims but is pseudoscience. Methodological naturalism? It rules out theology but also historical sciences that reconstruct unique past events. The hard problem is that "science" is a family-resemblance concept, not a neat category. Any bright-line rule you propose either excludes legitimate sciences or lets in obvious pseudoscience, revealing that demarcation is a social and philosophical negotiation, not a logical one.

Example: Is evolutionary biology science? It reconstructs unique past events (unfalsifiable in a strict lab sense). Yet it's a core science. Is phrenology pseudoscience? It used measurement and data (the "scientific method" of its day). The hard problem: We know the difference intuitively, but can't define it without circular logic ("It's science because scientists do it"). The demarcation criteria are like trying to nail jelly to a wall—the harder you try, the messier it gets, and you're left wondering why you're nailing jelly in the first place. Hard Problem of the Demarcation Problem.

by Nammugal January 24, 2026

Hard Problem of the Demarcation Problem

The prediction problem. Unlike in physics, where you can isolate variables and predict an eclipse to the second, social sciences (economics, political science, sociology) deal with complex, reflexive systems. Humans react to predictions, changing the outcome (the "Lucas Critique"). The hard problem is: Can you have a real science of human society if its core subjects alter their behavior upon hearing your findings? True scientific laws are supposed to be invariant. Social "laws" are more like trends that expire once people know about them, making the field perpetually one step behind a moving target.

Example: An economist develops a perfect model predicting stock market crashes. Once published, investors see it and adjust their behavior to avoid the predicted conditions, thereby preventing the very crash the model forecasted. The model is now wrong. The hard problem: The act of studying the system changes it. This makes falsification—the bedrock of science—incredibly tricky. Social science thus often ends up explaining the past very well (postdiction) but failing at predicting the future, which is what we usually want from a science. Hard Problem of the Demarcation Problem.

by Nammugal January 24, 2026

Hard Problem of Science & Pseudoscience

The problem of motivation, not method. Both can use data, jargon, and peer review (see creation "science"). The core difference might be the attitude toward evidence: science seeks to test and potentially disprove its ideas; pseudoscience seeks to defend a preordained conclusion. The hard problem is that this is a psychological distinction about the practitioners, not a methodological one. You can't look at a paper and always tell. A bad scientist (cherry-picking data) is using pseudoscientific tactics, while a clever pseudoscientist can mimic the form of science perfectly. The line is blurred because it's about internal intent, which is invisible.

Example: Flat Earthers run experiments (lasers over water) they claim prove no curvature. Scientists point out flawed methodology. The Flat Earthers dismiss it as part of the conspiracy. The hard problem: Their process looks scientific—hypothesis, test, observation. The breakdown is their refusal to accept counter-evidence as valid. But who decides what "valid" counter-evidence is? The scientific community. So, in practice, science is defined by social consensus of what counts as proper evidence, not by a pure, objective rulebook. Pseudoscience is simply what that consensus excludes. Hard Problem of Science & Pseudoscience.

by Nammugal January 24, 2026

Hard Problem of Epistemology

The problem of the criterion. To know which things we know (a theory of knowledge), we need a reliable method. But to justify that method, we need to know it leads to truth. This is a vicious circle: we need a method to identify knowledge, but we need knowledge to validate the method. Every foundational theory (empiricism, rationalism) starts with an unproven assumption. The hard problem is that epistemology, the study of knowledge, cannot get started without presupposing the very thing it seeks to justify. We are like a person searching for their glasses while needing their glasses to see.

Example: "I know the sun will rise tomorrow based on induction (past experience)." The epistemologist asks: "How do you know induction is reliable?" You might say, "It's always worked before." But that's using induction to justify induction—circular reasoning. Any other justification (e.g., it's logically necessary) would require its own justification. The hard problem: We clearly have functional knowledge, but we cannot construct a watertight, non-circular, non-arbitrary account of how we have it. Epistemology either ends in infinite regress, circularity, or an arbitrary stopping point ("just trust your senses, bro"). Hard Problem of Epistemology.

by Nammugal January 24, 2026