The Unmeasurable: What Values Cannot Be Probabilities and Why It Matters

Published

what values cannot be probabilities

Probability is the language of modern decision-making—from stock markets to medical diagnostics, algorithms now predict outcomes with surgical precision. Yet beneath this veneer of calculable certainty lies a quiet rebellion: certain values cannot be reduced to probabilities. They resist modeling, defy statistical regression, and persist as irreducible truths. These are the values that anchor human dignity, artistic integrity, and moral clarity in a world increasingly obsessed with predictability.

The tension arises when we ask: Can justice be assigned a confidence interval? Or love quantified as a beta distribution? Probability thrives on repeatable events, but human experience is rife with singularities—moments where meaning transcends data. The question isn’t just academic; it’s a crisis of modern epistemology. Governments use predictive policing to forecast crime, yet the value of an innocent life cannot be weighed against a risk score. Artists reject algorithms that "optimize" creativity, insisting genius emerges from chaos, not correlation.

What values cannot be probabilities? The answer lies in the fractures between logic and lived experience—where ethics outpaces utility, where beauty defies utility functions, and where the human spirit refuses to be a variable in someone else’s equation.

what values cannot be probabilities

The Complete Overview of What Values Cannot Be Probabilities

The phrase what values cannot be probabilities cuts to the heart of a philosophical and practical paradox: probability is a tool of efficiency, but not all worth is measurable. At its core, probability assumes a universe of repeatable events—where outcomes can be sampled, tested, and averaged. Yet human values often reside in the unrepeatable: the first time a child laughs, the irreversible act of forgiveness, or the quiet courage of standing alone against a crowd. These moments are not data points; they are events that demand moral or aesthetic judgment, not statistical inference.

The problem deepens when institutions—from courts to corporations—attempt to quantify the unquantifiable. A risk assessment might calculate the probability of recidivism, but it cannot assign a number to redemption. Similarly, an algorithm might predict which employees will quit, but it cannot measure loyalty. The gap between what can be predicted and what should matter is where the limits of probability become ethical boundaries. This is not a rejection of data; it’s a recognition that some values exist outside the framework of predictive modeling.

Historical Background and Evolution

The idea that certain values resist probability has roots in ancient skepticism toward quantification. The Stoics warned against reducing virtue to arithmetic, and Kant later argued that moral laws (like the categorical imperative) are a priori—universal and unconditional, not contingent on empirical probability. Fast-forward to the 20th century, and philosophers like Ludwig Wittgenstein explored how language itself imposes limits on what can be meaningfully discussed. His linguistic framework suggests that some concepts (e.g., "the sublime," "the sacred") are showable, not sayable—let alone calculable.

The rise of behavioral economics in the 1970s further exposed the flaws in treating humans as probabilistic agents. Daniel Kahneman’s System 1 and System 2 thinking revealed that intuitive judgments (e.g., trust, fairness) often override rational calculations. Probability assumes actors are utility maximizers, but real people act on values—some of which are inherently non-quantifiable. For example, the value of a "good death" in palliative care cannot be derived from survival statistics; it’s a qualitative judgment about dignity. This tension has only sharpened with AI, where machine learning models now "explain" human behavior while ignoring the values that drive it.

Core Mechanisms: How It Works

Probability functions by abstracting complexity into distributions—turning chaos into patterns. But this abstraction fails when the subject matter is essentially singular. Consider three domains where probability breaks down:

Table of Contents

1. Ethics: Utilitarianism might assign a probability to "the greatest good," but deontological ethics (e.g., "never lie") is a rule, not a calculation. The value of integrity cannot be averaged across scenarios.
2. Art: A painting’s "value" might correlate with auction prices, but its worth to an individual is subjective and irreversible. Probability can’t capture the moment a viewer’s breath catches at a brushstroke.
3. Justice: A court might calculate the probability of a defendant’s guilt, but the value of rehabilitation—or the harm of wrongful conviction—is not a number.

The mechanism is simple: probability requires reducibility. If an event or value cannot be decomposed into independent variables, it cannot be modeled. This is why existential questions ("Is life meaningful?") and aesthetic judgments ("Is this symphony beautiful?") remain beyond the reach of algorithms. They are first-person experiences, not third-person data.

Key Benefits and Crucial Impact

Understanding what values cannot be probabilities isn’t just an academic exercise—it’s a safeguard against dehumanization. When institutions treat people as variables in a regression, they risk eroding the very qualities that make humanity distinct: empathy, creativity, and moral agency. The benefits of recognizing these limits are profound: it preserves the space for judgment, artistry, and ethical autonomy in an increasingly algorithmic world.

Consider the alternative: a society where even grief is quantified ("The probability of mourning duration is 3.2 years ± 6 months"). Such a world would lose its capacity for spontaneity, reverence, and the kind of courage that defies odds. Probability is a tool, not a philosophy. Its power lies in its precision, but its weakness is its blindness to what matters most.

"Probability is to the mind what a map is to a territory: useful for navigation, but it cannot capture the essence of the land itself."
— John Maynard Keynes, A Treatise on Probability

Major Advantages

Recognizing the boundaries of probability yields critical advantages:
  • Ethical Clarity: Probability can’t resolve moral dilemmas (e.g., "Is it ethical to sacrifice one life to save five?"), but acknowledging this forces us to engage with values directly.
  • Artistic Integrity: Algorithms may generate "optimal" art, but they cannot create meaning—a distinction that matters when culture’s purpose is to evoke, not predict.
  • Legal Safeguards: Courts that rely solely on risk assessments (e.g., predictive policing) risk violating principles like innocent until proven guilty—a value that transcends statistical likelihood.
  • Personal Autonomy: If life choices were reduced to probability-weighted options, free will would evaporate. The ability to act on values (not just expected outcomes) is what makes us human.
  • Existential Resilience: Probability offers no answer to questions like "What is the point of suffering?" or "How do we find beauty in chaos?" These are values that demand philosophy, not data.

what values cannot be probabilities - Ilustrasi 2

Comparative Analysis

Domain Probability’s Role vs. Non-Quantifiable Values
Medicine Probability: Predicts disease risk (e.g., "10% chance of heart attack").
Non-Quantifiable: The value of a patient’s trust in their doctor—built on empathy, not actuarial tables.
Education Probability: Standardized test scores predict college success.
Non-Quantifiable: The value of curiosity, which cannot be graded.
Law Probability: Jury decisions based on statistical evidence.
Non-Quantifiable: The principle that punishment should fit the crime’s moral weight, not its predicted recidivism rate.
Technology Probability: Algorithms optimize user engagement.
Non-Quantifiable: The value of digital minimalism—choosing less convenience for more presence.
The tension between probability and non-quantifiable values will only intensify as AI and big data encroach on more aspects of life. One trend is the rise of post-quantitative ethics—movements that explicitly reject probabilistic morality in favor of rule-based or virtue ethics. For example, some legal systems are experimenting with "algorithmic impact assessments" that flag decisions where probability might override justice.

Another frontier is experiential AI—systems designed to augment human judgment rather than replace it. Imagine an AI that doesn’t just predict a patient’s recovery time but also flags moments where human connection (a non-quantifiable value) could alter the outcome. Similarly, in art, generative models might soon include "ethical guardrails" to prevent the erosion of creative intent.

Yet the greatest challenge lies in education. Future generations must learn to navigate a world where data is abundant but wisdom is scarce. This means teaching not just how to calculate probabilities, but how to recognize when a value cannot be one—and why that matters.

what values cannot be probabilities - Ilustrasi 3

Conclusion

The question what values cannot be probabilities is not a rejection of science or logic. It’s a reminder that some things are sacred in the original sense of the word: set apart, not for their utility, but for their intrinsic worth. Probability will continue to shape our tools, our markets, and even our laws, but it cannot—and should not—dictate what we hold dear.

The alternative is a world where love is measured in engagement metrics, courage is a risk score, and beauty is an optimization problem. That world would be efficient, but it would be hollow. The values that cannot be probabilities are the ones that keep us human.

Comprehensive FAQs

Q: Can emotions be reduced to probabilities?

A: Emotions are often expressed in probabilistic terms (e.g., "70% chance of happiness after this event"), but their essence—the raw experience of joy, grief, or awe—resists quantification. Probability can describe patterns of emotional responses, but it cannot capture the uniqueness of an individual’s feeling.

Q: How does this apply to religious or spiritual values?

A: Spiritual values (e.g., faith, enlightenment) are fundamentally non-quantifiable because they involve transcendent experiences—moments of connection to something beyond the measurable. Probability can’t assign a number to grace, divine love, or the sense of the sacred, because these are first-person truths, not third-person data.

Q: Are there fields where non-quantifiable values are already protected?

A: Yes. In law, principles like habeas corpus (protection against unlawful detention) and due process are non-quantifiable—they exist as moral absolutes, not statistical outcomes. Similarly, in art, copyright law protects creative expression not because it’s profitable, but because it’s intrinsically valuable.

Q: Can AI ever respect non-quantifiable values?

A: Current AI cannot, because it operates on probabilistic or rule-based logic. However, future systems might incorporate value-aligned constraints—for example, an AI that refuses to optimize for engagement if it detects manipulation of human trust (a non-quantifiable value). This would require a shift from pure utility maximization to ethical co-design.

Q: What’s the risk of ignoring these limits?

A: The risk is a society that confuses predictability with worth. When we treat people as variables, we lose sight of their dignity. When we reduce art to algorithms, we lose its soul. And when we quantify justice, we risk making the system more about efficiency than fairness. The limits of probability are not bugs—they’re features of what makes us human.

Leave a Comment

Comments are moderated before appearing. The data you submit is processed according to the Privacy Policy of Neip.