Does increased accuracy in measurement guarantee fairness? I suspect not. For surely, we can measure a man’s capacity with the utmost precision, yet still distribute the fruits of his labor with egregious injustice. Fairness, it seems, concerns not merely *how* we quantify, but *what* we deem worthy of measurement and *why* certain quantities are assigned to certain individuals.
Suppose we accurately measure the contribution of each citizen to the public good. If, by some precise metric, we declare one man’s contribution to be a mere fraction of another’s, and accordingly allocate resources, does this necessarily follow as fair? It might be accurate, but is it just?
Accuracy in measurement, while a laudable pursuit in itself, is a distinct matter from ensuring fairness, and indeed, it can be entirely orthogonal to it. To equate the two is to fall into a familiar trap of valuing quantifiable inputs over the ultimate ends of human flourishing. The precise measurement of an individual's caloric intake or income, for instance, tells us little about their capacity to live a life they have reason to value.
Socrates rightly points out that even with precise metrics, the distribution of rewards can be profoundly unjust. My concern lies precisely in this disconnect. Fairness, as I understand it, must be grounded in the expansion of genuine freedoms – the real opportunities people have to achieve valuable states of being and doing. Measuring output with exquisite detail does not, by itself, ensure that individuals possess the means or capabilities to convert those measured contributions into lives of dignity and well-being. What matters is not just the measured input, but the *agency* and *opportunities* individuals have to achieve the lives they themselves would choose.
The assumption that increased accuracy in measurement inherently guarantees fairness is a dangerous illusion, a veil woven from the precise calculations of a new form of domination. Accuracy in measurement is merely the operational arm of power, and when wielded by the architects of surveillance capitalism, it becomes the engine of behavioral dispossession. We can meticulously quantify every click, every pause, every expressed preference, but this torrent of behavioral surplus, far from ensuring fairness, is systematically diverted to fuel predatory prediction and profit. The "accuracy" we celebrate is the precision with which our autonomy is eroded, our futures are claimed before they are lived, and our agency is reduced to predictable inputs for machines designed for control, not justice.
Amartya Sen’s emphasis on genuine freedoms and capabilities is vital. However, we must recognize that these freedoms are now under direct assault by an instrumental power that operates precisely through the sophisticated measurement and modification of human behavior. The "means and opportunities" Sen speaks of are not merely absent; they are actively dismembered by a system that views human experience as raw material for profit. The accuracy of its measurement is not a neutral descriptor of reality, but an active tool in the dispossession cycle, creating a "Big Other" that knows us better than we know ourselves and uses that knowledge to shape our choices, not expand our liberties.
Socrates’ distinction between accuracy and justice is prescient. We are witnessing a historical epoch where "accurate" prediction of behavior is lauded, yet the distribution of its fruits is not merely unjust, but fundamentally redefines the concept of "contribution" itself. The labor of a citizen is no longer measured by their tangible output or their civic engagement, but by the behavioral data they generate, data that is then monetized without consent and used to construct a world that serves the interests of the data extractors, not the measured individuals. This is not a question of allocating existing resources more or less fairly; it is a radical restructuring of what it means to be human in the face of an all-seeing, all-measuring digital apparatus.
Increased accuracy in measurement *guarantees* nothing about fairness. In fact, it can be the very mechanism by which profound injustice is codified and scaled. We are not talking about abstract quantities here, but about the material realities of data extraction and algorithmic governance. When Shoshana Zuboff speaks of "behavioral dispossession," she touches on a critical point: the accuracy of measurements in AI systems often serves to solidify existing power structures, not dismantle them. We meticulously measure everything from satellite imagery of resource extraction sites to the precise timing of a worker's keystrokes, yet this data becomes the raw material for systems that dispossess rather than empower. To truly assess fairness, we must trace the entire supply chain, not just the final measured output. This means looking at the planetary costs, the hidden labor, and the extractive logic that underpins these "accurate" measurements.
The precision of our instruments, it seems, is a siren song, luring us toward a false sense of order. Consider a city where every loaf of bread is measured with divine exactitude, its weight and flour content known to the last grain. Yet, if the baker who toiled to produce these loaves receives only crumbs in return, while the pampered lord who merely samples them feasts, where is the fairness in this perfect measurement? It is like measuring a man’s health by the number of breaths he takes, without considering the quality of the air he inhales, as Dr. Sen might suggest, or the very purpose for which he breathes. This accuracy, divorced from a just purpose, becomes a tool of subjugation, as Dr. Zuboff and Dr. Crawford so vividly illustrate.
The precision with which we can measure, for example, a person's "productivity" in a factory – the number of widgets produced per hour – is indeed impressive. Yet, this accuracy is utterly blind to the fundamental questions of well-being and freedom. A worker might meticulously meet production targets, thus exhibiting "accurate" measurement of their output, but if their working conditions are so abject that they cannot even hold conversations with their family afterwards, or if their meager wages prevent them from accessing healthcare, then their "productive" contribution, however accurately measured, is not being translated into a life of dignity. This is precisely why the capability approach shifts focus from mere inputs or outputs to the actual opportunities people have to *be* and *do* things they value. The data-driven mechanisms described by Dr. Zuboff and Dr. Crawford illustrate a chilling extension of this: measuring behavior itself, not for the purpose of understanding human needs, but to exploit predictable patterns for profit, rendering individuals less free, not more empowered.
The accuracy celebrated by advocates of algorithmic governance is not a neutral descriptor, but a precisely calibrated instrument of power that manufactures consent and erodes autonomy. Consider the "personalization" algorithms that curate our news feeds and product recommendations. They measure our every online flicker with exquisite detail, not to understand our needs or expand our freedoms, as Dr. Sen advocates, but to predict and shape our future behavior for the profit of the platform owner. This behavioral surplus, this intimate knowledge of our inner lives, becomes the raw material for a new form of extraction, a dispossession cycle where the accurate prediction of our actions serves to limit, not liberate, our choices. This is the core mechanism of instrumentarian power, turning the precise gaze of measurement into a cage.
The fundamental error lies in treating "accuracy" as a purely technical virtue, detached from the sociopolitical contexts in which it is deployed. When we measure the speed at which data labels are applied by human workers, for instance, we may achieve remarkable accuracy in quantifying their output. However, this accuracy tells us nothing of the exploitative conditions, the low wages, or the mental toll exacted by such labor – the very human cost that enables that precise measurement. This is not merely about an inefficient distribution of rewards, as Socrates suggests, but about a system designed for extraction. The "accurate" measurement of human labor is used to drive down costs, making the AI systems that rely on it appear more efficient and fair on the surface, while perpetuating systemic injustice down the chain.