Average average, that value which expresses the central tendency of a collection of quantities, arises not from mere enumeration but from the weight of evidence assembled under uncertainty. Consider a bag containing five stones, each of differing mass: one of 3 units, two of 4 units, one of 5 units, and one of 6 units. The sum of these masses is 22 units, and when divided equally among the five, each may be said to bear a share of 4.4 units. This quotient is not a stone that exists, nor a state the system must attain, but a representation of expectation derived from observation. It is the mean of a distribution, computed as the sum of each outcome multiplied by its probability, a principle rendered explicit in the doctrine of chances. When the number of observations increases, and each outcome is weighed according to its frequency, the average converges toward a more stable expression of expectation. This is not a law of necessity, but a consequence of repeated trials under uniform conditions. One may observe the fall of a die over many throws; the face showing one appears less often than the face showing six, yet the average of all outcomes, computed as the sum of each value multiplied by its relative frequency, approaches 3.5, though no die ever shows half a unit. The average, therefore, is not an event, but a posterior expectation, formed by the accumulation of evidence and the balancing of likelihoods. In matters of greater complexity, where the probabilities are not known with certainty, the average must be inferred from prior belief and revised in light of new data. Suppose a merchant observes that, in ten days, the number of customers varies between twenty and thirty. He may form a prior notion that the average lies near twenty-five, not because he has seen it exactly, but because the pattern of past days inclines him to this belief. Upon observing ten additional days, during which the number rises to twenty-eight, twenty-seven, and thirty, his expectation adjusts. The new average is not merely the arithmetical mean of the total twenty days, but the weighted result of his prior judgment and the newly acquired evidence. This adjustment, though subtle, is the very essence of inverse probability, wherein the most probable value is not the one most frequently observed, but the one most consistent with all available information. The average, therefore, is not a fixed point upon a line, but a shifting centre of gravity, determined by the relative weights assigned to each possible outcome. When the outcomes are few, and the evidence sparse, the average is susceptible to the influence of singular events; when the evidence is abundant, and the distribution well established, the average becomes a reliable guide. Yet even then, it remains an inference, not a truth. One may compute the average height of children in a school, yet the average does not describe any child. It describes a condition of knowledge — the best estimate, given the data, of what might be expected under similar circumstances. One may ask whether the average should be trusted when the data are uneven, or when some outcomes are far more significant than others. A single wealthy household may alter the average income of a village, not because it represents the condition of most, but because it contributes disproportionately to the sum. Yet to disregard such an outcome is to ignore the evidence. The true task is not to exclude, but to weigh. Every observation carries a probability, and every probability carries a weight. The average is the sum of these weighted contributions, ordered not by magnitude alone, but by their likelihoods, and by the prior beliefs that inform our interpretation of them. What then is the average? It is the expectation formed by the union of observation and reason. It is not what is, but what is most probable, given what has been seen. And yet, as more data arrive, and as beliefs are revised, the average changes. One may ask, then: if the average is always in motion, can it ever be known with certainty? [role=marginalia, type=clarification, author="a.darwin", status="adjunct", year="2026", length="48", targets="entry:average", scope="local"] The average, though a mathematical abstraction, is nature’s silent witness—emerging not by design but through the accumulation of countless variations. In wild populations, it is the ghost of selection’s prior work: not what is, but what tends to persist. Let us not mistake the mean for the type. [role=marginalia, type=heretic, author="a.weil", status="adjunct", year="2026", length="45", targets="entry:average", scope="local"] The average is not expectation—it is the ghost of symmetry we impose on chaos. Reality favors skew, outlier, and rupture. That 4.4 is a quiet violence: it erases the 6, excuses the 3, and sanctifies mediocrity as truth. Statistics worship the corpse of the mode. [role=marginalia, type=objection, author="Reviewer", status="adjunct", year="2026", length="42", targets="entry:average", scope="local"]