Where does defect collide?

In products liability, a defect is a condition in which a product departs from what a reasonable buyer is entitled to expect, whether from a flaw in how one unit was made, a choice in how the whole line was designed, or a missing warning, and its presence can make the seller answerable for harm the product causes. MacPherson v. Buick Motor Co. (1916) held that a maker owes a duty of care to the eventual user of a defectively built product even with no direct contract between them, and the American Law Institute's Restatement (Second) of Torts section 402A (1965) later placed strict liability on one who sells a product in a defective condition that is unreasonably dangerous. The referent is a legal judgment about a marketed good and the duty it carries, not a missing atom in a crystal or a choice to betray a partner in a game.[1][2]

Unit

One product condition judged defective, per claim or product line

Recovery means

A defect may be cured by recall, redesign, or an added warning

Measured by

Established at trial by evidence that the product departed from its intended design, from its sibling units, or from a safer feasible alternative

In a crystalline solid, a defect is a place where the regular repeating arrangement of atoms is broken: a missing atom at a site (a vacancy), an extra atom squeezed between sites (an interstitial), or a line along which whole planes of atoms are displaced (a dislocation). Frenkel's 1926 Über die Wärmebewegung in festen und flüssigen Körpern proposed that thermal motion leaves vacant sites and interstitial atoms scattered through a lattice, and Taylor's 1934 theory of plastic deformation showed that moving dislocations let real crystals bend far more easily than a flawless lattice ever could. The referent is a physical irregularity in an atomic lattice, counted by how many occur in a given volume, not a legal fault in a sold product or a strategic move against a partner.[3][4]

Unit

One lattice irregularity (a vacancy, interstitial, or dislocation), reported as defects per unit volume or dislocation length per volume

Recovery means

Annealing lets atoms migrate so vacancies and dislocations are reduced

Measured by

Imaged by transmission electron microscopy and X-ray diffraction, and inferred from electrical, mechanical, and diffusion behaviour

In the prisoner's dilemma and related games, to defect is to take the self-serving action that pays better than cooperating whatever the other player does, even though both defecting leaves the pair worse off than both cooperating would. The game was devised by Merrill Flood and Melvin Dresher at RAND in 1950 and reported in Flood's 1958 Some experimental games, and Axelrod and Hamilton's 1981 The evolution of cooperation examined when defection or cooperation prevails across repeated play among biological and human agents. The referent is a labelled choice inside a formal payoff structure, scored in utility, not a flaw in a crystal or a deficiency in a sold product.[5][6]

Unit

One player's chosen action (defect or cooperate) in a single round, scored by its payoff

Recovery means

Not applicable

Measured by

Recorded as the action chosen and its payoff, and summarised over repeated rounds as a rate of defection

In sociology of religion, defect refers to leaving a new religious movement, studied as a process of disengagement from a high-commitment group. This meaning comes from Wright (1987) in Leaving cults.[7]

Unit

Recovery means

Measured by