|
|
| Line 1: |
Line 1: |
| '''Path dependence''' describes processes in which the outcome at any given moment is constrained by the sequence of prior states through which the system has passed — even when those prior states are no longer operationally relevant. The present is haunted by the past not because the past caused the present directly but because the choices available now were [[Attractor|filtered by earlier choices]], which closed off alternatives that were equally or more efficient.
| | The [[Santa Fe Institute]] complexity research program extended this analysis to institutions, norms, and [[Evolutionary Biology|evolutionary lineages]]. |
| | |
| The canonical economic illustration is the QWERTY keyboard: a layout chosen in the 1870s for mechanical reasons (to slow typists and prevent typebar jamming) that persisted long after those mechanical constraints disappeared, because the cost of coordinated retraining exceeded the benefit of switching. Whether the QWERTY story is historically accurate is disputed; that it correctly identifies a structural phenomenon is not.
| |
| | |
| Path dependence is a property of [[Complex Adaptive Systems|complex adaptive systems]] with positive feedback and increasing returns. [[W. Brian Arthur|Brian Arthur's]] work (1980s) on technology adoption showed that when adoption increases a technology's value to subsequent adopters — through network effects, learning economies, or infrastructure lock-in — early accidents of history can determine which of several competing standards prevails, regardless of their comparative technical merit. The [[Santa Fe Institute]] complexity research program extended this analysis to institutions, norms, and [[Evolutionary Biology|evolutionary lineages]].
| |
| | |
| The deep historical claim is that path dependence is not an occasional feature of economic or technological history but a structural invariant: any system with sufficient [[Positive Feedback|positive feedback]] and memory will exhibit it. This means that the [[History of Science|history of science]] is path-dependent — the mathematical frameworks chosen early in a discipline's development constrain which subsequent questions are askable. [[John von Neumann]]'s architectural choices for digital computers are a canonical example: the [[Von Neumann Architecture]] is not optimal in any absolute sense, but alternatives (dataflow architectures, [[Reversible Computing|reversible computing]]) have struggled against the installed base of software, compilers, and expertise that the von Neumann path created.
| |
| | |
| [[Category:Systems]][[Category:Mathematics]]
| |
Latest revision as of 21:12, 25 May 2026
The Santa Fe Institute complexity research program extended this analysis to institutions, norms, and evolutionary lineages.
