Skip to content

Progressive vs Degenerating Problemshifts

Definition

Lakatos's demarcation criterion: a programme is theoretically progressive if each new version predicts novel, unexpected facts; empirically progressive if some of those predictions are corroborated; and degenerating if its adjustments only accommodate known facts after the fact, explaining yesterday's anomaly while predicting nothing new. "A research programme is said to be progressing as long as its theoretical growth anticipates its empirical growth... it is stagnating if its theoretical growth lags behind its empirical growth." Science versus pseudoscience is a property of trajectory, not of a theory's content at a moment.

In the Book

The chapter's twin exhibits are Einstein and Marxism. Einstein's general relativity predicted light-bending before Eddington's 1919 expedition measured it — a stunning, corroborated novel fact, textbook progressive. Marxism, in Lakatos's harsh audit, made real predictions that failed (the immiseration of workers, revolution in the most industrialized nations first) and each failure was absorbed by a belt-patch ("the workers' rising living standard? false consciousness!") that predicted nothing further — degenerating, structurally identical to Newtonianism (hard core, belt, heuristic all present) but opposite in trajectory. The chapter stresses degeneration is not death: a lagging programme can recover, as Prout's did, which is why the verdict is always provisional and dated rather than final.

Why It Matters

This reframes "is this framework still valid?" from a content question into a growth-curve question, judged the way markets judge companies rather than the way a court judges a single claim. The portable audit: list a framework's last several adjustments — to a strategy, a model, a product thesis — and ask of each whether it predicted something new that came true, or merely explained away a miss. A string of retrospective explanations is degeneration whatever the sunk investment says, and the same audit unmasks hype: something whose every new version re-explains its unmet promises rather than calling a new one is a degenerating programme wearing a roadmap.