Larry Laudan — Progress, the Pessimistic Meta-Induction, and the Reticulated Model of Science (1941–2022)

Larry Laudan was an American philosopher of science whose career — at Pittsburgh, Virginia Tech, Hawaiʻi, and the National Autonomous University of Mexico — produced some of the most influential and most contested arguments in post-positivist philosophy of science: a problem-solving account of scientific progress, the reticulated model of scientific rationality, and the pessimistic meta-induction against scientific realism — an argument of such force and simplicity that it restructured the realism debate and has not been definitively answered in four decades.

Trained in physics at Kansas before philosophy at Princeton, he brought to philosophy of science both a scientist's respect for empirical constraint and a historian's appreciation for how scientific traditions actually develop over time — finding in that combination a position between the naive optimism of positivism and the radical relativism of the post-Kuhnian left.

His central concern: that neither scientific realism nor scientific relativism was adequately supported by the actual history of science — that science was genuinely progressive without being a march toward truth, and that understanding what that progress consisted in required a more sophisticated account of rationality than either camp had provided.

Progress and Its Problems — Science as Problem-Solving

Laudan's 1977 book "Progress and Its Problems" offered a fundamental reconceptualization of what scientific progress meant — and in doing so, challenged both the dominant realist account and the relativist alternative.

His argument: science is fundamentally a problem-solving activity. Scientists face both empirical problems — facts or regularities that need explaining — and conceptual problems — inconsistencies within or between theories. Progress consists in the increasing effectiveness of research traditions at solving problems of both kinds, and at reducing the anomalies and conceptual difficulties that every theory generates. A theory or research tradition is more progressive than its rivals if it solves more problems, more important problems, and generates fewer unsolved ones.

This account had the advantage of being historically tractable — one could actually evaluate competing research traditions by examining their problem-solving records — and it avoided both the realist assumption that progress required getting closer to truth and the relativist abandonment of any evaluative standard. Science could be genuinely progressive without its theories being true or approximately true.

"Science aims to maximize the scope of solved empirical problems while minimizing the scope of anomalous and conceptual problems — this is what progress consists in, and it can be assessed without any commitment to whether our theories are true."

The Pessimistic Meta-Induction — History Against Realism

Laudan's most influential and most debated contribution was his 1981 paper "A Confutation of Convergent Realism" — which articulated what has become known as the pessimistic meta-induction.

The argument is deceptively simple. Scientific realists claim that the success of our current theories is best explained by their being approximately true — that only if our theories are at least roughly right about the world can we explain why they work so well. This is Putnam's "no miracles" argument: it would be a miracle if our theories were successful and false.

Laudan's response: look at the history of science. The history of science is littered with theories that were empirically successful — that made accurate predictions, that guided fruitful research programs — and that we now know to be false. The caloric theory of heat was successful and false. Phlogiston theory was successful and false. The theory of crystalline spheres was successful and false. Laudan compiled a list and noted that for every past successful theory we now believe was genuinely referring, we could find half a dozen once-successful theories we now regard as substantially non-referring. If past successful theories were false, why should we believe our current successful theories are true? The same induction that tells us to trust current theories tells us they will likely prove false, just as their predecessors did.

"If past successful theories were found to be false, we have no reason to believe the realist's claim that our currently successful theories are approximately true — history of science is the graveyard of once-successful theories."

The Reticulated Model — Beyond Hierarchical Rationality

Laudan's "Science and Values" (1984) developed the reticulated model — his most sophisticated contribution to the theory of scientific rationality — as a response to what he saw as the inadequacy of both Popper's hierarchical model and Kuhn's apparent relativism.

The standard hierarchical view: when scientists disagree about theories, they appeal to methods to resolve the dispute; when they disagree about methods, they appeal to aims; aims (or values) constitute the fixed background that makes methodological and theoretical disputes resolvable. Laudan's critique: this model is historically false. Scientists' aims and methods change over time as much as theories do — and the changes are not irrational but responsive to theoretical results. New theories can legitimize new methods; new methods can advance new aims; new aims can justify new methods and theories. The traffic is two-way among all three.

On the reticulated model, there is no fixed Archimedean point — aims, methods, and theories form a web of mutual support and mutual revision — but this does not entail relativism. Rational adjudication is still possible: one can evaluate whether a set of aims is coherent, whether stated methods actually lead to stated aims, whether theories perform better than rivals by shared evaluative standards. Rationality is real but not algorithmic — it requires judgment rather than rule-application.

"That scientists sometimes disagree about which theory to accept is no more evidence for relativism than the fact that juries sometimes disagree is evidence that there is no truth about guilt or innocence."

Science and Relativism — Against the Post-Kuhnian Left

Laudan's "Science and Relativism" (1990) took the form of a dialogue between four fictional philosophers — a realist, a relativist, a pragmatist, and a positivist — and was designed to exhibit the weaknesses of each position while articulating his own. Its primary target was the relativist tradition — associated with Kuhn, Feyerabend, and Rorty — which had concluded from the genuine insights of historical epistemology that scientific rationality was incoherent or that science was just one epistemic tradition among others.

Laudan's position: the post-Kuhnians had correctly identified the inadequacy of naive positivist accounts of rationality — but had drawn the wrong conclusion from this. The failure of algorithmic rationality did not imply the failure of rationality as such. The history of science showed genuine problem-solving progress — genuine improvement in the effectiveness of research traditions — and this progress was real and evaluable even without convergence on truth and without fixed methodological rules.

"What distinguishes reliable belief-forming practices is not their conformity to a priori rules, but their long-run success in minimizing error."

Legal Epistemology — Evidence, Error, and Criminal Law

In his later career Laudan extended his philosophy of science framework to legal epistemology — asking what standards of evidence and proof were appropriate in criminal trials, and whether the existing legal standards actually minimized the risk of error in the ways they claimed to.

"Truth, Error, and Criminal Law" (2006) argued that existing legal standards — particularly "beyond reasonable doubt" — were philosophically confused and practically inadequate as guides to minimizing wrongful convictions. He applied his naturalistic methodology — the view that normative standards should be assessed by their actual effectiveness in achieving their stated aims — to the legal domain, with conclusions that were as critical of legal convention as his philosophy of science had been of positivist convention.

"What we want from legal epistemology is the same as what we want from scientific epistemology: reliable methods for minimizing error — and the question of whether current standards achieve this is an empirical question, not a philosophical one."

Legacy — The Argument That Wouldn't Go Away

Laudan died in 2022, having spent his final years in Austin, Texas — still engaging with the debates his work had generated. The pessimistic meta-induction remains one of the central arguments in philosophy of science — generating a literature of responses, modifications, and attempted refutations that has not yet produced a consensus against it. The realist responses — selective realism, structural realism, deployment realism — are all, in part, responses to Laudan's challenge.

His broader contribution was the demonstration that historical epistemology — the careful study of how science has actually developed — was an indispensable resource for philosophy of science, not an optional supplement. Neither a priori methodological rules nor abstract arguments about truth and reference could substitute for the historical record, which was the evidence base for any adequate account of scientific rationality.

On CivSim he belongs alongside Pierre Duhem, Harvey Brown, and I. Bernard Cohen — philosophers and historians of science who pressed hardest on the question of what scientific knowledge actually is and what its history tells us about its status. His pessimistic meta-induction is a direct challenge to any epistemology that relies on science's track record as evidence for its current theories' approximate truth — including the scientific foundations of Universal Humanism — and the challenge deserves engagement rather than dismissal.

"The history of science is the graveyard of once-successful theories — and this fact, if taken seriously, should make us far less confident that our current successful theories describe the world as it is."