Larry Laudan’s Philosophy of Science: Science as a Problem-Solving Activity

What if the goal of science isn’t to find “the truth” at all? That’s the provocative claim at the heart of Larry Laudan’s philosophy of science. Laudan (1941-2022), an American philosopher who spent decades studying how science actually works, argued that science is best understood not as a truth-seeking mission but as a problem-solving enterprise. His framework, laid out primarily in his landmark 1977 book Progress and Its Problems, challenged dominant views from Karl Popper, Thomas Kuhn, and Imre Lakatos – and offered a fresh, pragmatic way to understand scientific progress.

Why Laudan rejected truth as the aim of science 🔗

Before Laudan, most philosophers of science assumed some connection between science and truth. Logical empiricists and falsificationists believed science and truth were closely linked – that the scientific method steadily brings us closer to an accurate picture of reality. Even Kuhn, who introduced the idea of paradigm shifts and scientific revolutions, didn’t entirely abandon the idea that science makes progress in some meaningful direction.

Laudan took a different path. He argued that we simply cannot know whether our best theories correspond to some ultimate reality. Many once-successful scientific theories – Newtonian mechanics, caloric theory of heat, the ether theory of light – were later abandoned or significantly revised. If historically successful theories turned out to be wrong, then success alone cannot be evidence that a theory is true. This reasoning became central to what philosophers call the pessimistic induction argument against scientific realism.

For Laudan, this meant truth was neither a necessary nor a sufficient condition for evaluating theories. A theory might be true but trivial – failing to solve any important problems. Conversely, a theory might be strictly false yet enormously productive in solving problems. So instead of truth, Laudan proposed a different yardstick: problem-solving effectiveness.

The problem-solving model of science 🔗

At the core of Laudan’s philosophy is a simple but powerful idea: the fundamental unit of scientific progress is the solved problem. Science advances not by inching closer to truth but by solving more problems and solving them better than before. This is a thoroughly pragmatic view – what matters is whether a theory works, not whether it mirrors reality.

Laudan distinguished between two types of problems that science addresses, and both are equally important for understanding scientific progress.

Empirical problems 🔗

Empirical problems are questions about observable phenomena in the natural world. Why does a thrown ball follow a curved path? How do vaccines trigger immunity? What causes tides? These are the bread-and-butter questions that most people associate with scientific inquiry. When a theory provides a satisfactory explanation for an empirical phenomenon, it has “solved” that empirical problem.

Laudan further divided empirical problems into three categories. Solved problems are those that a given theory has adequately addressed. Unsolved problems are empirical questions that no existing theory has yet managed to explain. Anomalous problems are the most interesting: these are problems that a competing theory has solved but your theory has not. The concept of anomalous problems is distinctly comparative – a problem only becomes anomalous for a theory when a rival theory successfully addresses it.

Conceptual problems 🔗

Conceptual problems are higher-order difficulties that arise within or between theories themselves. These are not about explaining observable data but about the internal consistency and intellectual coherence of the theoretical framework. Laudan identified two subtypes here.

Internal conceptual problems occur when a theory contains logical inconsistencies, vague concepts, or circular reasoning. If a theory contradicts itself or relies on poorly defined terms, it faces an internal conceptual problem. External conceptual problems arise when a theory conflicts with other well-established theories or with prevailing methodological or philosophical commitments. The ongoing tension between quantum mechanics and general relativity is a classic example of an external conceptual problem in modern physics.

What made Laudan’s approach distinctive was his insistence that conceptual problems matter just as much as empirical ones. Traditional philosophy of science had focused almost entirely on empirical adequacy – whether a theory fits the data. Laudan argued this was a mistake. A theory that explains a great deal of data but generates severe conceptual difficulties is not necessarily more progressive than a theory with fewer empirical successes but greater conceptual coherence.

Measuring scientific progress 🔗

If science is about problem-solving, how do we measure whether it’s actually progressing? Laudan offered a straightforward formula. The overall problem-solving effectiveness of a theory is determined by weighing the number and significance of empirical problems it solves against the number and significance of anomalies and conceptual problems it generates.

A theory is progressive if it increases the total scope of solved empirical problems while reducing anomalies and conceptual difficulties. This means that, in principle, a shift from an empirically well-supported theory to a less well-supported one could still count as progressive – provided the newer theory resolved significant conceptual difficulties that plagued the older one. This is a strikingly flexible criterion, and it matches how scientific change has actually unfolded historically far better than rigid truth-based models.

Laudan also reversed the traditional relationship between progress and rationality. Most philosophers had defined scientific progress by first establishing what counts as rational inquiry and then measuring whether science meets that standard. Laudan flipped this: rationality is defined in terms of progress. A rational scientific choice is simply one that opts for the theory with greater problem-solving effectiveness. There is no need to invoke metaphysical notions of truth or verisimilitude to explain why scientists do what they do.

Research traditions: Laudan’s alternative to paradigms and research programmes 🔗

To understand how scientific problem-solving operates at a larger scale, Laudan developed the concept of research traditions (RTs). This concept builds upon – and deliberately modifies – Thomas Kuhn’s paradigms and Imre Lakatos’s scientific research programmes (SRPs).

A research tradition, for Laudan, is a set of general assumptions about the kinds of entities and processes that exist in a domain of study, along with the appropriate methods for investigating problems and building theories in that domain. Darwinism, Marxism, behaviourism, and the atomic theory of matter are all examples of broad research traditions that have guided generations of scientists.

How research traditions differ from Kuhn’s paradigms 🔗

Kuhn’s paradigms are monolithic and all-encompassing. During periods of “normal science,” a single paradigm dominates an entire field, and it can only be replaced through dramatic revolutionary upheaval. This model implies that scientists working under different paradigms are essentially working in different intellectual worlds – they cannot meaningfully communicate across the paradigm divide (the famous “incommensurability” thesis).

Laudan rejected this picture on several grounds. First, he argued that multiple research traditions can and do coexist within a single scientific field at the same time. Science is more pluralistic than Kuhn allowed. Physicists in the early twentieth century, for instance, were simultaneously working within classical, relativistic, and quantum frameworks. Second, Laudan denied that scientific change is always revolutionary. Traditions evolve gradually, and their core assumptions can shift over time without a dramatic “crisis” triggering a wholesale replacement. Scientific change is more often evolutionary than revolutionary.

How research traditions differ from Lakatos’s research programmes 🔗

Lakatos’s model was itself an attempt to improve on Kuhn. He proposed that every research programme has a “hard core” of unfalsifiable assumptions, protected by a “protective belt” of auxiliary hypotheses. The hard core never changes – if it does, you’re in a different research programme altogether.

Laudan found this too rigid. In his model, even the core assumptions of a research tradition can be revised over time. There is no absolute, untouchable hard core. This makes research traditions more flexible and historically realistic. The atomic theory, for example, has undergone dramatic changes – from Dalton’s solid spheres to Thomson’s plum pudding model to Rutherford’s nucleus to quantum mechanical orbitals – yet we can still recognise it as a single, evolving research tradition.

Additionally, Lakatos judged research programmes primarily by whether they were “progressive” (predicting new facts) or “degenerating” (only making ad hoc adjustments). Laudan broadened this criterion to include the resolution of conceptual problems, not just empirical predictions. A research tradition that resolves deep conceptual tensions might be more progressive than one that merely accumulates new empirical predictions.

The reticulated model of scientific rationality 🔗

Laudan didn’t stop at Progress and Its Problems. In his 1984 book Science and Values, he introduced what he called the reticulated model of scientific rationality – an important extension and refinement of his earlier work.

Before Laudan, the dominant view of scientific rationality was hierarchical. In this model, scientific aims sit at the top of a hierarchy. Aims justify methods, and methods justify theory choices. The flow of justification goes strictly one way – top-down. The problem with this model, Laudan argued, is that it cannot explain how or why scientific aims themselves change over time. If aims are foundational and unjustifiable by anything below them, then changes in scientific goals become irrational by definition.

Laudan’s reticulated model eliminates this rigid hierarchy. Instead, he proposed that theories, methods, and aims exist in a network of mutual influence and justification. Theories constrain which methods are viable. Methods constrain which aims are achievable. And aims shape which theories and methods are valued. All three levels can change, and changes at any one level can rationally prompt changes at the others.

This interconnected, web-like picture is what “reticulated” means – like a net. It allows for rational change at every level of scientific practice without requiring any fixed, unchangeable foundation. It also helps explain why scientific debates are often so complex: disagreements can occur simultaneously at the level of facts, methods, and goals.

Criticisms and legacy 🔗

Laudan’s framework, while influential, has faced its share of criticism. Some philosophers, notably John Worrall, argued that allowing change at all three levels – theories, methods, and aims – risks collapsing into a damaging relativism. If there is no fixed standard by which to measure progress, how can we avoid the conclusion that any change is as good as any other? Laudan’s response was characteristically pragmatic: we can measure progress from our current standpoint, using our best available standards, without needing to claim those standards are eternal or absolute.

Others have questioned whether problem-solving effectiveness can genuinely replace truth as the aim of science. If a theory solves many problems, doesn’t that give us at least some reason to think it’s getting something right about reality? Scientific realists have pushed back, arguing that Laudan’s pragmatism is too deflationary – that it doesn’t explain why problem-solving works if theories aren’t at least approximately true.

Despite these debates, Laudan’s contributions have left a lasting mark. His emphasis on conceptual problems broadened what counts as scientific progress. His research traditions concept offered a more historically faithful picture of science than either Kuhn’s paradigms or Lakatos’s research programmes. And his reticulated model remains an important reference point for anyone thinking about how scientific aims, methods, and theories interact.

Why Laudan’s ideas still matter 🔗

Laudan’s philosophy resonates beyond academic debate. In an era when public trust in science is frequently questioned, his framework offers a useful perspective. Science doesn’t need to promise absolute truth to be valuable. Its value lies in its demonstrated ability to solve problems – to cure diseases, build bridges, predict weather patterns, and understand the cosmos. If a new theory does this better than an old one, that constitutes genuine progress, regardless of whether we can prove it “truly” describes reality.

His pluralistic view of research traditions also speaks to how science actually operates today. Modern physics, biology, and the social sciences are all home to multiple competing frameworks. This isn’t a sign of crisis or failure – it’s a healthy, productive feature of scientific inquiry. Competition between research traditions drives innovation and problem-solving in ways that a monolithic paradigm never could.

Laudan gave us a philosophy of science that takes seriously what scientists actually do rather than what philosophers think they should do. That blend of historical sensitivity and philosophical rigour is perhaps his greatest legacy.

What do you think? Can science be genuinely progressive without aiming at truth, or does the success of problem-solving ultimately depend on our theories getting something right about reality? And does the coexistence of multiple research traditions within a single field strengthen or weaken scientific inquiry?

References

1. https://www.ucpress.edu/books/progress-and-its-problems/paper
2. https://plato.stanford.edu/entries/rationality-historicist/
3. https://en.wikipedia.org/wiki/Pessimistic_induction
4. https://plato.stanford.edu/entries/thomas-kuhn/
5. https://plato.stanford.edu/entries/lakatos/
6. https://link.springer.com/article/10.1007/BF01128903
7. https://www.scientowiki.com/Larry_Laudan
8. https://link.springer.com/article/10.1007/s10838-023-09670-5