Lakatos's Methodology of Scientific Research Programmes
Imre Lakatos's account of research programmes, how their theories change over time, and what distinguishes progressive revisions from degenerating ones.
PhilosophyAll chapters Reading tools Notes & comments
Evaluating a research programme › When Should a Failure Change the Programme?Full chapter · Sections in this chapter1 / 7 ↓
When Should a Failure Change the Programme?
Imagine two groups trying to explain a small but persistent discrepancy in a planet's orbit. One thinks the calculation needs to include the pull of another body. The other suspects that the laws used in the calculation have reached a limit. Both know there is a discrepancy. What separates them is which parts of the explanation they are willing to reconsider first, and what they expect to learn by doing so. The next observation might be much the same whoever makes it. It could have different implications once the groups have developed their proposals far enough to say what each would expect to find.
Imre Lakatos places questions like these at the center of his account of scientific research programmes. The central essay, “Falsification and the Methodology of Scientific Research Programmes,” was published in Criticism and the Growth of Knowledge and reprinted as the first chapter of the collection The Methodology of Scientific Research Programmes. His subject is the development of a sequence of theories. Someone who tries to judge only the latest one can miss the difference between a proposal that has opened a new line of evidence and a series of efforts to explain why each earlier difficulty can be set aside. Lakatos, “Falsification and the Methodology of Scientific Research Programmes,” §§2(c), 3 · Cambridge edition record
A Test Puts Several Claims at Risk
A familiar argument begins with a theory and a prediction that follows from it. If the prediction fails, something in the reasoning that produced it or in the premises ascribed to that situation has to give way. A real test, however, also relies on claims about the case to which it is applied. In the orbital example, the result of a calculation depends on which bodies are included, their starting positions, and the treatment of relevant forces. The report from the telescope depends on how measurements were made and interpreted. A contradiction limits what the participants can consistently accept together. It may leave room for several choices about what to revise. Lakatos, “Falsification,” §§2(b)–(c)
This is one reason Lakatos distinguishes dogmatic falsification from more sophisticated positions. The dogmatic approach expects evidence to provide a verdict that can conclusively disprove a theory. He also discusses versions of methodological falsification that accept the fallibility of reports and make decisions about which observations and background theories to use in a particular test. He builds on that recognition. His proposal is to judge a failed theory in relation to alternatives that can preserve some of its successful work, make further commitments, and put those commitments to a test. Lakatos, “Falsification,” §§2(a)–(c)
Consider a simpler case. A laboratory predicts that a liquid will change color in a sealed container at a specified temperature. The change does not appear. A researcher can still question the calibration of the thermometer, the seal, the exact composition of the liquid, or the predicted threshold. Carrying out another test requires treating some of those issues as settled for the time being. Retaining a familiar theory during the next investigation can be reasonable if the investigators know which checks would bear on the alternatives. It would be a different kind of decision to declare from the outset that every conceivable result would be compatible with the theory, whatever else had to change. Lakatos, “Falsification,” §§2(c), 3(a)
Where the Programme Directs the Work
Lakatos calls the set of central commitments a programme holds fixed its hard core. In his example from the Newtonian programme, it includes the three laws of dynamics and the law of gravitation. Participants take disagreements with data as reasons to look initially at other parts of their explanation. Lakatos describes this protection as a methodological decision by the participants. The decision gives the programme a continuity across successive versions. It also means the outlook must prove what it can do through the changing theories that develop around the core. Lakatos, “Falsification,” §3(a)
The assumptions and initial conditions open to revision form a protective belt. Someone applying the laws of dynamics may have to decide which bodies to treat as isolated, which forces matter, and which detailed description to start with. If the predicted and observed paths differ, fixing one of those assumptions can be a good answer even when the laws are kept. The price of making that adjustment intelligible is to be explicit about which version of the combined explanation made the failed prediction and what the new version claims in its place. Lakatos, “Falsification,” §3(a)
There are two sides to the advice a programme gives. The negative heuristic specifies lines of revision to avoid. It can tell the researchers to reconsider auxiliary hypotheses when a result conflicts with a prediction. The positive heuristic suggests a direction in which to build these auxiliary theories. A group might begin with an idealized calculation and then work out how to include bodies that were initially treated as absent, or factors first left out to make the problem tractable. Having a reason to take the next model in that direction distinguishes a continuing programme from a set of changes with no relation beyond their arrival after a failure. Lakatos, “Falsification,” §§3(a)–(b)
The hard core identifies commitments participants hold fixed while developing a programme. The protective belt includes the auxiliary assumptions they can alter and test. The positive heuristic offers ways to develop those assumptions that can be considered before the next result comes in. Lakatos, “Falsification,” §§3(a)–(b)
The positive heuristic also helps explain how work can continue in the presence of known difficulties. An early model may be too restricted to answer a question its proponents already understand. They can pursue a more detailed version without needing an answer to every objection to the present one. What counts in favor of doing that is the prospect that the sequence will develop testable claims. The existence of a difficulty alone may tell us little about whether a proposed way of proceeding will yield a new result or just keep work occupied. Lakatos, “Falsification,” §3(b)
Lakatos gives the Newtonian series of models a particular role in illustrating those constraints. Modelling an orbit can begin with an intentionally simplified arrangement of bodies. The approximation directs attention to some of the strongest effects and leaves others to be introduced in more detailed cases. Lakatos describes a transition from simpler systems, with bodies treated as point masses or interacting in idealized ways, toward models that can take account of more of their properties and mutual effects. An idealization can guide this work even while its details become inappropriate at a later stage. He distinguishes the comparatively flexible way of building models from the central laws that this sequence of models is intended to put to work. Lakatos, “Falsification,” §3(b), Newtonian models and the flexibility of the positive heuristic
That distinction changes what you could learn from an objection to a model. If the early estimate assumes an isolated pair of bodies, finding that a third body influences the motion is a reason to reconsider the estimate. It also supplies a possible next step that a researcher could take without rewriting the basic equations. The estimate has done real work if it helps someone locate the discrepancy precisely enough to build the next model. It may remain serviceable in cases where the omitted effects are sufficiently small. Assessing it requires identifying the stage of the inquiry and what the people using it expected it to explain. Lakatos, “Falsification,” §3(b)
When an Adjustment Counts as Progress
Lakatos distinguishes theoretical from empirical progress. A sequence is theoretically progressive when a new theory has excess empirical content over the preceding one. It puts forward additional claims that can be tested, including novel facts that the earlier account did not lead us to expect. Empirical progress requires that some of this excess content be corroborated. There can be a gap between the two. One can often identify what an improved theory predicts as soon as it is proposed, while it may take a long time to find out whether the prediction holds. Lakatos, “Falsification,” §2(c), discussion of progressive and degenerating problemshifts
In the orbital case, suppose the new calculation can reproduce what the earlier observations already show. That is a useful start. Suppose it also gives investigators a region of the sky to inspect at a stated time, and a way to tell the predicted body from several nearby objects. The result of the search can now bear on a claim they did not have to make to reproduce the original discrepancy. A verified body in the relevant location would still have to be identified and shown to have the appropriate properties. The illustration has done its job once it shows the difference between describing the data that motivated a change and using the change to reach an additional piece of evidence. Lakatos, “Falsification,” §2(c)
A sequence can go the other way. When a calculation fails, a group can introduce a new quantity chosen to recover the old result. The next discrepant case can call for another such quantity, with no claim about where else it can be measured or what else will change by a predictable amount. The revisions may bring the proposed laws into agreement with more records. If they do not add testable claims, that agreement does not show the kind of progress Lakatos wants. His account asks what each change has permitted the investigators to find that they could not have found by staying with the old version. Lakatos, “Falsification,” §2(c)
Judging novelty can require more than checking the dates on two documents. Compare two ways of developing the example. In one, the extra source of force is assigned whatever value will reproduce the orbit that was already measured. In the other, the value is constrained by observations of another system, and working out its consequences directs the investigators to a further feature of the original orbit. These choices determine which results the new account was designed to fit and which it puts at risk. Lakatos's distinction gives people a reason to record those dependencies, since simply adding another line of confirmed data cannot tell you which part of the new theory earned support. Lakatos, “Falsification,” §2(c), discussion of excess empirical content
What Can Be Decided at the Time?
Some scientific results acquire much of their significance later. Lakatos discusses the motion of Mercury's perihelion, the point of the orbit closest to the Sun. Its anomalous motion was already a difficulty in the Newtonian programme. In Lakatos's presentation, Einstein's success in accounting for the discrepancy gave the earlier observation a new role in comparing programmes. The observation did not change in the meantime. What changed was what people could use it to discriminate between, because another theory was available and had enough structure to account for it. Lakatos, “Falsification,” §3(d), Mercury discussion
It follows that comparing programmes takes time. People have to develop the rival account, see what it preserves, and pursue observations that bear on its additional claims. An experiment can help settle which of two specific predictions should be accepted under given conditions. Whether it is a decisive episode in the history of two programmes may depend on other work, some of it still to be done. Lakatos uses the term instant rationality for the expectation that a single result or rule can immediately give a final verdict on a developing theory. He thinks the history of work on competing programmes makes a slower judgment more realistic. Lakatos, “Falsification,” §3(d)
In the meantime, researchers can identify where they are placing their bets. A group might keep working on a programme that has performed poorly lately because it has a definite proposal for making a test no rival has offered. A rival group might be already making successful predictions in cases the first group has found hard to reach. Comparing those situations can help someone decide where to spend their time. It cannot supply a deadline by which a programme has to recover. The method lets people ask whether working on it now amounts to pursuing a serious possibility or maintaining agreement with a familiar set of past results. The questions are more assessable than a rule that people must have the same verdict as soon as they see the same table of measurements. Lakatos, “Falsification,” §§3(c)–(d), 4
One challenge for the account is that an investigator choosing what to do has to act on a record that the historian will later know to be incomplete. The first group may have an idea that gives it a plausible research plan. There can be uncertainty about whether the idea is testable and will preserve enough of what scientists already know. The second group can point to tests it has passed, but the area where the two accounts differ most may not yet be accessible. Specifying what each side would count as progress can sharpen the disagreement before either gets to claim the result. It also leaves a written record of whether the hoped-for work eventually happened. Lakatos, “Falsification,” §3(d)
A new line of work may also begin inside a programme that has recently gone through a series of failures. Lakatos allows that a change in its positive heuristic can renew its prospects without replacing its core. That possibility gives its proponents a burden they can describe concretely. They have to show what the new direction will let them investigate and what findings would give reason to think it was productive. A former success may justify listening to their proposal. The evidence about whether the change worked will have to come from what the renewed programme can do. Lakatos, “Falsification,” §3(b)
Reading a Sequence of Theories
Take the first orbital discrepancy again. A useful record would show what the original calculation assumed, how the deviation was established, which part of the assumptions each group put up for revision, and what else the resulting proposal committed it to. Preserving those details keeps one later success from being confused with a chain of successes that made it possible. It also makes it harder to mistake a story that fits everything in the record for an account that led investigators to a result they might never have sought.
The distinction matters when an older theory keeps working in areas where its competitor has changed the terms of the inquiry. The observations the theory helped people obtain will still need to be understood, and a later approach can earn some of its standing by making sense of them. The task is to see whether the newer sequence does that work and also takes on commitments that can fail. That is where the history of testing and the history of developing a programme begin to tell us something together. Lakatos, “Falsification,” §§2(c), 3(d)
Further Reading
- Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes”. Start with §2(c) on progressive problemshifts and §§3(a)–(b) on the two heuristics.
- Cambridge University Press, The Methodology of Scientific Research Programmes, Philosophical Papers Volume 1, edited by John Worrall and Gregory Currie. The collection also contains essays on historical reconstruction and on Popper.
Join the discussionAsk a question or share what you learned.
Keep exploring
Reference
Citation details
Cite or share this article.

Comments
Add to the conversation
A question, a correction, or an example from your own work.
No comments yet. Be the first to share your thoughts!