Other terms for Popper's "metaphysical research program"

Question

Are there synonyms or alternative terms for Poppers "metaphysical research program"? This seems like a useful concept for practicing scientists, but telling them that their work isn't "true" science but is "metaphysical" generally does not go over very well

Some context: Part of my graduate work in ecology investigates a phenomenon in plants, known as allelopathy, for which it is potentially impossible to conduct realistic experiments that achieve the ideals of Popperian falsification. Support for the existence and importance of the phenomenon is often built up from experiments done in simplified model systems (eg, in petri dishes, garden pots, green houses, etc) but experiments "in the field" in ecologically realistic scenarios are fraught with problems, and alternative hypotheses are difficult to falsify. Papers on the topic (including mine) are fraught with phrases with like "the data are consistent with the hypothesis of allelopathy" but neglect to mention that the data are not necessarily inconsistent with alternative hypotheses (eg, competition).

My elaborate questions are 1)is a realm of inquiry such as this related to Popper's "metaphysical research program" idea and 2)are there synonyms or alternative terms for this type of science? Telling biologists that there work isn't "true" science but is "metaphysical" generally does not go over very well!

Answer 1

Popper held that there were three ways in which a theory could be tested. One was to test whether is is self-consistent. The second is to check whether it is consistent with experimental results. The third is to check whether it solves the explanatory problems that it was invented to solve: Popper called this the "metaphysical research program". See "Quantum Theory and the Schism in Physics", Chapter IV.

All knowledge creation proceeds by conjecture and criticism. You notice a problem with current knowledge, propose solutions to that problem and criticise the solutions until only one is left and it has no outstanding criticisms. There is no such thing as support for a theory. A theory either passes or fails a test. If it passes then it lives to fight another day, but that doesn't change its status.

Now, if you have a theory that allelopathy might be important then the easiest way to test that is to see if it operates when no other effects play a role, such as in a garden pot. If you look in the wild, you might not be able to exclude the other effects, but you may still be able to tell whether some set of ideas can be ruled out, e.g. - allelopathy and competition. This does not make the theories metaphysical, since there are states of affairs you know of that would rule out both theories. There is then a separate question of whether there is a state of affairs that would rule out one theory but not the other, e.g. - a distribution of plants that would rule out allelopathy, say. The most important issue to bear in mind is that you should be clear about what a given test actually rules out.

Another issue to bear in mind is that the experiment itself is interpreted in the light of an explanation about what is going on during the experiment. You should be interested in looking for flaws in that explanation, or ways in which it might enable you to do other tests than what you originally intended to do. If that explanation implies that allelopathy implies X about the root system of plants and competition implies Y and they can be distinguished, then it is worth considering how that test could be done.

As for metaphysical research programs, I don't see any particular reason to adopt a fancy term for looking at your theory as an explanation and considering whether or not it cuts the mustard, although it is something that you should do.

For more on the creation of knowledge, see "Realism and the Aim of Science" by Popper, Chapter I and "The Beginning of Infinity" and "The Fabric of Reality" by David Deutsch.

Answer 2

I would suggest to use the expression research program, and to just drop the word metaphysical. Popper himself used the expression in this way sometimes, as can be seen in two passages that are quoted in the Wikipedia article on him:

I still believe that natural selection works in this way as a research programme. Nevertheless, I have changed my mind about the testability and logical status of the theory of natural selection.

and

Thus not all phenomena of evolution are explained by natural selection alone. Yet in every particular case it is a challenging research program to show how far natural selection can possibly be held responsible for the evolution of a particular organ or behavioral program.

Answer 3

First, it's good to have actual scientists (in training) asking questions on SE.Philosophy!

Does falsificationism apply to scientific practice?

You seem to be bothered by the observation that the scientific practice in your field doesn't conform to Popperian ideals of falsification(ism).

Well, the good news is that no scientific practice conforms to Popper's ideals, so you're not alone.

Indeed, Popper's theory has no pretence to describe scientific practice, it is a normative theory, its aim (at least starting with Logic of scientific discovery of 1934) is to postulate ideal methodological rules.

So

are there synonyms or alternative terms for this [i.e. my] type of science?

Yes, real science! :)

And there's nothing particularly metaphysical about that. The notion of "metaphysical research program" is not directly relevant to the problem you articulate in your question. (Feel free to ask another question if you want to know more about it.)

Falsificationism and the problem of competing hypotheses

Going into the specifics of your question, you write that in your field it is

potentially impossible to conduct realistic experiments that achieve the ideals of Popperian falsification

but the specific point you are concerned with is that

alternative hypotheses are difficult to falsify. Papers on the topic (including mine) are fraught with phrases with like "the data are consistent with the hypothesis of allelopathy" but neglect to mention that the data are not necessarily inconsistent with alternative hypotheses (eg, competition).

Your problem is difficult to address in the Popperian framework because falsificationism addresses theories, i.e. a conjunction of multiple hypotheses (plus other things), not a single hypothesis.

Popper was aware that data may be consistent with multiple competing hypotheses and that is why he thought that sophisticated falsification happens at the level of theories, not hypotheses.

While two competing theories may overlap in many of their predictions, they might not overlap in all. So there's at least one pair of competing hypotheses that predict two contrary experimental outcomes and that can be used to stage a crucial experiment. This experiment can rule out one hypothesis, thus falsifying one theory, while the other theory survives.

(Please note that the logical possibility or factual existence of crucial experiments has been disputed and debated for a long time. Popper's own theory was a response to Pierre Duhem's view that crucial experiments are a logical impossibility. This is called "contrastive underdetermination". See the problem of underdetermination of scientific theory by evidence for a background discussion.)

Conclusion

Even according to Popper, crucial experiments are rare. It is normal for scientists to work with data that is consistent with multiple hypotheses most of the time.

Curiously enough, there are also enough scientific problems for which scientists have struggled to find any viable theory and are happy when they find one. No wonder that they disregard the philosophical discussion about competing alternatives to be highly artificial. I wonder if this is specific to certain sciences.

---

Further reading:

• See the Standford Encyclopedia of Philosophy for a detailed introduction to the philosophical problem of underdetermination.