The New Experimentalists (and Deborah Mayo)

On Sunday, I chickened out of describing the New Experimentalists’ philosophy of science because I actually couldn’t remember very much of what it entailed. I promised myself I would come back to it and so here I am, exploring New Experimentalism.

This blog post is a bit silly because instead of starting at the beginning I’m actually describing here a recent addition to the philosophy of science. It’s also not very well sourced. Most of my understanding, in broad strokes, comes from Alan Chalmers’ What is This Thing Called Science?

*

Part of the problem with anything called “New” is that there is necessarily an old which probably needs to be understood before the new can be understood. In her 1994 paper which coined the name ‘New Experimentalist’ Deborah Mayo writes…

“[New Experimentalists] hope is to find ways to steer a path between the old logical empiricism, where observations were deemed relatively unproblematic, and the more pessimistic post-Kuhnians, who take the failure of logical empiricist models of appraisal as leading to underdetermination and holistic theory change, if not to denying outright the role of evidence in constraining appraisal”

Mayo, 1994

In a nutshell, the story of the Philosophy of Science (as I roughly understand it) goes like this: In the beginning, things were relatively straightforward. Scientists were thought to undertake experimentation which led to facts that joined together to make theories (inductive reasoning).

Thomas Kuhn, Science Philosopher

But then philosophers thought some more and started getting worried. Surely theory plays a part in the way that scientists see the world and which experiments they choose to carry out and how they carry them out. The interpretation of the results of experiment also relies on theory as scientists have to determine what results of an experiment are valuable and significant and they can often use their theoretical understanding to help them with this.  

Thomas Kuhn and his book The Structure of Scientific Revolutions is a significant marker in this change from a focus on experiment to one on theory theory. One of the ideas that made sense to me that I think comes from Kuhn is that aliens or other extra-theory people conducting the same experiments or viewing the same results might actually produce an understanding of the world very different from our own. There is nothing, argued these people, particularly definitive about experiment.

Between the 1950s and the 1970s, things got crazy. Experiment was dependent on theory which was clearly almost always wrong. People like Paul Feyerabend took this to extremes, eventually arguing that since theory was always wrong and experiments were dependent on theory, very little about science could be said to be more correct than it had been before. In short, science was paddling around in circles.

I don’t know for sure, but I get the sense that Feyerabend’s ideas were the trigger for everyone else to go, “Hold on a sec.” Enter the New Experimentalists who, according to Mayo’s footnotes, were writing in the 80s.

I’ve linked Deborah Mayo and New Experimentalists together, but in fact her 1994 paper actually attempts to fix some of the problems which she perceives to be part of the ideas of what she dubs the New Experimentalist movement. I will come back to those at the end.

At the beginning of the 1994 paper linked above, Mayo invokes an ‘apt slogan’ to explain the New Experimentalists that was used by New Experimentalist Ian Hacking in a 1983 paper: “Experiments have a life of their own”. This of course stands in opposition to earlier theories where experimentation was theory dominated.

She articulates three ways of interpreting this slogan, which I’ve summarized…

1. Much of experimentation is “local” to a specific problem, focusing on extracting results from background noise, for example, and bears little reference to theory.
2. If you are making up a philosophy of experimental science, “it cannot make the very concept of observation become suspect.” (Hacking again)
3. The results of experiment are not only valuable within the theory they were created in. After Newtonian physics fell, observations made under it were still valid and a new theory or theories had to take most of those observations into account.

Relieved experimenters (Clipart: Philip Martin)

New Experimentalism put experiment back into philosophy of science, which must have been a big relief to those people carrying out experiments.

Mayo goes on to discuss a problem with this new idea-- that this is all well and good, but the myriad possible errors of observation, perception and the evaluation of what the results actually are still pose a problem between the carrying out of experiments and the induction of fact.

For me, it is easier to understand Chalmers on Mayo than Mayo explaining herself. He describes Mayo thus:

“A key idea underlying her treatment is that a claim can only be said to be supported by experiment if the various ways in which the claim could be at fault have been investigated and eliminated. A claim can only be said to be borne out by experiment if it has been severely tested be experiment, and a severe test of a claim, as usefully construed by Mayo, must be such that the claim would be unlikely to pass it if it were false.” (pg 202)

She articulates how this might be done although it does appear that each individual experimental case might require a specific assessment of what the possible sources of error could be and then a specific elimination of each possible error, much like making up the dreaded, ‘risk assessment’ before going on a field trip (my words not theirs).

All in all, New Experimentalists allow for science to include facts known to a high level of probability and for those facts to be added to: it can be both (relatively) correct and progressive, which I think is a win for science.