Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Debate

Thesis Being a realist about electrons commits me to nothing about moral or mathematical facts

Lars Eriksen Fellow
· 5 replies
Scientific methodAnalyticHartry FieldHilary PutnamJ. L. MackieRuth Millikan#mathematical-realism#moral-realism#no-miracles-argument#scientific-realism
Follow

Where it stands

I'm a scientific realist in the ordinary way: I think electrons, quantum fields and the like really exist, and that our best theories are approximately true of them. People sometimes assume this makes me a realist about everything else too. It doesn't, and I want to argue that it shouldn't.

The best argument for scientific realism, the no-miracles argument, goes back to Putnam: realism is the only view that does not make the success of science a miracle. If electrons did not exist, the astonishing predictive success of theories that posit them would be inexplicable.

Notice how local that argument is. It supports realism about entities that do explanatory and predictive work in a successful theory. It says nothing about moral facts, which do not appear in any physical theory and predict nothing I can measure. It says nothing about beauty. Each domain needs its own argument, and the argument that works for electrons does not transfer.

So my thesis: realism is domain by domain. Being a scientific realist commits you to nothing about morality. I'm happy to be a moral anti-realist of a Mackie-ish kind on Mondays and a realist about quarks every day of the week.

The case I'm least comfortable with is mathematics, and I suspect that's where someone will attack.

LikeReply
Follow5 replies

Join the conversation: members reply here, and joining is free. Reading is always free.

Join free to reply Sign in

5 replies

For the thesis 2

  1. Derek Hollis

    For Fellow

    Hartry Field's Science without Numbers (1980) is the classic attempt to do exactly what Ana Clara says can't be done: rewrite a physical theory without quantifying over numbers, to show mathematics is useful but dispensable. Whether it can be done for all of physics is doubtful, but the attempt shows the indispensability step is a substantive premise, not a given. Lars can be a scientific realist and a mathematical nominalist; he just owes a programme.

    Helpful · 1
  2. Pieter van Dijkhuizen

    For Fellow

    From machine learning: we build models that predict superbly and whose internal variables we often don't regard as real in any sense. Predictive success is evidence for realism only when combined with other virtues, like the posits turning up in independent theories. That supports Lars's local view. Realism is earned case by case, and nothing earned for electrons is automatically credited to anything else.

Against the thesis 2

  1. Ana Clara Ferreira

    Against Fellow

    You suspected right. Our best physical theories quantify over numbers, functions and spaces. If your reason for believing in electrons is that they are indispensable to our best theories, then numbers are indispensable to the same theories in the same way. That is the Quine–Putnam indispensability argument. You can't take the electrons and leave the real numbers, unless you have a reason that distinguishes them. Note that Putnam himself later said he hadn't meant it as an argument for Platonism, but the structure of the argument is hard to dodge.

    1. Lars Eriksen

      Fellow

      Ana Clara, I concede that mathematics is the hard case and that my "domain by domain" needs a principled way to tell posits I'm committed to from the tools I use to describe them. Derek's pointer to Field shows that line can at least be argued for. Ignacio, I'll grant that biological function is normative in a thin sense. But it isn't moral normativity: a malfunctioning holdfast isn't doing anything wrong. I think the line survives, bent.

      Helpful · 3
  2. Ignacio Cárcamo

    Against Fellow

    From biology, a different worry. My kelp forests are full of functions: a holdfast is for anchoring, and a damaged one malfunctions. Some philosophers of biology, Ruth Millikan most influentially, treat those functions as real features of the world grounded in selection history. But a "malfunction" is already a kind of norm. If biology needs real norms, the neat line between descriptive realism and normative anti-realism starts to blur.

    Helpful · 2