Skip to content
Coordinates of ThoughtHome of Philosophy
Join
Reading Group

Reading group: van Fraassen’s The Scientific Image, on realism and explanation

Tomasz Szewczak Fellow
· 5 replies
Scientific methodAnalyticBas van FraassenIan Hacking#constructive-empiricism#empirical-adequacy#observability#van-fraassen
Follow

The Scientific Image (1980) is the book that made anti-realism respectable again after a period when scientific realism seemed to have won. It's short, dense and often witty. I'd like to read the parts that matter most for the realism debate.

Plan, three weeks:

  • Week 1. Chapter 2, “Arguments concerning scientific realism”. Van Fraassen states realism carefully and sets out his alternative, constructive empiricism: science aims at theories that are empirically adequate, and accepting a theory involves believing only that it gets the observable phenomena right.
  • Week 2. Chapter 3, on empirical adequacy and models, where the semantic view of theories does a lot of quiet work.
  • Week 3. Chapter 5, “The pragmatics of explanation”, where explanations become answers to contrastive why-questions, and with that the realist's favourite inference, to the best explanation, loses its grip.

Question to carry: where exactly does van Fraassen draw the line between observable and unobservable, and can the line bear the weight he puts on it?

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

  1. Derek Hollis

    Fellow

    Week 1. The observable/unobservable line is the pressure point. Van Fraassen's examples: the moons of Jupiter count as observable, because an astronaut could go and look; particles in a cloud chamber don't, because we only ever see their tracks. Grover Maxwell had argued in the early sixties that this line is hopelessly vague. Van Fraassen's answer is that vague predicates are still usable if there are clear cases on each side. I found that reply better than I expected.

    Helpful · 1
  2. Lars Eriksen

    Fellow

    Also week 1, from a physics reader who disagrees with nearly every page. Why should what an unaided human eye could see mark an epistemic boundary? It's a fact about our biology, not about evidence. Van Fraassen admits it's anthropocentric and says that's fine because science is a human enterprise. I don't think “fine” is an argument.

    Helpful · 2
  3. Ana Clara Ferreira

    Fellow

    Week 2. The semantic view, theories as families of models rather than sets of sentences, is what makes empirical adequacy precise. A theory is empirically adequate if the observable phenomena can be embedded in one of its models. As someone who works on mathematical structures, I find that more illuminating than the realism debate itself.

    Helpful · 3
  4. Emeka Anyaoha

    Fellow

    A thread to read alongside: Hacking's paper “Do we see through a microscope?” from the early eighties presses van Fraassen on instruments. If I see a cell through an optical microscope, then an electron microscope, then by fluorescence, and all agree, at what point did I stop observing? Hacking thinks the convergence of independent instruments gives us reason to believe in the structures, observable or not.

  5. Tomasz Szewczak

    Fellow

    Week 3. The move I hadn't seen coming: if explanation is answering a why-question relative to a contrast and an interest, then “this theory explains best” is partly a fact about our interests, and you can't infer truth from it. That cuts the no-miracles argument off at the root. It also accounts for the flagpole case neatly. Whether you accept it depends on whether you think explanatory goodness is ever more than pragmatic, and I'm still not sure I do.

    Helpful · 1