Hacker News new | past | comments | ask | show | jobs | submit login
What Is a Thought Experiment, Anyhow? (2006) [pdf] (philpapers.org)
25 points by lainon on Dec 20, 2016 | hide | past | favorite | 9 comments

The parent article is about 99% fluff, but for people interested in Philosophy of Science, John D. Norton (from the article) has a treasure trove of interesting stuff.



Unfortunately our universities are big failure when it comes to properly maintaining and archiving their websites. I guess the librarians have not been involved when it comes to matters of IT. Anyway, you have to search for his stuff on archive.org. For thought experiments, check out things like:

Thought Experiment Anti-Thought Experiment Pairs


and his papers on Maxwell's demon:




Thanks for the links, and I see that they are of Norton's work. Going solely on the article, I felt I could understand Norton's view, but in the author's outline of Brown's view, I haven't been able to make the jump from "the laws of gravity... are 'real' in that they exist, again, independently of human observers" to "thought experiments are genuine examples of how the human mind can 'perceive' laws of nature by simply thinking about reality." The first thing I want to ask about that (beyond just how it works) is how can you tell that you are not just making it up? Of course, one way would be to use reasoning to see if your ideas are consistent with what you take to be known, but does that not give you just Norton's view decorated with some metaphysical musings?

I realize that Brown's version may be too subtle to even be summarized in an article such as this, and if someone knows of a link to a fair outline of Brown's views, I would give it a go.

As an aside, I am also amused to see 'real' in quotes (and it is not for the reason I just put it there). It puts me in mind of George Burns' joke about faking sincerity.

If you are interested in thought experiments, this book does a great job of providing you "tools for thinking". The book is called intuition pumps and other tools for thinking by Daniel Dennett.


This was my immediate idea for a comment as well. His "Darwin's Dangerous Idea" book also has an extended discussion on the "Library of Mendel" which I forgot if he included in intuition pumps. But from a programming languages perspective I think the discussion of navigating "design space" in a generalized way is a very powerful abstraction.

Rather delightfully, the phrase "intuition pump" is itself an intuition pump :)

Perhaps I am missing something, but the example given here seems to have been loaded up with more insight than it can bear - it can refute "heavier things fall faster" but not conclude that all things' acceleration under gravity alone is independent of their mass (the article itself says "two objects of different weight must fall at the same speed" which is simply false.)

To reach the equal acceleration conclusion by way of this thought experiment, I think you have to have a quantitative and fairly comprehensive theory of drag, and a supercomputer on which to do the calculations. Of course, one alternative is to posit that drag is caused by movement through the surrounding fluid and then, without elaborating on that, perform an experiment - a real experiment - in a vacuum.

The major misconception I always had was that thought experiments were tools for teaching, i. e. cute little metaphors to make special relativity accessible to 16-year olds.

The reality (on the macro scale :) was apparently that especially Einstein strongly relied on them to allow him to use project intuition into these rather strange realms he was working on.

(The book I got this from added something like "because his math skills were relatively shallow", but people always misunderstand that because they don't see the unspoken context "... among physicists winning two or more Nobels")

Gedanken! https://www.britannica.com/science/Gedankenexperiment I love this word and it's associated meanings. Einstein is it's most famous practitioner, but I find it's a useful way to play out situations to see possible outcomes.

I think of thought experiments as consistency audits on a mental framework - they can't tell you anything new about the world directly, but they can reveal problems in the way you've broken the problem down.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact