What can we prove?

16 Oct

Evidence mounts in the strange case of proving that Kevin Bacon, the man who is seven degrees from everyone, has never been to the Liberty Bell in Philadelphia.

The hypothesis: Kevin Bacon has NEVER visited the Liberty Bell


Exhibit A: Kevin Bacon standing in front of a VW beetle – not the Liberty Bell


Exhibits B-E: The Liberty Bell with accompanying visitors, none of which are Kevin Bacon


Exhibit F: The Liberty Bell, yet again, without Kevin Bacon anywhere in sight


Conclusion: Hypothesis accepted – Kevin Bacon has NEVER been to the Liberty Bell


What’s wrong with this argument?

Special thanks to Paul Offit, MD, whose Coursera Vaccines Class inspired this investigation


Posted by on October 16, 2013 in Uncategorized


Tags: , , , , , , , , , , ,

4 responses to “What can we prove?

  1. Mehron

    October 17, 2013 at 11:40 pm

    What if the study relied on 24/7 security camera footage or a visitor logbook as evidence instead? Could one not prove the negative that way?

  2. downhousesoftware

    October 18, 2013 at 7:16 am

    Proofs are for triangles.
    You would certainly strengthen your argument though – especially if the question is modified to ‘Bacon hasn’t visited the liberty bell since …’
    There’s a taboo against using the word ‘proof’.

    One might argue that this taboo should be loosened when speaking to the public as a means of at least leveling the playing field of discourse. But that solution would lead to its own problems once ‘scientifically proven’ issues get new evidence suggesting they’re wrong.

  3. Mehron

    October 18, 2013 at 11:32 am

    True. Maybe I could just feel more confident that he hadn’t been there.

    “Scientifically proven” seems to a good red flag for spotting questionable claims: The Scientifically Proven Subterfuge

  4. downhousesoftware

    October 21, 2013 at 9:17 pm

    Bryn replied (via twitter): “Love this: Logic 101. Also, can you imagine a Bayesian solution?” I’m not so handy with stats, so I don’t know how one would do this a la Bayes. I think he would say that his method doesn’t address this question very well, but that the probability was inching towards zero.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: