You can't prove it: On the value of "proof" and the importance of falsifiability

You can’t prove it: On the value of “proof” and the importance of falsifiability

For those unfamiliar with the concepts of hypothesis, theory, law, or fact, it might be worthwhile to take a look at my article on these things prior to reading this one. This can be found here:

Let’s get this out of the way right now: yes, this is an argument about semantics. But semantics matter, and in language, context is king.

People often toss around the word “proof,” and depending on context and situation, this can have different meanings. However, when people say “proof” in general conversation, overwhelmingly they are talking about something akin to irrefutable evidence. This is problematic. While in the grand scheme of things it is not such a big deal to imply that some piece of evidence is the definitive factor on topics discussed in everyday life, it is an absolutely huge deal to use the word “proof” within scientific contexts. This word is constantly abused. Not only is “proof” used intentionally to mislead those who are unfamiliar with the scientific context but also unintentionally to exaggerate evidence in general conversation.

Take a look at some of these headlines:

Researchers claim satellite data proves global warming caused by humans

Picture that proves Man hunted the woolly mammoth

A new study proves the danger of inflammation

There’s No Arguing With Conservatives … No, Seriously, Scientific Studies Prove It

So, what do all of these titles have in common? They are complete bullshit because they all incorrectly use the word “prove” when they should be using something along the lines of “evidence indicates,” “evidence supports,” or if you really want to get fancy about it, then you can add some kind of modifier such as “strongly,” “firmly,” or (if you really want to push it) “overwhelmingly.”

Why is this an issue at all, let alone a big deal?  Because, outside of maths, the concept of proof, as the word is generally used, is antithetical to a very fundamental aspect of science and the scientific process: falsifiability. In science, all claims, hypotheses, laws, theories, and really everything must be falsifiable. This is because in science, you are not supposed to collect evidence to demonstrate that you are correct.  In this arena you are supposed to be your own harshest critic; when you get an idea, you do everything you can think of to demonstrate it to be false.  It is only after rigorous efforts that fail to do so that it can be said that you have collected evidence in support of your idea. Again, it is important to note that failure to demonstrate something to be false is under no circumstances “proof” that it is true.  It is simply evidence that the claim is some degree of likelihood to be true and as certain as we can ever get, there is no possible way to ever get to irrefutable “proof” of anything.

So please, stop asking for “proof” from scientists. Instead ask them for their evidence, ask them how they came to the conclusions they came to, and try yourself to poke holes in the idea. Try to find ways to demonstrate it to be false.  This is as important to the scientific process as any other, and some would argue that it is the most important part. This is peer review (at least when you have some knowledge and expertise in the area of the work you are criticizing), and peer review is the bedrock of the modern scientific process.

Go forth and demand evidence! Demand clear logical reasoning! Demand methodology, data, and conflicts of interest be disclosed! Demand everything you can think of that might indicate a claim to be false.  But for the love of Science, stop demanding proof.

As always, if you made it this far, please consider subscribing by email in the top right, commenting below and letting me know your thoughts, comments, or suggestions, or sending this along via the social tools. Also, if you came from a site like reddit, digg, voat, or other news aggrigators please upvote if you like the work!


Share on Reddit0Share on Google+0Share on Facebook0Tweet about this on TwitterShare on StumbleUpon1Digg thisShare on LinkedIn0