Arguing From First Principles

. lecture : 2 minutes

Last week, while listening to the "Hello Internet" podcast, I heard CGP Grey & Brady Haran talk about "arguing from 1st principles", which Grey opposed to arguing with analogies, which he finds less useful.

I found the discution very intriguing, since I'm toying with the idea of making a tool to represent arguments as links between ideas, theories and evidence. If there are different types of arguments, I want to know all about it!

Wikipedia has a page about 1st principle arguing, but I also found a nice guide for debate clubs: Tim Sonnreich's Guide for university debating (2010 edition), available as a 2 MB .pdf file at:

The guide explains that 1st principle arguing is:


I have to admit that at first, I thought this was a misprint; surely the "EVIDENCE" part should be at the beginning, but no:

Finally there is the EVIDENCE. I put it last for two reasons – first because it’s the least important, and second because it should be the last thing you worry about – focus first on having the right IDEAS about what your side needs to argue, and then spend your time coming up with smart analysis to make it sound reasonable. If after that you have time for thinking up evidence and examples, then that’s great.

EVIDENCE can be statistics (boring, but can be helpful – like the unemployment rate before and after a policy, or the percentage of people affected by a particular problem, or the costs of a proposal) or quotes (not direct quotes, but knowing what important people have said about an issue). But at university level evidence is more commonly presented by case study or analogy

Tim Sonnreich's Training Guide for University Debating, 2010 Edition (p. 12)

By then I was thinking "well this is the worst kind of debate, he's trying to convince people, and not trying to discover the truth". But I guess I'm the dumb one, because a bit later I find:

The argument chain is weakest at link three – EVIDENCE – since it’s always easy to dispute the evidence presented by your opposition. But attacking the argument here is a poor strategy. Because the opposition can repair the chain by providing more evidence (which you attack, then they give more and it’s a stalemate) or by simply haggling over whether ASH is a good source is evidence. So booooring…

Attacking the argument a little higher, at the ANALYSIS, is more difficult but also more effective. If you can demonstrate that the ANALYSIS is illogical or based on assumptions that are not true (or unlikely to be true) then you heavily damage the credibility of the whole argument.

Tim Sonnreich's Training Guide for University Debating, 2010 Edition (p. 54–55)


I think we can all recall arguing on the interwebs, where the other side just keeps adding links of blog posts and dubious articles each time you discredit those sources. Attacking the evidence is easy, but may be pointless (unless you're reading a scientific paper, of course).