Quite a few people have linked to this video of Eric Berlow from TED. His theme is “how complexity leads to simplicity”.
He looks at how various visualisations offer us new insights into complex systems. He asks us not to dismiss as too complicated diagrams which appear to be particularly complicated.
“The more you step back embrace complexity, the better chance you have of finding simple answers.. which are often different from the simple answer you started with.
I warm to that theme. But….
His chief exhibit is a visualisation of the factors affecting the conflict in Iraq, famously featured in this New York Times article. He shows a clip where Eric remaps the ugly 2D diagram in three dimensions (much more attractively) and then that this complicated diagram can be seen as an ordered network. If we then look up to 3 degrees away from the node of popular support for the government, he says we can quickly identify the 2 key things that can be used to support the desired result.
His moral is that when we see diagrams like this, instead of being relieved we should be excited because simple anwers may emerge.
I don’t buy this argument. His suggested simple answers seem to rely on assuming that the spiders wed diagram is somehow accurate in the first place. There’s a saying that the only model of a complex system is the system itself. Complicated diagrams easily risk beguiling us that we’ve captured reality but we haven’t. He doesn’t fully justify his choice of three degrees of separation from the node he’s most interested in affecting, but he does allude to other research suggesting this is a reasonable move – but it is a simplification that he has inserted, not one that emerges beautifully from the apparent mess.
So whilst I admire the search for the simplicity on the other side of complexity, I think we what see in this example is actually the insertion of simplification. I have a hunch this happens a lot with visualisations. Part of what happens is that Eric uses higher production valued than the ugly powerpoint he starts with, so his answer looks and feels more attractive but the beauty may be more in the graphics than in the content.
If he’s saying that we can play with the model to generate possibilities for further consideration, then I’m less concerned.. but I think the effect is to create a feeling of certainty that’s not really justified.