Roger Martin, Integrative Thinking and the Opposable Mind

I read Roger Martin's epic little 2008 book 'The Opposable Mind' at the same time as I was working through Sam Savage's 2012 book "The Flaw of Averages". Both authors used a similar analogy: the human hand.

Martin talked about the importance of the opposable thumb-and-forefinger as a master metaphor. Savage came up with the idea of a 'mindle' - like a handle, but for the mind. Since then I've collected a lot of mindles but all the while a part of me wanted to go back to The Opposable Mind and diagram it.

Re-reading Martin, I realised I'd internalised many of the ideas but I still couldn't summarise them well. When ideas are both simple and hard my rule of thumb is there'll be much wisdom contained within. It took a while but I managed to get it done.

First up, here's his cascade model. The thinker starts at the bottom, ladders up then ladders back down. He identifies a default thinking stance to contrast with his ideas. The default thinker is a mythic beast; but Martin does teach business for a living and is part of that little Canadian cabal that doesn't think much of the MBA thinker.

The first choice facing the thinker is salience: what's important to the decision problem. The default thinker limits the dimensionality of the problem space while the integrative thinker expands it.

I started getting early breakthroughs in my data governance work when I began expanding dimensions past the things I was told to care about (usually a form of the privacy vs security false dilemma).

There's still a decent chance that the two thinkers will align at the first stage. But at the second stage they diverge wildly. It's taken nearly two years for me to work through the next layer in causal modelling and it's led to a ton of great insight. The problem is now when I look over the fence at the default thinker all I see are logical deficiencies and formal fallacies.

I started digging into the default thinking mindset to look for something I could crash-start or swap in to help break the deadlock. But it got fairly depressing quite quickly as every time I peeled back a defective model, framework or premise I'd just find three more.

The third stage on the integrative thinker side leads you into design, art and math. All useful things to start bring holism into your work. Being able to visualise things turns out to be a super power humans have and machines don't. Why we haven't figured this out properly I don't know. On the other side of the fence are a set of classification structures often involving big arrows, circles that circle but don't go anywhere and the 'story on a page'. Nuff said.

At the fourth stage the integrative thinker has to use that opposable mind and come up with a third way. On the other side of the fence we can see a race to the preferred option, references to wicked problems and trade-off language. When I dig into the reasons, they're often dignified excuses or vague justifications. No wonder it's only now that we're starting to surface some decent data governance thinking.

But that's not the main model in the book: it's the stance-tools-experience structure that was the most interesting and which I couldn't visualise. Thus:

Ignore the green box: that's me mapping Martin to Floridi's information resource product target (RPT) model. Another post. Also ignore the causal arrows: I'm still working through them, figuring out the feedback loops between the three elements.

For stance, Martin distinguished between 'contented model defense' of the default thinker and the 'optimistic model seeking' of the integrative thinker. Setting your stance right and using a decent set of tools turns out to be really important. I haven't mocked it up yet, but my spidey sense tells me that a contented model defense stance will beat optimistic model seeking tools.

In other words, you probably can't fake it till you make it. You need to dig down into those six rules of opposable thumbware. I've seen people just get to the fourth one (better models can be bootstrapped into life) and stop there. I think it's something to do with the contented model defense. That and to get past that point means you need a bunch of things in place already (like knowing what evidence is). Aka this stuff is hard.

With tools, causal models were noted in the cascade, assertive inquiry is a specific questioning technique and the radial metaphor leverages analogical thinking, which isn't something people tend to get hired for. Oh dear, success seems vanishingly hard doesn't it.

Once we get to experience, I'm swapping out these terms for ones from Cesar Hidalgo "Why Information Grows" which says much the same thing but adds crunchy stuff like entropy. That's for later too, but the important bit is the originality piece. Contented model defenders just aren't going to rate well here. Mastery - yes and YouTube, TED, LinkedIn has a ton of stuff for that. But originality? That requires imagination and, like analogical reasoning, it's often not something people hire for.

And that's it. A few more diagrams fell out but these two are the main ones. I've got some more tinkering to do and will update this when I get there. Thanks for reading this far.