Re: [IxDA] Re: [IxDA] Design is not a “problem-solving techni que”

6 Jul 2010 - 10:41pm
824 reads
Christopher Fahey

On Jul 6, 2010, at 8:27 PM, David Malouf wrote: > Abductive = gut? hmm? never equated it that way. Abductive reasoning does have an approach of validation to it.

Sure, but the "validation" part of abductive reasoning comes from the gut. Don't knock the gut! And, to mix a metaphor here, let's not try to put lipstick on it either with words like "validation", as if that makes a gut decision more legit -- guts are guts.

As much as we might want to believe abductive reasoning is intellectually sound just like deductive and (to a lesser extent inductive) reasoning are, it's really not not even close to being a truly "logical" way of thinking. An abductive approach always (by definition!) falls back on a situation where there is more than one valid answer (sometimes an infinite number of valid answers) and the thinker/designer chooses one based on what we have to admit is nothing better than "gut".

I'm not arguing an opinion here, either: Every definition of abductive reasoning includes an element of choosing a solution without a certain logical reason for that solution.

You can use probability, heuristics, qualitative criteria, experience, rules of thumb, etc, to formulate solutions, but ultimately all of these flavors of abductive reasoning fall back on best guesses, hunches. Your gut.

No sense in running from it. If you don't endorse gut calls from demonstrably good designers, then you're simply not endorsing abductive reasoning as a design method (so I suppose that means you're also not endorsing design thinking).

Where abductive evolves into inductive and then to deductive is when you take your awesome abductive/gut design choice that you pulled out of your smart ass and subject it to some sort of comparative testing. Which as we all know (a) doesn't always happen, (b) probably ought to happen more than it does, and (c) is not always necessary or worth the cost.

Also, to be clear, Jared is right that abductive isn't the only game in town -- good design can come from purely inductive and deductive reasoning, too, especially if you are evolving an existing system. In such situations, design is truly "problem solving", just like an algebra equation or a crime scene. Whether such approaches are worth their expense compared to a good designer's gut in combination with good testing (i.e., abductive+deductive), well, that's another question entirely.

Cheers, -Cf

Christopher Fahey Principal / User Experience Director

Behavior biz: me:

Syndicate content Get the feed