Graphic by Mark Smith, Informal Education
I’ll get to Chris Argyris and "double-loop learning" in a moment, but first, what do I mean by "meta-work"? Putting it simply, meta-work is the work we have to do in order to work more effectively. Meta-work occurs anytime we step back from our regular activities to ask larger questions, like "Why do we do this task this way?" or even "Why do we do this task at all?" Meta-work is ultimately about challenging assumptions and not taking things for granted.
I started thinking about meta-work per se after reading Merlin Mann on various ways to improve personal productivity, but I’ve always been fascinated by the systems and tools we use to do our work–sometimes because I’m legitimately looking to solve a problem, and sometimes because research is a respectable form of procrastination.
Merlin’s Inbox Zero series encouraged me to do some serious meta-work and rethink how I manage my email. But my experience as a teaching assistant for High Performance Leadership and my ongoing interest in improving interpersonal skills has encouraged me to look at meta-work in a much larger context–and that thinking led me to Chris Argyris, professor emeritus at Harvard Business School and a major influence on such thinkers as Peter Senge.
One of Argyris’s key concepts is "double-loop learning," described very effectively by Mark Smith at Informal Education:
Single-loop learning seems to be
present when goals, values, frameworks and, to a significant extent,
strategies are taken for granted. The emphasis is on ‘techniques and
making techniques more efficient’ (Usher and Bryant: 1989: 87). Any
reflection is directed toward making the strategy more effective.
Double-loop learning, in contrast, ‘involves questioning the role of the
framing and learning systems which underlie actual goals and strategies’ (op.
cit.)… The former
involves following routines and some sort of preset plan – and is both
less risky for the individual and the organization, and affords greater
control. The latter is more creative and reflexive, and involves
consideration [of] notions of the good. Reflection here is more fundamental: the
basic assumptions behind ideas or policies are confronted… hypotheses are
publicly tested…(Argyris
1982: 103-4).
Double-loop learning, then, involves a substantial amount of meta-work and requires raising a number of potentially challenging questions: Not only "Why do we do this task this way?" and "Why do we do this task at all?" but also "What assumptions are embedded in our methods and our goals?"
As Smith notes, asking these questions can be risky–the phrase "opening a can of worms" comes to mind–but the potential reward is the possibility of being much more effective (and, I’d argue, much more fulfilled) because such questions not only create opportunities to do things in more innovative ways, but also because they challenge us to think about finding entirely new things to do.
It’s easy to come up with all sorts of reasons not to do our meta-work, starting with: We’re too busy! But that begs the question: Busy doing what? If all your learning is single-loop, at best you’re making marginal improvements in existing routines. More on double-loop learning and personal development tomorrow.
UPDATE: More on Double-Loop Learning and Executive Coaching
6 Responses
Help me out on this. I have read some Chris Argyris, and heard him speak a little, and have found wisdom on what he says–but it always seems like it could have been said more simply. Or so it seems to me.
What’s the difference between “double-loop learning” and “learning from your mistakes?”
Apparently it’s the depth of learning–do you examine fundamental assumptions, or are you just incrementally making improvements? This distinction sounds great, but in practical life it sometimes feels less clear.
When Tiger Woods practices, which is he doing? Is he doing single-loop learning when he hits 300 shots, and double-loop learning when he re-tools his entire swing? Or is he thinking about both all the time?
This is a question also raised by Geoffrey Colvin in the October 30,2006 issue of Fortune, where he discusses a British study that suggests success comes from practice, not from talent.
In Colvin’s article, he suggests the distinction lies between just “doing” things and doing things to “get better at them.” He suggests that “getting better” starts out easy, then gets harder as the low-hanging fruit is picked. The way Colvin talks, “getting better” is a continuum, unlike Argyris’s single/double dichotomy.
Colvin’s thesis might imply that higher achievement depends on mastering the basics. But there are other areas in life where an ability to see the big picture is sometimes enhanced by a distance from, even an ignorance of, some of the basics. Does double-loop learning work on the back of single-loop learning, or are they separable in some practical way?
Any thoughts? I’m being honest here, I’m confused.
Help me out on this. I have read some Chris Argyris, and heard him speak a little, and have found wisdom on what he says–but it always seems like it could have been said more simply. Or so it seems to me.
What’s the difference between “double-loop learning” and “learning from your mistakes?”
Apparently it’s the depth of learning–do you examine fundamental assumptions, or are you just incrementally making improvements? This distinction sounds great, but in practical life it sometimes feels less clear.
When Tiger Woods practices, which is he doing? Is he doing single-loop learning when he hits 300 shots, and double-loop learning when he re-tools his entire swing? Or is he thinking about both all the time?
This is a question also raised by Geoffrey Colvin in the October 30,2006 issue of Fortune, where he discusses a British study that suggests success comes from practice, not from talent.
In Colvin’s article, he suggests the distinction lies between just “doing” things and doing things to “get better at them.” He suggests that “getting better” starts out easy, then gets harder as the low-hanging fruit is picked. The way Colvin talks, “getting better” is a continuum, unlike Argyris’s single/double dichotomy.
Colvin’s thesis might imply that higher achievement depends on mastering the basics. But there are other areas in life where an ability to see the big picture is sometimes enhanced by a distance from, even an ignorance of, some of the basics. Does double-loop learning work on the back of single-loop learning, or are they separable in some practical way?
Any thoughts? I’m being honest here, I’m confused.
Help me out on this. I have read some Chris Argyris, and heard him speak a little, and have found wisdom on what he says–but it always seems like it could have been said more simply. Or so it seems to me.
What’s the difference between “double-loop learning” and “learning from your mistakes?”
Apparently it’s the depth of learning–do you examine fundamental assumptions, or are you just incrementally making improvements? This distinction sounds great, but in practical life it sometimes feels less clear.
When Tiger Woods practices, which is he doing? Is he doing single-loop learning when he hits 300 shots, and double-loop learning when he re-tools his entire swing? Or is he thinking about both all the time?
This is a question also raised by Geoffrey Colvin in the October 30,2006 issue of Fortune, where he discusses a British study that suggests success comes from practice, not from talent.
In Colvin’s article, he suggests the distinction lies between just “doing” things and doing things to “get better at them.” He suggests that “getting better” starts out easy, then gets harder as the low-hanging fruit is picked. The way Colvin talks, “getting better” is a continuum, unlike Argyris’s single/double dichotomy.
Colvin’s thesis might imply that higher achievement depends on mastering the basics. But there are other areas in life where an ability to see the big picture is sometimes enhanced by a distance from, even an ignorance of, some of the basics. Does double-loop learning work on the back of single-loop learning, or are they separable in some practical way?
Any thoughts? I’m being honest here, I’m confused.
Very thought-provoking, Charles–thanks. First of all, we should bear in mind that most academics could be more clear or straightforward, and Argyris is probably no exeception 😉
But I believe you’re looking in vain for a set of heuristics that will allow you to consistently distinguish between single-loop and double-loop learning. The distinction between the two processes is relative depending on who’s doing the analysis and what preconceived notions they bring to the table. The key to double-loop learning is asking whether there are any unchallenged assumptions built into your current goals and strategies, and whether it’s possible to pull back and include those assumptions in the frame of your analysis.
In most circumstances, the learning we undertake is aimed at improving our performance relative to a set of goals and other factors that are taken for granted. Feedback from our performance (or “learning from our mistakes”) typically cycles immediately back into our analysis of the strategies, tactics or techniques that led to our performance. This is important work, but it’s inherently limited by those initial factors that are taken for granted at the outset and that remain unchallenged by an assessment of the performance results.
In contrast, if we can pull back and expand the frame of our analysis, we begin to call into question some of the factors that we usually take for granted. Our performance results aren’t simply used to assess the strategies that have been derived from those factors–they question the factors themselves.
Think about the email management meta-work I mentioned in the post above. Prior to reading Merlin Mann, I filed all the email messages I saved in a series of nested folders, just as I do for other documents. The process of creating and maintaining a complex file folder system was taken for granted. And in trying to manage my email more effectively, I essentially limited myself to revising the strategies that were built on the assumption that a complex file folder system would always be present. I never stopped to challenge that assumption, and all the improvements I made in my email management were limited by that predetermined factor. Single-loop learning.
Reading Merlin Mann caused me to take a big step back and challenge the fundamental assumption that messages should be stored in a complex file system. “Desktop search changes everything,” Merlin said. “Scrap your nested files and keep all your messages in a single folder!” In a small way, it was a frame-breaking realization, and a great example of double-loop learning.
(I think an additional insight of Argyris’s was to note the interpersonal dynamics and organizational conditions that reinforce single-loop learning and make double-loop learning so rare.)
Hope this was helpful–thanks for inspiring some further thinking on the subject!
Ed
Great post Ed and nice conversation piece as well. I’ve been neglecting my news reader for a while and see I’ve missed some great posts…
I’m totally fascinated with the framing of “double-loop” learning and your link to metawork. I think I’m going to have to pick-up that book and have a go at it myself.
As I read your post, my mind kept wandering to triple-loop learning, i.e. “How do we decide what is right?” whereby the context in which one’s assumptions exist actually gets challenged and explored. On that, I’m curious to see how it might fit with the metawork concept… Could this be meta-metawork?
I’ve been an avid reader of your blog for a while now without ever posting. So, I’ll take this moment to thank you for the great work you share.
Be well.
Dan
Hi Dan,
I’m sorry for the delayed response to your very gracious note. I took about 6 weeks off from blogging right around the time you commented and only got back to it today. But thank you! It’s much appreciated.
But meta-meta-work? I think that’s one meta too far 😉 In a sense, though, once you identify the assumptions underlying your goals, values and strategies, you’re likely to integrate them, if you’re having any success at all–and then you’re (more or less) back in a single-loop mode. To inquire further is to return to double-loop learning.
So I actually don’t think of it as wheels within wheels, ad infinitum, but as oscillating between two different modes of inquiry–the single-loop state, where we take much for granted, just to be able to live and work fluidly, and the double-loop state, where we come to a stop of some sort, and peek behind the curtains.
Thanks again!
Ed