The process “giving feedback” is outdated. Or limited, at least.
Let’s do a post-mortem on a post-mortem, to find out why…
My friend Benjamin (who works with me on both Complice and the Upstart Collaboratory culture project) and I had an experience where we were making some nachos together, and… long story short: most of them burnt. We then spent a bit of time debriefing what had happened. What was the chain of events that led to the nachos burning? What had we each experienced? What did we notice?
This yielded a bunch of interesting and valuable observations. One thing that it did not yield was any plans or commitments for how we would do it differently in the future. Anything like “So when [this] happens next time, I’ll do [that], and that will act as a kind of preventative measure.”
Given that lack of future plans or commitments, one might ask: was it a waste of time? Did we not really learn anything? Will things just happen the same way again?
It depends. There are some circumstances where that might be true. But increasingly, the situations we face are more and more VUCA: Volatile, Uncertain, Complex, and Ambiguous (an acronym originally from the military). And the VUCA nature of things means that things wouldn’t happen quite the same way again anyway—even without a post-mortem or plans. So any plans would themselves have to be complex and contingent on a multitude of factors—otherwise having made a specific plan or commitment might actually impede effective action in the future, because of the unpredictable way in which the future would end up being different.
But furthermore, I want to question the assumption that plans or commitments are required for a post-mortem to be effective. (I’m aware that I’m the one who brought in that assumption. Maybe you, the reader, already don’t hold it! Or you sort of did but you’ve readily dropped it now that it’s under question. Well, this is an assumption that my past self from a few months ago was sort of making, so it seems probable that there are yet people out there making such an assumption.)
What Benjamin and I did in that debrief of the nacho incident was we trained our collective attention on various details of the incident, and made connections between things we’d noticed that we hadn’t made at the time. We don’t know quite what will happen in the future, but by individually and collectively pointing our attention at what we noticed (and what each other noticed) we create new possibilities for what we’ll notice in the future. We also have more collective capacity because we have a better ability to predict what each other will notice in the future.
And this process can include constructing causal models, or considering counterfactuals, or whatever, but in complex systems it can be hard to post-dict what “would have happened” if someone had made a particular different decision. Besides, noticing something new or understanding something differently is ultimately upstream of making a different decision in the first place. So we still need to train our noticing and understanding.
In general, humans are complex systems, so systems involving human interaction are complex systems. They can be approximated otherwise, but this too becomes less true as culture becomes more fragmented and as we deliberately try to escape from the control-and-predictability-driven structures of how-things-have-been. Which is what we’re doing, and is also what anyone trying to innovate is also doing, even if they don’t realize it.
In addition to helping us collectively integrate our learning from this incident, this debrief served another function: to tap into our sense of being on the same side about it, and our shared desire to coordinate effectively. This builds trust.
But it’s not just about post-mortems, but the whole concept of “giving feedback”.
“Feedback” as a basic systems concept is actually so important and fundamental that its its own thing. And the word for that is still feedback. That kind of feedback is any kind of information or system dynamic that flows in a loop. So a control system (like an automatic steering system) uses negative feedback: if the car veers to the left, it steers to the right to correct. And other systems have positive feedback: the more people share this blog post (hint, hint) the more people will see it, and the more will share it. So this is everywhere. If you give a presentation to a live audience, there’s already a ton of feedback available to you if you pay attention to their faces and the sounds (subtle or overt) they might make.
But explicit “feedback” as in “I have some feedback for you” is broken. And my recommended replacement is the kind of thing that I gestured at above: pointing our collective attention at each of our experiences. I’m currently calling this “co-attending”. So the point isn’t to tell the person what they did wrong, or what they did right, per se, but to more deeply appreciate the nuances of what happened, so that in future situations our general capacity is higher and we’re able to respond effectively together. Learning happens via attention. Group learning happens via co-attention.
Another issue is that “feedback”, particularly “negative feedback” is easily confused with criticism: negative feedback that has a quality (subtle or overt) of blame, or shame, or judgment, or punishment. (Or in the positive case, with praise.) These confusions get in the way of learning. Further elaboration in this deconstruction of criticism.
Trading feedback is a way for each individual to learn from what the other noticed, but it’s not enough to increase our multi-person system’s collective capacity. If we’re going to be able to work effectively together, we need to improve not just our individual performance but our attunement, and doing that means putting what we’re noticing in the center so we can look at it together.
That’s a huge part of what’s different about this is. It’s not “I’ve got feedback for you”. It’s “I’ve got a tune-up for us“.
Another phrasing shift: after some experience of working with someone, instead of asking “do you have feedback for me?” ask “what did you notice that might be worth looking at together?”
(There’s also “critique”, about which you can watch a really cute video. Critique, as a form, requires the critiquers to make specific suggestions for how the thing-being-critiqued could be improved. This is very valuable in technical situations where it’s the output that matters, but in complex interpersonal situations where the output is being generated by some underlying process that is profoundly buggy, giving critique on the level of the output may not really be that relevant. Instead, helping someone to pay more effective attention to what might be generating things can be much more useful.)
(Also, here’s a great article on blameless post-mortems that also focuses on revealing “what did I do? what did I notice?” versus telling someone else “what you should have done”.)
In summary:
Expired: Blame & Criticism
Tired: Giving Feedback
Wired: Co-Attending & Attunement
Future posts will generalize this:
(You are, of course, invited to consider these generalizations an exercise for the reader at present. )
Constantly consciously expanding the boundaries of thoughtspace and actionspace. Creator of Intend, a system for improvisationally & creatively staying in touch with what's most important to you, and taking action towards it.
Decius » 23 Apr 2017 »
>”We don’t know quite what will happen in the future, but by individually and collectively pointing our attention at what we noticed (and what each other noticed) we create new possibilities for what we’ll notice in the future. We also have more collective capacity because we have a better ability to predict what each other will notice in the future.”
That falls under the paramilitary aphorism “Plans are useless, but planning is indispensable”. By becoming aware of what the intended outcome (Nachos!) is, and how to handle a few of the things which are known to be likely to interfere with that, the agents involved are primed to be able to respond to unanticipated events and move the world towards the goal.
In symmetrical competitive contexts where prediction is advantageous and the players are skilled, the /state of the game/ is anti-inductive. The debrief there can consist of “These are the errors in implementation we identified, and this is what we believe the state of the game was, and this is what we our available information revealed about the state of the game this iteration, and now let’s model what our opponents learned and guess how they will adapt…
Have your say!