Complexity versus Lean versus Agile versus Me

While the experts are battling over each others’ ideas, I remind myself that all models are useful (though some fail faster than others).

Dave Snowden is a complexity researcher who has repeatedly attacked systems thinking in his keynotes, saying that it assumes there is an ideal future that it tries to realize, which he considers the wrong approach.

John Seddon styles himself a systems thinker, and I’ve heard he said Snowden misrepresents systems thinking. But then, in his own keynotes, John Seddon has attacked Lean, saying “I hate Lean. It destroys organizations!”

It is no surprise that the Lean community claims Seddon totally misrepresents Lean. But I’ve noticed some of those same people have been busy attacking Agile, saying that Agile sub-optimizes at the team level, and Lean doesn’t.

Of course, some Agile experts say that this is a misrepresentation of Agile, and that Agile scales quite well to the entire organization. But at the same time Agilists have been attacking traditional project management.

Glen Alleman, project manager and prolific blogger, has repeatedly called bullshit on Agile experts. He claims that the attacks from the Agile community are often just misrepresentations of proper project management.

Complexity Thinking in the Middle

And now we’ve come full circle.

Both David Snowden and Glen Alleman have claimed that I am misrepresenting their favorite ideas. Though, interestingly enough, both have said this from completely opposite positions. (It’s as if Wall Street and Occupy both agree that I’m wrong, for exactly the opposite reasons.)

And so I find myself in the middle.

I consider myself a complexity thinker. I find all perspectives interesting.

All models are useful, though some fail faster than others.

It takes multiple perspectives to figure out which idea fails where and when.

And, while a new battle in the war between Scrum and Kanban has just started, I prefer a view from above and in the middle. I learn most by watching the battle of ideas unfold beneath me. Though it means I have to take a bit of heat from all sides. But never that much.

It’s actually nice and warm up here.

The view is great too!

p.s. Being human I also might have misrepresented a thing or two in this blog post. I suggest you deal with it.

(Jurgen Appelo is author of Management 3.0, a best-selling management book for Agile developers. It has a picture of a monster in it.)

Related Posts
free book
GET MY FREE BOOK!
“How to Change the World”
  • http://profile.typepad.com/galleman Glen B. Alleman

    You at time co-op ideas without attribution.
    Like the quote “all models are useful…” is a rearrangement of George E. P. Box’s statement “all models are wrong, some are useful,” without acknowledging Box, “Robustness in the Strategy of Scientific Model Building” (May 1979)
    I’ve come to believe it’s just style not intent, but it will hopefully improve over time to acknowledge that “remix” and “tweaking” has limits.
    As Agile moves from being personal to being an enterprise process, the notion of an integrated system built on governance and systems will become more common, http://info.managementconcepts.com/agile_social
    But no problem, the discussion process is evolving as well, and repeating the ideas of others – or making inverted assertions of obvious mis-applications of well established processes will work its way out in the market place.
    The recent webinar, where your book was clearly referenced to ~300 federal and contractors looking for advice on applying agile to federal IT programs is an example. In that webinar the 12 Principles of Agile were used to show the inversion concept – “no one would even NOT want to implement those 12 agile principles on a program. They are simply good program management principles.”

  • http://profile.typepad.com/leanlooney Patrick Verheij

    People, in general, are biased. We tend to cling to a viewpoint once we understand it…or at least *think* we do. Once we’re up on the highest levels of Bateson’s pyramid, we are stuck.
    It is most obvious in the everlasting battle between the religions. Now agile, lean and the like have become religions too and hence we smash each other’s heads for not adapting to our point of view.
    On one side, we need these battles because we learn from them. On the other hand, we should make a treaty now and then, step over our differences and find something else to fight over.
    Alas, it doesn’t matter who the enemy is. Once we beat one, we search for another.
    So get ready for another round while I grab my popcorn and a coke 😉

  • http://seddonwatchblog.blogspot.com SeddonWatch

    If John Seddon hates Lean, then he somewhat hates himself. He used to do all of his work under the banner of “Lean Service,” until he decided it was better marketing to bash the Lean movement and to portray himself as the smarter-than-thou savior.
    His act is getting old.
    http://seddonwatch.blogspot.com/2010/11/john-seddon-site-admits-systems.html

  • http://www.linkedin.com/riccardobua Riccardo Bua

    I think some models need to work, thinking all of them fail is once again a religion, I consider the human element the weakest link and the one that would eventually cause the system to fail, but then we reached the moon and got back in one piece, so somehow some models worked.

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    I think you misunderstand. There is _no_ model that is always right. That’s not religion. That’s fundamental science. All models are context-dependent. They work in certain circumstances.
    The physical models they used to get people on the moon _don’t_ work at the quantum level.
    Therefore, ultimately, all models fail.

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    Glen,
    I have attributed George Box in my book. I also acknowledge him in my new Complexity Thinking slides:
    http://www.slideshare.net/jurgenappelo/complexity-thinking
    In fact, I get _compliments_ from people for the sheer amount of references I have in my book and presentations. Of course, my mind might slip once or twice. But not this time. The quote you refer to in this case is mine. No attribution is necessary.
    Let me take this opportunity to point out that you repeatedly refer to and complain about “self-styled gurus” without explicitly mentioning names or refering to examples. I would think some references are in order there as well…
    Thanks for promoting my book!

  • http://www.cognitive-edge.com Dave Snowden

    I’m not sure if my earlier comment just didn’t get posted or was deleted, so assuming accident let me repeat it.
    – I have not said that systems thinking is wrong, I have said it has methods and tools appropriate to complicated but not complex systems
    – I think you misuse the language of complexity theory (Cynthia made the same point to you) which causes confusion. Your 3by2 model for example. Its your call if you do that but expect some criticism
    – If you really want to take the post-modernist stance that all models fail, then it may be no surprise that you pick up and use multiple models (with different levels of understanding). However that approach does not really justify you arguing for a Management 3.0 which is science based.
    In britain we have a phrase – you can’t have the penny and the bun,

  • http://www.linkedin.com/riccardobua Riccardo Bua

    I think you are back to one big assumption that somebody/something would apply a fly_to_the_moon model to the quantum level.
    One model fits its need and purposes up until the human element changes those conditions.
    I found this quote that I like for this kind of scenarios:
    “The fundamental issue is circularity: embedding one’s assumptions as foundational “input” axioms in a model, then proceeding to “prove” that, indeed, the model’s “output” supports the validity of those assumptions. Such a model is consistent with similar models that have adopted those same assumptions. But is it consistent with reality? As with any scientific theory, empirical validation is needed, if we are to have any confidence in its predictive ability.”
    Hence if you take your position you would need to prove that all of them fail eventually even in their own scenarios, the ones that they were built for…
    I love Nietzsche, but your stand is too nihilist on this as somebody else commented as well :-))

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    Again, you misinterpret my meaning.
    I do _not_ assume that someone will try the fly-to-the-moon model to the quantum level. I just mean that people _don’t_ apply the model there because it doesn’t work there.
    I mean exactly the same as what you’ve been saying. Every model fits its need and purposes. But it will fail outside those boundaries. It’s just hard to determine where the boundaries are, but ultimately all models fail. That’s all.
    Seriously, I think we mean the same thing. We just use different words. This _is_ part of sense-making.

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    Hello Dave,
    Thanks for your comment!
    I have noticed you have been adapting many complexity terms in your keynotes and writings to serve your own purposes. Systems thinking, chaos, self-organization, emergence, fitness landscapes. Every time your definitions differ from the consensus I have found among experts in complexity literature. Of course, you are free to do so. But it doesn’t make it easier for simple people like me to make sense of it all.
    Now the question is _which_ complexity language do you think I misuse? Do I misuse the mainstream language? Or do I misuse your distorted and branded language?
    I’m puzzled.
    Jurgen
    p.s. You _misuse_ Cynthia’s name and feedback. What she actually wrote under my 3×2 model is this:
    “Hi, great post and great examples. I’ve been thinking about the same issues and have come up with remarkably similar solutions – great minds think alike”.

  • http://www.cognitive-edge.com Dave Snowden

    Cynthia also said very clearly that if you chose to use language in a different way from everyone else then you would get confusion so you are, as ever, being very selective in what you quote.
    Otherwise, re your fist paragraph I’m not impressed and I had thought better of you. I don’t remember many arguments when I last presented at Santa Fe. John Childs and Brian Arthur were also highly complementary at a specilaised event in Singapore this October. I spent some time with Kauffman on our use of fitness landscapes a couple of years ago; over dinner with Brian & Walter Freeman as it happens, they had also both sat through a half day seminar I gave introducing the subject which was intimidating but interesting.
    I think I’ll take their perspective on my use of language rather than yours, especially given your responses the first time I challenged some of your views at that event in Finland. The praise of the praiseworthy and all that, along with its corollory apply here.
    Oh, and as I remember it the HBR article with all that “distorted and branded language” got two Academy of Management awards. Its called peer review. Its a way of testing your understanding of a subject. I commend it to you, its not enough to have read many books, you also need to publish in a peer reviewed environment if you want to speak authoritatively on a subject
    Otherwise I will complete my review of your book next month and will summarise the issues as I see them there. It will be in a blog and well tweeted, it will also be a considered response. You might also want to look out for a blog next week on my concerns about the way the training/thought leadership market works in the whole Agile/Lean/Scrum ecology.
    Oh and you do yourself a disservice, you are far from simple, but you may well be puzzled.

  • http://www.cognitive-edge.com Dave Snowden

    Jurgen, your response here is a little disappointing.

    • You have a tendency to be selective in what you quote and this is evidenced by your reference to Cynthia. She made the point that you were using language in a very different way from other people and that it had the potential to cause confusion. Yes she said some nice things as well, but you need to pay attention to the more problematic comments as well as the nice ones.
    • Comments like “distorted and branded” language are a little silly. For a start the HBR article containing that language has won two awards from the Acadamy of language and has been published in several peer reviewed journals. That is the way you test your understanding of the subject. You might want to attempt to do that yourself.
    • Aside from peer review I don’t remember any similar comments when I presented at Santa Fe earlier this year. Also at a specialist event in Singapore in October Brian Arthur and John Child were with me and were not only not critical but complementary. A few years earlier I talked through our use of fitness landscapes with Stu Kauffman over dinner, along with Brian and Walter Freeman. The latter two has just spent the morning attending a seminar I ran as an introduction to the subject. That was intimidating, but again I don’t remember their commenting that I was distorting the language.

    In Tolkein’s Two Towers there is a wonderful statement by Faramir: The praise of the praiseworthy is above all rewards. Of course the opposite also applies. I think I can live with your somewhat snarky criticism.

    Otherwise I will provide considered comment on your use of the language when I review your book next month. It will pick up on concerns I expressed when I first heard you speak at that conference in Finland. You may also be interested in a blog I will post next week on the necessary interaction between theory, practice and training for aspirant thought leaders.

  • http://www.cognitive-edge.com Dave Snowden

    I reposted a revised version as I could not see the original, feel free to delete the first
    Its not a flame war, if you want to be critical in public expect a response. In a flame war I would be far more polemical in my response, but I really don’t see the point. I’m trying to give you some advise here.

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    I’ll take this off-line, since a public flame war is no use to us.

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    I decided to reply to Dave Snowden privately. I have great respect for his work, but it turns out we disagree on the appropriate ways to communicate on our own work and how to criticize the other.
    For now I will consider the matter closed.

  • http://www.systemswiki.org/ Gene Bellinger

    Jurgen,
    Multiple perspectives absolutely.
    And as I commented in the discussion where someone posted a link to this blog, “Isn’t it appropriate to employ critical thinking whenever one encounters someone selling something?”
    be well,
    Gene

  • http://www.appreciatingsystems.com Nicolas Stampf

    Excellent article!
    The small comment wars are… interesting? And amusing too 🙂
    Please continue postings!

  • Eirik M

    Jurgen, have you actually listened to what John Seddon says? He is not attacking lean as such, what he criticises is the fact that it goes by a name (because then managers think it comes in a box).
    I had for a long time associated Lean with what I read in books on Lean Software Development. Only recently did I learn that some consultancies (not software) teaches “lean techniques” to managers, and in the end it turns out to be a euphemism for downsizing. I think lean is good, but I avoid using the word.
    I think Seddon has some really good points. It would be good to hear some more opinions 🙂

  • http://profile.typepad.com/jurgenappelo Jurgen Appelo

    Yes, I totally understand that John Seddon’s issue was with the use of the word, not with the original thinking behind it.

How to Change the World - free Workout - free
CLOSE