Systemantics
A review of a book that has a lot to say about systems: how they works, how they fail, but most of all how we should build them
A couple of months ago, I don’t even remember where, I came across this principle called the Gall’s Law:
A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works and cannot be patched up to make it work. You have to start over with a working simple system.
At that time, I read it and made a lot of sense to me. I was starting to work on a project to introduce a Design System in my company, on the basis of a previously introduced Style Guide, so I saved it in my notes as a useful advice. But I didn’t fully understand the implications and how it would have impacted on my work a few months later.
Fast forward to one month ago, I was attending CSS/JS Conf EU when Ashley Williams — one of the speakers, during her talk A Brief History of Modularity — mentioned exactly the same law referring to the JavaScript ecosystem complexity and its modularity problems.
In one of her slides, she showed the cover of the book, and it struck me for its perfect design (the layout, the typography, the picture):
The same evening back in the hotel I looked online for the book and discovered that it was out of print. I was absolutely curious and eager to read it, and luckily enough I found a used copy on Amazon, so I bought it straight away. A few weeks later (the seller was in the USA) the book finally arrived, and the morning after I started to read it during my commute.
Something unexpected
The book is very small, a pocket edition with yellowed pages. It’s not long, a little bit more than 150 pages. I have a personal predilection for short and concise books (my philosophy when comes to authors and books is: “whatever you have to say, you can say it in 100 pages”) so I was already pleased by that.
The first surprise came as soon as I opened the book: on the very first page there were a name (“W Franklin”?) and a date: “21 June 1978”. Wait, what?!
Looking at the cover, at how it was modern in its design, I thought it was a quite recent book, so I didn’t actually look at when it was written. Suddenly I discovered that the book I had in my hands was at least 40 years old!
The second surprise came after the very first pages: the language and the tone were so unusual and facetious that it took me a while to understand if the author was ironic and was joking (even bullshitting the readers, at some point) or whether he was actually serious and really believed in what he wrote.
Just to give you an example: at a certain point he introduces the concept of Anergy as:
Any state or condition of the Universe, or any portion of it, that requires the expenditure of human effort or ingenuity to bring it into line with human desires, needs, or pleasures is defined as an Anergy-State. Anergy is measured in units of effort required to bring about the desired change.
Now we are in the position to state a Theorem of sweeping generality: “The Total Amount Of Anergy In The Universe Is Fixed”.
Now, taken at face value this assertion looks like the result of a distorted mind, right? I mean, is he really believing that there is such a thing as the Anergy? What is he thinking, that he has discovered a universal law like the Principle of Conservation of Energy or the Principle of Maximum Entropy? Crazy, isn’t it?
But at the same time, look at some of the words he uses: “expenditure of human effort or ingenuity”, “human desires, needs, or pleasures”, “units of effort”. Since when human ingenuity and pleasures have become units of measure? There is a hidden humorous irony behind it. He’s not taking himself seriously, to me it looks like that he wants to misdirect the readers in a sort of play, of Comedy of Errors.
Well, it didn’t take me long to enter in the mood of the book, to understand it required some kind of double level of interpretation to read it, and I totally fall in love with it.
The one quoted above is just one of the many disputable propositions and affirmations of the book: almost all the book is based on axioms and “universal rules”. The author doesn’t care to prove them in any way (on the other hand, since when one is supposed to prove an axiom?). He just enunciates them, gives them elaborated names and/or acronyms, briefly explains what observations and reasonings lead him to that conclusions, and moves on to the next one, building an entire compendium of laws and principles that rule every system (up to the entire universe!).
The author itself explains (only in the appendix of the book) why he has chosen this approach. Describing what the Axiomatic Method is, he says:
The logical and necessary approach to developing the science of General Systemantics. The traditional approaches of Observation and Experiment are clearly inadequate [bold is mine]; the former because progress bogs down in impenetrable swamps of data, the latter because experiments upon systems invariably distort them beyond recognition.
See, the author doesn’t bother to prove his ideas. He is above all of this. There is nothing that resembles Science in the book, at least how we all know it. Quoting again the author: “Euclid used the [axiomatic] method to advantage, so can we. And everyone knows how successful it was in the hands of Descartes”. Is he really considering himself at the same level as Euclid and Descartes? Of course (I think) not!
At that time when he wrote the book, John Gall was not an expert in Complex Systems or Systems Thinking or any other academic systems-related sciences. He was a pediatrician. His interests and research in systems-failures were a sideline.
But that’s the catch (and that’s what puts people off when reading the book, I suppose; I’ll come back to this later). Just because is not a canonical “scientific” book, or the author likes to play the role of the “father” of a discipline, or the text of all the book is a mixture of para-scientific terms and language on one side, and witty observations and ironic citations on the other side, it doesn’t mean there is no value in it, or truth in it, or we should just dismiss it as “pile of bullshit” (as a dear friend of mine said about it).
“A simple, basic primer that spells out the essential pragmatic facts of systems in the form if simple and easy-to-grasp axioms”, that’s how the author describes his book. Now that we know it, let’s look at what is actually written inside the book, and why it makes so much sense to me.
“No one, these days, can avoid contact with systems.”
Reading the phrase above was one of the first aha moments for me (and I was only at the introduction). It’s clearly obvious, especially nowadays. It’s so generic that is automatically true, self-evident. Even tautological I would say: every human society is a system, no one can avoid it. So what does the author actually means? Why is so important?
Well, for me is in the fact that implicitly John Gall has just declared the intent of his book: it is not an essay about systems (at least, not only), is a guide (his guide, from his perspective) for those who want to build, manage and organise systems. If the systems are everywhere, being able to understand and “control” them (we’ll see it is not possible, according to him) is extremely important. And in fact a few lines later he says “No one can afford not to understand the basic principles of How System Works”. Common ordinary people just live with/within the systems, while those who need to understand them are clearly those who want to control them, who need to work with them.
I have built systems for my entire career (complex websites, applications, and platforms before; more recently style guides and UI web frameworks). Now I’ll have to build an entire Design System that will (hopefully) impact on a whole company. This book is for me!
The second aha moment for me was a few pages later. This is somehow the base principle, once understood this assumption all the others somehow come consequentially:
“Systems In General Work Poorly Or Not At All”
or in other words
“Failure to function as expected is an intrinsic feature of the systems”.
Gall affirms, without showing any uncertainty or sign of doubt, that all the systems created by humans are intrinsically flawed. A bold statement, no doubts.
But if that is true, what one can do? Well, we see in a moment what the author suggests about approaching a system. But there is more to this than meets the eye. If that is true, it also means that if a system fails it’s not the fault of the one who has built it. Maybe, as Gall suggests, it’s the nature of the system itself.
For me, this has been a real revelation. After having read these pages, I realised that maybe I should have started to look at a system in a completely different (opposite) way: instead of looking at it as something to continuously try to perfect, and damn myself every time I was not able to have it working the way I wanted, I should have accepted its imperfection and get the best I could out of it. In other words, don’t look at what was missing or failing, but at what was working and functioning.
You have no idea how much liberatory it’s been for me: I was getting out of a period in which I have seen with my own eyes a system, on which I’ve worked almost one year, fail and crumble piece by piece. Being so emotionally attached to it, I could not see all the other parts that were actually working, and understand how I could have used them to rebuild the system in a different way. Changing perspective, looking at it as if it has always been “broken”, and not because of me but because of its “systemic” nature, relieved me of all the sadness and guilt I was undergoing.
The functioning of a system
The book then continues with other axioms and general laws like “New Systems Mean New Problems”, “Large Systems Usually Operate In Failure Mode”, “Complex Systems Exhibit Unexpected Behaviours”, “Systems Tend to Oppose To Their Own Proper Functions” and Gall explains — in his peculiar way — why things are so. He introduces concepts like Anergy (mentioned above), Operational Fallacy, Catalytic Managership, and The Fundamental Law of Administrative Workings (or F.L.A.W., see the irony?) but in doing so he reveals some pearls of wisdom like this one:
The Real World Is What Is Reported To The System
that may be an obvious observation as well but definitely struck a chord and resonated a lot with my recent personal experience.
In another page, at one point he tells:
“[…] the quality necessary for being elected president do not include the ability to run the country”
which is perfectly spot on when read today, isn’t it? :) or similarly:
When A Fail-Safe System Fails, It Fails By Failing To Fail Safe
which — I swear it — I read the same exact day in which the British Airways global IT system crashed (despite being “uninterruptible”) leaving for two days hundreds of flights grounded and around 75,000 passengers stranded and stuck in an airport.
Of course, one can dismiss all these observations as trivial, or simplistic, or to be lacking any foundation. I am sure that many of these concepts have already been discussed, debated, argued and counter-argued, and finally settled and “sorted out” in many other fields and disciplines, by many other people probably smarter and more qualified and known than Gall. I am thinking to General Systems Theory and Complex Systems Theory, to System Thinking and Modeling, to System Dynamics and Behaviors. And I understand that a lot of what Gall says can be explained by the simple fact that almost every system, when created by humans, is also as a consequence a complex system. But this, at least for me, doesn’t take away the value of what Gall says. On the contrary: his ability to expose these concepts in such a succinct and trenchant way makes them stand out and be easily understandable.
It is at this point that Gall introduces the observation (obvious, again) that “a simple system may or may not work”, from which he consequently deduces the law named after him:
In a recent conversation with Alla Kholmatova, she told me: “Gall’s Law, yes. That’s how living systems develop — from cell to organism. But interestingly, when we deconstruct, we should start with the big picture, not the cells.” which made me realise how fundamental this law is. And how important it is to pursue a living Style Guide or Design System (more on this later).
At this point of the book, you can see the two fundamental traits of a system: how it can work only as an evolution of an already existing working system, and how it never function as expected and intrinsically fails. And is the balance between these two aspects the interesting part of working with a system.
Working with a system
What does it all mean? That we should approach systems as non-working, as I’ve explained above, but at the same time work to get the maximum benefit out of them:
“[…] what can be done is to cope, and on rare and satisfying occasions, to prevail”.
And in fact, after going through more advanced concepts of systems-functions and discussing (again) the implications of systems-failures, Gall moves to more practical propositions trying to answer the question that runs in the reader’s head since the beginning: How to have a functioning system?
The great secret of Systems Design is to be able to sense what things can be done easily and elegantly by means of a system and what things are hard, and stay away from the hard things.
And just afterward he adds:
Systems Run Best When Designed to Run Downhill
that he explains in this way: “In other words, this means working with human tendencies rather than against them”. Maybe obvious (again) and easier said than done, I know, but recently I have found myself thinking more and more in terms of “pull, not push” and learned the hard way how to (try to) introduce changes and innovations in teams, processes, and companies.
In the last pages of the book Gall gives these other two pieces of advice:
Loose Systems Last Longer And Function Better
Do It Without A System If You Can
And they are both quite puzzling. The first one probably because of me being a perfectionist and (often) a control freak, and the idea of giving away “control” of a system scares me and makes me uncomfortable. The second one because is not just the classic “Keep it simple, stupid!” (K.I.S.S. principle). I sense there are deeper implications that I have not yet grasped yet behind these (obvious, again) declarations.
A controversial but enlightening book
I have spoken with more than one person about this book, and all those who have read it have also told me the same thing: they have started it, but after a bunch of pages they have put it down, essentially because it was just bullshit.
I totally get the feeling and understand where this impression comes from: the same happened to me when I started to read the book. Nevertheless, I have found the book very valuable and important for me. I would say life-changing (at least, until the next life-changing book). Probably it will be one of those books there every here and there I tend to read again and again.
Should you read it? Well, if you manage to get a copy definitely yes! Especially if you build or work with systems. And when I say systems, I mean every possible system. For me ‘a team’ is a system (a very complex one), ‘a codebase’ is a system (a complicated, often chaotic one) and of course, ‘a design system’ is a system.
I am building one of these, in these days, and reading this book have had a huge impact on how I am going to design, build, and look at it in the next months. Quoting Nathan Curtis: “A design system is a living, funded product with a roadmap & backlog, serving an ecosystem” and “A design system isn’t just WHAT’s made (style, components, etc). It’s HOW it’s decided, made, released, and used by people making products”. Simply put, a design system is a complex system, built by human(s) for humans and I am sure Gall’s book will be an invaluable resource for inspiration and guidance to me.
In conclusion, quoting John Gall’s obituary: this book is “a humorous but insightful study of the problems and pitfalls that afflict every type of system that human beings create — whether mechanical, economic, corporate, aeronautical, legal, linguistic, religious, medical or electronic”.
So go, find it, and read it: you won’t regret it.
At least, that’s my suggestion :)
But if that is true, what one can do? Well, we see in a moment what the author suggests about approaching a system. But there is more to this than meets the eye. If that is true, it also means that if a system fails it’s not the fault of the one who has built it. Maybe, as Gall suggests, it’s the nature of the system itself.
Originally published at didoo.net.
