I have been using information technology all my working life; systems analysis (of the kind used in business) has become second nature to me. This is a short overview of systems analysis and thinking with some comments of my own along the way.
What are systems?
Systems are made of components; each component is capable of doing something that allows it to interact with other components – the component has inputs, does something, has outputs and is subject to feedback in the sense that is outputs may modify the future state of its inputs. A component can of course be thought of as a system as well – when these are high level they may be referred to as sub-systems. In any systems view we need to consider what are the sub-systems and how do they interact to make up the whole – these form networks and can be nested like Russian Dolls.
Talking about components is mechanical, we can see the physical and living world in systems terms. We can view the cell as a system and a component, within part of the organism, say the liver and on up to a whole person. We can look at pretty much anything through the lens of systems - plants and trees, the sea, the weather, earth itself.
Network and hierarchy
It has become fashionable lately to challenge hierarchy and talk about networks. My view is that both are true. The reason for this is because of nature in the main part of this site. The short version is that all groups have a leader - this may be just the person the group defers to or it may be formal, chairmanship, president. The view I elaborate in the main sections of this site it that we can flatten the hierarchies that we have substantially and change the paradigm of leadership to facilitation. However it is unlikely that we will get rid of them altogether.
Large and small systems
If we consider the world with a systems view a good starting point is the small area referred to as the “bio-sphere”. The bio-sphere is the area we inhabit – what goes on here potentially impacts everyone. It is sufficiently enclosed, and self-contained for us to treat it as the basis for the analysis in this essay Note: The Biosphere is a Sub-System . The Biosphere receives inputs e.g. sunlight, has outputs e.g. radiated heat, contains a huge number of number of sub-systems that change things internally e.g. the weather – more or less sunlight is experienced by us on the surface, and is subject to feedback – the amount of rain is determined (in part) by the amount of water in the atmosphere which is influenced by the rate of evaporation linked to the amount of sunlight at the surface Note: Simplification
An example of a complex system which is part of a wider system is the carbon cycle because carbon in the atmosphere influences the weather.
In the picture of the carbon cycle the black numbers indicate how much carbon is stored in various reservoirs, in billions of tons ("GtC" stands for GigaTons of Carbon and figures are circa 2004). The dark blue numbers indicate how much carbon moves between reservoirs each year. The sediments, as defined in this diagram, do not include the ~70 million GtC of carbonate rock and kerogen Note: The Carbon Cycle .
At the other end of the scale we can look at an individual cell. The cell takes nutrients from the environment, transforms them in order to maintain itself and gets rid of waste (see Capra, Chapter 7)
System Characteristics
Systems have predictable characteristics; they have boundaries, inputs, outputs, have internal processes, create outputs and receive feedback, they have regulators as well as emergent properties (the whole being greater than the parts).
The implications of this for how we think about the political economy are profound. We need to develop institutions and processes that allow feedback, learning and look to create virtuous circles.
Many businesses have already understood this and recognise that change management involves rapid feedback to reinforce what works; this is often referred to in the literature of continuous improvement in learning organisations.
Emergent properties
We are all familiar with a murmuration of starlings – this collective behaviour cannot be predicted by looking at individual starlings, it emerges from (is an emergent property of) the collective behaviour of a the flock of starlings and is observed from outside; in systems terms the behaviour needs to be studied by looking at the system as a whole.
Life is an emergent property of the particular organisation of molecules “biological life is chemistry only…we also state that life as a property cannot be reduced to the properties of the single chemical components” Capra and Luisi, 7.2.3 p133
Human activity systems
When we consider what have been referred too as human activity systems Checkland the system boundaries are often less clear. We can define a City; within that we can define a Factory, Shop, Cinema, Theatre, homes – these have physical boundaries (walls) to help. But when it comes to Politics, Economics and Culture then the boundaries are more contested – there may agreement at the broad scale but as we dig into the detail it all become more judgemental and is not necessarily stable over time.
To take one example it used to be common place to talk about “broken families”, now it is common to talk about extended families, these are constructs based on a view of the world Checkland, p214-221 - in one case a view that a family consisting of 2 parents with a number of children, subject to divorce or separation was somehow broken, in the other an attempt to celebrate the freedom to end unworkable relationships and the resulting reshaping of familial, and other connections in a mature way. In other cultures the term extended family is used more traditionally to bring in cousins, aunts and uncles…the point here is not to make any value judgements but to highlight that the values we bring to the description of the world colour the way we look at it.
Locating people in the biosphere may seem a bit of geeky pedantry to complete the view of the system; actually it is a vitally important step.
What happens in the biosphere matters because it is essentially self balancing – except over a very long period of time Davies and Lovelock
When James Lovelock was building models of the world he made the surprising discovery that they can be self-correcting but that, subject to enough stress, they will flip – the change when it comes will be sudden and drastic Note: The Tipping Point . The reason to worry about global warming is not so much the gradual temperature rise (though resulting impacts in dry and costal areas can be devastating enough) but the tipping point is an unknown . It is almost certain that the self-balancing mechanisms are being stressed by human activity – with a range of consequences from disruptive climate change to existential threats in some extreme scenarios Lovelock, Klein .
The key point of this systems view is to locate people within a physical system – the world (actually the bio-sphere) within which we exist as individuals and groups. Some argue that we have become alienated from the natural world or lost touch with it. By cultivating systems thinking we can establish knowledge and awareness of our interconnectedness without having to resort to the wishful thinking that we need to get closer to nature. There are too many of us for that, we are using our brains to invent things that shape the industrial world we live in, there is no going back. Note: Legends of the Fall .
That thought brings us right to the critical point, what can we change and what do we have to live with? A corollary is if we change it, how much control do we have? It appears we that we are adding enough carbon to the atmosphere to trigger (or accelerate underlying) global warming; the resulting weather is outside our control.
On boundaries
Some sub systems are obvious and whilst we are dealing with the natural world we can see many clear boundaries; any surface Note: Water Cycle . When we come to animals, humans included, the boundary between the organism and environment is the skin. However even in the physical world care may be needed -the boundary may be look obvious until we take a closer look; fly out of a cloud slowly and what looks like a boundary from the ground may be a gradual change in the density of water droplets. What is the actual boundary of a cloud?
Boundaries are ubiquitous and important; in chemistry and materials science the boundary characteristics of materials, in business Coase wrote about the boundary of the firm which is important for making outsourcing decisions, in politics the boundary commission has the contentious job of fixing political constituencies and in sociology and psychology the whole subject of groups and the way we create the “other”outside our own group boundaries.
When it comes to social science we have become accustomed to using groups with boundaries that do not have any basis is systems, i.e. when the thing being studied is abstract. Where a boundary is proposed merely for the purposes of statistical analysis it may have no bearing on the natural or human systems world. Unlike the use of Maxwell’s statistics applied to physical things, people are different – the aggregate behaviour of gas doesn’t encounter the problem that only 70% of the gas particles behave in that way. However a statistical approach applied to people always faces the problem some of the people fit the profile whilst others do not. That is not to say this imagined grouping may not have any utility, for example, if we know the proportion of people who prefer vanilla ice-cream compared to raspberry, then we can produce flavours of ice cream in the right proportions. However there is no system boundary here. Class is therefore is a massive problem and the notion of class war even more problematical. By what input-process-output system does the middle class exist? Snobbery and exclusion based on wealth, manners, or “not being one of us” of course all exist (often with pernicious consequences) but that is not the same thing. We may say that Capitalism exists – being the mode of organisation for financing companies organised in a certain way but by that definition a CEO, even one with massive rewards is not a member of the capitalist class, she may be a swindler, confidence trickster or just plain greedy but that again is another thing altogether. Society does of course exist – it is simply the aggregate of all human activity sub-systems.
This is a fundamental observation and has implication for the practice of politics as we shall see.
Changing systems
According to Donella Meadows the ways to change a system starting with the least effective and ending with the most effective are these:
12. Constants, parameters, numbers (such as subsidies, taxes, standards).
11. The sizes of buffers and other stabilizing stocks, relative to their flows.
10. The structure of material stocks and flows (such as transport networks, population age structures).
9. The lengths of delays, relative to the rate of system change.
8. The strength of negative feedback loops, relative to the impacts they are trying to correct against.
7. The gain around driving positive feedback loops.
6. The structure of information flows (who does and does not have access to information).
5. The rules of the system (such as incentives, punishments, constraints).
4. The power to add, change, evolve, or self-organize system structure.
3. The goals of the system.
2. The mindset or paradigm out of which the system — its goals, structure, rules, delays, parameters — arises.
1. The power to transcend paradigms.
This can be found here Note: Places to Intervene in a System .
This finding supports the diagnosis that pursuing regulation is going to make change difficult, it concentrates on the least effective means of changing a system and is always open to being gamed. Much better to win the argument and create a climate of opinion that makes gaming the system unacceptable. We frequently hear that someone did not do anything wrong as an excuse for unacceptable but legal behaviour.
History of systems thinking
Capra and Luisi give a historical account of the development of systems thinking as it emerged from number of disciplines in the 1920’s, they highlight one source as the debates in biology between mechanism and vitalism, which is also pointed out in Checkland Checkland p75 , Capra and Luisi, Ch4 .
For a more discursive and normative discussion of the change needed in science summarised against the background history of science see Capra, The Turning Point. Capra is clear that the straight jacket of reductionism has had many (unintended) negative connotation for society as a whole.
Taken at the coarsest level it seems to me that systems thinking emerges at the point (differently depending on the discipline) where reductionism encounters some insuperable difficulty; it has no way of approaching emergent properties to give an explanation of clearly observable behaviours.
Systems thinking is a recent development, especially set against the history of human thought. This representation may not go back far enough but it does make the point - we have to understand something we have only just and are still working out right at the point where population pressures are going to peak (environment degradation, resource use, migration, anti-biotic resistance).
Systems thinking now
The biosphere viewed as a system which we also describe loosely as the environment has a massive literature on the environment which I have no intention (or ability) to summarise any better than what is out there already. The thing to take away is that the world we live in is a closed system and that we all operate within it. The holistic thinking used to comprehend this is a powerful tool allowing us to build up knowledge about how things work.
My takeaway from this is simple; we only have one so we had better look after it and as a follow on we cannot allow a small number of people ruin it for everyone else. The implication for political economy is clear - we have to adapt our political economy to this reality.
Within Science multidisciplinary work is now common. Scientists work increasingly in large teams with diverse backgrounds and specializations. A simple current example is from BBC Radio 4, The Life Scientific, Trevor Cox, whose research is in Acoustics, in interview with Jim Al-Khalili said that the field contained Physics, Biology and Psychology.
The focus of this site is holistic political economy. Economics is perhaps the most aggressive of the social sciences in the sense that it wants to want to see itself as scientific and as a result be able to proscribe actions, which can be applied in a technical manner to control the economy. Embedded within this are a whole raft of systems ideas usually split between macro and micro-economics. We can easily comprehend a business as a system – it takes in supplies and orders, carries out processes in order to produce outputs fulfilling orders. At the macro economic level the country has inward and outward flows resulting fro the economic activity within it. So what do we observe about the current state of economics.
Increasingly following the financial crash of 2008 economics is being challenged on a number of fronts. There is a lot of evidence for this, to pick just a few of the more widely known;
- For claiming to be objective when it is in fact value based look no further than 23 things they don’t tell you about capitalism, Ha-Joon Chang, Pengiun, 2011, ISBN 978-0141047973.
- For being little more than a collection of inconsistent theories that are based on insufficient evidence but which are nevertheless used to prop up the status quo (willing or unwitting depending on your view). From a systems thinking perspective on this see both Paul Mason, Post Capitalism; A guide our future, Penguin, 2016, 978-0141975290 – page 50 – “three general features of complex, adaptive systems…such systems tend to be ‘open’…they respond to challenge by innovating and transforming…they generate ‘emergent’ phenomena, which can be studied only at a higher level”. And Doughnut Economics, Kate Raworth, Random House, ISBN 978 1847941374, Chapter 4 Get Savvy with Systems. Raworth also provides a useful roundup of the way economic theories do not fit together and are value based (see introduction), they often lack a full evidence base – one example being the Kuznets Curve pp 208.
- For not allowing alternative views to be explored (although this is changing (to me) it seems to be a slow process). You can follow what has been happening in academia after student protests “calling for a more relevant curriculum that reflects different approaches” Financial Time Report May 2014 and then go to the site that was set up as a result.
- Sources here Note: Challenging Conventional Economics
Future implications of systems thinking
It was said to me long ago (by my old headmaster and later by a work colleague) that socialism was all very well but did not take human nature into account Note: A Quip About Socialism . I have been pondering that ever since – is human nature one of the givens that we cannot change? How much control do we have over its Note: How Much Control? . We can build lots of stuff, but we may not be as clever as we like to think - we have only recently started to design earthquake resistant buildings. When things we do react with things in the biosphere there may be unintended consequences which we don’t understand and cannot control. We argue over large damn projects, we hesitate to use nuclear power but we still rapaciously cut down trees, and mine for and burn coal and gas. We all know about critical mass when it comes to nuclear weapons, but don’t get critical mass when it comes to a possible environmental tipping point. We create cities and have culture so it follows that large parts of the environment we are responding to are in fact human creations.
It took a long time to fully appreciate the implications of systems thinking at the simplest level possible; human activity as a system from which behaviour is an emergent property. It is a central part of the arguments developed in this ebook that we can change the world if we change the way we learn and think about it (that is by concentrating on Donella Medows numbers 4,3,2 and 1 as ways to change a system).
The belief that I bring to this is that it is pretty urgent that we start doing it.
Further Reading
A really good place to start which does not require any specialist knowledge is;
- Donella H. Meadows, Thinking in Systems, a primer, Earthscan 2008, ISBN: 978-1-84407-725-0 (hb)
Influences
My key influences for systems analysis are listed below; my introduction to systems analysis was through the design of computer systems for business. Starting at Perkins Engines in Peterborough I was very fortunate to be learning about systems analysis not just from people with a programming background but also from people with an Organisation Methods background, many of these people had been coopted into the IT Department (at the time called Management Systems) and worked as systems analysts. I say fortunate because this was 1975-83, when we designed a system we designed it all - people, process and technology. This was well ahead of its time.
When I did my MBA I was again fortunate, I was able to do a dissertation on Information Systems Planning, the literature survey for that provided a useful link between my practical experiences and the theory behind systems, planing and change. Some sources that have also been very influential on my thinking, in particular the following:
- Kuhn, Thomas S: The Structure of Scientific Revolutions, Chicago University Press, 1970
- Mintzberg, Henry: The Rise and Fall of Strategic Planning, Prentice Hall, 1994
- Senge, P.M: The Fifth Discipline, the art and practice of the Learning Organisation, Doubleday, 1990
- Ward, John and Griffiths P: Strategic Planning for Information Systems, John Wiley, 1996
- Zuboff, Shoshan: In the Age of the Smart Machine, Heinemann, 1988
The following sources are getting on a bit now but the ideas are still in use.
- Peter Checkland, Systems Thinking, Systems Practice, John Wiley, 1981, ISBN; 0 471 27911 0.
This is the original text for Soft Systems Methodology (SSM) and is the source for my usage The Human Activity System, which I have depicted at the highest level of abstraction that I can. According to the action research approach within soft systems the steps a practitioner should go through when intervening are;
- Enter situation considered problematical
- Express the problem situation
- Formulate root definitions of relevant systems of purposeful activity
- Build conceptual models of the systems named in the root definitions
- Compare models with real world situations
- Define possible changes which are both possible and feasible
- Take action to improve the problem situation
A key aspect of this which does not jump out from the above list is that it is important to create, from the different viewpoints encountered a consensual model. This can take some time and effort. The roles of the various participants and some key concepts that it uses are these;
- Customers - Who are the beneficiaries of the highest level business process and how does the issue affect them?
- Actors - Who is involved in the situation, who will be involved in implementing solutions and what will impact their success?
- Transformation Process - What is the transformation that lies at the heart of the system - transforming grapes into wine, transforming unsold goods into sold goods, transforming a societal need into a societal need met?
- Weltanschauung (or Worldview) - What is the big picture and what are the wider impacts of the issue?
- Owner - Who owns the process or situation being investigated and what role will they play in the solution?
- Environmental Constraints - What are the constraints and limitations that will impact the solution and its success?
Other IT systems books
I have been influenced by other works notably;
- Chris Gane and Trish Sarson, Structured Sytems Analysis, Tools and Techniques, Prentice Hall, 1979, ISBN; 0-13-854547-2.
- James Martin; Information Engineering (3 Vols, Introduction, Planning and Analysis, Design and Construction), 1989, 1990 ISBN; 0-13-464462-X, 0-13-464885-4 and 0-13-4650501-X
- Ian Graham, Object Oriented Methods, 2nd Ed.,Addison-Wesley, 1993, ISBN 0-201-59371-8
- Frederick P. Brooks Jr, The Mythical Man Month, Essays on Software Engineering, (Anniversary Ed. with new chapters), Addison-Wesley, 1995 (1975), ISBN 0-201-83595-9
- John Ward and Pat Griffiths, Strategic Planning for Information Systems, 2nd Edition, John Wiley, 1996, ISBN 0-471-96183-3
- Yan-Ping Shan and Ralph H. Earle, Enterprise Computing with Objects, from Client-Server Environments to the Internet, Addison Wesley, 1998, ISBN 0-201-325-66-7