Policy analysts and systems analysts

Print Friendly, PDF & Email

I’ve finally got round to reading a book I have had for a while – the second edition of Beryl Ralin’s book ‘Beyond Machiavelli’.  The first 2000 edition is subtitled ‘Policy analysis comes of age’ and the second 2013 edition is subtitled ‘Policy analysis reaches midlife’.  It is entirely US based and traces the evolution of the policy analyst profession from its inception in 1960s american policy project to the present day.  There’s lots in there that I am mulling over, but I couldn’t move on without noting linkages with ‘systems analysts’ – another US profession that kicked off in the 1960s.  As Ralin says:

The imperatives of war had stimulated new analytic techniques – among them systems analysis and operational research – whose users sought to apply principles of rationality to strategic decision making (p.14)

Ralin cites the writings of Dror in his advocacy for the policy analyst profession.  Writing in the 60s/early 70s I see that it was Dror who argued for the name of ‘policy analyst’  because “it combines affinity with systems analysis with the concept of policy in the broadest sense” (p. 24).  The text implies that he wanted to draw attention to the political aspect of policy that was overlooked in systems analysis, yet retain the links to the notion of rationality and expertise.

Later on Ralin cites Dror again.  This time she references a 1984 chapter in an edited volume.  When I looked up the reference I found it was included in a book called “Rethinking the process of operational research and systems analysis” which amused me because I thought that he’d been trying to create a clear delineation between systems analysis and policy analysis.  And then I found a used copy on amazon for just 2 pence – I couldn’t resist!

In the introduction to the volume the editors – Istvan Kriss and Rolfe Tomlinson – describe its origins in a seminar held in 1980…

The seminar was, in the first place, the result of a common concern that systems analysis, and its sister disciplines such as operational research, too often failed to achieve successful implementation because of an incomplete, and often incorrect, methodology.   It was felt that analysts were making assumptions, sometimes sub-consciously, that did not stand up to serious examination.  One such assumption was that the ‘hard’ part of a problem – which could be expressed in mathematical terms – could usefully be isolated from the human and organisational elements which could thus be eliminated from the analysis.  Another false assumption was that implementation was an entirely separate activity from the analysis itself (p. xi)

When I was looking at the contents for the Dror chapter, I got distracted by seeing Peter Checkland’s name (and Churchman’s).  I started to flick through Checkland’s contribution.   Of course a lot of the chapter is relatively familiar material but with my ‘appreciative setting’ focused on links across to policy analysis two parts particularly stood out for me…


His distinction of different types of phenomena or situations.

  • Type 1 – those characterised by interconnections which are part of the regularities of the universe – (I’d refer to as “ecological” or “biological” systems)
  • Type 2 – those characterised by interconnections which derive from the logic of situations – (I’d refer to as mechanical systems)
  • Type 3 – those chacterised by interconnections which are cultural and are dominated by meanings attributed to their perceptions by observers – (I’d refer to as human activity systems)

He points out that the most frequent error in using a systems approach is to treat type 3 situations as if they are Type 1 or 2.  I’ve seen this distinction before in a paper by Jake Chapman, I hadn’t connected it back to Checkland.  But it reminds me how often people are currently using the notion of ‘complex adaptive system’ which developed from studies of Type 1 situations (the natural world) but applying it in Type 3 – “the NHS is a complex adaptive system” (aggregate of a variety of implicit or explicit statements I have seen), rather than “what do I learn if I think of the NHS as if it is a complex adaptive system”.

So to try and relate this back to the title of this blog – I’d have thought that the interests of policy are pretty much all Type 3.  No wonder those early day policy analysts working in a world where positivism and economic theory were (even more) dominant started to notice the limitations of what they were up to.


Checkland includes an appendix to his chapter which includes ‘CATWOE’ definitions of paradigms of systems analysis based on ‘hard systems thinking’ and ‘soft systems thinking’.

Hard systems thinking Soft systems thinking
C “Customer” victims and beneficiaries of what the system does decision makers who command real-world situations participants who debate the differences between the models and expression of the problem situation
A ” actors” those who carry out the system activities external analysts and engineers those who choose to take part: analysts and/or problem owners
T “transformation” what input does this system transform into what output information into advice to decision makers information into specific learning for the ‘actors’
W “weltanschauung” the worldview which makes it meaningful to consider this system Real world is systemic, Methodology is systematic, Optimisation is possible real world is problematical, methodology is systemic, Learning is possible
O “owners” those who could demolish this notional system, could prevent it from acting Decision-makers/clients “Actors” as defined above or the analyst
E “environmental constraints” the things in its environment that this system takes as given Power structures and value systems of the decision-maker/client as little as possible compatible with achieving change in the problem situation

(Source: page 65)

This re-conceptualisation of ‘systems analysis’ he proposes seems to me to be helpful as the start of a re-conceptualisation of ‘policy analysis’.

Still musing over these ideas…



Tomlinson, R and Kiss, I (1984), Rethinking the process of operational research and systems analysis, New York: Pergamon Press

Ralin, B.A. (2013), Beyond Machiavelli: Policy analysis reaches midlife Second Edition., Georgetown: Georgetown University Press.


Share what you think...

This site uses Akismet to reduce spam. Learn how your comment data is processed.