YOURS IS NOT TO QUESTION WHY

July 2016 marks the passing of 100 years since the Battle of the Somme was fought as a major campaign in the First World War. The Battle involved a huge number of British soldiers, but the detailed plans that were put in place for the operation failed with disastrous consequences. The assumptions and approach used at the time for the planning and execution of the battle have, with hindsight, been found to be at best flawed and at worse catastrophic. Taking a step back, it might seem alarming that many of the assumptions and features of the battle’s preparation and conduct can still be seen being applied in a similar way to modern-day business project delivery, particularly in the approaches to governance of IT projects.

The Battle of the Somme

Although the Battle of the Somme began on 1st July 1916, the British Fourth Army, having been specifically formed and tasked with conducting the operation, started to work on the plan for the attack at the beginning of 1916. As a consequence of Lord Kitchener’s[1] huge recruitment efforts much of the Fourth Army was made up of new recruits. The plans were therefore drawn up based on the suppositions that the new troops would not be skilled enough to conduct complex manoeuvres and that, as new and untested soldiers, they would lack the discipline of regular soldiers. These assumptions resulted in the belief that the troops would have to be tightly managed. In turn, this led to detailed specifications of ‘best practice’ and standards, including instructions that troops should walk across no-man’s land to make sure they stuck together under the control of their officers, thereby maintaining good order[2].  Additional controls that were put in place were “phase lines” which specified when troops should arrive at certain locations and objectives. This was necessary to coordinate the movement of battalions in with the artillery fire plan, which was being controlled at corps[3] level (3 levels up in the hierarchy), and therefore prevent “friendly fire” casualties.

The planning assumptions and highly specified and detailed plans all appeared very plausible, however as with any effort to produce a plan, the planners were affected by the same ever-present challenges: there was a gap between what the planners wanted to know and what they actually knew and therefore gaps between what they intended and what actually happened.[4] It is these gaps that contribute to the occurrence of Clausewitz’s concept of ‘friction’. Clausewitz wrote that “friction is the only concept that makes the difference between real war and war on paper” [5], a scholar of Clausewitz summarised the concept as the totality of “uncertainties, errors, accidents, technical difficulties, the unforeseen and their effect on decisions, morale and actions”[6]. Key gaps in knowledge at the Battle of the Somme were in the understanding of the structure and integrity of the German trench defences and in the understandable lack of knowledge about the ‘unprecedented’ week-long artillery barrage which preceded the infantry advance.  The intended effects of the barrage were that it would destroy the German dugouts, cut through the barbed wire and kill, injure or suppress a large proportion of the German troops. Haig[7] was so confident about the impact of the artillery he claimed that “not even a rat” would survive[8]. The reality of the artillery barrage was that it did not achieve any of the intended effects.

As the Battle progressed, middle ranking officers and frontline commanders faced circumstances that the planners had not foreseen. At their level, they had no decision-making authority and could therefore only try to communicate back to their corps headquarters in an effort to relay the ‘friction’ they were experiencing and to seek direction and new orders.  However the Fourth Army headquarters had given multiple objectives to their corps commanders and the overall intention of the Fourth Army was unclear because the two most senior officers had an unresolved disagreement as to what the operation should achieve[9]. As a consequence, the army and various corps headquarters were of no help to the middle-ranking officers. So with no direction, no change of orders and no authority to change the plan, these frontline commanders could only continue to follow the plan. The plan had essentially become the ultimate authority. The Fourth Army suffered over 60,000 casualties on the first day of the battle, making 1st July 1916 the bloodiest day in the whole history of the British Army, and with very little to show for the sacrifice of so many soldiers.

It is easy to conjure images of senior officers at that time being in the General Melchett[10] mould: making decisions and issuing orders in the role of mad and bad psychopaths. However these officers were employing a style of command that had been very successful from the time of the Roman Empire through to Wellington at Waterloo. This approach to command was based on disciplined soldiers following highly specified drills[11], resulting in formed units of individuals that functioned like proverbial cogs in a machine. This enabled the Commander to tightly control the units under his command and implement his intended plan. The Generals of the Fourth Army were in fact, unwittingly, applying a management approach being taken in modern industry at that time which was based on the logic of Fredrick Winslow Taylor’s ‘scientific management’. Although Taylor and the Generals were both applying a machine model of operation and management to their organisations, there is no suggestion that military commanders were guided by, or even aware of, scientific management[12]. Taylor’s assumptions (and therefore consequential actions) such as how to control workers and how to co-ordinate their activity therefore had striking similarities with the Fourth Army’s preparation and execution of their operations at the Battle of the Somme. The mechanisms implemented by the generals were an effort to achieve control, coordination and therefore assurance that the necessary outcomes would be achieved in order to enable a decisive victory for the allies and so bring an end to the war. Of course this legacy thinking was not the only contributing factor to the outcome of the Somme; factors such as the unprecedented numbers of soldiers under command, their wide dispersion on the battlefields and the scale of the attack demanded a steep learning curve[13] which, added to the command approach, resulted in such large numbers of casualties.

I.T. Project Delivery

The source of the word governance comes from the Greek kubernáo, meaning to steer. Governance is a term that has a great deal of currency in the contemporary management of our services and sectors, often used in relation to controlling and therefore assuring supervising bodies that organisations are functioning as expected. In the First World War the term governance was not in use, however characteristics of the planning and execution of the Battle of the Somme can be seen in the governance methods in modern project delivery. ‘Good’ governance often requires:

• Specification of good practice (such as processes and procedures)

• Creation of detailed schedules or plans

• Central management and control over activity

• Robust reporting structures

It follows then that the assumptions and logic that underpinned modern management thinking at the time of the First World War are still present in many ‘modern’ management approaches and actions[14]. Governance is highly prevalent in modern Information Technology (IT) project delivery, and unsurprisingly the characteristics of good governance are observable in the delivery of projects.

Despite all efforts to assure the outcomes and benefits from investments made in changes, IT projects in particular have had appalling failure[15]. The common response to projects showing signs of failing has been to increase oversight by committees, boards and sponsors through increased emphasis on reporting in order to regain control. This sort of response and behaviour is not specific to particular types of project or organisation; I have observed this in both the public and private sector and in large and small organisations. Atul Gawande cites examples in a wide variety of organisations and circumstances and observes that “in response to risk, most authorities tend to centralize power and decision-making”[16].  Gartner’s article ‘IT Projects Need Less Complexity, Not More Governance’ observes similar failings and advocates “less bureaucracy, greater focus on outcomes and simplify everything about the project”[17].

Efforts to reduce bureaucracy and simplify project delivery are laudable and on the face of it very plausible, however in themselves they are unlikely to provide a sustainable change in project delivery performance. The various artefacts and control mechanisms that may be the target of simplification efforts will not have appeared in organisations by chance or as a result of any physical laws of nature.  They will have been introduced by somebody at some point in time to satisfy an assumed, but often unstated and misunderstood, need such as maintaining control of activity through compliance. Chris Argyris identified this as a problem for organisations, and described it as “single-loop learning”, in which errors in organisations are detected and corrected in a way that permits the organisation to continue with its present policies or achieve its present objectives[18]. Many governance activities are associated with ensuring compliance to policies and maintaining control over the changes within the organisation, in order to provide an increased certainty of outcome. Therefore the efforts to ‘detect and correct errors’ in project bureaucracy and complexity will be undertaken without challenging the logic and assumptions that placed them there in the first instance. Assumptions, such as that observed by Gawande, that centralising power and decision making is an effective response to risk remains unchallenged even as issues and risks continue to recur[19]. Alternatively the “double-loop learning” that Argyris describes seeks to challenge and modify the assumptions, beliefs, behavioural norms and policies within organisations.

As a result of the Vanguard Method encompassing the nature of double-loop learning many people have identified unhelpful, outdated and flawed assumptions that are prevalent, but rarely recognised in their organisations. This learning often highlights that the ‘modern’ approaches to management are still grounded in the logic of the industrial production era. The consequences for the operational delivery of services in these organisations is that just as for IT project delivery and its governance, the operational delivery is full of artefacts and mechanisms which seek to maintain the delusion of control through specification, standardisation and compliance. Perversely this not only compromises the service experienced by customers and service users, but also results in performance problems being misunderstood. By then employing single loop learning organisations often seek to ‘do the wrong thing righter’[20], with the result that projects are launched to implement a change that will never achieve the desired outcome. This is explored further in John Seddon’s article for the Cutter IT journal, ‘Dissolving a Dangerous Enthusiasm: Taking a Systems Approach to IT Systems’. The consequence for project delivery is that a combination of single-loop learning and management assumptions derived from outdated industrial production means that IT projects might not only be mandated to tackle the wrong problem, but are then subjected to a governance approach that provides only an illusion of control.  The consequences of these two aspects mean that IT projects are often viewed as failures.

An alternative approach to IT project governance

Clausewitz’s concept of friction is not only relevant to the battlefield, but to any instance of planned and coordinated effort. At the outset of any project there are things that are not known, or even as Donald Rumsfeld[21] put it there are “unknown unknowns”. In the delivery of projects there are often differences between what the plan requires people to do and what they actually do. In the results of projects there is often a difference between what is expected from people’s actions versus the actual outcomes.[22] As a project is initiated it is therefore important not to assume that the plan is an accurate prediction of what will happen, but accept that friction will invariably affect the plan. It is equally important not to assume that just because a ‘risks and issues log’ exists, the friction will therefore be dealt with effectively.

Releasing project delivery from the traditional management assumptions, governance therefore becomes a connection between those with authority and influence into those with current and detailed knowledge.  Those responsible for governance therefore connect to the actions and work being undertaken to achieve the change, as opposed to sitting in a meeting room receiving RAG status, written reports and briefings.  All of these artefacts act as filters of the facts and the reality of what is actually occurring. This then ensures that the causes of friction are acted on in a timely and effective way, so that obstacles to the progression of the change are removed. Some of the causes may require greater spans of control or access to resources and budget than members of the change/project team possesses.

Rather than governance seeking obedience of actions to processes and best practice, it can instead seek to achieve an adherence to rigorous thought processes and decision-making. Evaluation therefore focuses on the logic and judgement of those undertaking the change. This is no different to the Toyota Production System’s use of the A3 in problem solving. The A3 is often misunderstood as a tool to standardise problem solving, whereas its actual purpose is for open reflection of the logic and approach that is being used to tackle the problem.[23] Again the early origins of this approach can be found in military history, as Stephen Bungay writes about Gen von Moltke[24] teaching his officers to “think independently and use their own judgement” and “in one exercise put officers in a position in which they had to disobey orders in order to be successful. The result was, von Moltke said, that in a given situation 99 out of 100 officers would react as he would himself”[25].  The alignment of decisions and actions that von Moltke achieved exemplifies the effective functioning and desired outcomes which modern-day governance seeks. It also mirrors the model of autonomous decision-making that Toyota seeks in its operations:

‘At Toyota, we began to think about how to install an autonomic nervous system in our own rapidly growing business organization. In our production plant, an autonomic nerve means making judgments autonomously at the lowest possible level; for example, when to stop production, what sequence to follow in making parts, or when overtime is necessary to produce the required amount.

These discussions can be made by factory workers themselves, without having to consult the production control or engineering departments that correspond to the brain in the human body. The plant should be a place where such judgments can be made by workers autonomously.’[26]

Clearly this approach is the complete antithesis of the machine model employed by both Haig and Taylor.

Summary

The prevailing and often accepted management thinking that underpins many of our modern organisations was born out of industrial manufacturing, but it was also prevalent in the approach taken by the British Fourth Army at the Battle of the Somme. The legacy of this management thinking is also widespread amongst those charged with the governance of change in our modern organisations. The alternate approach to IT project delivery and governance therefore requires a move away from the products, tools and mechanisms that add complication, bureaucracy and complexity to the environment in which project teams operate. Additionally, it also requires a move away from some unhelpful assumptions about the identification of what changes are required, about how to maintain control of the changes and a belief that it is possible to produce a perfect plan that assures certainty of outcomes from the change. No longer is it appropriate for us to seek blind obedience and compliance, now it requires an expectation that everyone should question and understand why, because in the words of von Moltke “no plan of operations extends with any certainty beyond the first contact with the main hostile force”[27].

---

Sources:

[1] Lord Kitchener was the British Secretary of State for War at the beginning of the First World War

[2] Martin Samuels, 1995 ‘Command or Control – Command: Training and Tactics in the British and German Armies 1888 – 1918’ London: Frank Cass

[3] A corps is a large military formation, made up of at least 2 Divisions. Typically each corps will have between 20,000 and 40,000 soldiers.  In World War 1 multiple corps were grouped together and formed as armies, as in the case of Fourth Army.

[4] Stephen Bungay, 2011 ‘The Art Of Action’. Boston, MA: Nicholas Brealey Pub.,

[5] Carl von Clausewitz 1832 ‘Vom Kriege – On War’

[6] Peter Paret, 1985 ‘Clausewitz and the State: The Man, His Theories, and His Times’ Princeton: Princeton University Press

[7] Field Marshall Douglas Haig was Commander-in-Chief of the British Expeditionary Force in 1916

[8] See BBC website http://www.bbc.co.uk/schools/worldwarone/hq/wfront3_01.shtml accessed 7/6/16

[9] Gen Rawlinson envisaged a “bite & hold” operation, whereas Gen Haig wanted a “breakthrough” operation.

[10] Fictional character from the BBC series ‘Blackadder Goes Fourth’, written as a caricature of the worst features of First World War generals

[11] In military terms drill is a specification of procedures and movements such as marching or use of weapons

[12] Martin Samuels, 2015 ‘Doctrine for orders and decentralization in the British and German Armies, 1885 – 1935’ War in History 2015, Vol. 22(4) 448–477.  In the article Samuels points out that “Nonetheless some of the professional managers brought into the British Army were familiar with Taylorism and applied it in their newfound military duties.  For example, Brigadier General E G Wace, who from 1916 headed the army’s military labour establishment, was an ‘enthusiastic advocate of scientific management and a follower of…Taylor’s system’

[13] Gary Sheffield 2001’Forgotten Victory: The First World War – Myths and Realities’ London, Headline

[14] See John Seddon  ‘The method and the madness’ The Observer 12 September 2004 https://www.theguardian.com/business/2004/sep/12/theobserver.observerbusiness11 accessed 07/06/16

[15] Robin Gauld and Shaun Goldfinch, 2006 ‘Dangerous Enthusiasms: E-government, Computer Failure and Information System Development’ Otago University Press: New Zealand, The authors summarise that for large-scale IT projects: 30% fail completely and a further 60% fail in that they require significantly more time, resources, and effort than planned, and even then they often fail to meet requirements.

[16] Atul Gawande, 2010. ‘The checklist manifesto: how to get things right’. New York: Metropolitan Books

[17] Susan Moore July 17, 2015 ‘IT Projects Need Less Complexity, Not More Governance’

http://www.gartner.com/smarterwithgartner/it-projects-need-less-complexity-not-more-governance/ accessed 7/6/16

[18] Chris Argyris & Donald Schön, 1978 ‘Organizational learning: A theory of action perspective’, Reading, Mass: Addison Wesley.

[19] As recognised in Cobb’s Paradox: “We know why projects fail; we know how to prevent their failure—so why do they still fail?”

[20] Quote originally from Russell Ackoff see Financial Times 09 November 2009 ‘Fond farewell to a brilliant thinker’ Byline Stefan Stern http://www.ft.com/cms/s/0/0168c7de-cd7e-11de-8162-00144feabdc0.html#axzz22Hyp8bBK accessed 7/6/16 “All of our problems arise out of doing the wrong thing righter,” [Ackoff] told me. “The more efficient you are at doing the wrong thing, the wronger you become. It is much better to do the right thing wronger than the wrong thing righter. If you do the right thing wrong and correct it, you get better.”

[21] Donald Rumsfeld, former US Secretary of Defense said “…we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns — the ones we don’t know we don’t know.” http://archive.defense.gov/Transcripts/Transcript.aspx?TranscriptID=2636 accessed 7/6/16

[22] Stephen Bungay, 2011 ‘The Art Of Action’. Boston, MA: Nicholas Brealey Pub.

[23] Jeffrey Liker & David Meier 2006 ‘The Toyota Way fieldbook: A practical guide for implementing Toyota’s 4Ps’. New York: McGraw-Hill pg 376 para 3

[24] Gen von Moltke was a highly influential Prussian Officer who became Chief of the General Staff in 1857 and went on to be the driving force that created the German Army at the end of the 19th Century.

[25] Bungay pg 180

[26] From Taiichi Ohno 1988 ‘Toyota Production System’ Productivity Press: Portland, Oregon. P45

[27] Originally in von Moltke 1892 ‘Militarische Werke’ vol. 2, part 2., pp. 33-40