People, Protection and Parameters: Comparing Flooding in the UK and the Netherlands

13 January 2015

People, Protection and Parameters: Comparing Flooding in the UK and the Netherlands

Dr Maurits Ertsen

Introduction

Recent floods in the United Kingdom have led to debates within the country on its flood protection policy. Experts from the Netherlands, a country often seen as the standard to follow on flood protection, are involved too. Comparison between the Dutch and the UK situations can help explain how differences in water system properties, economic development and ideas on national security result in different flooding policies and patterns in the two countries. 

It may appear strange to start a talk about the United Kingdom and the Netherlands – from the wet and cold part of Europe – with a scene from northern Nigeria, but that is where we will start the story. There, in the 1970s, we meet a proud Dutch engineer standing next to an irrigation structure known as Begemann gate he had just constructed in the Kano River Irrigation Project. Dutch engineers were involved in this project to irrigate some 24,000 hectares from a reservoir in the Kano River. How to arrange water distribution was one of the key questions. One of the Dutch team members proposed to apply Begemann gates to control water levels.

The Begemann gate was named after Dutch irrigation engineer Begemann, who had introduced the artefact in the 1920s in the Netherlands East Indies – modern Indonesia. [1] At the time, there was a bewildering amount of gates tested in the Indies. Automatic gates had the advantage that water management could “become independent of the alertness of the operating personnel”. [2] The gate that was proposed by Begemann was relatively simple, a hook-type gate with a counterweight on top.

Dutch irrigation engineers proposed to apply a structure in Nigeria that was developed within a former Dutch colony. The Nigerian government had contracted another engineering firm to assist them in evaluating the plans made by the Dutch consultants. This firm mainly employed engineers from Egypt and Pakistan. These engineers had different ideas about what was a proper structure to be used in irrigation, based on their own regional experience and knowledge.[3]

As such, the Kano River Project was an arena where different preferences for irrigation system design met. In this case, to determine which type of water distribution structures would be suitable, several candidate structures were tested at the experimental station of the project. In the Kano system the Begemann gate was selected and applied.[4]

Please do not get the impression here that I am planning to discuss Dutch-UK relations on flood policies and measures in terms of development cooperation or aid. Far from it. What I do want to argue, however, is that the process of expertise and experts being confronted with each other in Nigeria is also relevant when trying to understand the linkages established between the current flooding policies and processes in the two countries.

It is those similarities in the processes I would like to explore a little more in detail. First I discuss some theoretical notions to frame what we will discuss about floods and wet feet in our two countries. Then I will explore what floods actually are, or at least which types of floods we might want to distinguish between. I will also discuss human influence on flooding. Then I will explore the Dutch and UK situations, in order to compare them.

Technological regimes

The different ‘schools’ of irrigation development that clashed in Kano are examples of ‘traditions of practice’, which are communities of practitioners that physically embody information of that community plus the rules for action which these practitioners master. Traditions define accepted technical operations and encompass aspects of relevant scientific theory, engineering design formulae, accepted procedures, specialized instrumentation, and usually some kind of ideological rationale. [5] An important mechanism in this process of preference-guided selection is engineering education; graduating from engineering programs is like complying with the preparatory demands for community membership.

Thus, “invention and innovation are conditioned by such factors as earlier innovations, the search heuristics of engineers in an industry, available technical knowledge, market demand and industrial structure”. [6] This conditioning is guided by “formalized knowledge that can be traced through courses and treatises”, but needs to be reconfirmed by “the everyday decisions made by engineers”. [7]

The set of rules – which prescribe what to do in which way - exists, is continuously being reconfirmed and changed at the same time. The daily actions of those that use the rules shape what I define as a ‘technological regime’. [8] “Some rules will be explicitly laid down in requirements and technical norms. Other rules will be tacit and implicit and will be followed by the actors on the basis of habits or tacit knowledge. […] Rules in technological regimes can also be embodied in production apparatus or technological artefacts.” [9] The totality of relevant rules shapes the technological regime. Within a technological regime different categories of rules can be distinguished. I employ five of these categories. [10]

1.     Guiding principles relate the design of a technology to doctrines and values used to legitimize a tradition and its outcomes.

2.     Closely related to these principles are the promises and expectations about a future technology, which will be translated into more specific requirements for new technologies.

3.     I employ the term design requirements to describe functions to be fulfilled by an artefact and boundary conditions that are important in the design of a technology.

4.     To enable the fulfilment of requirements, design tools are employed, including scientific knowledge, design heuristics, technical models and formulas.

5.     Artefacts and operations are the result of any design activity, both in the meaning of physical objects and in the meaning of operation and management procedures. Artefacts can and certainly do function as examples: future designers still apply them because they are known or have been proven in practice.

The categories are shown in a suggested hierarchy; guiding principles seem to be on a higher level than design tools. In most contexts, a “higher level” refers to the more abstract nature of guiding principles, but also to the larger number of stakeholders involved in and the political connotation of formulating guiding principles. Debates on the appropriate foundations for water policy involve civil servants and engineers, government and private industry, citizens, and rulers. On the other hand, discussions which structures to use to realize this water policy are more exclusively situated within the civil engineering circle.

This extremely simplified description of regime development has some functionalistic connotations: rules on one level would shape rules on lower levels. Functionalism, however, is the last thing I want to defend; humans, not abstract forces, create flood policies and measures. I am more interested in conceptualizing technological traditions in the way Giddens – before he became a politician – discussed the concept of structure.[11]

“'Structure' refers to 'structural property', or more exactly, to 'structuring property', structuring properties providing the 'binding' of time and space in social systems. […] these properties can be understood as rules and resources, recursively implicated in the reproduction of social systems.” [12]

Structures do not exist; they manifest themselves through the constituting moments of social systems. [13] Regime development is a social activity; in social interaction human actors construct technological regimes as they construct society. Generally, in daily practice we tend to reproduce many existing, historically grown sets of rules by applying and slightly changing them. To know a rule is to implicitly know what one is supposed to do in particular situations and rules are widely used and sanctioned.

Rules show a tendency to be stable, but are not static.[14] Rules do not develop by themselves, nor are they followed simply because they are there. Actors, real people, make and break rules. Actors will follow the relevant rules – act within the technological regime – not just unconsciously or routinely, but also because they think they have something to lose by not acting in accordance with the rules, or something to win if they do. [15] The theoretical and methodological underpinning of my flood histories is that flood regimes including technologies and policies are local and constructed within networks of actors.[16] These networks are continuously created and recreated by human actors engaging with other human actors and non-human intermediaries, simultaneously at different localities. The micro shapes the macro while being created.

The actor “reveals the narrow space in which all of the grandiose ingredients of the world begin to be hatched”; the network explains “through which vehicles” the outside world is brought inside the local, how these vehicles are transformed, and then how they are “pumped back” to the world. [17] Let’s discuss how this pumping back works in our theme of this lecture: floods in the UK and the Netherlands.

Flooding in The Netherlands

First, I will discuss the Dutch way to deal with flooding. In other words, what does the technological regime of Dutch flooding policies look like? The general topography of the Netherlands shows the higher parts in the east and south, and the lower parts in the west. Parts of the Netherlands are actually below mean sea level. This is about 26% of the country’s surface area. The area under threat of flooding is larger, as much of the Netherlands is also prone to flooding from one of the larger rivers. 29% of the country is susceptible to river flooding. In addition, 4% of the Dutch land surface is situated outside protected areas, and therefore not protected by dunes, dykes, dams or artificial structures. So, a total of 59% of the Netherlands is susceptible for floods from the rivers or the sea.

I mention these figures, as they are often confused. In 2007, the Intergovernmental Panel on Climate Change mentioned that 55% of the Netherlands was below sea level. [18]It should have said that 55% is flood prone. Naturally, such mistakes are typically used to make political points on whether figures on climate change can be trusted or not. This is not the moment to go into these debates, but let me just say that the misconception that most of my country is below sea level is widespread, including amongst many people in the Netherlands itself…

Putting percentages aside, the simple fact is that without adequate protection from flooding, most of the Netherlands would be flooded for long times. This does not mean that all of the flood prone areas will be under meters of water all the time. To take the example of Delft, as Delft centre is exactly at mean sea level, a flood from the sea would mean that the area would be under water only part of the time. I live in an area some two meters below mean sea level, so my ground floor would be flooded all the time. Furthermore, the micro-topography influences how areas would flood exactly – sometimes it takes a long time for water to reach an area. There may be cases where there simply is not enough water to flood an area very deeply.

The flood threat from rivers and sea is translated in a national standard for flood safety. For each area, an accepted occurrence of flooding is determined. For areas with higher economic value and more people, higher standards are prescribed than for areas with less economic value and/or less people. These levels are currently under debate, for two reasons. First, higher flood threats and higher economic values to protect might ask for higher safety levels. Second, safety levels should perhaps be the same for the whole country.

The level of safety is translated into a statistical norm: which extreme event will be the limit? Will it be an event that occurs every year? That would result in relatively low safety levels. The less likely an event is, the higher the standard associated with it. At the moment, the Dutch safety standards range from a probability of one in 10,000 per year for the densely populated western provinces to one in 1250 for the less densely populated river areas more to the east. The levels express the desire to protect the densely-populated and economically important west. They also show that sea floods are treated as more threatening than river floods. These statistics on occurrence of flood events is translated into water levels along coast and rivers. Thus, a standard for safety is defined as design water levels. Protection measures are evaluated with these levels.

Much of the responsibility for complying with flood protection policies lies with the national Ministry of Infrastructure and Environment, mostly through the agency of Public Works. Public Works is responsible for the national waters. As such, Public Works is responsible for operation and maintenance of famous works on the Dutch coast, like the Oosterschelde barrier in the south west of the Netherlands or the more recent Maeslant barrier in Rotterdam harbour. Each of its two gates is as long as the Eiffel tower is high. Public Works is also responsible for construction of the many embankments along the main rivers in the Netherlands.

In executing its responsibility, Public Works meets another main player in Dutch flood policies: the water boards. These governmental institutions are responsible for maintaining and controlling flood defences along rivers and coast. As such, when embankments need to be dealt with, the national Public Works comes to the embankment from the river and the regional water board comes to the embankment from the land. The embankment is literally the meeting place between the two governmental institutions in realizing Dutch flood policy.

Dutch water boards are regional government bodies responsible for daily water management in their respective area. Currently, there are 23 of these boards. They are charged with managing water barriers, waterways, water levels, water quality, and sewage treatment in their respective regions. Water boards are often referred to as one of the oldest forms of local government in the Netherlands – even the oldest form of democracy. The history of water boards shows that they could be understood as feudal as well, and even as early capitalistic entities!

Complex as they might be in historical terms, water boards are clearly governmental institutions today, with elections and a clearly described position in the governmental hierarchy. Water boards are positioned alongside municipalities, and as such are a form of local government. Water boards are obviously much bigger in area than municipalities. For example, the city of Delft is located in the centre of the water board of Delfland.

Delfland is one of the older water boards in the Netherlands. In 2014, it celebrated its 725th birthday! Delfland has 81 polder areas and some 700 kilometres of embankments of different kinds. Within its area, many pumps and canals can be found bringing water from an area to a canal that is part of the belt canal system – the main canals bringing water outside the Delfland area. These belt canals are higher than much of the land around it; as such are a potential flood threat as well. Outside the Delfland area means bringing water to the sea through four main pumping stations, directly or indirectly – the latter through the Rotterdam harbour canal.

This immense infrastructure of canals and pumps is in place to drain away excess rainfall for the area. On average, all of the Netherlands has an annual rainfall excess. Basically, this means that in areas without much natural flow, this water needs to be removed in other ways, for example through pumping. Obviously, the average never occurs. Pumping is not needed every day – though seepage might require continuous pumping in lower areas – but sometimes there is much rainfall.

The water systems in the Netherlands are generally designed in such a way that pumping capacity equals something between 12 and 14 millimetres per day. This means that a rainfall event of 14 mm can be pumped out of an area. As soon as the rainfall event is more than that amount, water needs to be stored, or the water system will start to overflow. In case of severe rainfall, like in 1998 in the Delfland area, even Dutch water systems might start to fail. Much has been done since to prevent these events, especially in terms of temporary storage in locations with less problems and emergency pumping.

In summary, Dutch flooding policies are complex, in the sense that different flood sources, probabilities, safety levels, and institutions are mixed. For floods from the sea, with their tidal effects and huge amounts of water, the threat is highest for the coastal areas. These same areas are the economic core of the Netherlands. Floods from the rivers are also a problem, with large amounts of water, but these come in waves of three to four days and are relatively predictable – although not completely as we will discuss below. Obviously, as parts of the Dutch rivers have tidal influence, tidal levels and storm surges will affect river discharges. Finally, we have excessive rainfall which can occur anywhere in the Netherlands. This is, however, basically a drainage problem, with mainly economic damage only. It is serious enough as it is, but not really life-threatening.

The arrangements for formal responsibilities are also complex. For issues related to excessive rainfall, water boards are responsible, but municipalities obviously are involved as well. For sea and river floods, water boards and Public Works share a responsibility. For all water issues, provinces are involved in the planning process as well, as they oversee water boards. Furthermore, as soon as a disaster emerges, the mayor of a municipality is the first responsible governmental actor. As soon as a disaster involves more than one municipality, the mayors cooperate within so-called safety regions. Involvement from the national level – a minister – is extremely rare and would be highly contested.

For the sake of time and simplicity, I have left out the European levels of arrangements and legislation for now. Let me just say that European legislations, for example to several Directives on water quality and floods, have quite an impact on water policies, but do not change the responsibilities defined above. Going into the details of Europe, regardless of how relevant this might be for both our countries, would take another lecture! Even without the European details, Dutch water management is somewhat complex. Or perhaps it is better to say that Dutch water management is complex, but organized.

Let’s take a minute to arrange some of the elements I just discussed within the framework of a technological regime. In terms of guiding principles and/or the closely related promises and expectations, the Dutch flooding policies are shaped based on the idea that floods should be prevented. Water needs to stay out. A closely related guiding principle is that the western part of the country needs the highest protection levels. This preference for the west is now under debate, actually, but both the current safety levels as well as the history of Dutch flood management and design confirm the preferred position of the West. The main design requirements can be found in the safety levels defined in the Netherlands. The probability of flooding is prescribed in the embankment areas. I have not discussed the design tools – the scientific knowledge and design heuristics employed – but we have touched upon the artefacts used: the embankments, flood defences, and pumps. Furthermore, the Dutch policies include specified disaster protocols. There is obviously more to say, but for now this should suffice.

Comparing floods and policies

Let’s focus now on how this all compares to flooding issues and policies in the UK? Why does the UK flood more often than the Netherlands? Has it just been unlucky in recent years? Is there more to be said about structural properties of either system? And to what extent could Dutch ideas and approaches help the UK? At first glance, the Netherlands and the United Kingdom share many issues. Both countries could flood from different sources. In many circles, people are thinking about what climatic changes, planning issues, and urbanization mean for the adequacy of current policies, whatever these may be.

In both countries, the basic artefacts are the same. Both use pumping stations, including the emergency pumps so often mentioned. In the Netherlands and the UK, regional water managers do not own all the pumps needed in emergencies. Images like from the UK a year ago with pumps being brought in from external sources are not that different from the Dutch situation. However, Dutch water boards have contracts with private pump suppliers which have to deliver pumps in emergencies. There is even a national coordinating mechanism for supplying pumps in case many are needed.

River embankments are found in both countries as well. Please note that we know that flood safety levels in the Netherlands are higher than in the UK, but that does not necessarily make embankments higher. Embankment dimensions depend on the river it is supposed to keep out. The UK, or should I say London, has its own iconic barrier too, adding another tourist attraction to the city.

At first sight, responsibilities for flood management in the UK are as scattered between the local, regional, and national as in the Netherlands. However, looking at financial arrangements for water management, the image emerges that the scatter in the UK is higher than in the Netherlands. The number of different entities involved in financing UK flood measures is impressive. In the Netherlands, Public Works and the water boards are financed through taxes. The water boards raise their own dedicated taxes, which is a good base for their functioning – although it should be noted that financial safety for the water boards is relatively recent. [19]

Perhaps responsibilities are scattered differently, finances may be complex, and even embankments could be different, but in order to find out what is the case, we need to go a little further into the details. I will discuss two main differences between the Netherlands and the UK. The first difference deals with the hydrology of floods in both countries, the second difference deals with the principal basis for flood response in both countries in terms of investments and effects. Both issues will be presented in a somewhat simplified manner. For those interested in more, let me assure you there is plenty to read! [20]

A first major difference between Dutch and UK flooding issues can be found in the hydrology of the water systems in the areas. Basically, it concerns how rain is transformed into runoff. The main concept of relevance in this simple introduction to hydrology is the lag time, which is a measure of how quick river flow results from rain. When a storm begins, discharge and river levels do not increase immediately because only a small percentage of the rain falls directly into the river channel. The first water to reach the channel comes from surface runoff. Additional flow can come later from infiltrated water.  The gap in time between the time of peak rainfall and peak discharge is called the lag time. A river with a short lag time and high discharge is more likely to flood than a river with a longer lag time and low discharge. Dutch and UK rivers show different lag times.

The typical UK River combines a short lag time with a relatively large discharge. This is probably the most challenging combination of lag time and discharge, as it means that all the rain drains down very quickly from the hills. Seen from that perspective, flooding events like the ones we’ve seen a year ago are not that strange for the UK.

In contrast, the larger rivers in the Netherlands combine longer lag times with relatively large discharges. Floods can be severe, but they are slightly better to predict. Our floods originate in Germany and usually take a few days to arrive. The Netherlands has one rain-river, the Meuse in the south. Most rain for that river falls in Belgium, but the river combines a short lag time with relatively large discharges. It is the shape of the river bed plus the river improvements implemented within the technological regime described above which manages to keep the Meuse under control most of the time.

Knowing that our floods are German does not mean that they can be predicted with the accuracy associated with that origin. Exactly how rain, glaciers, landscape, and other features create discharges and flood patterns is not given. Some of the flood water might flood Germany first. One can imagine that being one day wrong in the prediction of when the flood wave passes can have a huge impact.

In January 1995, predicting the flood wave proved to be difficult. Combined with some river embankments under stress, this made the Dutch government decide that some 250,000 people needed to be evacuated from the central river area. I should know, as I helped to evacuate my parents, their chickens, and the goat. The goat had a great time, as he could spend his time in a stable with a very nice horse. 

It might also be good to remember that the historical expertise of the Dutch is mainly based on their defences against the sea and the rivers. In terms of rainfall, the Dutch reputation is not as strong, perhaps. Nevertheless, rain-induced floods like those in the UK last year are quite often directly compared with Dutch flood policies concerning rivers and the sea. The Dutch do not need a river to be flooded by rain.

A second main difference I would like to mention is the financial strategy in both countries, which is closely related to who assumes the risks and consequences of flooding. The basic approach of the two countries is quite similar. Both countries strive for cost-effective protection measures that offer adequate levels of risk. Despite being a recent concept, risk can be applied to both cases over a longer period to find the basic differences between them.

For an event, the associated risk is usually defined as probability of that event times the results of the event. Although – again – a little simple, for us this definition will work nicely. From the definition, it follows directly that an event with a low probability but with a big result may have a similar risk as an event with a high probability but with low result.

Thinking about flooding in terms of risk helps to explain differences between the UK and the Netherlands, although slightly simplistically. The basic question at stake is how to spend your money: to prevent the probability of an event, or to deal with the effects of that same event? Is one interested in preventing floods or in dealing with the consequences?

The Dutch have chosen for preventing floods. Well, let’s put it in a slightly more specific and correct way: over the years the Dutch flood policy has emerged as being based on preventing that large floods would destroy the economic heart of the country – the western provinces. If these would flood, the country would disappear and the remaining people could start doing something else or migrate. Please note that such a flood would still need to hit at specific places, as the water source would have to be able to flood large areas with much water in a short time. To give an example, which may be reassuring for many of you: even though Amsterdam International Airport is well below sea level, the amount of water available directly in the neighbourhood of the airport is not enough to flood it. The belt system in one of the Northern provinces, however, consists of many lakes and would have the volumes of water needed to fill complete polders.

The embankment areas of the Dutch I discussed before are an expression of that main goal. The Netherlands’ flood policies are based on a hierarchy in safety, grown in history. This hierarchy is now under debate, but it has been a guiding principle for long: protect the west, the economic power. This resulted in measures that either moved river water away from the economic heart or measures that prevented flood water to flow from the east to the west. In practice, this has often meant that people in the east drowned in higher flood water because the water was blocked by embankments controlled by those in the west.

In recent history, suggestions are regularly made that Dutch river flood management needs to be based on new principles. Room for the River, water storage, and nature-friendly design are mentioned. I would argue, however, the same principle has been held for long time and is still structuring Dutch flood policies. In terms of the technological regime theory, I would argue that the guiding principle of river flood management in the Netherlands remains what it has been: protecting the economically and demographically important Western Netherlands. Concerning rainfall floods, a similar idea could be developed: it is simply not allowed to happen anywhere, but especially not in the economic heart of the country. [21]

In contrast to the Dutch, the UK main policy is to allow more frequent flooding, and deal with the effects of flooding through insurance. In a way, quite some money is spent on floods, but some of that money is spent after flooding. I have not come across good cost estimates of these two systems. It is clear that the Netherlands spend more money on flood defences than the UK, but I have not seen how much is spent in total – flood defences and flood insurance.

Within the Dutch system of prescribed high standards, the different governments put quite some money in preventing damage. In the UK, there is obviously governmental investment, but much less, and additional investment is expected from those who benefit directly. Another obvious observation is that the UK has its focus areas for protection as well: protecting London from flooding appears to be more important than of a rural area, as protecting the urban west in the Netherlands is more important.[22] Nevertheless, the UK policy is more focusing on specific investments in localities. As such, the UK is world-leading in being able to maximise return on investment. This is where a major difference between the two countries can be found. [23]

In the UK and the Netherlands, design tools and artefacts are similar. In both countries, design requirements can be found in the safety levels. It is in the guiding principles and/or the closely related promises and expectations that the main differences can be found, with the result that safety levels in the two countries are different indeed.

Dutch flooding policies are shaped based on the idea that floods should be prevented, whereas in the UK, damage needs to be reduced. Both approaches are based on a cost-benefit assessment, which on its own is based on a specific risk assessment, but the results are quite different. In the Netherlands, the insurability of the areas under flood threat is pretty high. The values to be protected in the west – and more and more elsewhere too – would result in enormous insurance fees. Therefore, water needs to stay out, as flooding would typically mean the end of the country. In the UK, the flood risk is different; at least the country’s existence is less at stake. The UK approach is based on flooding being one of the major issues to deal with, not the major one.

Floods and the Dutch: an automatic privilege of knowledge?

This does not mean that flood policies in the UK are worse than in the Netherlands, or vice versa for the sake of argument. Whether policies are good or bad is difficult to determine with such historical contingencies as I tried to touch upon. At the same time, history cannot really be used convincingly to accept any current policies either. The fact that Dutch water policies have strong historical roots does not make them good per definition.

It is interesting to see that it is exactly the historical argument which tends to be used when defending Dutch involvement – both by Dutch and by others. Again, I would not like to downplay the importance of history – being a historian myself – but using the unique history of the Netherlands as evidence for the applicability of a single strategy seems to be strange. We only have one history, so who knows what would have happened in another version? [24] Furthermore, using something like a technological regime would offer as many options to stress the similarity between the UK and the Netherlands as the differences.

Please be assured that I would be the last one – well, one of the last ones – to argue against the Dutch knowing their water or having their water management arranged pretty well. Please be assured that I would be the last one – well, one of the last ones – to argue against the Dutch knowing their water or having their water management arranged pretty well. Especially the financial strength of Dutch water boards when executing their tasks appears to be a success factor. This is recognized internationally as well. During a meeting of the Water Governance Initiative (WGI) in Paris in 2013, an English consultant criticised a concept-report from the Organization for Economic Cooperation and Development on Dutch water governance for missing that point – the final report did mention it.[25]

 I would, however, not argue that this makes the Netherlands the promised land of water management or flood policies. Many Dutch involved in the waters sector would like this to be the case; quite a few complain when experts with other nationalities are asked to draft plans. It is even official Dutch policy to export Dutch water expertise and experts.

At the same time, I would argue that it makes sense to share knowledge and expertise, as indeed much is shared and differences between approaches can help to understand one’s own position. What I have tried to do is to suggest some first ideas, in terms of theory and actual flood management, to understand why Dutch water managers come from a different planet than their UK colleagues. Luckily, they are from the same solar system!

                                                                                                                                      © Dr Maurits Ertsen, 2015

[1]Begemann S.H.A.1926 Ontwikkeling automatische stuwkleppen type Doell-Beauchez in de sectie Modjokerto van de Irrigatie-afdeeling “Brantas”, DWI 9(14)301-310. See also Beauchez A.J. 1921 Automatische stuwkleppen, DWI 9(9)286-289; Kleijn C.H. 1925 Automatische stuwkleppen, DWI 6(13)171-174

[2]Beauchez(1921; 286)

[3]Ertsen M.W.2014 Step after step the ladder is ascended. The delicate balance between landscape and anti-landscapein irrigation. In: Nye D. and Elkind S. (eds) The Anti-Landscape, Nr. 1 Studies in Environmental Humanities, Rodopi, Amsterdam, 45-60; Ertsen M.W.2010 Locales of happiness. Colonial irrigation in the Netherlands East Indies and its remains, 1830 – 1980, VSSD Press, Delft; Ertsen M.W. 2007 The development of irrigation design schools or how history structures human action, Irrigation and Drainage, 56, 1-19; Geels F.W. 2002 Understanding the dynamics of technological transitions. A co-evolutionary and socio-technical analysis, PhD thesis Twente University; Glick T.L.1972The Old World Background of the Irrigation System of San Antonio, Texas, Texas Western Press, El Paso; Horst L.1996 Irrigation water division technology in Indonesia. A case of ambivalent development, Liquid Gold Paper 2, ILRI Special Report; Jurriens M., Mollinga P. and Wester P. 1996 Scarcity by design. Protective irrigation in India and Pakistan, Liquid Gold Paper 1, ILRI Special Report; Van Halsema G.E.2002 Trial and re-trial; the evolution of irrigation modernization in NWFP, Pakistan, Dissertation Wageningen University, the Netherlands

[4]Brouwer R.1987 Design and application of automatic check gate for tertiary turnouts, 13th International Congress on Irrigation and Drainage, International Commission on Irrigation and Drainage, Rabat, 671-683; Burt C.M. , R. Angold, M. Lehmkuhl and S. Styles 2001 Flap gate design for automatic upstream canal water level control, Irrigation and Drainage Engineering. 127(2)84-91.

[5]Constant II E.W.1980. The origins of the turbojet revolution, Johns Hopkins, Baltimore, page 10. See also Bijker W.E. 1995 Of bicycles, bakelites, and bulbs. Towards a theory of sociotechnical change, London, Cambridge, MIT Press; Downey G.L. and Lucena J.C.2004 Knowledge and professional identity in engineering: code-switching and the metrics of progress, History and Technology 20(4)393-420; Picon A. 1996 Towards a history of technological thought, in: Fox R. (ed) 1996 Technological change. Methods and themes in the history of technology, London, Harwood Academic Publishers, 37-49; Picon A. 2000 Technological traditions and national identities. A comparison between France and Great Britain during the XIXth century, in: Nicolaidis E. and Chatzis K. (eds) 2000 Science, technology and the 19th century state, Athens, Institut de Recherches Neohelleniques, 13-21; Picon A. 2004 Engineers and engineering history: problems and perspectives, History and Technology 20(4)421-436; Van de Poel I.2003 The transformation of technological regimes, Research Policy 32, 49-68; Van de Poel I. 1998. Changing technologies. A comparative study of eight processes of transformation of technological regimes, Twente University Press, Enschede

[6]Van de Poel(2003; 49)

[7]Picon(2004; 424)

[8]Two terms are applied: ‘technological tradition’ (Constant II 1980) or ‘technological regime’ (Van der Poel 1998; Van de Poel 2003). I have chosen to apply the regime concept as it is much wider used in discussions on technology development than the tradition concept.

[9]Van de Poel(1998; 16)

[10]From Van de Poel (1998; 17)

[11]See Sewell 2005 Logics of History. Social Theory and Social Transformation. Chicago University Press, Chicagofor a very nice elaboration of Giddens’ theories. As I only apply some restricted ideas of Giddens, basically the idea of rules having a social history and future, I do not include a detailed discussion of Sewell’s improvements of Giddens.

[12]Giddens A.1979 Central problems in social theory: action, structure and contradiction in social analysis, Contemporary social theory, MacMillan, London, page 64. See also Giddens A. 1984 The constitution of society: outline of the theory of structuration, Polity Press, Cambridge; Sewell (2005).

[13]Giddens(1979; 64)

[14]Giddens(1984; 17). See Sewell (2005) for a more elaborate discussion about rules and resources and their differences.

[15]Van de Poel (1998)

[17]Latour 2005, p179-180

[18]Report by Working group 2 (Climate change 2007: Impacts, adaptation and vulnerability), Intergovernmental Panel on Climate Change (IPCC)

[19]On December 4 1963, the Union of Water Boards sent a letter to the Dutch national government to support them with 30% of the exploitation costs. After developments which go too far to discuss here, water boards have developed a pretty sound financial basis, amongst others with their own bank. Havekes H. and Dekking W. 2014 60 Jaar. Hard gegaan. Unie van Waterschappen

[20]Within easy reach online are: Bennett O. and Hartwell-Naguib 2014 Flood defence spending in England. Library House of Commons, SN/SC/5755; Brinkhuis-Jak M., Holternman S.R., Kok M. and Jonkman S.N. (nd) Cost benefit analysis and flood damage mitigation in the Netherlands; Buijs F.A., Van Gelder H.A.J.M. and Hall J.W. (nd) Application of reliability-based flood defence design in the UK; Flikweert J., Hart P. and Spliethoff C. (nd) Space for water: sharing lessons on floodplain management policy between Queensland, the Netherlands and England; Institute for Water Resources 2011 Flood risk management approaches. As being practiced in Japan, Netherlands, United Kingdom, and United States. IWR Report 2011-08; Kolen B., Hommes S. and Huijskes E. (eds) 2012 Flood preparedness in The Netherlands: a US perspective. Netherlands US Water Crisis Research Network; Silva W., Dijkman J.P.M. and Loucks D.P. 2004 Flood management options for the Netherlands, International Journal of River Basin Management Vol2, No2, 101-112; Surminski S., Aerts J., Botzen W., Hudson P., Mysiak J. and Donisio Perez-Blanco C. 2014 Reflection on the current debate on how to link flood insurance and disaster reduction in the European Union, Centre for Climate Change Economics and policy 184, Grantham Research Institute on Climate Change and the Environment 162.

[21]Ertsen M.W. and Ten Horn van Nispen M.L.2005 Deconstructing Dutch Discourses. Modern flood management policies in a historical perspective. In: Ohlig C. (ed) 2005 Integrated Land and Water Resources Management in history. Deutschen Wasserhistorischen Gesellschaft, Sonderband 2, Siegburg. See also De Bruin D. and Schultz B.2003, A simple start with far-reaching consequences, in: Irrigation and Drainage, 52, 51-63; Van de Ven G. (ed), 1996, Man-made lowlands. History of water management and land reclamation in the Netherlands, Utrecht; Vis M., Klijn F., De Bruijn K.M. and Van Buuren M. 2003, Resilience strategies for flood risk management in the Netherlands, International Journal for River Basin Management, 1, 1, 33-40.

[22]Although it is good to realize than works like the Thames barrier – similar to the Maeslant barrier in Rotterdam – protect a larger area than just the large city close to it.

[23]Flooding in Focus 2014, Royal Society for the Protection of Birds. See also Business Reporter July 2014, A special report on flood protection.

[24]Virtual history in itself is interesting, see Ferguson N. (ed) 2011 Virtual history. Alternatives and counterfactuals, Penguin Books (first published 1997 by Picador).

[25]OECD 2014, Water Governance in the Netherlands: Fit for the Future?, OECD Studies on Water, OECD Publishing, p. 212. See Havekes H. and Hofstra M. Water Governance 02/2014, p. 16-27.