History final exam

The Johns Hopkins University Press and Society for the History of Technology are collaborating with JSTOR to digitize, preserve and extend access to Technology and Culture.

http://www.jstor.org

The Oosterschelde Storm Surge Barrier: A Test Case for Dutch Water Technology, Management, and Politics Author(s): Wiebe E. Bijker Source: Technology and Culture, Vol. 43, No. 3, Water Technology in the Netherlands (Jul.,

2002), pp. 569-584 Published by: and the The Johns Hopkins University Press Society for the History of

Technology Stable URL: http://www.jstor.org/stable/25147960 Accessed: 07-04-2015 14:10 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

ESSAYS

The Oosterschelde Storm Surge Barrier

A Test Case for Dutch Water Technology, Management, and Politics

WIEBE E. BIJKER

"God created the world, and the Dutch created the Netherlands." The old

adage summarizes?albeit in an immodest, not to say blasphemous, way?

the popular Dutch view of their relationship to water. There is some truth

in it: about half the country is below sea level and would be flooded with

out the dikes that hold back the waters of the rivers and the sea. But the

relationship is not as straightforward?humans dominating nature?as the

phrase suggests. It is, for example, mediated in complex ways by science and

technology. In this essay I will focus on one recent crisis in this relationship between the Dutch and the sea, the disastrous flood of 1953, and its resolu

tion through the Delta Plan, and in particular the building of the storm

surge barrier in the Oosterschelde.1

Dr. Bijker is professor of technology and society at the University of Maastricht, Faculty of Arts and Culture.

?2002 by the Society for the History of Technology. All rights reserved.

0040-165X/02/4303-0006$8.00

1. I am grateful to Martin Reuss and John Staudenmaier for inviting me to con

tribute this essay. It allows me to address Dutch coastal engineering more fully than I did

in two previous publications, which had a primarily methodological purpose. And, in a

way, it serves to fulfill an old dream. It is only because I did not want to sit in my father's

classes that I studied physics rather than civil engineering, but my fascination with the

water sorcerers never faded. This essay gives me an opportunity to return to this old fas

cination, albeit under the banner of the history of technology. The term "water sorcerers" was coined by Den Doolaard in Het verjaagde water. This

1948 novel gives an engaging and historically accurate account of the 1945 closures of

the dikes that were bombed by British planes to drive the Germans out of the polders in

the southwest of the Netherlands. The novel, which inspired Samuel Florman to write his

reflections on being an engineer, was translated into nine languages, and has recently

been republished by the Delft University of Technology with several appendices giving additional technical and historical information. A. den Doolaard, Het verjaagde water,

ed. Kees d'Angremond and Gerrit-Jan Schiereck (Delft, 2001); A. den Doolaard, Roll

Back the Sea, trans. lune Barrows Mussey (New York, 1948); Samuel Florman, The Exis

tential Pleasures of Engineering (New York, 1976).

569

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

JULY

2002

VOL. 43

Technology has always played a central role in the relationship between the Dutch and the sea. From the earliest mound constructions, built to keep farm houses and outbuildings dry during the frequent floods, to windmills and steam-driven pumping stations the Dutch have actively tried to control their environment with technology.2 But the science and technology needed for the Delta Plan, and especially the research and high-tech solu tions used in the construction of the Oosterschelde barrier, constituted a

radical departure from centuries-old traditions.

During the nineteenth century and the first half of the twentieth cen

tury, relations between government agencies and private construction com

panies involved in the building and maintenance of dikes, locks, sluices, and other water control structures were

subject to routines and procedures

that provided for adequate checks and balances. The central government agency responsible for the water control system, the Rijkswaterstaat, typi cally designed harbors, dikes, sluices, bridges, and so on, and then con

tracted the construction out to private companies. These companies sub

mitted bids, sometimes joining together in consortia when the project was

big and complicated, and the company or consortium with the lowest bid received the contract. Once construction

began, the Rijkswaterstaat moni

tored the process. This style of management was radically changed for the Oosterschelde project.3

The earliest forms of democracy in the Netherlands were related to dike and sluice maintenance and management. From the twelfth century

on

ward, specialized water boards (waterschappen), supervised by elected

councils, assumed responsibility for local dikes and sluices. These boards

constituted a highly decentralized form of democracy in which all land owners had voting rights, with the weight of each vote depending on the extent of the landowner's property. The Delta Plan can be seen as a funda

mental change in the balance between local and national water politics. The Delta Plan, and particularly the Oosterschelde project, precipitated

a crisis involving three aspects of the relationship between the Dutch and the sea: technology, management, and political culture. I will argue, how

ever, that in the end that crisis only reinforced the basic characteristics of this relationship.

The 1953 Flood

On 31 January 1953, a Saturday night, ebb tide did not bring a lowering of the water level as it always does. Then, as the tide began to come in, a

2. See Petra van Dam's, Arne Kaijser's, and William TeBrake's articles elsewhere in

this special issue for accounts of early sluice technology, the implications of windmill

development for political institutions, and drainage technology. 3. On the history of the Rijkswaterstaat, see Harry Lintsen's essay elsewhere in this

issue.

570

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

Bijker I The Oosterschelde Storm Surge Barrier

^^^^^^^^^^^^^^^^Hj^^^^^^^^^^^^^^^^GK ESSAYS

pHH^^!^ * ^^^W?pll^^"^MM|l^^HJH^^^^^^^^^^^^^^^^^^^^^BBB^^^^^^^^^^^B

FIG. 1 A broken dike, 1 February 1953. (De Ramp: Nationale uitgave [Amster

dam, 1953].)

storm pushed the water to higher than normal levels. In the early morning of 1 February the sea reached the top of the dikes in Zeeland, at the south ern end of the Dutch coast. Waves started to nibble at the back slopes of the

dikes, which are not armored by stones, undermining them from the rear,

and eventually the dikes broke. Quickly the breaches were scoured out by the seawater rushing into the polders, several meters below sea level (fig. 1).

Analyses later showed that it had been neither a particularly high spring tide nor an exceptionally strong

storm. It had, however, been a long-lasting

storm, and, crucially, one that had changed direction in a very particular manner at exactly the wrong moment. A northerly wind had first pushed the flood wave along the British coast toward the narrow channel between

England and the Netherlands. Just as this tidal wave reached the Dutch coast the wind veered to the west, sending the water more forcefully against the coast.4

It took several days before the extent of the disaster became clear to the rest of the Netherlands, as communications with the affected areas had bro

ken down and there were no helicopters and but few aircraft. In one week,

1,835 people drowned. More than 750,000 inhabitants were affected, and

200,000 hectares of land were inundated (fig. 2). The effects were trau

matic, both for individuals and for the Netherlands as a country. This

became particularly clear in the 1970s, when political discussions about water management

were cast in terms of safety versus

ecology.

4. Rijkswaterstaat and Koninklijk Nederlands Meterologisch Instituut, Verslag over

de stormvloed van 1953 (The Hague, 1961).

571

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

FIG. 2 Zeeland, In the southeast corner of the Netherlands. The white parts were flooded in 1953. (Courtesy Rijkswaterstaat Archief, the Hague.)

The whole gamut of technologies that had been developed during cen

turies of keeping the sea out were employed to reclaim the lost land.5 Time was a crucial factor. Tidal currents quickly widen any breach in a dike. The

largest breach in the 1953 disaster was 100 meters wide and 15 meters deep on 1 February, but within a few months it had grown to 200 meters by 20

meters. If the breaches were not closed before the next winter season, the

damage might become irreversible. Time was critical on the scale of min utes as well as months: currents rage at their fastest where breaches are at

their smallest, so the right moment to close off a breach in a dike is during the few minutes of slack water.

For centuries the key material used to strengthen and repair dikes has

been sand in jute bags. On the night of 1 February 1953, sandbags were

made available from emergency depots and played a crucial role in bat

tling the flood. Sand is readily available and very heavy, but unpacked sand would immediately be swept away by the water?hence the jute sacks. Only with the enclosure of the Zuider Zee in the 1920s did keileem, a heavy clay from glacial moraines, come to be used to build dikes so large

5. Johan van Veen, a Rijkswaterstaat engineer from the 1920s to the 1950s, gives a

historical review of early Dutch coastal engineering technologies in Dredge, Drain, Reclaim: The Art of

a Nation, 5th ed. (The Hague, 1962). Before 1940 van Veen developed several plans to close tidal inlets in Zeeland, and these played an important role after

1953. Since 1937 he had warned of the deplorable state of dike maintenance, to no avail.

He appended a critical analysis of the 1953 disaster?under the pseudonym "Cassan

dra"?to the fifth and last edition of his book.

572

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

Bijker I The Oosterschelde Storm Surge Barrier

that sandbags could no longer serve as the feasible core and beginning of

a dike.

In 1953, as in the centuries before, human power did most of the work, in combination with the skills needed to move the sandbags by human

chains and place them where they would do the most good. Muscle power was the only energy source distributed widely enough through the Dutch

coastal area to act adequately at short notice. Dredges, tugs, ships,

and

cranes would eventually be called in to close the breaches in the dikes, but

on that February night everything depended on human hands.

An armored foundation is necessary to build a dike in a gap where tidal

currents flow. For centuries fascine mattresses consisting of a net structure

about 50 centimeters thick, 100 meters long, and 20 meters wide have been

used for this purpose. A series of such mattresses lowered onto the seabed

provides a foundation for the dike. Until the 1970s the dikes in the Nether

lands were built on mattresses woven by hand from branches of willow

trees or similar material.6 The mattresses were fabricated on land, then

towed out to sea and sunk by carefully dumping quarry stone on them. This

was done by hand, to ensure that the mattress was lowered gradually and in

a controlled manner into the right position (fig. 3). These basic technologies were used to good effect in 1953. In the

decades that followed, however, radical innovations were developed and

new high-tech tools created for building dikes, sluices, and storm barriers.

When one looks carefully, though, the same basic techniques (usually

excepting manual labor) are still deployed in all hydrological projects.

Early Water Politics

There are such striking similarities between early water politics and the

present political culture in the Netherlands that it is illuminating to briefly review the history of the political systems that have governed Dutch water

management since the Middle Ages.7 Around the beginning of the previous millennium the first collective organizations developed to maintain dikes

and sluices. In the twelfth century the water boards were established, the

first democratic institutions in the Netherlands, which still exist today. These statutory organizations were (and still are) governed by councils

elected by landowners whose voting rights correspond to the size of their

ESSAYS

6. This is a Dutch technique that was transferred in the twentieth century to other

countries, where bamboo was often used in place of willow branches. Without this mat

tress technique dikes have to be built on a bed of gravel built up of several layers, each

using larger stones than the one below it, which is much more difficult to construct.

7. See, in particular, Frans van Waarden, notes from a lecture titled "Truth in the

Stereotypes? Or Hydraulics and Dutch Political Culture and Institutions," Wassenaar,

1999, copy in the author's possession.

573

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

july iggBHlHj^^^^^^^^H^^^^H^^^^H 2002 iIH^B^^H^^^^^^^^^^^^^^^^h

FIG. 3 A willow mattress being sunk. (Kees Slager, De Ramp: Een reconstructie

[Goes, 1992].)

properties. The duties of the boards included such communal tasks as

drainage, dike maintenance, and sluice management. They had the power

to levy taxes, and some acquired additional legislative, judicial, and execu

tive powers. A few times each year they conducted inspections, and when

parts of the hydraulic infrastructure were found to be out of order those

responsible were severely fined. Only during the eighteenth century did a

more centralized system of oversight gradually develop, and in 1798 the

first national agency, the Rijkswaterstaat, was established.8

Dutch political culture still exhibits several characteristics that can be

traced back to this early history of water politics. First, there is a certain trust

8. For more details on early Dutch politics and water management, see Kaijser's and

TeBrake's articles elsewhere in this issue.

574

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

Bijker I The Oosterschelde Storm Surge Barrier

in technical solutions and in technocracy?perhaps not as much as in

France, but more, for example, than is found in Germany. Indeed, close links

exist between policy makers and scientists (including social scientists) and

engineers. A sense of vulnerability, because of the centuries-long threat from

high water, is compensated for by a capacity to react swiftly to crises. In reac

tion to a crisis, Dutch politics will often take a pragmatic approach to find

ad hoc and flexible solutions, even when this means flexibly interpreting reg ulations.9 The Dutch have a long tradition of planning and actively shaping their environment. This applies not only to the physical landscape of the

Netherlands but also to society; Dutch political culture displays a general belief in the malleability of society. Finally, the political culture of the

Netherlands is distinctly consensual and oriented toward cooperation and

compromise.10 This is not to say that there are no opposing interests

or con

flicts. But in the end the Dutch need to cooperate with each other, under

penalty of being flooded.11 In the 1950s the restoration of the prewar polit ical culture strengthened many of these characteristics. In this essay I will

argue that this strengthening process culminated in the Delta Plan that was

adopted after the 1953 disaster. However, during the Oosterschelde enclo

sure, the final step in the Delta Plan, the process produced a crisis.

Since the end of the nineteenth century the construction of dikes and

other large infrastructural works had been organized in a straightforward manner: the Rijkswaterstaat designed projects and then contracted with

private companies to carry them out under the supervision of Rijkswater staat engineers. The distinct duties and responsibilities of Rijkswaterstaat and contractors were clear, and the dividing line between the two was

unambiguous. Numerous stories convey the almost sporting relationship

between Rijkswaterstaat inspectors and the chief engineers of the dredging companies, both trying

to get the better of the contract.12

ESSAYS

9. Examples that do not concern water are the contemporary policies related to

abortion, prostitution, and drugs. 10. For a discussion of the implications of this characteristic for housing politics after

World War II, see Wiebe E. Bijker and Karin Bijsterveld, "Women Walking through Plans:

Technology, Democracy and Gender Identity," Technology and Culture 41 (2000): 485-515.

11. An example of such pragmatic cooperation?and rule stretching?comes from

the final days of closing the breaches made in the dikes in 1953. It was the beginning of

autumn and time was running out; if the gaps were not closed quickly the autumn

storms would scour them out beyond repair. The engineers of the construction company wanted to make the final move on a Sunday, when the tidal currents would be at their

weakest. The workers from this region of very strict Calvinists initially refused, because

that would be breaking the Sabbath. After long talks, and when they recognized the

hydrological necessity, they decided to cooperate?but only on condition that they not

be paid. Eco W Bijker, interview by author, Maassluis, 29 June 2001. Eco Bijker, my

father, was one of the young engineers involved in the repair work; he later became

deputy director of the Delft Hydraulics Laboratory and professor of coastal engineering at the Delft University of Technology.

12. Although the distinction between the Rijkswaterstaat and contractors was clear

575

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

JULY

2002

VOL. 43

This transparent relationship had crystallized after the failure of

Rijkswaterstaat to manage the design and construction of the Rotterdamse

Waterweg (1857-77), connecting Rotterdam to the sea at Hoek van Hol

land. The history of the enclosure of the Zuider Zee, the inland sea east of

Amsterdam, in the 1920s and 1930s added two important elements to the

set of instruments that related the Rijkswaterstaat to contracting compa nies.13 The first was the creation of temporary consortia, lasting for the

duration of a project, large (and rich) enough to carry the risk of the proj ect. Four of the largest Dutch dredging and building companies joined forces and established the consortium Maatschappij tot uitvoering der

Zuiderzeewerken (MUZ) as a limited liability company for the duration of

the Zuider Zee project. The second innovation, closely related, was the use

of the raamcontract, or frame contract.14 In a frame contract the state

agency grants the construction of the whole project to the building con

sortium without specifying the details of the various individual structures.

These structures, which together constitute the whole project, are then

specified in separate contracts. The private companies thus receive

assur

ances of their long-term involvement, which they need to make the neces

sary technological investments, and the state agency is still able to specify the particulars of the separate subprojects, which is necessary if it is to exer

cise detailed oversight. The frame contract for the Zuider Zee project also

specified that the contracting consortium would take all of the first 6 per cent of profit or loss, while losses or profits exceeding that amount would

be shared with the state.

This combination of a legal framework and a culture of competitive

collaboration between engineers of the Rijkswaterstaat and the private

companies formed the starting point of the Delta Plan works, and indeed

culminated during the first phase. But, in concurrence with the crisis in the

political culture, the balance of power in this relationship shifted radically

during the Oosterschelde project.

cut, everyone also realized that they needed each other. Additionally, all civil engineers were trained in the same school?the Delft University of Technology?and many who

worked on opposite sides of these construction projects had been classmates in earlier

times.

13. The Zuider Zee project presents a discontinuity in the history of the Rijkswater staat. Instead of granting the Rijkswaterstaat oversight of this large national project,

a

separate Zuider Zee agency was established and given responsibility for its management. See D. M. Ligtermoet and H. De Visch Eybergen, Uitvoering

en uitbesteding: Ontwikke

lingen in de organisatie van waterbouwkundige werken bij de Rijkswaterstaat, vol. 52 (The

Hague, 1990). 14. The term raamcontract was not used in the 1930s. The character of the contract

used then, however, is the same as the one used during the Delta Plan, when the label

raamcontract was introduced. See Ligtermoet and Eybergen.

576

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

Bijker I The Oosterschelde Storm Surge Barrier

The Delta Plan

Three weeks after the 1953 flood, a governmental committee was

formed. One week later the committee put forward an interim version of

the Delta Plan that called for closure of all tidal outlets except the most

northerly and southerly ones, connecting Rotterdam and Antwerp to the

North Sea.15 As could be expected from the Dutch political culture?swiftly

reacting to a crisis with pragmatic solutions supported by a broad consen

sus?the implementation of the Delta Plan started even before proper

political procedures had been completed. In August 1955 the Delta Project

unofficially began with the building of two working harbors. On 1 May 1956 a new department within the Rijkswaterstaat that would be responsi

ble for carrying out the Delta Plan was established. Only in November 1957 was the Delta Law debated and adopted, by a great majority, in parliament, to take effect on 8 May 1958. Formal decisions ran almost three years behind material decisions.

When the Delta Law was adopted, some of the planned closures were

beyond the technical capabilities of the day. The Rijkswaterstaat engineers used the phrase "Delta school" to stress that in the course of the first phases of the Delta Plan the knowledge, skills, and technologies needed to make

the most ambitious closures in the last phase possible would have to be

acquired. One aspect of present Dutch hydrological practice came to

fruition during the Delta Plan: the integration of scientific research and

technological design. This development culminated in the Oosterschelde

enclosure, but crucial first steps were made in the first phases, and indeed

during the Zuider Zee enclosure.16

The first example of the integration of scientific research with hydrau lic engineering dates from the 1920s. The physicist Hendrik A. Lorentz

was

asked to make mathematical predictions about the tidal effects caused by a closure of the Zuider Zee. Empirical research using scale models began in the 1930s and intensified following the war. The Delft Hydraulics

Laboratory, center of this modeling research, received important financial

support under the Marshall Plan. Scale models developed there played a

crucial role in the closure of the last breaches of the 1953 flood. The closure at Zierikzee, for example,

was carried out many times in the labo

ESSAYS

15. The name "Delta Plan" was invented by the director general of the Rijkswater staat, A. G. Maris, renowned for his inventiveness in coining new words for new con

cepts. H. A. Ferguson, Delta-Visie: Een terugblik op 40 jaar natte waterbouw in Zuidwest

Nederland, vol. 49 (The Hague, 1988). It acquired such a magic ring of urgency, nationwide support, and effectiveness that decades later politicians could propose a

"Delta Plan" for art restoration or a "Delta Plan" for restoring the safety of river dikes.

16. For an internal history of Dutch coastal engineering, see Eco W. Bijker, "History and Heritage in Coastal Engineering in the Netherlands," in History and Heritage of Coastal Engineering, ed. Nicholas C. Kraus (New York, 1996), 390-412.

577

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

JULY

2002

VOL. 43

ratory.17 There researchers, using springs to measure tidal forces, held the

cables and played the winches holding the last caisson to be eased into the

gap by the last remnants of high tide.18 (If it was to be finished before the

ebb tide gained force, the operation had to commence while the flood was

still strong.) The day came for the breach to be closed, and the young engi neers who had practiced in the laboratory stood on deck behind the older

experienced workmen. When one of the cables snapped and control of the

caisson was about to be lost, they were able to intervene because they had seen that snapping rope a dozen times in the laboratory model and had

elaborated a scenario to save the caisson. With a series of unusual com

mands that took advantage of the queer characteristics of the currents

they had identified in the lab, the last caisson was eased down into the final

gap during the crucial few minutes of slack water. The breach was closed.19

During subsequent stages in the Delta school?from the closure of the

Veerse Gat with caissons (1961), to the closure of the Haringvliet with a

large system of discharge sluices (1971), to the closure of the Brouwer

shavense Gat in the Grevelingen with a combination of caissons and blocks

of concrete dumped by a cableway (1972)?new technologies developed hand in hand with further scientific research.20 Eventually only the last and

most difficult closure remained: the Oosterschelde, 8 kilometers wide at the

opening, 20 to 40 meters deep, with 1.1 billion cubic meters of water mov

ing in and out at each tide, four times a day. A site was selected for the dam that made use of two large sandbars in

the mouth of the Oosterschelde. The parts of the dam that would extend over the sandbars posed only minor problems, leaving three deep gaps to

be closed. In 1971 it was decided to close these using the technique that had

been employed in the Brouwershavense Gat: a huge cableway to drop the

large concrete blocks that would form the core of the dike with great preci

17. This was done with caissons?except that, since no sophisticated caissons were

available, old barges were used; these were, quite spectacularly, sunk with dynamite. 18. For a more general discussion of the

use of modeling in science and technology,

using the same case of Dutch hydraulic coastal models, see Bruno Latour, Science in

Action: How to Follow Scientists and Engineers Through Society (Cambridge, Mass., 1987).

19. One of these young engineers was my father, Eco W. Bijker. Model research is no

guarantee of success, however. For one thing, it depends on whether you have modeled

all relevant aspects. Though the Zierikzee closure first seemed a success, a few days later

the caissons started to shift. Since the Rijkswaterstaat and the building companies had

not wanted to lose time laying a fascine mattress foundation, the ground

was too slip

pery and the caissons were pushed out of the gap. 20.1 do not list the extra storm barriers, dikes, and locks that were built at the inland

side of the large tidal basins. These are necessary to control the water level while allow

ing for discharge of the Maas and the Rhine and ship traffic. See Ferguson, Delta-Visie,

and Dialoog met de Noordzee: 2000 jaar Deltawerken (Hippolytushoef, 1991); R.

Antonisse, De kroon op het Deltaplan: Stormvloedkering Oosterschelde?Het grootste

waterbouwproject aller tijden, rev. ed. (Amsterdam, 1986).

578

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

Bijker I The Oosterschelde Storm Surge Barrier

sion from a height of some 30 meters. Three independent cableways were

to be constructed for the three gaps. Twelve pylons were to be built, at a cost

of 17.5 million guilders, each designed for the conditions at its position in

the mouth, the tallest reaching some 80 meters skyward. The last pylon was

to be placed in July 1974.

But then the nationwide support that the Delta Plan had received in the 1950s started to wear thin. The special quality of the tidal ecology of the

Oosterschelde was valued more than before: the polluted waters of the

Rhine and Maas threatened to transform their closed estuaries from trans

parent lakes into huge sinks, and the butter and wheat "mountains" in the

European Community diminished the importance of providing freshwater

to benefit agriculture, since food production did not seem to be pressing a

problem as it had been immediately after World War II.21 Other societal

changes in the 1970s affected the project as well. As happened with so many other political institutions in the Netherlands, the Rijkswaterstaat's author

ity was challenged. During the general elections in 1972 the Oosterschelde

closure became a political issue, and an alternative plan,

to leave the

Oosterschelde open and increase the height of its 150 kilometers of dikes, was

proposed.

The new government, now

including the social-democratic and leftist

liberal parties, decided to investigate the possibility. A commission was

formed in August 1973, and in February 1974 it produced a report recom

mending that a porous flood barrier be built in the mouth of the Ooster

schelde, consisting of a dam of concrete blocks, that would allow seawater

to pass through but reduce the tidal difference in the Oosterschelde basin

by some 50 percent. The commission's report played a crucial role in opening up the dis

cussion, although it was criticized from all directions. Ecologists argued

that the commission had not seriously investigated the "null option" to

leave the Oosterschelde open. Several other groups concluded that Zeeland

was left unprotected against the sea for a much longer period than

was

promised in the Delta Law, and most engineers criticized the plan for being technically impossible.22 Whatever the report's technical merits and short

comings, the option of a

half-open Oosterschelde was now on the agenda.

The debate split the Netherlands completely, and the traumatic experience of the 1953 disaster only made the controversy more bitter. The consensual

political culture of the Dutch broke down, with fault lines running though all parts of society, from government and parliament through the commu

ESSAYS

21. For a comprehensive account, with special attention to the increasing role of

environmentalists and ecological scientists, see Cornells Disco, "Remaking 'Nature': The

Ecological Turn in Dutch Water Management" Science, Technology and Human Values

27, no. 2 (2002): 206-35.

22. For one thing, the gaps in the dam would quickly fill up with cockles and sedi

ments.

579

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

JULY

2002

VOL. 43

nity of engineers to provincial and local administrative centers and down

to the level of individual families.23

The Oosterschelde Barrier

In 1974, with a governmental crisis threatening, parliament reluctantly

accepted a compromise: to partly close the Oosterschelde with a storm bar

rier caisson dam?a dam consisting of caissons that are normally open but

close when a storm approaches. Three additional conditions were set: (1) the

plan should be technically sound, (2) the barrier should be finished not later

than 1985, and (3) the extra costs, as compared to a complete closure, should

not exceed twenty billion guilders. A countermotion to continue with a

complete closure was

rejected, 75 votes to 67. Those who favored closure

called this "a purely political decision."24 Quickly it became clear that the

political compromise was technically impossible. But at the same time, pop ular mistrust of the Rijkswaterstaat had reached its height. Since this agency had always been in favor of carrying out the original Delta Plan, including the complete closure of the Oosterschelde, the parliamentary decision was

viewed by friend and foe alike as a slap in the face of the Rijkswaterstaat

engineers. H. A. Ferguson, director of the Deltadienst, the department within the Rijkswaterstaat that carried out the Delta Plan, realized that his

department was?at least temporarily?sidetracked. The people rejoiced in

seeing the Rijkswaterstaat brought to its knees. It was a political drama.25

Then the dredging companies stepped in, and in a new way.26 They

were

given the contract to codesign the

new barrier, an unprecedented level of

involvement that further blurred the boundary between the state and pri vate contractors. This process had begun with the frame contracts, but

never before had the construction companies been so centrally involved in

designing a whole project. A single integrated project team was established

comprising engineers of four building companies, the Delft Hydraulics Laboratories, and the Rijkswaterstaat. The team started scientific modeling research into several alternative designs.

Model research had been accepted by the building companies since its

23. In my case: father still gave priority to safety and thus preferred a complete clo

sure; sons, young engineering students in Delft, sided with the environmentalists and

advocated an open Oosterschelde; and mother mediated to keep the family together. 24. This is of course a rather trivial label for a decision taken in parliament, but what

they meant was a technically uninformed decision.

25. H. A. Ferguson, interview by author, Voorburg, 15 March 1993.1 did this inter

view with Eduard Aibar and Rob Hagendijk. 26. Age J. Hoekstra, one of the directors of the large dredging and construction

com

pany Volker, commented on the plan to create a half-open Oosterschelde: "As a civil engi

neer I thought it a silly idea, but as a contractor I saw a great project down the road."

Interview by author (with Rob Hagendijk), Oostvoorne, 31 March 1998.

580

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

Bijker I The Oosterschelde Storm Surge Barrier

^^^^^^^H^Z^^^^^^^H^^^^^^^^^^G^^^^^^^I ESSAYS

FIG. 4 Left: The Delta Plan, 1958. Right: the revised Delta Plan, 1976. (Courtesy

Rijkswaterstaat Archief.)

contribution to the 1953 closures, and it played a crucial role in different

stages of the Delta Plan. In physical models, dimensions are scaled down by factors of one hundred and four hundred, time is scaled up by a factor of

forty, sand is scaled down by using finely ground Bakelite, and water

remains water at a scale of one to one.27 The most complicated models,

such as the Oosterschelde model, used a combination of salt and fresh water. For detailed studies of dikes and constructions, wind and wave

flumes were used. The organization of this model research was as difficult

and crucial as interpreting the scaling principles. Managing the relations

between the Rijkswaterstaat, the Delft Hydraulics Laboratory, and the pri vate construction firms was thus as much part of the Oosterschelde project

as the weaving of mattresses or the design of the

storm surge barrier.

A final plan was presented to the government and approved in June 1976 (fig. 4). Debate in parliament descended even to such details as the

size of the door openings in the construction, the construction schedule, and the budgetary controls. If ever a technological system deserved the

label "designed by committee," this was it. The core of the adopted solution was to build a permanent structure in the mouth of the Oosterschelde

through which the tide would flow four times each day, and which could be

closed completely in case of a large storm. The principles of this solution were in all details different from that which the parliament had approved in

1974, and even in 1976 most of the research and design work remained to

be done. The engineers of the Rijkswaterstaat and the construction compa nies worked in fully integrated teams toward this end. Next stages in the

27. Vertical downscaling, for example 100:1, cannot be as

large as the horizontal

downscaling, for example 400:1, because water's behavior changes fundamentally when

flowing in more shallow streams. This is one example of the complicated principles of

scaling involved in all technological modeling. Consequently, results from a model can

not be translated to full scale in any unambiguous or "objective" way, just as the results

of scientific experiments cannot be taken to provide unambiguous answers about the

state of Nature.

581

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

JULY

2002

VOL. 43

design process were discussed in parliament

as late as 1977. Outside parlia

ment, some crucial decisions?concerning, for example, the use of caissons

or pillars?were made

even later. One key decision was not to use caissons

with integrated sliding doors but rather to hang the sliding doors between

concrete pillars.28 These pillars, which numbered more than sixty, were of

cathedral-like dimensions: some 35 meters high, weighing 18,000 tons.

They were built in dry dock and moved to their final positions by a specially built vessel. This mode of transport was made possible by the pillars' buoy ancy; they were built with hollow interiors, which were filled with sand once the pillars were positioned. The accuracy of the whole operation could

be measured in centimeters.

In 1981-83 a series of further crises in the storm surge barrier project

developed. Although technological and scientific uncertainties lay at the

roots of these crises, they took the political shape of predicted budget over

runs. Clashes between parliament and government resulted in political

compromises?design changes to make the project cheaper combined with

acceptance of larger budget overruns. In a rather desperate last budget cut,

the minister of public works decided in 1984 to use one fewer pillar and one

fewer sliding door.29 The decision had undesirable ecological effects, but

budgetary problems had taken priority by that time. On 4 October 1986

Queen Beatrix of the Netherlands officially opened the Oosterschelde

Storm Surge Barrier (fig. 5). Since 1986 it has been used to counter storm

surges about once a year.30 And the thing still works.

Technology, Management, and Politics

The Oosterschelde barrier plunged the Netherlands and Dutch water

management into deep crisis. It generated a profound political conflict that

left no level of society untouched and revealed an unprecedented mistrust

in the central water authority, the Rijkswaterstaat, thereby temporarily

eroding an

important element in the institutional structure of water man

agement in the Netherlands. It also presented hydrological engineers with a

challenge they had no idea how to meet. Between 1974 and 1986 this

changed the world radically, or so it seems. Protection against flooding

28. Frank Spaargaren, chair of the Rijkswaterstaat Project Bureau Afsluiting until

1979, recalled how uncertainty about the special fluidity of the Oosterschelde seabed

tipped the balance in this case. Interview by author (with Rob Hagendijk), Garderen, 19

May 1998.

29. The pillar had already been built, and can still be seen standing in the dry dock,

next to the visitors center?called Neeltje Jans after its location on the former island of the

same name?on the barrier. Mountaineers now practice climbing on the walls of this

dinosaur-like remnant of techno-optimism. See www.neeltjejans.nl for the visitors center.

30. For an evaluation of the first five years, see Rijkswaterstaat Directie Zeeland,

Veilig Tij: Evaluatie van de Oosterschelde na 5 jaar stormvloedkering (The Hague, 1991).

582

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

Bijker I The Oosterschelde Storm Surge Barrier

J^^^^^BvYl'If^H ^|p 1 BP^ HB^Mg, ESSAYS

FIG. 5 The Oosterschelde Storm Surge Barrier. (Courtesy Rijkswaterstaat Directie

Zeeland.)

came to be weighed against ecological concerns. The Oosterschelde was not

closed, but defended with sliding doors. The Rijkswaterstaat lost its central

role in Dutch society. The balance of power between state and private sec

tor shifted, and a unique joint venture of the Rijkswaterstaat and private contractors took charge of the project. And, finally, the science and tech

nology required were so innovative that even after the barrier was finished

some engineers still could not believe it would really work.31

When we take a close look, however, we can see an argument to be made

for continuity as well. Nobody questioned the basic safety goals of the Delta

Law; ecological concerns were added to it. With the help of the 1972 Club

of Rome report The Limits to Growth, which had a particularly significant

impact in the Netherlands, ecological concerns could also be translated into

safety terms, but on a

larger scale.32 All parties involved, including the envi

31. In the beginning the fact that the barrier worked had surprised some engineers who were particularly suspicious of the Oosterschelde seabed. Although they had given the exceptionally fluid sand special treatment and used extra foundation mattresses, they remained afraid that the pillars would shift and the sliding doors would jam. Now confi

dence has risen, and the barrier is generally expected to hold up for at least two centuries.

32. Donella H. Meadows et al., The Limits to Growth: A Report for the Club of Rome's

Project on the Predicament of Mankind (New York, 1972). It sold more than two million

copies all over the world, but the Dutch translation sold more than a hundred thousand

copies in a single month. Maarten A. Hajer, The Politics of Environmental Discourse: Eco

logical Modernization and the Policy Process (Oxford, 1995).

583

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions

TECHNOLOGY AND CULTURE

JULY

2002

VOL. 43

ronmental action groups, were after solutions that could gain broad accept ance. And thus collaboration reemerged, not only between the Rijkswater staat and the building companies but also between the hydrological engi neers and the ecologists. The Rijkswaterstaat recuperated after the slap in

the face and regained control over the process, although for the contracting

companies the Oosterschelde barrier remained one of the sweetest projects ever. Afterward its revival continued, and by the end of the last century the

agency had recovered its central institutional position in integrated water

management. The hydrological science and technology deployed in the

project were indeed radically innovative, but could only be developed from

the basic techniques of previous centuries through the gradual learning process of the Delta school.

No surprise, then, that all involved?including the Rijkswaterstaat, the

construction companies, environmental action groups, and politicians?

are now happy with the barrier. Success has many fathers, and Dutch suc

cess even more so.

584

This content downloaded from 144.118.150.21 on Tue, 07 Apr 2015 14:10:05 UTC All use subject to JSTOR Terms and Conditions