Since the 1960's in Japan, troubles have often arisen between residents and the enterprise who undertaking construction of public works. Accordingly the construction sometimes involves much time and effort that can be very costly, and can even cause cancellation of the construction plan itself. Such problems of "public acceptance (PA)" are a very important issue in Japan. On the other hand, civil engineering researches have been systemized and grown fast in Japan since the Second World War. So it is important to study the mutual relationship and influences between those scientific researches and public acceptance. In this paper focusing on dam construction, the characteristics of research in this field are examined. The 497 reports concerning dam construction are selected from the Proceedings of the Japan Society of Civil Engineers, 1944-1993. By calculating the entropy of research, the changes in research activities in this field were studied. It was found that throughout the whole period there is a tendency of increasing entropy, and that the researches have "dynamic properties" such as variation and concentration of activities. Also, it was found that problems concerning the safety and environment of dam construction which cause problems with public acceptance, affect the dynamic properties of research.

There is no doubt that the main goal of engineering is human welfare, in other words, to contribute to the improvement of human living and industry. For instance, the Japan Society of Civil Engineers defines civil engineering as "technology to plan and to complete facilities necessary to construct land, to improve environment and to protect against various kinds of disasters."

Nevertheless, just when construction is about to be undertaken, troubles may often arise between residents and the enterprise who undertaking the construction. In the western powers like Japan, this can sometimes lead to a settlement in court. Under such circumstances, construction sometimes involves much time and effort that can be very costly, and can even cause cancellation of the construction plan itself. Such problems of "public acceptance" are a very important issue in Japan for civil engineers who work for human welfare.

Civil engineering research was systemized and grew rapidly in Japan after the Second World War, as universities expanded and private- and government-run research facilities were built. Taking a look at the number of university graduates from civil engineering department nationwide, only 650 people graduated in 1941, but in 1947 2,330 people graduated and the number increased to over 10,000 in 1975 in Japan. Also, there is data showing an increase in the number of members of the Japan Society of Civil Engineers and the number of papers released in the Proceedings of the Japan Society of Civil Engineers since the war. The increase in the number of research papers indicates active research in this field.

It is difficult to gain public acceptance simply by promoting the public benefits. The problems of safety and the environment, which are major issues affecting public acceptance, are always major concerns among civil engineering researchers as well. It is thus important to clarify the mutual relationship and influences between scientific researches and public acceptance. In this research there are two questions, the first is, "Does difficulty in public acceptance have any effect on research on civil engineering?" The second is "Does research on civil engineering have any effect in solving the difficulty of being accepted by the community?" The answer to these two questions will lead to a proposal on the ideal style of scientific researches in civil engineering.

First, consider question one. Focusing on research on dam construction, let us examine the characteristic of how construction and public acceptance by the community are related.

The present paper studies research on dam construction after the Second World War by analyzing research papers from the Proceedings of the Japan Society of Civil Engineers published since 1944.

Dam construction is one of the largest scale works undertaken by humans. It is largely influenced by nature, both by geological and hydrological aspects, but it also largely influences the natural and social environment. As a result, many researches on dams have been conducted and the influence of dams on safety and the environment has been discussed. Then, research on dam construction is taken here as being typical of civil engineering research, and is considered in detail.

The Proceedings of the Japan Society of Civil Engineers was first published in 1944 and has been published monthly since 1962. There were 289 papers printed in 1993 and since the first issue, the total number of papers reached to 5,665 in the fifty years up to 1993.

In this research, papers from the Proceedings of the Japan Society of Civil Engineers printed between 1944–1993 concerning dam construction are examined, totaling 497. Papers were chosen on the basis of whether the research paper applies to either of the following. First, the writer clearly talks about contributing to dam construction. Second, the theme of the paper is of dam construction. In this research, 16 themes are chosen as of dam construction, that is, topographic and geological survey, hydrological survey, water utilization and discharge planning static load to dam body, dynamic load to dam body, sedimentation (sand deposition in dam lake), design of dam structure, mechanical properties of dam body, hydraulic properties of dam body, thermal properties of dam body, mechanical properties of rock mass, hydraulic properties of rock mass, dam concrete, materials of fill-type dam, water quality of dam lake, and landscape design. The chosen 497 papers intend to use the results of research to improve the quality and safety of dam construction, and these papers reflect the research of dams in Japan in the past fifty years.

The changes in number of research papers concerning dam construction that were considered in this paper are shown in Fig. 1, together with the transition of the number of dam construction starts in Japan after the Second World War.

Research papers on dam construction steadily increased each year up to 1972, particularly in 1969–1972 there was a dramatic increase. There was some variation after 1972, approximately 15 papers were written annually. And after 1987, the number has increased again. On the other hand, the number of dam construction has its peak in 1972. The number, similar to research papers, increased until 1972, but in the years 1973–1976 the number decreased sharply, and after 1985 it is on the slow decrease.

The coefficient of correlation for the number of dam construction starts and dam related research papers calculated from Fig. 1 is low at 0.331. This is because, despite the decrease in the number of dam construction starts after the oil shock in 1973, the number of research papers decreased in only one year in 1973 and continued to increase though with fluctuations. Therefore, after 1973 the number of construction starts and research papers shows a negative correlation and this affects the coefficient of correlation. The coefficient of correlation from 1945 to 1972 is 0.786. In other words, in this period the number of research papers increased as the number of dam construction starts increased.

From the above observations, research on dam construction continued to be active not only during the stable period of the late 1950's and late 1960's, but also after 1973 when the number of dam construction starts retreated due to the adverse effect of the economy and social situation.

In order to examine the characteristics in the transition of research themes of dam construction quantitatively, an index measuring the variety of research was formulated by using the concept of entropy.

The concept of entropy appears in various fields of study in natural science and plays a central role in the field of thermodynamics and statistical dynamics, it is also used in social science as an indicator for measuring the volume of information that is contained in numerical statement. For example, by using the concept of entropy, Kodama sought an index for variations in research and development of industrial technology according to the type of industry [1].

In this paper, an index is formulated for measuring the variety of research of dam construction. The research themes are divided into 16 and papers classified into these 16 themes, then the composition ratio of research themes (Pij) is calculated (Fig. 2). Pij indicates the percentage of research themes of field j in year i and is given by:

Here, Eij is the number of research papers in field j in year i. Suppose the composition ratio of research themes each year has a probability distribution, the entropy of year i can be calculated by:

Here, Hi is the entropy of research of year i. In other words, this indicates that in the year where the entropy of research is large, research themes tend to vary and papers break up into various themes, while in the year where entropy is small, papers focus on particular themes. The entropy of research introduced from the composition ratio of Fig. 2 is shown in Fig. 3.

Until the mid 1950's entropy is scattered because not many papers were sent to the Proceedings of the Japan Society of Civil Engineers as shown in Fig. 1. Accordingly, looking at the entropy solely after 1956, there are several characteristics regarding the variety of research on dams.

The main features are a) an increase during the 1962–1963 period, b) an increase and decrease with a peak in 1974–1975, and c) increasing entropy throughout the whole period. Then, why does this "variation" or "concentration" occur?

Firstly, the major factor for the increase of entropy during the 1962–1963 period is the appearance of papers on the mechanical and hydraulic properties of rock mass. As shown in Fig. 4, the number of research papers concerning the mechanical properties of rock foundation increased after this period.

Second, as to the increase and decrease of entropy with a peak in 1974–1975, there was an increase in the variety of research up to 1975, and then the concentration peaks towards 1981. The increase of entropy in the first half of the 1970's is due to the initiation and promotion of research on the water quality, water utilization and discharge planning, and hydrological survey (Fig. 4). However, since 1975 the entropy of research decreased, hitting bottom in 1980–1981. In this period, the number of research papers on the water quality and hydrological survey supporting the increase of entropy in 1968–1975 further increased. Also, as shown in Fig. 4, the number of papers on hydraulic properties of dam body increased sharply during 1980–1981. Namely, owing to the increasing share of research papers in these fields, the tendency for the increase of variety temporarily retreated and was substituted by a tendency for concentration.

Finally, the increasing variety of research leading to the increase of entropy throughout the whole period has been brought about by the birth of new themes and the durability of themes such as studies on the water quality in the 1960's and studies on landscape design since 1988.

Thus, the dynamic nature of research activity such as the expansion of studies due to the birth of new themes or the concentration of studies on specific areas is well represented by the increase and decrease of entropy.

The analysis on the variety of research using the idea of entropy so far has mentioned some of the features of dam research in the postwar period. Here, the following features are examined and it is investigated what motivated researchers and why these themes have changed over time.

As shown in Fig. 4, the papers on the mechanical properties of rock mass have appeared since 1949. However, the focus of researches at the time was on the sliding of the contact surface of the bank with the rock foundation and few papers were published. But since the research by Takano [2], the emphasis was shifted to safety problems such as stress distribution and failure of the rock foundation itself, and the number of papers also increased. The peak came in 1966–1967 and in the years from 1962–1969, nine papers were published. In the introduction of each paper, the authors emphasize the importance of the study of rock mechanics. Three of them published in the early 1960's refer to the Malpasset Dam accident a. As Kimishima mentioned in his paper [3], " after the collapse of the Malpasset Dam, the main cause of which is said to be the shear collapse of the rock foundation, some other dams have suffered from disaster, and the safety problem of Japanese complex rock foundations especially of shear strength must urgently be studied by power plant engineers," uncertainty over the safety of dams following the accident of 1965 have influenced greatly the increase in the research on rock mechanics.

Papers on water quality began to published in 1968 and peaked in 1976–1983 (Fig. 4). The content of 18 papers published in the 16 years of 1968–1983 is as follows: 8 on turbid water, 5 on eutrophication, 4 on water temperature (of which 2 dealt with the relationship with turbid water), and 3 on water quality. In other words, the two big problems of the long-term turbidity of water and eutrophication which became serious since 1970 had drawn growing social concerns b, and activated the researches in this field.

Papers on hydrological survey began to be published in 1947 and increased rapidly in 1972 (Fig. 4). The objectives of hydrological survey are divided roughly into two groups; one is to predict the long-term trend of the volume of water of the river for efficient utilization of water resources, and the other is to predict the size of floods to help prevent disasters. Therefore, the research themes include hydrological observation relative to the volume of water of the river and the volume of rainfall, hydrological statistics that predict the probable volume of rainfall and floods by considering observation data using statistical methods, and run-off that predict the volume of water of the river. These researches began to be published in 1947, leading to the increase in the number of studies since 1972. Research papers published in 1972–1979 consisted of 2 studies of hydrological observations, 8 studies of hydrological statistics, 12 studies of run-off, and 3 others (such as reliability of water utilization plans). Of these 25 papers, 10 were related specifically to practical application of the study. They include seven papers on measures against the rising demand for water (including one that overlaps with measures against floods) and four papers on measures against disasters caused by floods. In other words, the problems of water shortage caused by the increase of demand for water for home and industrial usage and the increased frequency of flood influenced the increase in studies in this period.

On the other hand, researches on water utilization and discharge planning, which started to be published in 1966, increased after 1972, and peaked in 1974–1977, since when only one or so paper a year has been published (Fig. 4). Studies on plans for the water utilization and discharge planning include both plans for improving the efficiency of water distribution and plans for improving the efficiency of discharge control of dams. The 16 papers published in 1972–1979 include 8 on plans for water distribution and the other 8 on the water discharge operation of dams. In the case of multi-purpose dams, the function of dams "to always keep a sufficient volume of water to ensure the steady provision of water" contradicts "the need to control water by having enough idle capacity in case of flood." Among the research papers concerning water discharge operation, four written after 1973 discuss measures against floods, keeping a balance between water utilization and flood control, as the main theme of research. In other words, after 1973 the balance between the water utilization and flood control, which was previously ignored became an important theme following floods caused by dams that caused social problems ^c.

After a research on the hydraulic properties of dam body was first published in 1950, only one or two papers were published annually, but during the two years of 1980–1981, 8 were published (Fig. 4). Until the 1970's, study on the hydraulic properties of dam body mainly focused on water seepage analysis inside banks formed by borrow material. Taking a look at the eight research papers released during 1980–1981 when the number of research papers increased, there were four studies concerning water seepage and four on failure mechanisms by coupling water seepage and stress. Among the abstracts of these eight research papers, five were related to the importance of the failure mechanism induced by water seepage and three related to enhancing public awareness of the danger of accidents during initial impoundment of fill-type dams. One of the reasons for the concentration of papers concerning the hydraulic properties of dam body during this period was the Teton Dam accident in 1976 ^d.

As described above, accidents and environmental problems have influenced the variation and concentration of research on dams. Next, problems of public acceptance in Japan regarding dams are discussed.

Residents have acted against dam construction since the 1960's, but there are also some reports of disputes between residents and construction companies before this period. For instance, Hanayama [4] pointed out many difficulties involved in evacuation from Aganogawa dam due to insufficient compensation at an appropriate timing for those residents living in areas to be covered by water. Disagreements like these before the war were mainly related to compensation such as purchasing of land, the main cause of this was due to a lack of national policy. The compensation problem did not develop into a systematic campaign, but instead was mostly solved on an individual basis.

On the other hand, the necessity for facilities, safety and integrity of the natural environment became an important issue among residents after the 1960's. A major problem of public acceptance is the basic rights of human being. This refers to the right to live on the land where one's ancestors have lived generation after generation and the seizure of such land for construction. In addition, problems of safety and the environment caused by the construction of dams surfaced in the 1960's and influenced research activities.

Among safety and environment problems include dam accidents, sedimentation, flooding caused by dams and water quality. TABLE shows the period when these problems surfaced and their influence on research. The table shows that from an analysis of papers published by the Japan Society of Civil Engineers, safety and the environment, which are related to public acceptance, have greatly stimulated much research.

This research analyzed studies in the field of civil engineering, which have advanced greatly after the war, in order to suggest ways to gain public acceptance and to establish a proposal for this. Questions of, have problems of public acceptance influenced civil engineering research? and, has the result of research influenced on the solution of public acceptance? were investigated.

Dam related research papers published by the Proceedings of the Japan Society of Civil Engineers were analyzed, and the entropy calculated from the composition of the number of research papers. It was shown that the variation and concentration on research themes agreed closely with problems of public acceptance such as safety and the environment.

It is true that not only in the field of civil engineering, there is a wish to develop new fields as a nature of researcher, or to introduce new technology such as the finite element method (FEM) for analyzing rock mechanics that stimulates new research themes. In addition, this paper has shown that the problems of public acceptance exist as what is termed an "external factor" in sociology of science, and greatly stimulate research in the field of civil engineering. An answer to the first question was proposed.

Although the second question was not considered in this paper, it is a fact that dam related research has greatly helped solve problems of public acceptance. For instance, although water seepage analysis of banks of fill-type dams is of great importance when evaluating the safety of a dam, it has been shown that there is no better way than to follow long experience and appropriate decisions because of inadequate analysis methods. The advancement in research on the hydraulic properties of banks is linked to the design guideline on water seepage and failure by seepage that appeared in the Second Version of the Standard on the Design of Dams in 1978. At present, the analysis of water seepage has an important role in ensuring dam safety. In other words, the research spawned by the accident at Teton Dam, which strongly influenced public acceptance, has progressed the level of science and has led to a mathematical design technology for dam construction in Japan which used to rely mostly on experience. In order to clarify how these research efforts have contributed to the problems of public acceptance of construction, much remains to be studied.

Acknowledgments- I would like to express my sincere appreciation to Prof. Yoichiro Murakami of International Christian University and Prof. Tsuyoshi Hiromatsu of University of Tokyo for their useful advice.

a. Malpasset dam was constructed in southern France in 1954 as a concrete arch dam of 61 m high. The maximum thickness of concrete being 6.7 m, this dam was supposed to be the most economical dam in the world, with the thinnest body among those of similar height. However, it collapsed in 1959, because of the failure of dam foundation. Japanese paper quickly reported the accident, with a headline "Several Hundreds Deaths as Dam Collapsed". It stimulated public concerns over safety of arch dams. At that time, construction of arch dams was becoming standard in Japan, eleven construction of big arch dams taller than 70 m had just started, including the development of Kurobe dam (the biggest arch dam of 186 m high in Japan). In particular, there had already been concerns over the stability of foundation rock in Kurobe dam before the accident. Under such circumstances, Malpasset Dam accident was a huge shock to those involved in the arch dam construction.

b. Turbid water first become a serious problem in Sameura dam (completed in 1973). Right after the completion of the dam, typhoons caused turbid water in Yoshino River for a long period of time, and this become an object of serious public concern. Consequently, "Investigation Committee for Yoshino River System Turbid Water" published a conclusion that the development of Sameure dam caused the long-them turbid water. The problem of eutrophication started with the rise of public concern over the conditions of natural lakes such as Lake Biwa and Kasumigaura between 1975 and 1985. Meanwhile, livestock effluent and land drainage were flowing into reservoirs, causing eutrophication. Naka river in Shikoku experienced particularly serious red tides in Nagayasuguchi dam and Kominono dam in 1976. A numerous of sweetfish died as a result, which provoked much public concern over the drawbacks of dam development.

c. Prevention of flood damage is one of the major benefits that a dam can give to the local residents as a flood control facility. However, there are often cases where the development of dam itself became the cause for flood damages. Flood damages caused by dams are categorized into the following : those owing to functional deterioration in the upper stream of a dam because of sedimentation; and those owing to a mistake in discharge control during heavy rainfall. The flood damage that occurred in the Tenryu river system gave a rise to the problem of sedimentation. In 1961, large-scale flood damages (often referred to as "S36-disaster") occurred around Inadani valley located in the upper stream of a dam. Sedimentation in the upper stream of Yasuoka dam extended to 30 km, as is of ten said to be "running the river up wards". It was pointed out that the flood damage in this case was caused by the river bed rising as a result of such a sedimentation. Meanwhile, flood damages caused by mistakes in discharge control became an issue after the flood damage of Tadami river in 1969 and the "Heavy Rainfall of July" that hit Chugoku and Kyushu areas in 1972. "Flood Damages Investigation Council" was set up after the flood damage of Tadami river, and recognized that the dam was to blame. As for the Heavy Rainfall of July, residents of many affected places claimed that it was the structural defects of the dam and operational mistakes that were to blame for the flood damage. It eventually led them to bring a case before the court.

d. On 5^th June 1976, Teton dam in the U.S.A. (a zoned earth-fill dam of 123 m high with a central core) suddenly suffered hydraulic fracturing a moment before filling water during initial impoundment. At that time, the development of large-scale rock-fill dams was at its peak in Japan (more than twenty rock-fill dams were under construction including Takase Dam, the Japanese biggest rock-fill dam of 186 m high. The accident at Teton dam was published with heavy headlines and shocking photos, leaving a powerful negative image of dam development to the public.

1. H. Kodama, "Paradigm in High Technology Industry (in Japanese)" Tyuokohronsha, Tokyo, Japan, (1991).
2. M. Takano, "Experimental Examination of the Safety of Arch Dam Foundation (in Japanese)", Proceedings of the Japan Society of Civil Engineers, 78:1, 43-70, (1962).
3. H. Kimisima, "In-situ Experiment Concerning Failure Mechanism of Rock Mass (in Japanese)", Proceedings of the Japan Society of Civil Engineers, 120:8, 1-7, (1965).
4. K. Hanayama, "The Theory and Reality of Compensation (in Japanese)", Keiso-syobo, Tokyo, Japan, (1958).

BACK