The emergence of new scientific disciplines and new forms of cooperation between their representatives for further progress in knowledge are two mutually dependent tendencies in the development of the sciences, both in their systems of order and in the research and teaching profiles based on them. These two tendencies can be used, above all, to grasp the change in the relationship between the object area of research and the object area of social practice, which conditions scientific disciplines as a form of historically established and changeable boundaries of knowledge and knowledge production. Ultimately, the view of scientific disciplines expressed by Max Planck as early as the 1930s applies here: “Their separation according to different subjects is not, after all, rooted in the nature of things, but arises only from the limitations of human capacity, which inevitably leads to a division of labor” (Planck, 1944, p. 243, translated).

Scientific disciplines are historically conditioned and thus changeable forms of knowledge acquisition and knowledge reproduction in which both the manner of scientific questioning and the preference for certain methodological approaches are acquired and practiced by individual scientists, and in which scientists and scholars experience or can achieve social recognition and are institutionally established.

Based on the assumption that scientists have to refer to certain areas of theoretical knowledge both in formulating problems and in methodically working on problems, a distinction can be made between disciplinary and interdisciplinary research situations.

Disciplinary and Interdisciplinary Research Situations

Science develops through theoretical thinking and observational activity, be it mere or experimentally determined observational activity, in which researchers methodically solve epistemological problems by means of knowledge and research techniques. Every problem refers to knowledge of situations in mental or observational or practical–experimental activity, in which the available knowledge is not sufficient to achieve the goals and therefore must be expanded accordingly.

In a narrower sense, the awareness of such a knowledge deficit is only called a problem if the missing knowledge cannot be taken over by others, but has to be gained anew. A research problem exists, if for a system of statements and questions about or according to conditions of goal achievement no algorithm is known, by which the determined lack of knowledge can be eliminated in a finite number of steps. If an algorithm is known, then a task exists. The conceptual differentiation between problem and task has also been fruitful in work for the methodology of modeling.

With the scientific problem the questions are justified by the existing knowledge, but not answered. A problem dissolves to the extent that new knowledge as substantiated information answers the questions that represent a scientific problem. An important difference exists between the occurrence of a problem situation, which is seized and represented by the researcher in the problem, and the existence of a research situation. Thus, the creative scientist must have a feeling for the really crucial questions, but at the same time (s)he must also have the correct feeling for it: to what extent it will at all be possible, with the given conditions of the research technology, to master the problems with the instruments available or which can be developed. According to this, a research situation can be understood as such connections between problem areas and methodological structures that allow the scientist to methodically work on the problem areas by means of actual availability of knowledge and research technology.

Following the understanding of the methodological structure of research situations, in addition to the two entities of problem field and methodological structure and the relations between them, the actual availability of conceptual and physical means to deal with the problem on the one hand and the epistemological and social relevance of research problems on the other hand must also be taken into account (see Fig. 7.1). For if research situations with a novel relationship between problem and method as well as device (software and hardware) are to be brought about, only those research possibilities can be realized for which society provides the corresponding means and resources. Decisions on this, however, depend on the problem relevance shown. The problem relevance, i.e., the evaluation of the problems according to the contribution of their possible solution both for the progress of knowledge and for the solution of societal practical problems, ultimately regulates the actual availability of knowledge and equipment for problem processing.

Fig. 7.1
figure 1

Methodological structure of the research situation

Full size image

At the end of the 1970s, Wolfgang Stegmüller, in discussion with Thomas Kuhn, attempted to define the concept of normal science more precisely with the help of the concept of having a theory. The term we use, availability of knowledge and equipment for problem solving, is much more comprehensive than the concept of having a theory at one’s disposal, since it also includes practical feasibility in research. In a later version, for Stegmüller, “everything pushes towards a systematic pragmatics, in which non-logical terms are used, such as knowledge situation of persons and its change; subjective faith of persons at certain times; background knowledge available at a certain historical time and the like” (Stegmüller, 1983, p. 236, translated). In a further attempt in this direction, Stegmüller considers “additional pragmatic concepts that we have to build into the conceptual apparatus, because ‘person’, ‘historical time’, ‘available knowledge’, ‘standards for the acceptability of hypotheses’ are concepts of this kind” (Stegmüller, 1986, p. 109, translated).

If, in order to characterize research situations, the relationship between a problem field and a set of preconditions for problem solving is considered, then different research situations can be distinguished—at least according to the degree of the cognitive and social relevance of the respective problem and the degree of the actual availability of preconditions for solving the respective problem—but, above all, also according to their scientific and social integrity (see Fig. 7.2).

Fig. 7.2
figure 2

Structure of scientific and social integrity of the research situation. (Bottom: scientific integrity, top: social integrity)

Full size image

The relationship between the scientifically necessary disciplining of methodological problem-solving in research and the socially conditioned formulation of interdisciplinary problem fields for research leads to increased reflection on the distinction between disciplinary and interdisciplinary research situations: A research situation is disciplinary if both the problems formulated in it and the methods used in it refer to one and the same field of theoretical knowledge, and a research situation is interdisciplinary if the problem and method of research are formulated or substantiated in different theories.

Indicators of Interdisciplinary Work in Research Groups

As early as three decades ago, a comprehensive empirical study by UNESCO on the effectiveness of research groups asked, among other things: “In carrying out your research projects, do you borrow some methods, theories or other specific elements developed in other fields, not normally used in your research” (Andrews, 1979, p. 445). The first interpretations attempted to establish the comparability of the 1,200 groups studied by means of classification by discipline and interdisciplinary orientation in research. At the same time, it was assumed that the specific scope of cooperative relationships and thus co-authorship could be understood as a surrogate measure of the productivity of research groups working in interdisciplinary fields (cf. Steck, 1979).

The indicators of interdisciplinarity used by the current author in investigating 56 life science research groups in the years 1979–1981 were based on the assumption that the decisive factor for interdisciplinarity in research groups is whether at least one group member thinks in an interdisciplinary way, regardless of whether the group members are assigned to only one or several disciplines (cf. Parthey, 1982, 1983, 1990, 1997).

A first indicator of interdisciplinarity concerns the percentage of scientists in the research group who formulate their problems in terms of (different) scientific disciplines in an interdisciplinary manner. If all scientists in the group were to formulate problems in only one discipline, the percentage of scientists formulating problems across disciplines would be zero. Thus, groups that work on problem fields of the type mentioned are rightly classified as predominantly working in a disciplinary manner if, due to the derivation of sub-problems from a problem field, they are composed of representatives of different disciplines but work on these sub-problems via the means of their own discipline.

A second indicator of interdisciplinarity refers to the percentage of scientists in the group who need and use methods to work on their problem that are not based in the same field of knowledge as the problem itself. In this sense, our research enquired whether: “The methods used in the research group to work on your problem are: (A) grounded in the same area of knowledge in which your problem is formulated, [or] (B) are grounded in an area of knowledge that is different from the knowledge in which your problem is formulated.” The percentage of scientists who answered with (B) in relation to the group size was recorded as the degree to which the interdisciplinarity of problem and method is expressed in research groups.

On the basis of these studies, the following forms of scientific activity can be distinguished (see Fig. 7.3):

  1. 1.

    Firstly, monodisciplinary research (i.e., in scientific activity, no cross-disciplinary problem has been formulated and no interdisciplinarity of problem and method has been developed).

  2. 2.

    Second, multidisciplinary research (i.e., in scientific activity, problems are formulated across disciplines, but no interdisciplinarity of problem and method has been developed).

  3. 3.

    Thirdly, interdisciplinary treatment of disciplinary problems (i.e., no interdisciplinary problem has been formulated in scientific activity, but interdisciplinarity of problem and method does occur).

  4. 4.

    And finally, fourthly, interdisciplinary treatment of cross-disciplinary problem fields.

Using the frequency of this combination of problem formulation in a cross-disciplinary context on the one hand, with the interdisciplinarity of problem and method on the other hand, we found the frequencies shown in Table 7.1. As we see, interdisciplinarity of problem and method seems to be a necessary condition for full co-authorship in cross-disciplinary research groups.

Fig. 7.3
figure 3

Forms of scientific activity

Full size image
Table 7.1 Interdisciplinary problem formulation and interdisciplinarity of problem and method in 56 research groups from four non-university institutes of the life sciences in the early 1980s
Full size table

Table 7.2 shows the frequency of the characteristic coupling between the indicator “Percentage of scientists working interdisciplinary with problem and method in research groups” and the indicator “Composition of research groups according to diploma disciplines.” The first data column of Table 7.2 indicates that we find personal interdisciplinarity, even if the research group is monodisciplinary (i.e., the group members represent only one discipline). Interdisciplinarity and multidisciplinary composition of research groups do not coincide. In other words, personal interdisciplinarity does not require multidisciplinary composition in the research group. However, it can be assumed that the interdisciplinary work of individual scientists (understood as personal interdisciplinarity) is promoted by the composition of the research unit from representatives of different disciplines.

Table 7.2 Percentage of scientists working interdisciplinarily and group composition (education in different disciplines) in 41 research groups from three non-university institutes of the life sciences at the beginning of the 1980s (each”+” represents one group)
Full size table

In addition to the indicators for interdisciplinarity mentioned above, further indicators were taken into account in analyses of interdisciplinary work, such as the indicator “co-authorship in the group,” which is used as a surrogate measure of the productivity of interdisciplinary research groups and is based on bibliometric profiles of groups, as well as indicators for the “multidisciplinary composition of the group” (according to education) shown in Table 7.2 and the often used indicator for the “distribution of expertise by discipline.” Our search for correlations between these indicators is based on an analysis of the rankings of the respective group values using correlation coefficients (see Table 7.3).

Table 7.3 Interdisciplinarity and co-authorship
Full size table

The many positive rank correlations—as shown in Table 7.3—can be interpreted to mean that a multidisciplinary training and competence structure of the group offers favorable conditions for interdisciplinary work by individual scientists. However, the finding in Table 7.3 underlines that only practiced interdisciplinarity of problem and method (4) significantly correlates with co-authorship (6). We see concurrent rankings, i.e., the more/less that individual scientists within the group practice interdisciplinarity of problem and method, then the more co-authorship within the group increases/decreases. This corresponds to the finding we see in Table 7.1: encompassing co-authorship can only be expected in the case of interdisciplinarity of both problem and method.

Studies on personal interdisciplinarity in science touch on the analysis of collaborative work in research groups, especially regarding the influence of other group members on the performance of an interdisciplinarily working scientist. In good tradition, social science questions the influence of others on one’s own performance or the advantages/disadvantages of working in groups compared to individual work (already highlighted by Triplett in 1898). This question, applied to the scientific work itself, leads to analysis of the relationship between individual and cooperative performance in research groups and follows Max Planck’s view of science, as mentioned previously, that their separation into different subjects “is not based on the nature of the matter, but only on the limitations of human capacity, which inevitably leads to a division of labor” (Planck, 1944, p. 243, translated).

In this way, the analysis of research groups deals with a topic of long-standing interest to scientists from different disciplines—and especially science researchers—and that remains highly topical today. Research of this kind has existed worldwide since the 1930s. Such research tends to be based on different methods such as participating observation or historical reconstruction. The more or less standardized questioning of research groups only began in the 1960s. In particular, the assumptions and methods used in the 1960s and 1970s assume that the effectiveness of research groups is decisively influenced by the correspondence between the structure of the problem and the division of labor within the research group (e.g., Pelz & Andrews, 1966).

These studies examined the working relationships that researchers must enter into with each other when working on certain problem areas. By problem structure we mean, above all, the relationships between the primary, secondary, and sub-topics of a problem field. On the basis of numerous analyses from the 1960s and 1970s, the concept of a research group has developed that can be characterized by the following features (e.g., Swantes, 1970):

  • common concern, in the form of a problem field to be worked on together,

  • division of labor and cooperation in methodological problem solving, and

  • their coordination by leadership.

However, empirical tests reveal that the above-mentioned assumption (that the effectiveness of research groups is decisively influenced by the correspondence between the problem structure and the structure of the division of labor in the group) is only supported to a limited extent. Thus, our analyses—which are also recorded in larger overview studies on German science research (Woodward, 1985)—point to two fundamental considerations: On the one hand, the existence of a problem situation and correspondingly formulated research problems are certainly necessary for the development of cooperative relationships between researchers, but they are not sufficient. The necessary and sufficient condition for forms of cooperation between scientists to occur is the existence of a research situation with regard to a problem, i.e., above all, the creation and actual availability of conceptual and physical means to deal with the problem. On the other hand, different types of research situations exert different influences on the form of cooperation. Different degrees of availability of idealistic and material means for the treatment of research problems require different relations between researchers based on the division of labor.

Our empirical finding is that it is not the composition of a group of representatives from different scientific disciplines that is significantly correlated with co-authorship, but only the group share of scientists who practice interdisciplinarity of problem and method. According to our analyses, the decisive characteristic of interdisciplinary research situations is therefore not—as was often assumed in the first approach of sociological studies into interdisciplinarity—the multidisciplinary composition of the group according to education and competence in different disciplines, but rather the disciplinary lack of knowledge about problem solving among individual scientists and the resulting search for method transfer from other special fields.

Institutionalization of Research Situations

Research institutes were and are designed as self-organizing systems. It is also a goal to develop research–technical systems that have self-organizing properties in science. Self-organizing systems are constantly confronted with alternatives in which it is up to them to make a selection. In this sense, researchers are always in situations where they have to decide for or against performing certain actions. Description and explanation of scientific institutions can be based on the fact that there is a fundamental need for a social space for the creation and development of research situations, without which science cannot exist, as its history shows.

The researcher needs the institution, because this is the only way to ensure the necessary freedom for research. This freedom is created through appropriate funds, such as the personnel and material budgets, and through the institute’s own system of information, communication, and library. Scientific libraries as a component of scientific institutions become scientific workplaces to the extent that they make their publications available for further research with minimal redundancy as the sciences become increasingly differentiated. And the researchers themselves decide on the necessary and sufficient minimization of redundancy. The exchange of letters between scholars has shown—and continues to show—this in an exemplary manner at its time. Today, scientific journals have taken over the function as “libraries of scientific disciplines” (Parthey, 2003). Here, researchers are responsible in the function of editors, on behalf of scientific collectives (i.e., the networks of journals), of which at least two scientists (i.e., peers) judge the submitted research findings of others (i.e., peer review) according to whether (and after which revisions) they should be included in the respective special library of a “scientific journal.” Publications in scientific journals contain, at least in one structural part, something scientifically new, which is presented along with exact citations that form a comprehensible reference to the “old” in science. Since its emergence in the second half of the seventeenth century, the scientific journal has proved its worth as an organ in the communication and information system of original research papers.

Scientific disciplines differ according to the area of investigation of reality and the theory on which it is based; how further knowledge of the structure and laws of the world is sought; which of the problems and which methodological approaches are preferred for their scientific treatment. Disciplinarity in science can be increasingly differentiated. The reasons for this are the increasingly higher degree of specialization of this knowledge and the discipline-specific terminology created for its articulation, as well as the highly specialized research techniques required to further deepen this specialized knowledge. In this sense, it can be observed that new scientific disciplines have emerged at universities to the extent that, first, a chair has been created for each new scientific discipline; second, a textbook has been written for it; and, third, after the advent of letterpress printing, a new journal has become available for original papers by researchers in this new scientific discipline. Umstätter (2003) points to a comparatively “constant relation of journals and special fields.” Wilhelm Ostwald has described this process of organizing new journals (in the process of the development of a new scientific discipline, which he himself helped to promote) as follows:

That I then, after the textbook was finished, founded the Zeitschrift für Physikalische Chemie (Journal of Physical Chemistry) was just as natural a process… The fact that both forms of organizational work, the textbook and the journal, had a considerable influence on the further development of matters, is essentially due to the fact that at that time (in the eighties of the last century) a number of excellent collaborators in the field appeared at far-flung points in the cultural world, i.e., without mutual agreement or influence, who very soon made the scientific content of the field unusually rich and fruitful. They found the ground prepared by the above-mentioned works, and conversely, the new journal was able to prove its raison d’être to other circles by publishing groundbreaking works soon. (Ostwald, 1919, p. 10, translated)

Although the emergence of institutions is generally explained in terms of people’s demand for individual orientation and social order, considerations in institutional theory also point out that institutions are only accepted and supported in people’s demand for individual orientation and social order to the extent that they do not conflict with their interests. In this sense, forms of scientific institutions in their historical formation are of particular interest.

Securing Science and Its Freedom Through Institutions: Historical Forms of Institutionalization

As outlined above, the scientifically active person needs the institution, because only through this can the free space necessary for research be secured. This free space is created by appropriate funds, such as personnel and material budgets, and with an institute’s own system of information, communication, and library. In order to be attractive, the scientific institution must secure the researcher an appropriate status in society and itself be flexible enough to cope with the dynamics of the modern scientific enterprise.

Plato’s Academy Near Athens, Aristotle’s High School in Athens and State Research Center in Alexandria

Obviously, the history of scientific institutions begins with the fact that Plato gathered his students around him in a grove of the Academy near Athens since about 388 B.C.E. Thus, the Platonic Academy was also the first scientific institution. Aristotle was active in this academy for 19 years until Plato’s death. Afterwards, he was appointed by the Macedonian King Philip II as tutor to his son Alexander. Soon after Philip’s death, Aristotle returned to Athens and founded his own school, the Lyceum, as a second scientific institution for teaching young people.

As the third scientific institution, a state study center of the entire Hellenistic world was established in Alexandria in the third century B.C.E., consisting of the Mouseion research center (cf. Parthey, G., 1838) and the largest library of the ancient world. Euclid was among those who worked there, between 320 and 260 B.C.E. and Ptolemy from 127 to 141 B.C.E., who carried out the observations used in his work “Almagest” in the observatory. Alexandria was a center of scientific life for more than 700 years of history until about the beginning of the fifth century C.E. In the following centuries without any scientific institutions worth mentioning, hardly any scientific publications were published, sometimes not at all, i.e., for several centuries almost no scientists can be proven.Footnote 1

University Education of Science-Based Professions Since the Middle Ages

Even if the institutions that emerged in antiquity to ensure problematization and methodical problem-solving—such as the Platonic Academy, the Aristotelian Lyceum as a municipal grammar school, and the Alexandrian Mouseion as a state research institution—did not survive the centuries despite their research achievements, a new, sustainable scientific institution has endured since the twelfth century with the university, due to the increasing interest in the training of science-based professions (initially mainly for doctors and lawyers). From then on, the university has also been involved in the training of other emerging science-based professions and has thus become a fundamental institution of science all over the world. In addition to this, modern academies have also been established with worldwide success since the fifteenth century (following the Platonic Academy) as research institutions without the teaching obligations of universities.

Today, university education can enable people to carry out a scientific activity if, in addition to imparting a disciplinary field of knowledge that is subject to constant renewal, it aims above all at the ability to independently ask further questions, to develop these into knowledge problems with the available level of knowledge and to methodically gain problem-solving insights. This can only be achieved by teaching that presents and discusses the process of scientific knowledge in a model way and actively involves the students in this process. Research-based learning is thus an integral part of every scientific course of study (cf. Mieg et al., 2021).

Non-university Research Institutes Since the Emergence of Science-Based Economy

In the nineteenth century, the institutional form of science was still largely that of the academy and—increasingly—the university in the unity of teaching and research striven for by Wilhelm von Humboldt, whereby his great scientific plan called for independent research institutes as integrating parts of the overall scientific organism in addition to the Academy of Sciences and the university (cf. Humboldt, 1964). With the emergence of science-based industries such as the electrical industry, which could not have existed beforehand—not even as a trade—without the scientific theories of flowing electricity and electromagnetism and the discovery of the dynamoelectric principle (1866 by Werner von Siemens), and the transformation of traditional trades into science-based industries such as the chemical industry in the last third of the nineteenth century (cf. Zott, 1998), the university was able to establish itself as a center of scientific excellence. In the last third of the nineteenth century, the establishment of scientific institutions outside of universities grew to include large chemical research laboratories set up by the chemical industry, and state laboratories for basic research in physics that were intended to contribute to improving the scientific basis of precision measurement and materials testing. An example of the latter is the Physikalisch-Technische Reichsanstalt, founded in 1887 in Berlin-Charlottenburg (Imperial Physical-Technical Institute, cf. Förster, 1887; Cahan, 1989), which Wilhelm Ostwald still described as a “completely new type of scientific institution” two decades later (Ostwald, 1909, p. 294).

The Physikalisch-Technische Reichsanstalt consisted of two departments, the scientific and the technical. The first one is currently still trying to work on problems of physical precision measurement that are pending but urgently in need of a solution, especially those problems for which universities lack the necessary rooms and equipment or those that require a scientist to devote themselves entirely to research for a long period of time without the additional demands of teaching. The second department is intended to provide direct support to the precision trade, taking care of all the technical services that cannot be carried out by mechanics in small- and medium-sized enterprises, but also serving as an official testing institute for mechanical and technical instruments. The president of the institute is also the director of the scientific department.Footnote 2 The success of the Physikalisch-Technische Reichsanstalt triggered efforts to establish an analogous Chemisch-Technische Reichsanstalt (Imperial Chemical-Technical Institute). Driven by the developmental needs of science itself as well as of the state and the economy, which is also evident in studies of science policy in Germany since the eighteenth century (cf. McClelland, 1980), several research institutes independent of teaching were founded in Berlin within the framework of the Kaiser-Wilhelm-Gesellschaft zur Förderung der Wissenschaften (KWG, Kaiser Wilhelm Society for the Promotion of Science), which existed for more than three decades (1911–1945) and was financed by both the state and the economy. Today, the former Kaiser-Wilhelm-Gesellschaft (KWG) has been succeeded by the Max-Planck-Gesellschaft (Max Planck Society, MPG, www.mpg.de).

Interdisciplinary Research Situation in Non-University Research Institutes: Lessons from the Example of the Kaiser-Wilhelm-Gesellschaft (KWG)

As early as the last third of the nineteenth century, research directions developed “that no longer fit into the university framework at all, partly because they require such large mechanical and instrumental facilities that no university institute can afford them, and partly because they deal with problems that are far too advanced for students and can only be presented by young scholars.” (MPG, 1961, p. 82, translated) It also addresses novel relationships between research in government institutes and in the business world. For example, Adolf von Harnack, in his memorandum of November 1909, used as an example the situation in organic chemistry, “the leadership of which until not so long ago lay undisputedly in the chemical laboratories of German universities,” but which “today has almost completely migrated from there to the large laboratories of factories,” and concluded that “this whole field of research is to a large extent lost to pure science,” because “factories always continue research only to the extent that it promises practical results, and they keep these results as secrets or put them under patent. Therefore, the laboratories of the individual factories, which work with the greatest of means, can rarely be expected to promote science. The reverse has always been true: pure science has brought the greatest support to industry by opening up truly new areas.” (MPG, 1961, pp. 82–83, translated)

Thus, with the emergence of research-dependent industries, such as the chemical and electrical industries in the last third of the nineteenth century, there was an increase in the founding of scientific institutions outside the universities, for example, large chemical research laboratories, which the chemical industry built up, in addition to state laboratories for physical research that were to contribute to the improvement of the scientific basis of precision measurement and materials testing. There are three main reasons given for the establishment of research institutes that are independent of teaching (financed not only by the state but also by industry):

  1. 1.

    First, the rising costs of research technology (cf. Biedermann, 2002).

  2. 2.

    Second, the growing teaching obligations for university teachers, which make it difficult for them to work in the unity of teaching and research that Wilhelm von Humboldt aimed for.

  3. 3.

    And thirdly, the opportunities to create and work on many more interdisciplinary research situations, unhindered by the inevitably disciplinary teaching profiles at universities.

Therefore, in the founding history of the Kaiser-Wilhelm-Gesellschaft (KWG), reference was made to the fruitfulness of collaboration between researchers from different directions. From a later viewpoint of Adolf Butenandt, the founding of the KWG took place in 1911 in order to:

…close a gap in the German scientific structure. It was felt that working methods became necessary that were difficult to master in the conventional forms: It seemed urgently necessary to allow scholars who wanted to devote themselves primarily to pure research to work in complete freedom, to shield them to a large extent from all those things that might ultimately impair their ability to perform in the service of human progress. Secondly, it was necessary to give scholars working in newly developing border regions their very special working instrument, tailored to their needs, in order to strengthen and grow disciplines which had no or not yet had sufficient space in the structure of the universities and technical colleges. From the early days of the Kaiser Wilhelm Society, I mention as examples the physical chemistry of Haber, the radiochemistry of Hahn, the theoretical physics of Einstein, the biochemistry of Warburg. Thirdly, since the founding of the Kaiser-Wilhelm-Gesellschaft, the task had existed of developing and supervising new types of institutes. In order to solve some of the problems, very extensive personnel and material resources must be combined to form a structure that would have to go beyond the scope and technical complexity of any university structure. The institutes for iron research, coal research, and occupational physiology can be mentioned as examples. (MPG, 1961, pp. 7–8, translated)

August von Wassermann demanded at the inauguration of the Kaiser Wilhelm Institute for Experimental Therapy (as an institute of the KWG) in October 1913, that:

New ways of healing and all that is connected with it, especially the recognition of disease, should no longer be left here in this house to the more or less subjective experiences of the individual observer at the sickbed, as in earlier times, but should be explored on the basis of purposeful research with the help of the exact scientific auxiliary disciplines. (MPG, 1961, p. 158, translated)

Thus, in the founding history of the KWG, the fertility of a traffic of researchers from different directions was pointed out. Especially in the justifications for life science research without additional responsibility for teaching, the idea was developed that they should work outside the university in a more interdisciplinary way, which has also been scientifically profitable (cf. Jaekel, 1907). To this end, institutes for biochemistry and biophysics were founded in the KWG, among others. In most cases, a successfully proven “horizontal” interdisciplinarity led to the development of new disciplines with all the characteristics of an independent discipline, including later university teaching and training institutes.

Interdisciplinarity as a developmental form of science, which is institutionalized in a disciplined manner in the further scientific procedure, including within the university framework, is more of a “horizontal” interdisciplinarity, less of a “vertical” one,Footnote 3 as it was pursued outside of the university and instead within the framework of the KWG, especially in the institutes devoted to brain research, iron, metals, coal, leather, hydrobiology, silicates, fluid dynamics, and plant breeding programs. A comparison of the development of science in the USA and in Germany over the past decades shows that vertical interdisciplinarity is institutionalized and evaluated in non-university research institutions in the USA more quickly than elsewhere and, if scientific success continues, is also more rapidly introduced to university education programs.

A further lesson relates to the issue of research funding. In recent decades, the design of research situations has led to considerations that, in terms of their institutionalization, large-scale research should be set up in the form of umbrella organizations and so-called virtual research institutes. In Germany, for example, a new Joint Science Conference (Gemeinsame Wissenschaftskonferenz, GWK, www.gwk-bonn.de) was set up in 2008, which deals with the funding of the major research organizations such as the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG), the Max Planck Society (Max-Planck-Gesellschaft, MPG), the Helmholtz Association of German Research Centres (Helmholtz-Gemeinschaft Deutscher Forschungszentren) and the Leibniz Association (Leibniz Gemeinschaft), as well as academies such as the Leopoldina Academy of Sciences and the Wissenschaftskolleg in Berlin. This German science conference funds research projects, research buildings, and large-scale research equipment at universities that are of supraregional importance.

For a long time now, around two-thirds of all investment in research and development in Germany has come from the industrial sector. In order to tie in with this innovation momentum, the Helmholtz Association (www.helmholtz.de), Germany’s largest non-university scientific organization, is about to integrate the entire chain of effects from basic and applied research to future product maturity (vertical interdisciplinarity). In doing so, it too is relying on a strategic partnership with the universities. The fifteen Helmholtz centers are involved in areas of special university research and in priority programs funded by the German Research Foundation.

The discussion on science and financial policy in Germany since the beginning of the twentieth century shows that with the enabling financing of science by the innovative power of the economy, there is also a change in research in a science-integrated economy (cf. Spur, 2002), which may not solidify every new field of knowledge into a teachable discipline. In this context, we should to refer to the methodological structure of research situations as an invariant of knowledge production even in the twenty-first century but now with more focus on interdisciplinary research situations and their institutionalization than in the preceding centuries.

Ambivalence of the Experimental Method in Research

Science as published methodical problem solving has today for this purpose three large methodological structures: the experimental, the mathematical, and the historical method. At the birth of science, mainly the bare observation method, the mathematical method, and the historical method were used, because there was such a strict distinction between the epistemological and the technological that experimental methods for revealing truths were rejected in preference for only the bare observation without experiment.

Experimentation was excluded at the birth of science, due to the argument of ensuring scientific integrity in the methodological procedure of research. This remained true for science for one and a half millennia. Only with Galileo Galilei was there adoption of experimentally based observation—in all those cases where the truth value of statements cannot be determined directly by bare observation. For Galileo, this experimental basis for observation allowed more robust investigation and confirmation of hypothesized connections and claims of fact. Identifying an experimental problem, carrying out experiments, and finally interpreting experimental results for the verification of hypotheses were also introduced into research with Galileo, as three steps in the experimental method. In research, experiments are characterized by a system of conditions deliberately set by the experimenter so that essential relationships can be observed in a repeatable way under conditions of change and control.

Unlike mere observation, experimentation is based on an active intervention in natural and social contexts in the form of experimental technology, the ambivalence of which has now been discussed more intensively again since the twentieth century in various studies, following Aristotle’s rejection of experimentally conditioned observation in research.

Ambivalence, following its psychological usage, is used to describe an often-conflictual state in which opposing courses of action, such as affection-rejection, exist simultaneously with respect to the same object (see Bleuler, 1914/2017). Experimental research is increasingly ambivalent about its impact on society and science. A historically early example of the ambivalence of experimental research in the twentieth century is Lise Meitner’s rejection in July 1938 of Fritz Straßmann’s first laboratory notes on nuclear fission in uranium irradiated with neutrons (by chemical detection of barium in the irradiation products).Footnote 4 When Straßmann and Otto Hahn turned to this experiment again in December 1938 and had to conclude on uranium nuclear fission, they first communicated this to Meitner, who had emigrated in the meantime, with their publication submitted for printing (Hahn & Straßmann, 1939). Within a few days, Meitner had calculated the energy balance of this nuclear fission process in an article published jointly with Otto Frisch in January 1939 (Meitner & Frisch, 1939). Concerning the ambivalence of science, Karl Friedrich von Weizsäcker concluded:

Science cannot afford, under the motto that it seeks the truth and nothing else, not to consider the effects it has on life. Personally, I have never found it comprehensible that scientists felt that when what science produces in technology is used by politicians or by the military in such a way that scientists are unhappy with it, to say that here science has been misused. After all, science has provided these means, and it is, of course, responsible for the means it puts into other hands. If it supplies means into a political structure that is not adequate to these means—means that have a baleful effect in this structure—the least that is to be demanded of science is that it reflect on how the structure can be changed, which obviously cannot avoid producing these baleful effects. In this sense, then, self-reflection of science is a demand on science. (Weizsäcker, 1970, translated).

In resonance with the discussion about the scientific ambivalence of nuclear power, the German government pushed through a resolution (as early as 2000) to gradually shut down nuclear power plants. Finally, the disaster in Fukushima, Japan, led to the final legal nuclear phase-out in Germany, adopted in 2010. A new law on the nuclear phase-out in Germany is emerging for disposal of the radioactive legacies of nuclear power plants. Under the new law, the state will assume financial and organizational responsibility for nuclear waste disposal. Nevertheless, the revised draft law states that the polluter pays principle will be strictly adhered to in disposing of nuclear waste. Accordingly, energy utility companies will remain responsible for decommissioning and demolishing the nuclear power plants that they operated. For the disposal of nuclear waste, they are to pay a total of around 23.4 billion Euros into a public fund from existing reserves plus an additional risk surcharge.

In our century, embryo research, in particular, is increasingly ambivalent in its impact on society and science.Footnote 5 The German Embryo Protection Act prohibits the production or use of embryos for any purpose other than to induce pregnancy. Experimentation on human embryos is still a criminal offense in Germany.Footnote 6 Until now, active intervention in the human genome has also been ethically taboo internationally.

It is therefore not surprising that some researchers now fear that they have opened a Pandora’s box with the CRISPR gene-editing process. A technique that is suitable for transforming yeast cells, mice, or monkeys is also suitable for creating custom-made humans. Ethicists and lawyers have spoken out internationally, and of course the luminaries of the CRISPR guild: Jennifer Doudna, the method’s discoverer, and her colleague Emmanuelle Charpentier, who has moved to the Max-Planck-Institut für Infektionsbiologie (Max Planck Institute for Infection Biology) in Berlin. In order to equip mice with enhanced cancer protection, cell biologists introduced a mutation into their genome that activates a tumor-suppression gene (Morris et al., 2012). Only afterwards were they surprised to discover that the genetically manipulated animals showed side effects of premature ageing.

Such unexpected effects constitute perhaps the most cogent argument opposing human modification of our genome. The approximately twenty thousand human genes are interwoven into an immeasurably complex network of reciprocal influences. Any intervention will have consequences, and by no means all of them can be predicted. All over the world, biotechnologists and ethicists are discussing whether there can be circumstances in which it is acceptable to impose such potential consequences on future generations.