Technological "Paradigms:" Cognitive Traditions and Communities in Technological Change
Following Kuhn's seminal work paradigms were claimed to be discovered in many scientific fields including sociology, economics, psychology, mathematics, even literature, arts and history. It is well known that Kuhn himself was astonished to see that, for him unexpected, escalation. Garry Gutting rightly emphasized 198O that most of the applications of the paradigm conception led to nowhere but to insignificant, relatively trivial analogies, to assertions that "supertheories" exist. (1) But some application may have overcome trivial analogies. The story of technological "paradigms" is one case for this.
The trial to apply the paradigm conception to technological change came 1O-15 years later then the applications to other fields. (2) In an important case study for history of technology (published 198O), E. W. Constant II set up a general model for technological change. (3) In this model technological change is represented by knowledge change and put into an evolutionary epistemology perspective, overtaken from D. Campbell. Constant exploits philosophy of science, mainly Kuhn's paradigm conception. He finds a community structure in technological practice, traditions of practice, normal technology with its puzzle solving character and technological changes initiated by recognizing two types of failure. He claims that, from time to time, technological changes are technological revolutions. "We define a technological paradigm as an accepted mode of technical operation. . . . It is the conventional system as defined and accepted by a relevant community of technological practitioners. A technological paradigm . . . like a scientific paradigm, is also rationale, practice, procedure, method, instrumentation, and a particular shared way of perceiving a set of technology. It is a cognition." (4)
Analyst of historiographic conceptions in history of technology, Staudenmaier appropriately emphasizes the importance of Constant's findings. "Three aspects of Constants definition are significant. First, a technical tradition is primarily a cognitive reality. It is a specific way of perceiving of technology that integrates the pertinent devices and techniques into a meaningful whole, a 'governing conception'. As such the governing conception is more important to the tradition than any single element in it. Second, the governing conception is radically dependent on a community of practitioners who, recognizing themselves to be such, create the governing conception precisely because they must define the norms to be satisfied by aspirants to community membership. Third, the governing conception can be defined as the body of learning handed down from one generation to the other. Two functions of a tradition's governing concept are obvious: it defines the meaning of all devices and techniques that are components within it, and it defines the canons of membership in the community . . . we can identify a third function by which the concept fosters the inventor's creative insight. The knowledge handed down in a tradition creates . . . a frame of reference with recognizable boundaries, the right and wrong way to do the technical thing. That intellectual framework is the necessary background against which new insight emerges." (5)
While Constant made a profound use of the paradigm conception, he did not understand the absolutely essential role of an "exemplar" in Kuhn's conception, as Gutting recognizes. This hinders Constant to find a possibility of developing a common basis by the paradigm conception for including arts and crafts cognition into technology analysis. (6) Similar to M. Masterman, Gutting recognizes an essential role for "exemplars" when they can lead paradigm development without explicit theory.
Constant got immediate acknowledgment. The reason for this may mostly be similar to that of Kuhn's appreciation: the conception of technological "paradigms" challenged the prevailing idea of incremental growth occurring through steady individual efforts by providing for an impressive alternative insight with cognitive traditions and communities.
Constant's model can be taken as a paradigm change on the meta-, the historiographic level, bringing historians into a "different world." By developing this different world it not only emphasizes similarities to the paradigm model of scientific change but works out some differences too. Technological paradigms not only change because of self-willed "functional failures" (the mainly functional equivalent to Kuhn's anomalies) but also because of "presumptive" ones, something without analogy in scientific paradigms conception. "Presumptive anomalies" make an essential element of the pattern like technological change. To put it shortly, during the reigning of a previous technological paradigm and its normal science development a new scientific theory may already be able to show the limits of it, that it would not work under different or more stringent conditions. So a transcending step of the still working technological paradigm can already be undertaken through recognition of a "presumptive anomaly."
Beside the obvious difference to scientific change by the introduction into the model of "presumptive anomalies" Constant perceives three differences. These are the hierarchical structure of the technological paradigms, stronger than in science, the 'satisficing modes' (H. Simon) and the role of economic criteria. According to Constant the basic difference behind all these is that technology is "less vicarious exploration of the world than science." I think Constant is right by stating these differences.
Nevertheless, to my mind, he sweeps over some other differences too quickly, differences that do not stop the possibility of heuristic application of his model to technological change but make it impossible to use it in an unambigous way. The main problem with his model is that it fails to differentiate among some possible references, like change of technology, change of technological knowledge, further: functioning of an artifact and knowledge of functioning as true/false statements on this.
1) The first problem where this somewhat chaotic use of references is to be seen is with "technological communities" and their different roles. Constant's technological communities are mixed ones which can include manufacturers, civil and military users, governmental and community agencies, among many others. Technological communities, if they are composed of the participants of Constant's model, are first of all practical communities to produce some technological artifacts. They are, first of all, production-consumption-regulation communities. As practical communities they produce real life artifacts (and as byproduct learning/cognition integrated into it). So Constant's technological community is cognitive community through its byproduct of activity only. But it is misleading to identify the whole practical community with those who are mainly responsible for appropriate knowledge for technological change. This is so, even when first, technological knowledge is much more produced by involving of those who are not included into professionalized research institutions (like consumers) than knowledge in science, and second, social value changes (like environmental or aesthetic sensibility growth or reinterpretation of social functioning) have a (sometimes strong) impact on requirements against technological functioning of technological artifacts and this evolves the requirement of appropriate knowledge about them. At least the modern technological system has a functionally specialized cognitive institution and technology is changed, first of all, through to research professionalized institutions and experts.
2) The other side of the coin is that technologies are to be used outside the community of their professionalized developers, (7) and really, all the listed participants in the technological community are in position of special observation possibilities, and in this quality are relevant cogntive actors, who produce some sort of empirical knowledge. (8)
3/ Another difference to scientific cognition and its paradigm changes is to be found through the different purposes of science: technological research is goal-oriented. As such, any technological research faces a type of anomaly different from the scientific one. This can be called "anomaly through time pressure or lack of tools." (9) (Showing only apparent anomalies technological projects may fail, because this time pressure.)
There is an another constraint from the goal directedness. The end product must be something realizable as artifact (in a broad sense). Facing analogical difficulties a scientist more easily can change its activity by reducing the purpose of it by accommodating to the available tools, by bracketing the (temporarily) unapproachable. Technologists must stronger try to keep on the task they got. This is what Constant correctly recognizes.
4) Another difference to scientific research is, at least of magnitude, that to solve, even to fully put their tasks technologists have more heavily rely on something what can be called as "bricolage" by bringing together all the available knowledge that can support their research.
5) This way the cognitive community in technology is generally much more fuzzy than the scientific one. The reason is quite obvious for the reasons enumerated earlier for the fuzziness of the cognitive tradition. Together with this the perceived anomaly as real crash of something may be much more direct than in science, nevertheless less evidently something in need of to be solved in a different paradigm.
6) Constant correctly states that there are such participants in the cognitive arena of technological development like firms or other "aggregate" participants, but he does not refer to their special cognitive role. (10) There are aggregate participants in science too, but their cognitive role seems to be more pronounced and different in technological research. Firms, their labs and departments as - vehicles and proprietors of specialized verbal and tacit knowledge - are actors beside, or many times above individuals in technological cognition, more than in and differently from science.
An important milestone in the story of technological "paradigms" conception was the conference organized by Rachel Laudan 1981. Laudan's (and perhaps the other participants') intention was to contribute to "understanding change and development within technology itself." (11) This "within technology itself" as demarcationistic declaration of the tasks needs some remark. Laudan refers to economics analyses of his time that leave out technology as exogenous factor of their work and identifies the left out as "the nature of technology," "technology itself," "the internal development," "the internal dynamics." Having made this demarcationistic analytical step as if it were obviously acceptable she also suggests to model technological change by change of technological knowledge through comparison to the Kuhnian paradigm model. (12) In one point she defers from Constant. According to her anomalies can come from every corner not only as functional and presumptive ones. This claim is rightful if we recognize that changes in social organization of technology, new social values like environmental ones, have a consequence on what is taken as appropriate functioning of a technology and appropriate technological knowledge.
This way, showing a series of differences, even when having community and cognitive tradition patterns, is technological cognition a type of cognition in its own right from this point of view too. It shows patterns. I think it is better to think of a technological tradition as system of paradigms but not a paradigm itself. (13) A system can be changed through development of some elements that themselves are not results of paradigmatic revolutions. One possibility to this is a surprising technological development (measured by any parameter) which results from "pure" empirical trial. Coevolutional constraints can then effect paradigm crisis on other elements.
Technological traditions only partially posses cognitive authority and autonomy. According to Gutting the partial cognitive authority of technological community is due to the existence of 'presumptive anomalies'. But beside this, as we saw, technological communities are mixed communities. This mixed nature has two consequences. One is that quite many can contribute to technological innovation, at least through 'innovating by using' or by identifying weak places in the construction. In measure of this involvement the yshare the cognitive authority with experts.
Economists came to the idea of "nature of technology" because technologies appeared for them as special inertia against being integrated into economic conceptualization. Pace and direction of technological change as caused by its internal nature was to be asked about. Govanni Dosi begun a similar effort as Constant around 198O. Actually Nelson and Winter already wrote earlier in the spirit of accepting some sort of paradigm conception for technological cognition a by referring to "technological regimes." A "technological regime" may be characteristic for an evolutionary period in which the improvement of a guiding model (e.g. the DC-3 as theirs) defines the activity of realising a "technological trajectory." I think there is a clear difference bw the role of a "basic scientific achievement" with Kuhn and the guiding model for research of Nelson and Winter. While the development of mechanics should be seen, following Kuhn, as the realisation of the same principles in different subfields, of which one helps to fix the system, the development of a guiding model like the DC-3 led to variations of the same original concrete artifact.
Evolutionary economics fund technological "paradigms" appropriate for its goals. But these paradigms got a different characteristics in one more relation. They were put into a changing economic (and organizational) environment (as invariable patterns in an evolutionary learning process in a changing environmnet) in which paradigms stepped into a steady competition for survival. While there are no steadily competing paradigms in normal scientific change, technological "paradigms" were supposed to behave differently. (14)
To sum up: the analogical extension of the paradigm conception proved to be heuristic in its application to technology. The conception of technological "paradigms" could be worked out by accommodating the original model to the characteristics of technology. Technological cognition became clearer through both of its positive and negative analogies. It became a "stylized fact" for economic theory. Another task is still left behind. Boundaries stand and effect on cognition but it is farther to show how far they are maintained by repeated by constructive efforts and how knowledge is socially constructed within the boundaries of technological "paradigms." It is a task more for historians than economic theory.
(1) Garry Gutting (Ed.): Paradigms and Revolutions, Appraisals and Applications of Thomas Kuhn's Philosophy of Science, Univ. of Notre Dame Pr., Notre Dame, 198O
(2) The special way leading to this idea may be mainly responsible for the delay. Studying technological change through change of technological knowledge had to overcome some barriers like the problem of strong tacitness of this type of knowledge, etc., that made implausible this work at its very beginning. R. Laudan (Ed): The Nature of Technological Knowledge. Are Models of Scientific Change Relevant? Reidel, Dordrecht, 1984.
(3) Edward W. Constant II: The Origins of the Turbojet Revolution, The Johns Hopkins Univ. Press, Baltimore, 198O
(4) Edward W. Constant II: A model for technological change...in: Technology and Culture v. 14., 1973, p. 554. He adds to it: "A technological paradigm is passed on as a tradition of practice in the preparation of aspirants to its community membership."
(5) John M. Staudenmaier: Technology's Storytellers, Reweaving the Human Fabric, MIT Pr., Cambridge/Mass., 1985, p. 65.
(6) In: R. Laudan, 1984, p. 64.
(7) Where the community itself changes in history.
(8) One has to put here the "local knowledge," the "improvement by using," etc. and remember that it is simply impossible to standardize the whole environment and reduce the complexity where a technology should function. Hence it is less possible than in science to foresee the normal application sphere of any technology.
(9) That means that a hard anomaly in one point of time may be proven later as a curable one within the old system but this knowledge is senseless for the real historical process.
(10) With Kuhn the scientific community, while holding indivisible characteristics, is composed of individuals, at least the hierarchical group structure is not worked out.
(11) In: R. Laudan, 1984, p. 4-7.
(12) Laudan 3-6 old. "Opening the black box" can be done in two very different ways, first through fixed analytical categories and stating perpetuous boundaries for the black box or on a social constructivistic way that always asks the construction and reconstruction of boundaries and together with it the so called technical content inside the black box, as effected by 'inner' and 'outer' social conditions.
(13) Taken into account that a system is characterized by the abstract system relation while a paradigm is a full "world." Staudenmaier seems to think similarly. Compare Staudenmaier 1985, Ch. 2.
(14) Arie Rip and Henk van den Belt reacted on the first application of the paradigm conception to technology (by Nelson and Winter, Dosi 1981). They refine the evolutionary model by pointing to the notional competition and the importance of social "niches" that protect new paradigms "against the myopia of natural selection." In: Wiebe E. Bijker, Thomas P. Hughes and Trevor J. Pynch (Eds.): The Social construction of Technological systems. The MIT Press, Cambridge/Mass, 1987, p. 135-