Discussion and Conclusion
In this chapter, we have studied two cases of breakthrough inventions using a complexity-theoretic process model of technology invention in conjunction with analysis of the role of government, institutions, and co-location in flows and creation of technological knowledge. The model explains how invention is an iterative and interactive process of (1) gradual conceptualization and materialization of a system as a configuration of components embodying particular functions,
- (2) overcoming of technical challenges for the various components in piecemeal,
- (3) learning of efforts of others elsewhere, translating and combining their insights and technical solutions, and (4) being steered by governmental or institutional interventions.
From our analysis of two cases, we conclude that the model is general enough and does not omit crucial historical factors. However, a disadvantage is that the level of the detail in the descriptions of the government interventions, knowledge flows, and effects of institutional factors on inventive activities goes hand in hand with the depth of the technological decomposition of the system. In the second case, that of the jet engine, the technological decomposition is limited and so is, thus, the specificity of indications of knowledge flows, involvement of government, etc.
To illustrate the value of the process model as a descriptive tool, we discuss the first case in more detail and sum the main conclusions. In our analysis of the heavier-than-air aircraft, we observe that the process of invention is characterized by a decentralized search among different design paradigms, where inventors are engaged in experiments with (configurations of) component technology. In general, for the various designs, visionary and captivating images inspired new generations of inventors that access technical knowledge ‘shelved’ in books and articles, carried over and combined in public and private communications, and not uncommonly supervised by mentors that are proponents of a particular design paradigm. Critical may be the becoming available of research tools for systematic experiments, both to discriminate among design alternatives (if required), but also for optimization of component parameters and configurations. Specific institutions for the advancement of the technology have played an important role in absorbing and diffusing knowledge, funding research tools, and establishing credibility to the field. The involvement of national governments may seem to have been limited, notably because they invested in design paradigms and projects that ultimately failed, the pruning of research directions added to the dynamic efficiency nonetheless. Plus, even if particular designs fail, inventors engaged with other designs reap the benefits of the fundamental insights as to why it failed and the efforts devoted to improve components common to multiple designs. In our analysis of the jet engine technology, we again observe decentralized search across space and time, but with the rearmament, Germany promoted co-location and coordination of research and accelerated research.
Apart from this general characterization of the invention process, we are able to draw conclusions on the role of government, institutions, and the moderating effect of geographical distance. The first conclusion is that (both the heavier-than-air aircraft and to lesser extent the jet engine) invention processes feature substantial dynamic inefficiencies. Firstly, there are inefficiencies because inventors are unaware of other solutions, e.g. due to overlooked ‘shelved’ technological knowledge. For instance, the Wrights came up with wing warping in 1899, which was nonetheless technically inferior to the ailerons invented already in 1868 by Boulton. We have also seen how Chanute’s efforts to gather and communicate technological knowledge contributed to the Wrights’ success. Secondly, there were inefficiencies due to the co-existence of different design paradigms, e.g. fixed wing aircraft vs. ornithopters, or the “most ridiculous” slavish mimicking of animal wings deep into the 1890s versus the straight-line, structurally sound biplanes with clear precursors already in the 1840s. Although it constitutes a form of decentralized search preventing a collective lock-in at a (potentially poor) local optimum, the persistence of certain inventors turned out to be stubborn. Thirdly, part of the inefficiencies and failures in arriving at a working configuration stems from the fact that inventors lack a system perspective; (1) they tended to develop a complete aircraft without properly understanding the ramifications of design choices both on component performance and the interaction of components, and (2) improving specific components (e.g. wing design for lift, propeller optimization, increasing the power-to- weight ratio of steam engines) yet then relying on immediate real-world testing in configurations with ill-performing complementary components. In retrospect, the sensible order was followed by the Wrights: focusing on wing design for lift, adding and mastering mechanical control, and then add power. However, obviously, the Wrights could rely on so many inventions readily done by others before them, and start from a fairly standard configuration: the Pratt truss biplane of Chanute; camber, aspect ratio, and lift statistics computed before them; a basic propeller design; the mini internal combustion engine; and—given that so much had to be redesigned and re-measured by them—the wind tunnel. Working backward from the Wrights’ inventions would not only be a ‘presentistic chronicle’ misrepresenting technology evolution, it would also reveal only part of the inefficiencies, knowledge flows, and invention process. Only by focusing on various designs (even those ultimately deselected), the variants of a particular components in these designs, and studying the factors that have affected the course of history (other than technical selection) gives a more holistic perspective of the process of invention.
The second conclusion is that the role of government during the invention of heavier-than-air aircraft has been limited. There have been a few attempts of repairing the market for fundamental research in aerodynamics (the wind tunnel of Wenham, rig-tests by Maxim) and material science (e.g. for the turbines in the hot section), plus some support for experimental research (Ader, Langley, also Maxim). Ironically, the pilot projects that did get substantial public funding (Ader’s Eole, Langley’s Aerodrome, Maxim’s rig-test) did not form the foundation for the ultimate breakthrough of the Wrights, nor did they contribute to the subsequent system development in France. However, in retrospect, what actually did accelerate developments was the progressive institutionalization of communication of research findings and that inventors started to follow scientifically rigorous approaches. Clearly, public support of both is well possible, even in liberal market economies like the U.S.A. and U.K. Interestingly, both the French and U.S.A. governments were involved in funding development of prototypes of warfare flying machines for which the requirements were demanding but technologically non-prescriptive. In this, the public funding in fact coincided with substantial private funding rather than that it compensated for a lack thereof. Plausible causes are that (1) both the government and the private parties have the same favorable risk assessment, (2) one perceives the involvement of the other as risk decreasing, and (3) one acts on the presumptions that the other has alternative information.
The third conclusion is that, already in the mid-nineteenth century, there were knowledge flows across national borders, even while governments were aware of military application of aircraft technology. Given the early state of technology, little of the communicated knowledge was codified, nor did it concern technically concrete information. However, the writings did convey results on experiments, concepts behind design choices, and as such formed a (limited and partially wrong) basis of know-why and know-how knowledge on design features. Arguably, this is typical for the ‘technological uncertainty’ in early industries. Interestingly, the uncertainty mainly concerned the various components, as the design and breakdown into components itself was already fairly well established.
The present work has several shortcomings. Firstly, the level of technical detail in the second case (the jet engine) is limited such that it is hard to assess the invention process, let alone the role of government in the inventive activities. Future work should adorn the (multi)national innovation system description with the invention process details. Secondly, the process model of invention should be embedded more deeply in the innovation system literature as this will provide more factors and functions to take into account in historical analysis. In line with that, the level of detail in the cases should be adjusted to that, so as to shed light on how also these factors and functions have mattered.