A distinct definition that draws a sharp boundary between science and technology is difficult, if not impossible, to produce. Science and technology frequently overlap. Nevertheless, it is possible to produce working definitions of science and technology that make a broad distinction. Accordingly, science may be defined as 'attempts to produce "basic" knowledge about natural phenomena which does not necessarily have any immediate commercial applicability'. Technology can be defined as knowledge related to transformation of inputs into commercial outputs, including production of new or different outputs.

Figure 2.1 Configuration of industries and institutions involved in biotechnology

Technological knowledge may be embodied in people, hardware (plant and equipment) and software, and forms of organization.

According to these definitions, biotechnology is related to science in the sense that the knowledge which underlies its three main technologies-recombinant DNA, cell fusion, and bioprocessing-clearly emerged from the science system. In a detailed account, for example, Cherfas (1982) traces the origins of biotechnology from the first recognition of DNA by Miescher in 1869, to Watson and Crick's model of the double helix in 1953, to the breakthrough of Boyer and Cohen on the recombinant DNA technique in 1973, and the work by Millstein and Kohler on cell fusion in 1975. A good deal of this work was influenced by research on the behaviour of bacteria and viruses and by the war on cancer. Despite the ultimately pragmatic objectives of such research, the research remained for the most part 'basic' in nature. Rosenberg (1982) points out that in many cases 'basic' knowledge has resulted from research undertaken with 'applied' motivations. This makes it difficult to sustain a distinction between basic and applied research in terms of the motivation for such research.

In this respect, biotechnologies contrast sharply with semiconductors, which developed largely, though not entirely, in response to the war and early post-war military demands of the U.S. Department of Defense (Borrus and Millstein, 1984). In contrast to biotechnology, whose major breakthroughs have occurred in universities, the transistor was invented in 1947 at Bell Labs, a part of AT&T which purchased its telecommunications equipment from Western Electric, its manufacturing arm. In 1959 the integrated circuit was invented at Texas Instruments and Fairchild, two small commercial companies which had spun off from Bell Labs. The milieu within which semiconductor technology was developed was therefore oriented more towards practical objectives than in the case of biotechnology, where 'basic' university scientific research, albeit health-related, played a more significant role. However, Borrus and Millstein (1984) can probably fairly be accused of being overly simplistic when they conclude that 'In the development of biotechnology, "science push", rather than the "market pull" that gave impetus to the US semiconductor industry, was particularly important' (p. 533).

Nevertheless, this dichotomy does raise the important question of what role the science base plays in science-oriented industries, such as microelectronics/information technology and biotechnology. Clearly, it is inadequate to see science as a subsystem, autonomous from the rest of the economy and society, or scientists as uninfluenced seekers of the truth attempting to understand the basic nature of the universe. The emergence of microelectronics/information technology and biotechnology has had a good deal to do with the twin social concerns-one may almost say neuroses-of military defence and health. Furthermore, scientific controversy and the progression of scientific ideas have often been greatly influenced by the interests of scientists themselves as some sociologists of science have documented (see Barnes and Edge, 1982, and references therein).

Neither can basic science that forms the core of biotechnology be assumed to be uninfluenced by commercial considerations or at least by the possibility of technological applications. Cohen and Boyer, for example, were aware of the commercial applicability of their recombinant DNA technique and this awareness led Stanford University to apply for a patent for the recombinant DNA (rDNA) process technique within the statutory one year after initial publication of results. In December 1980, Patent No. 4.237.224 was granted, providing for an initial nonexclusive licence fee of $10,000 and an equivalent annual amount for using the technique in research and development. In addition, the patent granted a royalty of I % on sales up to $5 million, falling to 0.5% on sales over $10 million. Since this technique is fundamental to work in genetic engineering, the implications for Stanford University funding are enormous. Stanford University subsequently filed a second patent on the products produced by the rDNA technique. [The Cohen and Boyer patent is discussed, for example, in Yoxen (1983, pp. 95-97), and U.S. Congress, Office of Technology Assessment (1984, Chapter 16).]

Millstein, who together with Kohler developed the cell fusion technique in 1975, was also aware to some extent of the commercial implications of his research. Accordingly, he wrote to the Medical Research Council informing them of the possible implications, hoping the National Research and Development Corporation (NRDC), which was responsible for commercialization and protection of intellectual property rights of inventions coming out of public laboratories, might make the necessary arrangements for patents. However, the NRDC did not act and the key patents to work on monoclonal antibodies were eventually taken out by American researchers. In 1980, partly in response to this failure, Celltech was formed by the British Technology Group, which took over the role of the NRDC, and the National Enterprise Board. The company was partly publicly and partly privately funded and was given exclusive access to the research output of the Medical Research Council's laboratories (see Yoxen, 1983, pp. 128-132).

These examples and their implications make it clear that, while there is a functional, institutional, and organizational difference between scientific establishments on the one hand and commercially oriented establishments on the other, neither are entirely self-contained but rather exert a mutual influence on one another. This suggests that a more general approach, which will consider these and other interactions, is needed to understand and evaluate the function of these organizations. This has important implications for the study of factors like international competitiveness.