fects both the number and the grant size of researcher proposals funded. In 2004, for example, only 24% of all proposals to NSF were funded, the lowest proportion in 15 years.⁵

Ultimately, increases in research funding must be justified by the results that can be expected rather than by the establishment of overall budget targets. But there is a great deal of evidence today that agencies do not support high-potential research because funding will not allow it. Furthermore, because of lack of funds, NSF in 2004 declined to support $2.1 billion in proposals that its independent external reviewers rated as very good or excellent.⁶

The DOD research picture is particularly troubling in this regard. As the US Senate Committee on Armed Services has noted, “investment in basic research has remained stagnant and is too focused on near-term demands.”⁷ A 2005 National Research Council panel’s assessment is similar: “In real terms the resources provided for Department of Defense basic research have declined substantially over the past decade.”⁸ Reductions in funding for basic research at DOD—in the “6.1 programs”—have a particularly large influence outside the department. For example, DOD funds 40% of the engineering research performed at universities, including more than half of all research in electrical and mechanical engineering, and 17% of basic research in mathematics and computer science.⁹

The importance of DOD basic research is illustrated by its products—in defense areas these include night vision; stealth technology; near-realtime delivery of battlefield information; navigation, communication, and weather satellites; and precision munitions. But the investments pay off for civilian applications too. The Internet, communications and weather satellites, global positioning technology, the standards that became JPEG, and even the search technologies used by Google all had origins in DOD basic research. John Deutch and William Perry point out that “the [Department of Defense] technology base program has also had a major effect on American industry. Indeed, it is the primary reason that the United States leads the world today in information technology.”¹⁰