In his inaugural address, President Obama referred repeatedly to education – but exclusively to education in STEM disciplines, as if only those fields had a defensible public purpose. Sadly, this is no aberration: in December the White House issued a report entitled "Transformation and Opportunity: The Future of the U.S. Research Enterprise,"  which completely overlooked research in the humanities and social sciences, even in its brief history of the growth of research at American universities.
Such a narrow focus is surprising, as the president himself apparently consults historians  (and probably other scholars); and it is counterproductive, whether in strict dollars and cents terms or broader ones. Some politicians have gone further, aggressively asserting  that various humanities and social science disciplines are useless, and attempting to impose  higher tuitions on students who major in them, making it all the more important that those who know better actively affirm the value of teaching and research beyond the STEM fields.
I will focus here on the case for history: it is what I know best, and since history straddles the line between humanities and social sciences, many arguments for its importance apply to various allied fields. One might loosely group these into three categories, ranging from the most social scientific to the most humanistic. The first applies to lessons drawn from circumstances relatively close to our own; the second to learning about times and places we know are quite different. The third applies to research showing that some currently accepted ideas are actually fairly novel, and that people not so different from us saw did without them; engaging the concepts they used instead may help us see additional possibilities in the world, whether for good or ill.
Examples of the first category underlie almost any sound public policy debate, as well as many private deliberations. Take, for example, the 2009 stimulus bill. By itself, no mathematical calculation could assess the relative accuracy of the more-or-less Keynesian models suggesting that the stimulus would help the economy and the "real business cycle" models, which predicted that it would be an expensive waste. The difference lay in historical research about how various modern economies had responded to historically specific policy initiatives. Other examples abound, though most are less well-known: closest to home in this regard would be evaluating options for STEM investment in light of the vast literature on what has given rise to specific clusters of innovation in the past, and which innovations proved most beneficial. One would also expect development efforts to gain from examining research on past relationships among, say, education, urbanization, birthrates, and investment.
The benefits of research into the importance of understanding differences in the context of policy decisions abound, with special clarity emerging in what we might call "area studies" knowledge – an enormous part of the growth of U.S. research universities after WWII. Surely we could have saved lives and money had policy-makers known more about religious differences within Iraqi society, the political and social history of Afghanistan, or class relations and popular nationalism in Vietnam before military interventions in those places. The same, I would argue, goes for using research into the evolution of Chinese notions of ethnicity, nationality, race, and geopolitics to understand likely governmental and popular reactions to possible American policies on Tibet, trade, the Diaoyu/Senkaku Islands, and so on.
Perhaps less obvious, but equally important, is the usefulness of research that shows that many ideas we may take to be "natural," or at least of very long standing, are actually relatively new.. Some of these insights may be "just" a contribution to increased self-understanding, but others bear directly on public issues. Urgent debates over how fixed the concept of "marriage" has been come first to mind, but there are many more actual and potential examples. Recognizing that the term "ethnic group" is barely 75 years old reminds us how mutable are our understandings of the basis and implications of human groupings; that "gross national product" is of roughly the same vintage suggests maximizing that particular measurement is not inevitably the paramount goal of economic policy.
It hardly seems a stretch to think that a world facing our current challenges might benefit from awareness of other ways that people have thought about the relationship of work, citizenship, adult status, "independence" and dignity, or about consumption, economic growth, leisure and the nature of progress. Or to take some narrower examples, consider the implications of learning how relatively recently life insurance went from seeming like a morally dubious gambling on death to a taken-for-granted tool for managing risk. Or that, while (as Thomas Ricks noted  in a recent Atlantic) almost no U.S. generals were removed from their commands for poor performance during Vietnam, Afghanistan or Iraq, many were so removed during World War II – suggesting that the recent situation does not represent an inevitable feature of government, much less of hierarchy generally. Historical knowledge of this kind does not provide lessons as straightforward as “deficit spending can work,” but it can add significantly to our understandings of what is possible, for better or worse, and how things may become, or cease to be, unthinkable.
Research that produces these results, both testing earlier certainties and responding to new questions , thus seems a useful, even necessary complement to research in the STEM fields. Fortunately, most historical research is also relatively cheap, but it does not thrive on complete neglect.
Kenneth Pomeranz is University Professor of History at the University of Chicago and president of the American Historical Association. The views expressed here are his alone.