In lieu of an abstract, here is a brief excerpt of the content:

The study of immigration has its distinct vocabulary—incorporation, assimilation, mobilization, coalitions, conflict, identity, and so forth. The terms in play touch on the broad question of whether ethnic and racial boundaries are being hardened or blurred, and to what extent the recent immigrant flows contribute to some mixture of these outcomes . The small contribution I offer is the reminder that the boundaries themselves, or at least their accessibility to research, rest on the way in which official statistics label population groups—starting even with the labels foreign born and native born. Subdividing the population is as old as census taking itself. Numbers , the fourth book of the Hebrew Bible, has Yahweh instructing Moses: “Take ye the sum of all the congregation of the children of Israel , after their families, with the number of their names, every male . . . from twenty years old and upward.” Here the key categories are male/female and under/over twenty. The intent, of course, was to know how many of the Israelites are “able to go forth to war.” A census is never just a count; it is always also a series of classifications selected to serve policy decisions. America’s earliest national census in 1790 rested, first, on a geographical classification—needed, of course, to allocate seats in the new Congress proportionate to population size. This census also divided the population by civil status: taxed and untaxed, free and slave. These civil status categories generated a racial classification that separated those of European descent from those of African descent and from Native Americans. The earliest censuses did not even bother to distinguish between native born and foreign born. The nation’s first racial classification carried a lot of policy and political weight. Including slaves in the census counts, even at three- fifths, rewarded the South with approximately a dozen more congresImmigrants and the Changing Categories of Race Kenneth Prewitt 20 Kenneth Prewitt sional seats and votes in the Electoral College than a count limited to its white population would have provided. This population bonus was among the several compromises struck with the slave-owning states to secure ratification of the new federal constitution. It had immense consequences. Known by historians as the “slave power,” the bonus in the Electoral College put Thomas Jefferson in the White House, and then his Virginian compatriots, James Madison and James Monroe. As Gary Wills documents in detail, a steady stream of pro-slavery (and anti-Indian) acts by Congress can be traced to the “extra” congressional seats awarded to the Southern states by the three-fifths clause (Wills 2003). It is instructive to compare the ease with which a racial classification was introduced into our statistical system with the resistance resulting from an occupational classification. In preparing for the nation ’s first decennial census, James Madison proposed a question to classify America’s working population into agriculture, commerce, or manufacturing (Cohen 1982).1 The new Congress rebuffed his initiative , registering both a technical and a philosophical objection. Technically, said the congressional opponents, the categories were imprecise , because, after all, the same person could fall into all three sectors —being a farmer who manufactured nails on the side and traded those he did not need to a neighboring farmer who made ax handles. More philosophically, Madison’s critics held that an occupational classi fication would admit to, and perhaps even excite, differing economic interests. This very possibility challenged eighteenth-century thought that took society to be a harmonious whole, and viewed the task of governing as that of divining a common good rather than that of managing conflicting interests. The harmonious whole that was blind to occupational differences was not, of course, color-blind. In the colorcoded language that becomes prominent in the nineteenth century, the earliest census separates the black, red, and white population groups. I take from the 1790 census a larger lesson. To divide the population into its several race groups was unquestioned. The categories could change, but not the need for the classification itself (see, e.g., Nobles 2000). In 1820 “free colored persons” was added to the census form (as, by the way, was Madison’s occupation question). After the Civil War, interest in shades of color led the census to classify people as mulatto, quadroon, and octoroon, motivated by a race science that viewed race mixing as detrimental to the moral fiber of the nation itself. New immigrant groups began to appear...

Share