12 THENEWYORKER, MARCH 23, 2020
in “The Sum of the People: How the
Census Has Shaped Nations, from the
Ancient World to the Modern Age.”
Censuses abound in the Bible, in-
cluding one ordered by the Roman em-
peror Caesar Augustus and overseen by
Quirinius, the Roman governor of Syria.
“And it came to pass in those days, that
there went out a decree from Caesar Au-
gustus, that all the world should be taxed,”
according to the Gospel of Luke. “This
census first took place while Quirinius
was governing Syria.” Everyone was sup-
posed to register in the place of his or
her birth. That, supposedly, was why Jo-
seph made the journey from Nazareth
to Bethlehem, “to be taxed with Mary
his espoused wife, being great with child.”
(Quirinius’ census of Judea actually took
place years later, but it’s a good story.)
The first modern census—one that
counted everyone, not just men of fight-
ing age or taxpayers, and noted all their
names and ages—dates to 1703, and was
taken in Iceland, where astonishingly ac-
curate census-takers counted 50,366 peo-
ple. (They missed only one farm.) The
modern census is a function of the mod-
ern state, and also of the scientific revo-
lution. Modern demography began with
the study of births and deaths recorded
in parish registers and bills of mortality.
The Englishman John Graunt, extrapo-
lating from these records in the mid-sev-
enteenth century, worked out the popu-
lation of London, thereby founding the
field that his contemporary William Petty
called “political arithmetic.” Another way
to do this is to take a census. In 1753, Par-
liament considered a bill for “taking and
registering an annual Account of the total
number of people” in order to “ascertain
the collective strength of the nation.” This
measure was almost single-handedly de-
feated by the parliamentarian William
Thornton of York, who asked, “Can it be
pretended, that by the knowledge of our
number, or our wealth, either can be in-
creased?” He argued that a census would
reveal to England’s enemies the very in-
formation England sought to conceal:
the size and distribution of its popula-
tion. Also, it violated liberty. “If any officer,
by whatever authority, should demand of
me an account of the number and cir-
cumstance of my family, I would refuse
it,” he announced.
Two years later, in Pennsylvania,
Benjamin Franklin published “Obser-
vations Concerning the Increase of
Mankind.” Franklin had every reason
to want to count the people in Brit-
ain’s North American colonies. He cal-
culated that they numbered about a
million, roughly the population of Scot-
land, which had forty-five members in
the House of Commons and sixteen
peers in the House of Lords. How many
had the Americans? None.
To make this matter of representation
mathematical, enumeration of the peo-
ple, every ten years, is mandated by the
U.S. Constitution. There would be no
more than one member of Congress for
every thirty thousand people. The Con-
stitution also mandates that any direct
tax levied on the states must be propor-
tional to population. The federal govern-
ment hardly ever levies taxes directly,
though. Instead, it’s more likely to pro-
vide money and services to the states, and
these, too, are almost always allocated in
proportion to population. So the accu-
racy of the census has huge implications.
Wilbur Ross’s proposed citizenship ques-
tion, which was expected to reduce the
response rate in congressional districts
with large numbers of immigrants, would
have reduced the size of the congressio-
nal delegations from those districts, and
choked off services to them.
Under the terms of the Constitution,
everyone in the United States was to be
counted, except “Indians not taxed” (a
phrase that both excluded Native peo-
ples from U.S. citizenship and served
as a de-facto acknowledgment of the
sovereignty of Native nations). Every
person would be counted, and there
were three kinds: “free persons”; persons
“bound to service for a term of years”;
and “all other persons,” the last a sorry
euphemism for enslaved people, who
were to be counted as three-fifths of a
free person. It was a compromise be-
tween Northern delegates (who didn’t
want to count them at all, to thwart the
South from gaining additional seats in
Congress) and Southern delegates (who
wanted to count them, for the sake of
those seats)—a compromise, that is, be-
tween zero and one.I
t took six hundred and fifty census-
takers eighteen months to enumer-
ate the population in 1790. And then
Americans went census-crazy. There
were six questions on the first U.S. cen-sus. Then came questions that divided
people into native-born and foreign-
born. By 1840, when the questionnaires
were printed, rather than written by
hand, there were more than seventy
questions. Other questions, like one
about the ages of the enslaved popu-
lation (lobbied for by abolitionists),
were struck down. In the decades since,
questions have been added and dropped.
Most of them have involved sorting
people into categories, especially by
race. In the eighteen-forties, South-
erners in Congress defeated proposals
to record the names of people held in
bondage and their place of birth. Had
these proposals passed, the descen-
dants of those Americans would be
able to trace their ancestors far more
easily, and the scholarship on the his-
tory of the African diaspora would be
infinitely richer.
The 1850 census, the first conducted
by a new entity known as the Census
Board, was also the first to record in-
dividual-level rather than family data
(except for enslaved people), the first
to record an immigrant’s country of
birth, and the first to ask about “color,”
in column 6, a question that required
particular instructions, as Paul Schor
explains in “Counting Americans: How
the U.S. Census Classified the Nation”
(Oxford). “Under heading 6, entitled
‘Color,’ in all cases where the person is
white, leave the space blank; in all cases
where the person is black, insert the let-
ter B; if mulatto, insert M. It is very de-
sirable that these particulars be care-
fully regarded.”
The federal government had all kinds
of reasons for carefully regarding these
particulars. In 1860, the Census Board
added a new “color,” for indigenous peo-
ples who had become American citizens:
the federal government wanted more in-
formation about a population that it
sought to control. Although “Indians not
taxed” were still not to be counted, “the
families of Indians who have renounced
tribal rule, and who under State or Ter-
ritorial laws exercise the rights of citi-
zens, are to be enumerated. In all such
cases write ‘Ind.’ Opposite their names,
in column 6, under heading ‘Color.’”
Americans designated as “Ind.” could
“exercise the rights of citizens” but were
not, in 1860, deemed to be “white.”
The government’s interest in counting