36 THENEWYORKER,AUGUST17, 2020
maintained by the city government. In
China, such case histories are often avail-
able, as resources for local residents. Liu’s
case history notes that, during the first
three days after he is unknowingly in-
fected, he visits a bar, a store, two phar-
macies, three gas stations, and six restau-
rants. Liu’s tastes are eclectic, ranging
from a pancake restaurant to a frog-and-
fish-head restaurant. He picks up a friend
named Huang, and he visits his elderly
parents. He goes to work. He gets a fever.
Post-fever, Liu hops over to a few more
pharmacies, and then he keeps going:
he picks up a friend named Li; he vis-
its his parents again; he goes to another
party. On the WeChat account, Liu is
the Liupold Bloom of northeastern Si-
chuan, with every step of his urban od-
yssey recorded in terrifying detail. When
is this guy going to stop?
S
uch meticulous case histories were
prepared by contact tracers who
worked under the direction of the
Chinese Center for Disease Control.
There are about three thousand C.D.C.
branches in China, each branch con-
taining roughly a hundred to a hundred
and fifty staff members. Despite these
numbers, the Chinese C.D.C. has tra-
ditionally been underfunded, like Chi-
nese public health in general.
Approximately ten thousand contact
tracers worked in Wuhan, where more
than eighty per cent of China’s deaths
occurred. Epidemiologists told me that
the tracers were divided into teams of
between five and seven, with each group
directed by an individual who had for-
mal training in public health. Other
team members might have no health
background, but they came out of the
same detail-oriented national educa-
tional system that had produced my
students, and they often had local knowl-
edge. Many tracers worked for neigh-
borhood committees or other govern-
ment organizations, including the police.
As the virus spread, tracing teams were
established across the country, and the
C.D.C. recruited others who had tech-
nical expertise.
In Shanghai, a twenty-four-year-old
named Jiang Xilin was contracted to
work on various projects for the C.D.C.
and the Gates Foundation. Jiang is in
his third year of a doctoral program at
the University of Oxford, where he stud-
ies genomic medicine and statistics. He
had won a Rhodes scholarship to Ox-
ford after studying at Fudan University,
in Shanghai. In early March, Jiang wor-
ried about the initially complacent Brit-
ish response to the coronavirus, and he
asked his advisers for permission to re-
turn to China and study remotely. “They
all thought I was crazy to want to go
back at that time,” he told me.
In Shanghai, Jiang helped the C.D.C.
with modelling, computer program-
ming, and writing proposals. “The first
weekend, I got a call at 12 A.M. on a
Sunday evening,” he told me. “Nobody
said, ‘I’m sorry to disturb you so late.’
They said, ‘Did you get that proposal
done?’ I said, ‘No,’ and they said, ‘We
need that report by noon.’” He quickly
became accustomed to such demands.
Jiang also learned that, if a late-night
call went silent, it often meant that the
person on the other end had fallen asleep
from exhaustion.
By then, many overseas students and
others were coming home. It would
have been useful to know exactly where
they had been, so Jiang wrote a pro-
posal requesting that Tencent, the com-
pany that owns WeChat, provide the
I.P. log-in information for returnees.
“They rejected me because of the data
privacy,” he said. He was told that Ten-
cent was adamantly opposed to its da-
ta’s being used in this fashion.
Once, when Jiang and I met for din-
ner in Shanghai, he showed me how
our phones automatically sensed each
other via Bluetooth. Such data could be
used to figure out who had been in close
proximity to an infected person. In an-
other C.D.C. work meeting, a colleague
of Jiang’s suggested using this tool. But
her idea was quickly dismissed. “They
said, ‘This is a violation of data protec-
tion. We can’t do that,’” Jiang explained.
“It was surprising to me.”
It surprised me, too—given the
heavy-handed tactics of many lockdown
policies, I had assumed that the gov-
ernment used any tools available. But
there seemed to have been some resis-
tance from prominent tech companies.
Tencent and Alibaba helped the gov-
ernment develop “health code” apps that
assist in monitoring and controlling the
virus’s spread among citizens, but these
tools are much less sophisticated than
programs used in South Korea and Sin-
gapore. In Europe, virus-alert apps based
on software developed by Google and
Apple have been downloaded by mil-
lions of users, and the apps rely on Blue-
tooth signals to detect close contact
with infected individuals.
In some parts of China, the health-
code apps register a change in a user’s
location largely through a manual data
transfer: if the user checks in with his
I.D. at an airport, for example, or if his
license plate is recorded at a toll booth.
An epidemiologist in Shanghai told me
that one Chinese city with a flourish-
ing tech industry had commissioned
the development of a much better tool
that combines G.P.S. data and artificial
intelligence to alert anyone who comes
into the proximity of an infected per-
son. “But that system was never imple-
mented, even in that city,” the epidemi-
ologist, who asked not to be identified,
said. “It could not get approval from
somewhere in the government system
because of data privacy.” He noted that
while some of the apps track location
through cell-phone towers, they don’t
use the more accurate G.P.S. data.
“One can argue that what was most
useful for COVID was old science,” he
continued. “The methodology is from
fifty or seventy years ago. It has not
changed.” Jiang Xilin told me that,
when the proposals to use automated
data collection were rejected, the other
C.D.C. researchers grumbled. But then
they buckled down and continued to
do the hard legwork of phone calls and
face-to-face interviews. The C.D.C.
policy is that, whenever a new case ap-
pears, contact tracers are called imme-
diately, even in the middle of the night.
They are given eight hours to complete
the tracing.
In June, after Beijing had reported
no locally transmitted cases for fifty-six
days, there was a sudden outbreak at a
wholesale produce market called Xin-
fadi. The epidemiologist in Shanghai
told me that the place was well man-
aged: masks were required, and anybody
who entered had to show his health
code and have his temperature taken.
Even so, more than three hundred peo-
ple were infected, and all the warning
systems had failed to catch it in the
early stages. The first alert came when
a man in his fifties felt sick and went
to a hospital to request a test. It was an-
