IEEE Technology and Society Magazine - September 2020 - 27

for scientific research would shrivel.
Economists categorize the situation
as a "market failure." Not all market
failures involve public goods, but all
public goods are examples of market failures, because the market -
private actors pursuing profits for
themselves - won't make the necessary investments.
For science, the solution to "market failure" is well known and uncontroversial: the government fills the
gap, investing taxpayer's monies
into research. In the U.S., two agencies alone, the National Institutes
of Health (NIH) and the National
Science Foundation (NSF), have a
combined annual budget of more
than $40 billion. Each agency, while
enjoying strong bi-partisan support,
essentially operates as an association or club: practitioners decide
which of their peers receive funding,
for how much, and to what end.
The mechanism, called "peer
review," allows scientists to judge
the value of each other's work and
choose to fund projects they value.
The public goes along with this system of accountability because scientists produce beneficial results over
the long run and because neither
politicians nor typical citizens have
any the know-how or skill to decide
what good science is or how to go
about producing it.
"Peer review" is a robust model
for funding serious journalism. The
method is not perfect, and is subject to criticism and improvement.
Yet despite decades of complaints
about bias and inefficiency in peer
review, the model has survived.4
Let's look a bit more closely at how
the model works - to gain a clearer sense of how direct government
funding of independent journalism,
powered by the engine of "peer

review," can work to support quality
Uganda and each time interviewed
journalism at scale.
dozens of computer scientists,
The mechanism has three piland observed them at work. So far
lars. First, the NIH and the NSF
I've published two articles in peeremploy a cadre of experienced scireviewed academic journals drawentists and researchers, many on
ing on the research. I've never heard
loan from universities but some
a word from the government about
part of the government's
civil service structure. They
are usually called program
managers, and they help to
create "calls," or requests for
Not all market failures involve
funding. So if the NIH wants
to learn more about memory
public goods, but all public goods
loss in the elderly, or the NSF
are examples of market failures,
wants to learn more about
m a t e r i a l s fo r c o mpu tbecause the market won't make the
e r switches that can work
faster and use less energy
necessary investments.
than silicon, a program
officer creates a panel of
expert researchers who then evaluwhat I've written, and what I might
ate requests for money from peers
write in future. Based on a proposal
in their engineering and scientific
drawn from my knowledge of computer research being done in Africommunities.
ca, an NSF panel awarded me (and
To prevent rigidity and bias, protwo others at ASU) ample fund for
gram officers usually serve for a few
our study.
years and then return to their origiTo be sure, peer review is only as
nal universities. Some come from
strong as the panels themselves and
industry. Some do multiple stints as
as the language and concepts in the
program officers, though often seprequests for funding. And the model
arated by many years.
is inherently conservative, biased
Government-funded researchers,
against novelty in the manner histormeanwhile, are free to do their work
ically documented and conceptually
in their home locations and publish
outlined by Thomas Kuhn in his The
their work anywhere. Since they've
Structure of Scientific Revolutions.
already been paid to carry out their
Yet whatever its shortcomings, the
research, they need not modify their
mechanism nicely maps onto the
results or findings in order to satisfy
world of serious journalism, where
commercial interests. Having been
beats are akin to research fields and
paid once, they have a civic and
past publications and the potential
legal obligation to share their knowlvalue of research aims resemble the
edge widely. They can even get paid
way journalists long have obtained
a second time for their work - for
support: they advocate for plans
instance by publishing a book or givand projects, and whether they get
ing a lecture for a fee - though they
support from editors depends on
aren't required to do so.
what they know, the quality of their
Full disclosure: I myself have
past work, and the feasibility of their
received an NSF grant for field
future goals. Substitute editors for
research in the East African nation
program managers and the process
of Uganda. For three years runcould seem seamless.
ning, earlier this decade, I visited

4

The literature on the merits and defects of
peer review is large. For a comprehensive, early
survey, see [5].

SEPTEMBER 2020

∕

IEEE TECHNOLOGY AND SOCIETY MAGAZINE

27
IEEE Technology and Society Magazine - September 2020

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - September 2020
