7 December 2019 | New Scientist | 45evidence that the positive feelings people get
from donating to a cause they feel strongly
about or volunteering with a charity, where
there is engagement with the people you seek
to help, can encourage more persistent giving.
MacAskill doesn’t buy this. “I think there’s
only a superficial tension between empathy
and effective altruism,” he says. “It’s really
a failure of empathy if you can only empathise
with people you’ve interacted with.”
What he and other effective altruists
do accept, however, is that GiveWell’s
recommended charities should serve merely
as a baseline. Indeed, MacAskill points out
that finding new and unusual approaches
to maximising your impact, even where the
chances of success are less than perfect, is
increasingly important to the movement.
The trouble is that thinking in such terms
can lead to some peculiar conclusions.Are all lives equal?
To see why things can get strange, remember
the thought experiment about the child in the
pond. It taught us that two lives separated in
space ought to be treated equally. You can apply
similar thinking to the time dimension. If you
could save one life now or 10 lives in 30 years,
should you not choose the latter?
Effective altruists increasingly seem to think
so. In 2012, GiveWell partnered with Good
Ventures, a foundation set up to give away
some of the huge wealth of Dustin Moskovitz,
a co-founder of Facebook, and his wife Cari
Tuna. This, in turn, led to the formation of the
Open Philanthropy Project, which is focused
on unusual modes of giving – and committed
to offering long-term support to high-risk
projects, provided the potential pay-offs are
sufficiently large. In practice, that has come
to include funding research that could prevent
existential threats to humanity, such as
pandemics, catastrophic climate change
and – most controversially – an apocalypse
caused by rogue artificial intelligence. Indeed,
in 2017, Open Philanthropy agreed to give
$30 million over three years to San Francisco-
based OpenAI, whose stated mission is nothing
short of “discovering and enacting the path to
safe artificial general intelligence”.
The reason for this is that many effective
altruists have become convinced that they can
do vastly more good in the future. They argue
that, in terms of sheer numbers of lives, the
stakes are enormous. The most important
moral imperative, therefore, is to prevent
anything that could kill vast numbers of
people in the centuries to come.ImpactMatters. “What they do, they do well.
But we want to do 1000 impact estimates
by the end of 2019, whereas they do more
like two or three assessments a year.”
Making more evidence available has to be
a good thing. But I wonder if relying on these
sorts of evaluations makes us too passive.
More broadly, what if divorcing altruism fromemotion, as many effective altruists
recommend, is ultimately unhelpful?
Jamil Zaki, a psychologist at Stanford
University in California, has argued that
effective altruism is “misguided” because it fails
to account for empathy’s role in motivating
and maintaining philanthropy. He points tocentralised form of giving, where a select
few charities monopolise donations.
GiveWell has recently said it is aiming
to begin investigating philanthropy where
the outcomes are less easy to measure.
Meanwhile, Will MacAskill, a philosopher
at the University of Oxford, insists that
effective altruism in general is broadening
its scope. “Perhaps even most of the effort in
the community now is on things where it’s
much harder to measure impact,” he says.
GiveWell is certainly not the only game in
town. In 2015, Dean Karlan, an economist
now at Northwestern University in Illinois,
helped set up ImpactMatters in an attempt to
give a broader range of charities the chance to
have the amount of good they do accurately
evaluated. Instead of insisting on the most
rigorous research, it looks at the charity’s
own data on the outcomes of its programmes
and checks to see if its general approach
is supported by independent studies
elsewhere. This allows it to assess all manner
of charities. “We have no criticism of GiveWell,”
says Michael Weinstein, the president of
“ Thinking in terms
of maximum impact
can lead to some
strange conclusions”
Funding mosquito
nets is recognised
as one of the most
effective ways to
save lives>