[Review and notes] How Democracy Ends—David Runciman

Word count: 981

Read­ing time: 4 minutes

Key­words: democ­racy, au­thor­i­tar­i­anism, pop­ulism, lock-in, tech­nol­ogy, policy, global catas­trophic risk, cli­mate change, ex­is­ten­tial risk

Au­dio­book length: 7hrs 39mins (at 1x)


David Runci­man is an English aca­demic who teaches poli­tics and his­tory at Cam­bridge Univer­sity, where he is Head of the Depart­ment of Poli­tics and In­ter­na­tional Stud­ies. He is the au­thor of sev­eral books, in­clud­ing The Con­fi­dence Trap: A His­tory of Democ­racy in Cri­sis from World War I to the Pre­sent, and is host of the weekly Talk­ing Poli­tics pod­cast.

I re­cently en­joyed read­ing this book and I think it had many rele­vant and in­ter­est­ing ideas for effec­tive al­tru­ists and peo­ple in­ter­ested in pos­i­tively shap­ing the fu­ture. I think the book is ex­cel­lent, and if you liked this sum­mary, I would recom­mend sup­port­ing it by go­ing out and buy­ing it!


1. Coup?

We should not fear a re­turn to the 1930s. We should not ex­pect coups in the fu­ture to look like the past for three rea­sons: vi­o­lence has de­clined, we are much richer than be­fore, and our in­sti­tu­tions have learnt the les­sons of the past.

The big­ger threats are from the de­cay of our in­sti­tu­tions. In­stead, Nancy Ber­meo ar­gues that democ­rac­tic back­slid­ing has be­come more preva­lent since the Cold War. In par­tic­u­lar, Runci­man ar­gues that ‘ex­ec­u­tive ag­gran­dize­ment—when elected strong­men chip away at democ­racy while pay­ing lip ser­vice to it—looks like the biggest threat to democ­racy in the twenty-first cen­tury’.

Con­spir­acy the­o­ries, post-truth, and pop­ulism are likely to in­crease. Pop­ulism claims that democ­racy has been stolen by elites, and that the peo­ple need to claim it back. This view be­comes more preva­lent when peo­ple feel that they have lost out, and is as­so­ci­ated with rises in in­equal­ity.

2. Catas­tro­phe?

There are three main ways mod­ern civil­i­sa­tion could de­stroy it­self: through weapons of mass de­struc­tion, from fatal en­vi­ron­men­tal degra­da­tion, or by the lock-in of bad val­ues such as in to­tal­i­tar­ian states. Runci­man gives as ex­am­ples for each three: Silent Spring, Hiroshima, and Han­nah Arendt’s fa­mous piece ‘on the ba­nal­ity of evil’.

Th­ese three is­sues are caused more by inat­ten­tion than at­ten­tion, and so demo­cratic sys­tems may strug­gle to ad­dress them:

  • Cli­mate change poses a se­ri­ous threat to fu­ture civil­i­sa­tion, but it is too diffuse and un­even for peo­ple to feel its effects clearly and re­spond to them. This makes con­scious­ness-rais­ing harder.

  • Though dur­ing the Cold War, the nu­clear threat was a pop­u­lar topic of dis­cus­sion. The Cam­paign for Nu­clear Disar­ma­ment (CND) had two mil­lion mem­bers at its peak, but now has fewer than a few thousand

  • The threat of to­tal­i­tar­i­anism re­mains pre­sent, and his­to­ri­ans such as Ti­mothy Sny­der, have ar­gued that democ­racy com­pla­cency led to Weimar Germany

Runci­man ar­gues that these three threats re­main real, and that democ­racy may not be able to ad­dress them. It might be that the end of civil­i­sa­tion comes be­fore the end of democ­racy. He quotes Bostrom:

“democ­ra­cies will find it difficult to act de­ci­sively be­fore there has been any visi­ble demon­stra­tion of what is at stake. Wait­ing for such a demon­stra­tion is de­cid­edly not an op­tion, be­cause it might it­self be the end”
Nick Bostrom, Ex­is­ten­tial Risks: An­a­lyz­ing Hu­man Ex­tinc­tion Sce­nar­ios and Re­lated Hazards

Runci­man then dis­cusses the ‘rar­efied at­mo­sphere’ of ex­is­ten­tial risk, mak­ing refer­ence to Parfit, Bostrom, and the AI al­ign­ment prob­lem. Some peo­ple ar­gued that democ­racy ended when the atomic bomb was cre­ated—it was the biggest lever on the fu­ture, and it was out­side of the hands of in­di­vi­d­u­als. Runci­man ar­gues that ‘democ­racy can­not con­trol ex­is­ten­tial risk’.

3. Tech­nolog­i­cal takeover?

In this chap­ter, Runci­man looks at mod­els of the state as a ma­chine, go­ing back to Hobbes’ Le­viathan, and con­sid­ers how the in­ter­net could fa­cil­i­tate di­rect democ­racy.

Runci­man ar­gues that the states’ leviathan is what is needed to ad­dress cor­po­rate cap­ture and reg­u­late en­tities such as Face­book and big oil com­pa­nies. In­di­vi­d­u­als have a limited abil­ity to tackle these be­he­moths.

In re­al­ity, David can­not beat go­liath—you need an­other go­liath in the form of the state. The leviathan of the peo­ple as a col­lec­tive is the only thing big enough to take on the large or­gani­sa­tions and ex­ter­nal­ities threat­en­ing civil­i­sa­tion. [I’m re­minded of Founders Pledge’s em­pha­sis on policy as the best route to mak­ing progress on cli­mate change, and also Bruce Sch­neier’s quote in the 80,000 Hours pod­cast, which I’ve pul­led out be­low, where he says that policy re­sponses push­ing back on banks was the only way to im­prove se­cu­rity for in­di­vi­d­u­als.]

How­ever, Runci­man ar­gues that the in­for­ma­tion age has tended to re­in­force au­thor­i­tar­ian states rather than un­der­mine them. For ex­am­ple, China’s uni­ver­sal credit sys­tem, and al­gorith­mic bias and dis­crim­i­na­tion risk en­trench­ing bad val­ues and in­equal­ity.

While di­rect democ­racy is ap­peal­ing, it can have ter­rify­ing and dis­as­trous con­se­quences (for in­stance the gen­eral pub­lic sup­ports the re­turn of the death penalty). There is an open ques­tion of how more re­spon­sive tech­nol­ogy can in­te­grate with democ­racy to de­liver bet­ter out­comes.

4. Some­thing bet­ter?

Some con­tem­po­rary in­tel­lec­tu­als claim that democ­racy is lead­ing to highly re­ac­tive and jin­go­is­tic poli­tics, and that maybe we should con­sider al­ter­na­tives, namely:

  • Prag­matic au­thor­i­tar­i­anism—one of the strengths of au­thor­i­tar­ian states is that they can move quickly and de­ci­sively, and this could be a good way to avoid im­mi­nent threats. But util­i­tar­i­anism can quickly be­come au­thor­i­tar­i­anism, and the jus­tifi­ca­tion of ex­pe­di­ency creeps quickly to en­abling rulers and tyrants be­yond re­proach.

  • Epis­toc­racy—defined as rule by knowl­edge­able peo­ple, this is a var­i­ant on tech­noc­racy. Rule by tech­nocrats is de­rided as be­ing all about the ma­chine and not about hu­man lives. Epis­toc­racy, by con­trast, claims to ap­pre­ci­ate the im­por­tant things, but to be gov­erned by wise peo­ple. The prob­lem is that wise peo­ple are just as prey to cog­ni­tive bi­ases as ev­ery­one else.

  • Liber­ated tech­nol­ogy. Ru­nic­man dis­cusses ac­cel­er­a­tionism—a sort of pro-growth fu­tur­ism dis­t­in­guished from de-growth en­vi­ron­men­tal­ists. This could greatly ex­pand our moral con­cern. Parfit ar­gued that our cling­ing to per­sonal iden­tity led to a lack of com­pas­sion for those dis­tant from us in space and time. But strong re­vi­sion­ism risks tear­ing apart in­sti­tu­tions and sen­si­bil­ities that are there for a rea­son. In 1919, the Ital­ian Par­tito Poli­tica Fu­tur­ista merged with Mus­soli’s fas­cists.

Runci­man con­cludes that, of the three op­tions, democ­racy main­tains a good op­tion. The ran­dom­ness and change in democ­racy avoids lock-in of bad situ­a­tions, such as au­thor­i­tar­ian states.

Longer-term, liber­ated tech­nol­ogy may offer us bet­ter ex­pe­riences than democ­racy cur­rently does, but that the down­sides could be more ter­rible than any­thing we’ve known. Which route we choose, and what the fu­ture looks like, is up to us to de­cide.

Fur­ther reading

Bruce Sch­neier: But you, Mr Economist, un­der­stand the no­tion of ex­ter­nal­ities and a bank is not go­ing to fix the prob­lem if some­one else has the prob­lem. So, I mean, in 1978, in the United States, we passed the Fair Credit Re­port­ing Act. And one of the things it did, is it limited li­a­bil­ities for credit card losses to the in­di­vi­d­ual to $50. And this was a game-changer in credit card se­cu­rity. Be­fore that law, credit card com­pa­nies would ba­si­cally charge the user for fraud. Your credit card got stolen or lost and you were stuck with the bill un­til like the two weeks un­til the com­pany could print the new lit­tle book with bad num­bers. When Congress passed that law, sud­denly the credit card com­pa­nies were ab­sorb­ing all the losses. They couldn’t pass it to the con­sumer.
Robert Wiblin: And they fixed it very fast.
Bruce Sch­neier: But they did so many things that the con­sumer could never do. So think of what they did. Real-time ver­ifi­ca­tion of card val­idity. Micro­print­ing on the cards. And the holo­gram to make them less forge­able. Ship­ping the cards and a PIN to the user in sep­a­rate en­velopes. Re­quiring ac­ti­va­tion from a phone that was rec­og­nized. Now, if you’re a user and you’re get­ting those losses you couldn’t im­ple­ment any of those things.
But the credit card com­pany could. They just never did be­cause they never suffered the losses.
Robert Wiblin: Give the cost to the group that can do the most to fix the prob­lem is just an ob­vi­ous ap­proach.
Bruce Sch­neier on the 80,000 Hours podcast