Which nuclear wars should worry us most?

Summary

  • A nu­clear ex­change may have the po­ten­tial to kill mil­lions or billions of peo­ple, and pos­si­bly lead to hu­man ex­tinc­tion.

  • In this post, I rank plau­si­ble nu­clear ex­change sce­nar­ios in terms of their po­ten­tial to cause harm based on three fac­tors: 1) The size of the in­volved coun­tries’ nu­clear ar­se­nals; 2) The size of the in­volved coun­tries’ pop­u­la­tions; 3) The prob­a­bil­ity of the given nu­clear ex­change sce­nario.

  • Based on my rough pri­ori­ti­za­tion, I ex­pect the fol­low­ing nu­clear ex­change sce­nar­ios have the high­est po­ten­tial for harm:

  1. Rus­sia and the US

  2. In­dia and Pakistan

  3. China and ei­ther the United States, In­dia, or Russia

Pro­ject Overview

This is the first post in Re­think Pri­ori­ties’ se­ries on nu­clear risks. In this post, I look into which plau­si­ble nu­clear ex­change sce­nar­ios should worry us most, rank­ing them based on their po­ten­tial to cause harm. In the sec­ond post, I ex­plore the make-up and sur­viv­abil­ity of the US and Rus­sian nu­clear ar­se­nals. In the third post, I es­ti­mate the num­ber of peo­ple that would die as a di­rect re­sult of a nu­clear ex­change be­tween NATO states and Rus­sia. In the fourth post, I es­ti­mate the sever­ity of the nu­clear famine we might ex­pect to re­sult from a NATO-Rus­sia nu­clear war. In the fifth post, I get a rough sense of the prob­a­bil­ity of nu­clear war by look­ing at his­tor­i­cal ev­i­dence, the views of ex­perts, and pre­dic­tions made by fore­cast­ers. Fu­ture work will ex­plore sce­nar­ios for In­dia and Pak­istan, sce­nar­ios for China, the con­tra­dic­tory re­search around nu­clear win­ter, the im­pact of sev­eral nu­clear arms con­trol treaties, and the case for and against fund­ing par­tic­u­lar or­ga­ni­za­tions work­ing on re­duc­ing nu­clear risks.

Toward a bet­ter un­der­stand­ing of nu­clear risks

A nu­clear ex­change may have the po­ten­tial to kill mil­lions or billions of peo­ple, and pos­si­bly lead to hu­man ex­tinc­tion. There have been many cases where nu­clear weapons have al­most been launched by mis­take (Baum, de Neufville & Bar­rett, 2018).[1] And if a nu­clear ex­change — started by ac­ci­dent or on pur­pose — were to es­ca­late to a full-scale nu­clear war, the nu­clear deto­na­tions could lead to a nu­clear win­ter, a state where soot launched into the at­mo­sphere blocks out enough sun­light to cause a famine so se­vere and long-last­ing that al­most ev­ery­one on Earth could starve to death be­fore its end.

Be­cause there seems to be a non-neg­ligible prob­a­bil­ity of a large-scale nu­clear ex­change, and be­cause the stakes would be so high in the event that a nu­clear ex­change did es­ca­late, many effec­tive al­tru­ists be­lieve re­duc­ing nu­clear risks should be among the top pri­ori­ties for the EA com­mu­nity. For ex­am­ple, 80,000 Hours pub­lished a prob­lem pro­file on nu­clear se­cu­rity, giv­ing a score of 15 out of 16 on Scale (though it scores rel­a­tively low on Solv­abil­ity and Ne­glect­ed­ness).[2]

But my sense is that some de­tails of the nu­clear risks prob­lem area aren’t well-un­der­stood by most EAs — for ex­am­ple, how bad nu­clear war would ac­tu­ally be, the mechanisms be­hind nu­clear win­ter, and where EAs that pri­ori­tize re­duc­ing nu­clear risks should donate. In a num­ber of up­com­ing posts, I’ll try to un­der­stand, in some­what con­crete terms, how much harm nu­clear war would cause and how plau­si­ble nu­clear risks are. One of the things I’ll do to bet­ter un­der­stand the risks posed by nu­clear win­ter is re­view the im­pli­ca­tions of re­cent aca­demic liter­a­ture that is in­ter­preted by some as cast­ing doubt on the sci­ence be­hind the nu­clear win­ter phe­nomenon. Fi­nally, I’ll also eval­u­ate some of the work be­ing done to re­duce nu­clear risks. In par­tic­u­lar, I’ll fo­cus on a re­cent treaty that’s been adopted by the United Na­tions, the Treaty on the Pro­hi­bi­tion of Nu­clear Weapons (TPNW), which would make the re­search and use of nu­clear weapons ille­gal in all coun­tries that rat­ify the treaty.

Fo­cus­ing on the most trou­bling nu­clear risks

In the first few posts, I’ll con­sider the prob­a­bil­ity and sever­ity of sev­eral nu­clear war sce­nar­ios, look­ing sep­a­rately at the amount of harm that would be caused from both the short-ter­mist per­spec­tive and the long-ter­mist one.

It would be in­tractable to es­ti­mate the im­pacts of ev­ery imag­in­able nu­clear war sce­nario. So I in­stead fo­cus on nu­clear war sce­nar­ios which I ex­pect make up the ma­jor­ity of the ex­pected harm that would be caused by nu­clear war. Th­ese are sce­nar­ios in which:

  1. The speci­fied coun­tries have rel­a­tively large nu­clear ar­se­nals. This an im­por­tant con­sid­er­a­tion for two rea­sons: first, be­cause the size of the nu­clear ar­se­nal is a ma­jor fac­tor in how se­vere the di­rect effects of a nu­clear ex­change are — more nu­clear weapons can cause many more deaths. Se­cond, whether a nu­clear ex­change leads to a nu­clear win­ter de­pends in large part on the num­ber and size of nu­clear weapons deto­nated. It would take a lot of nu­clear weapons to pro­duce a nu­clear win­ter se­vere enough to cause a wor­ld­wide famine that could lead to hu­man ex­tinc­tion. Given that I’m most wor­ried about nu­clear win­ter sce­nar­ios that pose an ex­tinc­tion risk, I be­lieve we should fo­cus on nu­clear ex­change sce­nar­ios that would in­volve large nu­clear ar­se­nals.

  2. The coun­tries in­volved have large pop­u­la­tions. This is an im­por­tant con­sid­er­a­tion for similar rea­sons: first, be­cause the pop­u­la­tion size of a coun­try is an im­por­tant fac­tor in de­ter­min­ing how many peo­ple could die as a di­rect re­sult of the nu­clear deto­na­tions. Se­cond, whether a given nu­clear ex­change would lead to a nu­clear win­ter de­pends on how much smoke is pro­duced from the burn­ing of cities dur­ing the ex­change. Coun­tries with larger, densely pop­u­lated cities have much more flammable ma­te­rial. This means that a nu­clear ex­change in­volv­ing densely pop­u­lated coun­tries would be more likely to lead to nu­clear win­ter, all else equal.

  3. The spe­cific con­flict sce­nar­ios are rea­son­ably prob­a­ble. This is im­por­tant be­cause the ex­pected harm is higher in nu­clear ex­change sce­nar­ios that are more likely to ac­tu­ally hap­pen, all else equal.

When I con­sider these fac­tors for all 9 nu­clear weapons pos­ses­sor states, I get a rough rank­ing of var­i­ous plau­si­ble nu­clear ex­change sce­nar­ios in terms of their ex­pected dis-value:

The rank­ing is based on a crude scor­ing sys­tem that ap­prox­i­mates the ex­pected harm that would be caused by each sce­nario. Each of the fac­tors that in­forms how ter­rible a given nu­clear ex­change would be — the size of the in­volved coun­tries’ nu­clear ar­se­nals, the size of the pop­u­la­tions of the in­volved coun­tries, and the prob­a­bil­ity of the spe­cific sce­nario — were as­signed a score of 1 (shaded in green) , 2 (yel­low), or 3 (red), where a fac­tor with a score of 1 should worry us less, and a fac­tor with a score of 3 should worry us a lot. For each sce­nario, the fac­tor scores were summed to­gether to pro­duce the Ex­pected Harm score.

Note that some end­notes are em­bed­ded in the table image but can be seen here: source for ar­se­nal size[3]; source for me­dian war prob­a­bil­ity[4]; note on non-state ac­tors[5].

Next Steps

Based on this rough pri­ori­ti­za­tion, I’ll spend sev­eral posts look­ing at the amount of harm we would ex­pect to see caused by the fol­low­ing nu­clear ex­change sce­nar­ios:

  1. A sce­nario where the US and Rus­sia use nu­clear weapons, in­ten­tion­ally or ac­ci­den­tally, in a con­ven­tional nu­clear ex­change.

  2. A sce­nario where In­dia and Pak­istan use nu­clear weapons, in­ten­tion­ally or ac­ci­den­tally, in a con­ven­tional nu­clear ex­change.

  3. A sce­nario where China uses nu­clear weapons, in­ten­tion­ally or ac­ci­den­tally, in a con­ven­tional nu­clear ex­change with ei­ther the United States, In­dia, or Rus­sia.

Edits and Corrections

July 17 2019 — I re­placed a sim­plified ver­sion of the table rank­ing po­ten­tial con­flict sce­nar­ios on the ba­sis of nu­clear ar­se­nal size, pop­u­la­tion size, and sce­nario prob­a­bil­ity with a more de­tailed ver­sion. I also added a brief ex­pla­na­tion of how the Ex­pected Harm scores were calcu­lated.

Credits

This es­say is a pro­ject of Re­think Pri­ori­ties. It was writ­ten by Luisa Ro­driguez. Thanks to Peter Hur­ford, Marinella Capriati, Ida Sprengers, Mar­cus A. Davis, and Neil Dul­laghan for their valuable com­ments. Thanks also to Matt Gentzel, Seth Baum, and Carl Shul­man for pro­vid­ing guidance and feed­back on the larger pro­ject. If you like our work, please con­sider sub­scribing to our newslet­ter. You can see all our work to date here.

Bibliography

Apps, P. (2015). PS21 Great Power Con­flict Re­port (Rep.). Pro­ject for the Study of the 21st Cen­tury. Retrieved from: https://​​www.scribd.com/​​doc­u­ment/​​289407938/​​PS21-Great-Power-Con­flict-Report

Baum, S., de Neufville, R., & Bar­rett, A. (2018). A model for the prob­a­bil­ity of nu­clear war. So­cial Science Re­search Net­work (SSRN). https://​​doi.org/​​10.2139/​​ssrn.3137081

Daniel Ells­berg on the cre­ation of nu­clear dooms­day ma­chines, the in­sti­tu­tional in­san­ity that main­tains them, and a prac­ti­cal plan for dis­man­tling them [Au­dio blog in­ter­view]. (2018, Septem­ber 24). Retrieved April 14, 2019, from https://​​80000hours.org/​​pod­cast/​​epi­sodes/​​daniel-el­ls­berg-dooms­day-ma­chines/​​

Kristensen, H. M., & Korda, M. (2019). Rus­sian nu­clear forces, 2019. Bul­letin of the Atomic Scien­tists. https://​​doi.org/​​10.1080/​​00963402.2019.1580891

Kristensen, H. M., & Nor­ris, R. S. (2014). Is­raeli nu­clear weapons, 2014. Bul­letin of the Atomic Scien­tists. https://​​doi.org/​​10.1177/​​0096340214555409

Kristensen, H. M., & Nor­ris, R. S. (2016). Pak­istani nu­clear forces, 2016. Bul­letin of the Atomic Scien­tists. https://​​doi.org/​​10.1080/​​00963402.2016.1241520

Kristensen, H. M., & Nor­ris, R. S. (2017). In­dian nu­clear forces, 2017. Bul­letin of the Atomic Scien­tists. https://​​doi.org/​​10.1080/​​00963402.2017.1337998

Kristensen, H. M., & Nor­ris, R. S. (2018a). Chi­nese nu­clear forces, 2018. Bul­letin of the Atomic Scien­tists. https://​​doi.org/​​10.1080/​​00963402.2018.1486620

Kristensen, H. M., & Nor­ris, R. S. (2018b). North Korean nu­clear ca­pa­bil­ities, 2018. Bul­letin of the Atomic Scien­tists. https://​​doi.org/​​10.1080/​​00963402.2017.1413062

Kristensen, H. M., & Nor­ris, R. S. (2018c). United States nu­clear forces, 2018. Bul­letin of the Atomic Scien­tists. https://​​doi.org/​​10.1080/​​00963402.2018.1438219

McIn­tyre, P. (2016, April 12). How you can lower the risk of a catas­trophic nu­clear war. Retrieved from https://​​80000hours.org/​​prob­lem-pro­files/​​nu­clear-se­cu­rity/​​

Talley, I. (2019, March 22). U.S. says Iran poised to re­sume work on nu­clear weapons. Wall Street Jour­nal. Retrieved May 19, 2019, from https://​​www.wsj.com/​​ar­ti­cles/​​u-s-says-iran-poised-to-re­sume-work-on-nu­clear-weapons-11553263221

Treaty on the pro­hi­bi­tion of nu­clear weapons – UNODA. (n.d.). Retrieved May 19, 2019, from https://​​www.un.org/​​disar­ma­ment/​​wmd/​​nu­clear/​​tpnw/​​


  1. See pages 27-32 for an item­ized list of the US-Rus­sia near misses (Baum, de Neufville & Bar­rett, 2018). ↩︎

  2. Note, they now recom­mend peo­ple use their in-depth in­ter­view with Daniel Ells­berg as a source of in­for­ma­tion on nu­clear se­cu­rity. ↩︎

  3. Arse­nal data from the Bul­letin of the Atomic Scien­tists’ Nu­clear Note­books. ↩︎

  4. I be­lieve the Pro­ject for the Study of the 21st Cen­tury (PS21) Great Power Con­flict Re­port has sev­eral ty­pos (Apps, 2015). I pre­sent what I be­lieve to be the cor­rect val­ues (and the val­ues I use in my anal­y­sis) here. ↩︎

  5. While a nu­clear deto­na­tion by a non-state ac­tor (ter­ror­ist) looks plau­si­bly quite harm­ful in ex­pec­ta­tion, it’d be very difficult to an­a­lyze, as there’s no sin­gle ter­ror­ism sce­nario to con­sider. I there­fore leave a dis­cus­sion of the po­ten­tial harm caused by nu­clear ter­ror­ism for fu­ture work. ↩︎