War and technological progress

One of the big benefits of blogging for me is the chance to try out my ideas on an audience I couldn’t easily reach (or at least hear back from) in any other way. That’s particularly true when I’m writing a book, which is always a difficult process for me. My last post, on the opportunity cost of war produced a great comments thread. Particularly useful was a discussion, started by Chris Bertram at Crooked Timber, of the oft-heard claim that war stimulates scientific and technological progress. I’ve used my response, along with points appropriated from commenters to draft a new section for the book, pointing out how this claim ignores the problem of opportunity cost.

As always, comments of (nearly) all kinds are appreciated, and useful ones may be recycled.

Despite, or perhaps because of, the obvious waste and destruction of war, it’s often claimed that war has economic benefits, and even that it’s necessary to the successful functioning of the economy. One version of this argument, based on the idea of ‘military Keynesianism’ will be discussed later.

In this section, we’ll look at another popular argument, that war is a spur to research and development (R &D), and therefore to peacetime prosperity. This idea has some superficial appeal. Penicillin, nuclear energy, computers and jet aircraft are examples of technologies that were developed, or advanced rapidly, during World War II, and played a major role in postwar prosperity.

In all of these cases, the underlying research had been undertaken in the 1920s and 1930s. The outbreak of war led to a massive push to apply this research on an industrial scale, producing millions of doses of penicillin, hundreds of thousands of jet airplanes, and of course the atomic bomb. ENIAC, the first electronic general-purpose computer was commissioned to compute artillery tables, but did not appear until 1946, when it was used in computations to produce the first hydrogen bomb.

Opportunity cost reasoning leads us to ask what was foregone to release the resources. In large part, the answer is ‘research of the kind that made these developments possible’. War gives great urgency to the “D” part of R&D, at the expense of R. This can produce some impressive short run payoffs.

To be counted against that is the loss arising when scientists are shifted from fundamental research to activities more directly relevant to the war effort, much of it with very little value beyond the immediate needs of the military. The there are the vast numbers of young scientists whose careers were interrupted because of military service, and older scientists.

For quite a few scientists war service has been more than a career interruption. Harry Moseley, widely regarded as the greatest experimental physicist of the twentieth century, was killed at Gallipoli in 1915. [Bohr (I think) said that even if no one else had died, the death of Harry Moseley alone was enough to make the First World War an unbearable tragedy.]

Henry Moseley: the single most costly death of the war


The great theoretical physicist Karl Schwarzschild died the following year. Many more died before having any chance to contribute. one can think of the 50 fatality rate suffered by the class of 1914 at the École Normale Supérieure https://books.google.com.au/books?id=EjZHLXRKjtEC&pg=PA329&lpg=PA329&dq=ecole+nationale+superieure+casualties+world+war+i&source=bl&ots=asLFDx9V5p&sig=gr4l5-65JgNhXGRaCHkEz39xzmk&hl=en&sa=X&redir_esc=y#v=onepage&q=ecole%20nationale%20superieure%20casualties%20world%20war%20i&f=false.

A tragic and heroic story from World War II is that of the scientists of the Pavlovsk Experimental Station near Leningrad (now St Petersburg), twelve of whom starved to death while protecing the station’s seed bank during the siege of the city in 1941. Other losses include the mathematicians Jean Cavailles, shot by the Gestapo, and Wolfgang Doblin, one of thousands of Jewish scientists and doctors who perished in the Holocaust.

As this example shows, scientific projects themselves were not immune from the destruction. The first programmable computer to be built was not ENIAC, but the Z1 designed by German Konrad Zuse. This computer and its successors, the Z2 and Z3 were destroyed by Allied bombing raids, and Zuse’s work was not resumed for years.

Yet again, the idea of opportunity cost as ‘that which is not seen’ provides a corrective against any attempt to minimise the costs of destruction.

46 thoughts on “War and technological progress

  1. Ikonoclast @12,

    You can say those things but I doubt that they are true. There is a huge amount of development work done by industry that you will never hear about. An example, the Rev Bill Crews being interviewed reveals that as a young out of uni Phd he was employed by AWA and for this company was the first person in Australia to produce ultra pure moncrystaline silicon and make integrated circuits from this material. Not the CSIRO, but AWA. Who would have known. Intel, Microsoft, Ford, German chemical companies

    http://www.gtai.de/GTAI/Content/EN/Invest/_SharedDocs/Downloads/GTAI/Brochures/Industries/the-german-chemical-industry.pdf

    , IBM, Bell telephone Laboratories, …..Huge and long list of Private industry doing ground breaking research.

    I don’t believe the argument that most research today is done by government would stand up to scrutiny.

  2. A rather different point. You can make a better case that cold wars – conditions of intense political rivalry without actual fighting – do accelerate technical progress more cheaply. Consider not only the US-USSR conflict in 1945-1990, but the rivalry between Imperial Germany and Britain and France in the decades before 1914. Dreadnoughts, aircraft, chemicals, radio. The absence of overt warfare made the cash and opportunity cost of the military spending, and military-related research, much lower. It’s difficult to imagine DARPA and the first Internet without the Cold War. Mind you, the second Internet – the Web – was invented at CERN, an entirely civilian research centre.

  3. PS: An earlier example is the Longitude Prize and Harrison’s marine chronometer. The impulse for the enormous prize was the loss of an entire British fleet by Cloudesley Shovell on the Scillies in 1707. The background was the intense maritime competition with France throughout the 18th century. The problem was extremely difficult and required numerous advances in precision clockmaking to solve.

    It’s also noteworthy because of the success of the prize mechanism, now being revived on a small scale by DARPA and Google, and considered by mavericks as an alternative to pharmaceutical patents. Also by the openness of the prize to all comers. The great Swiss mathematician Leonhard Euler won an award of £300 (£50,000 today? – a bourgeois income for a year) for advancing the astronomical method.

  4. PPS: there is a Greek saying attributed to Heraclitus: “Strife is the father of all things”. It is sometimes translated as “war is …” but polemos is apparently wider, and the epigram works better in the more general sense.

  5. @BilB

    “I don’t believe the argument that most research today is done by government would stand up to scrutiny.” – BilB.

    But that wasn’t my argument was it?

    I wrote, to quote directly from my comment 12: “Much R&D now is only going to happen at government level or transnational corporation level.”

    My point really was that modern research, for the most part, is complicated, expensive and highly technical. Only large organisations with large resources, for the most part, are going to be able to fund it.

    This is a quandary for those who dislike big government and/or big corporations. We seem to need vast conglomerates to run big research and build big projects. How do we square that off against the desire to avoid oligopolies and monopolies be they state or private?

    For my part, if government is genuinely social democratic I am happy for it to be big and to run natural monopolies. Further, we need to explore how to run worker cooperative conglomerates (worker owned and managed rather than capitalist owned and managed). There also needs to be a better stakeholder model which recognizes all citizens are impacted by big projects and all citizens need a say in them.

  6. @James Wimberley

    In the early days of the scientific and industrial revolutions, the lone “natural philosopher” (proto-scientist), researcher, maker or inventor could make discoveries and advances. This is much less the case today.

    “John Harrison (3 April [O.S. 24 March] 1693– 24 March 1776) was a self-educated English carpenter and clockmaker.” – Wikipedia.

    I haven’t noticed any self-educated, home-spun physicist building CERN or the equivalent on his/her own.

    This goes back to my discussion with BilB in posts 12 and 30. Big organisations make the main advances now for the most part. The question essentially is how do we operate and regulate these big organizations? For private benefit or social benefit? The private benefit model concentrates wealth, creates large inequalities and ignores serious and damaging negative externalities.

  7. Ikonoclast, whereas you are more than likely correct in claiming that high expenditure R&D is undertaken by Government Institutions (universities predominately) and transnational businesses on the dollars spent scale, I feel that the broader variety of R&D is undertaken by small business and private individuals. This other level of research is done with low cost techniques, but that should not be seen as being ineffective.

    The point that I am making is that the amount of research being undertaken has ballooned way beyond the scale that a war could stimulate. We are in a technology age which would only be suppressed by war.

  8. For example

    http://blog.cafefoundation.org/2015-british-human-powered-flying-club-rally/

    not to mention the CAFE (Comparative Air Flight Efficiency) Foundation itself. In aviation the majority of primary innovations have come from the efforts of individuals. Corporate Aviation expands and refines those features for commercial exploitation.

    Electric aviation began with hobbyists with and model air planes. The first electric powered man carrying aircraft was a microlight form a paddock in a suburb. The first fully functional 2 seater solar only powered aircraft was made by a husband and wife team. Airbus are now moving into the field of electric aviation with their eFan project (which is largely driven by the design efforts of one person). The first micro gas turbine was an individual who built a working engine using parts such as a camping gas canister for the combustion chamber. That technology is now a booming industry.

    A thorough study of R&D would reveal some huge surprises, I believe.

  9. It’s nice to say that a rational government could have given us all penicillin, jet airplanes and computers with the right employment creating investment in the 30s, without any need for a war. It’s probably true. But by that standard, we didn’t have a rational government, we don’t have a rational government, and we never will have a rational government. So the question seems irrelevant as posed.

    A better question for political economists would not be “could we have gotten all this great stuff earlier if we didn’t have warmongering, economically irrational governments” but “given that we will always have warmongering, economically irrational governments, does having an actual war lead to better economic outcomes than not having one (e.g. having a cold war military industrial complex, or a 1930s economy with no state investment).”

    An important effect of WWII not discussed here is the impetus it gave to anticolonial struggles worldwide, especially Asia.

  10. As much as I love the Internet, computers, video players and other high-tech products, I think the price we have paid for this is far too high: around 16 million dead in the First World War, 60 million dead in the Second World War, millions more dead in the Korean and Vietnam Wars, many hundreds of thousands dead in Iraq, Afghanistan, Libya, Syria, Ukraine, Yemen etc., the enslavement of most of the Irish nation since the time of Cromwell, the enslavement of the British working class in the satanic mills of the 18th and 19th centuries, the enslavement of much of black Africa and the destruction of stable indigenous societies in Australia and the Pacific.

    This is only some of what has occurred in human history in the second millennium AD. I think humanity would be much better off without the high technology and the terrible price paid for it. Whilst we consider ourselves smart in the 21st century, much knowledge of past stable societies has been lost.

    There is a small hope that some of today’s world leaders, who have shown themselves to be far more humane and enlightened than Abbott, Obama, Cameron and Hollande and Merkel, may find a way out of the mess faced by the world today.

  11. @James
    There was plenty of enslavement before the high technology came along, James. There’s also not a lot of point lamenting what might have been. We can only deal with what is, and try to influence what is to come.

  12. @Nevil Kingston-Brown
    I think the lionisation of “rational government” is misguided. Reason is the slave of the passions, as David Hume (I believe) said. Many extraordinary wasteful and destructive programs (including at least some of the wars mentioned in this thread) have been prosecuted in a highly rational, technocratic manner, albeit in pursuit of the wrong ends. What we need is more like humane government. Technocracy can easily become oppressive in itself, if not restrained by some humanitarian principle.

  13. @Ikonoclast
    In the Olden Days, groups like the PMG/Telecom Australia were government funded and put a big budget in place for technological research. CSIRO were also government funded to conduct research of national significance, especially in agriculture and mining. The USA used to pursue a similar strategy, until it decided to privatise and break up the biggest such labs.

    On the Cold War, I think the accelerated scientific progress happened, but at the expense of other research which could have been done, but wasn’t. Anything which wasn’t relevant to superiority in the Cold War would have been relegated to last place in the funding trough. In other words, the Cold War placed a substantial bias upon the research topics which could easily win funding, essentially reducing the funding available for research of no special importance in the context of the Cold War.

  14. The war on accumulation type super accounts is on again, aided by ‘financial products technologies’. Superfunds invest in financial securities more or less continuously, which includes buying at prices leveraged by borrowing. Then some event, actual or imagined, triggers a deleveraging and here we go, the selling season starts and prices fall a lot. An opportunity cost (not sure it is ‘the’ opportunity cost’ because I don’t know the rankings of alternatives) for people, more so for some than for others, is higher mortgages. Just a thought on the daily struggle, which, at times looks like a trench warfare.

  15. @BilB

    You might be right and I might be out of date somewhat. The miniaturisation and electronics trends might have taken some aspects of innovation back to the “lone wolf” inventor.

  16. @Ernestine Gross

    Years ago when compulsory super was brought in, a friend of mine said to me, “That’s a big new honey-pot. The robber bees will be after that in no time.”

    I see no need for the existence of private super. This is not the same as saying “I see no need for private investment.” If people want to invest privately for super or any other purpose they should be able to do so. But there is no reason a private investment for super should get differential treatment from any other investment.

    In addition, a sensible, economically viable nation really does not need any kind of dedicated private super system (separate from private investment as such). A dedicated national public super system for all would fit the bill. It would supply reliable super for all more cheaply and efficiently and with a 100% government guarantee.

    Private super is just a honey pot for the robber bees. I am expecting, sooner or later, a huge stock market crash which will wipe out most super funds. Many people under about 50 today are unlikely (in my opinion) to ever see their super money.

  17. It is hard to read Richard Feynman’s account of his time as Los Alamos designing the atomic bomb without being impressed at the focus and intensity that was given to that effort, also the bottomless resources – intellectual and material – that the manhattan project had. The story is in the ‘los alamos from below’ lecture – check youtube. There was an urgency to the effort which i don’t think you could get under many other circumstances. I wonder if anyone would have even thought it feasible to develop a bomb in peacetime (to do so would be an act of aggression in itself?)

  18. “…mathematicians Jean Cavailles, shot by the Gestapo, and Wolfgang Doblin, one of thousands of Jewish scientists and doctors who perished in the Holocaust…”

    I would be wary of citing these as examples of the damage “done by war*. Certainly, you can argue that any violent death is as much a tragedy as any other violent death. On the other hand, most people will see a distinction between the death of somebody executed by a criminal dictatorship and the death of a soldier in battle or even civilian “collateral damage”.
    Why weaken a good argument by citing examples that people feel do not fit the discussed topic?

    Also, if you aim your text mainly at present-day situations, murderous regimes similar to the Nazis (or Stalin’s Soviet Union) are few and weak. The Islamic State and North Korea are the only more or less valid examples that I can think of. So these examples are not too relevant to a discussion related to present policies.

  19. Pr Q said:

    Particularly useful was a discussion, started by Chris Bertram at Crooked Timber, of the oft-heard claim that war stimulates scientific and technological progress…that war is a spur to research and development (R &D), and therefore to peacetime prosperity…


    War-preparing, rather than war-waging, was the dynamo driving 20th C sci-tech progress and therefore economic growth. As the Romans used to say: Si vis pacem, para bellum. The Cold War gave sci-tech the best of both worlds, a focused nation-wide drive for high-powered machinery without the massive opportunity costs of an actual shooting war. Sorry for banging on endlessly but on this blog, JUN 2013:

    The Internet, that is the TCP/IP protocol, was invented by military computer scientists – particulalry Paul Baran & Vince Cerf – working for DARPA, that is the R& D agency for the DoD. DARPA also pioneered GPS.The USA-USSR Arms Race, which includes the Space Race, was the greatest boon to sci-tech research ever. Period.


    To repeat: I am not arguing for a state policy of aggressive militarism, merely for the proven utility of national development through competitive military R & D. But the opportunity cost of our Davos-globalist “piping times of peace” seems to be a form of national intellectual stagnation.

    Recent research supports the view that, from the late 19th C through late 20th C, national military preparedness has been the most powerful spur to sci-tech progress. Tyler Cowen (NYT JUN 2014) makes the (Toynbeean) point out that civilizations that face survival challenges use their brains to over-come adversity:

    Counterintuitive though it may sound, the greater peacefulness of the world may make the attainment of higher rates of economic growth less urgent and thus less likely. This view does not claim that fighting wars improves economies, as of course the actual conflict brings death and destruction. The claim is also distinct from the Keynesian argument that preparing for war lifts government spending and puts people to work. Rather, the very possibility of war focuses the attention of governments on getting some basic decisions right — whether investing in science or simply liberalizing the economy.


    There are political and technical reasons why Military-Industrialism (M-I) turbo-charges sci-tech: On the political demand side, keeping the “the dogs of war” on a short leash mobilize public (and elite) opinion for major R & D efforts. Fear of hanging concentrates the mind wonderfully. They don’t call them Arms Race for nothing.

    On the technical supply side, military R & D drives research to fundamental levels in order to “reach for the sky”, literally the case with the A-bomb and Space Race. The computer and digital revolution were both a product of military scientists, Turing and von Neumann both worked for the DoD. Cowen elaborates:

    Ian Morris…has revived the hypothesis that war is a significant factor behind economic growth in his recent book, “War! What Is it Good For? Conflict and the Progress of Civilization From Primates to Robots.” Morris considers a wide variety of cases, including the Roman Empire, the European state during its Renaissance rise and the contemporary United States. In each case there is good evidence that the desire to prepare for war spurred technological invention and also brought a higher degree of internal social order.


    Cowen points to research which suggests the prevalence of the Occident over the Orient stems from a kind of militaristic “keeping up with the Joness” driven by the Alpha-males striving for status-dominance jockeying for postion at the helm of nationalistic ships of state:

    a recent working paper by the economists Chiu Yu Ko, Mark Koyama and Tuan-Hwee Sng…argues that Europe evolved as more politically fragmented than China because China’s risk of conquest from its western flank led it toward political centralization for purposes of defense…The European countries invested more in technology and modernization, precisely because they were afraid of being taken over by their nearby rivals.


    Its no accident that the German state was both the world’s leading sci-tech Researcher and Developer and the world’s leading military power. The combination of Prussian militarism and Protestant industrialism developed the world’s first M-I complex. Modern higher education was launched by Prussia in the 19th C, combining the Humboldtian Model of pure research with the polytechnic’s pursuit of applied development. But the German political elite jumped the gun, focusing more on war-waging more than war-preparing and in so doing gave the M-I complex a bad name.

    By contrast the US managed to place constitutional controls on elite militarism and channel scientific research into dual use technology, Ruttan points out that industrial standardisation grew out of the Yankee Civil War effort:

    In the United States, what came to be termed the American system of manufacturing emerged from the New England armory system of gun manufacture.


    Oddly, given his much quoted valediction, Ike was a poster child for the M-I complex. He appointed the von Neumann, the worlds best scientist and an ardent supporter of the military, as his sci-tech advisor. He then amped up the military’s role in sci/tech leading to a staggering double barrelled success with NASA winning the Space Race and DARPA establishing the Internet.

    Since the end of the Cold War the rate of (non-IT) innovation has declined, a fact lamented by ntellectuals as disparate as Peter Thiel, Robert Gordon and Vernon Ruttan. The main reason is that capitalist economics – such as privatisation, deficit reduction, IP policing and commercialisation – have trumped nationalist politics. Ruttan (Is War Necessary for Economic Growth 2006) is pessimistic that the end of Cold War-inspired R & D has taken the edge off sci-tech progress:

    A major objective in this paper is to demonstrate that military and defense related research, development and procurement have been major sources of technology development across a broad spectrum of industries that account for an important share of United States industrial production. I argue that the United States and the global technological landscape would be vastly different in the absence of the contribution of military and defense related research, development and procurement. I also argue that as we look to the future the contribution of defense and defense related technology research, development and procurement to United States industrial production will be smaller than in the last half century.
    An implication is that in the future the rate of productivity and income growth in the United States economy will be slower than during the first two post-World War I decades or than during the information technology bubble that began in the early 1990s.

    The one outstanding exception to this disappointing trend has been the spectacular development of AI. In 2005 “AI winter” effectively ended when Stanford won the robot desert navigation challenge. But this exception proves the M-I complex rule because the challenge was sponsored by…DARPA.

  20. @alex

    Since you raise the point, I’ll just mention that lots of development of new weapons technology has taken place in peacetime. The hydrogen bomb was a product of peacetime development; so, in earlier ages, were the dreadnought and the Maxim gun. I’m not clear on what larger significance the point has.

  21. Because Quiggin decided to be a jerk and close comments for the opportunities cost column about war, I ll make a comment about that column inasmuch as there s nowhere else to do so. And if Quiggin doesn t like that, he can learn to like it. A recent study by the Policy Economic Research Institute finds that military spennding is the worst way to create jobs in the U.S. Military spending produces half as many jobs per dollar as spending on education.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s