If it were not for weapons research, there would be no predator drones or smart bombs or improvised explosive devices or assault rifles. The insurgents in the Middle East and elsewhere would have no means to fight, and there would be no wars, large or small. Even more importantly, there would be no vast arsenals of thermonuclear weapons capable of ending much of the sentient life on the planet. The world would then most certainly be a safer place. But weapons research is not something new: the gunpowder weaponry of the early modern period was the product of research, as were the torsion catapults in Greece at the time of Philip and Alexander of Macedon. Whatever else is true about weapons research, it is clear that it introduces new (or improved) means of killing and destruction, and this is sufficient to define the activity. [1] This would appear to be a very weighty matter, something that one might imagine philosophers, and others who think about such things, would have had a lot to say; surprisingly, not much at all has been written on the subject, though some explanation of this neglect will be given in this chapter.


The main issue for ethics and weapons research centres on the ethical or moral evaluation of the activity: Is it ever morally justified to design the means to kill, harm and destroy, and if so, under precisely what circumstances? Turning to science and its relation to weapons research, the question here is the role that science plays in weapons research. Perhaps weapons research is a wholly (applied) scientific endeavour or perhaps science is a part of weapons research? Bringing ethics back in, if weapons research is deemed morally wrong, then is it the case that whatever role science plays is also wrong? To answer these questions, three examples will be given which will help to clarify the roles that science can play in weapons research. If weapons research itself is understood as applied science, as it is by Arrigo for instance(Arrigo 2000: 303), then one might expect this to entail the application of theory to the design for new weapons, for true or radical innovation. But there are other ways in which science can inform weapons research, as will be seen presently. Before moving on to these examples, it is worth making some general, and very brief, comments about ethics and the way it can apply to an intellectual activity such as science. This is worth doing because it cannot be assumed that the audience for the present topic is familiar with philosophy or ethics, but it is necessary to have a framework.[2] 




A straightforward way to describe ethics is to say that it is a study which deals with what persons ought and ought not to do. It is thus to do with the choices, actions and behaviour of mature competent people. Some of the things that people do do not affect others, other humans, other sentient beings, in any significant way and hence these do not come under the purview of ethics. Those actions that do affect others are, however, open to moral or ethical evaluation: are they right or are they wrong? To resolve that question, one needs to appeal to a moral system. All such systems forbid certain actions, namely those that inflict unjustified harm on others. This is surely intuitive and obvious: no one wants to be harmed. It is almost by definition that no sentient being wants to feel pain - assuming that the pain does not indicate that some medical treatment is working or some suchand to be in pain is one form of being harmed.


Some moral systems require people not only to refrain from harming others but also to provide some positive benefit for them. Jeremy Bentham and John Stuart Mill, the nineteenth century English philosophers, famously believed that one ought to strive to increase the amount of happiness in the world. However, morality is supposed to be impartial in the sense that it forbids discrimination in regard to moral action. Prohibitions on harming do not end with family or friends or community or country: nobody should be harmed, no one at all. Some critics of the style of morality advocated by Bentham and Mill have pointed out that it is impossible to increase the amount of happiness in the world impartially: no one can make everyone happy! Just how serious this objection is is a matter of ongoing debate. But it is only necessary here to note that this kind of moral system shares the prohibition on harming with the former kind: for the topic at hand, it is clear that the moral evaluation of weapons research, whatever else it might involve, will not be such as to see it as an activity which aims to increase the amount of happiness in the world.


Most philosophers do not believe that the dictates of morality are absolute and cannot be broken in any circumstances. For example, most accept that a moral rule such as “Do not cause pain” has justified exceptions. Clearly, a dentist who inflicts pain on her patient to save his teeth has not done something morally wrongprovided that the patient understands and assents to the treatment. Also, it is generally agreed that it is permissible to cause pain in self-defence, if that is the only way to defend oneself. This leads to the view that justifiable exceptions to the overall moral prohibition against harming will be such as to show that the harm inflicted will prevent other harms. Just how this is worked out will vary from case to case, and it is here that much of the hard work in ethical reflection and evaluation takes place. One might think as a basic principle that the harms prevented should be at least as much or many or as great as the harms caused if there is to be justification, and that therefore it must be necessary to be able to make some informed assessment of what these might be. This brings the discussion back to weapons research, an activity that aims to provide the means to harm. It is now necessary to look at some examples.


1. It is well-known that both kinds of nuclear weapon, fission and fusion weapons, were the direct results of the application of scientific theory to design. Without advances in nuclear physics in the 1930s and 1940s, the very idea of a nuclear weapon would not have been dreamed up. Rhodes gives an excellent account of the genesis of the idea of a fission weapon, from the speculations of Szilard from 1933 to his work with Fermi in the US, and thememorandumwritten by the émigré scientists Frisch and Peierls in England which made predictions about crucial nuclear parameters. The designs of the two kinds of fission bombs made during the Second World War were determined by the properties of the fissile materials used, matters that were uncovered by painstaking research.[3] The idea for a thermonuclear weapon emerged from the atomic bomb project, and was also driven by science and by scientists. Indeed, the fusion reactions that power these weapons must be initiated byfission triggers, so not only did the research into fission or atomic bombs lead to thermonuclear weapons, the products of the former were integral elements of the latter. [4]


The research leading to nuclear weapons was exceptional in that scientists, including all of those mentioned above, took the initiative in agitating for programmes to investigate the possibility of using the recently discovered fissile materials for a weapon of terrible destruction, to the subsequent regret of some of them. The military is now well-aware of the power of science and has set up agencies to oversee the application of scientific theory to the design of new weapons systems. The best-known is the USs Defense Advanced Research Projects Agency, or DARPA; notice that, in common with armed forces the world over, the word “defense” appears in its name. There are equivalents in Russia, China, the UK, France, Israel and other countries.


2. In the nineteenth century the weapons that caused the majority of fatalities and injuries in war, artillery and the infantrymans longarm, were transformed by the scientific analysis of existing armaments. Smoothbore cannon and muskets were transformed into steel, breech-loading guns whose recoil was damped and rifles firing clips of cartridges using smokeless powder.[5] These are instances of science improving existing designs. One example will be given as an illustration.


All smoothbore weapons which fire balls, musket and cannon balls, are inherently inaccurate because of uncontrollable forces which act on balls as they spin in the air after leaving a gun barrel. This is called the Magnus Effect or Magnus Force, after the German physicist Heinrich Magnus who described it in 1852. It was also investigated by the English physicist Benjamin Robbins, in his book The New Principles of Gunnery, published a century before Magnuswork (Steele 1994: 359). The effect is this: unless a ball only spins about the axis in the direction it is travelling, what is called the roll axis in aeronautics, it will experience unequal pressure on either side in the direction of its motion, and hence be deflected in the direction of the axis of spin. In other words, it will not travel straight, and if it actually hits (anywhere near) its target, then this is a matter of luck.


This analysis showed that if only the bullet could be made to spin about the roll axis, accuracy would be greatly increased. The first attempts to do so involvedriflingmusket barrels, fitting them with raisedlandsspiralling down the inside of the barrel which caused the ball to spin about the required axis. There were two problems: the rifled musket was very slow to load because the ball had to be hammered down the barrel, and the barrel became fouled after only a few shots because the powder residue collected around the lands. These problems were solved by three innovations. There were, in the first place, advances in metal working that allowed relatively efficient breech-loading riflesleakage of gas from around the breech was eventually dealt with by gunsmiths such as Paul Mauser. Then there was progressive improvement in propellant until eventually smoke-less powder became available. Finally, bullets were designed that would not stick in the barrel on the lands. The idea was that the bullet would have two parts; a base that was made out of a relatively soft material that would deform in the lands and impart a twist to a hard core made narrower than the base so that it would not stick in the barrel. After two attempts in England in the 1820s, eventually a French innovation by Captain Claude-Étienne Minié of the French Chasseurs was successful in 1847. [6] This led to many orders of magnitudes increase in lethality: in the American Civil War over half a million causalities, including two hundred thousand killed, were caused byMinié rifles.


3. The application of scientific theory, either to produce quite new weapon designs or to improve existing ones, is a relatively new phenomena, beginning with the study of  ballistics in the seventeenth century (see Hall 2009 ). However, systematic research into weapons design took place at least as far back as the fourth century BCE. Remarkable work was done in that century in relation to the design of the torsion catapult, which marked the beginning of effective siege artillery. A torsion catapult works because of the restoring force exerted byspringymaterial, like rope, which has been stretched. Wrapping rope around a stout piece of wood fixed to a frame, embedding a handle in the mass of rope and then pulling the handle back will create a torque: the handle will fly back to its resting position when released. It can be therefore seen how, in principle, such a device can be used to cast stones when placed in a basket attached to the handle.


It is not known how the Greek engineers discovered the torsion principle and applied it to designing catapults, but this must have been done empirically as the theory was not formulated until much later. However, the design of torsion catapults was codified in treatises which gave general instructions how to build them. These instructions were general because formulae, known as the hole formulae, gave the key parameter for casting stones of a given weight or arrows of a given length. The Greek engineers had discovered formulae that related the optimal size of the hole in the catapult frame in which the wood about which the rope or springs was to be wound to the weight of the stone or length of the arrow. The former is more complex and involves a cube root, which gives an equation that Greek mathematics at the time could not solve exactly. Perhaps still more remarkably, the treatises gave approximation methods for solving the equations in question.[7]


Future Directions


If weapons research is morally wrong, then it would appear that the actions of any scientist who contributes to weapons research are also morally wrong. But the matter is not quite so straightforward. A scientist who knowingly, willingly and intentionally engages in weapons research is as open to moral censure as any other weapons researcher, assuming that weapons research is a proscribed activity. It is not so clear, however, that a scientist who does not intend that her work be used to further the ends of weapons research, but sees that it may well in fact do so is also blameworthy. It is less clear still that a scientist who does not foresee this outcome, assuming it comes about, is to blame. It would not be necessary to address any of these issues if weapons research was not morally wrong, so this matter must now be discussed.


It was mentioned above that the topic has received much less attention than it deserves. Forge, who has addressed the topic at some length and argued that weapons research is morally wrong under all circumstances, believes that the question of the morality of weapons research has been assimilated to questions about the justness of war, about which a great deal has been written of late, and he believes this is a mistake (Forge 2013: ix)). The consensus about war is that war is just or morally permissibleno blame attachesif it resists aggression. This is themodern viewwhich sees just war as more than self-defence, other defence being allowed if the other is the victim of aggression. The implicit corollary is that measures such as weapons research are justified if they aid in the prosecution of a just war or in the pursuit of self-defence or deterrence. [8] This is the standard rationale or standard justification of weapons research. Forge denies that this justification is convincing. However, Kemp, one of the few who have discussed the ethics of weapons research, claims that it is not only permissible to undertake weapons research for defence, and maintains that it is a civic duty imposed on scientists to do so (Kemp 1994). Forge and Kemp therefore adopt diametrically opposed viewpoints, and the topic can be explored by comparing and assessing their respective positions.


Kemp claims that since the state has a duty to protect its citizens, and since Hobbes many political philosophers believe that this is the raison detre of the state, then those who are able to assist in this vital endeavour by designing the means for defence have a duty to do so. Note that a civic duty in this sense is not a moral duty, as it only applies to a part of the moral community, namely ones fellow citizens, and hence lacks the (completely) impartial character of morality. Expressed in these terms, Forges view is that it is moral duty not to undertake weapons research. Forge does not deny that states have a right to defend themselves, or defend others from aggression, nor does he deny that weapons research can help to achieve these ends. What he does deny is that this is all that weapons research does, namely provide the means to resist aggression, whatever the intend behind it. Note also that Kemps position is compatible with weapons research being permissible while Forges is not. Which is correct? Kemps view seems to be convincing, but when weapons research is analysed in more depth, this initial reaction is seen to be wrong.


Hacking maintains that what weapons research produces in the first instance is knowledge, instructions for making new or improved weapons, not hardware (Hacking 1986).[9] Weapons research is not the same thing as weapons manufacture. A scientist who successfully discharges her civic duty does not therefore provide a unique array of defensive weapons for her country, but a design, and any institution with the requisite skills and resources could produce identical weapons if it possesses the design. For instance, the Kalashnikov, the AK-47, the most widely made weapon of all time, has been manufactured from 1947 by the Soviet Union, Russia, Poland, Hungary, East Germany, Bulgaria, Romania, China, North Korea, Yugoslavia, Iraq, Egypt and India, and variants by Finland, Israel and South Africa.[10] Mikhail Kalashnikov invented a highly effective assault rifle, but should he be praised for discharging his civic duty or condemned for introducing into the world a terrible new means to kill? Clearly, his innovation spread beyond his own country. Successful weapons tend to become ubiquitous. Both sides in the American Civil War used ‘Minié rifles, as did both sides in the Franco-Prussian War of 1871. By the First World War, all protagonists had machine guns, quick-firing artillery as well as reliable rifles. There was less uniformity in the armaments of the Second World War, though all sides had tanks, bombers, fighters, field artillery and so forth, and there is still less uniformity today. This trend is due to the vastly increased pace of weapons research and it is to be expected that states will strive to copy successful weapons innovations as much as they can. This may not be easy and it may not always be clear how it can be done, but the very fact that many different countries come up with weapons that are remarkable similar shows that it happens. Weapons innovation cannot be contained.


If successful weapons innovation does not lead to a unique array of weapons, perhaps this does not matter if leads to weapons that can only be used to defend against aggressionperhaps Kemps position can be resurrected along the lines that designing defensive weapons is a civic duty? The role of a defensive weapon, or weapon that is postured defensively, is to actively protect an asset, but this does not necessarily amount to defending against aggression. The work of Luttwak on the levels of strategy shows the relationship between defence and its strictoppositeoffence is complex and multi-faceted.[11] One implication of Luttwaks analysis is that a state that would embark on a war of aggression needs a range of defensive options. It needs to be able to defend itself, itshomeland, from retaliation, it needs to defend captured territory and its forces in enemy territory from counter-attack, etc. Thus, if the overall theatre or grand strategy is one of aggression, with the long-term goal of capturing territory or otherwise exerting hegemony, atlowerlevels defensive tactics and even defensive operations may be necessary at certain times and places, and conversely. Fighting on the Eastern Front in the Second World War, after Hitler invaded the Soviet Union, exemplified this dynamic. At the level of grand strategy, Hitler was waging an aggressive war of conquest, while Stalin was defending. But even as early as the third month of the war, the Germans were mostly on the defensive, with their infantry holding positions while their tanks, the main instrument of the blitzkrieg operations, were being repaired and refitted.[12] Thereafter, the German offensive re-commenced, stalling finally in November 1941 in front of Moscow, and the next two years the protagonists defended and attacked one another in turn until the Soviet Union began to prevail. All the weapons both sides had at their disposal were used in all forms of fighting, including tanks that were deployed in mobile defence: there were no weapons only used for defence. Kemps position therefore cannot be salvaged.


Forge maintains that since weapons are, by their very nature, the means to harm, then research which aims to introduce new or improved weapons into the world requires justification, and that the only possible justification is that the weapon in question will prevent more harm than would be the case were it not available.[13] This kind of justification lies behind all references to defence in relation to weapons research, military spending and so forth. But this demand cannot be satisfied: weapons research produces knowledge in the form of designs which can be implemented in the future in circumstances that are unknowable – as the examples discussed above have shown - and since there is no such thing as a weapon that cannot be used to aid aggression, there can be no assurance that an episode of weapons research will produce a weapon that prevents more harm than it causes. It is precisely this assurance that is needed if weapons research is to be justified, hence Forge argues that weapons research is always unjustified. The historical record shows that wars have become more frequent and more deadly since scientific advances began to be applied systematically to weapons research, and so the facts seem to support Forges position. But these important issues surely need more discussion, not just from philosophers and others who reflect on weapons research, but from the scientists and other experts who actually take part.




Returning to focus again explicitly to the role of scientists, if weapons research is morally wrong and without justification, then scientists, just like everyone else, should not take part. A scientist who knowingly and deliberately takes part in an activity that is morally wrong is as blameworthy as any other participant. However, there are different ways in which scientists can be said to participate. For instance, a scientist can receive a grant from the military even though she does not intend to undertake weapons research and does not see how her work might provide a basis for weapons research.  But, the military only sponsors research it thinks might be useful for its own ends, including weapons research, and this is a good reason why scientist should not accept such funding. It is also possible that a scientist knows she is conducting research that could have military application without being sponsored by the military. This happened in the late 1930s when some members at least among the scientific community believed that current advances in nuclear physics could contribute to a weapon of mass destruction. The majority of scientists working at the time, including Hahn and Strassman, the discoverers of nuclear fission, saw themselves aspureresearch scientists. Szilard sought a moratorium on publishing papers on nuclear physics in order the try to prevent such nuclear weapons research, with limited success (Rose 1998; 92). This raised complex questions about the responsibility of scientists for applications of their work that they do not intend, or even foresee, but these are questions beyond the scope of this chapter.




Acheson, D (1969) Present at the Creation. New York: Norton.

Arrigo, J. (2000) “The Ethics of Weapons Research: A Framework for Discourse between Insiders and Outsiders” Journal of Power and Ethics, 1.

Forge, J. (2007) “No Consolation for Kalashnikov” Philosophy Now, 59 (Jan.).

Forge, J. (2008) The Responsible Scientist. Pittsburgh: Pittsburgh Univ. Press.

Forge, J. (2013) Designed to Kill: The Case Against Weapons Research. Dordrecht: Springer.

Fotion, N. (1990) Military ethics. Hoover Institution Press: Stanford, California.

Grandy, D. (1996) Leo Szilard. Lantham,. Maryland: Univ. Press of America, 1996.

Greener, W. (1910) The Gun and Its Development. London: Cassell.

Grice, M. (2009) On Gunnery. Charleston, South Carolina: Booksurge Publishing.

Hacking, I. (1986) “Weapons Research and the Form of Scientific Knowledge” Canadian Journal Of Philosophy, Supplementary Volume 1,2.

Hall, A (2009) Ballistics in the Seventeenth Century. Cambridge: Cambridge Univ. Press.

Hall, B. (1997) Weapons and Warfare in Renaissance Europe. Baltimore: Johns Hopkins Univ. Press.

Hoddeson, L., P. Henriksen, R. Meade and C. Westfall (1993) Critical Assembly. Cambridge: Cambridge Univ. Press.

Kanaher, L. (2007) AK-47. New Jersey: Wiley.

Kalashnikov M. (2006) The Gun that Changed the World. Cambridge: Polity Press.

Kemp, K. (1994) “Conducting Scientific Research for the Military as a Civic Duty” In Erwin et al. (eds) Ethical Issues in Scientific Research. New York: Garland.

Liddell Hart, B. (1973) History of the Second World War. London: Pan.

Luttwak, E. (1987) Strategy: The Logic of War and Peace. Cambridge, Mass.: Harvard Univ. Press

Manchester, W. (2003) The Arms of Krupp. Boston: Back Bay Books.

Marsden, E. (1969) Greek and Roman Artillery: Historical Development. Oxford: Oxford Univ. Press.

Marsden, E. (1971) Greek and Roman Artillery: Technical Treatises. Oxford: Oxford Univ. Press

Rhodes, R (1986) The Making of the Atomic Bomb. Harmonsworth: Penguin.

Rhodes, R. (1995) Dark Sun: The Making of the Hydrogen Bomb. New York: Touchstone.

Rihill, T. (2007) The Catapult. Yardley, Pennsylvania Westholme.

Rose, P. (1998) Heisenberg. Berkeley: California Univ, Press.

Stankiewicz, R. (2000) “The Concept of Design Space” in Ziman, J. (ed.) Technological Innovation as an Evolutionary Process. Cambridge: Cambridge Univ, Press.

Stahel, D. (2009) Operation Barbarossa and Germanys Defeat in the East. Cambridge: Cambridge Univ. Press.

Steele, B. (1994) “Muskets and Pendulums: Benjamin Robins, Leonhard Euler and the Ballistics Revolution” in Technology and Culture, 35.

Walter, J. (1979) The German Rifle. London: Arms and Armour Press.

Warren, K (1989) Armstrongs of Elswick. London: MacMillan.

Weart, S. and G. Szilard (1978) Leo Szilard: His Version of the Facts, Vol. 2. Cambridge Mass: MIT Press.

Zabicki, D. ( 2006 ) “The French 75” in S. Tucker and P. Roberts (eds) World War 1. Santa Barbara: ABC-CLIO.


Further Reading


MacNeill gives an excellent account of the relations between weapons and the pursuit of power from antiquity to 1945 see W. McNeill The Pursuit of Power. Chicago: Chicago Univ. Press, 1982. An introduction to the view of ethics favoured by the author of this chapter see B. Gert Common Morality. Oxford: Oxford Univ. Press, 2004; for a more advanced treatment see B. Gert Morality: Its Nature and Justification. Revised Edition. Oxford: Oxford Univ. Press, 2005. The ‘classic’ statement of modern Just War Theory is M. Walzer Just and Unjust Wars. London: Allen Lane 1977; much more readable, though along the same lines, is B. Orend The Morality of War. Peterborough, Ontario: Broadview 2005. For more on the present author’s work on weapons research, see for example "The Morality of Weapons Research", Science and Engineering Ethics, 10, July 2004, “What are the Moral Limits of Weapons Research?”, Philosophy in the Contemporary World, 14, Spring, 2007, "Proportionality, Just War Theory and Weapons Innovation", Science and Engineering Ethics, 15, March 2009 “The Morality of Weapons Research” Wiley-Blackwell International Encyclopaedia of Ethics, 2011 and “The Case Against Weapons Research”. International Journal of Technoethics. 5, 2014.

[1] Forge has given the following definition: Weapons research is research carried out with the intention of designing new weaponns, or improving the design of existing weapons, or designing or improving the means for carrying out activities associated with the use of weapons, Forge, 2013: 14.

[2]  The writer’s own webpage,, contains some introductory material, together with more on the topic of this chapter, and also more references.

[3]  See especially Rhodes 1986; for his discussion of Szilard, see 28-30, and for the Frisch-Peirels Memorandum see 321-325. Weart edited Szilard’s papers, which were published in 1978, see Weart and Szilard 1978. Grandy has a more objective account of Szilard’s contribution. Forge 2008 Chapter 8 is a short accessible overview of the Manhattan Project and the events leading up to it. For a more technical discussion, see Hodderon et al. 1993.

[4]  Rhodes’ second book, Rhodes 1995, takes up where his first book ends, at the conclusion of the Second World War, and begins with a discussion of the Soviet atomic bomb project and moves on to the decision by the US to then proceed with thermonuclear weapons. Both books are important because they give background to the decisions to actually go ahead with the respective nuclear weapons research programmes, surely among the most momentous decisions of all time. For more on the decision to proceed with thermonuclear weapons research by one who was actually involved, see Acheson 1969, especially 344-349.

[5]  It was necessary to develop special steels for guns that used the new high explosives, both of which were due to advances in chemistry in the nineteenth century, see Manchester 2003. Manchester’s book gives an excellent account of Krupps, the famous, or notorious, weapons manufacturer and innovator. Warren 1989 tells the story of George Armstrong, the English equivalent of Alfred Krupp. The problem of recoil, which had plagued gunners from the very beginnings of artillery, was solved at the end of the century by the French, see Zabecki 2004.

[6]  See Greener 1910: 629-632 for the two British attempts. An excellent account of Minié is in Hall 1997: 135-147. Walter 1979 is a comprehensive account of the nineteenth century German rifle innovation, especially in regard to the Mauser.

[7]  Marsden wrote two important books on this topic, which have become classics. One gives the historical development of Greek and Roman artillery, while the second reproduces four technical treatises, from the fourth century BCE to the first century CE, which give instructions for building catapults. Rihill is a more up to date, and perhaps more readable, general text on the topic, which does not always agree with Marsden.

[8] Some Just War Theorists, and some others, have argued that certain classes of weapons, especially weapons of mass destruction, are immoral but have not (even) gone on to state the corollary to that proposition that the weapons research that aims to produce weapons of mass destruction is immoral. Fotion, Fotion 1990, gives one of the best discussion of the issue from this resticted viewpoint.

[9] This knowledge, in the form of designs, can be more or less well articulated, from the most basic and sketchy idea – for instance, using a chain reaction in a uranium assembly – to an engineering specification that could,  in principle, be implemented entirely by machine. For more on this issue, see Stankiewicz 2000.

[10]  For Kalashnikov’s story in his own words, which must be taken with a grain of salt, see Kalashnikov 2006. For more details, and more objectivity, see Kanaher 2007.

[11]  Luttwak 1987 is the main source. But see also Liddell Hart 1973 for various ideas about what could count as a defensive weapon.

[12]  See Stahel 200, especially Chapter 7 and the introduction.

[13]  Forge’s main work is Forge 2013, which contains many further references.