John Forge

Just wars are supposed to be proportional responses to aggression: the costs of war must not greatly exceed the benefits. This proportionality principle raises a corresponding ‘interpretation problem’: what are the costs and benefits of war, how are they to be determined, and a ‘measurement problem’: how are costs and benefits to be balanced? And it raises a problem about scope: how far into the future do the states of affairs to be measured stretch? It is argued here that weapons innovation always introduces costs, and that these costs cannot be determined in advance of going to war. Three examples, the atomic bomb, the AK-47 and the ancient Greek catapult, are given as examples. It is therefore argued that the proportionality principle is inapplicable prospectively. Some replies to the argument are discussed and rejected. Some more general defences of the proportionality principle are considered and also rejected. Finally, the significance of the argument for Just War Theory as a whole is discussed.

This paper is about structuralism as a form of reconstructing theories, associated with the work Sneed, Balzar and Moulines among others, and not about "structuralism" is any of its other manifold senses. The paper is a reflection in that it looks back on some earlier work of my own on the subject of structuralism and explanation, in which I argued that structuralism and my 'instance view' of explanation go well together, with structuralism providing the means to develop the idea of a theoretical instance. Bartelborth has suggested a view that has some similarity with my early ideas, so I reflect on those as well. I suggest, in opposition to both positions, that a causal account of explanation might also sit well with structuralism. This paper will appear in a special edition of Synthese edited by Moulines and devoted to structuralism themes

In his new book What is a Law of Nature?, David Armstrong gives an account of functional laws on the basis of the theory, originally proposed independently by Armstrong himself, Dretske, and Tooley, and further developed in this work, which asserts that laws are relations of necessitation between properties. On the theory, properties and relations are universals, and so a law is a relation between universals and is itself a universal. There are two reasons why Armstrong's account of functional laws is worth discussing. First, any theory that purports to be about laws that are investigated by science–I assume that we are only interested in such theories–must cover functional laws and not merely be adequate for generalizations about ravens or grue. Second, Armstrong claims that the main rival to his theory, the regularity theory of laws, cannot give a satisfactory account of functional laws. If this is true, and if Armstrong's theory can provide a satisfactory account of functional laws, then this may incline us towards his point of view.

While there has been much interest in this topic, no generally accepted definition of dual use has been forthcoming. As a contribution to this issue, it is maintained that three related kinds of things comprise the category of dual use: research, technologies and artefacts. In regard to all three kinds, difficulties are identified in making clear distinctions between those that are and are not dual use. It is suggested that our classification should take account of actual capacities and willingness to make use of these objects for ‘bad ends’ and not the mere possibility that this could be done, and here three ‘contextual factors’ are identified. A (provisional) definition is proposed that takes account of threats and risks.

I ask whether weapons research is ever justified. Weapons research is identified as the business of the engineer. It is argued that the engineer has responsibility for the uses to which the tools that he designs can be put, and that responsibility extends to the use of weapons. It is maintained that there are no inherently defensive weapons, and hence there is no such thing as ‘defensive’ weapons research. The issue then is what responsibilities as a professional the engineer has in regard to such research. An account is given to ground the injunction not to provide the means to harm as a duty for the engineers. This account is not, however, absolutist, and as such it allows justifiable exceptions. The answer to my question is thus not that weapons research is never justified but there must be a strong assurance that the results will only be used as a just means in a just cause.

The fact that corporate responsibility supervenes on human action implies that there are two possible kinds of account of the former, namely reductive accounts in which the responsibility of the corporation devolves down without remainder to its officers, and those in which it does not. Two versions of the latter are discussed here. The first, due to Peter French, tries to satisfy the supervenience requirement by defining corporate action in terms of human action. It is argued that the corresponding view of intention, intentions as plans, does not serve to show how the defined notion of corporate action also brings with it attributions of responsibility. An alternative account, taking its point of departure from Feinberg’s ideas of vicarious and collective responsibility, is therefore proposed. It is argued that when officers of a corporation substitute the “decision-making mechanism” of the corporation for their own, then responsibility, but not action, can transfer to the corporation. Furthermore, it is argued that this nonreductivist account can be defended against the reductivist charge that attributions of moral responsibility to corporations is a category mistake.

An account of physical explanation derived from the instance view of scientific explanation is outlined, and it is shown that this account does not cover explanations by theories which contain theoretical functions. An alternative account, also derived from the instance view, is proposed on the basis of Sneed's account of theories. It is shown that this account does cover theoretical explanations. Finally, it is shown that this account can accommodate explananda that record errors of measurement.

Abstract This paper argues that there are good reasons to adopt a non-reductive account of states when it comes to quantum mechanics. That is to say, it is argued that there are advantages to thinking about states as sui generis, as reducible to classes of values of quantities, when it comes to the quantum domain. One reason for holding this view is that it seems to improve the prospects for explanation. In more detail, it is argued that there is an 'explanatory shortfall' in the quantum domain owing to the failure of value definiteness. To remedy this situation, two proposals are put forward about the nature of the quantum state: Proposal A, that the quantum state is the only first-order property of quantum systems, Proposal B, that the quantum state is one first-order property among many. These proposals seem equally good

Some features of physical science relevant for a discussion of physical explanation are mentioned. The D-N account of physical explanation is discussed, and it is seen to restrict the scope of explanation in physical science because it imposes the requirement that the explanandum must be deducible from the explanans. Analysis shows that an alternative view of scientific explanation, called the instance view, allows a wider range of physical explanations. The view is seen to be free from a certain class of counter examples to the D-N theory

This is a challenging book and it addresses important questions. This review has focused on what I think is the most important question of all: just what is the relationship between the ‘strategies’ which drive modern science and the social values which guide the societies we live in. I have much sympathy with the way in which Lacey tries to answer this question and how he tries to open up alternative possibilities and give us a view of the future which is not just a continuing catalogue of environmental disaster ‘for the relief of man's estate’. However, I have expressed doubts as to whether he has been entirely successful in this regard, especially concerning the alternatives to the materialist strategies

Glymour's account of confirmation is seen to have paradoxical consequences when applied to the confirmation of theories containing theoretical functions. An alternative conception of instances derived from Sneed's reconstruction of physical theories is conjoined with the instance view of confirmation to produce an account of confirmation that avoids these problems. The topic of selective confirmation is discussed, and it is argued that theories containing theoretical functions are not selectively confirmable

The paper tries to locate the moral limits of weapons research, an issue that comes about because weapons harm and unjustified harms are wrong.Doing research does not itself harm, so first it is shown that a means principle holds. Weapons research then needs to be justified, and two ways to do this arecanvassed, historical and a historical. The former takes account of the context in which the work is done and the circumstances the products used. It is arguedthat there can only be historical justifications, given that there are no inherently defensive, deterrent or humane weapons. However, weapons designs live onbeyond the circumstances in which they were created, and even if these amount to ‘just war’ there can be an assurance that the products will not be used unjustlyin the future. A radical solution is suggested for this problem.

The purpose of this paper is to provide an analysis of the concept of model as it is applied in the physical sciences, and to show that this analysis is fruitful insofar as it can be used as an acceptable account of the role of models in physical explanation.A realist interpretation of theories is adopted as a point of departure. A distinction between theories and models is drawn on the basis of this interpretation. The relation between model and prototype is expressed in terms of the concepts of access and accessibility, and four conditions are proposed as an analysis of the concept of model. It is concluded that models are introduced when approximate methods are used.

By using the concept of a uniformity, the Structuralists have given us a most useful means of representing approximations. In the second section of this paper, I have made use of this technique to show how we can deal with errors of measurement — imprecise explananda — in the context of theoretical explanation. As well as (I hope) providing further demonstration of the power of the Structuralist approach, this also serves to support the ontic conception of explanation by showing that it can help us resolve substantial problems in the theory of explanation

ConclusionI think it is clear that Bqm and Oqm are quite different theories, even if they have the same empirical consequences. This is, of course, to adopt something like the realist perspective. Oqm is not normally interpreted realistically by physicists (the survey still has not been done!) but it can be, and what it says things are like is by no means the same as what Bqm says. One of the most puzzling features of Oqm is the status of the particle when it is in an eigenstate of momentum: if it has no precise position how can it be anywhere at all, and if it isn’t anywhere how can it be said to exist? The great advantage of Bqm is that it restores definite trajectories.The books under review, on which I have based this exposition, are all to be recommended. I am particularly impressed with Peter Holland’s book and assure the reader that it will repay study. The collection by Cushinget al. is, on the whole, better than most in that it is consistently focused on the advertised theme, but it is mostly technical. Cushing’s book is the most accessible of the three volumes.