Such an overview necessarily demands an analysis of military production and thus of the more general problem of non-productive labour.
The panorama of crises presented by capitalism during the period 1914-46 requires no commentary: two world wars and a depression on the scale of 1929, in less than thirty years, speak eloquently enough. Two remarks may, nonetheless, be of use. Because it was an economic crisis in the most traditional meaning of the term, the depression of 1929-34 is often cited as the crisis of modern capitalism. But the two world wars also constituted crises of the system in no way less profound. In the same way as the economic depression of the inter-war period, the world conflicts expressed in the most brutal fashion the incapacity of capitalist relations of production to continue to reproduce themselves in the normal way. 'Pure' economic crises eliminate excess capital through bankruptcy; war achieves the same end through physical destruction and armed violence. But in both cases, the content is the same: capital is forced to destroy itself under the pressure of the contradictions of its own mode of functioning. In both cases, the system gives rise to violent convulsions provoked by the definitive inability of the dominant relations of production to adapt to the needs and potentialities of society. In a world dominated by capital, there is nothing which is not economic: these 'international political crises' called world wars, far from being extra-economic phenomena, have only been the barbarous manifestation of the most profound economic crises.
Unlike the crises of the ascendant phase of capitalism, the convulsions of decadent capitalism bring to the fore a general and accentuated movement of systematic deterioration. In this very fact lies the proof of their different nature. The crisis of 1929 brought about a fall in production greater than that caused by World War I. World War 2 was far more destructive than World War I, and was a disaster incomparably greater than that of 1929. This progression in intensity indicates the irreversible character of a general movement of decline.
From 1914 until the end of World War 2 the importance and meaning of these convulsions became more and more obvious. But the absence of serious crises during the twenty years following World War 2 was enough to convince commentators of the definitive disappearance of crises within capitalism, and consequently to relegate the crises of the preceding period to the never-never land of "crises of growth".
And yet, in the period immediately after World War 2, capitalism began a new cycle on foundations which were no more sound than those which had precipitated the war. Economic expansion depended more than ever on two crutches which had just demonstrated their inherent weaknesses in terms of any long-term solution to the crisis. These crutches were the mechanisms of "reconstruction" and production for military ends.
Of course, the destruction was far more extensive this time, and the reconstruction would last longer than it did after World War I. War became permanent through constantly maintained local conflicts, and the production of war material achieved higher levels in "peace time" than during the two world conflagrations. The state learned better how to handle the effects of internal contradictions which progressively undermine the system, and tended to emerge more and more as the system's controlling force. In short, capital had begun a new phase of its decline with an array of particularly effective palliatives, but the purely palliative nature of these measures has never been concealed for a moment.
The limits of reconstruction
This point has been dealt with at length in the article "La crise", which appeared in Révolution Internationale (old series) nos. 6 and 7, and we will not repeat it here. Rather we will content ourselves with a review of some of the principal economic phenomena which, since the second half of the 60s, unequivocally demonstrate the limits of these palliatives.
The completion of the reconstruction of the European countries, and of Japan, around 1965 gave rise to profound dislocation in the international economic channels which had presided over capitalist "expansion" since the war. All the international exchange mechanisms entered into crisis. The former outlets for American industry suddenly began to display a rude, commercial aggression. In 1967 the United States recorded its first balance of trade deficit since 1893. In short, what had for decades propped up the system was now itself exhausted, and with no possible remedy anywhere in sight. The cycle whereby capitalism had "developed" since the beginning of its decline (crisis - war - reconstruction), had once again arrived at a fatal turning point: the end of reconstruction.
The limits of the economic stimulus of military production
We saw in the previous sections the "virtues" of the outlet constituted by military needs and the importance of this factor in the capitalist economy since World War 2. The limits of this kind of outlet, universally recognised as one of the major "underpinnings of capitalist expansion" in our era, must be analysed if one wishes to understand the basis for the inevitability of the system's next major crisis. We dwell at length on this subject not only because it forms part of the answer to the problem of crises during the decadence of capitalism, but also because it is one of the most spectacular illustrations of a particularly significant phenomenon of this decadence: the precipitous development of non-productive sectors at the expense of the productive sector.
The general confusion which prevails with regard to military production is not foreign to our attempt to clarify the problem: it is indeed not rare to hear "marxists" claim that any distinction between production such as that of armaments and the means of production or means of subsistence corresponds to "ethical criteria" completely foreign to marxist economic criteria .
Armaments: a pure loss for global capital
Capital lives only for and by its self-expansion, that is to say from that process which begins with the exploitation of living labour - the extraction of surplus value - and concludes with the increase of capital due to the transformation of a part of this surplus value into new capital.
Capital is not synonymous with accumulated wealth - even if it is also that. Its specific characteristic lies in its capacity to extract surplus labour and in the goal of this exploitation: the growth of capital. Capital is first and foremost a social relationship.
But armaments have the unique characteristic of possessing a use value which does not allow them to enter, in any form, into the process of production. A washing machine can contribute to the reproduction of labour power, just like a loaf of bread or a shirt. Through the content of their use value, these goods can serve as capital in the form of variable capital. A computer, a ton of iron, or a steam engine, insofar as they are means or objects of labour can function as capital in the form of constant capital. But arms can only destroy or rust (we are discounting for the moment the indirect productive effect of arms production - cf the second part of this article).
Obviously, arms can furnish capital by metabolism. Once sold, they are transformed into money, and with the sum obtained, the capitalist who sold them can buy means of production or subsistence. But the arms themselves do not become capital, for all that. The buyer of arms pays with capital and receives in exchange a product which can never become capital. What global capital earns in the person of the seller of arms is lost in the person of their purchaser. The global result of the operation is zero.
Let us take as an example Marcel Dassault, one of the world's leading arms manufacturers, who sells arms to foreign countries and also to the French state. Consider an example of the first type of sale: Mirage jets sold to Peru. Dassault receives in payment a sum of money, which he can use to replace the capital expended and to increase the capital of his enterprise through the purchase of new machines and new quantities of labour power. For that faction of capital represented by Dassault the exchange is a normal and productive affair. For French capital, the same holds true.
What does this exchange represent, however, for Peruvian capital? To get the necessary money for the purchase of these Mirages, Peru must, for example, sell an equivalent quantity of fish-meal. It could have used this value to increase its own power of exploitation, its capital (fish-meal factories, or fishing boats). Instead, Peru has transformed that value into supersonic fighter-bombers. Such bombers might inflate the pride of the "progressive" military clique in power, but the national capital will never be able to use them directly to extract a penny of surplus value (short of establishing a civil transport service of limited profitability). For Peruvian capital, then, this operation is paid for by a destruction of capital.
For world capital, the unproductive nature of armaments has not disappeared: it has simply been displaced from one point to another within its own sphere. Whatever the exchanges that have taken place, the fact remains that somewhere in the world productive labour has been sterilised as far as the development of capital goes. The place where the weight of that unproductive activity makes itself felt changes nothing about the problem; it is not because armaments cross a national border that their nature in relation to global capital has been transformed.
Now consider the second case: Dassault sells his Mirages to the French state. For the seller of armaments, the operation is still just as profitable and productive for his capital. But for French capital as a whole, it no longer is. Whether through indirect taxes, taxes on capital or taxes on labour, it is always surplus value extracted from living labour that the state, Dassault's customer, uses to buy the Mirages. For the national capital, it is a question of wasted surplus value.Therefore, each time that the national capital sacrifices, through this waste of surplus value, a certain amount of labour for the production of arms consumed by the state, there is that much more surplus value which will be unproductively consumed and which will not expand at all the value of capital as a whole.
 See for example Henri Weber's polemical pamphlet on the AJS