in ‘home’ disciplines. IR has lacked attention to ‘fundamental issues concerning the nature and purpose of social science theory and research, as raised in the philosophy and sociology of science literature’ (Godard, 1993: 284).
In positive terms, IR had a clear intellectual basis in institutional economics, one of whose key founders, John R. Commons, was a leader in IR (Kaufman, 2003). The depth of this basis is illustrated by Kaufman’s (2004b: 95-116) analysis of the ‘science building’ component of IR in its country of first development, the US: Kaufman details in the space of 21 pages what the early institutionalists said in relation to theory and methodology and how this shaped IR. Yet, as he admits, theoretical development was often ‘inchoate’, and a leading work of institutional analysis, Commons’s Institutional Economics (1934), had a ‘negligible influence’ on IR (pp. 101, 102). Hodgson (2006) similarly notes the lack of a ‘systematic statement’ of institutionalism, and the inability of the approach to address such key issues as the operation of markets and the nature of incentive structures. It was thus in a weak position to respond to more rigorous and mathematical models.
Several scholars have analysed ‘old’ institutionalist analysis, often stressing its strengths as against newer approaches using the same label (Jacoby, 1990; Hodgson, 2006). In the words of Stinchcombe (1997: 6), new approaches focus on what is easy to ‘mathematicize’, to the neglect of the ‘guts of the causal processes of institutional influence’, that is, the complex and uncertain negotiation of order. The purpose here is not to repeat these analyses, but to use them as a bridge between empirical IR and the philosophy of CR.
Institutional economics, with its emphasis on historical variation and the embeddedness of economic processes in social institutions, is strongly compatible with both IR and CR. Jacoby’s (1990) important essay on ‘institutional labour economics’ (ILE), which is contrasted with new efficiency oriented ILE (NEO-ILE), makes several key points. The older ILE tradition stressed four explanatory tenets: indeterminacy of the labour contract, and the importance of local custom or bargaining; endogeneity; behavioural realism (meaning realistic assumptions, not realism in a CR sense); and diachronic analysis. These features, says Jacoby, provide a better explanation than does NEO-ILE, which seeks an explanation of observed phenomena in terms of efficiency: American internal labour markets for example are seen in NEO-ILE as an efficient means to induce worker commitment. Yet this functional and synchronic analysis cannot explain some key facts, such as that the alleged efficient features emerged many years after the appearance of the large firms that apparently stood to gain from these features: why did it take so long for efficiency to be identified? And other national IR systems operate with very different forms of organization. An explanation sensitive to history and context works better although, as Jacoby admits, it lends itself less well to clear hypothesis testing than does work in the NEO-ILE style. In short, a search for the complex and embedded causal powers of institutions drives ILE, but there is also the problem, shared with CR, that explanations can appear to be idiosyncratic and ad hoc.
Hodgson (2006) has further crystallized some of the core tenets of old institutionalists.
First, an institution is not necessarily anything as formal as a firm or a trade union; the term also embraces ways of thinking or acting that have a degree of permanence.