Structure and Process in Consociationalism and Federalism

It is relatively easy to characterize the institutional structureof a country as consociational or federal. The task is muchmore difficult if we turn from structure to the decision process.Within a given structure, there may be a great deal of variationwith regard to the decision process. Data for such variationare given for the decision process in Switzerland for whichthe following four types are distinguished: (1) consociationaland federal, (2) consociational and unitary, (3) competitiveand federal, and (4) competitive and unitary. Difficulties arediscussed in aggregating from such a variation to a measurefor the prevailing decision process in a country as a whole.It is suggested that the problem be disaggregated and that decisionprocesses be compared for particular issues. This can be doneboth across and within countries.     

Information flow and party influence: Theory,measurement, and policy consequences

In this paper I present a theory of party influence, based on Bayesian decision theory, as a process in which the voting decisions of individual legislators are influenced by information concerning the votes or intended votes of their rank-and-file colleagues. Procedures derived from the theory are then used to estimate the magnitude of party influence relative to the influence of the president and of party leaders, committee cue givers, and constituents on roll call voting in three policy domains in the U.S. House of Representatives. The results imply that party influence has important short-run and long-run consequences for public policy.     

Hyperplasia of islets of langerhans and low serum insulin in cot deaths

Of the fifty-seven cases of cot death studied two-thirds were younger than 3 months, which is also the peak age of infantile hypoglycaemia. Findings from routine necropsy and histology were scarce; in eleven cases they could be regarded as potentially fatal. About half of the infants had had a mild virus-type infection approximately one week before death. Special attention was paid to endocrine pancreas. Insulitis or lymphocytes in the septa were discovered in twelve cases. Hyperplasia of the islets of Langerhans was a common observation; the hyperplasia being either nesidioblastosis-like with clusters of islets around ducti, or diffuse. The average proportion of islet tissue in the whole pancreas parenchyma was around 5% in infants aged 1–6 months, the percentage being significantly greater than in age-matched controls (4.3%). The pancreatic insulin content was also higher in the cot death cases. Serum insulin values were low (mean 4.8 ± 1.2 μU/ml) in cot deaths; in the controls they were twice as high (mean 11.6 ± 1.6 μU/ml) (p < 0.005). The cause of death in this group of cot deaths could thus be (congenital?) hyperplasia of the islets, possibly combined with a lesion in the B-cells caused by a virus. The mechanism of death would be hypoglycaemia.