Discourse about policy tends to be cast in a form that assumes ‘the government’ – either the national government itself, or a body operating on its behalf – which is comprised of authorized leaders who determine goals or set norms, which are then applied in practice by officials. In these accounts, policy can be traced back to sovereign authority.

Increasingly, however, policy work is conducted across national boundaries. New regional bodies are emerging whose ability to make rules can override national government authority, such as the European Union. There are also global regimes such as the World Trade Organization, to which governments yield some of their sovereign authority. There are also areas where sovereignty is disputed, like the Antarctic, but governments have sought to construct acceptable regimes of governing. Two of our case studies concern policy work within the European Union, which is perhaps the most developed of these trans-national regimes. One study is by an official (Woeltjes) reflecting on her own work; the other (Geuijen and 't Hart) reports on research on Dutch officials who deal directly with the EU. They offer a particularly interesting perspective because the policy workers involved in these cases can draw on a number of identities – as officials of functional national agencies, as part of a national engagement with the EU, as ‘good Europeans,’ constructing the new regime, and often, also, as professional experts whose professional knowledge is shared across national boundaries.

There are a number of common themes that emerge from these studies. As Woeltjes points out, the problems that this policy activity is concerned with cannot be contained within national boundaries, and we can see the development of a body of specialized knowledge regarding this problem and its management. Geuijen and 't Hart suggest that the institutionalization of these concerns creates new sites for policy practice, with their own norms, rules and specific practices. This gives rise to ambiguity in the relationship between these locations for international rule-generation and the constitutional formulations, which validate rules by referring back to the national political leadership. In these cases, political leadership remains ambiguous because the leaders may have no particular views on the issues under discussion, or may even avoid taking a position when the issue is considered a ‘hot potato.’ Questions may be described as ‘technical’ when leaders approve of what the officials are doing, and ‘political’ where they disagree.