Path Dependencies and Critical Junctures: A Response to Baudenbacher

I am writing this article in English in response to earlier contributions by Carl Baudenbacher and Smári McCarthy, both of which were also published in English.

Ahead of Saturday’s historic referendum, prominent voices in the no-camp have argued that reopening the EU accession negotiations would create a path dependence that would become increasingly difficult for Iceland to resist in the course of the negotiation process. In other words, by restarting the negotiations, we will embark on a path that will allegedly lead us into the EU whether we want it or not, regardless of the substantive content of the “package” that many of us would very much like to see. This is nonsense. Even in the event of a victory for the yes-camp on Saturday, the hurdles for Icelandic EU membership will remain high, especially at the domestic level. Talk about an alleged path dependence created by accession negotiations is as implausible as it is ahistorical.

Path dependence is a central concept in historical institutionalist theories in the social sciences, especially in the fields of political science and EU studies. But in such theories, the term is also complemented by the concept of critical junctures, a term that – while conspicuously absent in Baudenbacher’s argument – has shaped the trajectory of European integration on numerous occasions. Baudenbacher’s argument relies entirely on the deterministic assumption that once you opt for one path, this choice will make it difficult to change course down the road. In principle, this is a valid point that has been applied in many influential studies on the development of European integration: institutional choices made at the outset of the integration process have indeed had a profound impact on the subsequent development of the European project – and continue to do so to this day. But historical institutionalism also emphasizes the conditions under which path dependence is resisted, that is: under which conditions a departure from an initially chosen path becomes necessary or even inevitable. The path dependence of a yes in Saturday’s referendum is implausibly deterministic because we can already predict which forks in the road we would encounter at different stages in the accession process: the negotiations themselves, a second referendum, parliamentary elections, not least the required change to the constitution. One path may indeed lead to full membership. But domestic constraints, not least the fierce opposition from some interests in this country – which have succeeded in selling the public the outlandish narrative that we are deciding on membership now – will make full membership very difficult to achieve. We should be grateful for this democratic process, despite the fact that the debate may not have turned out to be the “festival of democracy” that some had hoped for. Smári McCarthy made an important point when he argued that democracy requires the possibility of investigating something “without being accused of deciding it beforehand.”

But the path-dependency argument is so unconvincing also because it appears to be entirely ahistorical. Indeed, it would be very interesting to see at least some empirical indication of an historical precedent – a country whose path has led it into the European Union purely because of the strength of the path dependence of the accession negotiations. On the other hand, evidence to the contrary abounds. In fact, Iceland is a obvious case in point: it only took a change in government, not even the referendum promised in the election campaign, for Iceland to reverse course in 2013/2014. What happened to the path dependence there? Norway is another example, where the path dependency of an already existing accession treaty was evidently too weak to convince the public to vote for membership. And even Switzerland’s 1992 referendum on EEA membership may be taken as reminder that path dependence is not at all as deterministic as some seem to assume.

More examples can be drawn from the history of EU treaty reform. These remind us of the power that even small member states in the treaty reform process (to which all member states have to agree). Denmark’s opt-out from the common currency (amongst others) was possible only because the Danes rejected the Treaty on European Union (better known as the “Maastricht Treaty”) in its first referendum. The French and Dutch “no” to the Constitutional Treaty in 2005 created a ratification crisis that defied the international expectations that, according to Carl Baudenbacher, create an important part of an essentially inescapable path. And even the renegotiated Treaty of Lisbon, in force since late 2009, required changes because Ireland departed from the path-dependent process and said no in its referendum in 2008.

I would be very curious to hear the case for why the force of the presumed path dependence should be so much stronger in the case of Iceland, especially given that “we” have already changed course at earlier stages. But maybe the path-dependence argument is to be understood simply as political argument rather than as academic analysis. In this case, we should treat it as such.

Höfundur er prófessor í stjórnmálafræði við Háskóla Íslands.