Three points stand out in relation to the EU referendum, and which partly explain the territorial differentiation of the results in a fragmented Britain. First, remember what Britain has become over recent decades – a more complex political system vis-à-vis elections, parties, identities and governments. We may retain the first-past-the-post electoral system for one level of elections but we now have multi-level politics and multi-party politics with numerous devolved institutions, weak regional representation for England in British structure (‘northern powerhouses’ have yet to emerge) and a lot more than two-party politics at elections. The referendum helped to give voice to some of these differences during the campaign and importantly, in the post-result period too.

UKIP election successes in 2014 and 2015 in England were actually portents of what was to come rather than sporadic developments. Results in Scotland from 2007 onwards have shown a more fragmented British picture that became catastrophic for Labour as a Britain-wide party. Northern Ireland, meantime, has always had a different party pattern, whilst Wales became more Eurosceptic with the 2016 election whilst still having a prominent nationalist party.

Two-party Britain has been under threat for some time and recent results – even the 2010-15 coalition – demonstrate that. At the same time, there has been a developing territorial politics around identity, devolution and Europe that was on display during and after the EU referendum – with EVEL after the 2014 referendum – and seems set to be a prominent feature of British politics for some time.

Second, referendums are not like elections. There is a binary choice and voters can vote according to party or be free and ignore what their party tells them – though this is complicated when parties campaign on both sides of an issue. Of course, when voters become detached from parties this can have continuing effects beyond a referendum – which we recognize from Scottish Labour’s electoral problems after 2014 and in the current period too in parts of England.

Third, it is important to ignore simplistic arguments about voting behaviour at the referendum, especially around race and immigration. Sections of the media have been more than happy to blame working class Labour voters in the North for the ‘leave’ vote, without actually inquiring why this was the case or why working class voters in some other parts of the UK voted ‘remain’. Focusing narrowly on immigration in itself as a cultural phenomenon is also not particularly instructive. Rather, immigration needs to be situated in the context of political disengagement and poor representation, open labour markets, insecure employment opportunities, austerity and lack of public investment in communities when it comes to housing, training opportunities etcetera.

So, in reality there are all sorts of patterns of voters supporting ‘remain’ and ‘Brexit’ though, as yet, we don’t have good voter data on the result in Scotland. We do have some basic details of the demographics at the British level in relation to social class, age etc., which all tells a mixed picture (see Lord Ashcroft polling and YouGov). Look at the social group AB who voted 43% ‘leave’, versus the 36% of DE group who voted ‘remain’, and you realise voting patterns are much more complex than media portrayal. Scotland voted convincingly ‘remain’ but a large section of it voted ‘leave’ and they were spread across parties and demographic groups too.

Of course, where the territorial patterning of results does matter is in the post-referendum political landscape of a Scotland and Northern Ireland that were ‘remain’ (along with London) and England and Wales that were clearly ‘leave’. How this plays out is difficult to tell but as all three ‘remain’ areas have elected assemblies and governments, with representation in the British and Irish Council and some international presence, then you realise there’s an awful lot of politics to come to determine who actually leaves the EU and whether Britain remains united or sees more referendums or special status areas that defy Brexit.

A final point is worth keeping in mind too. Westminster elections are not in the gift of the Prime Minister. Instead, they are determined by the Fixed Term Parliaments Act of 2011 and a ‘snap’ general election requires a 2/3s majority in the House of Commons. So, for all the talk of an early election, it would require cross-party support in a period in which both the Conservatives and Labour have much to lose. The Conservatives have a slim majority and are in a fragile condition following a divisive referendum. Labour has been in complete meltdown and faces the prospect of losing northern seats to UKIP in significant numbers due to the EU referendum and the fact it has become so out of tune with its core voters. There is definitely trouble ahead for Labour in its heartlands – with 27% of its voters from 2015 intending to desert the party – though Labour’s problems were evident before 23 June.

Peter Lynch is senior lecturer in politics at the University of Stirling and is currently working on a book looking at referendums.