One of the most popular explanations for Labour’s poor electoral performances in both Hartlepool and Batley and Spen is that a mythical group of working-class people voted solidly for Labour until Jeremy Corbyn and Brexit came along. This idea is stubbornly persistent in political analysis, yet it’s bunk. Labour has seen its vote share collapse, both nationally and in these seats, three times over the last 50 years. Indeed, the belief that working-class people traditionally voted Labour has only been true (and barely so) for a mere 25 years of British history, and a long time ago. The working classes of the Edwardian years, from where our cloth-cap-and-shawl images of miners, shipbuilders and textile workers come, mostly voted Liberal and Tory. In the interwar years when Labour first stood as a national party, the Tories always got more votes. Labour’s best performance was in 1935, when it secured 38.6% of the vote. But at the time, the working classes accounted for 75% of Britain’s population, and many of them voted Tory. At a reasonable guess, the only period when even a slim majority of the working class consistently voted Labour was during the 25-year period between 1945 and 1970. Even though Britain was an overwhelmingly working-class nation, Labour never managed to secure even 50% of the total vote. In these years, Labour’s vote share hovered between 43% and 48%. Its highest-ever share of the vote was in 1951, the point at which – as Eric Hobsbawm memorably put it – the forward march of Labour halted. The period from 1945 into the 1970s was without question Labour’s moment of success. Not only did it win elections, but it transformed British society. But over the last half century, as Labour’s halted forward march went into reverse, its vote share fell and oscillated wildly. Between the 1970s and 1983 the party’s share of the vote collapsed to 28%. A similar fall took place between 1997 and 2010 (when the Labour vote was reduced to 29%), and between 2017 and 2019. In fact Labour did better in 2019 than in 1983 or 2010, with 32% of the vote. Scotland dramatically dumped Labour, the largest party in the country, in 2015. In 2021, the party’s vote share continues to fall: in the recent byelections in Hartlepool and Batley and Spen, Labour had a lower share of the vote than it had in these seats at the 1983 election under Michael Foot. That is not obvious from the case of Batley and Spen, where Kim Leadbeater won the seat by a slim margin (as a result of our first-past-the-post system, she won with only 35% of the vote). At the national level we also need to distinguish clearly between what share of the overall vote a party gets, and whether they win. In 1935 Labour got 38% of the vote but only 25% of seats. In 2019 it got a lower share of votes but a higher share of seats. The perversity does not end there: although Labour was more popular with voters in 2019, when it got a larger share of the vote than in 1983, 1987, 2010 or 2015, it got a lower number of seats in 2019 than in any of these elections. Indeed, once we look beyond the fallacious, Westminster-centric view of elections, which measures the popularity of a party according to how many seats they get rather than their share of the vote, a very different story of Labour leaders’ political success emerges. In 1997 Tony Blair did as well (but no better) than Hugh Gaitskill did in 1959, but Blair won a huge parliamentary majority whereas Gaitskell lost seats. In 2017, Jeremy Corbyn did slightly better in terms of vote share than Harold Wilson in 1974; but while Wilson won a parliamentary majority, Corbyn was only able to deprive Theresa May of hers. Nor should we think of the so-called red-wall seats as if they were stuck until recently in some unthinking Labourist la-la land. They followed national trends, not least electorally. And since 1970 there have been two powerful upswings in Labour’s vote share, which increased from 1983 to a peak in 1997 (44%), and from a nadir in 2010 to the second peak in 2017 (41%). Blair and Corbyn both increased Labour’s share nationally, and in all kinds of seats. In Hartlepool Labour’s recent vote share peaked in 1997 and 2017, in Batley and Spen in 2001 and 2017. There were deep troughs in 1983 and 2010. It is, tellingly, the same story in southern Chesham and Amersham – troughs in 1983 and 2010 and peaks in 1997 and 2017. Even more absurd is the notion of an unchanging working class recently betrayed by Labour. The working class has changed radically since Labour’s heyday. Where once it was made up of miners and factory workers, today it includes (among others) health service, education and hospitality workers. There are no miners left, and far fewer factory workers today. The working class is more female, better educated and not captured by the 43% of the population whom advertisers class as “C2DE”, both because a quarter think of themselves as middle class but also because workers are not limited to the old “manual” occupations this archaic definition rests on. The phenomenon of a working-class red wall is an ideological concoction that benefits Labour’s enemies. It makes little sociological or psephological sense today, and the fragment of the past it reflects is one of Tory working classes. Yet this group has come to define how Labour thinks of the working class. That the party views this Tory analysis as a bellwether of its fortunes speaks to its collapse as an independent, transformative political force. If it is ever to win significant support today among real English people, Labour needs to understand its own history, celebrate its successes and love itself, its members and its voters. Labour undoubtedly still needs the working-class vote. Winning this means creating a Labour party for workers and trade unionists in the present day, not those of a mythologised past. Doing better among those workers than Labour did in its heyday would also be necessary for electoral success. The party needs to relearn not only how to get votes, but how to keep them too, which it has failed to do for decades. To make all this possible it needs to present a real alternative with vigour and confidence, and to stop acting as if it believed that this uniquely dangerous Conservative government had the British past, present and future in its bones. David Edgerton is Hans Rausing Professor of the history of science and technology and professor of modern British history at King’s College London. He is the author of The Rise and Fall of the British Nation: a Twentieth Century History