The Norman Yoke

George Orwell spotted the symptom before anyone had named the disease. In The Lion and the Unicorn, written in 1941 while German bombers worked overhead, he observed that England is perhaps the only great country whose intellectuals are ashamed of their own nationality. In left-wing circles, he wrote, it was felt there was something slightly disgraceful in being an Englishman, and a duty to snigger at every English institution. He was describing his own side. He was not joking.

What Orwell may not have known is that the condition he described - a ruling stratum estranged from the people beneath it, holding them in faint contempt - already had a name in English politics. It had carried it for three hundred years. The name was the Norman Yoke.

The oldest argument in England

The theory ran like this. Before 1066, the English governed themselves - through their own institutions, in their own language, under their own law. Then came the conquest. A foreign aristocracy seized the land, imposed its law in a language the people could not speak, and settled in as a permanent ruling caste. Everything wrong with England since - the great estates, the rigged courts, the distance between rulers and ruled - descended from that original theft.

Here is the part that has been memory-holed so thoroughly that saying it now sounds like a provocation: this was a left-wing argument. The founding one. The Levellers argued it in the 1640s, demanding the law be conducted in English and the commons restored. The Diggers dug on St George's Hill because the earth had been stolen from the people at the conquest. John Lilburne stood trial on it. Gerrard Winstanley built a theology on it. Tom Paine was still swinging it at the aristocracy in the 1790s. The historian Christopher Hill - a Marxist, for the avoidance of doubt - made it the radical tradition's origin story in his standard essay on the subject in 1954. For three centuries, if you argued that England's elite was alien to England, you were standing on the left.

Now listen to the same argument today. Our elite is estranged from us. Our institutions have been captured. They prefer their own kind, across borders, to their own neighbours. They hold the people they govern in contempt. Word for word the Leveller frame. And in today's vocabulary it codes as far-right populism. The argument did not move. The label moved around it.

The finding that refuses to go away

It would be convenient to file all this under romantic myth-making, and much of it belongs there - we will come to that. But one piece of evidence is awkwardly solid. The economists Gregory Clark and Neil Cummins tracked rare Norman surnames - names that arrived with the Domesday survey of 1086, like Darcy, Mandeville, Percy, Talbot and Baskerville - through nine centuries of English records, using Oxford and Cambridge enrolment as the measure of elite membership. In 1170, the conquerors' descendants were over-represented among the English elite by a factor of about sixteen. By 1470, four times. In 2012 - not a misprint - still roughly twenty-five per cent over-represented, against a control group of artisan surnames like Smith and Carpenter that have sat in the middle of the distribution the whole time. Elite persistence, on their measure, runs at about 0.9 per generation. Nine hundred and fifty years, and the tide has not finished going out.

One honesty is owed here. Clark's explanation of his own finding - that social competence is substantially heritable - is contested, and reviewers have called it overreach. This piece takes no position on that and does not need to. The finding is the record. The explanation is an argument about the record. They are different things, and the reader is invited to check both.

What the myth got wrong

The steelman against all of this deserves its full weight, because it is mostly true. Anglo-Saxon England was no free commonwealth. Roughly one person in ten recorded in Domesday Book was a slave, and the thegns did not ask the churls their opinion of anything. The golden age was invented by the people who missed it. In that sense the Norman Yoke was always bad history.

But look at what the best version of the argument actually claimed. The Levellers' grievance was not about blood. It was about capture and jurisdiction: law conducted in a language the people could not speak, land held by a class that answered elsewhere, institutions that had stopped belonging to the country they governed. That claim does not need the golden age to be real. It only needs the capture to be real - and the surname data, whatever explains it, suggests the capture had remarkable staying power.

Normans by disposition

Which brings the argument home. Nobody serious thinks today's establishment matters because of who its great-great-grandfathers were. The interesting inheritance is not genetic but structural: a stratum that shares more - in education, in outlook, in loyalty, in where it would rather live - with its international peers than with the people it governs, and that finds attachment to the country beneath it faintly embarrassing. David Goodhart calls the two camps Anywheres and Somewheres. The Levellers had a blunter name for the same shape.

Orwell's ashamed intellectuals, in this frame, are not traitors and need no conspiracy. They are simply the current occupants of a very old position - Normans by disposition, not by blood. And the reason our political vocabulary keeps misfiring may be that it is describing a 950-year-old structure with labels invented the day before yesterday. An argument the left owned for three centuries now belongs, apparently, to the right. When a coding flips that completely while the evidence stands still, the useful conclusion is not that the argument changed sides. It is that the labels were never load-bearing in the first place - which regular readers will recognise as the recurring finding of this desk.

The records, as ever, are linked. The reader is invited to disagree - in English, which the courts have permitted since 1650.

An opinion of the house. The argument is ours; the record beneath it belongs to no one.