The dispute is called the investiture contest, and stated plainly it sounds like nothing: a disagreement about whether a king could hand a newly appointed bishop the ring and staff that symbolised his office.
Here is why it was not nothing. A bishop in eleventh century Germany was a great landowner, often the largest in his region. He held estates, collected dues, ran courts, controlled towns, and owed the king military contingents. He was also, unlike a count, unable to found a dynasty, which from a king's point of view made him the ideal official: powerful in life, and vacant on death.
So the German kings had built their government on churchmen. Something like a third to a half of the realm's resources ran through bishops and abbots the king had appointed. If the king could no longer choose them, he did not lose a ceremony. He lost his administration.
And from the reforming clergy's side the objection was equally concrete. A bishopric chosen by a king went to the king's relatives and servants, was frequently sold, and produced a church staffed by men whose first loyalty was secular. If the church was going to be an independent moral authority, it could not be a branch of royal patronage.
Both sides were right about what was at stake, which is why it took fifty years and two civil wars.
The pressure came from a movement inside the church that had been building for a century and a half.
It started with monastic reform. A monastery founded in Burgundy in 910 was granted, unusually, direct subordination to the pope rather than to any local lord or bishop, which insulated it from lay interference. It became the head of a network of hundreds of houses across Europe, and it made independence from lay control into an ideal with a large and disciplined constituency.
The reformers' targets were three practices they regarded as corruption: the buying and selling of church offices, clerical marriage and concubinage, which put church property at risk of becoming family property, and lay investiture. The first two were popular causes. The third was the one with a state attached to it.
In 1059 the reformers took the decisive institutional step and changed how popes were chosen, transferring the election to the cardinals and cutting out both the Roman nobility and the emperor, who had been appointing and deposing popes for a century. The rule has been modified since and it is still, in essence, how it is done.
In 1073 a monk who had been the movement's organiser for twenty years became pope as Gregory the Seventh, and he was not a man for gradualism.
In the papal register for 1075 there is a list of twenty seven short propositions. Nobody knows exactly what it was for, and it may be headings for a projected collection of canon law, but it states the maximal position with startling clarity.
That the Roman church was founded by God alone. That only the pope may depose bishops. That his title is unique in the world. That he may depose emperors. That he may absolve subjects from their allegiance to unjust men. That the Roman church has never erred and will never err. That a pope, if canonically ordained, is undoubtedly made holy.
Read the fourth and fifth of those again. If a pope can depose a ruler and release his subjects from their oaths, then every king in Europe holds his office conditionally. That is not a claim about ceremony. It is a claim to be the final judge of legitimate government.
The German king, Henry the Fourth, aged twenty five and already fighting a Saxon rebellion, responded by having his bishops declare Gregory deposed, in a letter addressed to Hildebrand, not now pope but false monk. Gregory excommunicated him, deposed him in turn, and released his subjects from their obedience.
And the subjects took it. The German princes, who had their own reasons to want a weak king, announced that if Henry were not absolved within a year they would elect somebody else. That is the moment the theory became a weapon.
In January 1077 Henry crossed the Alps in winter with a small party, found the pope staying at a castle at Canossa in the Apennines, and stood outside the gate for three days as a barefoot penitent in the snow, asking for absolution.
The scene has been painted, dramatised and used as shorthand for humiliation ever since. A German chancellor in the 1870s, fighting his own quarrel with the Catholic church, declared that we shall not go to Canossa, and the phrase entered the language.
The reading as humiliation is at best half right. Gregory was a priest confronted by a penitent, and a priest cannot refuse absolution to a penitent who is sincere. By making the approach as a penitent rather than as a king, Henry put him in a position where refusal would have cost him the moral high ground he had built everything on. Gregory absolved him. Henry went back to Germany with the excommunication lifted, defeated the rival the princes had elected anyway, and seven years later marched on Rome, installed his own pope, and drove Gregory out.
Gregory died in exile in Salerno, having been rescued by Norman allies who sacked Rome so thoroughly in the process that parts of the city never recovered. His reported last words, that he had loved justice and hated iniquity and therefore died in exile, are probably not authentic.
So who won at Canossa. Tactically, Henry. Structurally, Gregory, because the spectacle established in everyone's mind that a king might have to answer to somebody, and that image did more work over the following centuries than the outcome of any campaign.
The formal end came in 1122, at Worms, with a compromise that separated the two things the ring had confused. The church would invest a bishop with the spiritual symbols, the ring and staff. The emperor would invest him with the temporal ones, the regalia, the lands and rights that came with the office. Elections would be canonical, but in Germany they would happen in the emperor's presence, which in practice preserved a good deal of his influence.
It is a lawyer's solution, and it worked because it acknowledged that a bishop genuinely was two things at once. It also established the habit of settling such disputes by written concordat, which is how church and state relations have been arranged ever since.
The papacy went on rising for another century. Under Innocent the Third, from 1198, it arbitrated a disputed imperial election, laid England under interdict until its king surrendered the kingdom and received it back as a papal fief, called the crusade that ended up sacking Constantinople, launched another against heretics in southern France, and summoned a council that legislated for the whole western church, from confession to the treatment of Jews.
The reversal came fast and against a different opponent. In 1296 the king of France began taxing the clergy without permission. Boniface the Eighth resisted, escalated, and in 1302 issued the most absolute statement of papal supremacy ever made, that it is altogether necessary to salvation for every human creature to be subject to the Roman pontiff. The king's response was to send an armed party to seize him. The old man was held for three days at Anagni and died a month later.
Six years after that the papacy moved to Avignon, on the edge of France, and stayed for nearly seventy years. Then a disputed election produced two popes at once, and later three, for another forty. The institution that had claimed to depose emperors spent a century unable to establish who it was.
Germany and Italy paid for the quarrel in a way France and England did not, and the difference shows on the map for the next seven hundred years.
The struggle ran on into the thirteenth century against Frederick the Second, a Hohenstaufen who was king of Germany, king of Sicily and emperor at once, raised in multilingual Palermo, fluent in Arabic, sceptical enough that his enemies called him the baptised sultan. He was excommunicated four times, once for delaying a crusade and again for going on it while excommunicated, and he recovered Jerusalem by negotiation rather than battle, which satisfied nobody.
The papacy fought him with every instrument it had, including preaching a crusade against him, and after his death in 1250 it pursued his family until the line was extinguished, the last of them beheaded in a Naples square at sixteen.
The cost of winning was that Germany had no effective central monarchy. Each round of the conflict had been settled by buying the German princes, and what they were bought with was rights: to their own courts, their own coinage, their own castles, their own succession. By the time the dust settled the emperor was elected by a handful of them and ruled little beyond his own family lands. Italy, meanwhile, had spent two centuries as the battlefield, and its cities had used the space between the two claimants to make themselves independent, which is where chapter sixteen and chapter twenty two begin.
France and England, unbothered by an imperial claim on their soil, spent the same centuries doing the opposite: accumulating royal courts, royal taxes and royal officials. Germany and Italy would not be unified states until the 1870s, and the chapter on that is number forty three.
Firm: the documents. This is one of the best evidenced disputes of the Middle Ages, because both sides wrote constantly and both sides kept archives. We have the letters, the decrees, the concordat text, and a large polemical literature arguing the theory on both sides.
Soft: Canossa itself. The main accounts are by partisans, including Gregory's own letter describing it, and the famous details of the snow, the bare feet and the three days come from writers with an interest in the shape of the scene.
Soft: the twenty seven propositions. Their status is genuinely unclear. They appear in the papal register without explanation, were never promulgated as law, and historians disagree about whether they represent policy, a draft, or an aspiration.
Contested: what it all amounted to. One long tradition treats this as the origin of the western separation of church and state, and therefore of limited government: two jurisdictions, neither able to absorb the other, leaving a permanent gap in which liberties can grow. Another points out that neither side wanted separation, both wanted supremacy, and the outcome was an accident of stalemate.
The accident, if that is what it was, produced something concrete. Both sides needed to argue from texts, so both sides built up systematic law. Canon law was collected, harmonised and taught as a discipline from the 1140s; Roman law was recovered and taught alongside it. The professional lawyer, the law faculty and the idea that authority must be justified by argument from a written rule all come out of this quarrel, and they are the subject of the chapter after next.