Visigothic Kingdom

A treaty and a province

Drafted from Heather, The Peoples of Europe (1996)formation

In 418 the Roman government settled a Gothic army in Aquitaine. The Goths had been inside the empire since crossing the Danube in 376, had destroyed an imperial army and killed an emperor at Adrianople in 378, and had sacked Rome in 410. The settlement gave them land and a treaty, and made them federates: an army in imperial service, under its own king, holding territory in a province.

418 is the earliest point at which there is a polity here rather than an army. What it is not is a migration of a people out of Scandinavia into Gaul, which is the account the Gothic historian Jordanes gives and which Heather argues against at length. Gothic identity on that reading was assembled inside the empire in the fourth and fifth centuries out of several groups, under pressure from the Huns, and the Visigoths of 418 are a coalition that had been in existence for about a generation.

What kind of succession this was

It is worth being careful about what kind of succession this is, because none of the ordinary words fits.

Conquest would be false: the Goths did not conquer Rome and the settlement was a Roman decision. Secession would be false: they were never part of the Roman state in the way a province was. What happened is that a group inside the empire took over the administration of territory as the central government's ability to pay and to direct it disappeared, and then kept it. The Vandals and the Ostrogoths acquired their kingdoms the same way.

That is a different relation from one state succeeding another. These are polities assembled out of a dead empire's provinces by people who had been inside the empire, in its pay, for a generation before they were on their own.

Losing the half it started with

Everything the kingdom is remembered for happened after it lost Gaul.

In 507 Clovis destroyed the Visigothic army near Poitiers and killed King Alaric II, and Aquitaine went to the Franks. The kingdom's centre moved across the Pyrenees, and Toledo became the capital under Leovigild in the 570s. The country the kingdom is remembered for holding is a peninsula it had held only partially before 507 and did not fully control until Leovigild finished the Suevic kingdom in the north-west in 585 and the Byzantines were pushed out of the south-east in 624.

So a kingdom that lasted two hundred and ninety-three years spent its first eighty-nine in a country it lost, and the Spain it is remembered for is the second half of its life.

As was their custom

Drafted from Tours, History of the Franks (1916)formation

The Visigothic kingdom people mean when they say Visigothic Spain was made in an afternoon in 507, by losing a battle in France.

For the previous ninety years it had been a Gallic kingdom centred on Toulouse, with Spain as an appendage. Everything that makes it a Spanish story — Toledo, the councils, the conversion, the law, 711 — happens because of one day near Poitiers. The account of how comes from a bishop of Tours writing eighty years later, and it has to be read with his commitments in front of you: Gregory is a Catholic writing about Arians and a partisan of the Frankish house, and he does not pretend otherwise.

He begins with a meeting. Alaric II, watching Clovis conquering steadily, sent envoys proposing that the two of them meet. Clovis agreed. They met on an island in the Loire near Amboise, in the territory of Tours, and they talked and ate and drank together, and plighted friendship, and departed in peace.

Then Gregory adds the sentence that tells you what the meeting was worth: even at that time many in the Gauls desired greatly to have the Franks as masters.

The bishop of Rodez

The next paragraph is the mechanism.

Quintian, bishop of Rodez, was driven out of his city because people said it was his desire that Frankish rule should extend over the country. A quarrel followed; the Goths living in Rodez became suspicious at the citizens' charge that their bishop wanted to submit himself to the Franks; and they took counsel and decided to kill him. He was warned, got up in the night and left with his most faithful servants for Clermont, where the bishop took him in and gave him houses, fields and vineyards.

That is the Visigothic problem in one anecdote. An Arian Gothic garrison governing a Catholic Gallo-Roman population, in which the local bishop is plausibly suspected of preferring the Catholic king across the border, and the garrison's response is to consider murdering him.

Vouillé

Clovis met Alaric on the plain of Vouillé, ten miles from Poitiers. One army wanted to fight at a distance and the other tried to close.

Gregory's account of the outcome is four words long and contemptuous: when the Goths had fled, as was their custom, Clovis won the victory by God's aid.

Alaric was killed. Clovis was nearly killed himself, struck at by two lances at once and saved by his mail and a fast horse. A great many of the Auvergnats who had come with Apollinaris and the leading senators died there — Roman provincials fighting on the Gothic side, which is worth noticing against Gregory's framing.

Alaric's son Amalaric got away to Spain and, Gregory says, wisely seized his father's kingdom. Clovis wintered at Bordeaux, took all of Alaric's treasure at Toulouse, and went on to Angoulême, where — Gregory reports without a flicker — the Lord gave him such grace that the walls fell down of their own accord when he looked at them.

What was actually lost, and what was made

Everything north of the Pyrenees except a coastal strip.

The kingdom had been centred on Toulouse and its heartland was the country between the Loire and the Pyrenees. After 507 it was a Spanish kingdom with a Gallic remnant, ruled at first from Barcelona, and it never recovered the ground.

The interesting question is why it lost so completely and so fast, and Gregory — against his own intention — supplies the answer in the first two anecdotes. The Gothic kings were a small Arian military aristocracy governing a Catholic population whose bishops were the effective leaders of every city, at a moment when there was a Catholic king next door. It took a century and a half after Vouillé for a Visigothic king to convert and close that gap, and by then the Gallic provinces were gone.

The baptismal pool

Drafted from Tours, The History of the Franks (1927)expansion

The Visigothic kingdom that people mean when they say Visigothic Spain was made by Leovigild, who reigned from 568 and spent it doing what none of his predecessors had managed: he put down the Suevic kingdom in the north-west, pushed at the imperial enclave in the south-east, reformed the law, minted in his own name and made Toledo a capital. Almost none of that reached the one contemporary historian who was writing about him.

Gregory, bishop of Tours, was interested in Leovigild for a family reason. A Frankish princess had married his elder son, and the quarrel that followed was about the faith Gregory shared with her. So the fullest surviving account of the reign that built the kingdom is an account of a domestic disaster inside it, written by a hostile Catholic across the Pyrenees, and it has to be read that way.

Ingund

Leovigild had married Goiswintha, the widow of a previous king, and Gregory introduces her as the chief agent of the persecution of Catholics in Spain — noting with satisfaction that divine vengeance had marked her in return, since a white cloud had covered one of her eyes and driven the light from it.

Into that household came Ingund, daughter of the Frankish king Sigibert, sent into Spain with a great train to marry Leovigild's elder son. Goiswintha was her grandmother, and received her joyfully, and then began with persuasive words to entice her to a second baptism into the Arian faith.

Ingund refused. It is enough for me, Gregory has her say, to have been washed once for all from original sin by the baptism that maketh whole, and to have confessed the Persons of the Holy Trinity one and equal.

At these words the queen, aflame with fury of wrath, seized the girl by the hair of her head and dashed her to the ground; there she spurned her for a long time with her feet, and, all bleeding as she was, had her stripped and plunged in the piscina.

The piscina is the baptismal pool. A queen who could not argue a thirteen-year-old into conversion had her rebaptised by force in the font. Gregory adds, in the careful voice he uses when he is repeating rather than asserting, that as many maintain, Ingund never suffered her mind to swerve.

The rebellion

Leovigild gave the couple a city of their own — Seville, on the best reconstruction — and there Ingund did to her husband what her grandmother had failed to do to her. She pressed him for a long time, and eventually Hermenegild renounced Arianism, was baptised a Catholic, and took the name John.

His father began looking for a way to destroy him. Hermenegild learned of it and went over to the emperor's side, making friends with the imperial commander whose army was in Spain. Leovigild sent for him to talk; the son answered that he would not come, since his father was his enemy because he was a Catholic.

What settled it was money. Leovigild gave thirty thousand pieces of gold to the imperial prefect to withdraw his help, and then marched. When the armies met, the men who should have supported Hermenegild abandoned him, and he fled into a church nearby, saying that his father must not come against him, since it is impious that a father should slay his son or a son his father.

Leovigild sent in the younger brother — Reccared, who would be king six years later — with an oath that the prince should not be abased, and a promise that if he threw himself at their father's feet all would be forgiven. Hermenegild asked that the king be summoned, and prostrated himself. Leovigild raised him up and kissed him, spoke kindly to him, and walked him back to the camp. There, disregarding the oath, he made a sign to his men, who stripped the prince of his own clothing and put vile raiment on him; and at Toledo he took away his son's servants and sent him into exile attended by one slave.

Gregory records the ending in a single clause, later and elsewhere, as a thing that had happened: her lord Hermenegild was slain by his father Leovigild.

Ingund never got home. Left with the imperial army when her husband fled, she was being conducted to Constantinople with her small son when she died in Africa, and was buried there.

What the story is doing

Gregory is not a neutral witness, and the shape the story took afterwards is a martyrdom: a Catholic prince killed by an Arian father for the faith. Gregory the Great put a full martyr's account of it in his Dialogues — Dalton, who translated the Frankish chronicle, calls that version fictitious and suggests it came from Leander of Seville, who was in exile at Constantinople while it happened and did not dare go home to Spain until Leovigild was dead. By the eighth century Paul the Deacon assumes the martyrdom as a matter of course. The Spanish chronicle written closest to the events says only that Hermenegild was killed at Tarragona by a man named Sisbert.

What Gregory wrote down makes the other reading possible, and he wrote it down without appearing to notice: the thirty thousand gold pieces paid to buy off the imperial prefect, the imperial army in the field on the son's side, the picked men, the troops who changed sides. That is a rebellion in alliance with a foreign empire holding territory on Spanish soil, and the conversion may have served it rather than caused it. A martyr does not usually need a general.

What is not in doubt is the arithmetic of the outcome. Leovigild won his war, killed his heir, and died within a year or two of it — some say, Gregory reports carefully, that he repented of his heresy at the end, forbade anyone to consent to Arian error, and wept continually for seven days for all the evil he had done against God, and then gave up the ghost. Whether or not that happened, the son who inherited was the one who had been sent to carry the false oath, and within three years of his accession he did in a council at Toledo what his brother had been destroyed for doing alone.

One church

Drafted from Collins, A History of Spain (2004)expansion

The Goths were Arians. The population of Spain was Catholic. That is the structural problem of every post-Roman kingdom in the west — the Vandals in Africa and the Ostrogoths in Italy had the same one — and this is the only kingdom that solved it.

In 589, at the Third Council of Toledo, King Reccared renounced Arianism and brought the Gothic ruling class into the church of the majority. Nothing else the kingdom did mattered as much.

What it removed and what it created

It removed the line between rulers and ruled at a stroke, and everything that had followed from it: separate hierarchies, separate churches, a nobility that could not intermarry with the provincial aristocracy without one side converting.

Compare the two kingdoms that did not. The Vandals in Africa enforced Arianism against the Catholic hierarchy with exiles and confiscations, and when Belisarius landed in 534 the African provinces did not fight for it. Ostrogothic Italy did not persecute and did not convert either, and Theodoric's execution of Boethius in 524 is the moment its Roman aristocracy discovered it had somewhere else to look. This kingdom converted, and survived another hundred and twenty-two years.

What it created was a state run through the church. The councils of Toledo — there were eighteen of them — legislated, judged, and from 633 claimed the right to choose the king. The bishops were the kingdom's civil service, its intellectual class and its constitutional check, and after 589 a Visigothic king's right to rule came from an assembly of them rather than from anything he asserted on his own.

And the cost

It also produced, within a generation, the most sustained legal campaign against Jews anywhere in early medieval Europe: forced baptism decreed and then partly retracted, children removed from parents, property confiscated, and a body of conciliar and royal legislation on the subject that runs through the seventh century and gets worse.

The connection is not incidental and does not belong in a footnote. A kingdom that has just abolished the confessional distinction between Goth and Roman, and defined itself as one Catholic people under one law, has produced a category of people who do not fit and has given itself an instrument — the council — for legislating about them. The great code that followed applies to everybody in the kingdom without distinction of descent, and the anti-Jewish laws are part of the same programme of making everybody the same.

Isidore

The other product of the arrangement is the reason this kingdom appears in histories of everything else.

Isidore of Seville, archbishop and the presiding figure at the Fourth Council of Toledo in 633, wrote the Etymologiae — twenty books attempting to organise all secular and religious knowledge by explaining the origins of the words for it. It is unreliable, enormously influential, and one of the most copied books in medieval Europe; something like a thousand manuscripts survive, a good many of them copied in Carolingian scriptoria two centuries later.

That is a strange thing for a kingdom to be measured by and there is no obvious way to weigh it. A state's territory, its army and its revenue can all be compared with another state's. A compendium that half of literate Europe used for eight hundred years cannot be, and it may be the most consequential thing the Visigoths produced.

The law and the election

Drafted from Collins, A History of Spain (2004)peak

In 654 Recceswinth promulgated the Liber Iudiciorum, a code of some five hundred laws applying to every inhabitant of the kingdom without distinction of descent. It abolished the separate legal treatment of Goths and Romans, which the earlier Visigothic legislation had maintained, and it went on being used in Spain — as the Fuero Juzgo in Castilian translation — for six hundred years after the kingdom that issued it had ceased to exist.

It is the most comprehensive body of legislation produced in post-Roman western Europe, and the contrast with the Lombard law of the same generation is exact and instructive.

Two codes, opposite answers

Rothari's Edict of 643, eleven years earlier, is three hundred and eighty-eight chapters of Lombard law in Latin, built on compensation tariffs, binding Lombards and leaving Romans to Roman law. It is personal law: the rule that applies to you follows your descent.

The Liber Iudiciorum is territorial: the rule that applies to you follows the ground you are standing on. It is built on Roman legal categories rather than on wergild schedules, and it treats the kingdom as a single jurisdiction.

Two post-Roman kingdoms, a decade apart, both legislating in Latin for a Germanic minority ruling a Roman majority, arriving at opposite answers to the question of whether law is a property of persons or of places. The Lombards kept two laws for two peoples. The Visigoths made one law for one territory, which is the answer every European state eventually arrived at, six hundred years early.

An elective monarchy that nobody wanted

The kingdom's rule of succession was election, and it comes from the Fourth Council of Toledo in 633, which legislated that a king should be chosen by the bishops and the magnates on the death of his predecessor, that nobody should seize the throne by force, and that anyone who did was excommunicate.

That is a written constitutional rule, and there are not many of them: Telipinu's edict, Esarhaddon's succession oath, the Hasmonean decree of 140, the Venetian ballot, the Golden Bull of 1356.

They have something in common. A state writes a succession law after being badly hurt by not having one, and this case is as clear as any. The council legislated in 633 after a century in which Visigothic kings were routinely murdered by their successors; a Frankish chronicler had already coined a phrase for it, the morbus Gothorum, the Gothic disease.

It did not work either. Kings spent their reigns trying to have their sons elected in their own lifetimes, which is the manoeuvre the rule exists to prevent; several succeeded and several were deposed for trying. Between 633 and 711 the throne changed hands eleven times and the transfer was contested more often than not.

The rule and the outcome are about as far apart here as they were in the Roman third century, when an empire with a formal procedure for making emperors made twenty of them in fifty years by acclamation at swordpoint and ratified every one.

Rulers made by a vote

It is also worth noticing how common this is in the post-Roman west. Venice elected its doge, Genoa its doge, the Holy Roman Empire its emperor, the Lombards their king, the Visigoths theirs. Whatever else the Latin west was, it is the part of the world where rulers were most often made by a body rather than by a birth — which is what happens downstream of an empire whose own emperors were, in constitutional theory, given their powers by a vote.

Book twelve

Drafted from Judicum), The Visigothic Code (Forum Judicum) (1910)contraction

The last book of the Visigothic Code is about heresy, and almost all of it is about Jews.

Its titles run: concerning the wickedness of the Jews; concerning the extirpation of the errors of the Jews. The provisions convert, in stages across the seventh century, from disabilities into a programme — forced baptism, confiscation of property, the seizure of children to be raised as Christians, the prohibition of observance, and the reduction of the unconverted to slavery. Those who returned to their own practice after baptism were to be stoned.

The most drastic of these were enacted under Ervig, in the last part of the seventh century.

What a law like that indicates

Legislation of this intensity is not a sign of a confident state.

The Visigothic monarchy of the seventh century was elective, and it was elective in the worst way: a crown that any noble with a following might get, and that the holder had therefore to defend continuously. Kings were deposed, tonsured, blinded and murdered with a regularity the councils of Toledo kept trying to legislate away, binding the nobility by oath to the reigning king and then, when he was removed, binding them to the next one.

An insecure elective monarchy needs the church, because the church consecrates the king and the oath is sworn on it. What it can offer the church in exchange is enforcement. Each round of anti-Jewish law comes out of a council of Toledo and is issued by a king who has just taken the throne and needs the bishops.

The argument the translator makes

Scott, writing in 1910, draws a causal line from this to 711, and it is worth setting out because it is a real argument even though he states it with more confidence than it can carry.

The Jews of the peninsula, he says, had long-standing relations with their co-religionists on the north African coast. The oppression, the confiscations, the seizure of children, the enforced conversion and the prospect of the annihilation of their community tightened those bonds, established an understanding between traders and the Muslim conquerors of the west, and thus invited the conquest of Spain. The harshest laws were passed under Ervig; less than forty years later the monarchy was gone.

That is a monocausal explanation of 711 and it should not be taken at face value. What can be said without straining is narrower and worse: a state facing an invasion had spent the preceding generation making a substantial, literate, commercially connected part of its own population into enemies by statute, and had left them no reason at all to prefer it to whatever came next.

Isidore of Seville, who was no friend of Judaism, said of the policy of driving Jews to baptism that it was non secundum scientiam — not according to knowledge. He said it early in the century, and nobody stopped.

Eighteen months

Drafted from Collins, A History of Spain (2004)end

In 711 an army under Tariq ibn Ziyad crossed from Tangier. King Roderic marched south, and somewhere near the Guadalete or the Barbate — the site is not known — his army was destroyed and he was killed. Within four years the peninsula was held, and a great deal of it had been taken by treaty rather than by fighting: the surrender agreement of Theodemir of Murcia survives in an Arabic text, guaranteeing his people their lives, property and churches in exchange for tribute.

Almost nothing ends this fast. The Sasanian empire took fifteen years to unwind; the Vandal kingdom went in a campaigning season and had been under sustained attack for months. This one had two and a half centuries of continuous existence, a law code, a functioning church and a tax system, and it was gone in eighteen months.

The explanations, and why Collins distrusts them

The traditional accounts are Arabic and Christian and both are late — the earliest substantial ones are a century or more after the events — and both are explanations for something the writer already knew had happened.

Treachery is the commonest: a faction of the previous king's family, dispossessed by Roderic, is said to have invited the invaders in and to have deserted at the battle. Count Julian of Ceuta opens the straits out of a private grievance. The Jews of the cities, driven to it by a century of legislation against them, are said to have opened the gates.

Decadence is the other: a kingdom weakened by luxury, doctrinal quarrels and a century of contested successions.

Collins's position is that most of these are literary and that the honest answer involves less narrative. A kingdom whose army was a levy on landholders — an obligation the seventh-century laws had to keep re-imposing, which tells you how well it was being met — loses one battle, loses the king and most of the senior nobility in it, and has no second army and no institution capable of raising one. Election by bishops and magnates meant there was no uncontested heir. And a great many cities preferred a treaty guaranteeing their property to a siege.

None of that is decadence. It is a state with a single point of failure, failing at it.

What was standing on the ground afterwards

A great deal of later history points back at this event. The Umayyad caliphate counted the conquest of Spain among its campaigns; the emirate of Cordoba begins with an Umayyad prince arriving in a province taken in 711; every Andalusi state that followed stood on ground this kingdom had held.

What claimed it afterwards

One polity claimed to be this one continued, and it is the kingdom of Asturias.

Asturias was founded in the northern mountains within a decade of 711 and from the ninth century presented itself as this kingdom continued: its chronicles run the Gothic king list straight into the Asturian one, its kings adopted Visigothic titles and law, and the reconquest of the peninsula was framed for seven hundred years as the restoration of a Gothic realm rather than as a conquest.

That is a claim asserted publicly and continuously, and it is the whole ideological basis of the reconquest: not a new Christian kingdom taking Muslim territory, but the Gothic kingdom recovering its own. A state that lost a peninsula in eighteen months was worth claiming to be for the next seven hundred years.

The code

Drafted from Judicum), The Visigothic Code (Forum Judicum) (1910)afterlife

The Visigothic kingdom ended in 711 and its law did not.

The Liber Iudiciorum, promulgated in the middle of the seventh century, was the kingdom's great achievement: a single territorial law for Goths and Romans alike, replacing the older arrangement in which each people carried its own, and covering procedure, inheritance, property, marriage, debt and criminal offences in something over five hundred titles.

After the conquest it went on being used. The Christian communities of the north and the Mozarabs living under Muslim rule kept it as their own law, and it was translated into Castilian in the thirteenth century as the Fuero Juzgo and granted as a municipal charter to reconquered towns.

Still being cited

Its English translator, writing in 1910, offers a set of concrete claims about how far the survival went, and they are more useful than his general enthusiasms, which run well past the evidence.

Some of the laws, he says, were still recognised as binding in the Iberian peninsula in his own day. He cites a case in the royal chancery court of Granada in which the Visigothic provisions were ordered to prevail in a dispute over inheritance, where the property of a dead monk was claimed both by his monastery and by his relatives.

And the laws were long used in southern France, where the capitularies of the early kings carry unmistakable internal evidence of having been drawn from the same source.

A law written for a Gothic king in Toledo was deciding a monastery's claim to a dead man's estate in a Spanish court more than a thousand years after the kingdom that made it had been overrun.

The other survival

The translation itself did something the legislators cannot have intended.

Scott is a partisan and his prose about the Code's beauties should be discounted, but on this point he is describing a real thing. The Castilian version — its coarseness, its ambiguities and its errors included — is one of the earliest substantial bodies of prose in the language, and it shows Spanish being made: corrupt Latin that had lost its classical grammar and Romance that had not yet found its own, abbreviated words, crabbed constructions, misquoted scripture, provincial peculiarities, archaic terms whose meaning was already lost, and words of Arabic derivation entering the language for the first time as the names of animals and objects.

What that adds up to

The Visigoths are usually remembered for the one afternoon at Guadalete and the eighteen months that followed, which is the shortest possible summary of three centuries.

What they actually left was a book. Not a claim, not a dynasty, not a church — the kingdom's Arian church was abandoned by its own king in 589 and the Catholic one it joined was there already. A book of laws, which outlived the state by twelve hundred years, was translated into a language that did not exist when it was written, and taught that language how to say things it had not had to say before.