Few places suffered more at the hands of the Vikings than Ireland. For the best part of 200 years the Vikings systematically milked Ireland of its people to supply the slave trade, yet, for all their military success they failed to conquer and settle in any territory besides a few fortified coastal enclaves. This is the conundrum of Viking Age Ireland; it was a land that looked weak but was in reality strong and resilient.

Superficially, Ireland must have looked to the Vikings like an easy target. There is no doubt that in England and Francia internal divisions worked to the Vikings’ advantage, and if there, why not even more so in Ireland, which was the most divided country in western Europe? Early medieval Ireland was a complex mosaic of around 150 local kingdoms and a dozen over-kingdoms. The local kingdoms or túatha were usually very small – often less than 100 square miles with populations of only a few thousand – and were defined as a ‘people’ or ‘community’, rather than as territorial units. The people of a túath were, in theory at least, an extended kinship group, or clan, and the king was the head of the senior lineage. The king (rí túathe) was responsible to his people for the fertility of their land and cattle, hence their prosperity: this was a legacy of pagan times when a king who failed to deliver would be sacrificed to the gods. Kings also had duties of lawmaking, judgement and leadership in war. In return all the free families of the túath owed the king taxes (paid in kind) and military service. Local kings might themselves owe tribute (usually in cattle), hospitality and military service to an over-king (ruirí), who in turn might owe it to a high king (rí ruirech). Over-kings, therefore, did not exercise direct rule outside their own túath, their power rested upon their ability to call on the resources and services of their client kings. The most powerful over-king of the day might be described as High King of Ireland (rí Érenn), but this was not really a formal institution with defined rules of succession. The relationships between kingdoms were not fixed. A local king with military ability and ambition could build a strong war band and use it to make himself an over-king by forcing other local kings to become his tributaries. Nevertheless, by the eighth century some stable dynasties of over-kings had emerged, the most powerful of which were the Northern and Southern Uí Néill dynasties of north-east Ulster and Meath respectively. To an outsider, early medieval Ireland would have appeared to be a chaotic and deeply divided country and, indeed, small-scale warfare between its kingdoms was endemic. Yet this highly decentralised political structure was to prove incredibly resilient, well able to absorb the shock of Viking invasions and constantly renew resistance.

In contrast to England and Francia where the Danes dominated, these raids were mainly the work of Norwegians, sailing to Ireland via the Northern Isles and the Hebrides. Viking activity in Ireland developed at first in much the same way as it did in England and Francia, beginning with small-scale hit-and-run raids on exposed coastal monasteries gradually escalating until the Vikings founded permanent bases and became a year-round presence plundering and captive-taking across the whole country. The first recorded Viking raids in Ireland took place in 795 when the same Viking band that sacked Iona sacked a monastery on Rechru, which may either be Lambay Island north of Dublin, or Rathlin Island off the northern Irish coast. In the 830s, larger fleets, numbering around sixty ships, began to arrive. Once its island monasteries had been plundered, Ireland’s wild and mountainous west coast, so similar to the west coast of Scotland, was generally shunned by the Vikings because of its poverty. The Vikings concentrated their efforts on the more fertile and densely populated east coast and the great midland plain. In 836, a fleet sailed for the first time far inland along Ireland’s longest river, the Shannon, and sacked the wealthy monasteries of Clonmacnoise and Clonfert. The following year, a Viking fleet sailed from Donegal Bay into Lough Erne to plunder monasteries around its shores. Another sacked the monastery of Áth Cliath – on the site of modern Dublin – while a third army ravaged on the Boyne, and a fourth was on the Shannon again. Nowhere was safe: ‘the sea cast floods of foreigners into Ireland, so there was not a point thereof that was without a fleet’, wrote one chronicler.

Although the Irish often fought fiercely, the Vikings’ advantage of mobility meant that they often escaped unchallenged: the saints slept and did not protect their monasteries. Monks trembled in their cells and prayed for bad weather to keep the Vikings off the seas. As kings were rarely inclined to help their rivals, the Vikings often benefited from the divisions between the Irish kingdoms. Indeed, most kings took a thoroughly pragmatic view of the Vikings, treating them as just another element in their country’s complex political geography, often welcoming them as allies who could help weaken a rival kingdom. Some bands of Irishmen took advantage of the disorder created by the Vikings to go plundering themselves ‘in the manner of the heathens’. One such band was destroyed by Máel Sechnaill mac Máele Ruanaid (r. 845 – 62), the powerful Southern Uí Néill high king of Meath, in 847.

The first longphuirt

In 839 there was a step-change in Viking activity. A Viking fleet sailed up the River Bann into Lough Neagh. Instead of plundering and leaving, the Vikings built a fortified ship camp on the lakeshore, which they used as a base to plunder the heart of Ulster for three successive summers. This was the first of many such bases – known as longphuirt by the Irish – that Viking armies were to build in Ireland over the next few years as they intensified their raids. The foundation of the longphuirt subtly changed the dynamics of Viking activity in Ireland. The Vikings were now a permanent presence in Ireland and could raid all year round, but at the same time, they lost some of their mobility, making them more vulnerable to Irish counterattack.

The leader of the fleet on Lough Neagh was a warlord who the Irish called Turgeis, that is probably Thórgestr or Thórgils in Old Norse. Turgeis’ origins are not known, but he may have come from the Hebrides as he had as his allies the Gall-Gaedhil, those ‘foreign Gaels’ who were the product of marriages between Norse settlers and the local Gaelic-speaking population. Turgeis’ greatest coup was plundering St Patrick’s monastery at Armagh three times in 840: after his final attack he burned it down for good measure. Armagh was an especially rich prize; apart from its precious reliquaries and sacred vessels, many Irish kings had their royal treasuries there, hoping that they would enjoy the protection of its powerful patron saint. It would not only have been monks who suffered in these attacks. Armagh was surrounded by a small town of craftsmen, merchants, estate managers and others who serviced the needs of this most prestigious of all Irish ecclesiastical centres. Turgeis’ activities are uncertain for the next few years, but he is thought by some historians to have been the leader of the Vikings who in 841 founded what would become the most successful of all the longphuirt at Dublin. In 844, Turgeis led his fleet up the River Shannon as far as Lough Ree, where he built another longphort from which he plundered widely in the midlands. The following year, in the first serious reverse suffered by the Vikings in Ireland, he was captured by Máel Sechnaill, who drowned him in Lough Owel in County Westmeath.

Turgeis’ reputation grew with the telling and after his death he became a symbol of everything that was wicked about the Vikings. In the colourful but unreliable twelfth-century history of Ireland’s Viking wars, Cogadh Gaedhel re Gallaibh (‘The War of the Irish with the Foreigners’), Turgeis has become the king of all the Vikings in Ireland, bent on conquering the whole island. This Turgeis is a militant pagan who expels the abbot from Armagh and sets himself up as a pagan high priest. His wife Ota (probably Auðr) is just as bad, performing acts of witchcraft on the altar of the abbey at Clonmacnoise. This story might not be wholly improbable as Ota may have been a völva, a Viking seeress with powers to predict the future. According to the Welsh churchman Gerald of Wales, who travelled in Ireland during the 1180s, Turgeis actually conquered Ireland but was lured to his death by his weakness for women. Turgeis took a fancy to Máel Sechnaill’s daughter. The king, ‘hiding his hatred in his heart’, agreed to hand her over to Turgeis on an island in Lough Owel along with fifteen other beautiful girls. Turgeis was delighted and went to the rendezvous with fifteen of his leading warriors, all of them expecting amorous encounters. But Máel Sechnaill had laid a trap for them. His daughter was waiting for Turgeis on the island not with fifteen girls but with fifteen hand-picked young men, all clean shaven and dressed in women’s clothing, under which they carried knives. Turgeis and his unsuspecting warriors were stabbed to death ‘in the midst of their embraces’. Gerald probably recorded the story not to flatter the Irish for their cunning but because it chimed comfortably with his own prejudices: he regarded the Irish as a thoroughly deceitful and untrustworthy bunch who always negotiated in bad faith.

More reverses for the Vikings followed. In 848 the Irish won four major battles against the Vikings, killing over 2,000 of them in the process, according the Annals of Ulster. Irish annalists described these battle casualties as ‘heads’: Irish warriors still practiced the ancient Celtic custom of taking enemy heads as war trophies and rarely took prisoners. Then, in 849, Máel Sechnaill captured and plundered Dublin. Discouraged by their defeats, many Vikings left to seek easier pickings in Francia. The Norwegians suffered another blow in 851when a large force of Danish Vikings expelled them from Dublin. The following year the Norwegians suffered another crushing defeat by the Danes in a three-day battle at Carlingford Lough in County Down. The Danish intervention in Ireland was short-lived. In 853 two brothers, Olaf and Ivar, recaptured Dublin for the Norwegians and expelled the Danes.

The kingdom of Dublin

The arrival of Olaf and Ivar at Dublin in 853 was a decisive moment in Ireland’s Viking Age. Olaf and Ivar (who are called Amláib and Ímhar in Irish annals) became the first kings of Dublin and under their rule it developed from a rough ship-camp into the dominant Viking power centre of the whole Irish Sea area. Irish sources describe Olaf and Ivar as sons of King Gofraid of Lochlann, which is the usual Gaelic name for Norway, but their origins remain uncertain. Most modern historians identify Olaf with Olaf the White, a king of Dublin who features in Icelandic saga traditions. Attempts to identify Ivar with the legendary Viking Ivar the Boneless are unconvincing: Ivar the Boneless’s father was the equally legendary Viking Ragnar Lodbrok who, if he existed at all, was most likely a Dane. What is more certain is that the descendants of Olaf and Ivar, known to the Irish as the Uí Ímair, would dominate the Irish Sea for the next 200 years.

There is not enough evidence about the careers of Turgeis and Tomrair to be sure of their motives: did they aspire to found Viking states in Ireland or were they really just out for the plunder? It is clear, however, that Olaf and Ivar were trying to create a kingdom for themselves because their first actions were to impose tribute on all the Viking armies operating in Ireland. It is hard to work out from the Irish annals exactly how many of these there were but there must have been at least three or four. In their efforts to build a secure power base, the brothers took full advantage of the complex political rivalries of the Irish kingdoms. In 859 Olaf and Ivar allied with Cerball mac Dúnlainge (r. 842 – 880), king of Osraige, against his overlord Máel Sechnaill. According to saga traditions, the alliance was sealed by a marriage between Olaf and one of Cerball’s daughters. A Christian king is unlikely to have married his daughters to pagans, so, if the tradition is true, it is likely that Olaf had at least been baptised. In 858, Ivar and Cerball campaigned together in Leinster, and in Munster against the Gall-Gaedhil. The next year Olaf, Ivar and Cerball together invaded Máel Sechnaill’s kingdom of Meath. After Cerball came to terms with Máel Sechnaill, he dropped his Norse allies. Olaf and Ivar soon found a new ally in Áed Finnliath (c. 855 – 79), the northern Uí Néill king of Ulster. Together they plundered Máel Sechnaill’s kingdom in 861 and 862. After Máel Sechnaill’s death in 862, Olaf and Ivar switched to supporting his successor Lorcán against Áed. The brothers did Lorcán’s standing no good at all when, in 863, they dug open the great Neolithic burial mounds at Knowth on the River Boyne to look for treasure. Although pagan in origin, these ancient mounds were rich in mythological significance for the Irish and this desecration was thought to be shocking behaviour even by the Viking’s low standards. The following year Áed captured the discredited Lorcán, blinded him and forced him to abdicate.

Olaf and his brothers had now run out of willing allies in Ireland and, in 866, they took their fleet across the Irish Sea to raid Pictland in alliance with the Gall-Gaedhil. Áed, now high king, took advantage of their absence to plunder and destroy all the Viking longphuirt in Ulster. After a victory over the Vikings on Lough Foyle, Áed took 240 heads home as trophies. The limited extent of Viking territorial control was starkly demonstrated in 867 when Áed’s ally Cennétig king of Loigis, destroyed Olaf’s border fortress at Clondalkin just 5 miles from Dublin, which he then went on to plunder. Olaf now allied with the southern Uí Néill and Leinster against Áed. Áed crushed the alliance at the Battle of Killineer (Co. Louth) in 868: among the dead was one of Olaf’s sons. Olaf struck back at Áed in 869, brutally sacking Armagh and leading off 1,000 captives for the slave markets. This was a severe blow to Áed’s prestige – he was supposed to be the monastery’s protector. After this success, Olaf and Ivar crossed the Irish Sea to Strathclyde and laid siege to its capital, Alt Clut, on the summit of Dumbarton Rock, overlooking the River Clyde. Alt Clut fell after four months and the brothers returned to Dublin with a hoard of treasure. They went back to Strathclyde for more the following year and this time returned ‘with a great prey of Angles, Britons and Picts’. Olaf and Ivar were back plundering in Meath in 872, but in the next year Ivar died of ‘a sudden, horrible disease’. Olaf survived until 874 or 875: he was killed in battle with Constantine I of Scotland at Dollar in Clackmannanshire.

The deaths of Ivar and Olaf began what the Cogadh Gaedhel re Gallaibh dubbed the ‘Forty Years’ Rest’, a long period of reduced Viking activity in Ireland that lasted until 914. Deprived of the strong military leadership provided by Olaf and Ivar, Dublin became politically unstable under a succession of short-lived successors. Olaf’s first successor as king of Dublin, his son Oystín (Eystein), lasted barely a year: he was killed when Dublin was captured by a Danish Viking who Irish annalists called Alband. Alband is most likely to have been Halfdan, the Danish king of York. Áed Finnliath came to the rescue of his Viking allies, quickly expelling Alband and placing Ivar’s son Bárðr on the throne. Alband returned to Ireland in 877, but was killed fighting the Dublin Vikings at Strangford Lough. However, his dream of uniting Dublin and York into a trans-Irish Sea kingdom survived. Bárðr died in 881 and was followed by six short-lived kings, none of whom was able to arrest the kingdom’s decline. In 902, Cerball mac Muirecáin, king of Leinster and Máel Finnia of Brega launched a co-ordinated pincer attack on Dublin from the north and south, forcing the Norse to flee for their ships after a fierce battle. The refugees fled mainly to North Wales and north-west England. Ireland’s first Viking Age was over.

From longphort to town

Most of the Vikings’ longphuirt were either abandoned, or were destroyed by the Irish, after relatively short periods of occupation. Dublin was one of a small group of longphuirt, which also included Wexford, Waterford, Cork and Limerick, which developed into permanent towns. These longphuirt all had in common good tidal harbours. The exact location of the original Viking longphort at Dublin now lies buried beneath later buildings. This has necessarily limited archaeological investigation of the city’s origins to rescue excavations on sites that have been temporarily cleared for redevelopment. Evidence for early Viking occupation, including warrior burials, buildings, ship rivets and a possible defensive rampart, excavated from sites at Ship Street Great and South Great George’s Street, suggest that the longphort was probably in the area where Dublin Castle now stands, close to the Dubhlinn, the ‘black pool’ from which the city got its English name. This was a now-vanished tidal pool at the confluence of the River Liffey and its small tributary the Poddle. Dublin was already a place of some importance before the longphort was built as a monastic centre and the site of the lowest ford across the River Liffey: its Gaelic name Áth Cliath means ‘the ford of the hurdles’. This ford made Dublin a natural focus of overland routes and, with its good harbour and short sailing distances to Wales, north-west England, Galloway and the Isle of Man, it was ideally situated to become a successful port and trading centre. The same geographical advantages also made Dublin an ideal base for raiding, not only in eastern Ireland but around the whole Irish Sea region. No other longphort in Ireland had the same combination of advantages: it was almost inevitable that Dublin would become Ireland’s dominant Viking centre.

Early Dublin was probably similar to the well-preserved longphort at Linn Duchaill, about 40 miles further north, near the village of Annagassan in County Louth. Founded in the same year as Dublin, this longphort was built on the site of a minor monastery on the banks of the River Glyde, close to its estuary into the Irish Sea. The Vikings occupied the longphort until 891, when the Irish expelled them. Vikings reoccupied the site c. 914 only for it to be abandoned for good in 927. The site has been open farmland ever since so, unlike Dublin, this longphort’s remains have seen little disturbance. Covering about 40 acres (16 hectares), the longphort at Linn Duachaill was large enough to accommodate an army that was several thousand strong. A rampart and ditch, ¾ of a mile long, protected the landward side of the fort and there was a small citadel on higher ground within the fort. Excavations yielded large numbers of ships’ rivets, testifying to ship repair and perhaps shipbuilding on the site. Pieces of hacksilver and the remains of scales show that loot was divided up here and an iron slave chain dredged from the river is evidence of slave raiding. A shuttle and spindle whorl provide evidence of spinning and weaving in the fort. As these were not occupations for Viking warriors, women must have lived there. Geophysical surveys suggest that the waterfront was densely built-up but this has not yet been confirmed by excavations. Linn Duachaill did not have the good harbour that Dublin had, and it was that which probably prevented it ever developing into a permanent town.


The Viking slave trade

Archaeological evidence indicates that by 902 Dublin had begun to outgrow the longphort and become a true town rather than an armed camp. Significantly, following the Irish conquest in that year, Dublin was not abandoned: there is clear evidence of continuity of settlement through to its recapture by the Norse in 917. That there was an exodus of Scandinavians from Ireland at that time is not in doubt, so this is probably evidence that Dublin had a significant Irish population living alongside the Norse and that they were allowed to remain: they may even have been the majority because genetic studies have found scant evidence of Scandinavian DNA in the modern Irish population.

Dublin owed its transformation to a town to trade. Pre-Viking Ireland did not play a large part in international trade so it had no trading towns to compare with the likes of Dorestad or York. Coinage was not used either. Ireland was not poor, however. The hoards of magnificent gold and silver liturgical vessels from Ardagh and Derrynaflan stand testimony to the wealth of Ireland’s monasteries in the early Middle Ages. Major Irish monasteries like Armagh or Clonmacnoise were much more than communities of monks, they were also centres of political power and economic activity. Secular communities of craftsmen and merchants grew up around the more important monasteries and by the eighth century a few were becoming small towns. Kings, seeking the authority and safety that close association with the saints was believed to confer, often had residences, treasuries and garrisons in these monastic towns. All of this was more than enough to justify the Vikings’ attentions, but their main interest was in Ireland’s people.

Crude estimates based on a count of known settlements suggest that Ireland’s population was about half a million when the Viking Age began. Thanks to the country’s mild winters, cool summers and reliable rainfall, grass grew all year round so cattle and sheep did not have to be kept inside during the winter. The Irish did not bother to gather hay in the summer as it was so rarely necessary. Despite occasional famines caused by cattle epidemics and severe weather, the Irish population was generally well nourished and very few people were desperately poor. The Vikings rounded up these people in their thousands to be ransomed or sold as slaves according to their wealth and status. Slavery was rare in pre-Viking Ireland – it was used mainly as a form of debt bondage – so there was no slave trade. Plundering in wars between the Irish was usually confined to cattle rustling, so Viking slaving added a new form of suffering to the experience of warfare. Perhaps inevitably, Irish kings soon began to take captives during their wars and sell them to the Vikings. Irish captives who were not lucky enough to be ransomed by their relatives could expect to be sold abroad. Anglo-Saxon England and the Frankish kingdoms both had active slave trades but most Irish captives probably finished up in Scandinavia or the Moorish kingdoms in Spain and North Africa. Through developing the slave trade, the Vikings drew Ireland into fuller participation in the international trade networks. This is usually presented as one of the positive impacts that the Vikings had on Ireland, but it is unlikely that their victims were quite so sanguine about it.

We know enough about the horrors of the trans-Atlantic slave trade of the eighteenth century to guess at the human misery Viking slaving must have caused. Its economic impact is harder to estimate but it is likely that Vikings targeted the young and healthy rather than infants and the elderly. The kidnapping and breaking-up of communities of learned monks must have had a far more serious impact on Ireland’s flourishing monastic culture than ever the destruction of books, sacred vessels and buildings did. As mere commodities, the voices of slaves are rarely heard in the historical record, but two remarkable accounts have survived about the experiences of Irishmen who were captured by Viking slavers. One relates to a Leinsterman called Findan whose sister was captured by Viking raiders some time around the middle of the ninth century. Findan’s father sent him to the Vikings to arrange his sister’s ransom, but they immediately clapped him in irons and carried him off to their ship too. After keeping him without food and water for two days, the Vikings discussed what to do with him. Luckily, his captors decided that it was wrong to capture people who had come to pay ransom, no doubt because it would discourage others from doing so, and they let him, and presumably his sister, go. A short time afterwards Findan got caught up in another Viking raid but evaded capture by hiding behind the door of a hut. For Findan it was third time unlucky, because in his next encounter with the Vikings he was taken prisoner and sold into slavery. After changing hands several times, Findan finished up on a ship bound for Scandinavia. Findan gained his owner’s confidence by helping the crew fight off some pirates and he was released from his leg irons. When his owner made a stop-over in Orkney, Findan seized the opportunity to jump ship and escape. Findan eventually made his way to Rome as a pilgrim and ended his life as a monk at the monastery of Rheinau in Switzerland: one of his fellow monks recorded his life story shortly after he died.

The second story concerns an Irishman called Murchad, a married man with a daughter, who was captured by Vikings and taken to Northumbria, where he was sold as a slave to a nunnery, with comical consequences. After he had seduced several of the nuns and turned the nunnery into a brothel, Murchad was expelled and cast adrift on the sea in a boat without oars or a sail as a punishment for his impiety. Murchad was rescued by Vikings, who took him to Germany and sold him to a roguish widow, who paid for him with counterfeit money. Murchad seduced her too, of course. After many more adventures, Murchad eventually returned to Ireland, was reunited with his family, and took up a career teaching Latin grammar. How much real history there is in this tale is hard to tell; perhaps it is really about making the best of hard times. It is unlikely that many captives were as lucky as Findan and Murchad but neither is it likely that all came to bad ends: most of the thousands of Irish slaves who were taken to Iceland later in the ninth century were eventually freed and became tenant farmers, for example.

Division is strength

By the early tenth century, Vikings had conquered and colonised substantial parts of England, Scotland and Francia, as well as the uninhabited Faeroe Islands and Iceland. Yet for all the fury of their onslaught, in Ireland the Vikings had not even been able to retain a toehold. Appearances can be deceptive. Ireland’s divisions might have been a handicap in combating plundering raids but they also made it all but impossible for the Vikings to conquer and hold territory. On the face of it, it would have seemed that Ireland’s disunity should have made it more vulnerable to conquest by the Vikings than England, which was divided into only four powerful centralised kingdoms. In fact the opposite was true. In early medieval Europe it was always the centralised kingdoms that got conquered most easily. After the ‘Great Army’ of Danish Vikings invaded England in 865, the kingdoms of Northumbria and East Anglia both collapsed as soon as their kings had been killed in battle. Mercia too collapsed when its king decided he would prefer not to get killed and fled the country. Only Wessex survived to prevent England becoming Daneland. The centralised nature of the Anglo-Saxon kingdoms meant that it was relatively easy for the Vikings to destroy the small ruling class and take over; just one battle might do the trick, as it did, more or less, in 1066. Little trouble would then be expected from the leaderless peasantry. Ireland, however, had dozens of kings and even more lineages from which new kings could be chosen. No victory, therefore, could ever have the decisive knockout effect it could in a country like England. Nor was there much chance of a lasting peace agreement with so many kings to negotiate with because what one agreed was not binding on the others.

The military resources of the Irish should not be underestimated either. Most Irish local kingdoms could raise armies of around 300 men. This was inadequate to deal with anything but a small Viking raiding party, but there were a great many local kingdoms. Local kings owed military service to their over-kings, so an over-king who could enforce the obedience of his vassals could raise a very large army indeed. However, in a clash of shield walls an Irish warrior was no match for a well-equipped Viking. The Irish fought almost naked without armour or iron helmets, armed with spears and using only bucklers (small round shields) for protection. The Irish recognised the superiority of the Viking warrior and they usually avoided formal battle in favour of irregular tactics, harassing raiding parties and wearing them down with sudden ambushes before melting away into the woods and bogs. In this kind of fighting, their lack of armour was an advantage to the Irish, making them more agile than a mail-clad Viking. A weary Viking raiding party returning home burdened with loot, captives and stolen cattle would have been particularly vulnerable to these tactics.

The Irish countryside was scattered with as many as 50,000 ringforts, but these were probably less of a hindrance to Viking raiders than Ireland’s warriors. Ringforts varied in size according to the status of their inhabitants. An over-king might have a substantial stone structure like the Grianán of Aileach in Donegal, a stronghold of the Northern Uí Néill dynasty. Built in the eighth century, the Grianán’s 15 foot (4.5 m) thick stone walls enclose an area 75 feet (23 m) in diameter. But although it is an impressive structure, modern experiments have shown that it would not have been at all easy to defend so it may have been built mainly as a ceremonial centre rather than to withstand a siege. Local kings and aristocrats had more modest forts, sometimes as little as 30 feet (9 m) in diameter, with earth ramparts and a palisade, containing its owner’s house and ancillary buildings. The ramparts of these small forts were primarily markers of status, for they were barely adequate for keeping livestock in, never mind keeping raiders out. More secure were crannogs, high status dwellings built on artificial islands in the middle of lakes. Communal fortifications like the English burhs, intended to provide refuges for the general population, were unknown.

During the course of the Viking Age, monks began to provide their monasteries with tall, slender, stone round towers. These were primarily used as bell towers and treasuries but they were also refuges against Viking raids. Over eighty round towers are known to have been built: the tallest surviving round tower, at Kilmacduagh in Galway, is 113 feet (34.5 m) tall. The towers’ entrances were set well above ground level so that they could only be entered with a ladder. The entrances of some towers show signs of fire damage, which is likely a result of Viking attacks. Having no source of water, or battlements from which to fight off attackers, round towers could not withstand a long siege but a small Viking raiding party could not really afford any delay.

The Vikings return

During Ireland’s ‘Forty Years’ Rest’, the bulk of Viking forces were busying themselves plundering England and Francia. By the first decade of the tenth century, the English and Franks were finally getting the measure of the Vikings so Ireland once more began to look attractive to them. In 914 Ragnald, a grandson of Ivar I, appeared in the Irish Sea and defeated a rival Viking leader in a sea battle off the coast of the Isle of Man before going to set up a longphort at Waterford in south-east Ireland. The Vikings were back and with a vengeance. In 917, Ragnald’s brother Sihtric Cáech (‘squinty’) recaptured Dublin and in 919 smashed an Irish counter-attack at Islandbridge, killing the Úi Néill High King Niall Glúndubh and five other kings. In 922, Tomar mac Ailche (Thormódr Helgason) re-established Viking occupation at Limerick and around the same time other Viking leaders established themselves at Cork and Wexford. As was the case in the ninth century, the Vikings made no extensive territorial conquests or settlements outside their heavily fortified towns. Dublin came to control the most extensive territory: known as Dyflinnarskíri or ‘Dublinshire’, it extended along the coast from Wicklow (Vikinglo) in the south to Skerries (from Old Norse sker meaning a ‘reef’) in the north, and as far inland as Leixlip (Old Norse lax hlaup meaning ‘salmon leap’) on the River Liffey. A dearth of Norse place-names in the countryside of Dublinshire supports the conclusion that there was little or no Viking settlement outside Dublin and its immediate environs.

The history of the revived Viking kingdom of Dublin is frequently entangled with that of the Viking kingdom of York across the Irish Sea. While the Norse had been exiled from Ireland, Ragnald had briefly held power in York and now he wanted it back. Using Dublin as a base to campaign in northern England, Ragnald recaptured York in 919. York must have seemed a greater prize than Dublin because when Ragnald died in 921, Sihtric gave up the kingship of Dublin to another brother, Guthfrith, and took up the kingship of York. An aggressive ruler, Guthfrith immediately launched a furious campaign of plundering and slaving raids against the Irish, culminating in a curiously respectful sack of Armagh in November. Guthfrith spared the monks, the sick and the monastic buildings, ‘save for a few dwellings which were burned through carelessness.’ It may be that Guthfrith was a Christian. If so, Guthfrith’s show of respect for St Patrick did him no good because he was intercepted on his way home by Muirchertach of the Leather Cloaks (r. 919 – 43), the king of the Northern Uí Néill, and heavily defeated. This set the tone for Ireland’s second Viking Age: the days when Vikings might criss-cross Ireland without meeting serious opposition were gone. Muirchertach won another victory over the Dublin Vikings at Carlingford Lough in 925, when 200 of them were captured and beheaded, and the following year he killed Guthfrith’s son Alpthann (Halfdan) in another battle at Linn Duchaill. Muirchertach besieged the survivors in the longphort there until Guthfrith brought an army north from Dublin to rescue them. The longphort was afterwards permanently abandoned.

What had changed since Ireland’s first Viking Age? The shock of Viking raiding had forced change upon the Irish. Irish society became increasingly militarised and those kings who offered the most effective military leadership against the Vikings enhanced their status and power and, as they tightened their grip over their sub-kings, they could raise larger armies and enhance their power even more. It was the same virtuous circle of success that was driving political centralisation and state formation in contemporary Scandinavia. Irish kingship was gradually becoming more territorial and many local kings found themselves reduced to the status of local chieftains. At the same time, the Irish had learned from the Vikings, making greater use of swords and axes in battle. Though they still lacked armour, this went some way to evening the odds on the battlefield. War was also waged with a new ruthlessness, against both the Vikings and other Irish kingdoms. Ravaging and burning had been rare before the Viking Age, but now Irish kings used it routinely as a weapon against their foes irrespective of whether they were Irish or Norse.

After Sihtric’s death in 927, Guthfrith went to York, whether to claim the throne for himself or to support his brother’s son Olaf Cuarán is not known. Both were quickly expelled by Æthelstan of Wessex. Guthfrith returned to lay siege to York, but was forced to surrender to Æthelstan, who allowed him to return to Dublin, which he ruled until his death in 934. Guthfrith’s son and successor, Olaf Guthfrithsson, established dominance over all the Norse in Ireland when he defeated the Limerick Vikings in a naval battle on Lough Ree in 937. It was in the same year that he allied with the Scots and the Welsh of Strathclyde in another attempt to win the kingdom of York only to be defeated by Æthelstan at the Battle of Brunanburh (see p. 124). Muirchertach sacked Dublin the following year, taking advantage of its weakness after Olaf’s defeat in England. However, Æthelstan’s death in 939 finally gave Olaf the chance to seize York and unite it with Dublin in a single kingdom. Olaf did not enjoy his success for long: he died shortly after raiding the Northumbrian monasteries at Tyninghame and Auldhame in 941, a victim, it was said, of divine displeasure. A tenth-century Viking burial discovered in the monastic cemetery at Auldhame almost certainly belongs to a high-status Viking who was involved in these raids. It has been speculated that the burial was even that of Olaf himself. As Olaf was married to the daughter of King Constantine II of Scotland he must have been at least a nominal Christian. The king might therefore have been buried on consecrated ground as a posthumous act of penance. Following Olaf’s death his cousin Olaf Cuarán became king of York, while his brother Blácaire succeeded him at Dublin.

Blácaire was an active raider. On 26 February 943 he defeated and killed Muirchertach at the Battle of Glas Liatháin and five days later sacked Armagh. Muirchertach’s death was mourned by the Irish, the Annals of Ulster described him as ‘the Hector of the western world’ and lamented that his death had left the ‘land of the Irish orphaned’. Irish retaliation was swift. The following year, the newly acknowledged High King Congalach Cnogba captured and burned Dublin, carrying away a vast amount of booty. Four hundred Vikings were said to have been killed in the fighting and Blácaire fled into exile. In his absence, Congalach installed Olaf Cuarán, recently expelled from York by the English, as king of Dublin. Olaf’s dependence on Congalach was such that when the pair were defeated by a rival for the high kingship in 947, Blácaire was able to depose him and reclaim his throne. After his death in battle against Congalach in 948, Blácaire was succeeded by his cousin Godfred, another son of Sihtric Cáech. In 951 Godfred led an enormously successful expedition in the Irish midlands, plundering half a dozen monasteries including Kells. According to the Annals of Ulster, ‘three thousand men or more were taken captive and a great spoil of cattle and horses and gold and silver was taken away’. Divine vengeance followed swiftly, of course. A severe epidemic, described in the annals as dysentery and leprosy, broke out in Dublin on Godfred’s return and the king was one of its victims.

While Godfred had been plundering in Ireland his brother Olaf had briefly regained control of York before being expelled by the Norwegian Erik Bloodaxe in 952. Olaf now succeeded as king of Dublin but the dream of uniting Dublin and York was dead. The Dublin Vikings would never be a power in England again. It is doubtful that a Dublin-York axis was ever really viable in the long term. York is much more remote from Dublin than a casual glance at a map would suggest. As York could only be reached by ship from the North Sea, sailing there from Dublin involved a long, dangerous and time-consuming voyage around the north of Britain. The only alternative would have been to sail from Dublin to north-west England and then trek across the Pennine Hills to York. However, it is far from clear how much, if any, control the kings of York actually exercised west of the Pennines. And, fighting off the English and the Irish at the same time must have been way beyond the resources of the Dublin Vikings.

Olaf was not a peaceable king but neither was he a traditional freebooting Viking, as he rarely raided unless he was acting in alliance with an Irish king. Olaf was also closely linked to Irish dynasties by marriage – made possible by his baptism in England as part of a peace deal with king Edmund in 943. Olaf’s first wife was Dúnlaith, the sister of the high king Domnall ua Néill (r. 956 – 80) and, after her death, he married Gormflaith, daughter of Murchad mac Finn, king of Leinster. Olaf seems to have gained little, if any, political advantage from his marriages because his reign was dominated by conflicts with Domnall and with successive kings of Leinster (some of whom Olaf held hostage in Dublin). On Domnall’s death in 980, Dúnlaith’s son by an earlier marriage, Máel Sechnaill mac Domnall, the king of Meath, succeeded his uncle as high king. Máel Sechnaill clearly had no love lost for his stepfather as he had begun his reign as king of Meath in 975 with an attack on Dublin, in which he burned ‘Thor’s Wood’ (a pagan sacred grove) outside the city. Shortly after becoming high king, Máel Sechnaill heavily defeated a force of Vikings from Dublin, Man and the Hebrides in battle at the Hill of Tara, the traditional inauguration place of the high kings. Máel Sechnaill followed up his victory by laying siege to Dublin, which surrendered after three days. Máel Sechnaill imposed a heavy tribute on the citizens and deposed Olaf, who went into retirement as a monk on Iona, where he died soon afterwards. In his place, Máel Sechnaill appointed his half-brother Jarnkné (‘iron knee’) (r. 980 – 9), Olaf’s son by Dúnlaith, as tributary king. There was no disguising Dublin’s loss of independence.


The Vikings in Wales

One side effect of the strength of Irish resistance was to increase Viking interest in Wales. At its closest points, Wales was only a day’s sail away from Dublin, Waterford and Wexford, and from the Viking colony in the Isle of Man, but despite this had so far suffered relatively little from Viking raids. A combination of strong military rulers such as Rhodri Mawr (r. 844 – 78) of Gwynedd, difficult mountainous terrain, and Wales’ poverty compared to England, Ireland and Francia, seem to have deterred any major Viking invasions in the ninth century. Only a dozen Viking raids are recorded in the period 793 – 920, compared to over 130 in Ireland in the same period. This was fewer than the number of English invasions of Wales in the same period. Place-name evidence points to areas of Viking settlement in the south-west, in Pembrokeshire and Gower, but, as they are undocumented, it is not known when they were made. There was also a small area of Viking settlement in the far north-east, modern Flintshire, most probably by refugees from Dublin following its capture by the Irish in 902. This was probably overspill from the successful Viking colony a few miles away across the estuary of the River Dee in Wirral.

In the first half of the tenth century, Wales was dominated by Hywel Dda (r. 915-50), the king of Deheubarth in the south-west. During his long reign Hywel came close to uniting all of Wales under his rule but his death in 950 was followed by a civil war and the break-up of his dominion. This was a signal to Vikings based in Ireland, the Isle of Man and the Hebrides to launch a wave of attacks on Wales. The area most exposed to Viking raiding was the large and fertile island of Anglesey off the coast of North Wales, which lay only 70 miles due east of Dublin and just 45 miles south of the Isle of Man: raids are recorded in 961, 971, 972, 979, 980, 987 and 993. Another place hit hard was St David’s monastery on the Pembrokeshire coast, Wales’ most important ecclesiastical centre. Founded c. 500 by St David, the monastery became the seat of the archbishops of Wales in 519. Only 60 miles from Wexford, St David’s was first sacked by Vikings in 967, then again in 982, 988 and 998, when they killed archbishop Morgeneu. St David’s would be sacked at least another six times before the end of the eleventh century. In 989 the raids had become so bad that King Maredudd of Deheubarth paid tribute to the Vikings at the rate of one silver penny for each of his subjects. Viking raids declined quickly after 1000, perhaps because the Viking towns in Ireland had come under the control of Irish rulers, but raids from the Hebrides and Orkney continued into the twelfth century. Vikings from Ireland also continued to come to Wales, but they did so mainly as mercenaries signing on with Welsh kings to fight in their wars with one another and with the English.

The Rock of Cashel

The end of Ireland’s Viking Age is traditionally associated with the rise of the O’Brien (Ua Briain) dynasty of Munster, and of its greatest king Brian Boru (r. 976 – 1014) in particular. Brian’s career certainly had an epic quality about it. Brian was a younger son of Cennétig mac Lorcáin (d. 951), king of the Dál Cais, whose kingdom, which was roughly equivalent to modern County Clare, was subject to the kings of Munster. As a younger son Brian probably never expected to rule and his early life was spent in the shadow of his elder brother Mathgamain. Even Brian’s date of birth is uncertain. Some Irish sources claim that he was eighty-eight when he died in 1014, which would mean he was born in 926 or 927, but other sources give dates as early as 923 and as late as 942. Brian’s first experience of war came in 967 when he fought alongside his brother at the Battle of Sulcoit against Ivar, king of the Limerick Vikings. The following year the brothers captured and sacked Limerick, executing all male prisoners of fighting age. The rest were sold as slaves. Ivar, however, escaped to Britain and in 969 he returned with a new fleet and regained control of Limerick only to be expelled again by the Dal Cais in 972.

Probably in 970, Mathgamain expelled his nominal overlord, Máel Muad the king of Munster, from his stronghold on the Rock of Cashel. The rock is a natural fortress, a craggy limestone hill rising abruptly and offering a magnificent view over the fertile plains of County Tipperary. The rock is now crowned by the ruins of a medieval cathedral and one of Ireland’s tallest surviving round towers, so little evidence of earlier structures survives. In legend, Máel Muad’s ancestor Conall Corc made Cashel the capital of Munster after two swineherds told him of a vision in which an angel prophesised that whoever was the first to light a bonfire on the rock would win the kingship of Munster. Conall needed no more encouragement and had hurried to Cashel and lit a fire. This was supposed to have happened around sixty years before St Patrick visited around 453 and converted Munster’s then king Óengus to Christianity. During the baptismal ceremony the saint accidentally pierced Óengus’ foot with the sharp end of his crozier. The king, thinking it was part of the ritual, suffered in silence.

Mathgamain’s success in capturing Cashel promised to make the Dál Cais a major power as Munster was one of the most important of Ireland’s over-kingdoms, covering the whole of the south-west of the island. However, before Mathgamain could win effective control of Munster, Máel Muad murdered him and recaptured Cashel. Brian now unexpectedly found himself king of the Dál Cais and quickly proved himself to be a fine soldier. After his expulsion from Limerick in 972, Ivar established a new base on Scattery Island, close to the mouth of the Shannon, from where he could still easily threaten Dál Cais. This sort of tactic had served Vikings well since the 840s, but no more. Brian had learned the importance of naval power from the Vikings and in 977 he led a fleet to Scattery Island, surprising and killing Ivar. A year later Brian defeated and killed his brother’s murderer to regain control of Cashel. Very shortly afterwards he defeated his last serious rival for control of Munster, Donnubán of the Uí Fidgente, and the remnants of the Limerick Vikings under Ivar’s son Harald. Both Donnubán and Harald were killed. This spelled the end of Viking Limerick. The town now effectively became the capital of Dál Cais, but Brian allowed its Norse inhabitants to remain in return for their valuable military and naval support. In the years that followed, Brian also became overlord of the Viking towns of Cork, Wexford and Waterford.

Now secure in his control of Munster, Brian began to impose his authority on the neighbouring provinces of Connacht and Leinster. Brian’s ambitions inevitably brought him into conflict with Dublin’s overlord, Máel Sechnaill. Almost every year, Brian campaigned in either Leinster, Meath or Connacht. Limerick and other Viking towns provided Brian with fleets, which he sent up the River Shannon to ravage the lands of Connacht and Meath on either side. When Donchad mac Domnaill, the king of Leinster, submitted to Brian in 996 Máel Sechnaill recognised him as overlord of all of the southern half of Ireland, including Dublin. Brian almost immediately faced a rebellion by Donchad’s successor in Leinster, Máel Morda, and the king of Dublin, Sihtric Silkbeard (r. 989 – 1036). Sihtric was another son of Olaf Cuarán, by his second wife Gormflaith, who was Máel Morda’s sister. Brian’s crushing victory over the allies at the battle of Glen Mama in 999 left him unchallenged in the south. Brian dealt generously with Sihtric, allowing him to remain king, and marrying Gormflaith, so making him his son-in-law. There was a brief peace before Brian, his sights now set on the high kingship itself, went back onto the offensive against Máel Sechnaill. Sihtric played a full part in these campaigns, providing troops and warships. Finally defeated in 1002, Máel Sechnaill resigned his title in favour of Brian and accepted him as his overlord: it was the first time that anyone other than an Uí Néill had been high king. Two more years of campaigning and every kingdom in Ireland had become tributary to Brian, hence his nickname bóraime, ‘of the tributes’.

The Battle of Clontarf

Brian’s achievement was a considerable one but he did not in any meaningful sense unite Ireland: outside his own kingdom of Dál Cais, Brian exercised authority indirectly, through his tributary kings, and he created no national institutions of government. Nor was the obedience of Brian’s tributaries assured: he faced, and put down, several rebellions. The most serious of these rebellions began in 1013 when Máel Mórda of Leinster renewed his alliance with Sihtric Silkbeard, who, despite Brian’s conciliatory approach, still hoped to recover Dublin’s independence. To strengthen Dublin’s forces, Sihtric called in an army of Vikings under Sigurd the Stout, the jarl of Orkney, and Brodir, a Dane from the Isle of Man, which arrived at Dublin just before Easter 1014. Brian quickly raised an army that included several of his tributary kings, including Maél Sechnaill, and a contingent of Vikings under Brodir’s brother Óspak. The two armies met in battle at Clontarf, a few miles north of Dublin on Good Friday (23 April) 1014. Neither Brian nor Sihtric fought in the battle. Sihtric watched the battle from the walls of Dublin, where he had remained with a small garrison to defend the city if the battle was lost. Now in his seventies or eighties, Brian was too frail to take any part in the fighting and spent the battle in his tent. The exact size of the rival armies is unknown but Brian’s was probably the larger of the two.

The battle opened around daybreak in heroic style with a single combat between two champion warriors, both of whom died in a deadly embrace, their swords piercing one another’s hearts. The fighting was exceptionally fierce but Brian’s army eventually gained the upper hand and began to inflict severe casualties on the Vikings and the Leinstermen. Brian’s son and designated successor, Murchad, led the attack and was said personally to have killed 100 of the enemy, fifty holding his sword in his right hand and fifty holding his sword in his left hand, before he was himself cut down and killed. Among Murchad’s victims was jarl Sigurd. Of the Dublin Vikings fighting in the army, only twenty are said to have survived the battle and the Leinster-Dublin army as a whole suffered as many as 6,000 casualties. By evening, the Leinster-Dublin army was disintegrating in flight and many Vikings drowned as they tried desperately to reach their ships anchored in Dublin Bay. At this moment of victory, Brodir and a handful of Viking warriors broke through the enemy lines and killed Brian as he prayed in his tent. Brodir’s men were quickly killed by Brian’s bodyguards and, according to Icelandic saga traditions, Brodir was captured and put to a terrible death. His stomach was cut open and he was walked round and round a tree until all his entrails had been wound out. Máel Mórda and one of his tributary kings were also killed in the fighting, as too were two tributary kings on Brian’s side.

For the anonymous author of the Cogadh Gaedhel re Gallaibh, Clontarf was the decisive battle of Ireland’s Viking wars, but this exaggerates its importance. The author of the Coghad was essentially a propagandist for Brian Boru’s Ua Briain dynasty and he intended, by glorifying his achievements, to bolster his descendents’ claim to the high kingship of Ireland, which they contested with the Uí Néills. The true impact of the battle was rather different. The deaths of Brian and Murchad caused a succession crisis in Dál Cais that brought the rise of the Ua Briain dynasty to a crashing halt. Brian’s hard-won hegemony immediately disintegrated, Cashel reverted to its traditional rulers, and Máel Sechnaill reclaimed the high kingship. Sihtric found himself back where he had started his reign, a sub-king to Máel Sechnaill. There could have been no clearer way to demonstrate how far gone in decline Viking power in Ireland already was. Sihtric continued to take part in Ireland’s internecine conflicts but his defeats outnumbered his successes, and Dublin’s decline into political and military irrelevance continued. Dublin continued to prosper as Ireland’s most important port, however, making Sihtric a wealthy ruler. In 1029 he ransomed his son Olaf, who had been captured by the king of Brega, for 1,200 cows, 120 Welsh ponies, 60 ounces (1.7 kg) of gold, 60 ounces of silver, hostages, and another eighty cattle for the man who had conducted the negotiations. Though he was quite willing to sack monasteries when it suited him, Sihtric was a devout Christian and in 1028 he made a pilgrimage to Rome. Such journeys were primarily penitential and, as an active Viking, Sihtric no doubt had much to be penitential about. On his return to Dublin he founded Christ Church cathedral but pointedly placed it under the authority of the Archbishopric of Canterbury in England, then ruled by the Danish King Cnut. It was not until 1152 that the diocese of Dublin became part of the Irish church. His alliance with Cnut briefly resurrected Dublin as a power in the Irish Sea, but Cnut’s death in 1035 left Sihtric in a weak position. In 1036 Echmarcach mac Ragnaill, a Norse-Gaelic king of the Hebrides, captured Dublin and forced Sihtric into exile: he died in 1042, possibly murdered during another pilgrimage to Rome.

Echmarcach never succeeded in securing his hold on Dublin and in 1052 he was expelled by Diarmait mac Máel, the king of Leinster, who ruled the city directly as an integral part of his kingdom. For the next century Dublin became a prize to be fought over by rival Irish dynasties interspersed with periods of rule by Norse kings from the Isle of Man, the Hebrides, and even Norway.

Ostman Dublin

By the eleventh century the Viking towns had become fully integrated into Irish political life, accepted for the trade they brought and the taxes they paid on it, and their fleets of warships, which made them valuable allies in the wars of the Irish kings. Pagan burial customs had died out during the second half of the tenth century so it is likely that by now the Irish Vikings were mostly converted to Christianity. It was not only kings like Sihtric Silkbeard who had taken Irish wives, and in many cases the children of these mixed marriages were Gaelic speaking. It is even possible that people of Norse descent were minorities in the Viking towns among a population of slaves, servants, labourers and craftsmen that was mostly Irish. The Irish Vikings had become sufficiently different from ‘real’ Scandinavians to have acquired a new name, the Ostmen, meaning ‘men of the east’ (of Ireland). The name seems to have been coined by the English, who by this time had had ample opportunities to learn how to distinguish between different types of Scandinavian.

In its general appearance, Ostman Dublin was probably much like other Viking towns of the period, such as York or Hedeby in Denmark. In the tenth century the site was divided up by post-and-wattle fences into long narrow plots along streets. Sub-rectangular houses built of wood, wattles, clay and thatch were built end-on to the streets, with doors at both ends. Though the houses were often rebuilt, the boundaries of the plots themselves remained unchanged for centuries. Irish kings used these plots as the basic unit for levying tribute on Dublin, as Máel Sechnaill did in 989 when he levied an ounce (28 g) of gold for every plot. Paths around the houses were covered with split logs, or gravel and paving slabs. The streets of Ostman Dublin lie under the modern streets, so it is not known what they were surfaced with, but split logs were used in other Viking towns like York. Different quarters of the town were assigned to different activities. Comb-makers and cobblers were concentrated in the area of High Street, while wood-carvers and merchants occupied Fishamble Street, for example. Other crafts, like blacksmithing and shipbuilding, were probably carried out outside the town. The wreck of a Viking longship discovered at Skuldelev near Roskilde in Denmark proved to have been built of oak felled near Glendalough, 22 miles south of Dublin, in 1042.

The town was surrounded by an earth rampart, which was probably surmounted by a wooden palisade. By 1000, Dublin had begun to spread outside its walls and a new rampart was built to protect the new suburbs. By 1100 it had proved necessary to extend the defences again, this time with a stone wall that was up to 12 feet high. This was such a novelty that a poem of 1120 considered Dublin to be one of the wonders of Ireland. Dublin probably lacked any impressive public buildings – even the cathedral founded by Sihtric Silkbeard was built of wood and it would not be rebuilt in stone until the end of the twelfth century. Dublin’s four other known churches were probably also wooden structures. The basic institution of Dublin’s government, as in all Viking Age Scandinavian communities, was the thing, the meeting of all freemen. The thing met at the 40-foot (12 m) high thingmote (‘thing mound’), which was sited near the medieval castle. This survived until 1685, when it was levelled to make way for new buildings. Of the other Ostman towns, only Waterford has seen extensive archaeological investigations. Like Dublin it was a town of wooden buildings laid out in orderly plots within stout defences.

The end of Viking Dublin

Viking Dublin was finally brought to an end not by the Irish but by the Anglo-Normans. In 1167, Diarmait MacMurchada, exiled to England from his kingdom of Leinster, recruited a band of Anglo-Norman mercenaries to help him win back his kingdom. Reinforcements arrived in Leinster in 1169 and, with their help, Diarmait captured the Ostman town of Wexford. In 1170, Richard Fitzgilbert de Clare, popularly known as Strongbow, the earl of Pembroke, brought an army of 200 knights and 1,000 archers to support Diarmait, and within days he had captured another Ostman town, Waterford. On 21 September in the same year, Diarmait and Strongbow captured Dublin. The city’s last Norse king Asculf Ragnaldsson (r. 1160 – 71) fled to Orkney where he raised an army to help him win it back. In June 1171, Asculf returned with a fleet of sixty ships and attempted to storm Dublin’s east gate. The Norman garrison sallied out on horseback and scattered Asculf’s men. Asculf was captured as he fled back to his ships. The Normans generously offered to release Asculf if he paid a ransom, but when he foolishly boasted that he would return next time with a much larger army, they thought better of it and chopped his head off instead. Cork was the last Ostman town to fall to the Anglo-Normans, following the defeat of its fleet in 1173.

The Anglo-Norman conquest was a far more decisive event in Irish history than the advent of the Vikings. Despite their long presence in the country, the Viking impact on Ireland was surprisingly slight. Viking art styles influenced Irish art styles, and the Irish adopted Viking weapons and shipbuilding methods, and borrowed many Norse words relating to ships and seafaring into the Gaelic language, but that was about it. The Vikings certainly drew Ireland more closely into European trade networks and by the tenth century this had stimulated the development of regular trade fairs at the monastic towns. However, on the eve of the Anglo-Norman conquest, the Viking towns were still Ireland’s only fully developed urban communities. In contrast over fifty new towns were founded in the century after the Anglo-Norman conquest. Sihtric Silkbeard was the first ruler in Ireland to issue coinage in c. 997, but no native Irish ruler imitated his example: coinage only came into common use in Ireland after the Anglo-Norman conquest. The impact of Viking raiding did accelerate the slow process of political centralisation in Ireland, but even in 1169 the country still lacked any national government institutions. The high kings still exercised authority outside their personal domains indirectly through their tributary kings (though there were many fewer of them than when the Vikings had first arrived). A true national kingship would likely have emerged eventually, but the Anglo-Norman conquest brought this internal process to a sudden end. English governmental institutions were imposed in those areas controlled by the Anglo-Normans, while in those areas still controlled by the Irish, kingship degenerated into warlordism. There were no more high kings.

Dublin prospered after the Anglo-Norman conquest, becoming the centre of English rule in Ireland. England’s King Henry II (r. 1154 – 89) granted Dublin a charter of liberties based on those of the important English West Country port of Bristol. This gave Dublin privileged access to Henry’s vast British and French lands, spurring a period of rapid growth. One of Henry’s edicts took the Ostmen of Dublin and the other Norse towns under royal protection: their skills as merchants and seafarers made them far too useful to expel (though some chose to leave voluntarily). An influx of English settlers gradually made the Ostmen a minority in the city, however. The Ostmen also found that they did not always receive the privileges they had been granted because of the difficulty in distinguishing so many of them from the native Irish. In 1263, the dissatisfied Ostmen appealed to the Norwegian king Håkon IV to help them expel the English, but the collapse of Norse power that followed his death later that year ended any possibility that Dublin would recover its independence. Norse names soon fell out of use and by c. 1300, the Ostmen had been completely assimilated with either the native Irish or the immigrant English communities. A last vestige of the Viking domination of the city survives in the suburb of Oxmantown, a corruption of Ostmantown.

The Mongol Wars and the Evolution of the Gun, 1211–1279

The rise of the Mongols was a key event in the evolution of gunpowder technology. Their wars drove military developments in East Asia and spread gunpowder technology westward. You’d think this statement would be uncontroversial. After all, the Mongols created the world’s largest empire, connecting East Asia to South Asia, Western Asia, the Middle East, and Eastern Europe. Mongol commanders excelled at incorporating foreign experts into their forces, and Chinese artisans of all kinds followed Mongol armies far from home. Yet oddly, experts disagree about the extent to which—or even whether—Mongols used gunpowder weapons in their warfare, and some deny them a role in the dissemination of the technology.

How can there be disagreement about such a fundamental question? One reason is that most historians have a poor understanding about what early gunpowder weapons were like and what they were used for—they expect to find gunpowder weapons blasting down stone walls, as cannons would eventually come to do in the West. As we’ve seen, that’s not how gunpowder weapons worked in this period. Even the iron bombs of the Jin—the most powerful gunpowder weapons yet invented—were used not to batter walls but to kill people or, at most, to help destroy wooden structures. Moreover, at the time of the Mongol Wars, the most common gunpowder weapon was still the gunpowder arrow, used primarily as an incendiary. There’s no shortage of accounts referring to blazing arrows and fiery orbs hurled by Mongol catapults, but historians have argued that these were not gunpowder weapons on the grounds that gunpowder weapons would have attracted much more attention.

Another problem is that the Mongols left few historical documents to posterity. Even the records left by the Mongols’ regime in China—the Yuan dynasty—are fragmentary, and China is a place that takes its history seriously. The official History of the Yuan Dynasty, compiled by scholars in China after the fall of the Yuan in 1368, is sloppy and patchy compared to other official histories in the Chinese canon, and Sinologists have noted that Yuan documents are particularly reticent about military details. Scholars must piece together Mongols’ history from the sources of their beleaguered enemies, whose records tended not to survive burning cities. So although we can paint a fairly clear picture of the development of firearms technology during the Song-Jin Wars, our understanding of the more intense and catalytic Mongol Wars is less complete.

Even so, there seems to be little doubt that the Mongols were proficient in gunpowder weapons. No one fighting in the Chinese context—and the Mongols met their most determined resistance in the Chinese realm—could remain unconvinced about the power of gunpowder, which by the early 1200s had come to play an essential role in warfare.

Indeed, the Mongols had a chance to learn about gunpowder weapons from the masters of their use, the Jin dynasty.

The Mongol-Jin Wars

Genghis Khan launched his first concerted invasion of the Jin in 1211, and it wasn’t long before the Mongols were deploying gunpowder weapons themselves, for example in 1232, when they besieged the Jin capital of Kaifeng. By this point they understood that sieges required careful preparation, and they built a hundred kilometers of stockades around the city, stout and elaborate ones, equipped with watchtowers, trenches, and guardhouses, forcing Chinese captives—men, women, and children—to haul supplies and fill in moats. Then they began launching gunpowder bombs. Jin scholar Liu Qi recalled in a mournful memoir, how “the attack against the city walls grew increasingly intense, and bombs rained down as [the enemy] advanced.”

The Jin responded in kind. “From within the walls,” Liu Qi writes, “the defenders responded with a gunpowder bomb called the heaven-shaking-thunder bomb. Whenever the [Mongol] troops encountered one, several men at a time would be turned into ashes.” The official Jin History contains a clear description of the weapon: “The heaven-shaking-thunder bomb is an iron vessel filled with gunpowder. When lighted with fire and shot off, it goes off like a crash of thunder that can be heard for a hundred li [thirty miles], burning an expanse of land more than half a mu, a mu is a sixth of an acre], and the fire can even penetrate iron armor.” Three centuries later, a Ming official named He Mengchun ( 1474–1536) found an old cache of them in the Xi’an area: “When I went on official business to Shaanxi Province, I saw on top of Xi’an’s city walls an old stockpile of iron bombs. They were called ‘heaven-shaking-thunder’ bombs, and they were like an enclosed rice bowl with a hole at the top, just big enough to put your finger in. The troops said they hadn’t been used for a very long time.” Possibly he saw the bombs in action, because he wrote, “When the powder goes off, the bomb rips open, and the iron pieces fly in all directions. That is how it is able to kill people and horses from far away.”

Heaven-shaking-thunder bombs seem to have first appeared in 1231 (the year before the Mongol Siege of Kaifeng) when a Jin general had used them to destroy a Mongol warship. But it was during the Siege of Kaifeng of 1232 that they saw their most intense use. The Mongols tried to protect themselves by constructing elaborate screens of thick leather, which they used to cover workers who were undermining the city walls. In this way the workers managed to get right up to the walls, where they began excavating protective niches. Jin defenders found this exceedingly worrisome, so according to the official Jin History they “took iron cords and attached them to heaven-shaking-thunder bombs. The bombs were lowered down the walls, and when they reached the place where the miners were working, the [bombs were set off] and the excavators and their leather screens were together blown up, obliterated without a trace.”

The Jin defenders also deployed other gunpowder weapons, including a new and improved version of the fire lance, called the flying fire lance. This version seems to have been more effective than the one used by Chen Gui a century before. The official Jin History contains an unusually detailed description:

To make the lance, use chi-huang paper, sixteen layers of it for the tube, and make it a bit longer than two feet. Stuff it with willow charcoal, iron fragments, magnet ends, sulfur, white arsenic [probably an error that should mean saltpeter], and other ingredients, and put a fuse to the end. Each troop has hanging on him a little iron pot to keep fire [probably hot coals], and when it’s time to do battle, the flames shoot out the front of the lance more than ten feet, and when the gunpowder is depleted, the tube isn’t destroyed.

When wielded and set alight, it was fearsome weapon: “no one dared go near.” Apparently Mongol soldiers, although disdainful of most Jin weapons, greatly feared the flying fire lance and the heaven-shaking-thunder bomb.

Kaifeng held out for a year, during which hundreds of thousands died of starvation, but ultimately it capitulated. The Jin emperor fled. Many hoped the Jin might reconstitute the dynasty elsewhere, and here and there Jin troops still scored successes, as when a Jin commander led four hundred fifty fire lance troops against a Mongol encampment: “They couldn’t stand up against this and were completely routed, and three thousand five hundred were drowned.” But these isolated victories couldn’t break Mongol momentum, especially after the Jin emperor committed suicide in 1234. Although some Jin troops—many of them Chinese—continued to resist (one loyalist gathered all the metal that could be found in the city he was defending, even gold and silver, and made explosive shells to lob against the Mongols), the Jin were finished. The Mongols had conquered two of the three great states of the Song Warring States Period, the Xi Xia and the Jin. Now they turned to the Song.

The Song-Mongol Wars

It’s striking that the Song, this supposedly weak dynasty, held the Mongols off for forty-five years. As an eminent Sinologist wrote more than sixty years ago, “unquestionably in the Chinese the Mongols encountered more stubborn opposition and better defense than any of their other opponents in Europe and Asia.”

Gunpowder weapons were central to the fighting. In 1237, for example, a Mongol army attacked the Song city of Anfeng, “using gunpowder bombs [huo pao] to burn the [defensive] towers.” (Anfeng is modern-day Shouxian, in Anhui Province.) “Several hundred men hurled one bomb, and if it hit the tower it would immediately smash it to pieces.” The Song defending commander, Du Gao, fought back resourcefully, rebuilding towers, equipping his archers with special small arrows to shoot through the eye slits of Mongol’s thick armor (normal arrows were too thick), and, most important, deploying powerful gunpowder weapons, such as a bomb called the “Elipao,” named after a famous local pear. He prevailed. The Mongols withdrew, suffering heavy casualties.

Gunpowder technology evolved quickly, and although sources are sketchy, scattered references to arsenals show that gunpowder weapons were considered central to the war effort. For example, in 1257, a Song official named Li Zengbo was ordered to inspect border cities’ arsenals. He believed that a city should have several hundred thousand iron bombshells, and a good production facility should produce at least a couple thousand a month. But his tour was disheartening. He wrote that in one arsenal he found “no more than 85 iron bomb-shells, large and small, 95 fire-arrows, and 105 fire-lances. This is not sufficient for a mere hundred men, let alone a thousand, to use against an attack by the … barbarians. The government supposedly wants to make preparations for the defense of its fortified cities, and to furnish them with military supplies against the enemy (yet this is all they give us). What chilling indifference!”

Fortunately, the Mongol advance paused after the great khan died in 1259. When it resumed in 1268, fighting was extremely intense, and gunpowder weapons played significant roles. Blocking the Mongols’ advance were the twin fortress cities of Xiangyang and Fancheng, which guarded the passage southward to the Yangze River. The Mongol investment of these cities was one of the longest sieges of world history, lasting from 1268 to 1273. The details are too numerous to examine here, but two episodes are salient, each of which involved a pair of heroes.

The first was a bold relief mission carried out by the so-called Two Zhangs. For the first three years of the siege, the Song had been able to receive food, clothing, and reinforcements by water, but in late 1271 the Mongols had tightened their blockade, and the inhabitants had become desperate. Two men surnamed Zhang determined to run the blockade and take supplies to the cities. With a hundred paddle wheel boats they traveled toward the twin cities, moving by night when possible, red lanterns helping them recognize each other in the darkness. But a commander on the Mongol side learned about their plans and prepared a trap. As they approached the cities they found his “vessels spread out, filling the entire surface of the river, and there was no gap for them to enter.” Thick iron chains stretched across the water.

According to the official Song History, the two Zhangs had armed their boats with “fire-lances, fire-bombs, glowing charcoal, huge axes, and powerful crossbows.” Their flotilla opened fire, and, according to a source recorded from the Mongol side, “bomb-shells were hurled with great noise and loud reports.”30 Wang Zhaochun suggests that the fire bombs used on the two Zhangs’ boats were not hurled by catapults but were shot off like rockets, using the fiery coals the vessels carried. This would be exciting, but unfortunately the evidence is inconclusive. Historian Stephen Haw suggests that the vessels carried guns, which is also possible, but again the evidence is inconclusive.

In any case, the fight was brutal and long. The Zhangs’ soldiers had been told that “this voyage promises only death,” and many indeed died as they tried to cut through chains, pull up stakes, hurl bombs. A source from the Mongol side notes that “on their ships they were up to the ankles in blood.” But around dawn, the Zhangs’ vessels made it to the city walls. The citizens “leapt up a hundred times in joy.” When the men from the boats were mustered on shore, one Zhang was missing. His fate remains a mystery. The official Yuan History says one Zhang was captured alive. The official Song History has a more interesting story. A few days after the battle, it says, “a corpse came floating upstream, covered in armor and gripping a bow-and-arrow.… It was Zhang Shun, his body pierced by four lances and six arrows. The expression of anger [on his face] was so vigorous it was as though he were still alive. The troops were surprised and thought it miraculous, and they made a grave and prepared the body for burial, erected a temple, and made sacrifices.” Other sources suggest that Zhang Shun was indeed killed in battle. He was later immortalized in the famous novel The Water Margin.

Alas, the supplies didn’t save Xiangyang, because the Mongols had a pair of heroes of their own. Two Muslim artillery specialists—one from Persia and one from Syria—helped construct counterweight trebuchets whose advanced design allowed larger missiles to be hurled farther. They came to be known in China as “Muslim catapults” or “Xiangyang catapults,” and they were devastating. As one account notes, “when the machinery went off the noise shook heaven and earth; every thing that [the missile] hit was broken and destroyed.” Xiangyang’s tall drum tower, for example, was destroyed in one thundering crash. Did these trebuchets hurl explosive shells? There’s no conclusive evidence, but it would be surprising if they didn’t, since, as we’ve seen, bombs hurled by catapults had been a core component of siege warfare for a century or more. In any case, Xiangyang surrendered in 1273.

The Mongols moved south. A famous Mongol general named Bayan led the campaign, commanding an army of two hundred thousand, most of whom were Chinese. It was probably the largest army the Mongols had commanded, and gunpowder weapons were key arms. In the 1274 Siege of Shayang, for example, Bayan, having failed to storm the walls, waited for the wind to blow from the north and then ordered his artillerists to attack with molten metal bombs. With each strike, “the buildings were burned up and the smoke and flames rose up to heaven. What kind of bomb was this? The sources on the Battle of Shayang don’t provide details, but earlier references suggest that it was a type of gunpowder bomb. A reference to it appears in an account of a battle of 1129, when Song general Li Yanxian was defending a strategic pass against Jin troops. At one point, the Jin attacked the walls day and night with all manner of siege carts, fire carts, sky bridges, and so on, and General Li “resisted at each occasion, and also used molten metal bombs. Wherever the gunpowder touched, everything would disintegrate without a trace.” The molten metal bomb was a probably a catapult projectile that contained gunpowder and molten metal, a frightening combination. It didn’t work for General Li in 1129: he lost the battle and either committed suicide or was killed, depending on which account you believe, but it did work for Bayan in 1274. He captured Shayang and massacred the inhabitants.

Gunpowder bombs were also present at a more famous Mongol massacre, the Siege of Changzhou of 1275, the last major battle of the Mongol-Song Wars.46 Bayan arrived there with his army and informed the inhabitants that “if you … resist us … we shall drain your carcasses of blood and use them for pillows.” His warnings were ignored. His troops bombarded the town day and night with fire bombs and then stormed the walls and began slaughtering people. Perhaps a quarter million were killed. Did his troops get new pillows? Sources don’t say, but it seems that a huge earthen mound filled with dead bodies lasted for centuries. Bones from the massacre were still being discovered into the twentieth century.

The Song held out for another four years, often with mortal bravery, sometimes even blowing themselves up to avoid capture, as when, in 1276, a Song garrison managed to hold the city of Jingjiang in Guangxi Province against a much larger Mongol force for three months before the enemy stormed the walls. Two hundred fifty defenders held a redoubt until it was hopeless and then, instead of surrendering, set off a huge iron bomb. According to the official Song History, “the noise was like a tremendous thunderclap, shaking the walls and ground, and the smoke filled up the heavens outside. Many of troops [outside] were startled to death. When the fire was extinguished they went in to see. There were just ashes, not a trace left.”

Bombs like this one were the most significant gunpowder weapons in the Song-Mongol Wars, but in retrospect the most important development was the birth of the gun.

The Gun

What is a gun? The efficiency of a projectile-propelling firearm is directly related to how much of the expanding gas from the gunpowder reaction can get past the projectile. The technical term is “windage,” and less windage means more energy imparted to the projectile. A true gun therefore has a bullet that fits the barrel. During the Jin-Song Wars, fire lances were loaded with bits of shrapnel, such as ceramics and iron. Since they didn’t occlude the barrel, Joseph Needham calls them “coviatives”: they were simply swept along in the discharge. Although they could do damage, their accuracy, range, and power were relatively low.

In the late 1100s and the 1200s, the fire lance proliferated into a baffling array of weapons that spewed sparks and flames and ceramics and anything else people thought to put in them. This Cambrian Explosion of forms is similar to that found in the early gunpowder period itself—the fire birds, rolling rocket logs, and so on—and a famous military manual known as the Book of the Fire Dragon, compiled in the Ming period but partially written in the late 1200s, describes and illustrates many of these weapons, which historians have called, as a general category, “eruptors.”

These eruptors had fantastic names. The “filling-the-sky erupting tube” spewed out poisonous gas and fragments of porcelain. The “orifice-penetrating flying sand magic mist tube” spewed forth sand and poisonous chemicals, apparently into orifices. The “phalanx-charging fire gourd” shot out lead pellets and laid waste to enemy battle formations. We find these and other weapons jumbled together in the Book of the Fire Dragon, which makes it difficult to determine when they emerged and how they were used. But unfortunately, we must use whatever sources we can find, because starting in the Song-Mongol Wars, our documentary record becomes sparse, and it remains so through the Mongol period that followed, whose leaders, as I’ve noted, left unusually poor documentation relative to other Chinese dynasties.

It is clear that fire lances became common during the Mongol-Song Wars. In 1257, a production report for an arsenal in Jiankang Prefecture refers to the manufacture of 333 “fire-emitting tubes”, and two years later the Song History refers to the production of something quite similar, a “fire-emitting lance”, which emitted more than just fire: “It is made from a large bamboo tube, and inside is stuffed a pellet wad. Once the fire goes off it completely spews the rear pellet wad forth, and the sound is like a bomb that can be heard for five hundred or more paces.” Some consider this “pellet wad” to be the first true bullet in recorded history, because although the pellets themselves probably did not occlude the barrel, the wad did.

Yet a truly effective gun must be made of something stronger than bamboo. Traditionally, historians have argued that metal guns emerged after the Mongols defeated the Song and founded the Yuan dynasty in 1279. Researcher Liu Xu, for instance, writes, “It was the Yuan who completed the transition from the bamboo- (or wood- or paper-) barreled firearm to the metal-barreled firearm, and the first firearms in history appeared in China in the very earliest part of the Yuan.” Similarly, other scholars, including Joseph Needham, have suggested a date of around 1280.

Archaeological evidence tends to corroborate this view. Take, for instance, the Xanadu gun, so named because it was found in the ruins of Xanadu, the Mongol summer palace in Inner Mongolia. It is at present the oldest extant gun whose dating is unequivocal, corresponding to 1298. Like all early guns, it is small: just over six kilograms, thirty-five centimeters long. Archaeological context and the straightforward inscription leave little room for controversy about the dating, but it was certainly not the first of its kind. The inscription includes a serial number and other manufacturing information that together indicate that gun manufacture had already been codified and systematized by the time of its fabrication. Moreover, the gun has axial holes at the back that scholars have suggested served to affix it to a mount, allowing it to be elevated or lowered easily for aiming purposes. This, too, suggests that this gun was the product of considerable prior experimentation.

The Xanadu gun is the earliest dated gun, but undated finds may predate it. One famous candidate is a piece discovered in 1970 in the province of Heilongjiang, in northeastern China. Historians believe, based on contextual evidence, that it is from around 1288. One careful analysis argues persuasively that it was likely used by Yuan forces to quash a rebellion by a Mongol prince named Nayan. Like the Xanadu gun, it is small and light, three and a half kilograms, thirty-four centimeters, a bore of approximately two and a half centimeters.

Yet archaeologists in China have found evidence that may force us to move back the date of the first metal firearms. In 1980, a 108-kilogram bronze gun was discovered in a cellar in Gansu Province. There is no inscription, but contextual evidence suggests that it may be from the late Xi Xia period, from after 1214 but before the end of the Xi Xia in 1227 (Gansu was part of Xi Xia territory). What’s intriguing is that it was discovered with an iron ball and a tenth of a kilogram of gunpowder in it. The ball, about nine centimeters in diameter, is a bit smaller than the muzzle diameter of the gun (twelve centimeters), which indicates that it may have been a coviative rather than a true bullet-type projectile. In 1997, a bronze firearm of similar structure but much smaller size (just a kilogram and a half) was unearthed not far away, and the context of its discovery seems to suggest a similar date of origin. Both weapons seem more primitive than the Xanadu gun and other early Yuan guns, rougher in appearance, with uneven casting. Future archaeological discoveries will develop our understanding with greater certitude, but for now, it does seem possible that the earliest metal proto-guns were created in the late Xi Xia state, in the early 1200s.

Although historians debate the precise date of the gun’s origin, at present the disputes are in terms of decades. It seems likely that the gun was born during the 1200s and that the Mongols and their enemies aimed guns at each other. After defeating the Song dynasty in 1279 and founding the Yuan dynasty, the Mongols and their Chinese troops invaded Japan, Vietnam, Burma, and Java, wars that stimulated further innovation, although, alas, records are few and say little about gunpowder weapons.

Equally important, although the Yuan brought relative peace within the borders of the Middle Kingdom itself, it was not a lasting peace. As the Yuan dynasty dissolved during the early 1350s, guns played a central role in the bloody wars that followed. The most successful gunpowder lord was a poor monk named Zhu Yuanzhang, whose gunmen succeeded in establishing one of the most impressive dynasties in China’s history, the great Ming, which scholars now call the world’s first gunpowder empire.

Douglas AD (BT2D, A-1) Skyraider 1945–1972/5 Part I

Too late for World War II, the Douglas AD series went on to achieve a stunning combat record in both Korea and Vietnam during a career that stretched over two decades. Its story began in September 1943, when BUAER circulated the new requirement for a single-seat bomber-torpedo (BT) aircraft intended to replace SBDs, SB2Cs, and TBMs. Douglas originally submitted a proposal for the XBTD-1, which was basically a rehash of its less than successful XSB2D design; then in June 1944, the company surprised BUAER by asking for cancellation of the BTD program in favor of a totally new concept. Its proposed XBT2D-1 was much closer to BUAER’s bomber-torpedo criteria: a simple design with a tailwheel layout in which weapons were carried on external racks beneath a bottom-mounted wing. For dive-bombing, Douglas introduced a new type of dive brake system consisting of flat panels that extended from the sides and belly of the fuselage. BUAER was sufficiently interested in the new concept to award Douglas a contract for 25 pre-production BT2D-1s, and after the first prototype flew on 15 March 1945, increased the order to 548 production aircraft.

The huge government cutbacks imposed after V-J Day resulted in the BT2D contract being reduced to 277 aircraft, and when the BT designation was changed to A for attack in early 1946, the plane became the AD-1. Service trials were completed in late 1946 and by early 1947, production AD-1s began replacing SB2Cs and TBMs in fleet units. The final 25 aircraft were delivered as AD-1Qs, a specialized ECM sub-variant that featured a separate compartment aft of the pilot for a radar operator. When the AD-1 had been in service less than a year, BUAER selected it as the Fleet’s standard single-seat attack type and made plans to acquire improved versions. Deliveries of 152 AD-2s having the more powerful R-3350-26W engine, stronger wings, a new canopy design, and fully enclosed wheel covers began in mid-1948 and were joined by an additional 21 AD-2Qs and one AD-2QU target tug. During 1948-1949 the Navy took delivery of 127 AD-3s possessing even more airframe strengthening, longer stroke landing gear, and a redesigned tailwheel, plus 15 three-seat night attack AD-3Ns, 31 three-seat early-warning AD-3Ws fitted with belly radomes, and 21 two-seat AD-3Qs.

The AD-4, introduced in 1949 with increased takeoff weight, a stronger tailhook, and a P-1 autopilot, also came in night attack, early-warning, and ECM sub-variants.

BUAER originally anticipated AD production would end in 1950 when the last of 180 AD- 4 variants were delivered, but naval involvement in the Korean War, which began in June 1950, had the unexpected effect of continuing AD production nonstop and led to demand for development of new versions. By the end of 1952, 1,051 AD-4s (all variants) had been delivered, and they were followed by 165 AD-4Bs armed with four 20-mm cannons and also configured to carry a tactical nuclear weapon, the first single-seat naval aircraft to have such capability.

Originally envisaged as a four-seat ASW platform, the AD-5 emerged with a fuselage lengthened by two feet and widened to permit side-by-side seating for a pilot and three crewmembers under an elongated canopy. Fin area was increased and the dive-brakes on the sides of the fuselage were deleted. But even before the first AD-5 flew in August 1951, BUAER changed its mind and earmarked it for production as an attack aircraft. The 212 standard attack versions subsequently built came with conversion kits, which, in addition to its basic attack function, allowed the type to be used either as a transport (12 seats), cargo carrier, ambulance, or target tug. Production AD-5s began entering service in late 1953 and were followed by 218 AD-5W early-warning and 239 AD-5N night/all-weather attack sub-variants, 54 of which were later modified as AD-5Q ECM aircraft.

The refinements of the AD-4B, plus LABS (low-altitude bombing system), new bomb racks, a jettisonable canopy, and a hydraulic tailhook were standardized in the single-seat AD- 6, which flew in 1953 and replaced AD-4s during 1954-1956. After delivery of 713 AD-6s, the final model was the single-seat AD-7, which differed in having a more powerful R-3350-26WB engine, stronger landing gear, and stronger outer wing panels. AD production finally ended in February 1957 when the last of 72 AD-7s rolled off El Segundo’s assembly line.

ADs were destined to remain in active naval service for 22 years-considerably longer than BUAER expected. In late 1946-early 1947, VA-19A, VA-3B, and VA-4B were the first squadrons to receive ADs, and by the eve of the Korean war, the type was equipping sixteen Navy and two Marine attack squadrons. In Korea, ADs operating from both carriers and land bases earned a reputation as the best all-around attack aircraft in the combat zone. Besides flying day attack, night attack, countermeasures, and early-warning missions, it was the only aircraft in the theater capable of delivering 2,000-lb. bombs against hardened targets (like bridges and dams) with dive-bomber precision. After Korea, AD’s carried on as naval aviation’s standard single-seat attack type and reached their peak in the mid-1950s when they equipped 29 Navy and 13 Marine squadrons. Even though a gradual phase-out of the type began in 1956 with the arrival of A4Ds, BUAER still planned to keep its ADs in service until the early 1960s. Moving somewhat faster, the Marine Corps retired its last AD-6 at the end of 1959.

When the tri-service system was adopted in September 1962, those ADs remaining in service were re-designated as follows: AD-5=A-1E; AD-5W= EA-1E; AD-5Q=EA-1E; AD- 5N=A-1G; AD-6=A-1H; and AD-7=A-1J. In 1964, plans to retire the type were postponed by military developments in Southeast Asia, where A-1s subsequently flew hundreds of combat sorties as part of the ongoing carrier task force stationed off the coast of Vietnam. Owing to their slower speed and excellent loiter range, A-1s were considered the best tactical aircraft available for escorting troop-laden helicopters and ground-fire suppression in rescue combat air patrol (RESCAP) operations. Though never intended for air-to-air confrontations, Navy A- 1Hs were in fact credited with the downing of North Vietnamese MiG-17s on two occasions. The type’s active naval career ended in 1968 when the last single-seat combat sortie was flown by an A-1H of VA-25 in February and the final ECM mission by an EA-1E of VAQ-33 in December.

In the early 1960s, as U. S. military involvement in Southeast Asia increased, the USAF found itself without any type of attack aircraft that could be adapted to slow, close-in operations like counter-insurgency (i. e., COIN: suppression and interdiction of guerilla troops and supplies) or RESCAP. Several different types of aircraft, all prop-driven, were evaluated at the Special Warfare Center located at Eglin AFB in Florida, including several ex-Navy A-1 Skyraiders. Once the tests were concluded, USAF officials immediately made plans to acquire 150 surplus wide-body A-1Es from Navy stocks to be overhauled for expected service in Vietnam. Modifications included addition of dual controls and weapons racks not normally carried on Navy E models. Actual combat operations commenced in early 1964 with the 34th Tactical Group based at Ben Hoa AB in South Vietnam. A-1E sorties were initially flown with a Vietnamese observer in the right seat for the purpose of target identification, but for most of its service the type was flown as a single-pilot attack aircraft. As they became available from Navy stocks in the mid and late 1960s, the USAF also began to operate single-seat A-1Hs and Js which became especially well-known for their “Sandy” operations-RESCAPs escorting the “Jolly Greens” (i. e., Sikorsky HH-53 helicopters) deep into North Vietnamese or Laotian airspace to rescue downed American pilots and aircrew. The USAF operated A-1s on Sandy missions up until November 7, 1972, when American involvement in Vietnam was almost at an end.

Skyraiders were probably the most numerically important aircraft to operate with the Air Force of the Republic of Vietnam (VNAF). In fact, the American government began the transfer of surplus Navy AD-6s (A-1Hs) to the VNAF in 1960, and as they became available, more followed. VNAF Skyraider pilots were initially trained by the Navy at NAS Corpus Christi and later by the Air Force at Hurlburt AFB. During the so-called “Vietnamization” of the war between 1969-1972, many Air Force A-1s were simply turned over to the VNAF when U. S. forces left the country. During this time the VNAF reached a peak strength of eight Skyraider squadrons, and they were operated in combat all the way up to the fall of Saigon in April 1975. Although a number of VNAF A-1s are known to have fallen into North Vietnamese hands, no apparent effort was made to put them back into service. Beginning in 1951, the British Royal Navy Fleet Air Arm (FAA) began taking delivery of AD-4Ws for use aboard its carriers in the airborne early-warning role. The first 20 aircraft were new, but the remaining 30 of 50 delivered were supplied from U. S. Navy stocks. Known as the Skyraider AEW. 1, the type remained in frontline service with the FAA until replaced by Fairey Gannets in 1962. A number of these aircraft were thereafter converted to target tugs and operated by the Swedish Air Force until the early 1970s. Forty surplus AD-4s were sold to the French Air Force in 1959 and thereafter flew combat in support of French forces in Algeria and Chad (1960) and in French Somaliland and Madagascar (1963). French Skyraiders remained in service in small numbers until the early 1970s, and some of these were turned over to the Cambodian Air Force, where they were briefly used against Viet Cong and North Vietnamese troops.

Korean War Skyraiders

By 1950, the TBM Avenger torpedo-bomber, SB2C Helldiver dive-bomber and the post-war AM Mauler had been replaced on fleet carriers by a new type in the form of the single-seat Douglas AD (formerly BT2D) Skyraider attack aircraft. A rugged, powerful aeroplane capable of carrying a 10,000-lb ordnance load of bombs, rockets and torpedoes, as well as 20 mm cannon, the AD, or `Able Dog’, as it was often called, prevailed over several competing designs and sidelined the similar AM Mauler to become the standard attack aircraft in the fleet.

More than 865 ADs had been built in four basic versions by the time hostilities broke out in Korea, including a wide range of specialised variants. The AD-1, a production version of the XBT2D-1 prototype, had been superseded by later versions by the time the conflict commenced in June 1950, and none of the 242 built saw combat. The AD-2, of which 156 were built, featuring greater structural strength, greater internal fuel capacity and a revised cockpit, saw extensive combat, however. The 125 AD-3 versions, which featured still more strengthening, a redesigned canopy, improved cooling of the engine and improved landing gear, also helped to equip US Navy attack (VA) squadrons during war deployments to Korea.

The AD-4 version and its sub-variants – the production standard in 1950 – were the most numerous to serve in the Korean War, equipping attack squadrons in 17 of the 25 combat deployments undertaken by Skyraider units. The AD-4 featured an uprated engine, an improved cockpit windscreen, a modified tailhook and a P-1 autopilot. Production totalled 372 examples, of which 63 were modified specifically for service in the harsh Korean winters. Designated AD-4Ls, they boasted anti-icing equipment and de-icer boots on the leading edges of the wings. The AD-4Ls were also fitted with an additional two cannon. When the `Able Dog’ was assigned the role of nuclear strike, 28 AD-4s were structurally strengthened for loft-bombing and designated AD-4Bs. An additional 165 AD-4Bs were built as such at the factory. One attack squadron and one composite (VC) squadron deployed to the Korean war zone with the AD-4B.

The versatility of the TBM was carried over into the AD, as the Skyraider was modified to perform a variety of specialised missions. The aft fuselage of an XBT2D-1 was converted to accommodate an electronic countermeasures (ECM) operator crew station and an access door to produce an XBT2D-1Q prototype. This change was made operational with the creation of 35 AD-1Qs. The modification followed with the production of 21 AD-2Qs, 23 AD-3Qs and 39 AD-4Qs. The AD-2Q, AD-3Q and AD-4Q saw combat over Korea with several attack and fighter AD units, plus VCs -33 and -35. Early Q-models had only an electronic surveillance measures capability, with ECM – jamming – coming later.

Similar in configuration to the Q versions were the night-attack variants, the AD-3N and AD-4N. Unlike the Q, the N featured two crew stations – one for a radar operator – in the aft fuselage, along with two access doors, but left no room for dive brakes. The AD-3N carried the APS-19A radar pod, while AD-4Ns were equipped with wing-mounted APS-31 radar and a searchlight. They also boasted an electronic surveillance measures (ESM) capability similar to the Q. Production of the AD-3N totalled only 15 aircraft, and the type saw combat on just two Korea deployments with VC-3 and VC-35. The AD-4N was much more abundant, with 307 built, of which 37 were modified with the AD-4L’s cold-weather upgrades and extra 20 mm cannon. These became AD-4NLs. Because of the increased demand for straight attack aircraft in Korea, 100 AD-4Ns were stripped of their two aft crew stations, fitted with the extra cannon and given the designation AD-4NA.

Succeeding the TBM-3W in the airborne early warning role were the unarmed AD-3W and AD-4W Skyraiders, modified with a similar belly radome housing an APS-20 search radar. The radar was operated by two crewmen housed in the aft fuselage under a turtleback extension of the cockpit. The W variants were primarily used to warn carrier battle groups of approaching aircraft, although they also performed the anti-submarine search role. The US Navy procured 31 AD-3Ws and 118 AD-4Ws.

The US Navy’s carrier force in June 1950 included only nine carrier air groups (CVGs), just three of which were based in the Pacific. This was principally because the Cold War was in full swing and support of the North Atlantic Treaty Organisation facing the Soviet Union in Europe was the top priority for US naval forces in early 1950. The US Navy had nine AD attack squadrons in service at the outbreak of the Korean War, one per carrier air wing.

By the end of the Korean War, the US Navy fielded 16 attack and two frontline fighter squadrons equipped with Ads.

Attack squadrons would initially deploy with a single attack version of the AD, but as the war progressed and attrition occurred other models were sent as replacement aircraft. Some VA squadrons flew AD-2/3/4/4Q versions during a single deployment, and AD-4L/NL/NAs entered the mix later in the war. When a carrier departed station for home, it would transfer some aeroplanes to other carriers or to the aircraft replacement pool at Atsugi, Japan. Some AD attack squadrons (typically equipped with 18 aircraft of all types) also included a few Q-models in their line-up. In addition, special mission composite squadrons (VC) for night attack, ESM/ECM and early warning included VC-3, VC-11 and VC-35 in the Pacific Fleet and VC-4, VC-12 and VC-33 in the Atlantic Fleet.

During this period, CVG staffs were also occasionally equipped with one or two ADs, usually including the Q versions. For administrative efficiency, the CVG staff and VC dets were organised into temporary squadrons, with the senior officer of the various VC detachments as the `commanding officer’. Although discontinued in June 1949, the practice of giving these temporary units a designation continued unofficially in some cases. `VC-110′ was such a unit in CVG-11, for example. For the purposes of this book, the dets will be discussed in terms of their parent units.

The numerous Naval Air Reserve Training Units had yet to receive any Skyraiders by June 1950, being equipped instead with AM-1s and TBMs.

Also of note, at the beginning of the Korean War the US Marine Corps did not field any attack squadrons (VMA), nor was it equipped with any Skyraiders. It relied instead on F4U-4/5/5N variants of the Corsair and on nightfighter versions of the F7F Tigercat. Although the Marine Corps was slated to begin receiving W- and Q-models of the AD in 1950, its Skyraiders did not reach Korea until mid-1951.


The units that flew Skyraiders in both the USAF and the Vietnamese Air Force (VNAF) during the Vietnam War. The 12 squadrons of these two air forces that were equipped with the Douglas aircraft saw extensive combat from 1960 to 1975. And this 15-year period is but five years short of spanning the entire existence of the VNAF. History will show that with the introduction of the AD-6 Skyraider in 1960, the VNAF truly had a capable, albeit demanding, aircraft – demanding in that it required a pilot’s full attention all of the time, whether in the air or on the ground. That it lasted 15 years as the VNAF’s frontline attack aircraft speaks volumes for its capabilities, and those of the men who flew it.

These capabilities, however, did not come without a price. Of the approximately 350 Skyraiders operated by the VNAF, only 70 remained by the end of 1973. And by the time the North Vietnamese Army (NVA) invaded South Vietnam in April 1975, just 40 Skyraiders were left at various VNAF bases for the enemy to use as they saw fit. It was the end of not only the VNAF, but also of the country the Skyraider units had fought so hard to defend.

Nestled inside this 15-year timeframe was the eight-year period that the USAF operated various models of the A-1 Skyraider in Southeast Asia. Commencing operations in-theatre in mid-1964, Skyraiders were the premier close air support (CAS) aircraft for the USAF until the end of 1972. The A-1 also became synonymous with the search and rescue (SAR) mission, and many a downed airman gave thanks when he heard the voice of a `Sandy’ on his survival radio, followed shortly after by the din of the Wright R-3350 radial engine as the Skyraider roared overhead. But make no mistake, the A-1 served well in all of its roles, from Special Forces fort defence to Military Assistance Command, Vietnam – Studies and Observations Group (MACV-SOG) support.

All Skyraider pilots gave some, but far too many gave their all. Of the approximately 330 A-1s operated by the USAF in Southeast Asia, nearly 200 were lost. More than 100 USAF Skyraider pilots were either killed in action or listed as missing in action.

Douglas AD (BT2D, A-1) Skyraider 1945–1972/5 Part II

USAF Skyraiders

If you thought that USAF Skyraiders were the same as US Navy Skyraiders except for their exterior colour schemes, you would be wrong. Two A-1Es on loan from the US Navy (BuNos 132417 and 132439) were evaluated by Tactical Air Command’s Special Air Warfare Center (SAWC) at Eglin AFB, Florida, from August 1962 to January 1963. The stated purpose of the test was to `evaluate the A-1E Skyraider for possible use in counterinsurgency warfare, and gauge its maintenance supportability and requirements’.

The conclusions reached were that the A-1E was an aircraft in the operational inventory that could perform many roles peculiar to counterinsurgency warfare, and after completion of minor modifications it would be capable of carrying all conventional ordnance of the 2000-lb or smaller class either then in the inventory or programmed for production.

The following items required modification or new installation:

1. Newest version R-3350 engine (R-3350-26WD)

2. Landing and taxi lights

3. Parking brake

4. Speed-brake well doors (never implemented)

5. N-9 gun camera to replace installed N-6 camera

6. Dual controls to include rudder/wheel brakes and control column with trim controls. Engine controls were listed as not required (a requirement for a throttle was added at a later time)

7. The right-hand side of the glare shield required modification to prevent the blocking of important warning lights from view

8. A pneumatic tailwheel to replace the existing hard solid rubber wheel to allow operations on a variety of runway and parking ramp surfaces.

9. Exterior paint and markings consistent with applicable USAF regulations

10. Aircraft technical order revisions to reflect modifications made

Although these modifications were based solely on the testing of the A-1E aircraft in 1962-63, many of them applied to the other models of Skyraider that would be procured in the future.

The first aircraft delivered to the USAF were A-1Es in mid-1964. After numerous programme changes regarding the distribution of these first machines, 25 went to Tactical Air Command (TAC) to be used for Skyraider upgrade training at Hurlburt Field, Florida. A further 48 USAF A-1Es were at Bien Hoa AB by the end of 1964, by which point eight Skyraiders had been lost with the death of six American pilots and two Vietnamese observers.

In mid-1967 the USAF was able to acquire single-seat A-1H/Js from the US Navy, and these aircraft underwent a similar modification programme to that undertaken with the A-1E, except of course for the changes relating to the second set of flight controls – H- and J-models were single-seat Skyraiders. Once completed, the aircraft were transported by ship to Southeast Asia, arriving at their respective units about a month after the modifications had been completed. These deliveries began in late 1967, and were largely complete by the end of 1968.

The A-1E was a multipurpose version of the Skyraider developed to permit greater versatility either as an attack aircraft or in the utility role. It departed from previous variants in that it had side-by-side seating for two crewmembers. The A-1E was powered by the R-3350-26WA engine and fully equipped to carry bombs, rockets, torpedoes, mines and other stores on external racks. Four M3 20 mm cannons were installed in the wings. The aircraft could also be equipped with auxiliary tanks both internally and externally for long-range operations. For utility purposes, the aircraft could quickly be equipped with seating for passengers, as well as facilities for the carriage of litter patients or provisions.

USAF A-1Es were produced from four different US Navy variants, namely the AD-5, AD-5N, AD-5Q and AD-5W. These A-1Es subsequently proved to be the mainstay of the first group of Skyraiders used by the USAF and, later, by the VNAF. Gone were the bulbous radomes and electronics pods carried on the inner stations of the AD-5Q/W, as well as the opaque rear canopies with a single viewing port on each side. The latter were eventually replaced with the blue plastic enclosures that gave rise to the nickname `blue room’ for the space behind the two front side-by-side seats of the USAF’s A-1Es.

A close variant of the A-1E was the A-1E-5, which differed from the USAF’s standard E-model through its lack of right-seat controls. In order to expedite the delivery of additional A-1Es to Southeast Asia in the 1965-66 timeframe, the installation of right seat controls for these aircraft was not accomplished. By mid-1966 the training of VNAF pilots by USAF units in Southeast Asia had ended, thus removing the need for dual-control A-1Es. However, no E-5s were sent to the `Skyraider school’ at Hurlburt AFB, in Florida, for obvious reasons. Other than by looking at the tail number, there was no easy way an observer could tell the difference between an A-1E and an A-1E-5 from the outside. It would be a gross understatement to say that no self-respecting Skyraider pilot wanted to be in the right seat of an A-1E-5 in combat!

There were three A-1E-5s assigned to the 1st Special Operations Squadron (SOS) when I arrived at Nakhon Phanom (NKP) Royal Thai Air Force Base (RTAFB) in October 1971. There was a check-out programme in the 1st SOS at this time that required all new pilots to first ride in the right seat of the E or E-5, then get in the left seat for a few more flights with an IP (instructor pilot), before going solo in either the E-, H- or J-model Skyraider. The IPs hated to be in the right seat of the E-5, but there they were.

The USAF’s A-1G closely resembled the E-model, being formerly designated the AD-5N in US Navy service. This aircraft was designed as the three-seat night-attack variant of the AD-5, and for all intents and purposes the A-1G was different from the E-model only in ways we pilots could not detect. Without looking up the serial number or searching through the aircraft’s maintenance paperwork, there really was no way of telling a USAF A-1E from an A-1G.

Because of what became termed the `USAF A-1E standard’, many US Navy-designated A-1Gs became Air Force A-1Es. A `standard A-1E’ was produced when all the USAF-stipulated modifications had been made prior to an aircraft seeing frontline service. Some A-1Gs were only partially modified due their urgent requirement as attrition replacements in Southeast Asia, which in turn meant that they kept their US Navy designations.

In August 1965, Headquarters Pacific Air Forces made the decision to camouflage all USAF aircraft in Southeast Asia. This of course affected all Skyraiders then in-theatre, plus those undergoing modification for shipment to Southeast Asia. A goal was set to complete the camouflage of all Skyraiders in South Vietnam by the end of 1966. The result was a profound change in the appearance of all aircraft in-theatre. This camouflage scheme would become standard for all tactical combat aircraft in the USAF well into the 1980s.

As with most piston-engined aircraft designed and built prior to 1960, the Skyraider had no means for the pilot to escape should the need arise when operating at low altitudes – essentially below 2000 ft. With the A-1 typically operating well below the recommended safe bailout altitude of 2000 ft while performing its mission, the only choice available for most pilots was to crash-land the aircraft if they could. Many could not, however, and the loss rate for aircraft and pilots proved to be unacceptably high as a result.

In an effort to improve a pilot’s chances of survival, the USAF contracted Stanley Aviation Corporation in 1965 to develop an automated escape system for the Skyraider. The company’s answer was the extraction seat. The seat would remain in the aircraft, and the pilot would be pulled out in a standing position, attached to a rocket-propelled tether. By 20 April 1967, the task of installing the Yankee Extraction System in all USAF A-1Es in Southeast Asia had been completed.

It did not take long for the newly installed system to prove its worth, for on 21 May 1967 Maj James Holler’s Skyraider (133855) of the 1st Air Commando Squadron (ACS) was hit by ground fire shortly after departing Pleiku AB. Holler was about 1000 ft above the ground when he activated the extraction system, and although he subsequently landed on rocky ground and broke both ankles, this was the first successful use of the Yankee Extraction System. Shortly thereafter, on 11 June, Majs James Rauch and Robert Russell became the second and third satisfied customers of the Stanley Aviation product when their Skyraider (132408) experienced a loss of power during their air strike in northern Laos (possibly due to battle damage). Forced to extract about 1500 ft above the ground, both men landed safely and were rescued by a USAF Jolly Green HH-3 helicopter.

However, it should be noted that there were subsequently some problems with the Yankee system, which had many safety features that required `man-in-the-loop’ inspection and preparation. Some easy-to-miss items in the checklist were predictably overlooked, with disastrous results. Following each failure, there were modifications made to either procedures or equipment, or perhaps both. There is no way of knowing how many lives could have been saved if the Yankee Extraction System had been fitted in the A-1 from the very beginning, but certainly it would have been a significant number.

VNAF Skyraiders

The VNAF was provided with 25 AD-6 Skyraiders in 1960 to replace its ageing F8F Bearcats through the Military Assistance Program. The first of these aircraft arrived in Saigon on 24 September 1960, and after processing and flight testing they were flown from Tan Son Nhut AB to Bien Hoa AB to enter service with the VNAF. Over the next six years, further deliveries added a sufficient number of A-1s to allow the equipment of four additional squadrons. By January 1966 the VNAF had 146 Skyraiders assigned to it.

As far as is known, the only modifications made to Skyraiders transferred to the VNAF from the US Navy were the removal of all equipment associated with the delivery of nuclear weapons and the tailhook. The earliest A-1H Skyraiders even kept the US Navy paint scheme, but with US markings replaced by those of the VNAF. In one of the many ironies of the Vietnam War, the first USAF A-1Es based at Bien Hoa bore VNAF markings from June 1964 until February 1965 in an effort to mask the presence of American combat aircraft in South Vietnam. Many photographs exist of these early USAF Skyraiders incorrectly identified as belonging to the VNAF in various books, magazines and journals.

A significant change to the appearance of VNAF Skyraiders (and all their aircraft for that matter) occurred with the introduction of camouflage paint in 1966. From the very start, VNAF Skyraider markings had been flamboyant and eye-catching, and the addition of camouflage did not change this. During this period A-1s exhibited a mixture of flamboyance and stealth – a seeming contradictory combination for a combat aircraft. However, during the later stages of the war, VNAF Skyraiders were much more subdued in their overall appearance.

In 1967, Stanley Aviation Corporation’s Yankee Extraction System was installed in all VNAF Skyraiders. This system functioned by means of an extraction rocket similar in principle to the drogue gun system on a normal ejection seat. Once the catapult charge fired, the spin-stabilised rocket was fired when the pendant lines reached full stretch. Actuation of the system was effected after the canopy had been jettisoned. The rocket was then erected by means of a pyrotechnic piston and lever under the erector/launcher. The rocket launched from the rear wall of the cockpit, and by means of a pair of Perlon pendants (rope-like straps), the pilot was pulled up and out of the cockpit. His parachute was rigged with an automatic opening system which activated after the rocket pendants separated from the parachute risers. The system included a set of rails to allow the seat back to rise up, while the seat pan was articulated to assist in the positioning of the pilot to the vertical as the rocket extracted him from the cockpit.

By the late 1960s losses and ongoing conversion of some VNAF A-1 units to the A-37 Dragonfly meant that there were just 69 operational Skyraiders available to oppose the surprise communist Tet Offensive of January 1968. USAF Skyraiders began to be transferred to the VNAF through MAP at around this time too, these aircraft being configured slightly differently to the Skyraiders procured directly from the US Navy – the USAF A-1s were still fitted with tailhooks, for example.

In total, the VNAF operated 329 Skyraiders, of which 240 came from the US Navy and the remaining 89 from the USAF as MAP transfers (most of the latter were supplied between 1970 and 1972). According to one account, the VNAF lost a total of 242 Skyraiders either in combat or to non-combat related accidents. However, it could be said that in the end all the Skyraiders supplied to the VNAF were lost since the air force ceased to exist following the fall of Saigon at the end of April 1975.

One of nine pre-production A2D-1s over southern California in 1953, the only turboprop-powered attack type ever seriously considered by the Navy. Only five of the pre-production aircraft were actually flown.

Douglas A2D Skyshark 1950–1954

Power plant: One 5,035-shp Allison XT40-A-2 double turboprop engine driving a six-bladed Aeroproducts contra-rotating, constant-speed propeller. Armament: Four fixed forward-firing 20-millimeter cannons and up to 5,500-lbs. of mixed ordnance carried externally. Performance: Max. speed 501-mph at 25,000 ft.; cruise 276-mph; ceiling 48,100 ft.; range 637 mi. loaded. Weights: 12,944-lbs. empty, 18,720-lbs. loaded. Dimensions: Span 50 ft., length 41 ft. 3 in., wing area 400 sq. ft.

The A2D was the only type of turboprop-powered attack aircraft to receive serious consideration for production by the Navy. The project was originally begun by BUAER in 1947 as an effort to adapt Douglas’s proven AD design to turboprop power, and two prototypes were ordered under the designation XA2D-1. But as development progressed, the design bore little similarity to the AD other than a superficial resemblance. A completely new fuselage was created around the complex XT40 powerplant and drive system, which comprised two T38 engines connected via drive shafts to a common transmission. In the case of a shutdown, each engine could operate independently and drive one or both propellers. The wing was also new, having a thinner-section airfoil and large wing root extensions. On paper, the XA2D-1’s projected performance was so promising that BUAER ordered 10 pre-production models, and soon followed with an order for 339 production A2D-1s on three separate contracts.

The first flight of the XA2D-1 was made on May 26, 1950 from Edwards Air Base. Early testing immediately revealed severe problems with the double engine and drive system. And to make matters worse, the prototype crashed in mid-December 1950 due to engine failure, killing the Navy test pilot. When testing resumed after the second XA2D-1 flew in April 1952, problems with the T40 engines and drive system continued, and shortly thereafter, BUAER cancelled the contract on all but 10 pre-production examples. The first pre-production A2D-1 flew in June 1953 and nine more were completed, but the last four were never flown. A Douglas test pilot safely ejected from the second pre-production A2D-1 in August 1954 following a gearbox failure. No further development was undertaken after 1954.

Technical Specification

Aircraft: Douglas A-1H


Type: attack

Manufacturer: Douglas Aircraft Co.

Engine: Wright R-3350-26WA, radial, 18 cyl., air cooled

Power: 2738hp

Wingspan: 50ft (15.24m)

Length: 39ft 2in (11.83m)

Height: 15ft 8in (4.77m)

Wing area: 400.33sq ft (37.19m²)

Max take-off weight: 25,000 lb (11,340 kg)

Empty weight: 11,968 lb (5,429 kg)

Max speed: 322mph at 18,000ft (518km/h at 5,846m)

Service ceiling: 28,510ft (8,690m)

Range: 1,142mi (1,840km)

Crew: 1

Load-armament: 4x20mm cannon; 7,960 lb (3,630 kg)

Takeda clan

Hideo Takeda, battle in the Fuji River.

Battle of Nagashino, a painted screen of the XVII-XVIII centuries.

Battle of Nagashino, (1575)

Battle fought by Nobunaga Oda (1534–1582) and his ally Ieyasu Tokugawa (1543–1616) with Takeda Natsunori, around the strategic fortress of Nagashino. In this encounter, the forces of Tokugawa and Nobunaga Oda were the first to rely primarily on massed firepower in the form of Western armaments, helping to transform samurai warfare while pushing both houses closer to hegemony over Japan.

Ieyasu Tokugawa had actually forged a familial alliance with the Takedas, whose territories bordered his own in central Honshu. He married both a son and daughter into the Takeda household in the 1560s, but in the world of shifting alliances and steady warfare that characterized Japan at the time, the alliance quickly foundered. The Takedas were soon at war with the Tokugawa again.

The death of the elder Takeda (Shingen) in 1573, at the hands of a sniper in battle, placed his son Natsunori at the head of the Takeda house. The rising fortunes of the Tokugawa had made them fierce rivals of the Takedas, and when in 1575 a traitor to Tokugawa offered to hand over the vitally strategic castle of Ozaki to the Takedas, Natsunori Takeda jumped at the opportunity. Ozaki was the capital of Mikawa Province, the heart of Tokugawa territory, and its castle was guarded by Tokugawa’s own son.

Takeda led a force of 15,000 warriors in what was expected to be a near-bloodless seizure of Ozaki Castle. Instead, they discovered en route that the treachery had been discovered by Tokugawa. Rather than face a humiliating retreat, Takeda opted to send his troops instead against the nearby fortress of Nagashino, another strategic castle sitting at the convergence of three rivers and guarding the entrance to Mikawa and Totomi Provinces.

Takeda began his siege of the castle in May 1575 but was still unsuccessful when word came that relief forces led by Tokugawa and Oda were on their way. Takeda opted to stand his ground near Nagashino and engage the approaching allied armies, though his forces were outnumbered more than two to one. At the Battle of Nagashino in June 1575, the alliance’s greater numbers and, more important, overwhelming firepower, including musket volley fire by alternating ranks (the first time that this technique is known to have been employed in warfare), carried the day. Takeda lost almost two-thirds of his men and generals, and the mortally wounded Takeda clan would linger only until 1582, when it was overrun for good.

References and further reading: Parker, Geoffrey. The Military Revolution: Military Innovation and the Rise of the West, 1500–1800. Cambridge, UK: Cambridge University Press, 1988. Sadler,A. L. The Maker of Modern Japan: The Life of Tokugawa Ieyasu. Tokyo: Charles E. Tuttle,1937.


The Takeda were descendants of Emperor Seiwa (858-876) and are a branch of the Minamoto clan (Seiwa Genji), by Minamoto no Yoshimitsu (1056-1127), brother to the Chinjufu-shogun Minamoto no Yoshiie (1039-1106). Minamoto no Yoshikiyo (c.?1075 – c.?1149), son of Yoshimitsu, was the first to take the name of Takeda.

For much of the Sengoku Period the provinces controlled by the Sengoku Daimyo were quite well defined and ruled as a stable economic unit. There is little evidence of civil war within these territories except where the Ikko-ikki sectarians were involved. Warfare tended to be confined to clashes between daimyo, particularly at sensitive areas where two territories met. Thus the border between the Takeda, Uesugi and Hojo lands was frequently contested. Kawanakajima, an area of flatland which was effectively a no man’s land for the Takeda and Uesugi, saw no fewer than five battles across its fields. It was such conflicts, coupled with their geographical remoteness from the capital, that acted as a counterweight to whatever pretensions these daimyo might have had to becoming Shogun. Many possessed the necessary military power, but few were fated to exercise it in this direction.

Shingen Takeda, (1521-1573)

A prominent warlord (daimyo) of Japan’s Sengoku period (“the Age of the Country at War”). Shingen Takeda was born Harunobu Takeda in 1521, the eldest son of Katsuyori Takeda, ruler of Kai Province in north-central Japan. Young Takeda overthrew his father in 1541 and installed himself as the provincial shugo (military governor). He then embarked on the conquest of neighboring Shinano Province, which was secured by 1555. However, this action brought him into direct conflict with Kenshin Uesugi (1530-1578) of Eichigo Province, another young and dynamic military figure. For nearly two decades, the two leaders clashed at the battlefield of Kawanakajima, with especially severe encounters in 1553, 1554, 1556, and 1563.

At length, neither side could gain a decisive advantage over the other, and both turned their territorial ambitions elsewhere. During this period, Takeda shaved his head, became a Buddhist priest, and assumed the more familiar name of Shingen.

At this time, Japan was seething with conflict as major families of samurai battled for control of the country. In 1568, Takeda attacked the Imagawa family and drove it from Surguga Province. However, the ever-shifting balance of power forced him to ally with the Hojo, Asakura, and Asai families to oppose the growing strength of Nobunaga Oda. In 1573, Takeda attacked the combined forces of Oda and his surrogate, Ieyasu Tokugawa, at Mikatagahara, driving them from the field. This defeat had the effect of inducing the weakened shogun, Yoshiaki Ashikaga, to denounce Oda, a feat that ultimately led to the shogunate’s downfall. However, Takeda became distracted by events elsewhere and, by failing to follow up this impressive victory, allowed his enemies to consolidate.

In the spring of 1573, Takeda again advanced against Tokugawa and besieged one of his castles in Noda. Events are not clear, but he died either of disease or a gunshot wound on 13 May 1573. The Takeda clan did not outlive his demise and were eliminated as a military threat by Oda at Nagashino in 1575.

Beyond his military prowess, Takeda was also renowned for his administrative and organizational abilities. He placed Kai Province on a very high order of efficiency and was affectionately regarded by the populace. Takeda was also celebrated for his calligraphy and poetry, military guile, and capacity for great acts of both chivalry and cruelty.

The Armies of the Sengoku Jidai

The armies of the Sengoku Jidai were manifestations of the feudal social structure of Japan, which revolved around kinsmen and vassals. The head of the clan and its army was the daimyō, literally translated as “great name”. He was supported by the kashindan. These were a group of blood relatives and retainers associated by family ties, marriage, filial oaths, and hereditary vassalage. The retainers were given land to govern and were expected to provide military support during times of war. 

A standing army was uncommon but was popularised during the later years of the Sengoku Jidai. For the majority of the period, armies were composed of farmers who needed to stand down during the planting and harvesting seasons. Fighting a campaign during idle periods would offer an opportunity for the peasants to earn extra income from looting and possibly get promoted to samurai.

Typically, when a call to arms was issued, each landowning samurai was required to muster a pre-determined quantity of troops and equipment based on his wealth. Troops from all around the province would then converge at a designated place where they would be reorganised into battalions wielding similar weaponry and start practicing drills. The daimyō determined the chain of command for the campaign.  The prominent retainers would act as bushō (general). A taishō (field marshal, commander-in-chief) would be appointed if the daimyō did not intend to take the role himself.

Each general commanded a division comprised of specialised battalions of cavalry, missile and melee troops mustered from their fiefs. These troops were only loyal to their direct lord and the daimyo, not the taishō or other generals. To reflect this, Japanese commanders who are not assigned as the Commander-in-Chief are classified as Ally-Generals. Their units cannot receive any command effects from other generals except the C-in-C.

The Japanese wielded a variety of weapons, the prominent ones being the katana (sword), yari (spear), naginata (polearm), yumi (bow) and teppō (matchlock).  Contrary to popular depictions, the katana was just a sidearm and the yari was the weapon of choice due to its range and versatility. All classes of soldier, from the lowly ashigaru to the elite samurai, wore armour of lamellar construction.  

Before 1530, mounted samurai would primarily use bows, similar to other East Asian cavalry. The switch to the yari and shock tactics happened around the 1530s, pioneered by the Takeda clan.

The main fighting force was foot samurai, augmented by ashigaru. Due to the rugged terrain, the Japanese utilised loose formations and fighting was done man-to-man, as depicted in martial arts and samurai films. Hence they are classified as Warriors.

In 1543, Portuguese merchants introduced matchlock firearms (teppō) to the Japanese. Teppō ashigaru infantry were deployed, but there weren’t enough firearms available to equip large units. These small units are classified as Light Foot and are primarily used as skirmish troops.

By 1551, as battles grew larger, more and more ashigaru infantry were being mustered, as a result of which the proportion of foot samurai in the army was somewhat reduced. The Battle of Nagashino in 1575 showed the Japanese that massed volley fire from firearms behind field defences could defeat samurai cavalry From then on, teppō ashigaru formations were larger and did not engage in mere skirmishing tactics.

By 1577, samurai cavalry had lost its appeal due to changes in battlefield technology and tactics. And by 1592, ashigaru infantry tactics evolved into fighting in close formation. They would receive better training and form the backbone of the Late Sengoku Era army. Ashigaru infantry, including yumi and teppō armed units, are now classified as Medium Foot. A century of fighting also depleted the numbers of available samurai. Just like their mounted counterparts, foot samurai, who still fought man-to-man, were finding it harder to dominate the battlefield against organized peasant foot troops. The 1590s also introduced some other elements of modern warfare such as light artillery, but these were not used as extensively as on the Asian mainland.

Buddhist monks of various temples also trained for combat. They had to take up arms in order to protect their temples from rival sects. These warrior monks were called sōhei. During the Gempei War (1180-1185), the sōhei eventually became embroiled in secular politics as they joined the lords that supported their temple.  This was repeated during the Sengoku Jidai and the daimyō were able to gain the support of sōhei from their local temples.

The monks’ weapon of choice was the naginata, a long-bladed polearm. They also used bows and matchlocks. Occasionally, they can be seen wearing armour underneath their robes but the majority were unarmoured.

The Ikkō-ikki revolution gave some sōhei a new purpose. Instead of fighting for their temples and patrons, they fought under an ideology of equality and independence from the daimyō. Ikkō-ikki rebel armies were mostly made up of sōhei and supported by armed peasant mobs. Samurai who shared their ideals also joined but did not form separate units. The samurai fought alongside the monks and peasants and provided leadership as well as training.

The “Fahrgerät” FG1250 IR Night Vision equipment

“Solution B” It’s only a model!”

The Germans started experimenting with Night Vision technology in a date as early as 1936.

The first night driving tests by Wa Pruf 8/I occur in 1937. The Army considers the device ludicrously expensive for driving around at night.

Before the war the use of IR sights for gun aiming is discussed with the Army who set its IR sight requirement as the ability to hit bunker embrasure at 1000 meters with a 30cm / 100 watt searchlight. (Searchlight size and power consumption dictates the size/weight and battery life span of the equipment).

IR driving and gun aiming sights were again demonstrated to the Army in 1940, just after the French campaign, fitted to a Panzer I. Here it can be said the military were just uninterested. This subject was not discussed for two years however the Wa Pruf 8/I still continued work on other devices for the military particularly thermal detecting applications for Army coastal artillery.

Summer 1942 the Army came knocking on door of Wa Purf 8/I asking for a way to take on the Russian night time tank attacks (which greatly impressed the Germans). Dr Gaertner shows them the current gear can hit targets at night using a 36cm searchlight out to a range of 300 meters.

The man on the spot at the Wa Pruf General der Artillerie Heinz ZIEGLER (RK; DKiG) decides that range is sufficient and approves development. This is the well know Pak 40 ZG 1221 sight pictured on a Marder II. That shot dates from about November 1942 when firing tests confirm the set-up will work. Development continued and Gaertner notes development on IR in general increases at this point.

However, it was 1943 when one of such devices was finally tested on a Panther tank. It was named “Fahrgerät” (German for “driving equipment”) FG1250 although it was referred to as “Puma” by the allies in the post-war and also known as “Sperber”, a name deriving from the composite “Sperber” units that would be formed by mixing different Night Fighting equipped troops (Vampir, Falke, Uhu and Panthers ).

This equipment comprised an Infra-Red 200-Watt Headlamp and a IR receiver/sight capable of “seeing” in and converting the IR wave length to visible light. It is reported of having a range of up to 500 meters. The set was mounted on a mounting base installed in the commander’s copula and it could be used in a fixed 12 o’clock position traversing with the turret, or unlocked to provide 360° of rotation.

When installed the Commander’s copula, the hatch would not close and the equipment was fed from a battery/generator set installed inside the turret in the place used by the aft right munition rack, with a cost of 3 rounds less in total munition for the main gun.

During daytime the whole equipment would be taken down for protection and installed in a characteristic armoured box in place of the right standard storage bin on the back of the hull.

Such a setup was known as “Solution A” and offered Night Vision capability only to the Commander, the rest of the crew had to rely on his instructions for movement. For targeting the device was locked to the turret’s 12 o’clock position. The Commander would search for targets traversing the turret. Once acquired the crew had to calculate gun elevation by means of a steel “tape” riveted to the IR device mounting base that entered the turret by a small opening. Such a tape would slide up or down depending on the elevation of the IR headlamp/receiver and by measuring the size of the tape the crew would determine the main gun elevation from within the turret. Once everything was in place the shot was ordered by the Commander.

Soon enough a “Solution B” was promptly drafted, allowing the driver to use another set of IR beam/receiver in order to drive the tank under low light conditions, just like it was done with the “Falke”. The “Solution B” is just a fantasy that could maybe only have existed in the very beginning on some training Panther perhaps, but never in active service.

According to the “Germany’s Panther Tank: The Quest for Combat Supremacy” by Tom Jentz,book some 44 Panther tanks and its crews from at least 4 different units were equipped and trained to use Night Fighting gear by the end of WW2, however there are no records or evidence that they were actually used in combat. Interestingly enough some sources state that a trailing “F” was added to the serial number of all of those Panther tanks prepared of mounting Night Vision equipment.

The active units that fought in April 1945 used ONLY the IR-equipment for the commander – BUT one unit (1. Komp. Pz. Rgt 29 ) with 10 IR-Panthers used the – fahrgerät 1253 – for the driver also. He could just simply switch from normal periscope to the F. G. 1253 and use the same “halterung”. The F. G. 1253 was stored beside (on the left) the driver during the day. Nothing was needed to be rebuild outside. He could see the ground with help from the Commander’s spotlight when he had the IR-filter on, or from the light of the UHUs. It is correct that the commander could turn his IR in 360 degrees – the steel band was easy to remove and put on a holder over the periscope guard. Then he could focus on a target and the turret was turning to this position and the commander install the steel-band again, and begins the complicated communication with the gunner that saw nothing outside.

The Allies of course had their own IR and the Germans knew about it, for that reason IR was not used in the west. The British actively looked for IR on the east bank of the Rhine with specially equipped IR Mosquitos before the March crossings, finding none. The British also manufactured and held large numbers of cheap IR detectors close to the front should the Germans deploy their IR.