Ireland’s Wars: The Battle Of Limerick

Almost from the moment that the split in the republican movement became manifest, it had one of its major fault lines through the city of Limerick. The IRA in the area, if one includes Michael Brennan’s forces to the north in Co Clare, was divided very evenly, the city had been one of the major pivot points of the handover crisis and now, with the war beginning, the city constituted the western edge of the very loose line that the anti-Treaty IRA were holding at the border of the “Munster Republic”. With both National Army and IRA units holding positions within the city in force, it was inevitable that the first large-scale combat of the Civil War outside of Dublin would take place there.

The truce that had been agreed back in March, that left the various barracks and other strong-points of the city split between pro and anti-Treaty forces, had barely held together up to the start of the Civil War proper. The IRA Volunteers in the city were mostly taken from the Mid-Limerick Brigade of Liam Forde, and came more directly under the command of Liam Lynch, who had come to Limerick after his unlikely exit from Dublin. They held the majority of the city, with garrisons in all four of its military barracks: the Ordnance Barracks to the south-east of the city centre, the Strand Barracks on the north bank of the Shannon, the Castle Barracks inside King John’s Castle to the north and the New Barracks (the modern day Sarsfield Barracks) to the south. They had several other key positions, mostly in the south of the city, like the RIC Barracks and Daly’s Mill near the docks, the Glentworth Hotel and the Limerick GPO. They also held the majority of O’Connell Street, the city’s main thoroughfare, that runs south-west to north-east and intersects the other major streets of the city centre.

On the other side were National Army troops under General W.R.E. Murphy within Limerick, and then the the 1st Western Division, under Brennan, and 4th Southern under Donnacha O’Hannigan. They held a number of non-military positions like the town hall, customs house, courthouse and Cruises Hotel, and also the William Street RIC barracks. William Street, a thoroughfare that cuts north-west to south-east through the city centre, with Sarsfield Bridge on one end and avenues to the train station on the other, was a vital area of National Army control, from which their other garrisons expanded out from. The provisional government can generally be seen as holding the very centre of the city and north of there, on King’s Island.

In the first week of the Civil War there was no fighting in Limerick, the result of a hastily arranged continuation of the truce. Both sides seized additional buildings, but an open outbreak of hostilities did not occur. Neither side seemingly thought they were in a strong enough position to oust the other, or they genuinely felt that a more larger-scale truce was in the offing. Brennan and O’Hannigan were able to move a large number of soldiers, maybe as many as 800, into the city at this time, though not all of them were armed. The pro-Treaty side probably had the better of this shadow-boxing, able to take buildings that were strategically useful, like St Marys Cathedral and the Mary Street RIC barracks, that overlooked the Strand Barracks, the city prison next to the Ordnance Barracks, and the womens prison near the Castle, which disrupted the IRA garrisons’ ability to operate and communicate with each other. They were able to string together positions by knocking holes in adjoining buildings in William Street, and then also covered Sarsfield Bridge in the city centre, which would severely isolate the Castle garrison. Further, they maintained a hold on Athlunkard Bridge to the north, which secured a vital line to the 1st Western’s heartland in Clare. Provisional government barricades divided up key parts of the city centre, and lines of supply or retreat were kept open to the countryside.

That Brennan was able to do thus is testament to his own subterfuge work and to the hesitance of Lynch. In the first instance Brennan was able to successfully convince his counterpart that he was in a better position than he was: mocked up “Lewis Guns” – maybe a tube suitably accoutered – peered out of pro-Treaty positions and reinforcements arriving from the direction of Ennis were given guns outside the city before they marched in, before these were then taken off them, driven back out, and given to the next group of reinforcements. The impression was that the National Army was heavily armed, when the reality was anything but. In the second instance Lynch, despite the fact that the IRA had been able to rapidly overrun the few pro-Treaty positions in County Limerick, and were numerically superior to their opponents in the city, did not want to commit to an attack. He worried about the strength of his opponents, but also believed that a peace between the warring factions was possible. At worst, he thought a truce could bottle up a large number of pro-Treaty soldiers in Limerick while the war was fought elsewhere. His approach was criticised by other officers, like Sean Moylan, who urged that the city should be flooded with as many Volunteers as possible.

Brennan made sure to string negotiations out as long as he could, terrified that Lynch would realise that the IRA both outnumbered their opponents in terms of men and guns. In the end both sides came to a formal agreement that essentially meant they would hold what they had already taken and not fire on the other. Provisional government leaders, political and military, were unhappy with the truce, especially when Lynch made it public knowledge, and feared Brennan could not be relied on. This distrust was so acute that representatives from command were sent to Limerick to scout out the situation. This officer, Diarmuid MacManus, went as far as attempting to take command and telling Lynch he was about to begin hostilities, but eventually backed down, perhaps when he realised how dire the National Army position in Limerick City really was (and the possibility that, if he was to take over Brennan’s command, the soldiers would not obey his orders or, worse, join the republicans).

Beggers Bush realised how important Limerick was, and Eoin O’Duffy was appointed to take command of the situation there, with clear orders to bring additional men and material, and use this to expel the IRA from the city. It would be some time before he could take personal command, but the reinforcements were dispatched immediately. The first group of this, around 150 men with armoured cars under Sean Hogan, had to fight their way into the city by way of Galway and Ennis, but made it through to hook up with Brennan on the 11th. Hogan had secret instructions to evaluate Brennan’s situation and loyalty, and to relieve him if he felt the need, but the two were compatriots from the War of Independence and Hogan was satisfied that Brennan’s hands had been tied in Limerick up to that point. The new men, guns and vehicles tipped the balance, and Brennan was now more confident that he could go on the attack. An altercation in the city on the same day as the reinforcements’ arrival, where a National Army private was shot and killed, was the incident Brennan needed to end the truce and commence operations. The resulting fight would last the better part of two weeks, and constitute one of the longest sustained examples of urban conflict in Irish history.

National Army troops in the William Street area opened fire on IRA Volunteers in the Ordnance Barracks, which was the signal for a number of low intensity firefights to break out across the city. Much of the pattern of what would be the Battle of Limerick was set in those moments, as the various garrisons and positions fired back and forth, it being more a war of snipers than anything else. Most of the fighting was to take place in the parallel lines of William, Thomas and Roches Street in the city centre, with Roches and Thomas Street serving as the closet the battle had to an no man’s land between the opposing forces. The immediate lack of continuous lines in the city meant this was all that could be done in the short-term, just an exchange of bullets from a distance. Civilians hunkered down as best they could as a series of confused, brief engagements occurred all around the city centre and beyond. Both sides seized some additional buildings and the National Army threw up more barricades, seeking to limit IRA movement as much as possible, but there were no immediate gains of territory.

The IRA, many of whose officers were never informed of the truce’s cessation officially, were caught off guard. Lynch vacillated between conflicting emotions: upon hearing of the truce being broken he apparently thought the IRA would be inevitably forced to retreat and form into columns, but when he spoke to other officers in the New Barracks later that night his overly-optimistic assessment of the situation surprised others. These included Liam Deasy, then the O/C of the anti-Treaty 1st Southern Division, who already thought the situation in Limerick was hopeless. There was little sense of proper direction to what was occurring, mirroring the situation elsewhere. The provisional government knew what they wanted: to clear Limerick of the enemy by taking out their garrisons, one-by-one if need be. The republicans wanted to clear the National Army out, but had a less tangible strategy with which to do so, leading to sedentary tactics and confusion at every level.

On the 12th, as Lynch moved his command to Clonmel, the pro-Treaty side went on the attack in a somewhat firmer way, targeting anti-Treaty positions in the city’s docklands and quays. Movement became increasingly difficult in the streets owing to snipers from either side. The GPO, on Cecil Street, was a particular target for the National Army, but the IRA there managed to hold form in the face of repeated attacks, as did the garrison at the Ordnance Barracks. These buildings had thick, and high, enough walls to deflect infantry attacks, and the National Army as of yet had no access to artillery. Two IRA members were killed that day, and others on both sides wounded. Elsewhere the IRA attempted to catch up on their opponents in terms of gaining advantageous position, occupying buildings further up on O’Connell Street. It was not a bad strategy, but the necessity of having men in more buildings stretched IRA manpower to the breaking point. This was crucial as only one side was gaining men, through various batches of reinforcements were heading to National Army positions. One such column was turned away by IRA units guarding southern approaches to the city after a brief firefight, but they could not be kept out forever. Additional forces, most notably a mobile force under Tom Flood that included elements of the Dublin Guards, were coming towards Limerick from multiple directions.

All the while, the National Army kept up as much pressure as they could. On the 13th the focus went back to the Ordnance Barracks, with the armoured car Danny Boy utilised as much as possible, smashing through IRA barricades nearby and trying to clear mined areas ahead of infantry advances. Finding the barracks still too hard of a nut to crack, Brennan instead sent the Danny Boy and its attached infantry to attack the Munster Fair Tavern, an IRA position in far south-east of the city, important becasue it was a position that could monitor approaches to the city from that direction. The Danny Boy’s machine gun compelled the defenders to surrender relatively quickly, and 14 Volunteers went into captivity.

The IRA, especially the men of the 1st Southern Division, were not to be undone, and on that evening sent into the fighting their own improvised armoured car. Dubbed “The Hooded Terror”, it was a lorry that they had attached armour plating, and two revolving turret chambers armed with machine guns. Operating out of the New Barracks, it got involved in one of the few instances of armour-on-armour combat of the war when it engaged one of its counterparts of the National Army near Roches Corner, while infantry from both sides fired back and forth. On this occasion the anti-Treat side was successful, driving off the National Army, albeit temporarily, and securing buildings in the area. Outside of the city, the IRA ring was tightened further by the capture of Kilmallock on the same evening, a town that would soon become the focus of the last major military confrontation of the conventional phase.

On the fourth day of the fighting the National Army began to make more obvious gains inside the city. The IRA position at Daly’s bakery, near Sarsfield Bridge, was stormed and taken that morning, before a gradual, sustained assault on the Russell Mill’s garrison resulted in that buildings abandonment by the republicans, unable to stay put under the weight of machine gun and small arms fire. The nearby GPO was now hopelessly isolated, but its defenders hung grimly on for a few more hours, eventually forcing the pro-Treaty side to storm it. With its capture complete, the IRA ability to communicate with units outside the city, especially Lynch in Clonmel, was severely effected. Worse perhaps, the National Army was now in a better position to sweep the docks and quays. A short, failing, attack was carried out against King John’s Castle that evening, perhaps more a test of what defence it had, while Roches Corner continued to see sporadic exchanges of fire. The IRA held there, for the time being.

By Saturday the 15th, the fighting in Limerick was having an effect on the situation outside of the city and country, with anti-Treaty units taking advantage of the absence of the National Army in Clare in order to attack various barracks throughout the county, with mixed success: several villages and towns were temporarily captured, but two Volunteers were killed in the process. In the end it was all these Volunteers could to disrupt National Army movements towards Limerick, and in that they could never completely isolate the city. That morning the SS Arovina was able to sail up the Shannon and dock at the city, and disgorge a substantial supply of additional arms and ammunition for Brennan and his men. It was almost immediately put into use.

That evening a simultaneous attack was made on the two key anti-Treaty positions in the north-west of the city, at the Strand Barracks and King John’s Castle, with National Army troops utilising rifles, grenades, armoured cars, machine guns and mortar weapons in the attempt. Some damage was able to be dealt to both buildings but their garrisons, commanded by men with some experience in the attacking of such building during the War of Independence, stood firm. Efforts to place a mine against the wall of the Strand Barracks were deflected by carefully placed troops and a steady laying down of fire to interfere with the work. Both attacks had to be called off before nightfall, with the IRA still in possession of these key points.

On Sunday morning, another large scale effort was made to take the Strand Barracks, with National Army soldiers again able to fight their way up to its walls and doors, but again efforts to break in were beaten off. The IRA responded by attempting to make headway up William Street, but their limited infantry numbers and over-reliance on “The Hooded Terror” meant they too could make little headway, with pro-Treaty barricades doing their job at keeping the improvised armoured car hemmed in to a narrow area. Across the city, isolated firefights and exchanges of sniper fire continued, as the larger situation turned more and more to stalemate: the National Army seemed incapable of taking barracks buildings, and the IRA was not strong enough to expand outside of them. This kind or urban street fighting was far outside the experience of men on either side, save the very small number with memories of 1916, and getting used to the labourious, and dangerous, process of securing streets and buildings took time.

O’Duffy arrived in the area the following day with enough men and material to decisively swing things in the favour of the provisional government. His journey west had been slow, delayed by constant blocking of roads and the need to clear IRA units away from bridges, but that was all the IRA was able to do: pro-Treaty control of Dublin and the midlands told here. Basing himself in Killaloe, roughly 15 km’s north of the city, he was soon sending new infantry units, with more armoured cars, machine guns and rifles, into Limerick. Most importantly of all, his arrival came with the National Army’s trump card: an 18-pounder artillery piece. Suddenly the IRA’s holding of the various barracks in the city was no longer as terrible an obstacle.

The IRA were not ignorant of this danger, and on the 18th attempted their largest offensive operation of the battle, hooping to somehow sweep the National Army from the city centre before O’Duffy’s reinforcements could make an impact. 50 Volunteers moved into the centre of Limerick from the New Barracks, aiming to surprise pro-treaty positions in William Street. Their efforts were in vain: in trying to gain a march on their opponents by attacking throw a side-street, Fox’s Bow, they inadvertently put themselves in the firing line of well placed infantry units on William Street itself, and from machine gun crews on the other side of the Shannon, 300 yards away or more, who happened to have the perfect line of sight on the republican avenue of attack. The IRA retreated under the hail of fire, and then the National Army counter-attacked. The soldiers were so close to each other that anti-Treaty units in Roches Street found themselves unable to fire for fear of hitting their comrades, and after a brief and bitter fight, the effort to expand Republican control of the city resulted in the National Army expanding their own position.

The National Army next started to expand their segment of O’Connell Street, surprising an IRA garrison in a Cafe roughly half-way between William Street and the O’Connell monument by tunnelling into its basement and expelling its occupants. They fell back to the nearby Glentworth Hotel, but the loss of this outpost severely effected republican ability to project themselves into the city centre. The imminent arrival of provisional government artillery – an open secret in many ways – led to a mass exodus of civilians who had spent the previous week hunkered down, and a reappraisal of the situation by IRA officers. Fearful that the old walls of the Castle Barracks would be especially vulnerable, men started to be transferred to the seemingly more secure Strand Barracks, under the cover of a Red Cross ambulance (suspicious National Army machine gunners would eventually realise the ruse and subject the vehicle to fire). The Strand Barracks, under an enterprising officer named Connie McNamara, built up barricades in and around the building and expanded their position into neighbouring structures by tunnelling, but a degree of fatalism was also evident: McNamara ordered his younger brother out of the Barracks at this time, fearful of his parents having to bury two sons after expected assaults.

In desperation to turn things to their advantage before O’Duffy’s arrival made the situation untenable, the IRA briefly considered a scheme to use a fire engine to break through National Army lines and spray the William Street RIC Barracks – still Brennan’s HQ – with petrol and then light it. No fire engine could be sourced for what was probably a completely madcap idea with little chance of success, and now a creeping inevitability seemed to be becoming clear when it came to the possibility of IRA success in the city. The National Army had no need for such extremes. On the night of the 19th, having spent two days organising things in Killaloe, O’Duffy sent a huge portion of his reinforcements into the city: four lorries of troops, three armoured cars, lots of guns and the 18-pounder. Under cover of darkness it was set up at Arthur’s Quay, staring across at the walls of the Strand Barracks on the other side of the Shannon.

The final stages of the battle began on the 20th. A team of former British Army Royal Artillerymen operated the big gun, as another failing effort to attack the Strand Barracks was made by infantry on its rear. McNamara was offered the chance to surrender before it opened fire, and refused. Over the course of an hour 19 shells were fired at the Barracks, with a large crowd of civilians congregating to view the spectacle: the bombardment was enough to create a sizable breach, with the defenders obligated to contract their position. The gunners moved the 18-pounder to a different position so it could strike the rear of the Strand Barracks, and over the course of the afternoon another breach was made. Anti-Treaty units, hearing the shelling, tried to break through to the area by attacking down O’Connell Street, but were driven off by a sustained mass of rifle and machine gun fire. They no longer had the numbers to prosecute such an assault.

McNamara still refused to surrender, so a storming party was sent into one of the breaches. It was a perilous thing to do, as he breach was not that wide, and it was easy to cover it with small arms. The two leading officers, a David Reynolds and a Con O’Halloran, were wounded in the attempt, before the men they led fell back, unable to make any headway. Despite this success, McNamara now realised his position was finished. The barracks itself was on fire, and many of his men now took the opportunity to exit via a field hospital and flee north in the direction of Clare. The twenty or so left under McNamara put up a white flag and surrendered. Brennan allegedly offered McNamara a National Army commission on the spot, but the anti-Treaty leader refused.

The first capture of a barracks in the city was as clear a signal as could be made about how things were going to go, and the National Army wasted little time. The 18-pounder was re-positioned again, and now concentrated its fire on the Castle Barracks which soon caught fire, though this may have been started by the garrison. There was to be no epic siege this time: the IRA officer there realised quickly enough that the game was up, and ordered his Volunteers to withdraw as best they could, with some ferried over to the north bank of the Shannon by sympathetic local fishermen. Very rapidly, it was a case of two down and two to go for the provisional government.

The anti-Treaty position in Limerick was collapsing rapidly, and Lynch finally had enough. Deasy, in his account, claims that Lynch was unwilling to continue fighting in an urban environment now that artillery was a factor, and thought the IRA would have a better chance in the countryside. He wasn’t wrong, but it is still remarkable how quickly the Executives gave up in Limerick once the big gun came into play. Before the 20th was out, Lynch ordered the remaining garrisons in the city to set their buildings on fire and withdraw south. The Ordnance and New Barracks were duly lit up, both largely gutted, with the IRA units that had garrisoned them and the few remaining positions taking the Ballinacurra Road south and deep into the Limerick countryside, stopping only long enough to engage briefly with National Army troops attempting a belated encirclement. With that, the battle was over.

The Battle of Limerick had lasted ten difficult days, which makes the low death toll all the more surprising. Six pro-Treaty soldiers and five IRA Volunteers are recorded as being killed in the fighting, with many more injured for the National Army and for the republicans. Twelve civilians lost their lives in the crossfire. Given the length and extent of the fighting this seems like a remarkably small butchers bill, and reflects the inexperience of the soldiers on either side, hesitance to make fatal shots and the nature of the battle itself, which so often involved ineffective attacks on fortified positions: it’s hard to kill people when they are behind a barracks wall.

Brennan deserves enormous credit for his work ahead of the fighting in improving the pro-Treaty position given his limitations in men and guns. Taking buildings that were near the main anti-Treaty barracks positions interfered with IRA operations and communications without the need for risky assaults on the barracks’ themselves, and his work in convincing Lynch that the National Army was far stronger in the city than it really was is probably the only reason O’Duffy did not have to start from scratch when he arrived in the area. The erecting of barricades to impede movements, the creation of tunnels to eliminate small outposts and expand the provisional government position, and the holding of key bridges to ensure that the pro-Treaty side had routes for reinforcements to flood the city were all successes that Brennan, unfairly maligned by many leading political and military figures for the truce, could point to as proof of his own competence and commitment.

After that, it was a matter of the pro-Treaty side utilising their advantages and overpowering their enemy. Whether it was armoured cars, machine guns, grenades or artillery, the National Army had resources that the anti-Treaty side could not adequately replicate, and their correct application meant that victory in Limerick was a matter of when, and not if. Artillery was the final card that made the IRA’s position terminal in the city, and the only criticism that could be levelled at O’Duffy in terms of its use is how long it took for it to make an impact. There were wasted efforts aplenty in the fighting, and pro-Treaty superiority in guns counted for much, but it was a victory.

On the other side, though it was a battle that lasted longer than any thus far in the Civil War, Limerick was another catastrophe for the IRA, worse than Dublin or Blessington. In Limerick they had been in the ascendancy, and had then conspired to throw away that position of strength bit-by-bit, in not securing the bridges, in falling for Brennan’s ruses and drawn-out truce negotiations, in relying too much on their isolated barracks positions and in not being capable of forcing the National Army out of the city centre when they tried. One does not wish to judge the anti-Treaty personnel too harshly as this was, in many ways, a case two semi-professional militaries firing at each other in a fashion that was hardly an epitome of warcraft, but in critical moments, like the attempted advance into William Street for example, it was the National Army troops that succeeded.

The anti-Treaty side in Limerick could have been more aggressive from the moment the Four Courts was shelled, could have seized more buildings, could have pressed the city centre harder and could simply have been less sedentary and more pro-active. In the end, the IRA held on longer than they should have in Limerick, with a withdrawal after the arrival of artillery probably a better strategy than the several days holdout that transpired ahead of a withdrawal anyway. Perhaps it was a difference in officers, but the IRA was out-thought as well as outgunned in Limerick.

In many ways, the Battle of Limerick was the decisive conflict of the entire Civil War. When the war started the IRA could have taken the city, if not easily, then without enormous difficulty, given the difference in men and guns that was in effect at the time. If they had done, the National Army may have been able to re-take it, but it would have been a much more costly battle, and a more pro-active republican leadership could even have used the city as a jumping off point for an advance into the midlands. Instead, through early prevarication and then an inability to turn circumstance to advantage during then fighting, the IRA had lost one of their few key centres left. With Limerick’s fall to the provisional government, the defending line of the Munster Republic was fatally punctured at its western end. Numerous republican officers have stated the opinion that the war was lost as soon as Limerick City was, and that may well have been true. It certainly continued a run of defeats. In the next entry, we will turn to the eastern end of the line, also soon to be fatally punctured: the National Army was about to advance on Waterford.

To read the rest of the entries in this series, click here to go to the index.