Even though Griffith had already indicated a preference for Collins, Brugha was still given first choice; he declined the offer, however. Even so, typical of his conflicted ways, he could not help feeling anxious that Collins’ promotion, even though it lasted a mere five weeks, was an added complication to the civil–military question, a sentiment he voiced to de Valera on his return from the USA on 23 December. In response, while he accepted the reasoning behind Brugha’s anxiety, especially when Brugha’s reasoning was a contributory factor in his own unexpected return home, de Valera felt that Brugha, having refused the offer, had little grounds for complaint.54
Nevertheless, in the most violent circumstances of the winter of 1920, Mulcahy was perhaps far too worried about the issue of security to give such matters his full attention. Naturally enough, like all of his comrades, he was intent on protecting his own skin: ‘the civilian worker – the Dáil Minister or official, the Irish Volunteers’ Headquarters’ Executive, for instance – cooped up in Dublin, felt the strain’.55 Besides, he strongly held the view that the members of GHQ, himself included, would be very difficult to replace. For that reason, he systematically rearranged GHQ’s working routine in order to conform to a code of optimum safety:
I arranged that the staff would never meet as a whole. A schedule was made out grouping the members of the staff into groups of three, for various activities, one person being starred as the spearhead of these activities and the membership generally of the groups being such as to echelon in such a way that there never was any group that didn’t understand what the other groups were doing – so that there was a common mind.56
Furthermore, he, as CS, must have been acutely aware of just how precarious the entire military venture was, because at no time did the IRA’s numbers exceed much more than 3,000 in active service from a willing panel of 15,000.57 For instance, by mid-March 1921, while Tom Barry’s West Cork flying column had 104 men, there were 5,000 British troops in the same area and they usually travelled in units of not less than 300.58 One could surmise that morale in that situation would have depended greatly upon ruthlessly successful operations.
Yet this was not entirely true because success brought its own problems in the form of an insidious retreat by some of these self-same activists. Bear in mind the following. In July 1920, nearly three months after the arrival of the Black and Tans and at roughly the time when the Auxiliaries entered the scene, the commander of the Third Tipperary Brigade, an extremely active one, complained to Mulcahy about cowardice and desertions, advocating that, in order to stop the rot, ‘the terror behind [should be] greater than the terror in front’.59 And, after Bloody Sunday, during a phase of the war when the IRA was probably at its most effective, some officers were beginning to complain that too many of their men were getting themselves locked up for their own personal safety.60
Quite possibly, therefore, in order to cope with these stress-related consequences, GHQ, in April 1921, began to partially decentralise the chain of command by introducing the divisional idea. A GHQ memorandum gave the reason that ‘the work of co-ordinating operations between adjacent Brigade areas was becoming too difficult to be handled by the General Staff’.61 Significantly, however, another reason, which originated with Liam Lynch as far back as 2 January 1920, was to take pressure away from the most active districts within each province. Lynch wanted larger groups of men to make incursions into inactive areas, thereby hopefully stretching the enemy’s resources and thinning out their line of attack.62
The plan was that the country, outside the metropolitan area, would be divided into fifteen divisions under the overall command of GHQ: five northern; three western; three southern; and one midland.63 GHQ presented that idea as a ‘very definite advance in our development’.64 Yet, at virtually the exact same time, some of their own internal staff memoranda strongly argued the case for centralisation based on the strategic supremacy of the Dublin theatre:
the grip of our forces on Dublin must be maintained and strengthened at all costs, and our forces there must be reinforced by strong flanking units to bring the capital into closer touch with the county … but [nonetheless] it cannot be too clearly stated that no number nor any magnitude of victories in any distant provincial areas have any value if Dublin is lost in a military sense.65
Clearly there were some misgivings within GHQ about how far, and in what manner, to proceed with decentralisation in order that the army would still be able to protect and maintain Dublin as the most important politico–military theatre: ‘make the enemy exert itself greatly, without seriously draining our resources or weakening our striking power elsewhere’,66 and, ‘questionable if gains resulting were worth the trouble’.67
Retrospectively, it is difficult to see how GHQ could have been truly confident about the division idea. In essence it was a terrible gamble because, with failure, there would have been little hope of easily returning to the original vertical command structure. In point of fact, the likelihood was that collapse would have ensued had not the Truce intervened.68 Staff officers had already, on very many occasions, conceded that they encountered problems trying to get battalions to organise into regular half-companies; to comply with paper work; and not to be continually pre-empting the campaign.69 Yet GHQ was now about to hand over some of the initiative to men whom they already knew had failed to sympathise with a clerical type of military leadership: ‘[The divisional commandant] must not try to run his Division as he used to run his brigade. If he does chaos will result … Divisional Administration calls for suitable machinery – a Divisional Headquarters and a Despatch system.’70
On the credit side, however, the divisional experiment would probably have shortened the chain of command and, thereby, might have answered the usual complaint of absenteeism by GHQ. But the Achilles’ heel of GHQ’s hopes for success through the use of larger raiding formations was the poor supply of arms and ammunition, as well as the prospect of massive demoralising defeats. Essentially that was the gamble which Tom Barry knew he was taking at both Kilmichael (28 November 1920) and Crossbarry (19 March 1921). In fact, he strongly opposed the divisional idea. He argued that ‘the Divisional unit and the guerrilla army of the Irish Republic were, in times of war, a contradiction in terms’.71 By this he almost certainly meant that the original qualities of intuition and localism, these both as interpersonal bonding agents and as shock tactical instruments, could be applied with great difficulty after membership had been enlarged because, at that point in time, the structure of command would have been made more complicated. Besides, physical combat would have to be entered into on foreign soil, as it were. (This was recognised as a particular problem for Tom Ennis’s handpicked Second Battalion men, who had to leave the familiar surroundings of their own area in order to partake in the assassination of the Cairo Gang which was dispersed in various lodgings on the south side of the Liffey.72) In the light of such complexities, therefore, GHQ’s hopes for the initiative were probably far too sanguine.
In any event, as has just been said, the divisional idea was a decision which GHQ quite possibly made as a means of dealing with the extraordinary pressure which the IRA itself was experiencing during late 1920 and early 1921. However, this pressure was not confined to the army alone. It adversely affected relationships within the revolutionary elite as well, leading to a deepening of pre-existing personal, ideological and strategical fractures. British military intelligence was aware of the disagreements.73 Discovering such information was probably not too difficult for them because, so serious were the confrontations that An tÓglach was forced, in two consecutive issues, to try to reassure readers that rumours about trouble were unfounded: ‘We stated then [in the last issue] that there was no difference of opinion among those entrusted with the Government of the Irish Republic; that the Minister responsible for the Army of the Irish Republic was interpreting the unanimous wish of his fellow-Ministers in pushing on the guerrilla warfare against the enemy as vigorously as possible.’74
However, it is significant that Brugha was referred to here. First of all, despite his commitment as MD to making the army, especially army GHQ, become accountable to parliament, he himself, not having quite ditched his own brand of Volunteering, ignored another, and equally essential, part of any democratically based, civil–military relationship, namely the concept of a separation of powers, when he intruded into the realm of military