Louth For Ever ★ writing in Medium on 17-August-2026.
Everyone is arguing about why Sinn Féin’s Stormont term has failed. The answer, and the only way out, both run through South Africa.
| Photo by Osama Madlom on Unsplash |
Designed to Exist, Not to Excel
In late July 2026, at the Patrick MacGill Summer School in Donegal, the leader of the Ulster Unionist Party said something truer than he knew. Jon Burrows said Stormont had been built by the architects of the Good Friday Agreement to exist rather than to excel, loaded with every check and balance its makers could devise, so that it would run without collapsing rather than govern well. He meant it as an explanation for why nothing much gets done there. The numbers bear him out: the political correspondent Gavan Reilly has calculated that the institutions have been suspended for almost 10 cumulative years since 1998, 37% of their existence spent not existing. Burrows was right about the machine. He was only wrong about what the machine holds.
Because that is the question consuming many columns, blogs, podcasts and brains thinking about the north this summer: why nothing much gets done, and specifically why Sinn Féin, nine months from an Assembly election and freshly arrived at a five-year low in the polls, has so little to show for a full term at the head of the Executive.
Two answers are on the table. The first, and by now the settled verdict of the commentariat, is incompetence. Muiris Ó Súilleabháin argued it most fully in The Pensive Quill this month; Wee Sambo McBride, in the Sunday Independent, put it flattest, that in a couple of areas the party can blame the DUP but the main problem is deep incompetence. Bills badly drafted, ministers floundering under questioning, a flagship policy reversed inside a day, ambitions shrunk to potholes and road resurfacing schemes. The second answer, offered nightly by the party’s opponents, is sabotage: the old wreckers, in office at last, wrecking still.
There is a third answer. It is the one nobody making the running wants, because it is the only one whose remedy is not a change of the personnel.
The Two Cures That Are the Same Cure
Notice what the incompetence reading and the sabotage reading have in common. Both put the failure down to the people. If the problem is incompetence, the cure is better people. If the problem is sabotage, the cure is removing the saboteurs, which the party’s opponents have always wanted. Two diagnoses, one prescription: change who is in the room. And that is precisely the conversation Sinn Féin is now having with itself, with people who defended every leadership decision for years now calling for a change of leader, a change of strategy, or both.
The third answer says it would change nothing. It says a party holding these particular offices, under these particular vetoes, invested as deeply as this one is in institutions it can neither transform nor leave, produces this record under any leadership, because the design produces it and not the personalities. Burrows was right that the machine was built to exist rather than excel. What he did not say, is that the same design does something to the party that leads it. It does not only stop the institutions working. It holds the party still.
The Specimen
Watch the mechanism work, in a single document.
This month the party’s national chairperson, Declan Kearney, published a long indictment of the DUP in the Irish News. It is a competent piece of work, and I choose that word carefully, because it is not the writing of a party that cannot make its case. It names DUP arrogance, sectarianism and what it calls economic illiteracy. It sets out the blocking of the Good Jobs Bill as an assault on workers’ rights. It cites the loss of European funding, the cut to Invest NI’s budget, the RHI scandal, the whole ledger, thorough and dated and signed.
And it asks for nothing. After the entire indictment, the demand is that the DUP change course, choose power-sharing, abandon its wrecker’s agenda. There is no consequence named for refusal, no leverage identified, no account of what Sinn Féin will do if the other party simply declines. The chairperson of the largest party in the north writes a full charge sheet against the second largest and closes by asking it, please, to behave differently.
This is not incompetence at writing indictments. It is the opposite. It is a party grown fluent in diagnosis precisely because diagnosis is the move fully available to it, and the remedies are not. You can read the shape of the enclosure in the shape of the complaint: total complaint, tiny ask, because everything in between, every step that might actually change the outcome, runs into a wall the party itself helped to build.
The Machine, Not The Man
Set two more documents beside Kearney’s, from the same fortnight.
At Hillsborough at the end of July, the Sinn Féin senator Conor Murphy, until last year the 6C Economy Minister, warned that the institutions were in a precarious place and the wheels could come off if the DUP would not engage. Days later the First Minister, Michelle O’Neill, accused the DUP of an approach that was anti-Irish and bent on frustrating progress. Then, on the Tuesday, the 6C Finance Minister John O’Dowd was asked the obvious question and answered that his party had no intent, plan or otherwise to walk away from the Executive.
Read those three in sequence. The wheels could come off; the DUP is anti-Irish; we are going nowhere. The threat and its disavowal, issued by the same party in the same news cycle. This is not a party deciding whether to bring the institutions down. It is a party performing the consideration of a thing it has already ruled out, because the performance is the only pressure available to it, and even the performance must be recalled within the week in case anyone mistakes it for a plan.
There is truth in the party’s complaint, and it deserves its full weight. The DUP does obstruct. It sat on the Good Jobs Bill. The vetoes are real, and Brian Feeney, no enemy of Sinn Féin, puts the strongest version of the party’s defence: that when the DUP 6C Education Minister diverts money from Irish-medium schools to the mainstream sector and the Sinn Féin base demands retaliation, the leadership’s refusal to fight a discriminatory tit-for-tat produces a genuine sense of powerlessness. That is fair, and it is real.
But powerlessness that is structurally produced is not the refutation of the third answer. It is the third answer. A party that cannot retaliate without becoming the thing it opposes, and cannot deliver without the consent of the party it denounces, and cannot leave without owning the wreckage, is a contained party whether it walked into the trap knowingly or not. Feeney describes the walls exactly. He simply does not call them a cage.
And the deeper point is the one the party’s own framing hides. Sinn Féin presents DUP obstruction as a thing done to it. But the antagonism serves both front benches. The DUP needs a nationalist threat to rally a fracturing unionism; Sinn Féin needs a unionist wrecker to explain a barren term and hold a slipping base. Each party’s electoral interest is served by the other’s intransigence, which is why the managed hostility survives two decades and never quite resolves and never quite collapses. It is not a malfunction of the system. It is the system, working as designed. And it has a start date. As Reilly notes, when Sinn Féin collapsed the Executive in 2017 over the cash-for-ash scandal, it changed the rules of the place: no longer would the institutions fall only at some genuine constitutional impasse, but, from then on, for the same everyday electoral reasons as any coalition anywhere. Both elections since have been triggered by a governing party pulling the plug for advantage. The threat is now a routine instrument, and both parties know how to reach for it.
The strongest objection to all this is now argued on the nationalist left: that the equilibrium is breaking, because the DUP no longer wants the institutions to survive. With the Donaldson scandal unresolved and the census trending against unionism, the argument runs, the party has decided Stormont is a losing proposition, and is hammering the cultural flashpoints, Irish language signage, Casement, precisely to goad Sinn Féin into collapsing it, since the party that pulls the plug is the party the electorate punishes. Grant it in full, and notice it does not loosen the cage. It tightens it. The gambit only works if Sinn Féin will not walk, which O’Dowd confirmed within the week; you can only bait a party into collapse if you know it is desperate not to. The escalation is not evidence against containment. It is containment weaponised: one party has read the other’s cage and begun rattling the bars, confident the door will stay shut from the inside.
What the Veto Does Not Explain
The obstruction is the party’s alibi, and the way to test an alibi is to find the cases where there is no obstructor at all.
Take the A5, the road between Derry and Aughnacloy where more than 50 people have died since 2006. It has not been built, and the reason usually given is a court ruling: a judge found in 2025 that the scheme breached the north’s Climate Change Act, the Act the Executive itself passed in 2022. Sinn Féin’s 6C infrastructure minister is appealing. But strip the case to its frame and there is no unionist in it. Every party in the Executive has offered to amend the climate law to let the road proceed. The obstacle is that Sinn Féin passed a climate Act it is proud of and promised a road it cannot now build without unpicking that Act, and either choice costs it with a different half of its own coalition: the urban, younger, greener voters who want the targets, and the rural western voters who want the road. So it appeals to the court and waits, because the court is the one place a party can send a decision it does not want to make itself. It is not the DUP. It is a party colliding with the incompatibility of its own promises, and choosing paralysis over either bill. Strip the unionist out of almost any of these rows, the student-number cap that no party will lift because none will raise fees, the budget that cannot be set, and the same shape recurs: the machine refusing to move, with a Sinn Féin hand resting on a control that will not turn. Not incompetence, and not sabotage.
The Falsifier That Did Not Fire
If the third answer is wrong, there is a clean way to prove it. If the failure is really a matter of personnel, a change of course would follow from the diagnosis the party itself has published, and the remedy that fits its own charge sheet has now been named, in print, by a sympathetic hand. Brian Feeney has argued that after next year’s election Sinn Féin should renew its mandate as the largest party but decline to form an executive until the British government publishes the criteria by which it would call a border poll. Not a walkout, which Feeney explicitly rejects as leverage thrown away, but a condition: no return to office until the terms of the constitutional question are on paper.
It is the move that follows if you take the diagnosis seriously, and it is the move the party will not make. O’Neill has spent the summer publicly recommitting to the very institutions her chairman describes as wrecked. O’Dowd has ruled out even the leverage. The party has looked directly at the one remedy consistent with its own analysis and declined it, in front of its own base, in the worst polling of its northern life.
Why? Because the remedy is not a personnel move. It is a decision about the institutions themselves, and a party contained by those institutions cannot make one. It has invested too much to transform them and cannot leave them without owning the instability that follows. That is not a failure of nerve a braver leader or a sharper minister would fix. It is the structural position, and it would hold whoever stood in O’Neill’s place.
What New People Cannot Fix
So the conversation the party is having with itself, about leadership and strategy and whether McDonald should stay and who might do the northern job better, is a conversation about the wrong thing. It assumes the record is a personnel problem. Change the people, sharpen the message, turn up to the studios, and the term after next will look different.
It will not. The party will produce new faces and the same record, because the faces were never the constraint. The constraint is a set of institutions built to exist and not to excel, entered by a party that can now neither transform them nor leave them. Ó Súilleabháin, who has watched this from far closer than most ever will, puts it in a phrase I will borrow once and not try to improve: the revolutionary has become the custodian. What his incompetence reading stops short of is why. Not a loss of edge or talent, but the plain fact that the institutions reward custodianship and punish everything else, and the party chose to walk in and hold office rather than stay out and hold nothing.
That is the diagnosis. It would be a poor essay that stopped there, because a diagnosis with no way out is just a more sophisticated despair. So the harder question: is there a way out of the cage, and has anyone ever found one?
The Ones Who Tried
There is a way out, but the record of everyone who has reached for it is a warning before it is an encouragement, and the warning has to be heard first.
What Sinn Féin is ultimately trying to do, change the constitutional status of a territory by consent, through the ballot box, from inside the institutions of the state it wants to leave, has been attempted before and has mostly failed. Quebec is the sharpest case. The Parti Québécois governed the province and controlled the timing of its own referendums, advantages Sinn Féin can only dream of, and in 1995 it lost by half a percentage point on a turnout of 93%. Then the defeat did its work: the party slid from 43% in 1998 to 15% and 3 seats by 2022, its worst ever. Coming within a whisker of a state and losing did not just fail to deliver the country. It broke the movement for a generation. Scotland tells the same story: the referendum won in 2014, lost 55% to 45%, and then the question frozen, because Westminster declined a second vote and the courts confirmed there was no route to one without London’s consent.
Lay the northern numbers over that pattern. Support for a united Ireland in the north sits at around 34% on the most careful polling, risen from 27% but pointing, on the pollsters’ own reading, to a referendum that would be soundly defeated. The newest single survey put it lower and falling. A border poll held now loses, and a border poll lost freezes the question for a generation. This is the trap folded inside the cage: the movement most desperate for the vote is the movement most tempted to call it before it can be won, because the base wants it and the story requires it. That temptation is how you turn a bad decade into a lost generation.
The One Who Did It
There is one movement that reached this exact objective and held it, and everything worth learning is in how.
South Africa. A liberation movement that had waged an armed struggle achieved the total constitutional transformation of its state and took power without civil war. Republicans reach for the comparison often, and usually for the wrong part of it. The lesson is not the struggle and it is not the moral victory. It is the settlement, and specifically the concessions. The African National Congress did not win by defeating its enemy. Joe Slovo, the guerrilla leader who did more than anyone to design the transition, wrote that they were negotiating with a regime that could no longer rule as before but was “not a defeated enemy”. Neither side could impose, so the movement won the other side’s consent by guaranteeing the minority a stake: a government of national unity for a fixed term, posts shared by proportion of the vote, protected employment for the overwhelmingly white civil service. Slovo sold it to a furious base by insisting that a concession which looked like a trap was the price of a transition they could not otherwise force. The movement guaranteed the very people it had every reason to punish, because the guarantee was what made majority rule acceptable to them.
Two other things carried it. It was never only the party: the accord that opened the road was signed by the government and a wide front of unions and churches and civic bodies, far larger than the ANC alone. And Slovo understood the danger inside his own movement, warning that a leadership which hid the necessary compromises from its supporters would in the end be accused of selling out. The honesty was part of the strategy.
What Would Have to Be Given
Hold South Africa against the northern case and the way out takes a shape the party will not like, which is exactly why it is the way out and not another comfort.
First, the referendum is the finish line, not the starting gun, and it is run once. Everything the movement can control happens before the vote. A concrete, costed prospectus for what a new Ireland would be, its health service, its pensions, its constitution, its guarantees, is not a distraction from the campaign. It is the campaign, and it is the work the office leaves no time for, because the office rewards the daily performance of grievance over the slow labour of preparation.
Second, and this is the sunset-clause lesson, you win by guaranteeing the minority, not by outnumbering it. The one number in the northern polling genuinely moving in the right direction is not the headline unity figure, which is flat or falling. It is unionist consent: the willingness of unionists even to accept a result they would lose has begun, measurably, to rise. That is the South African mechanism showing its first Irish pulse, and it is the only variable that describes a durable yes. An offer addressed to unionists with real guarantees, of their institutions and their identity and their place, is not a betrayal of the objective. It is the only version of it that has ever worked anywhere.
Third, it cannot be the party’s possession. The South African transformation was carried by a front far wider than the ANC, and the northern unity project is understood, fatally, as Sinn Féin’s property. The party that fought the war cannot be the party that reassures the people it fought. A civic movement, cross-party and cross-community, holding the preparation and making the guarantees, is not a threat to the goal. It is the only vehicle for it.
Notice what every one of those three requires. Preparation over agitation, guarantees over numbers, a front instead of a party: each is the precise opposite of what the office rewards, which is administration and grievance, the daily holding of the base. The revolutionary became the custodian, and the custodian cannot prepare the revolution, because being the custodian is a full-time job.
The Front It Can Only Build of Itself
This is not a hypothetical, and to the party’s credit it is not asleep. In the middle of August, in An Phoblacht, Declan Kearney published the fullest statement of intent the unity project has produced in years, and it should be met as one. It names the Fleadh as proof the north has changed, invites unionists to join the debate, insists that Protestant and unionist people must have a fundamental influence in the creation and governance of a united Ireland, and calls for something concrete: a single forum in which, in his own phrase, all strands of “Irish national democratic opinion” would gather to deliberate, agree, and then speak as one on how to plan the transition. He models it, more than once, on the Hume-Adams initiative of the late 1980s.
Take it seriously and it is the first real attempt to do what the third lesson demands, to move the project beyond the party and formalise a shared position. And the honest objection is not that self-organisation is illegitimate. It is exactly what the liberation movement in South Africa did first: the Harare Declaration, the internal front, the unions and civics agreeing their own position years before they sat across from the regime. A movement is entitled to work out what it thinks before it negotiates. So far, Kearney is on solid ground, and an essay that jeered at a movement for organising itself would be a dishonest one.
The distinction is what the self-organisation is a step toward, and here the two roads part completely. In South Africa the movement’s own front was never, from within, a front of one community: the Congress tradition was non-racial by principle, so the voice that spoke already contained members of the minority it would later reassure. And that front was a prelude, not the venue. It gave way to a genuine multi-party negotiation, where the National Party and Inkatha and white interests sat as full parties with real standing, and to a peace accord the outgoing regime itself signed. The minority co-authored the settlement; it was not its audience. The single voice dissolved, deliberately, into a table where the other side could say no.
There is no second stage in Kearney’s account. The forum gathers, agrees, and speaks as one, and that single voice is offered to unionism from across a threshold, not worked out with unionism inside the room. Protestants are invited to the debate; the body that deliberates and decides is defined, in his own phrase, by the one tradition. And this is not a generosity he could add by inviting more people, which is the point that turns the whole essay. The ANC could hold the minority within its own ranks because its founding principle was a single non-racial people. Nationalism cannot hold unionists within a forum of national democratic opinion without dissolving the very thing unionism is, because unionism is precisely the refusal of that category. The boundary of the forum is not a failure of reach. It is structural. When the unity project sets out to build a front, the only front it can convene is a front of itself.
So here, at the level of the whole movement, is the cage one last time, and its deepest instance. The guarantee that would win, the one made with the minority at the table, cannot be issued by a party whose forum the minority stands outside by definition. The party that fought the war reaches, sincerely, for the front that wins, and its hand closes on a convention of its own tradition, chaired in its own paper, over its own president’s meetings, speaking with one voice that can only ever be its own. It has built the first half of the South African road and named it the whole journey. Not because Declan Kearney lacks the wit to see the difference, but because the office and the tradition that hold him leave no other front within reach.
The way out is real, then, and narrow, and no comparison is a template: South Africa was a demographic majority claiming power it was owed, and northern nationalism is not a majority, not yet anyway, and maybe not even soon. The path has been walked once, in far harder conditions than these, by a movement that guaranteed its enemy and prepared for years and made itself larger than a party. The keys are in Sinn Féin’s own hand. They have always been able to open the door out. What they cannot do, while the office holds them, is stop mistaking the warmth of the cell for the freedom of the house.
References
1. Jon Burrows, address to the Patrick MacGill Summer School, 25 July 2026 (“designed to exist, not to excel”). Reported in the Irish Times.
2. Gavan Reilly, “37% of the time, Stormont hasn’t existed,” The Journal, “Politics by Numbers” series, August 2026 (the 37% cumulative-suspension figure; the 2017 Rubicon framing).
3. Muiris Ó Súilleabháin, “Hope Springs Eternal,” The Pensive Quill, August 2026 (the incompetence reading; “the revolutionary has become the custodian”).
4. Sam McBride, Sunday Independent, August 2026 (competence as the “main problem”; discussed in Cedar Lounge Revolution).
5. Declan Kearney, “Time for a new initiative on Irish unity,” An Phoblacht, 14 August 2026 (the DUP indictment; the call for a single forum of “Irish national democratic opinion”; the Hume-Adams model; unionist and Protestant influence in a united Ireland).
6. Conor Murphy, remarks at Hillsborough, late July 2026 (institutions “precarious,” “wheels could come off”).
7. Michelle O’Neill, remarks on the DUP, 6 August 2026 (“anti-Irish,” “frustrating progress”).
8. John O’Dowd, 11 August 2026 (“no intent, plan or otherwise”).
9. Brian Feeney, Irish News, August 2026 (the “sense of powerlessness”; the conditional-re-entry proposal tied to publication of border-poll criteria).
10. A5 Western Transport Corridor: High Court ruling on Climate Change Act (2022) grounds, 2025; ministerial appeal.
11. ARINS / Irish Times North-South Survey (North 34% for unity, released February 2025); Institute of Irish Studies, University of Liverpool poll (June 2026, North 35.8%); the rising-unionist-consent finding, ARINS.
12. Quebec referendum, 30 October 1995 (49.42% Yes); Parti Québécois decline to 2022. Scottish independence referendum, 18 September 2014 (44.7% Yes).
13. Joe Slovo, “Negotiations: What Room for Compromise?” (African Communist, 1992), and the sunset-clause / Government of National Unity framework; the National Peace Accord, 1991.
⏩Louth For Ever writes on Irish politics and constitutional change. Follow for analysis of Ireland’s democratic future as it’s constructed by those actually engaged in the work.
