There is no way for NI to remain in the EU after Brexit. Alternatives exist, but they all come at a price.

Last week, Kevin O’Rourke wrote in the Irish Times:

It is logically coherent, if lunatic, to argue that Ireland should quit the EU and join the UK customs union. (Leaving the EU would on its own obviously not suffice to avoid a North-South border: our exit from the EU would have to be of the red, white, and blue variety.)

It is also logically coherent to argue that Northern Ireland should remain within the EU, and I wish it would. That seems like something worth arguing for. But it is logically incoherent to argue that if we remain in the EU and its customs union, and the North leaves both, there can be some special deal that will avoid the need for a customs frontier on the island.

Those who say Ireland should leave the EU know they are in a small minority. Many will not come out and argue for their position particularly strongly for fear of being laughed out of court. The evidence that our prosperity is based on EU membership is overwhelming. Still, expect them, in the months and years ahead, to claim that the return of a customs frontier somehow shows that “the EU” has let Ireland down.

The Brexit campaign shows that such dishonesty can pay. Which is why it is so important that everyone understand that if the North leaves the EU and its customs union, and we remain inside it, there is nothing that the EU or anyone else can do to prevent the return of such a frontier.

The article suffers from muddled thinking when it proposes solutions though. It is not legally possible for NI to remain in the EU while it also remains part of a post Brexit UK. Only sovereign states are members of the EU. This is not a matter of geographical exemptions. It is a matter of political representation, sovereignty and treaty law.

The Greenland precedent is not helpful – in that case Greenland gained an exemption from EU law while the sovereign state (Denmark) remained a member. If that were to be applied to Brexit, then England, Scotland and Wales would have an exemption from the EU while the UK as a whole remained a member, but only in order to represent the people of NI. Such an arrangement would be preposterous.

NI could remain de facto part of the single market but only by a) breaking the UK internal market instead and b) submitting to fax democracy, where NI would have its laws written for it in the EU without any representation at the table (short of what the Irish government might provide gratis). This would require all EU competencies to be devolved to NI, including employment and immigration law. It would likely also require full fiscal autonomy. This would not be devolution, but something closer to Gibraltar-style internal autonomy. Given how deeply the Irish government would then be embedded in the domestic legislative process, it would probably be both simpler and more honest to just have a united Ireland. At least in that scenario, EU agricultural and structural subsidies would continue (but the block grant would vanish).

An easier sell would be to keep NI de facto inside the EU customs union by shifting the RoI’s customs frontier to Larne. This would be an arrangement similar to that of Monaco, which has no official treaty with the EU, but which by bilateral agreement allows France to treat Monaco’s sea border as a French border for all practical purposes. This of course means that the UK’s internal customs union would be severed instead, but such arrangements have precedent. Büsingen am Hochrhein is a small exclave of Germany entirely surrounded by Switzerland, and has long been treated as part of Switzerland for customs purposes. Whether this would be sufficient to save the cross-border economy would depend on how generous and how quickly the new EU/UK free trade deal turns out (don’t hold your breath).

The big unknown is the Common Travel Area. It is just possible that the EU may allow it to continue with the current spot check enforcement regime, considering that the UK’s immigration law is likely to be stricter than the EU’s for the foreseeable future. But this would require a generous helping of goodwill from all sides. Alternatively, it may be possible to move the RoI’s hard immigration frontier to Larne.

None of the above are cost-free options. The emotional cost to Unionists in particular would be heavy no matter what sort of deal is struck, but any arrangement that keeps an open border would surely be cheaper and less disruptive than the default alternative.

A united Ireland is as far off as ever, for the usual well-worn reasons. That may change if after Brexit the UK decides to alter economic policy to NI’s disadvantage (e.g. by cutting agricultural subsidies) or if Indyref2 comes to pass. But NI is still incapable of supporting itself, and it is proportionately less of a burden on London than it would be on Dublin (which is still in primary deficit, let us not forget).

The best case scenario for Ireland (north and south) remains for Brexit to magically go away. But barring a political upset of Canadian proportions that’s a pipe dream.

(This post was first published on Facebook, and then on Slugger)

After Brexit: the options

As time went on, speculation reached fever pitch. The internet was full of crazy theories about developments that would overturn everything we knew, seemingly convincing timelines of events that would soon unfold, and ingenious analyses proving that things were either nowhere near as bad as they seemed, or much worse than anyone imagined.

I am of course talking about Game of Thrones.

And in the end (no spoilers, I promise!) the vast majority of these excitable scribblings fell away, shown to be the half-baked wish-fulfillment that they always were. The best plot twists in a good story, whether fictional or historical, are the ones that hindsight tells us we should have been able to predict, but didn’t because we were distracted by something else. Yet given enough monkeys with typewriters, there will always be somebody who called it right, whether by divine inspiration or by dumb luck.

The same will apply to Brexit. Fevered speculation at the moment is concentrating on a second referendum to reverse the outcome of the first, Article 50 never being invoked, a federal UK being both inside the EU and outside it simultaneously, England declaring independence from Scotland and other such clever, clever schemes. I’ve even indulged in it myself, enjoying the endless horizon of possibility. But there are constraints on the seemingly infinite selection of options available, and it is vastly more likely that the outcome will be disappointingly prosaic. So, in that spirit I’ll try to lay out the less clever scenarios that have some chance of actually happening.


The “referendum, what referendum?” option.

It is entirely possible that nothing will happen, that Article 50 will never be invoked, that some excuse will be found to backtrack on the referendum result – but this would be a risky option to take, both domestically and in the EU. Domestically, simply ignoring a referendum result (no matter how many times you explain “non-binding”) would call into question the legitimacy of the government. And the UK’s position in the EU is already compromised – no political leader would be able to deal with the UK in good faith while the Damoclean sword of a surprise Brexit remained dangling overhead. Cooler heads will prevail, and talk of forcing the UK’s hand will quiet down, but toothpaste cannot be squeezed back into the tube, and nobody’s patience is infinite.

The only theoretical scenario where UK/EU relations could return to the status quo ante would be following a LibDem general election landslide on a manifesto commitment to grovel for forgiveness. Much more likely that the government would collapse, perhaps repeatedly, and the UK’s chair in Europe sits empty while the rest of the world moves on without it.

Likelihood: **… Tempting, but unsustainable.

Customs Union / Common Market

The Turkish option.

The Customs Union is one of the oldest parts of the EU, dating back to the EEC’s founding in the 1950s. It is likely that the UK will retain CU access in any negotiations, as its terms are relatively uncontroversial. It allows for tariff-free trade between its members, a common external tariff rate and a common negotiating position in trade talks with third countries (e.g. TTIP). There is no requirement for freedom of movement in the CU, so this would allow for the immigration changes that were the central issue in the Leave campaign.

The only wrinkle would be that some Leave campaigners (UKIP in particular) were in favour of “negotiating our own trade deals”, which is incompatible with CU membership, but of all the promises to take back this would be one of the easiest. And it could be argued that “access to the Single Market” would be satisfied by the Customs Union (“access to” being distinct from “membership of”).

Turkey, Andorra , San Marino and Monaco already have this relationship with the EU. Norway, Iceland, Liechtenstein and Switzerland however do not, the main sticking points being agriculture and fisheries which are protected by (mutually) high tariff barriers in each case.

Likelihood: ***** With or without SM, see below.

Single Market

The Swiss option.

People often talk about the Single Market as if it and the Common Market were synonymous, but the SM is a more recent development, coming in to force in stages leading up to (and beyond) the original target date of 31 Dec 1992. Its main innovation is the delegation of market regulation to the European institutions, in order to more quickly remove non-tariff barriers to trade between its members – earlier efforts having ground to a halt under repeated national vetoes. The SM is what most people think of when you say “EU”, and it is the creation of the SM that first caused the UK to rethink its relationship with the (then) EEC, leading to Thatcher’s handbag deployments and the Sun’s infamous “Up Yours Delors” front page.

It cannot be repeated often enough that it is the conditions of the Single Market (freedom of movement, laws written in Brussels, budget contributions) that the Leave campaign was aimed squarely at.

It is possible for the UK to stay in the Single Market, either through EEA membership (this would need the consent of Norway, Iceland and Liechtenstein, which would complicate negotiations) or more likely through a Swiss-style bilateral treaty. In either case, it would mean reneging on some or all of the main promises of the Leave campaign. It is worth noting that the Swiss deal does not include financial services – the UK would be desperate to have these included, making its negotiating position in other areas weaker.

(There seems to be one school of thought among Leavers that the UK’s current predicament is the fault of weak politicians who won’t bang the table hard enough during negotiations, and that all the UK needs is a tougher stance to extract more concessions. To that hubris one can only say: if Thatcher’s handbag couldn’t get the concessions you want, what chance do you think you have?)

There is currently no non-EU country that is a member of both the Customs Union and the Single Market, but there is no reason in principle why this could not happen. In practice, simultaneous membership of the CU and SM is effectively full EU membership without the CAP and without a vote. It would be hard to sell any Single Market option (either inside or outside the CU) as a victory for democracy, and a Conservative government that came back with a deal that failed to address the three main Leave complaints (immigration, regulation, taxation) would be vulnerable to attack from UKIP (who are already positioning themselves in anticipation).

Despite those concerns CU+SM is the most likely scenario, one that most Remainers and a sufficient number of Leavers will accept. But because of those concerns, it will probably not be the last word on the subject.

Likelihood: ****. (with or without CU) Minimum economic chaos, maximum political chaos.

Scottish Independence

If Scotland wants to remain in the EU after Brexit, then the only way to do so is to declare independence from the UK. EU members are by definition signatories to the Treaty on European Union (aka the Maastricht Treaty), and only sovereign states under international law have the capacity to sign treaties. The Greenland precedent does not apply, because in that case it is Denmark who is the signatory and therefore the member. Opt-outs and opt-ins are not mirror images. Any territory wanting to be a member of the EU must do so via a sovereign signatory, whether that be itself or its parent state. That means that clever workarounds such as a federal UK will have no bearing on EU membership. You’re either a sovereign state or you’re not.

And this hard choice may be the undoing of Scottish independence, at least in the short term. Scottish resentment may have grown, but its economic predicament has just become more difficult – even setting aside the oil price. Before Brexit, it was a plank of the independence movement that an open border would remain between Scotland and the rUK on the same basis as its Irish counterpart. If the rUK tightens its immigration rules and/or leaves the customs union after Brexit, then a hard(er) border would have to be drawn.

It should be noted however that one commonly-cited impediment to Scottish independence within the EU is nonsense. Scotland cannot be forced to adopt the Euro if it does not want to, even if it is made to sign away the UK’s treaty exemption. Sweden shows the way:

Sweden maintains that joining the ERM II (a requirement for euro adoption) is voluntary, and has chosen to remain outside pending public approval by a referendum, thereby intentionally avoiding the fulfilment of the adoption requirements.

Likelihood: **… (Short term) ****. (Eventually)

The Catalan Precedent

The big “if”.

If both Brexit and Scottish independence happen, and if they happen within a short enough timeframe, it may be possible to strike a deal to allow Scotland to continue its EU membership – either by inheriting the UK’s membership at the moment of independence (which would require a feat of choreography and an innovative interpretation of international law) or by preparing well in advance so that Scotland could be admitted at the stroke of a pen afterwards. Scotland is currently enjoying a moment of remarkable goodwill, but this will only translate into a special arrangement if it can be written in a way that does not set a Catalan precedent, and therefore escapes a Spanish (or Belgian) veto.

A Catalan precedent would be set if the EU allowed a region of a member state to secede and afterwards join the EU without going through the full accession process from scratch (which under Article 49 can be vetoed by any member state). Spain fears that Catalonia and the Basque country would be encouraged to secede if they were able to join the EU afterwards, so threatens to use its Article 49 veto on their (hypothetical) membership applications. If a precedent were to be set that Article 49 could be bypassed, then this veto threat would be worthless. So Spain (and for similar reasons, Belgium) will in turn veto anything that might set such a precedent.

This was brought up in the original indyref as a stumbling block. Scottish independence would have to be followed by the full membership process, just like everyone else. This process would be shorter than for most countries, as Scotland already applies all relevant EU law, but would still be onerous. The UK might be able to persuade Spain and Belgium not to veto Scottish membership outright, on the basis that an amicable divorce would not set a precedent for a contentious one. But Article 49 would have to be obeyed.

In the case of Brexit, the UK would no longer be a member of the EU after Scottish independence – unlike the Catalan case where Spain would absolutely want to stay in – so it could (theoretically) be argued that this would not set a Catalan precedent either. But this decision would ultimately be a political one, and would depend heavily on the mood music. Expect utter carnage if Gibraltar comes up at any point in the discussions.

Likelihood: ????? This bookie is refusing to take bets.

The Irish Border

If the UK stays in the Customs Union then there will be no need for customs controls, and the smugglers will be disappointed – more so if the UK also stays in the Single Market. The great unknown then boils down mostly to immigration control. The current Common Travel Area arrangement relies on the immigration rules for the UK and Ireland remaining broadly equivalent. Shared membership (or non-membership) of the EU, and shared non-membership of Schengen meant that the two states were mostly free to arrange this between themselves. But if the UK tightens its rules regarding EU citizens then Ireland will, for the first time since independence, be unable to follow suit due to its EU obligations.

The only recourse in that event will be for the UK to impose immigration controls on travel to the UK from Ireland, to prevent EU citizens from entering the UK by the back door. Ireland may be able to get away without imposing controls, so long as UK immigration rules are consistently stricter (you only need to impose controls going from a less-strict jurisdiction to a more-strict one, not the other way).

If that is the case, then the most likely arrangement will be to impose controls at the sea/air border, on travel from the island of Ireland to GB. Security checks take place on all sea and air routes already, and these will be further hardened into de-facto immigration controls (but without admitting it, for fear of upsetting the unionists). This effectively leaves NI outside the UK border (and inside the EU) for immigration purposes, but inside the UK for all other purposes.

The price will be more sharing of data between the two border agencies, tolerating EU citizens entering NI so long as they don’t try to get a job, and annoying the unionists who see through it.

Likelihood: ****.

A Border Poll

There will be no border poll for the foreseeable future. The GFA is clear on the conditions, and any decision made by the SoS without supporting evidence would be subject to judicial review.

(And on a side note, Sinn Fein might take a leaf from Cameron’s book and stop banging on about about a border poll – it would do no harm to their unionist outreach programme, if they care about it that is…)

In the longer term however, there may be hints of movement. Young people-of-a-unionist-background of a liberal or left-wing persuasion have been openly questioning the UK government’s (and English voters’) ability to act in NI’s best interests, and their own commitment to a UK-wide identity. While this will not translate into any significant change on the border issue in the near future, it does demonstrate that young people-of-a-unionist-background (we really need better names for these things) do not have infinite patience with the UK government either.

And don’t forget, the practical arguments against a UI haven’t gone away you know.

Likelihood: ….. (Short term) ***.. (Eventually)


The Hail Mary pass.

It is of course possible that more countries will threaten to follow the UK out the door. If this happens before the UK actually leaves, then there is a slim chance that they could collectively demand treaty change as a price for changing their minds. The UK by itself has already got as much as the EU will give, and has vanishing hope of yet another renegotiation (ah, go on). But with France, Netherlands, Denmark joining in?

Europe has always proceeded by crisis, but a treaty renegotiation in the middle of the current one may expose gulfs too wide to bridge. See the next option.

Likelihood: *…. Stranger things have happened.


The hospital pass.

On the other hand, the spectacle of a second, and then a third country voting to leave the EU could trigger a mass loss of confidence and a sudden collapse. Morsels such as the Customs Union might be salvaged from the wreckage, but otherwise it would be a blank slate and a step into the great unknown. Governments freed from the corpus of EU law would be quick to enact reforms in the narrow national interest to deal with the economic and social fallout. Some of these reforms will be more effective than others, some will be less liberal than others and some will be downright nasty.

Likelihood: *…. Terrifyingly possible

(This post was first published over on Slugger)

The cure for creeping: four-state traffic lights

Since moving to Dublin, I’ve noticed a phenomenon that I don’t remember seeing in the north, and is rare even in Galway. At traffic lights, the first car in the queue will regularly start moving forwards long before the lights turn green. In extreme cases, presumably because the driver has misremembered the order of the phases, this can result in a car being left stranded in the middle of the junction while traffic from another direction attempts to negotiate around it.

I’m convinced (albeit unscientifically) that this results from the combination of two main factors: the desire, particularly common in cities, to be prompt in clearing the junction; and the lack of advance notice that a three-state traffic light provides. This results in a hair-trigger disposition in many drivers and a high rate of premature action.

To clarify, a three-state traffic light goes from green to amber to red on stop, but straight from red to green on start. This system is used in France, the US and the RoI (amongst others). The amber “prepare to stop” light is necessary to avoid a sudden change from green to red, which could obviously result in driver panic and serious accidents, but no similar facility exists going from red to green.

A four-state light, as found in the UK and Germany, adds a red+amber “prepare to go” state between red and green so that drivers do not need to remain in a prolonged state of readiness. This is useful for any vehicle that takes time to put into gear, or has a handbrake, or whose driver’s mind is likely to wander.

If the RoI were to introduce this fourth state into its traffic lights, I’m confident the incidence of creeping through red could be significantly reduced. And since all our new traffic lights are made in Germany by Siemens, the change is likely nothing more than a setting in the software, so could be introduced at minimal cost.

The flaxen revolution

I find myself in the unaccustomed, even uncomfortable position of agreeing with Jamie Bryson. This may be because in a democratic system at least some of the people are happy, while an undemocratic system is offensive to everyone except those in power (and in the case of NI, perhaps not even them).

NI is not a democracy, and grafting standard-issue “democratic institutions” on top of the current state of society will not magically make it one – we had those before in Old Stormont and it did not work. There are preconditions for majority rule to succeed as a democratic system – the willingness of voters to hold the elected to account by withholding or changing their vote; the commitment of the elected to act in the interest of those who did not vote for them; the resolve of both electors and elected to submit to the rule of law and uphold it impartially. None of these conditions prevailed in 1922 and few of them prevail today.

All this is well understood. What is less well understood is that the current system is not a route out of the undemocratic rut but merely a different, less violent rut. We must ask ourselves what we are trying to build – is it a society where power changes hands in the polling booth and the strong are accountable to the weak? Or is it merely a society where anyone likely to make trouble is kept placated with the spoils of office? We have achieved the latter, but seem as far away as ever from the former.

The usual argument against majority rule in NI is that the communal parties can’t be trusted with sole power, an argument given force by history. It is clear that whatever comes after must include safeguards against abuse of power at least as strong as the ones we have today. Those who wish to abolish mandatory coalition have proposed several alternatives, none entirely satisfactory. The most common are weighted or dual majorities, which require a majority of the entire Assembly and a minimum percentage of each community in order for a bill to pass (and in some variants, for the Executive to be elected). The problem comes when that minimum percentage is defined. Too low, and it raises the prospect that one particular party could be ganged up on and excluded forever, even if it were the largest single party. Too high, and it looks little different from mandatory coalition. It is not clear if a sweet spot exists, or if such a system would be sufficiently radical to be worth the effort.

The problem in all these proposals is that they treat the symptoms rather than the cause. Sectarian violence, cultural conflict and ethnic nationalism are far from unique problems – most countries have suffered to some degree from the same. What makes the few hard cases such as NI, Bosnia, and Lebanon special is the way these problems have so comprehensively infiltrated the political system. In each case, structures have been created to manage the symptoms of a dysfunctional politics – but the relationship between the system and its participants runs both ways. We soon find that because the structures of state have been shaped to fit the dysfunctional, only the dysfunctional fit into them.

On paper, there is little wrong with the Assembly as it stands. Everyone is represented, everyone gets a turn, nobody is excluded – and business can still continue if the participants are willing to compromise. The problem is prima facie not with the rules, but the players. However, there is little incentive for parties to compromise when the system rewards brinkmanship and extremism, so maybe the problem really is with the rules.

If we look at how Stormont differs from other, more successful systems, we see that the main innovations are motivated by the desire to prevent exclusion of one ethno-national grouping by another. This is understandable, as this is exactly the problem Old Stormont had, and the one that nationalists in particular still rightly fear. But it leaves as an unstated assumption that ethno-national groupings and political parties are more or less the same thing – and this is a peculiarly NI phenomenon. If we compare to the USA, where the treatment of the black minority was incomparably more brutal than anything dreamed up under unionist rule, we do not see a “black party” in permanent conflict with a “white party”. There may be imbalances in the percentages of different ethnic groups in the political parties, but these are as much a reflection of socioeconomic status as they are of identity. Before the Democrat Barack Obama, the most successful black politician in America was the Republican Colin Powell. The progress of black politicians is a bipartisan success story, and it is also in many ways a colour-blind one. Obama’s voting demographic was correlated with ethnic origin, but by nowhere near the degree we have come to expect in North Belfast.

Of course, black Americans do not seek the overthrow of the state. But if the polls are to be believed, neither do most Irish Nationalists. And in any case this in itself cannot be the reason for NI’s special problems, because Lebanon has no significant separatist movement and still suffered an ethno-religious civil war far worse than ours.

The common thread that separates Lebanon, Bosnia and ourselves from the US and other countries that are making progress with their divided past is the primacy of identity politics over practical considerations. Catholic Unionists and Protestant Shinners do exist, but they remain remarkable exceptions. Half-hearted “outreach” attempts are undertaken not in the hope of converting the doubtful, but of soothing those whose consciences are troubled by this obvious lack of progress.

Any attempt to fix NI must address this immovable object. Removing the Border from the competency of the Assembly was supposed to make normal politics possible. It did not. The constitutional question “went away”, but the parties merely found something else to fight over. Identity politics abhors a vacuum – no doubt if Westminster took flags and emblems off the hands of local politicians, they’d find some other reason to despise each other.

Because it’s not about the Border. It’s not about the flag, or a parade, or the Irish language. These things are excuses, rationalisations, bedtime stories we tell ourselves to make our own fears sound less preposterous. Greater problems have been solved over dinner and cigars. We can’t stand to make common cause with each other because in our gut we know we are still sharing a small box with someone, some thing that once hurt us. This is not going to be solved quickly. But we can make a conscious, rational decision to stop picking at the scab and making it worse.

The long term solution is the integration of our divided society. This will take generations, and it would be foolish to prejudge the details of the world that our grandchildren will build without us. The most important, the only important condition is that it is a whole society, where tales of unionists and nationalists are as misty and abstract as those of roundheads and cavaliers, and where children have to search the archives to discover which sides their ancestors were on.

The short term solution is to cut the Gordian knot of mutual suspicion and veto so that we can at least take the first steps. The current Assembly couldn’t even agree to save itself a fortune by merging two teacher training colleges, so there is no hope of it ever addressing an actual problem. Despite their protestations, the communal parties are too comfortable where they are right now, because getting where they are right now did not require them to grow, or learn. When SF and the DUP had a problem working together they fiddled the system so that they could avoid doing anything really difficult, such as voting for the same First Minister.

No, change is not going to come from within the Assembly – at least not with the current balance of parties. The first alternative is for the governments, perhaps again with the help of the Americans, to step in and impose order upon chaos. This seems unlikely, given the increasing signs of crisis fatigue amongst outsiders. And in the long term, this would perpetuate the infantilisation of NI, where the locals can’t be relied upon to sort themselves out and so will require feeding, changing and babysitting indefinitely. The other alternative is change imposed from below, from the electorate. But with turnout approaching 50% and new political parties failing to make headway, the only outlet remaining would appear to be the street.

And here I find myself agreeing with Bryson again, although for different reasons. Street politics comes in a variety of forms – sometimes joyful, sometimes angry, sometimes from the margins of society and sometimes pouring from every door and window. Sometimes the only way to fix a stagnant, failing system is with the collective will of millions. Sometimes all we need to do great things is the courage that comes from knowing that we are not alone. Sometimes society needs a little bit of revolution. A joyous, multi-coloured revolution.

A flaxen revolution?

Road numbering in NI is a mess

Any outsider trying to get from A to B in Northern Ireland would be best advised to invest in a satnav. People who regularly commute from Craigavon to Belfast are well accustomed to seeing cars with Dublin license plates barrelling westward on the M1 with panicked-looking drivers inside. This is because in the Republic, the powers that be have (sensibly) decided that a motorway should have the same number as the route that it replaces, which then reverts to a regional road with a different number. The concept of an M1 and an A1 going in completely different directions therefore takes people by complete surprise.

But that’s not all. Travelling from Belfast to Derry, one takes the M2, which is inexplicably renamed to the M22 shortly before downgrading to the A6*. One end of the M1 becomes the A4, but the other end becomes the A12. When driving from Derry to Coleraine one should not stay on the A2 for the full distance unless a scenic detour is desired. And nobody in their right mind would drive from Armagh to Newtownhamilton using the A29 via Keady. Then there are several routes that fork so that they have three or more ends (A2 in Derry, B2 in Lurgan, B3 in Gilford, B95 every half a mile…) and two that share the same number (A37 at Coleraine and also at Crossmaglen). It’s almost as if the person who was dishing out the numbers came in with a hangover one morning and forgot what he’d done the night before.

Several decades ago, the Republic completely renumbered all of its roads. The new system is consistent, logical and future proof. I would propose a limited renumbering of NI’s trunk routes on a similar basis, with the primary aim of making life easier for drivers of strategic traffic, who may not be familiar with the idiosyncrasies of the current system.

  • The renumbering would initially apply only to motorways, primary strategic routes and a few other border-crossing roads that connect to the Republic’s N routes.
  • Where routes cross the border they would have the same number on both sides.
  • Routes would use the same number for their entire length, the only change being an M to indicate a motorway.
  • The ultimate destination of each route would be clearly signposted, even if it was on the other side of the border (with one exception).
  • Route numbers would not need to be unique across the island, but would be chosen to minimise confusion and represent the relative importance of the routes.

The last point is important – this is an NI-specific scheme, and would be defined independently of the Republic’s system (making it easier to implement both logistically and politically). It would however be coordinated on a best effort basis, in the same way that neighbouring counties in the U.S. often coordinate the numbers of their border-crossing county roads while freely duplicating others. Strategic routes in NI would use a novel prefix (S for “strategic”, say) to avoid any ambiguity.

This scheme would entail significant costs, so would probably only happen as part of a larger project, perhaps including bilingual signage and/or metrication.

The new routes would be:

Primary Strategic Routes

  • S1 – A1 Sprucefield to Dublin (=N1)
  • S2 – A5 Derry to Dublin (=N2)
  • S3 – A509, A46 Dublin to Ballyshannon (via Enniskillen) (=N3)
  • M10, S10 – M2, M22, A6, A514 Belfast to Derry
  • M11, S11 – A8(M), A8 Glengormley to Larne
  • M12, S12 – M12, A3 Craigavon to Monaghan (=N12)
  • M13, S13 – A26, M2, A26, A37(N), A2, A515, A2 Antrim to Letterkenny (via Coleraine and Derry) (=N13)
  • M16, S16 – A12, M1, A4 Belfast to Sligo (=N16)
  • M31, S31 – M5, A2 Belfast to Carrick
  • M32, S32 – M3, A2 Belfast to Bangor

Concession Routes

These short (<5km each) border-crossing sections are not considered strategic routes by NI but are by the Republic and should have consistent signage (and quality) within NI for driver convenience (and safety). It might be best if any signs on the NI side showed only the Republic’s N designations, to avoid unnecessary confusion.

  • S14 – A38 Strabane to Letterkenny (=N14)
  • S53 – A37(S) Dundalk to Castleblayney (=N53)
  • S54 – A3 Monaghan to Cavan (=N54)

S15 is reserved. The N15 joins the N14 immediately on the Republic side of the border, and so may also be signposted from within NI.

All the numbers above match those used in the Republic apart from:

  • N10 is a short route in Kilkenny that arguably doesn’t justify a low number now that the M9 exists.
  • N11 is used for the Dublin to Rosslare route, but both N11 and S11 are part of the greater Larne to Rosslare strategic route, so the numbers do not conflict.
  • N31 is a short route used for access to the port of Dun Laoghaire.
  • N32 was once used for a very short section of road near Malahide that has since been detrunked.

The numbers have been chosen so that major strategic routes have satisfyingly low numbers (fortunately the Republic numbers roads from north to south so all border-crossing roads have low numbers already) and also so as not to use a number that the Republic might choose in the future for a road close to the border (it hasn’t yet reused a deallocated number).

The S1 does not extend the entire way to Belfast but terminates on the M16. The M12 also violates the rule of thumb that side branches should have higher numbers than the parent route. I consider this preferable to having motorways change number partway down their length. The S13 is not the preferred route between its end points (and so should not be signposted end to end), however it is the preferred route between any other two significant towns along its length.**

Future application to other trunk routes

This scheme could easily be extended to all other trunk routes in NI. If so, then they might be assigned numbers greater than 100, as south of the border these are reserved for minor roads. Currently trunk roads in NI are defined chronologically rather than geographically, and follow some bizarre routes which appear designed to use the fewest route numbers rather than define routes that a driver would find useful. Fortunately these trunk numbers are not used on signage, but it would be sensible to redefine trunk roads using the new numbering to minimise the number of different schemes in use. The 15 additional trunk routes would then be (roughly anti-clockwise around Belfast):

  • S100 – A55 Outer Ring
  • S101 – A44 Ballymena to Ballycastle
  • S102 – A36 Ballymena to Larne
  • S103 – A57 Larne to Aldergrove
  • S104 – A26 Antrim to Moira
  • S105 – A31, A29, A505, A32 Castledawson to Enniskillen
  • S106 – A29 Coleraine to Armagh
  • S107 – A28 Aughnacloy to Newry
  • S108 – A27 Portadown to Newry
  • S109 – A2 Newry to Warrenpoint
  • S110 – A24, A2 Belfast to Newcastle
  • S111 – A7 Belfast to Downpatrick
  • S112 – A22 Belfast to Comber
  • S113 – A20 Belfast to Newtownards
  • S114 – A21 Bangor to Comber

Trunk roads leading into Belfast (other than the three main motorways) would terminate on the Outer Ring. There are also several roads that use green signage (such as the A11 Belfast Inner Ring) but which are not legally defined as trunk roads. To minimise confusion, this practice should be discontinued.

Minor routes

The inconsistencies with minor routes could be individually addressed. There is little justification for a costly wholesale renumbering of non-trunk routes. The division into A, B and C classes is broadly sensible as it reflects the importance of a route and thereby implies the standard of road a driver should expect. Usually.

Some suggestions:

  • Renumber the B32 to A29 so that the A29 runs straight up the high street of Keady and out the Castleblayney road. The old A29 from Keady to Newtownhamilton could then be an extension of the B132, and the B31 would continue on the mainline from Armagh to Dundalk (no need for an A classification as the preferred route goes via Newry).
  • The Tandragee-Keady-Monaghan, Lurgan-Scarva and Loughbrickland-Rathfriland branches of the B3 should get separate numbers. The Scarva-Loughbrickland section is of such poor quality that it should probably be downgraded.
  • The B2 is a shattered mess. There are several identifiable sections: 1. a sensible section from Downpatrick to Lurgan. 2. a hairpin shaped section from Lurgan to Portadown via Charlestown. This should be renumbered. 3. A section from Lurgan to Tandragee which never matched the signposted route between the two towns and has anyway since been partially overbuilt. This should be both renumbered and rerouted so that it uses the Tandragee Road to go to Tandragee. 4. The Clare Road beyond Tandragee should be renumbered as an extension of the B131.
  • The B95 also has several distinct branches: 1. Antrim to Ballyclare 2. Antrim to M2 J6 3. Templepatrick to Glengormley. 4. Scullions Road, Mayfield Link, Upper Hightown Road 5. Hightown Road (due for demotion once Mayfield Link is complete?). Each of these should have a distinct number.

One would think numbers were a scarce resource, observing the absurdities that the Roads Service creates in trying to avoid assigning new ones. Bypass schemes have historically resulted in the bypassed sections being awkwardly renumbered as extensions of other, unrelated roads. There does seem to be an emerging standard whereby downgraded sections of road are given new codes by appending digits to the old ones but this is not consistently applied. It should be.

(*) there is a good historical reason for this, which will never be relevant again

(**) A good general principle for route numbering would be to adopt the notion of a geodesic from geometry, thereby ensuring that a numbered route will travel in a broadly consistent direction along its entire length (which is obviously not always the case at the moment). We could define a “geodesic condition” so that a numbered route should describe a best path across the landscape that cannot be improved by incremental detours along roads of similar quality. For example, the B3 between Gilford and Loughbrickland is not a geodesic, as it does not follow the the signposted “best” route between those towns. A geodesic road may not always be the preferred route between its end points, as the existence of a high quality road does not negate the need for a parallel minor road. A ring road may be a geodesic, if it beats driving through the centre of town.

We assumed the EU’s enemies were all internal. We were wrong.

There seem to be two main narratives of the Ukrainian conflict doing the rounds in the Western media. On one extreme is the thesis that Putin is determined on expanding Russian territory. On the other, his actions are a logical and understandable response to the reckless expansion of NATO. Neither of these is quite true.

Crimea aside, Putin has resisted officially expanding the territory of the Russian Federation. The separatist statelets of Abkhazia, South Ossetia and Transdnistria have each asked to be admitted to the federation only to be politely ignored. The military logic of this is impeccable: the cost of incorporating and defending territory is unnecessary if the same aims can be achieved by sowing chaos. Better to keep one’s neighbours worried enough that they don’t want to provoke you, but not worried enough that they act decisively against you. The ideal outcome from Russia’s point of view is a near-abroad of buffer states that know their place in the pecking order, and for nearly two decades that’s more or less what it had.

We instinctively understand that military alliances are threatening, and it’s tempting therefore to point the finger at NATO. But to blame NATO for upsetting the balance is disingenuous. The Yushchenko government suggested in 2008 that it might join NATO, and was rebuffed after Russia (understandably) objected. The post-Maidan government initially made clear that it had no intention of repeating that mistake. NATO membership was overwhelmingly unpopular in Ukraine before the Crimean crisis, polling consistently under 20%. The threat to Russia came not from NATO, but from the EU.

Unlike NATO, the EU is consistently popular in Ukraine. The catalyst for the Maidan protests was the choice of the government to abandon a long-promised free trade agreement with the EU in favour of a Russian counterproposal. Russia, Belarus and Kazakhstan, three of the five largest post-Soviet economies, had recently agreed to create the Eurasian Economic Union (EEU), a rival economic bloc to the EU with two distinct advantages for Putin: Russian hegemony and fewer restrictions on the exercise of state power.

Ukraine, Moldova, Georgia and Armenia were all close to signing free trade agreements with the EU when the EEU counter-offers started coming in. Armenia was persuaded to switch, but as the second largest economy after Russia itself, Ukraine was the real objective. When the Ukrainian people chose the EU over Russia, it not only stunted the viability of the EEU project, but also spurned what most Russians assumed was still a uniquely special friendship, even despite a decade of increasing political distance between the two countries. (The English and the Scots would surely understand.)

Ukraine is uniquely important to Russia economically. Since Soviet times, the industrial supply chain has straddled the border. Ukraine’s Donbass region relies heavily on Russian customers, and the military is one of the largest. With their relationship now probably damaged beyond repair, Ukraine and Russia have been working furiously to disentangle their economies – the most likely explanation for the mysteriously half-empty aid convoys is that they had enough capacity to asset strip entire factories and transport key machinery back to Russia. If this is true, then it demonstrates Russia has understood that it has lost Ukraine to the EU economically. But this has only increased its determination to salvage what it can and keep NATO out. The differing levels of outside commitment in Ukraine prove that Russia still claims a military sphere of influence, and NATO has little intention of challenging it.

Crimea was taken because it was easy, because it was strategic, because the Ukrainians know (even if they won’t admit) that they’ll never get it back, and because that’s what happens when you cross the wrong people. Unlike the other post-Soviet frozen conflicts, Crimea had to be annexed – Sevastopol is far too precious to be left in the hands of amateurs. The Donbass rebellion was facilitated to keep the Ukrainians tied down fighting a second insurrection while Russia quietly consolidated the first one. Now that a frozen conflict has been successfully engineered in the east, nobody is going to suggest an armed liberation of Crimea in the face of a hostile populace.

This leaves Ukraine in a nasty position – economically and politically committed to the EU, but militarily overshadowed by old-school Russian hard power. It also reminds us that the EU is not expanding into a vacuum. We have always thought about European integration as if it were a contract, where the only countries that matter are the contracting parties, and any agreement between them would naturally have the blessing of the world. The only enemies of European integration in this scenario are internal factions who disagree with their own government’s policy. The worst reaction we ever received from a third country was jealousy.

We now know that this was complacent. There is an external enemy of European integration, and he sits in the Kremlin. Some countries have decided for themselves that they don’t want to be involved in the EU project. Putin has decided that he doesn’t want other people’s countries being involved in the EU project. He has his own alternative, and clear ideas about who should be a member of which one. Talk of a new Iron Curtain is overblown, but there is an increasingly clear demarcation between the EU and Russian spheres of influence.

Dictionary definitions


Darach MacDonald (June 12th) makes a spirited defence of the dictionary definition of “unionism”, and rightly points out that purity of blood is a fiction in the modern world. But culture is not transmitted through the genes, and anyone observing an Orange parade should be left in no doubt of the existence of “unionist” culture.

The constitutional question is far from the “single common identifying policy” that unites Unionist political parties. Support for monarchism, the Orange Order and Scottish cultural heritage, together with disdain for the Irish language, Gaelic sports and (historically) the Catholic church have long been commonly held positions. None of these follow automatically from the dictionary definition, so the dictionary definition must be incomplete.

It is unfortunate that a political term has come to have a non-political meaning. “Protestant” is equally inaccurate, and “Hun” is potentially offensive. But whatever name we decide to use, most people understand it to mean more than just a single policy position. It also identifies a distinct, shared worldview that can be difficult to fully appreciate from the outside, leading to a gulf in understanding that perpetuates conflict.

It has never been just about the Border.


Andrew Gallagher

The ethnic basis of “Unionism” and “Nationalism”

Seems I’m on a roll. I wrote another letter to the Irish Times…


Cian Carlin (June 10th) repeats one of the cardinal errors of Irish politics when he reduces “Unionism” to a mere political preference. The divisions in Northern Ireland span not only politics but also culture, religion, history and ancestry. “Unionist” and “Nationalist” have become shorthand names for Ulster-British and Gaelic-Irish ethnic groups, each with their distinct mythology and cultural norms. Pretending that a word when uttered by someone else means only that which you would prefer it to mean is dishonest and serves only to derail the argument.

To believe that one ceases to become “Unionist” if one votes for a United Ireland is to reduce the entirety of a culture to a single issue. If changing your mind about a particular policy also implies wholesale abandonment of your culture and history, then it is no wonder that Northern Ireland politics is so dysfunctional. For too long we have pretended that a struggle for ethnic supremacy is a mere political disagreement, perhaps because we fear the implications of admitting that our problems are not amenable to quick fix solutions.

More thoughtful politicians and commentators prefer to use “pro-union” for the political viewpoint in order to clearly distinguish it from cultural “Unionism”. It is quite possible to mix and match political and cultural labels – there is a distinct body of “Unionist” opinion that would prefer an independent Northern Ireland state, and many “Nationalists” are content to be part of the UK.

So many fruitless arguments hinge on the misinterpretation of ambiguous terms. Just as “Ireland” can mean either the 32-county island or the 26-county republic, so can “Unionism” and “Nationalism” have multiple, distinct meanings depending on context. Debates devolve into slanging matches where opponents aim their rage past each other, each using the same words but meaning different things by them.

Perhaps it’s time we stopped using the words “Unionism” and “Nationalism” altogether, as they seem to create more confusion than enlightenment.


Andrew Gallagher

In praise of secular education

I wrote a letter to the Irish Times today.


Patrick Davey (May 30th) claims the right to have his children educated in a religious school. This is as unattainable a “right” as the right to live in a religious town, or to work in a religious factory. A “right” that can only be realised if others are denied their own rights is no right at all. The idea that the majority have more “rights” than minorities do is tyranny, and the removal of a right through effective unavailability is as much a violation as if it were explicitly denied in law.

Secular education is compulsory, and religious education cannot be. By conflating the roles of state school and church school we have created inequality between those of the majority faith and those of other faiths or none. The only way to respect everyone’s rights equally is to separate the roles of Church and State, leaving schools to teach a full secular curriculum to all regardless of faith, and allowing each church to supplement this with its own particular teachings outside school hours as parents wish and free from state interference.

Separation of Church and State is not an attack on religion. It releases everyone, religious and irreligious alike, from the shackles of pretence and hypocrisy. People of faith should follow the example of their brethren other countries and embrace a secular state as the means of their own liberation.

Russian special forces caught red handed on camera

Russia continues to deny that it has armed forces in Crimea, but it was only a matter of time before someone slipped up. Not only do the mysterious “self-defence forces” use Russian guns, uniforms and vehicles (complete with Russian military number plates) but at least one of them has forgotten to take all the identifying labels off his army uniform (Russian language original), leading to a social media profile naming his special forces unit.

Of course Putin doesn’t expect us to believe him. His plausible deniability is entirely for domestic consumption. His own polls show disapproval of his current actions running at an unprecedented 73%. Russians have no appetite for a shooting war with Ukraine, with whose citizens many have close personal and family ties. With chinks starting to show in his story, how long can Putin keep domestic opposition at bay?

(With thanks to @captsolo for the tip)