Once you eliminate the impossible, what’s left? A second referendum

Sherlock Holmes would have understood. In The Sign of Four, he famously told Dr Watson: “When you have eliminated the impossible, whatever remains, however improbable, must be the truth.”

The Brexit saga is not a murder mystery, but Holmes’s logic remains: if MPs reject Theresa May’s withdrawal agreement decisively, all the options for leaving the European Union on 29^thMarch will be impossible—politically, if not theoretically. The only sane option that will remain is for parliament to seek an extension of the March deadline and ask the people whether they still want Brexit after all.

Last night’s debate on Channel 4 underlined this point. One by one, the alternatives crumbled.

- James Cleverley, for the government, could not answer Jacob Rees-Mogg’s argument that the Withdrawal Agreement contained 68 pages of EU rules that would put Northern Ireland on a different footing than the rest of the UK. When Labour’s Barry Gardiner said the prime minister’s deal would be worse for the economy than continued EU membership, Cleverley did not contest the point: he said merely that the economy would continue to grow—as if anaemic under-performance would be something to boast about.

- Gardiner managed to keep an impressively straight face when he said that Labour would be able to negotiate a deal that maintained all the benefits of the single market and customs union without incurring the costs and responsibilities of EU membership. Asked directly whether Labour’s approach would be better for the economy than EU membership, he was honest enough to duck the question, rather than lie and say “yes” or commit political suicide and say “no.”

- Rees-Mogg evaded the practical consequences of a hard Brexit altogether. He talked of sovereignty and trust and the need to honour the result of the 2016 referendum. Challenged on the difficulties of UK exporters and importers trading under World Trade Organisation rules, he struggled to give a plausible account of how this would work. Wisely, he did not even try to challenge the estimates that his form of Brexit would leave many people worse off—not least the millions of voters who, suffering from austerity, voted for Brexit two years ago.

(The programme did not include an advocate of the Norway-plus option. Had it done so, s/he would have faced an enhanced version of one of the criticisms levelled at both Cleverley and Gardiner: that they were offering a future in which the UK would be bound by EU rules without having any say in them.)

As for Caroline Lucas, who made the case for Remain and a “People’s Vote,” the striking thing is that Cleverley, Gardiner and Rees-Mogg challenged her solely on process and democracy—not the case for staying in the EU. Lucas had a response to the democracy argument (albeit one that will not convince everyone), whereas her three opponents each tried to hide from specific challenges to their case. In essence, she said that voters should now have the chance to think again—either to reverse their decision to Leave or, as doctors put it, to give “informed consent” to Brexit and all its consequences.

What now? The answer depends in part on how the drama of Tuesday’s votes and their immediate aftermath play out. If Theresa May resigns, or is forced out, then we shall have to wait for her successor to take over. A hardline Brexiter might prefer no deal, but the attempt would be defeated by MPs—and even to attempt it at all could see the government fall. A referendum, including no deal as a choice on the ballot paper, could give them a way forward that might get through parliament.

Alternatively, Theresa May’s deep sense of duty could lead her to stay on and fight any attempt by Conservative MPs to remove her. But, again, a referendum seems the best way forward. So far, she has ruled this out—just as early last year she ruled out an early general election. However, she knows she is heading for a big defeat in parliament on her Withdrawal Agreement. Her denial that she is pondering Plan B is implausible. Such blinkered stubbornness would be both stupid and irresponsible. May is neither.

From her point of view, two referendum options look attractive. One would be a straight choice: Remain versus her deal. She could rule out no deal as a referendum choice on the grounds that its economic impact would be so catastrophic that it is simply not a sensible choice.

Alternatively—if she decides that her position within her party is so weak that she must include a no deal option—she could propose a two stage referendum. In stage one, voters would be asked to repeat the 2016 choice: Remain versus Leave. If Remain wins, then we stay in the EU. If Leave wins again, a second-stage vote would take place, say two weeks later. This time, we would be asked to choose which kind of Brexit—May’s deal or no deal.

Either way, Labour is likely to fall in line with the principle of a referendum, although it might haggle over the choice offered to voters. Like any self-respecting opposition facing a government in crisis, it wants to bring the government down. If it fails in this, then support for a referendum seems almost certain.

One thing is certain: any kind of referendum would need the EU to agree to extend Article 50, so we remain a member of the EU at least until the referendum is held. The European Court has ruled that Britain can revoke the notice unilaterally but that does not apply to extension. While it is likely that the other 27 countries would approve an extension (and the approval must be unanimous), there are doubts in Brussels that approval would be forthcoming if no deal were a possible outcome. This points to a straight choice: Remain versus May’s (or some other compromise) deal.

Logically, then, the likeliest outcome is a referendum, and the likeliest kind of referendum is a choice of Remain versus compromise.

“Logically…” There’s the rub. Logic is generally a better guide to politics than sceptics might think. But it is not infallible. In Holmes’s terms, “the truth” is that a referendum is the only sane option. It is likely to prevail. But we cannot completely rule out a victory for lunacy.