The wall and the watchman

The case now being made is straightforward and not unreasonable on its face. Islamic State’s Sahel affiliate is cooperating across the Niger frontier with fighters already operating inside Nigeria. That affiliate has, for the first time, claimed attacks inside Sokoto and Kebbi. Nigeria has spent a very large sum on defence over a long period without a security dividend to show for it. Put those three facts together and a physical barrier along the northern border looks, to a great many reasonable people, like the obvious next step.

We think the first two facts are broadly right, the third is right but frequently mis-stated, and the conclusion does not follow from any of them. This is worth setting out carefully, because the underlying problem, the replenishment of these groups faster than they can be attrited, is real, and the country deserves an answer to it that will actually work.

What the case gets right

Cross-border cooperation between the Islamic State’s Sahel and West Africa affiliates is not a claim that needs to be taken on faith. In late May 2026, Islamic State’s Sahel Province claimed its first attacks inside Nigeria’s northwest, against army positions in Sokoto and Kebbi states, publicised through the group’s usual propaganda channel. We would flag that the claim itself, and the casualty figures attached to it, are contested between sources and have not been independently verified — but the direction of travel, an eastward push along the Sahelian frontier towards a part of Nigeria that had not previously featured in this group’s claims, is consistent with a wider pattern regional analysts have been describing for over a year.

The spending point is also broadly correct, though the figure usually attached to it needs a correction. Nigeria has spent in the region of ₦32.9 trillion on defence, not over the past decade as is often stated, but over roughly the past fifteen years — a figure independently reported by multiple Nigerian outlets. Over that period more than a hundred thousand people are estimated to have been killed and 3.7 million remain displaced. The trend in outcomes has not tracked the trend in spending, and 2025’s defence budget, a record ₦6.57 trillion, was followed by continued and in places worsening violence rather than by a turning point.

The spending argument is real. It argues against another large, single, capital-heavy purchase of the kind that has already produced fifteen years of exactly that.

What does not follow

A fence is built to stop a static threat crossing a fixed line. Islamic State’s Sahel and West Africa affiliates are not a static threat. They are networks that recruit, resupply and move along routes that shift in response to whatever obstacle is placed in front of them, across a border that already runs for roughly 1,600 kilometres and through terrain and kinship networks that predate the border by generations.

The clearest evidence available on what a hardened land border actually does to a network-based threat comes from Tunisia’s frontier with Libya. An analysis by West Point’s Combating Terrorism Center found the barrier there to be counterproductive on balance: it burdened small-scale, legitimate cross-border traders far more than it burdened organised smuggling and militant networks, which adapted, and the resulting economic marginalisation is judged to have played into the hands of the jihadist groups the barrier was meant to exclude.

We looked at four barrier projects built at least partly against a security threat — the India–Bangladesh and India–Myanmar fencing, Saudi Arabia’s frontier with Iraq, Tunisia’s with Libya, and Turkey’s with Syria. Not one of them is supported by a rigorous, published assessment showing how much infiltration the barrier actually prevented. Where independent evidence exists at all, as in Tunisia, it points the other way. That absence should weigh as heavily in this argument as any case made in the barrier’s favour.

A detail that matters more than it looks

The officer most associated with the idea of fencing Nigeria’s borders is General Christopher Musa, who raised it in June 2025 while serving as Chief of Defence Staff, asking why Nigeria should not do what Pakistan and Saudi Arabia have done. Fact-checks at the time confirmed he had proposed fencing and control, not the closure some coverage implied.

Musa left the CDS post at the end of October 2025. He was confirmed as Minister of Defence in December 2025, and in that capacity now sits above the National Boundary Commission’s own border-management review, which reported in June 2026. What that review actually recommended is worth reading closely, because it is not what the current public argument describes.

The proposal on the table is not a wall

The National Boundary Commission’s June 2026 workshop set out a technology-led concept: drones carrying thermal, radar and lidar sensing, satellite integration, biometrics, geospatial analysis, and a proposed National Border Data Fusion and Intelligence Centre with a named national border coordinator. No cost estimate or timeline has been published for it. The only cost figures circulating for a Nigerian border wall, in the region of six to twelve billion dollars, are analyst extrapolations from other countries’ fencing programmes, not government numbers, and they should not be treated as one.

This is, in our judgement, the right kind of instrument for a border of this length and character. Detection is genuinely useful, and closing the gap between the roughly 364 approved border control points and the approximately 1,497 illegal or unmanned routes identified at the same workshop is a real and serious goal. The question is not whether to build a sensing and fusion capability. It is whether Nigeria has, until now, built the parts of a border security system that a sensor cannot supply on its own.

Nigeria has already run this experiment

The e-border surveillance programme approved in 2019, at a reported ₦52 billion, is the clearest domestic precedent, and it is not encouraging. Contracted to Huawei in 2022, three years behind schedule, independent reporting by The ICIR found it still incomplete at the last review available to us, with equipment installed at only seventeen of eighty-six legal border posts. The roughly 1,497 informal routes were never within its scope at all. A funded, built and switched-on technology programme still under-delivered, because detection without the response, prosecution and sustainment layers behind it does not reduce a threat. It relocates the point at which the state notices it.

This is not a Nigerian peculiarity. Two independent public audit bodies, the European Court of Auditors reviewing Frontex and the United States Government Accountability Office reviewing the Coast Guard, have separately examined two unrelated border systems on two continents. Neither recommended more sensors. Both pointed to weaknesses in information-sharing, performance measurement and institutional coordination — organisational failures, not technical ones.

What is actually missing

• A funded response chain behind the sensors. Detect, intercept, identify, process, charge, court, disposal. A wall, like a drone feed, buys the first link. The delay, on every published example we could find, sits in the other seven.
• A named institutional owner. Eight Nigerian bodies already hold a piece of border security: the Army, Immigration, Customs, the Police, the NSCDC, the DSS, the National Boundary Commission and the Office of the National Security Adviser. Credible commentary on the 2026 Homeland Security debate is explicit that the problem is coordination between these bodies, not a missing one, and warns against adding a ninth.
• A regional picture that has changed. Niger’s departure from ECOWAS and the Multinational Joint Task Force in early 2025 measurably weakened the regional response that had, between 2015 and 2021, cut area fatalities from over eight thousand to under two thousand a year. Any national design that ignores this is designed against last year’s map.
• A sense of where the threat actually is. The northwest’s exposure is recent, specific and moving. A programme spread evenly across all four of Nigeria’s land borders spends against the map rather than against the evidence.

The honest version of the argument

None of this is a case against the National Boundary Commission’s direction, and it is not a case that Nigeria’s northern frontier is somehow fine as it is. The underlying problem the wall argument is reaching for, that replenishment currently outpaces attrition on this frontier, is real, and the fifteen years of defence spending behind it should trouble anyone reviewing the next big-ticket proposal. The honest reading of that spending record argues against another large, single, capital-heavy purchase, not for one.

A wall is easy to picture, easy to announce and easy to photograph at a ribbon-cutting. A funded response chain, a named institutional owner and a sustainment budget line are none of those things, and that is precisely why they are the parts most often left out of a proposal and most responsible for whether the proposal works. The National Boundary Commission has already proposed the right instrument. What would make it worth building is not a taller fence. It is finishing the part of the plan that decides what happens after the sensor reports a crossing.

Our interest

We advise on exactly this kind of programme, and we would be glad to do that work in Nigeria. An article arguing that the answer is a response chain and an institutional design rather than a fence is an argument for the kind of work we sell, and it should be read with that in mind.

The recommendation we would make first costs nothing to accept: engage with the National Boundary Commission’s own review rather than proposing a rival concept beside it, and settle who owns each new function before any of it is funded. Everything else is sequencing.

Read next · Ground holding

Holding ground

A patrol base is a manning commitment with buildings attached. The buildings are the cheap part.