The Argument Nobody Wants to Have

We often encounter a certain type of silence that falls over a room when someone says the word “migration” with any intention behind it. Not the casual version… “ooh, immigration policy, fascinating”… but the real one, the one where you can feel everyone doing a quick mental risk assessment before they speak. Are we doing the compassionate version of this conversation, or the concerned one? Which one keeps me safe?

I’ve noticed most people pick neither. They pick silence, dressed up as nuance. “It’s complicated,” they say, and then change the subject to something safer, like death, or money, or their mother-in-law. And look, it is complicated. But “complicated” has become a polite way of saying “I haven’t actually thought about this properly and I’d rather you didn’t notice.”

So let’s notice. Properly, for once.

What’s actually happening, because the noise drowns the signal

Two things are true at once, and they sit awkwardly together.

First: anti-migrant violence has been climbing in parts of Europe and across southern Africa, and it isn’t a vibe, it’s documented. In July 2025, the Spanish town of Torre-Pacheco saw days of organised violence against North African residents after a pensioner was beaten, with the BBC reporting fourteen arrests. Euronews later found that disinformation, not the original incident itself, had done most of the work whipping the town into a mob. The summer before that, Britain had its own version: riots outside hotels housing asylum seekers, stoked initially by false claims about a stabbing attacker’s identity, with researchers tracking a documented rise in hate crimes against Muslim communities in the months that followed. Germany had its own flashpoint after the Magdeburg car-ramming attack, with migrant advocacy groups reporting a sharp jump in racist incidents in the weeks after. None of this is fringe anymore. Vigilante patrol groups, the kind that stage their own little theatre of border enforcement, are documented as resurging across Europe roughly a decade after the 2015-16 crisis first produced them.

Second, and this is the bit that should needle anyone with a strong opinion either way: actual asylum applications to the EU fell by roughly 27% in 2025 compared to the year before, down to around 669,000 first-time applicants, the lowest the bloc has seen in years. Syria, the name that dominated this conversation for over a decade, has been overtaken by Venezuela as the leading country of origin. The fear is rising. The thing people say they’re afraid of, in raw numerical terms, is shrinking.

That contradiction is the whole article, if I’m honest. Everything else is detail.

Southern Africa is running the same script with different actors

If you think this is a European phenomenon, the news out of South Africa this year should disabuse you of that fairly quickly. Operation Dudula, whose name translates roughly to “push out” in Zulu, has gone from grassroots township movement to genuine political force, with its former leader resigning to take up a candidacy for a mainstream party. A newer group, March and March, has been organising alongside it. Between April and June 2026, waves of xenophobic violence swept Johannesburg, Pretoria, Durban and Mossel Bay, displacing migrants into temporary shelters. The World Health Organization condemned the violence directly after the deaths of Ethiopian and Mozambican nationals; the South African presidency pushed back, framing the deaths as the work of organised crime and trafficking networks rather than evidence of a national mood. Human Rights Watch has documented attacks on foreign-owned shops and individuals attributed to groups affiliated with these movements throughout the spring.

The language matters here too. Reporting on the unrest has noted the use of “kwerekwere”, a slur used against African migrants that implies something less than human, deployed not just on the street but reportedly by figures with real institutional standing. That’s not background noise. That’s the permission structure that makes the street violence feel sanctioned rather than shameful to the people doing it.

And here’s the uncomfortable bit, the one that should complicate anyone’s tidy narrative: South Africa’s own unemployment rate has hovered around a third of the working population for years. Commentators in the South African press, including some explicitly making the case for African solidarity, have argued the root crisis isn’t migration at all, it’s decades of governance failure and economic exclusion, with migrants simply functioning as the visible, blameable symptom. Which is a very different read to “South Africans are uniquely intolerant”, and probably a more honest one.

Let me make the case you don’t want to hear, whichever side you’re on

If I only gave you the version of this essay that flatters one tribe, I’d be doing exactly what I said I was rebelling against at the start. So here’s the discomfort, evenly distributed.

The case for taking restrictionist anxiety seriously, not dismissing it as bigotry by default: there’s actual peer-reviewed research, from sociologists Maureen Eger and Susan Olzak, showing that after anti-immigrant violence in Germany, support for far-right parties rose not just among people who already held anti-immigrant views, but among previously neutral voters too. That’s a genuinely inconvenient finding for anyone who assumes more exposure to the “issue” naturally produces more compassion. Sometimes it produces the opposite, because the violence makes migration newly salient to people who hadn’t been thinking about it at all. If you’re someone who believes hostility is purely a moral failing in the hostile person, that finding should at least make you pause. Communities under genuine economic strain, with services stretched and wages stagnant, are not inventing their anxiety from nothing, even when the people exploiting that anxiety for votes very much are doing something cynical.

The case against treating “concern” as automatically reasonable just because it’s widely felt: widely felt isn’t the same as well-founded. The Torre-Pacheco riots were traceable, directly, to disinformation rather than to any real escalation in crime. The numbers I just gave you show EU asylum applications dropping by over a quarter year on year, at precisely the moment vigilante activity is described as resurging. If the fear were tracking the reality, it should be falling too. It isn’t. That’s worth sitting with if you’re inclined to nod along sympathetically every time someone tells you migration is “out of control.” Out of control compared to what, exactly, and according to whose count?

The case for taking the South African framing of governance failure seriously: it would be too easy, and a bit lazy, to read Operation Dudula purely as racism with no material cause. A third of the country has been unemployed for years. People do not generally turn on their neighbours from a place of comfort and security. Naming the economic root doesn’t excuse the violence, not remotely, but it does mean the activists profiting politically from “push them out” rhetoric are offering a solution that solves nothing structural, which is its own kind of cynicism worth calling out specifically.

The case against letting that economic framing become a full pardon: “poverty made them do it” has a habit of sliding into “so it’s understandable”, and from there into something uncomfortably close to permission. The slur, the targeted shop burnings, the mob that killed Elvis Nyathi in Diepsloot back in 2022, none of that required poverty to exist. Poverty explains a recruiting pool. It doesn’t explain a machete.

The thing almost nobody says out loud

Here’s my actual, unguarded opinion on the meta-question, the one about why people avoid having an opinion at all: I think a lot of the avoidance isn’t really about not knowing enough. It’s about not wanting to be filed into a category. Say one sympathetic sentence about migrants and you’re “naive”, say one sentence about strain on services and you’re “the other thing”, and most people, quite reasonably, don’t fancy either label enough to risk the sentence. So they retreat into “it’s complicated” and let the loudest, least careful voices on each side do all the talking instead. Which is, not coincidentally, exactly how you end up with disinformation-fuelled riots in small Spanish towns and a slur becoming normalised enough to use from a position of institutional power.

The two things this article opened with, rising violence and falling numbers, aren’t actually a contradiction once you stop pretending the fear is purely about migration itself. It rarely is. It’s about wages that haven’t moved, services that feel thinner than they used to, a sense of control that left the room some years ago and hasn’t come back. Migrants make a more visible target than an interest rate decision or a hollowed-out local economy. They always have.

None of which means the violence is forgivable, or that genuine concerns about integration, housing pressure, or border policy should be waved away as automatically illegitimate either. Both of those things can be true simultaneously, and most people, when you actually ask them properly rather than baiting them into a soundbite, hold a more nuanced position than the algorithm-friendly version of either side ever credits them with.

You’re allowed an opinion on this. You’re even allowed an uncertain one. What you’re not really allowed, not if you want to be honest with yourself, is the comfortable non-position of having no view at all whilst the actual numbers, and the actual people caught in the middle of them, keep moving regardless.

Until Next Time