Ask someone for an Australian accent and they'll have a go. Ask for a South African one and the question falls apart — because there isn’t one. There are many, and the reasons go back two centuries.

That was the starting point for LW Lanham, professor emeritus at Rhodes University, whose 1996 survey of English pronunciation in South Africa is still one of the clearest maps of how a single country ended up speaking English in so many different ways at once. Australia, he pointed out, produced one broadly national accent. South Africa never had the conditions for it.

The first genuinely local English accent grew in the mouths of children — specifically the children of the 1820 Settlers in the Eastern Cape, whose parents arrived speaking a jumble of English dialects. What survived was mostly south-eastern English with an unmistakable Cockney streak. It is why, in what Lanham called Cape English, Martin can sound like Morton and garden like Gordon, and why my drifts toward “moy”. Living beside and marrying into Dutch-speaking families added more: a tapped or rolled r, and p, t, and k pronounced without the small puff of breath that English speakers elsewhere give them.

Then came Natal. Settlers arriving between 1848 and 1862 brought Midlands and northern English habits, and one in particular has since gone national: the flattened i that turns five times five into faav taams faav. A related move pulls the tongue back before an l, so cold edges toward cock and bell toward Alf. Lanham was hearing both in television commercials by 1995 — a Sta-soft ad saying faav, Melrose cheese pronounced “Malrose” — evidence the features had reached small children.

Over all of this, from about 1880, sat the accent of the British upper middle class. The diamond and gold rushes filled cities such as Johannesburg and Kimberley with immigrants and a rigid class ladder, and standard southern British pronunciation became the sound of authority and aspiration. It held on remarkably long. News readers for the South African Broadcasting Corporation were still expected to use it through the 1960s and 70s, and a British Broadcasting Corporation correspondent in 1973 described their accents as “impeccably upper middle class”. A professor of English at the University of the Witwatersrand complained two years earlier that the only thing the English-speaking population was thin-skinned about was its unquestioning reverence for received pronunciation.

Afrikaans English left its own marks, and the most durable is easy to miss: a soft, voiced h slipped between two vowels, so that chaos becomes cha-hos and react becomes re-hact. Lanham treated it as a shibboleth: a tiny quirk of pronunciation that betrays where a speaker comes from. The word shibboleth, by the way, is biblical — the Gileadites used it as a password because their enemies could not manage its opening sound – and the principle has not changed. Here the giveaway is often the last trace of a first language in speakers whose English is otherwise indistinguishable from a native one.

African English, used by the largest body of English speakers in the country, began in nineteenth-century mission schools and became the accent we are familiar with today. Vowel length distinctions level out, so heat and hit converge; nurse arrives as ness; every syllable takes equal weight, so father has two identical halves.

Indian English travelled the opposite way. Carried to Natal from 1860 by indentured labourers, it shed its Indian-language features so thoroughly that Lanham identified its defining trait as how few of the strongest local markers it used. And the English of the Coloured community is recognised less by vowels than by melody — a sustained high pitch on the final syllable, in statements as much as questions.

Lanham’s snapshot is thirty years old, and accents do not stand still. But his central point has outlasted the details: this is a country where you can hear its history through how its inhabitants speak.