2

The Polyphony of Colonial Ports: The Social Space of the Vernacular Music Revolution

How do we understand these recording sessions that took place around the globe? Where did these vernacular phonograph musics come from? What if anything have they in common? Are there patterns and regularities amid the accidents and contingencies that brought these musics and musicians to regional, national, and even global renown? Was this a musical revolution, or was it simply a commercial and technical event? Perhaps the very novelty of sound recording distorts an underlying continuity in ordinary musicking. Perhaps these were not new musics, but rather new recordings of old musics that had existed for centuries before the onset of recording. This is certainly how the recordings of the 1920s were heard during their first rediscovery in the midst of the “folk music” revivals of 1950s and 1960s. To the scholars and listeners of the folk revivals, the scratchy 78s were not really artifacts of the 1920s: they were earwitnesses, capturing archaic if not timeless traditional musics just before their disappearance. The commercial 78s differed from the “field recordings” made by noncommercial folk-song collectors of the same era only in their lesser degree of authenticity, their ear to the market.

Though this remains a common perception, it is not true: as the music scholarship of the last two decades shows, virtually all the musics that assume canonic form in the recordings of the late 1920s—son, hula, tango, kroncong, samba, blues, ṭarab, jazz, palm-wine—were a product of “modern times,” of the generations that lived between the 1890s and World War I. Though they came to be heard as “roots” musics in the century to come, they were as much “modern” musics as those of the Parisian and Viennese avant-gardes. The gramophone amplified a musical revolution that was already taking place in urban streets and dance halls around the world.

Because vernacular phonograph musics were not, for the most part, the musics of rural peasants and sharecroppers: they were not the “folk” musics that were coming to be collected and studied by folklorists. Nor were they the musics of the industrial working classes that had emerged in the capitalisms of the North Atlantic over the previous century: they were neither the work songs of weavers and miners, nor the parlor songs and minstrel tunes of London music halls and New York vaudeville houses. Rather, the new vernacular musics of the era of electrical recording emerged on the edges and borders of the empires of global capitalism, in the barrios, bidonvilles, barrack-yards, arrabales, and favelas of an archipelago of colonial ports, linked by steamship routes, railway lines, and telegraph cables, moving commodities and people across and between empires.

The geography of this archipelago was charted on the steamship company maps of the time; it might be figured as three great arcs linked by the Suez Canal, the Panama Canal, the Strait of Malacca, and the great fueling port of Honolulu (see map). One arc spanned the littorals of the Atlantic and the Caribbean, where fusions of African and European musics produced Havana’s son sextetos, New Orleans’s jazz bands, Port of Spain’s calypso tents, and Fort-de-France’s beguine orchestras. To the north, it stretched up the Mississippi River and the Great Lakes to the blues and gospel ports of Memphis, St Louis, and Chicago; to the south it followed Latin America’s Atlantic coast to the samba schools of Rio de Janeiro and the tango orquestas típicas of Buenos Aires and Montevideo. These New World musics echoed across the black Atlantic to West African and South African musical worlds linked by British colonial ties and Kru sailors, as highlife and palm-wine music was taking shape in the ports of Accra and Lagos, and marabi in the ports of Cape Town and Durban and the mining capital of Johannesburg.

A second arc stretched from the coastal towns of Iberia across the Mediterranean and through the Suez Canal to the Arabian Sea and the Indian Ocean. Here the fusion of Iberian, Arabic, and Roma musical traditions—the Andalusian-Arabic as it has been called—took various timbres: from the fado of Lisbon to the flamenco of Cádiz and Seville; from the rebetika of Piraeus, played by refugees from Smyrna (Izmir), to the Turkish, Arabic, Jewish, Greek, and Armenian musics of Constantinople (Istanbul), a key musical crossroads; from the chaabi of the North African ports of Tunis, Algiers, and Oran, to the ṭarab of Cairo, and the Swahili taarab of the East African port of Zanzibar.

A third arc stretched from the Pacific deep water fueling port of Honolulu across the Pacific to the South China Sea, and through the Strait of Malacca to the Bay of Bengal and the Indian Ocean. Here a maritime silk road spread the ‘ukuleles and steel guitars of Honolulu, and linked the huangse yinyue of Shanghai, the kundiman of Manila, the kroncong of Batavia (Jakarta) and Singapore, and the popular theater musics of Bombay (Mumbai), Madras (Chennai), and Calcutta (Kolkata).

Each of these arcs was marked by a social diaspora that became a musical diaspora: the Atlantic arc by an African diaspora, the product of centuries of the Atlantic slave trade, that made “black music” central to the ports of the Americas and Africa; the Mediterranean arc by a Roma diaspora that made “gypsy music” a ubiquitous element in popular musicking; and the Pacific arc by a Polynesian diaspora that circulated “Hawaiian music” from Calcutta to San Francisco.

Why were these musics first heard in these ports? Why was samba born in Rio, hula ku‘i in Honolulu, tango in Buenos Aires and Montevideo, son in Havana, palm-wine in Accra, huangse yinyue in Shanghai, kroncong in Jakarta, jazz in New Orleans? The answer lies in the peculiar social and cultural formation of the colonial port: a volatile mix of millions of new migrants living in waterfront neighborhoods imbricated with the racial and ethnic logics of settler regimes and imperial conquests; a population dense enough to provide the critical mass to support the emerging institutions of commercial musicking, the urban industry of theaters, brothels and dance halls; a physical and cultural distance from the cultural capitals and centers of artistic prestige and power; and finally, a peculiar encounter and alliance between the “ear” musicians among the rural migrants, playing local musics on cheap, mass-produced horns, guitars, and concertinas as well as on hand-crafted drums and fiddles, and the “reading” musicians among the port’s subordinated but educated elite, a “talented tenth” playing waltzes and polkas as well as sacred hymns and calls to prayer.

One might see this as a “peripheral modernity,” the term Beatriz Sarlo coined in her pioneering study of Buenos Aires.¹ It was a musical culture dramatically unlike that of the learned musics of court and concert hall, the agrarian musicking of sharecroppers and tenant farmers, or even the popular parlor songs of piano sheet music. As the boundaries between sacred and secular, rural and urban, participatory and performative, polite and popular, folk and professional, were crossed or obscured, the music cultures of the colonial ports reverberated with sounds out of place, discordant noises. They embodied contradictions: urban country musics, commercial religious musics, professional folk musics, popular musics celebrated by elite intellectuals. The recorded vernacular musics were the product of this polyphony of colonial ports.

An Archipelago of Colonial Ports

This archipelago of colonial ports linked two distinct zones of the emerging world economy: the rapidly expanding settler colonial countries of the Americas, Southern Africa and Australia, and the overseas colonies and informal spheres of influence of the European and US empires—the Indian subcontinent, the Chinese treaty ports like Shanghai and Canton, French North and West Africa, British West Africa, British and German East Africa, the Belgian Congo, Portuguese Africa, the Dutch East Indies, French, British and US Polynesia in the Pacific, and the British, French, and Spanish West Indies in the Caribbean Sea.

The settler zone was for the most part formally independent. Settler colonial nations had industrial cities built around metalworking factories and textile mills, as well as vast forests, plains, and mountain ranges with mines and plantations; they each imagined their peoples divided “racially,” with formal and informal distinctions between northern and western European settlers, newly recruited southern and eastern European laborers, indentured Asian workers, indigenous peoples, and the recently emancipated descendants of enslaved Africans. In many ways, Buenos Aires was a cousin to New York, Johannesburg to Chicago, and Rio de Janeiro to New Orleans.

The colonial zone was based on varieties of direct and indirect rule, backed by warships and, by the 1920s, aerial bombers. These enabled capitalist investments in colonial plantations and mines, supported by an infrastructure of railroads, ports, and telegraph cables. Color lines were firmly drawn between the relatively small European communities and the local populations, as well as between both of these and the growing numbers of indentured and migrant workers recruited from other parts of the colonial world, particularly India and China.²

Despite their differences, the two zones were linked by commodity chains of trade and migration. For even if these were not all colonized ports, they were all colonial ports, part of a maritime network that moved the products and people of empire around the globe. Indeed, it was in this period that geographers first began to see “maritime routes, not as autonomous links between one port and another but as clusters tied together at great ports which remained the most competitive in world shipping.” Moreover, the colonial ports came to share economic, social, and cultural characteristics “because the industrialization of shipping … was common to them all.”³

At the center of this industrialization was the steamship. Its steel hull, steam engine, superstructure of decks, and machinery for handling freight and for steering “defined the image of modernity for contemporaries.” One might amend this to say that the steamship defined two different modernities: for if the emblem of metropolitan modernity was the spectacular ocean liner—like the celebrated and doomed Titanic—that “offered premium service” and “sailed between a given pair of ports on a fixed schedule, usually on the North Atlantic,” the emblem of peripheral modernity was the tramp steamer, which “carried the bulk of traffic between the industrialized nations of the North Atlantic and the rest of the world.”⁴

From Buenos Aires to Shanghai, New Orleans to Zanzibar, Marseilles to Singapore, tramp steamers handled “colonial goods,” the agricultural and mineral resources of the rural hinterlands of empire. Lands that had sustained indigenous communities were seized and put on the market—a worldwide “enclosure” of the commons—for the industrial cultivation of common foods and fibers—wheat, rice, and cotton—as well as a host of tropical plants that became everyday commodities in Europe and North America: rubber, coffee, tea, tobacco, sugar, cocoa, and bananas. Though separated by oceans and cultures, ports like Durban, Honolulu, and Havana shared the culture of sugar, as they became export centers for the plantations of Natal, Hawaii, and Cuba; similarly, the slaughterhouses of Chicago and Buenos Aires were pivotal points in a global trade in mass-produced canned meats.

The tramp steamers moved people as well. As a result, these colonial ports became crossroads in the imperial trafficking in labor, way stations in the massive migrations that historian Frank Thistlethwaite once called “proletarian globe-hopping.”⁵ Millions of ordinary migrants passed through these colonial ports: western and northern Europeans as settler colonists to the temperate grain belts of the world; eastern and southern Europeans recruited to steel mills and slaughterhouses; south Asians and Chinese indentured to tropical plantations, which had earlier been cultivated by slave labor.

Unlike the postmodern container ports of the present that are “huge consumers of space but employ comparatively few workers,”⁶ the colonial ports of modern times were labor-intensive. The docks depended on a circulating pool of male maritime workers—crews of seamen and stokers who manned the tramp steamers, gangs of longshoremen and stevedores who loaded and unloaded goods, as well as artisans and machinists who maintained and repaired the ships and trains. Small factories grew up amid the docks and warehouses, manufacturing consumer goods for the local market: fabrics and garments, beer, cigarettes, and soaps. As centers of commerce, the port cities hosted an intricate division of service trades: domestic servants in the town-houses of the merchant and planter elites, cooks and cleaners, street sweepers and prostitutes, taxi dancers and musicians who kept the city’s entertainment and leisure establishments running.

Thus, as multitudes of migrants passed through these cities, many stayed. Established ports grew by half over the two decades between 1910 and 1930. This included both the largest ones like Shanghai, Calcutta, Bombay, Buenos Aires, and Rio de Janeiro, which became cities of more than a million, and the medium-sized ones: Singapore and Batavia in the Pacific, Piraeus and Alexandria in the Mediterranean, New Orleans and Havana in the Caribbean. Smaller ports including Honolulu, Durban, Algiers, and Veracruz doubled in population (see Table 1).⁷

Table 1: Port Populations

City	1910	1930
Havana, Cuba	275,000	425,000
Port of Spain, Trinidad	no estimate	63,954
Fort de France, Martinique	20,000	27,070
New Orleans, USA	305,000	400,000
Chicago, USA	2,000,000	3,300,000
Veracruz, Mexico	30,000	70,000
Rio de Janeiro, Brazil	790,000	1,130,000
Buenos Aires, Argentina	1,000,000	1,690,881
Accra, Gold Coast (Ghana)	21,000	21,000
Lagos, Nigeria	no estimate	107,763
Durban, South Africa	69,244	127,579
Cape Town, South Africa	97,240	198,006
Piraeus, Greece	70,000	400,000
Smyrna	300,000	350,000
Istanbul, Turkey	750,000	699,602
Lisbon, Portugal	500,000	435,359
Seville, Spain	148,000	207,791
Marseilles, France	540,000	653,000
Alexandria, Egypt	230,000	444,650
Tunis, Tunisia	180,000	185,996
Algiers, Algeria	97,460	203,000
Zanzibar	150,000	187,000
Honolulu, Hawaii	39,306	113,000
Manila, Philippines	220,000	315,000
Shanghai, China	523,000	1,607,000
Batavia ( Jakarta), Indonesia	105,126	384,152
Singapore, Straits Settlement	228,555	425,912
Saigon, Vietnam	200,000	200,000
Bombay, India	860,000	1,176,000
Calcutta, India	847,796	1,300,000
Madras, India	500,000	500,000

The waterfront districts—La Boca in Buenos Aires, La Fosse (the Ditch) in Marseilles, the Trenches and Blood Alley in Shanghai—were dominated by the flophouses and dives that fed and fleeced transients. In addition, tenement neighborhoods with a host of new names—slumyards, ghettos, barrios, favelas, arrabales—grew up near factories and slaughterhouses, often radically segregated from the older city centers. Many of their inhabitants were newly arrived from the rural provinces, dispossessed peasants as well as the children of recently emancipated slaves and serfs. If Chicago was full of “southerners” from the Mississippi River Delta and the hills of Appalachia, divided by race, Rio attracted the nordestinos, “northeasterners,” themselves divided between the coastal Bahians and those of the arid interior, the sertão. By 1931, more than a third of Lagos’s people were Yoruba migrants from the Nigerian hinterlands, but there were also neighborhoods that had been settled by repatriated Yoruba ex-slaves from Cuba and Brazil.⁸ In Honolulu, native Hawaiians displaced from rural taro-patch farms and immigrant Japanese and Filipino workers moving off sugar plantations settled in Kaka‘ako, “an area of salt and duck ponds and mud flats … bustling with factories, workers’ tenements and homes” for those who worked as seamen, fishermen, longshoremen, stevedores, laundry workers, and cannery workers.⁹

In virtually every port, the history of settler or colonial occupation had produced an urban geography segregated into districts by race and ethnicity: native towns and settler towns, as Frantz Fanon was to call them in The Wretched of the Earth. If this divide was starkest in the settler colonies of Northern and Southern Africa, from the Casbah of Algiers to the “townships” of Johannesburg, it nevertheless could be seen throughout the archipelago of colonial ports: in the ports of the Indian Ocean (“In Mombassa,” one historian writes, “the new port city was created side by side to the traditional Swahili centre—a physical arrangement which resembled so many of the European plantings along the Asian littoral”¹⁰); in the treaty ports of China (Shanghai was divided between the International Settlement, the French Concession, and the Chinese municipality); and the post-slavery cities of the Americas, with African-American neighborhoods from New York’s Harlem and Chicago’s Bronzeville to Rio’s Praça Onze, Montevideo’s Barrio Sur,¹¹ and Havana’s Belén, Jesús María, and Cayo Hueso.¹²

The circulation of peoples through the archipelago of ports also created districts dominated by diasporas (the global archipelago of Chinatowns from Havana to Honolulu, the Kru-Town settlements of Kru seamen in the coastal ports of West Africa, the gitano neighborhood of Triana in Seville and the gypsy ghetto of Süle Küle in Istanbul), by refugees (the post-1922 settlements of Anatolian Greeks in New Smyrna and New Ionia on the outskirts of Athens), and by mixed communities (the Creole Downtown of New Orleans, the Alfama quarter of Lisbon where the black slaves of Portugal had settled after emancipation, Batavia’s Eurasian quarter of Kemayoran). In addition, the common laborers circulating around the globe on the ships of the maritime trade were a mix of “lascars” (Indian seamen on British ships), “kanakas” (Pacific Islander sailors), “Krumen” (West African sailors), and “Manilamen” (Filipino sailors). Thus these colonial ports were not only a counterpoint of products—like the Cuban counterpoint of tobacco and sugar that Fernando Ortiz figured in 1940—but a polyphony of peoples, elaborately distinguished by color, caste, language, and religion. “As a port,” Rodney Gallop wrote in his pioneering 1932 essay on fado, Lisbon “has always been particularly receptive to exotic influences and, as is generally admitted, to the admixture of foreign, even non-European, blood”; “the lisboeta has always shown a marked liking for exotic song and dance, and in particular for those of the negroid races with which the Portuguese came into contact in Africa, and whom they transplanted as slaves not only to Brazil, but also to extensive regions in Southern Portugal which had been left empty first by the Moors and later by the American colonists.”¹³

Thus, what Tan Tai Yong writes of Singapore is true across much of the archipelago of colonial ports in this period:

From the late nineteenth century onwards, the port city was not only bustling with commerce but it stood as a centre of Malay culture and literature, of Chinese diasporic intellectual and political ferment, and of Indian debates on cultural and religious reformism. As Singapore became the centre of overlapping migrant worlds, incorporating networks of trade, labour and cultures, it developed as a key economic and intellectual node … The port city became a centre of cultural and nationalist movement … a diasporic public sphere.¹⁴

One might call this cosmopolitanism, but that misses the historical contingency of this cosmopolitanism—it has been persuasively argued that this cosmopolitanism is not inherent to cities or ports, but is the result of the peculiar breathing room that certain ports had as a result of the competition between empires.¹⁵ It also misses the violence that constituted these cities. Popular memory and popular song were scarred by a host of episodes of racial violence, lynchings and massacres that triggered race riots across the first two decades of the century. These included the 1900 “bubonic plague” burning of Honolulu’s Chinatown, the 1900 Robert Charles riot in New Orleans, the 1903 water riots in Port of Spain, the 1906 Dinshawai incident in Egypt when villagers were executed in retaliation for attack on a British soldier hunting pigeons on local farmers’ land, and the 1912 Oriente massacre in Cuba.¹⁶

If this archipelago of colonial ports was created by the industrialization of shipping in the decades between the opening of the Suez Canal in 1869 and that of the Panama Canal in 1914, it was reshaped by the strikes and urban uprisings that erupted after the Great War. The worldwide crisis of 1919 was an extraordinary unsettling of social relations, as czars, kaisers, emperors, and the Sublime Porte collapsed in the wake of the bloody trench war, and revolutionary regimes emerged from the ashes of Czarist Russia, Bismarkian Germany, Hapsburg Austro-Hungary, Qing China, and the Ottoman Empire. Perry Anderson has famously argued that the “imaginative proximity of social revolution” was one of the three decisive coordinates of modernism.¹⁷ However, if we look at the peripheral modernism of tango and ṭarab, kroncong and marabi, the imaginative proximity of revolution must be understood to include not simply the remarkable European uprisings that produced “soviets” and “councils” in the cities of eastern and central Europe, but also the worldwide wave of anti-colonial rebellions that stretched from the May Fourth Movement in China in 1919 and the non-cooperation movement Gandhi launched in the wake of the 1919 Amritsar massacre to the 1919 Wafd rebellion in Egypt, as well as the general strikes across port cities, mining towns, and plantations: the Semana Trágica of January 1919 in Buenos Aires, the 1920 strike of Japanese sugar plantation workers on Hawaii’s Oahu, the 1922 Rand rebellion on South Africa’s Witswatersrand, and the 1925 killing of protesters in Shanghai that provoked the May 30 movement.

Musical Cultures of the Colonial Ports

The millions of migrants who arrived in the colonial ports were a critical mass that enabled the explosion of new plebeian musical cultures in the first two decades of the twentieth century. The “migration of former slaves to uptown New Orleans created … a huge mass of patronage for the new music, an essential element of any artistic flowering,” Thomas Brothers writes of the decade when the music that would come to be called jazz took shape.

By one count there were ten to fifteen dance halls uptown alone; between them they produced a function every night. A step or two below the dance halls were the ubiquitous honky tonks … Twenty-five cents per dance-hall patron—that was enough, when spread through a population of some forty thousand immigrants, to nurture the new style.¹⁸

Throughout this archipelago of colonial ports, young musicians found that they could make a living from music in the proliferating cafés, taverns, shebeens, brothels, cabarets, “black and tans,” dance halls, hotels, and vaudeville theaters catering to waterfront transients as well as well-to-do tourists, to young mill workers as well as students and clerks aspiring to middle-class respectability.

The prism of patrons produced a colorful spectrum of venues. At one end of the spectrum were the fugitive locales of plebeian drinking and dancing, like the shebeens of the South African slumyards where home-brew was sold illegally to the migrant African miners and laborers, and where the music and dances known as marabi and ndunduma developed. “Marabi: that was the environment!” the Jazz Maniacs saxophonist Wilson “King Force” Silgee recalled.

It was either organ but mostly piano. You get there, you pay your ten cents. You get your scale of whatever concoction there is, then you dance. It used to start from Friday night right through Sunday evening. You get tired, you go home, go and sleep, come back again: bob a time, each time you get in. The piano and with the audience making a lot of noise. Trying to make some theme out of what is playing.¹⁹

“Ndunduma concerts were real refuse dump affairs, musically and morally,” the pioneering South African playwright Herbert Dhlomo recalled in 1953.

They were attended by degenerate young elements, the uninitiated newly arrived country bumpkins and the morbidly curious. The people danced to the accompaniment of an organ and a most cacophonic “orchestra” of small tins filled with pebbles. The atmosphere was obscene. For the first time in the history of Bantu entertainments, liquor was introduced. The functions were like nightclubs of the lowest order.

“And yet,” he adds, “what naturally talented players the ragtime and ndunduma concerts had! Vampers (as they were called) who improvised many ‘hot’ original dance and singing numbers at the spur of the moment, and who play and accompany any piece after hearing the melody once, and did so in any key.”²⁰

Similar venues emerged in ports around the world, based on the cash wages of young workers. The tango violinist and bandleader Francisco Canaro recalled the cluster of bands in the Buenos Aires waterfront district of La Boca:

We played on a narrow stage where there was barely room for the three of us and the piano … The Café Royal, like similar establishments, was served by “waitresses” dressed in black, with white aprons, and they were much in demand by the customers. … Opposite the Royal was a café of similar style and importance, where the brothers Vicente and Domingo Greco played. Round the corner, in Suárez Street, about thirty metres away, was the “La Marina” café. … Roberto Firpo was playing at another café opposite “La Marina.”²¹

In Honolulu, sugar plantation workers would dance at Heinie’s Tavern—a well-known waterfront dive that Jack London compared to San Francisco’s Barbary Coast—to the small Hawaiian band of Sonny Cunha and Johnny Noble.²² Palm-wine music took shape in the bars of West Africa’s Gold Coast ports: “the name ‘palm-wine’ itself,” highlife historian John Collins notes, “was derived from the low-class dockside palm-wine bars where foreign and local sailors, stevedores and dockers congregated to drink the fermented juice of the palm-tree.”²³

At the other end of the spectrum were the “hotel orchestras” that played for local elites, well-heeled tourists, and white European communities (like the “Shanghailanders” of Shanghai’s International Settlement and French Concession). In elaborate ballrooms that echoed the opulence of the movie palaces of the 1920s, dance orchestras with celebrity leaders alternated established nineteenth-century dances—waltzes, polkas, schottisches—with fashionable foxtrots, tangos, and rumbas. In Shanghai, the large hotel ballrooms often featured foreign musicians, like the white American bandleader Whitey Smith (who claimed to have taught “China to dance”) and the black American bandleader Jack Carter.²⁴ Similarly in Havana, hotels featured foreign or white musicians.²⁵ In Buenos Aires, tango’s New Guard included hotel orchestra leaders like Julio de Caro, the “porteño Paul Whiteman,”²⁶ who experimented with a “symphonic tango” just as Whiteman had manufactured a “symphonic jazz.” In Honolulu, Johnny Noble moved from playing for sugar plantation workers to playing for tourists when he took over the Moana Hotel Orchestra in 1919.²⁷

Alongside the working-class cafés and the elite hotels were a variety of dance halls and dance academies, where local vernacular musicians played as patrons purchased tickets to dance with the young women instructors dubbed “taxi dancers” in the United States and wunü— dance hostesses—in Shanghai. Havana’s academias de baile (ballroom dancing schools) mixed prostitution with taxi dancing, accompanied by Afrocuban ensembles playing son: the Septeto Nacional came to prominence performing at Habana Sport, the largest and best-known academia de baile, which employed more than a hundred women.²⁸ In Singapore,

amusement parks became famous for their dance halls, which included stages for cabaret and joget (a popular Malay dance) that catered to members of the working class, invariably men. For a dance, the single man gave a ticket (costing 10 cents) to the dance hostess he fancied. The latest dances, such as the foxtrot, waltz, rumba, and slowfox, were performed in these dance halls to live music.²⁹

In Rio, they were known as dancings, adopting the newly coined French Anglicism that distinguished them from cabarets.³⁰

These venues for social dancing and commercial entertainment were distinguished not only by the laws and regulations that licensed and classified them—Havana’s cabarets were sorted into first, second, or third class³¹—but also by intricate if informal markers of class and color, vulgarity and refinement, even modes of dress and forms of drink. In his 1933 account of Lisbon’s “popular cafes such as the ‘Luso’ and the ‘Victoria’ where [fado] is regularly performed by semi-professional fadistas,” Rodney Gallop noted that

the social standing of these places seems to be largely a matter of headgear. Entrance to the first is forbidden to those wearing caps or bérets (a fine distinction). The patrons of the second, on the other hand, many of whom are seafaring folk, seem to wear no other head-covering. The spacious rectangle of the “Luso” and the low-vaulted room of the “Victoria” are alike crowded with tables and chairs at which many men (but few women) sit drinking coffee, beer, or soft drinks with exotic names such as Maracuja or Guarana.³²

In Lagos, the ethnomusicologist Afolabi Alaja-Browne was told, the working-class “palmwine ‘depots’” were distinguished from the “exclusive gathering of the dignitaries of Lagos—men who could afford to drink ‘schnapps’ and ‘cased beer,’ not ‘palmwine.’”³³ In Cuba, the world of respectable Afro-Cuban social clubs—the sociedades de color— forbade the working-class son and drums.³⁴ As Langston Hughes wrote of his visit to Havana’s leading “club of color,” the Club Atenas, “Then no rumbas were danced within the walls of the Atenas, for in Cuba in 1930 the rumba was not a respectable dance among persons of good breeding.”³⁵

Racial lines were drawn in both colonial and settler colonial cities. In Shanghai, there was a clear divide between the established hotel ballrooms of the foreign settlements that hosted the national balls of European and American Shanghailanders, and the cabarets and dance halls with Chinese hostesses and a Chinese clientele that emerged amid the “dance madness” of 1927.³⁶ In Havana, hotels and first-class cabarets refused black patrons and rarely hired Afro-Cuban performers.³⁷ In Chicago, where the large commercial dance halls were closed to African-Americans, the cabaret blues and hot jazz played by migrants from the south took shape in the “black and tans,” Bronzeville nightclubs that were open to both black and white patrons, and featured black entertainers; in the decade after 1919, the Dreamland Cafe, the Royal Gardens, and the Sunset Cafe featured Alberta Hunter, King Oliver, and Louis Armstrong.³⁸

If the dance halls and cabarets were the privileged venues for the early instrumentalists of son, jazz, and tango, the popular vaudeville theaters of the colonial ports held a similar place for singers, who developed their craft and their following through stage performances which combined melodramatic theater, comic sketches, puppet shows, and lyric song, all accompanied by small orchestras that combined European and indigenous instruments. In many places, theater singers were the first recording stars: Miss Riboet had been a diva in the Orion company in Batavia’s stambul theater before recording kroncong songs;³⁹ the cai luong actress Dào Nha was one of the early singers recorded in Hanoi; the actress Habiba Messika, “a major sex symbol, the Tunisian epitome of the ‘roaring twenties,’” was recorded by Pathé in their pioneering 1926 Tunis session; and Adelina Fernandes was a star of Lisbon’s revistas before recording fados.⁴⁰ In India, the gramophone boom of the 1920s was built on the recordings of singers from the popular vernacular theater; indeed, even before film music came to dominate the popular music industry in India, recording companies were issuing elaborate multi-disc sets of songs from popular stage productions.⁴¹ Similarly, across the Americas, many of the early recording stars came out of vaudeville, and particularly blackface, theater. In Cuba, Rita Montaner, who made her theatrical debut in 1928 singing “Ay, Mamá Inés” in blackface, became famous for her subsequent recording of that song, which made son respectable by combining it with danzón.⁴²

The musical world of the colonial ports is vividly depicted in Claude McKay’s 1929 novel, Banjo, a picaresque tale of young black mariner-musicians living in the Mediterranean port of Marseilles. McKay, a Jamaican poet who had spent more than a decade traveling between New York, London, Moscow, and Paris, lived in Marseilles in 1927 and 1928, haunting its waterfront district, the Vieux Port.

There any day he might meet with picturesque proletarians from far waters whose names were warm with romance: the Caribbean, the Gulf of Guinea, the Persian Gulf, the Bay of Bengal, the China Seas, the Indian Archipelago. And, oh, the earthy mingled smells of the docks! Grain from Canada, rice from India, rubber from the Congo, tea from China, brown sugar from Cuba, bananas from Guinea, lumber from the Soudan, coffee from Brazil, skins from the Argentine, palm-oil from Nigeria, pimento from Jamaica, wool from Australia, oranges from Spain, and oranges from Jerusalem.⁴³

McKay’s portrait of the Marseilles waterfront captured the polyphony of the colonial ports. In every bar, restaurant, and café, “music was supplied by a tin-panny pianola and half of the night was jazzed away to its noise.” The music of the cafés spilled out into the streets, “the loud voice of the pianola kicking out a popular trot rushed across the square,” attracting audiences—“before the Monkey Bar, a crowd was collected in admiration of a new jangling jazz”—and clashing: “Automatically the piano-panning jumped madly out of the Anglo-American Bar to clash rioting in the square with that of the Monkey Bar.”⁴⁴

McKay carefully counterpoints the sounds, instruments, and patrons of the different venues whose names map the West Indies, North Africa, and West Africa. At the Antilles Restaurant, “whose customers were colored seamen” from the West Indies and Madagascar, “a brown jolly-faced soldier played an accordion while a Martinique guide and sweetman … was shaking a steel pipe, about the size of a rolling-pin, containing something like beans or sand grains … [which] went beautifully with the accordion.” “They played the ‘beguin,’ which was just a Martinique variant of the ‘jelly-roll’ or the Jamaican ‘burru’ or the Senegalese ‘bombé.’” At the Cairo Café, patronized by North Africans, a “slightly built Algerian rattled the drum and banged the cymbals,” accompanying a woman pianist playing “a tragic imitation” of a popular Spanish cuplé singer, Raquel Meller. And, at the African Bar, “the player-piano was spitting out a ‘Charleston’ recently arrived in Marseilles, while Martinique, Madagascan, and Senegalese soldiers, dockers, maquereaux—and, breaking the thick dark mass in spots, a white soldier or docker—were jazzing with one another and with the girls of the Ditch.”⁴⁵

In the midst of this musical cacophony, McKay narrates the daily life of four “colored seamen,” temporarily on shore, living by panhandling, working the docks, and playing music: Malty, from the British West Indies by way of New Orleans, the “best drummer on the beach,” who could also “play the guitar right splendid, but he had no instrument”; the race man Goosey, a New Jersey high-school graduate from a “Cotton Belt” migrant family, who plays the flute, arguing that the banjo is “the instrument of slavery”; the guitarist Taloufa, a Garveyite from the “Nigerian bush” by way of Lagos, Cardiff, and New York, where he had been arrested and deported after jumping ship; and Banjo himself, who had worked “at all the easily-picked-up jobs—longshoreman, porter, factory worker, farm hand, seaman … sailing down through the Panama Canal to New Zealand and Australia, cruising cargo around the island continent and up along the coast of Africa.” “A child of the Cotton Belt” and “vagabond of the lowly life,” Banjo now “patrolled the magnificent length of the great breakwater of Marseilles, a banjo in hand.”⁴⁶

Banjo “dreamed constantly of forming an orchestra,” and the novel—whose subtitle is “A Story without a Plot”—is loosely held together by the fleeting bands Banjo forms. The novel opens with Banjo playing “Yes Sir, That’s My Baby”* (a 1925 Tin Pan Alley hit recorded by vaudeville blues diva Sara Martin for OKeh’s race record label) alone at a bistro on the quay, but he thinks that the banjo “wasn’t adequate for the occasion. It would need an orchestry to fix them right … I wouldn’t mind starting one up in this burg.” His first chance comes when he and Malty meet “a crew of four music-making colored boys, with banjo, ukulele, mandolin, guitar, and horn” from a cargo boat; their impromptu band plays “Shake That Thing”* (a popular race record of 1925 by Chicago banjo songster Papa Charlie Jackson) at the Café African where the “Senegalese boys crowd the floor, dancing with one another,” then at a “love shop” where “all the shop was out on the floor … strutting, jigging, shimmying, shuffling, humping, standing-swaying, dogging, doing, shaking that thing,” eventually ending up at a “drinking hole-in-the-wall … chock-full of a mixed crowd of girls, seamen, and dockers.”⁴⁷

This band dissolves into the night: as Banjo says, “the fellows with instruments never stay long in port.”⁴⁸ However, when the patrone of a café gives Banjo and his friends “a free option on the comfortable space at the rear for the use of their orchestra,” a new ensemble of guitar, banjo, tin horn, flute, singers and dancers takes shape, playing “a rollicking West African song, whose music was altogether more insinuating than that of ‘Shake That Thing’: ‘Stay, Carolina, Stay’”* (a song later recorded in Sierra Leone by Freetown’s Leading Sextet). “It was, perhaps,” the narrator writes, “the nearest that Banjo, quite unconscious of it, ever came to an aesthetic realization of his orchestra. If it had been possible to transfer him and his playing pals and dancing boys just as they were to some Metropolitan stage, he might have made a bigger thing than any of his dreams.”⁴⁹

“The bigger thing” remains elusive; “the group began to break up, every man to his own dream!”⁵⁰ Banjo even loses his instrument—his livelihood—and has to get a job on the docks, working in coal. By the novel’s end, even though Banjo has managed to buy a “second-hand instrument” and the band briefly reunites at the café to reprise “Shake That Thing,” the dream of an orchestra is as distant as ever.

None of McKay’s musicians made it to the “Metropolitan stage,” and none of them are brought into a recording studio (it was not until three years later, in 1931, that Odeon and Pathé went to Marseilles to record musicians from Madagascar and East Africa⁵¹). However, McKay’s portrait of music-making in Marseilles captures the world of cafés and dance halls in the colonial ports that produced the recording artists of the vernacular music boom of the late 1920s.

For most of these recording artists, we know little of their lives; some are only a name on a record label and a voice preserved in shellac. The lives of a handful of the most celebrated have been tracked, though their stories are often a fusion of history and legend: figures like Cairo’s Umm Kulthūm, Buenos Aires’s Carlos Gardel, New Orleans’s Louis Armstrong, Zanzibar’s Siti binti Saad, Havana’s Ignacio Piñeiro, Rio’s Ismael Silva, Port of Spain’s Atilla the Hun, and Athens’s Salonikiós.

They were a generation born near the turn of the twentieth century (see Table 2). Some grew up in the barrios of the ports: steel guitarist Sol Ho‘opi‘i in Honolulu’s Kaka‘ako, the son of a rat catcher and neighborhood minister;⁵² jazz trumpeter Louis Armstrong in New Orleans’s Storyville, the son of a domestic servant; tango singer Rosita Quiroga in La Boca of Buenos Aires, the daughter of a seamstress;⁵³ palm-wine guitarist Kwame Asare in Cape Coast, the son of a storekeeper who sold carpenter’s tools;⁵⁴ and Cuban bassist and songwriter Ignacio Piñeiro in Pueblo Nuevo on the outskirts of Havana.⁵⁵ Others were attracted from the rural hinterlands: ṭarab singer Umm Kulthūm moved from a poor rural village in the Nile Delta to Cairo; taarab singer Siti binti Saad, the daughter of slaves, was a migrant to Zanzibar.

Table 2: Birthdates of Musicians (and Associated City, Region, or Country)

1880: Siti binti Saad (Zanzibar), Ramón Montoya (Madrid), Rabbit Brown (New Orleans)

1883: Salonikiós (Anatolia)

1885: Sadettin Kaynak (Istanbul), Alexandre Stellio (Martinique)

1888: Ignacio Piñeiro (Havana)

1889: Cheikh Hamada (Mostaganem)

1890: La Niña de los Peines (Seville), Cirilo Marmolejo (Mexico City), Carlos Gardel (Buenos Aires)

1892: Atilla the Hun (Trinidad), Dalgás (Istanbul), Johnny Noble (Honolulu)

1894: Bessie Smith (Chattanooga)

1895: María Teresa Vera (Havana), Wilmoth Houdini (Trinidad), Reuben Tholakele Caluza (South Africa), Antonio Menano (Coìmbra)

1896: Adelina Fernandes (Lisbon), Griffiths Motsieloa (Johannesburg), M. K. Moke (Honolulu), Rosita Quiroga (Buenos Aires)

1897: Hafiz Burhan Bey (Istanbul), Jimmie Rodgers (Mississippi), Cheikh El-Afrit (Tunis), Blind Willie Johnson (Texas)

1898: Mike Hanapi (Honolulu), Francisco Alves (Rio), Sara Doughtery Carter (Virginia), Heitor dos Prazeres (Rio)

1900: Miss Riboet (Jakarta)

1901: Udi Hrant (Istanbul), Louis Armstrong (New Orleans), Paulo da Portela (Rio)

1902: Alcebiades Barcelos “Bide” (Rio), Sol Ho‘opi‘i (Honolulu)

1903: Kwame Asare (Kumasi, Ghana)

1904: Umm Kulthūm (Cairo), Niño Ricardo (Seville)

1905: Ismael Silva (Rio), Guty Cárdenas (Yucatán), Hirobai Barodekar (Bombay), Márkos Vamvakáris (Piraeus)

1907: Sofka Nikolić (Belgrade)

1908: K. B. Sundarambal (Tamil Nadu), Cartola (Rio)

1909: Oscar Alemán (Argentina), Li Minghui (Shanghai)

1910: Django Reinhardt (Roma encampments near Paris)

1918: Zhou Xuan (Shanghai)

Many had worked on the docks. Piñeiro had been a stevedore on Havana’s docks; the rebetika bouzouki pioneer Márkos Vamvakáris loaded coal on the docks of Piraeus, and the Hawaiian steel guitarist Freddie Tavares was hired to the orchestra at Honolulu’s Royal Hawaiian Hotel “from a job of sacking onions, down at the pier.”⁵⁶ One researcher tracking the West African musicians who were recorded by Zonophone in the late 1920s notes that “the only address entered on the recording contracts is that of Daniel Acquaah (from [Ben] Simmons’ ensemble), in Liverpool—Caryl Street, Toxteth, hard by the docks.”⁵⁷ Others, like the calypsonians Wilmoth Houdini and Atilla the Hun, had been merchant seamen; Atilla had also worked as a clerk in a coal company on the Port of Spain wharfs.⁵⁸

And many worked in the mills, factories, and small workshops of the ports: Vamvakáris worked in a slaughterhouse, Piñeiro had been a cooper, smelter, cigar maker, and mason, and Gardel recalled his time “as a cardboard maker, in a jeweler’s, and as an apprentice linotypist.”⁵⁹ In his account of early country music, Patrick Huber notes that “Piedmont textile workers made up the single largest occupational group to sing and play in front of radio and recording studio microphones before World War II.”⁶⁰ In Africa, Chris Waterman concluded, popular styles were often “pioneered by members of an intermediate urban wage force that includes laborers, artisans, drivers, sailors, railway workers, clerks, and teachers … cosmopolitan individuals … adept at interpreting multiple languages, cultural codes, and value systems.”⁶¹

These young musicians were shaped not only by the dance musics of cafés and dance halls, the proliferating venues, refined and risqué, of the commercial entertainment districts, but by forms of subaltern neighborhood musicking, sacred and secular. Many first learned their craft in storefront churches, mosques, and religious ceremonies, leading to the dialectic between the sacred and the secular that one finds throughout these musics. Umm Kulthūm began singing in a family musical group—led by her father, an imam in a village mosque—that performed pious tawshih and qasida about the life of Muhammad at weddings and celebrations; she developed her phrasing during three years studying Qur’anic recitation.⁶² Similarly, Siti began by learning to recite the Qur’an. In Istanbul, Sadettin Kaynak not only studied theology, but also worked in mosques throughout his career, and made the first recording of the Turkish call to prayer.⁶³

Several early African recording artists came out of mission school choirs and recorded hymns: this was true of the pioneering South African recording artist Reuben Caluza. The first West African Zonophone recording, of Christian hymns in Yoruba, was made in 1922 in London by the Rev. J. J. Ransome-Kuti (the grandfather of Fela, the Nigerian singer of the 1970s); they were followed by Roland Nathaniels’s 1925 recordings of Christian hymns, which he sang in Ewe and Latin to his own organ accompaniment.

In the Gold Coast, the Fante language hymns of the “singing band movement” associated with the pioneering Ghanaian composer Ephraim Amu not only “offered a new platform for Christians both to perform music outside the strict confines of the church and to Africanize church music,” but “extensively influenced the development of popular music in Ghana” and the recording business on the Gold Coast: many early Odeon and Parlophone recordings from the Gold Coast were of these Christian “singing bands.”⁶⁴

In Hawaii, North American missionaries had been translating hymns into Hawaiian and organizing church choirs since they arrived in the 1830s. As a result, songs based on hymn forms—himeni—had become a major part of Hawaiian music by the early part of the twentieth century. Not only were the late nineteenth-century Sankey-style gospel hymns composed by the Reverend Lorenzo Lyons—collected in Ka Buke Himeni Poepoe of 1872—common throughout Hawaii, but “by the 1870s, elite Hawaiians in Honolulu had begun to compose secular Hawaiian songs whose melodies were modeled on the gospel hymns then in vogue.”⁶⁵ As a result, many of the early recording artists were well versed in hymn forms, and had often been part of church choirs. An example is Helen Desha Beamer, member of the well-known Desha Beamer ‘ohana, or kin group, of hula dancers, musicians and composers, who was recorded by Columbia in the landmark 1928 Honolulu sessions; she was also an organist for the Haili Church Choir of Hilo, which regularly won choir contests through the 1910s and 1920s.⁶⁶

Many of the musicians were also shaped by the plebeian parading associations that made music in the streets of the colonial ports: brass bands, dance troupes, comparsas, calypso tents, and samba schools. Many were informal associations like those of Banjo and his friends in Marseilles. In the ports of the Gold Coast, osibisaaba was a street dance that combined local percussion with guitars and accordions: “osibisaaba outings began in the town center, where revelers organized together, formed a small mobile caravan, and paraded through town in order to attract other participants. Eventually, the gathered group settled in an open space on the town periphery where musicians played with unabated energy and dancers organized into a ring.”⁶⁷ Similarly, in Batavia, street or “walking” kroncong groups—kroncong jalan—“wandered the kampung (residential neighborhoods) at night, singing verses to the accompaniment of a kroncong lute and whatever other instruments they had available—violin, flute, guitar, and tambourine are usually mentioned.”⁶⁸

Marching brass bands emerged throughout the colonial ports. In South Africa, every mission station had a brass band, and the Tswana and Pedi had developed autonomous brass bands in the late nineteenth century; by the 1920s and 1930s, local brass bands marched through the townships on Sundays, in support of the stokvels and shebeens.⁶⁹ “In the 1920s,” one historian notes of the West African ports, “many prominent trading firms … set up brass bands and used them to signify their wealth and advertise their businesses.”⁷⁰ In Lagos, the Calabar Brass Band “utilized Western instruments to perform a music essentially modeled on European military band tradition, and spiced with African and Afro-American features.”⁷¹

There were also large choral groups, like the Cuban coro de guaguancó, Los Roncos, which was led by the young Ignacio Piñeiro,⁷² and dance troupes, like the ronggeng of Malaya, described by the editor of the Straits Echo in 1932:

Every public ceremony … concludes with a ronggeng. People sit round a brightly illuminated open space under the shade of trees in a large garden, close to the main road. A native orchestra, consisting of two violinists, a drummer, a tambourine player, and gong beater take their work seriously, while two Malay girls, well dressed and made up, sing and dance for hours and hours. Occasionally they dance towards a Malay or Chinese and challenge him to keep step with them. The dancing is painfully monotonous to watch, but the Orientals greatly enjoy the ronggengs.⁷³

In West Africa, the sákàrá and aṣíkò dances of Lagos (the former associated largely with Muslims, the latter with Christians) were described by the Yoruba writer Isaac Delano in 1937:

The dancing of “Sakara” is very easy and graceful, and is usually performed by groups of four men on one side and four women on the other. They dance together, meeting in the center of the ring, and sometimes returning, sometimes crossing. The expert dancers enjoy the variations of music introduced by the drummer, who watches the dancer, and is in turn watched by him … The drum is the major instrument, the diameter of which is not more than nine inches. The opposite side is open, and a variety of tunes is obtained by pressing the face of the leather with the left-hand fingers while beating the drum with the right … In the band also are three calabashes and three beaters, and the string instrument … [which] resembles a violin in appearance but is only about a quarter the size.

In the aṣíkò dance, “the dancing is done by pairs, two ladies and two gentlemen facing each other … No stringed instruments are employed, only drums and a carpenter’s saw, used occasionally to make a kind of noise on its sharp edge, as an embellishment. Sometimes a bottle is also used, a nail beating time on it, for the same purpose.”⁷⁴ These percussion-based dances of Lagos were a “local variant of a type of syncretic street drumming found in port towns throughout Anglophone West Africa.”⁷⁵ In South Africa, a variety of competitive dance troupes were organized by migrant workers: the IsiBhaca gumboot dance done in Wellington boots by dockers, the Ukukomika dance parodying military drills, and the ingoma ritual stick fighting.⁷⁶

Perhaps the most celebrated of these musical street associations were the barrio-based carnival bands throughout the Americas. In Cuba, carnival activities had been prohibited during the war of independence; when they were reinstated in 1902, Afro-Cuban carnival street bands organized by barrio, known as comparsas, took part using “stave drums of various sizes and shapes, bells, frying pans, tire rims, trumpets, and other brass instruments, and the corneta china”: “The inclusion of comparsas de cabildo in pre-Lenten carnival seems to have been considered one of their newly won ‘rights’ as citizens of Cuba libre.”⁷⁷ In Trinidad, calypso emerged as a carnival music, as “syndicates” set up tents made of bamboo and palm fronds for their bands and singers—chantwells—to practice in the weeks before the carnival that preceded the Christian season of Lent. In 1919, a reporter for the Trinidad Guardian covered two of these syndicates: a Woodford Street tent with a string band that “consisted of four pieces—a violin, flute, cuatro, and guitar, their blending soft and low being very good indeed”—and a George Street tent with a “bamboo band” where “the musical paraphernalia were confined to the popular ‘instruments’ of the proletariat, which consisted of lengths of hollow reeds of bamboo, a small grater operated on by a musician with a stick, a ‘chac-chac,’ and the inevitable empty gin flask with a tin spoon as the beater.”⁷⁸

In Brazil, the celebrated samba “schools” of Rio de Janeiro—the favela-based music and dance associations that paraded during carnival—were first formed in the late 1920s in the neighborhoods of Afro-Brazilian migrants from the coastal cities of Recife and Salvador in the northeastern states of Bahia and Pernambuco. In 1928, the young Afro-Brazilian sambistas of the neighborhood of Estácio—Ismael Silva, Nilton Bastos, and Alcebíades Barcelos (Bide)—formed Deixa Falar (Let Them Talk), which, historian Carlos Sandroni writes, “is believed to have been the first to parade in the carnival to the sound of a percussion orchestra made up of surdos (bass drums), tamborins (sharp drums), and cuícas (friction drums), to which tambourines and rattles were added,” an ensemble that came be called a “bateria.”⁷⁹ This Estácio samba became “Carioca samba par excellence” and led to competing samba schools: a year later, in 1929, the second samba school—Estação Primeira de Mangueira—was formed by Cartola (Angenor de Oliveria) and Carlos Cachaça, and Mangueira’s great rival, Portela, was founded in 1935 by Paulo da Portela and Heitor dos Prazeres.⁸⁰ Samba school competitions took place in Praça Onze, which Heitor called a “Little Africa,” which bordered on the Cidade Nova, a quarter that had been built around 1870 to house emancipated slaves and immigrants from inland.⁸¹

The competitions between these parading associations and dance troupes often embodied and expressed the social antagonisms within the colonial ports: rivalries and animosities between musical associations were tied to the intricate social differentiations based on color and class, religion and occupation, ethnicity and language that structured the plebeian neighborhoods of ports and mill towns. “Ingoma dancing,” the musicologist Veit Erlmann writes, “was the cultural correlative of the ‘political economy of tribal animosity’ among migrant workers.”⁸² “There had always been a most pronounced system of class differentiation in the choice of chantwells for carnival bands,” Atilla the Hun writes in his autobiographical account of the kaiso tents. In an analysis that echoes C. L. R. James’s famous account of Trinidad’s cricket clubs, Atilla delineates the spectrum of chantwells and bands: “Norman Le Blanc, a store walker, who earned fame as the first to sing a complete kaiso in English, was chantwell to Shamrock Syndicate, a band of whites and persons described as near-whites. The Duke of Marlborough, George Adilla, who was a senior shop assistant, was chantwell to Crescent, a band composed from the coloured middle class and persons described as near-whites … Kaisonians like Red Box, Lord Baden-Powell, and Conquerer could not in their wildest dreams aspire to become chantwells of such bands.”⁸³

The parading associations were sometimes mobilized by competing political parties: in Cuba, white political candidates hired Afro-Cuban comparsas to perform campaign songs like “La Chambelona”: “In addition to soliciting and promoting their own political songs, each major party frequently paid musicians to parade through the streets playing parodies of the rival party’s songs.”⁸⁴ Despite these occasional involvements with urban politics, the parading associations usually found themselves engaged in an unending struggle with the local and colonial state for the right to the streets. In Trinidad, the canboulay riots during the 1881 carnival led to two decades of struggle, as the government tried to regulate and ban drum dances, “noisy instruments,” and “indecent ballads” during carnival;⁸⁵ in Cuba, a 1900 Havana ordinance prohibited the use of all “drums of African origin, in all types of public meetings,” though it was “enforced only sporadically.”⁸⁶ Nonetheless, the participation of Afro-Cuban comparsas in carnival remained controversial, and “by 1916, the suppression of the comparsa ensembles in Havana was nearly total,” a ban that Machado extended throughout Cuba in 1925.⁸⁷ In South Africa, ingoma dances were banned after the 1929 beer riots.⁸⁸

However, in the midst of these struggles, the parading associations were also becoming part of the world of commercial entertainment in the 1920s. In his autobiography, Atilla the Hun recalled the “first time that visitors were charged a fee in a Syndicate’s tent” (probably 1926): “From the moment that charges were imposed by King Fanto to defray expenses of his tent, the public has had to pay to attend kaiso performances. A small fee of one penny was first charged for admission to Fanto’s tent. Other Syndicates at once emulated the vogue started by Red Dragon [Fanto’s Syndicate] and before the season was over, the price of admission had climbed to four cents.” This led to

a complete reversal of the kaisonian’s status in relation to the carnival band. Now instead of being under the dominance of the band, he emerged as the dominant partner in the chantwell-carnival band alliance. Whereas before the leaders of the band gave a miserable stipend to the kaisonian, for example a few charitable shillings and a carnival band costume for his services on carnival day, the shoe now found itself on the other foot.

“By the end of the thirties,” this had led to the separation of the chant-well and the carnival band.⁸⁹

Counterpoint of Musicians

Thus, across this archipelago of ports, music-making had become a way of making a living, part of the urban division of labor. However, for the most part, music-making in the colonial ports was not yet the routinized job it was becoming in the musical factories of London, Paris, New York, and Berlin, where instrumentalists held regular chairs in dance bands and theater pit orchestras (and, by the late 1920s, in radio orchestras and recording-studio house bands) and were beginning to form powerful unions to set standards for their profession and fight the competition of recorded, “canned,” music.⁹⁰ In contrast, in the colonial ports music-making was either a casual, itinerant street occupation, a way of supplementing day labor and occupying periods of unemployment, or a traditional skilled trade, a craft whose knowledges and secrets were handed down through families, ethnic networks, and informal guilds.

In many ways, the musical revolution that produced the vernacular phonograph musics of the late 1920s was a product of the counterpoint between these two classes of musicians: on the one hand, an array of ear-trained street musicians, the children of migrants from agricultural plantations, tenant farms, and mining towns, adapting folk tunes and playing styles to urban streets and dance halls; and, on the other hand, a caste of formally trained, often ethnically defined, artisan musicians—“reading musicians”—who occupied a niche in early twentieth-century ports not unlike other artisans in fine and luxury crafts, serving the needs of the city’s merchant elite.

The accounts of rural migrants bringing new musics to the city run throughout this period. Legend has it that son arrived in Havana from the rural eastern provinces of Cuba—the Oriente—together with the army. In the ports and mining towns of the Gold Coast, migrants “set up their own bands for the enjoyment of their own forms of music. Such immigrant bands are to be found in Accra, Kumasi, in mining and commercial areas, and in many Zongo quarters.”⁹¹ Umm Kulthūm recalled that her own success lay in part in her rural origins: “Compared to the Cairo singers, I was more well known to the rural audience. When someone from the country came to Cairo it was natural that he would buy the recording of a singer whom he had heard and seen before.”⁹²

The artisan musicians, like other artisan trades, often came from families that had long practiced the craft: the Hawaiian steel guitarist Mike Hanapi came from a family of musicians; Kwame Asare’s father played the concertina⁹³; the father and grandfather of Salonikiós were both Ottoman violin craftsmen, and his father was also a popular wedding musician.⁹⁴ Just as other trades and occupations were ethnicized by a combination of restricted labor markets and family traditions, so musical trades often became associated with particular peoples. Throughout Europe, the itinerant Roma people—known variously in this period as gypsies, manouches, Sinti, gitanos—were not only associated with several musical idioms from flamenco to tzigane, but Roma musicians were regularly recruited to popular dance orchestras. Similarly, across the Pacific, there was a diaspora of versatile Filipino musicians, who played in Hawaiian orchestras and jazz bands in Shanghai, Japan, the Dutch East Indies, and Malaya: “Filipino orchestras are the interpreters of jazz on the Pacific Ocean liners,” a journalist wrote in 1922. “Where music is concerned, Filipinos are known as the Italians of the East.”⁹⁵

If particular ethnic groups became associated with the musical trades, playing a wide range of idioms, it was also the case that musical genres became ethnicized by the recording industry. Musics from rumba to hula became so associated with particular racial, ethnic, or national groups that some musicians “passed” as that nationality: there were Puerto Rican musicians who were known as “Cubans,” Tahitians who performed as “Hawaiians.”⁹⁶

Like other artisans, these musicians constantly faced the threat of “deskilling,” as musical machines from the player piano to the gramophone, the radio and the sound film, automated their work and displaced live performances. They also faced new struggles over the legal ownership of their music, as the forms of musical copyright developed in the first decades of the century.

In many circumstances, these two classes of musicians would have little to do with each other; indeed the early metropolitan musicians’ unions rarely included “unskilled” ear musicians. But in the colonial ports, the artisan “reading” musicians were often themselves part of a subordinated elite, either the “talented tenth” of the city’s colonized or racially subordinated community or members of a mixed-race or ethnically defined in-between community. In some, though not all, cases, this led to collaborations that brought together musical idioms, rough and respectable, rural and urban.

It is a story vividly played out across the Caribbean. In New Orleans, there was often great tension between the established Downtown Creole musicians and the street musicians of the growing Uptown community of African-American migrants from Delta plantations. As Thomas Brothers has argued, the two classes of musicians were distinguished less by repertoire than by sound. For the Creoles of color, music was an artisanal trade, and their occupational, even class, identity was based on their clean, pure tones, the product of a polished and professional technique. The ear-trained street players, in contrast, distinguished themselves by their powerful volume and idiosyncratic “freak” effects.⁹⁷ New Orleans’s distinctive jazz was the product of those figures who crossed the urban boundary and counterpointed the styles: Creole musicians like Sidney Bechet and Jelly Roll Morton, and Uptown African-Americans like King Oliver and his young protégé, Louis Armstrong.

In Cuba, the national recognition of Afro-Cuban musics was, Robin Moore has argued, the result of the efforts of two distinct groups: “professional, conservatory-trained musicians”—salon composers and singers like Eliseo Grenet and Rita Montaner—and “working-class and underclass Afro-Cuban performers” like the son composer Miguel Matamoros.⁹⁸ In Trinidad, calypso grew out of a similar counterpoint, as carnival celebrations set the songs—“calindas”—of the rough stick-fighting bands of the barrack yards alongside the fancy masquerade bands of the English-speaking Creole middle class.⁹⁹

A similar counterpoint was central to the emergence of urban vernacular musics in the ports of Africa. In South Africa, there was, as David Coplan has argued, a tension between the mission-based African culture, encompassing clerks and small farmers, that developed from Christian hymns accompanied by the harmonium, and the black urban proletarian culture that adopted the guitar and concertina but not Christianity. By the 1930s, however, the prestige of African-American jazz led to the fusion, in the township dance bands, of the mission music of the educated African elite and the marabi of the shebeens.¹⁰⁰ In Lagos, Chris Waterman argued, “the distinction between ‘refined’ and ‘crude,’ or ‘civilized’ and ‘bush’ musical styles” not only reflected “a social boundary between a literate bourgeoisie and a large population of migrant workers from the rural hinterland,” but also served to protect the cultural and economic position of a Creole elite “vis-à-vis the European colonists.” In the face of this divide, “the early jùjú practitioners were cultural brokers … [who] fashioned an expressive code that linked clerks and laborers, immigrants and indigenes, the modern and the traditional, within a rhetorical framework deeply grounded in Yoruba values.”¹⁰¹ And in the Gold Coast, highlife music grew out of three streams of musical and social life: the working-class guitar bands of the palm-wine bars, the adaha brass bands that had developed from colonial military bands, and the dance orchestras of the Christianized black elite, which played waltzes, polkas, foxtrots, and tangos. By the mid-1920s, even the music of the palm-wine bars and the mission schools began to connect, as the mission-school composer Ephraim Amu wrote new lyrics to the popular palm-wine song, “Yaa Amponsah,” and the dance orchestras added palm-wine tunes to their repertoire: “the poor people who congregated outside prestigious black elite clubs in Accra like the Merry Villas and the Palladium Cinema suddenly began to hear their own local street music being orchestrated by sophisticated bands and they gave this music the name ‘highlife.’”¹⁰²

It is not a surprise that the colonial ports, with their mix of classes, races, and ethnicities, became a privileged site for the encounter between these two classes of musicians, self-taught ear players (“routine” musicians, as they were called in New Orleans) and skilled musical artisans reading scores and crafting instruments. But the colonial ports were also a privileged site for these musical counterpoints because of their distance from the major cultural capitals, the consecrated hubs of established metropolitan culture like Paris, London, Vienna, Berlin, and New York. Though the new vernacular musics would travel to the cabarets and recording studios of these imperial capitals, they did not develop in them, nor were they rooted in the popular styles of the metropolitan working classes—the Parisian café-concert, the London music hall, the Berlin cabaret, the New York vaudeville.¹⁰³

This is a crucial point about the vernacular revolution of the 1920s: the musics of the metropolitan working classes had little resonance outside their home cities, in marked contrast to the earlier worldwide impact of the European dance crazes of the revolutionary era of 1789–1848 that made the waltz and the polka ubiquitous. By the late nineteenth and early twentieth century, the plebeian “popular style” of the metropolitan working classes had been subordinated to the concert musics of the metropolitan elites, the world of Italian opera and Viennese sonata, and did not have wide-ranging reverberations.

In the colonial ports, in contrast, the plebeian musical culture that emerged from commercial dance halls and theaters as well as parading associations and street festivals attracted the attention of city elites, who were all too aware of their own distance and isolation from the centers of imperial culture. Though many were repelled by the music of the streets, and allied themselves ever more closely to metropolitan standards of “civilized” and “respectable” culture—accompanied by the carrot of uplift and the stick of censorship—this “colonial mimicry,” as it has been called, was always aware of the artificial, abstract, and “misplaced” character of the transplanted metropolitan culture. In retrospect, those members of the urban elite who embraced the music of the streets and the dance halls in the name of an independent cultural identity—creating modernist urban renaissances, often nationalist and populist in tenor—have seemed more significant. A classic instance comes from Brazil with the legendary 1926 “evening of guitar music”—analyzed in Hermano Vianna’s The Mystery of Samba—at which a group of modernist intellectuals including the anthropologist Gilberto Freyre and the composer Heitor Villa-Lobos met the sambistas Pixinguinha, Donga, and Patrício Teixeira.¹⁰⁴ For a decade after the celebrated Modern Art Week of 1922, many of Brazil’s artists and writers—most notably Mário de Andrade—heard the vernacular musics of choro and samba as the sound of a Brazilian nationality.¹⁰⁵

This meeting of subaltern musical worlds and the vanguardist intellectuals and artists of modernist renaissances is repeated around the world: in the Afrocubanismo of Havana, associated with the musicologist Fernando Ortiz and the writer Alejo Carpentier; in the invocations of blues and jazz among the writers and artists of the Harlem Renaissance, particularly Langston Hughes and Alain Locke; in Shanghai’s radical New Culture Movement, which attracted musicians like Li Jinhui, Ren Guang, and Nie Er; and in the organization of the 1922 Granada Concurso de Cante Jondo by the composer Manuel de Falla and the poet Federico García Lorca, which presented La Niña de los Peines, among others, in an attempt to revive flamenco’s deep song, the cante jondo. Each of these encounters triggered debates about music, race and nation that would, as we will see, surround the vernacular music recording boom.

To visitors, the soundscape of the colonial ports—street festivals and calls to prayer, refined hotel orchestras and dance-hall bands—seemed more like cacophony than polyphony. A characteristic impression is that of the French novelist Paul Morand, who visited Cholon, the Chinese section of Saigon in what was then French Cochin-China, in 1925: “The Chinese inns—full of noise, outcries, murderous turmoil, and the mah-jong tiles clattering all night like hail on a tin roof. The theatres—cymbals clanking, gramophones wailing, operetta singers adding their strident notes to the infernal din.”¹⁰⁶

But the voices each had their own logic, and as a result, the polyphony of the ports produced a music that was, I would suggest, richer and more complex than the segmented high and low musics of the metropolitan capitals, the isolated musics of rural peasantries, or those of the single-class mining and mill towns dominated by an industrial logic. This, I would suggest, is the answer to the oft-pondered mysteries of why samba was born in Rio¹⁰⁷ or jazz in New Orleans, as well as the equally mysterious roots of rumba and hula, highlife and marabi.¹⁰⁸

These were the local urban musical cultures that the recording engineers of the great multinational record companies encountered when they began to lug their equipment from port to port in the mid-1920s. Out of the counterpoint of recording engineer and vernacular musician came a worldwide dissemination of vernacular phonograph musics on disc, and a revolutionary remaking of the world’s musical space.