A story through the atlas
The chilli, followed
One plant, thirteen food worlds, and only seven dates anyone wrote down. What the chilli meant in each place it reached, in the words of the people who were there, ending each time on a dish you can cook.
Last changed 12 September 2026
How to read this
The chilli is the atlas's argument told through one plant. It was domesticated in the Americas, crossed the Atlantic on the first ships back, and within a century and a half was being written down in Nahuatl, Spanish, Portuguese, Chinese, Korean and Thai. Every stop below is a food world the atlas has written, and every stop ends on a dish the atlas carries with its quantities and timings.
Two rules hold throughout, and they are the reason this reads differently from the chilli story you will find elsewhere. A date rests only on a text that names the plant. Where no such text was found, the stop says so in its first sentence, gives the best evidenced bracket, and says whose it is. Six of the thirteen stops have no such text: West Africa, the Maghreb, Goa, the Deccan, Java and Guam. The gaps are not failures of research; the research found them, in the languages of each place, and they are part of what the plant's history is. What is widely repeated and could not be stood up is gathered at the end under "Deliberately not asserted", with the names of who tells it.
before 1521, written down 1575 to 1577
1. Central Mexico and Puebla
The plant is at home here, and the language shows it. In Nahuatl the word is chilli, and it is a root that grows compounds the way the plant grows pods: chilmolli, a chilli sauce; chilatolli, a maize gruel thickened with it. The Franciscan Alonso de Molina, compiling his dictionary in 1571, glossed chilli as "axi o pimienta de las indias", pepper of the Indies, because the Spanish had already met the plant in the Caribbean under its Taíno name and were carrying that name inland.
The text that names it is the Florentine Codex, Bernardino de Sahagún's Historia General de las Cosas de Nueva España, completed around 1575 to 1577 in Nahuatl and Spanish with Nahua scribes and informants. Chilli appears across at least six of its twelve books. Book X holds a market passage on the good and the bad chilli seller, naming varieties: chiltecpin, chilcoztli. The chilli was food, and it was also a marker of sacred time. During the feast of Macuilxóchitl a four day fast excluded it; the penitential priests, the tlamacazque, went without it too. Its absence marked the holy days as surely as its presence marked ordinary ones.
The dish is mole poblano, from the volume on the Valley of Mexico, Puebla and Oaxaca. Several separate preparations, each right on its own terms, brought together only at the end; the chilli base comes straight down from that vocabulary. What nobody wrote down: no precontact Nahuatl text survives on its own. Everything above is a 1570s compilation of remembered Mexica practice, made under a friar's direction, and the market passage is cited here at the level of the dictionary's index, not the manuscript page.
15 January 1493
2. Haiti
The earliest date anyone wrote down for the chilli outside the mainland is 15 January 1493, and it is at Hispaniola. Columbus's diary, in the transcription Bartolomé de las Casas made because the original is lost, records: "También hay mucho ají, que es su pimienta, della que vale más que pimienta, y toda la gente no come sin ella, que la halla muy sana." There is also much ají, which is their pepper, worth more than pepper, and all the people do not eat without it, finding it very wholesome. Ají is the Taíno word, from haxi. The plant was cultivated, valued above the imported black pepper it was being compared to, and eaten with everything.
What it means in Haiti now is a condiment and a feast. Pikliz is placed among the offerings at Fèt Gede, the Vodou feast of the ancestral dead on 2 November, when the living bring favourite foods to graves and altars. The dishes are pikliz, the condiment that goes with almost every fried thing, and griot, the fried pork at the centre of fritay, both from the volume on Haiti.
What nobody wrote down, or what could not be found: no source published in Haiti itself, in French or in Kreyòl, on the history of piment, of pikliz or of griot as composed dishes. The Dominican historian Bernardo Vega records Taíno fighters using chilli smoke against Spanish forces, which is the shared island's story rather than Haiti's. The leap from "the chilli was here and prized in 1493" to "pikliz as it is made today is old" is not supported by anything found, and it is not made here.
1535
3. Iberia
The chilli arrived in Spain as a curiosity before it was a food. Nicolás Monardes, the Sevillian physician who wrote up the plants of the Indies, said of it: "no hay jardín ni huerta, ni macetón que no la tenga sembrada por la hermosura del fruto que lleva." There is no garden, plot or pot that does not have it planted for the beauty of the fruit it bears. He credited it with medicinal virtues, and then noted, almost as an afterthought, that in stews it made a better taste than the ordinary pepper.
The text that names it with a date is Gonzalo Fernández de Oviedo's Historia General y Natural de las Indias, printed in Seville in 1535, which already distinguishes varieties of ají by shape and colour: "Ají que echa granos o vainas coloradas y redondos y tan gruesos como guindas", ají that bears red, round pods as fat as sour cherries. That quotation reaches this page through an academic chapter from the Universidad de Extremadura rather than through a digitised copy of the 1535 printing, which could not be read in this pass; it is said so plainly.
The same chapter gives pimentón de la Vera a firmer foundation than the story usually told about it. Not a monastery receiving Columbus's seeds, but the 1753 and 1791 interrogatorios of the Real Audiencia de Extremadura and the Catastro de Ensenada, which record production in La Vera rising from a thousand arrobas at mid century to three thousand by 1791. Pimentón as an industry is eighteenth century and documented.
The dishes are all i pebre, the eel stew of the Albufera, garlic, pimentón, eel and nothing else, from Catalonia, Valencia, Murcia and the Balearics, and chanfaina extremeña, the pimentón dish of the western Meseta, from Castile, Madrid, Extremadura and Aragón.
No text names the plant
no text names the plant; by the turn of the sixteenth century at the Portuguese footholds, an estimate
4. West Africa
No text names the plant here with a date. That is the first thing to say, because the one usually offered does not hold. Duarte Pacheco Pereira's Esmeraldo de Situ Orbis, written between 1505 and 1508, uses "Malagueta" again and again, and it means Aframomum melegueta, grains of paradise, the pod of the ginger family that gave the Grain Coast its name in a trade already running under Afonso V, decades before Columbus. It cannot be the chilli. The best that can be said is general Columbian Exchange scholarship: that Capsicum reached the Portuguese footholds at São Tomé, Príncipe, Cape Verde and Angola by the turn of the sixteenth century. That is an estimate about islands and the southern coast, not a date for the Bight of Benin or the Hausa interior, and it is given here as an estimate.
What the languages say is more useful than the missing date. In Hausa the chilli is barkono, the spice mix for grilled meat is yaji and the skewer is tsire. In Yoruba the word ata now covers the chillies in daily use, ata wewe the bird's eye, ata rodo the Scotch bonnet, and it is distinct from ata ire, the grains of paradise still used in naming ceremonies. The new plant took the old word for heat and the old plant kept a qualified name. That is the pattern this story will meet again in Thailand and Java.
The dish is suya, from Ghana and Nigeria: the yaji rub, chilli and ground peanut on grilled beef, the Hausa street food that carries the plant's West African name. What nobody wrote down: no Hausa or Yoruba language source, and no Nigerian scholarly one, dated and specific enough to date the chilli's arrival or the age of yaji and tsire. Searches for barkono and yaji history returned nothing on point. The gap is stated because it is real.
between 1520 and 1770, Tewolde Berhan Gebre Egziabher's bracket
5. The Horn
One text names the plant, and it is an absence followed by a presence, read by an Ethiopian scientist. Tewolde Berhan Gebre Egziabher, Ethiopia's first qualified plant ecologist, argued in 1984 that Francisco Álvares, in Ethiopia from 1520 to 1526 and published in 1540, describes the royal kitchen's spicing in detail and never names a chilli, an omission that matters for a plant so demanding to grow; and that James Bruce, in Ethiopia from 1769 to 1771, is the first to name "Cayenne pepper", mixed with black pepper and spicing raw meat. Tewolde's own conclusion: "it can be concluded that chili pepper was introduced into Ethiopia in the two and a half centuries between 1520 and 1770." A bracket of two hundred and fifty years, and it is his, not settled fact.
The word is older than the plant. Berbere, barbaré in Ge'ez and Amharic, is glossed as pepper or hot, and Tewolde records an epithet form, quenedo barbaré, which suggests a Ge'ez word for pepper existed before it was qualified for the newcomer. The blend that carries the name now is chilli at its base, with garlic, ginger, sacred basil, cloves and fenugreek.
The dish is doro wat, dorho tsebhi, the feast dish of both countries, from Ethiopia and Eritrea, built on berbere as its structure rather than its seasoning. What nobody wrote down: no Amharic or Tigrinya language source on berbere's own history was found despite direct searches in both, and the Eritrean side of the story rests on Ethiopia side scholarship. Manuel de Almeida's seventeenth century Historia de Ethiopia is often suggested; no passage naming the pepper could be found in it, so it grounds nothing here.
No text names the plant
no text names the plant
6. The Maghreb
No text names the plant here with a date. Two claims circulate for harissa's arrival in Tunisia, and neither is cited to a document: a sixteenth century arrival under the Spanish occupation of 1535 to 1574, and a seventeenth century arrival with the Moriscos expelled from Spain who settled Cap Bon and Nabeul from 1609. A Doha based outlet names a researcher tracing the Morisco route through "travel literature" and gives no title, author or date for it. Neither claim survives the rule this story holds itself to, and neither is asserted.
What can be said with certainty is what harissa is and what its name means. The name is technique, not plant: "L'appellation même de la harissa provient du verbe arabe harsa qui veut dire broyer ou écraser", the name comes from the Arabic verb to pound or crush, as La Presse de Tunisie put it in 2021. UNESCO inscribed it on the Representative List of the Intangible Cultural Heritage of Humanity on 1 December 2022, and the inscription text names salt, garlic and coriander with the dried red pepper, and no governorate, no Cap Bon and no variety. The chilli was folded into a mortar tradition that already existed, and took the mortar's name.
The dish is harissa arbi, from Algeria, Tunisia and Libya: sun dried red peppers, deseeded, pounded with salt, garlic and coriander, the paste that is a condiment and the base of half the region's stews. What nobody wrote down: a Tunisian or Maghrebi text, in Arabic or French, naming the pepper with a date.
No text names the plant
no confirmed text; mid sixteenth century, a Western secondary claim
7. Goa
The famous text does not say what it is usually said to say. Garcia de Orta's Colóquios dos Simples e Drogas da India, printed in Goa on 10 April 1563, is exactly the kind of source this story wants: local, dated, a physician writing about plants where he lived. It discusses pimenta across many of its dialogues. Read directly for this story, and checked against a 2013 survey written to answer precisely this question, Orta's pimenta is Piper nigrum, Malabar black pepper, the subject of his pepper dialogues. No passage naming the American chilli could be found. Orta's relevance to the chilli is not established, and it is not asserted here.
What the stop rests on instead is weaker and is named as weaker: the claim, traced to Lizzie Collingham's Curry, that by the mid sixteenth century chillies were eaten in Goa as "Pernambuco pepper", which would put established varieties within about thirty years of Vasco da Gama's 1498 landfall. Her own primary citation for it could not be located. It is the next earliest claim available, and that is all it is.
What is certain is the dish and what it carries. Vindalho is Portuguese vinha d'alhos, the wine vinegar and garlic marinade that kept meat on long voyages; in Goa palm vinegar took the wine's place and the chilli was added. Fatima da Silva Gracias, the Goan historian, writing in 2004: "chillies which are widely used in Goan and Indian cuisine were a stranger to our continent until the Portuguese introduced them from the Americas." The dish is vindalho, from the volume on Goa, preservation turned into flavour. What nobody wrote down: any Konkani or Marathi source, in its own script, on the shift from maricha, black pepper, to mirchi, the chilli.
No text names the plant
no text names the plant
8. The Deccan
No text names the plant here with a date. The one candidate is famous and could not be found. The story goes that Purandara Dasa, the Kannada composer of the fifteenth and sixteenth centuries, wrote a song, "vah vahre menasinakaayi", describing the chilli turning from green to red, tasty in a dish and too hot in excess. In everything found, that rests on the food historian K. T. Achaya alone, without a page reference to his own text. For this story the pada was searched for directly in Kannada, across two full text archives of Purandara Dasa's collected compositions, once by its opening words and once by the word menasinakayi. It was not there. The claim is named, with Achaya's name on it, under "Deliberately not asserted", and it dates nothing here.
What the language says instead is the same pattern as West Africa. The Telugu word for the chilli, mirapakaya, comes through miriyapukayalu from the Sanskrit marīca, the word for black pepper. The new plant was folded into the old word for the older spice rather than given one of its own. Guntur's rise as a chilli centre, the fact behind the region's modern reputation, could not be dated from anything better than an unsourced "about three decades" for the research station at Lam.
The dish is naatu kodi pulusu, country chicken in a roasted spice paste and tamarind gravy, the everyday grammar of Andhra meat cookery, from Andhra, Telangana and Karnataka. What nobody wrote down: any Vijayanagara or Golconda era Telugu text naming the pepper.
China 1614 to 1629; Sichuan 1749; adopted 1820s to 1860s
9. Sichuan and Chongqing
This is the stop where the story most often told is most clearly wrong, and where the Chinese scholarship is best. The usual claim is that Gao Lian's Zunsheng Bajian, the Eight Discourses on the Art of Living of 1591, is China's earliest chilli text: "番椒:丛生,白花,子俨秃笔头,味辣,色红,甚可观,子种", foreign pepper, growing in clumps, white flowers, pods like the tip of a worn writing brush, hot, red, very fine to look at, grown from seed. Cheng Jie of Nanjing Normal University compared three surviving copies of the 1591 Yashangzhai printing in the National Library and found the entry absent from the actual first printing, present only in a Chongzhen era reprint decades later. On that evidence 1591 is not defensible.
The safer text, on Cheng's authority, is Wang Xiangjin's Qunfangpu, compiled between 1614 and 1621 and printed in 1629: "番椒,亦名秦椒。白花,子如秃笔头,色红鲜可观,味甚辣." Cheng argues further, from the timing of the Korean records, that the plant's first confirmed entry to China was through the Korean peninsula into Shandong, a separate route from the usual Zhejiang one. The word itself was foreign first: fānjiāo, foreign pepper, on the pattern of the foreign sweet potato; hǎijiāo, sea pepper, in the Sichuan dialect; làjiāo, hot pepper, displacing both over the Qing.
Sichuan's own date is kept apart from the national one, because it is a century and a half later. Liang Pinghan of the Southwest University of Finance and Economics gives the earliest Sichuan record as the Dayi county gazetteer of 1749, "荤菜类:秦椒,又名海椒", and notes that the provincial gazetteers of the Yongzheng and Jiaqing eras carry no chilli entry at all. General adoption in Sichuan cooking dates only from the late Jiaqing era and becomes common in the Tongzhi era, the 1860s. "辣椒虽然最早自东南沿海传入,但是直到它传入西南地区才找到了自己大显身手的舞台": although the chilli first entered by the southeast coast, it only found its true stage on arriving in the southwest. The dish is mapo tofu, from Sichuan and Chongqing, the pairing of the American plant's straight heat with the native peppercorn's numbing that the province arrived at last, and made its own.
1614 for the plant, 1766 for the paste
10. Korea
Two dates, kept apart because they are different things. The plant is named in 1614, in Yi Su-gwang's Jibong yuseol, as 남만초, namman-cho, the southern barbarian pepper, also called 왜개자, waegaeja, Japanese mustard, because it was believed to have come by way of Japan. The names before the word settled tell the story of a stranger being placed: 번초, foreign pepper; 고초, bitter pepper, the direct ancestor of the modern 고추, gochu. The 1614 folio itself could not be read for this story; the date and the naming are corroborated by the Academy of Korean Studies' Encyclopedia of Korean Culture and by the National Institute of Korean History.
The paste is later, and the text usually cited for it is the wrong one. Sallim gyeongje, attributed to Hong Man-seon who died in 1715, is often called the earliest gochujang text; 1715 is his death year, not a date on the book, and the work carries no gochujang recipe. The recipe first appears in Jeungbo Sallim gyeongje, compiled by Yu Jung-rim in 1766: one mal of meju, three hop of red chilli powder, one doe of sweet rice flour. Gyuhap chongseo of 1809 names Sunchang and Cheonan as gochujang places, corroborating rather than predating it.
The dish is Jeonju bibimbap, from the Korean Peninsula, gochujang as the sauce that binds everything in the bowl. There is a dispute here and it is named rather than smoothed over. Kwon Dae-young, formerly of the Korea Food Research Institute, argued at a 2018 seminar that no record shows the chilli arriving through Japan during the Imjin War, and that a genetic divergence between two Korean varieties points to a native origin. The scholars who responded held the standard transmission account plausible or called for more study; none endorsed a native origin, and no peer reviewed genomic work places any Capsicum lineage in East Asia. The 1527 dictionary Kwon cites glosses a syllable used for peppery fruits in general, not the chilli, according to the Kyung Hee University etymological study that is the mainstream correction.
1558 to 1657, on Krit Luealamyai's reading of Ayutthaya tax records
11. Thailand
Before the chilli, the Thai word phrik meant black pepper, Piper nigrum, from Mon and from Pali and Sanskrit marica. When the American plant arrived it was named phrik thet, foreign pepper, and over time it took the bare word for itself, so that black pepper had to become phrik thai, Thai pepper, to stay distinct. The plant that came last now owns the name, and the native one wears the qualifier. This is Krit Luealamyai's account, a Thai food history writer who works from old textual evidence, in an essay of 2017 whose title asks whether the wrong pepper has been named.
The date is his too, and it is given as his. Krit places the arrival in the twenty second Buddhist century, roughly 1558 to 1657, "พริกเข้ามากับเรือสำเภาในช่วงพุทธศตวรรษที่ 22", the chilli came in with the trading junks, citing Ayutthaya period tax records that name phrik thet specifically. The tax record itself could not be checked for this story, only his citation of it. What is confirmed unambiguously is later: Sunthorn Phu's verse of around 1800, "พระไม่เคยจะเสวยพริกเทศเผ็ด", the lord had never eaten the hot phrik thet, names the plant beyond doubt. Simon de la Loubère's account of Siam of 1691 is read by Krit as describing the dipping sauces without naming the chilli, a reading not verified here against the French. A Thai food history site put the position plainly in 2026: there is no primary evidence that definitively states the earliest period of Capsicum in Siam.
The dish is tom yum goong, from Angkor and Ayutthaya, the sour and hot broth that could not exist as it is without the plant that took the pepper's name.
No text names the plant
no text names the plant
12. Java
No text names the plant here with a date. What is repeated across the Indonesian press, "circa 1522, via Sunda Kelapa", comes with no document naming the plant at that date, and it is not asserted. The obvious candidate, Tomé Pires's Suma Oriental of 1512 to 1515, written from visits to Java, could not be confirmed to name Capsicum, and the timing argues against it: those years predate most estimates for the plant's spread this far. Rumphius's Herbarium Amboinense, written between the 1660s and 1690s and printed in 1741, genuinely names Capsicum in the archipelago, but its island is Ambon, not Java.
The language, once more, says what the documents do not. Cabya, Piper retrofractum, a relative of betel pepper, is attested in Old Javanese inscriptions from the tenth century. When Capsicum arrived, cabai was applied to it, and the older native plant was pushed to the marked term cabai jawa, Javanese chilli. The same displacement as in Thailand and in Yoruba, a third time.
The dish is rujak cingur, from Java and Madura, recognised as Indonesian intangible cultural heritage in 2021. Its own origin is a genuine three way dispute, and the story keeps it that way: an Arekan Surabaya identity claim, carried in the strongest academic work found and dating popular memory only to "the Dutch colonial period"; a 1930s Madurese migrant popularisation, attributed to a named Surabaya history activist, in which cakalang petis was adapted to shrimp petis for a Javanese palate; and an Abdul Rozak folk legend that Indonesian food media itself has flagged as circulated and retracted. What nobody wrote down: any Javanese language chronicle reference to cabe from the sixteenth or seventeenth century.
No text names the plant
no text names the plant; the 1600s, uncited
13. Guam
No text names the plant here with a date, and here the reason is structural. CHamoru had no writing before Spanish contact in 1668, so there can be no precontact textual reference, only archaeology, genetics and what was passed down. Guampedia's own inventory of the crops the CHamoru grew before and at contact, rice, taro, yams, breadfruit, coconut, bananas, sugarcane and ginger, does not include donne', which is consistent with the plant arriving after. Guampedia says it is "thought to have been introduced by the Spanish in the 1600s", alongside the Manila galleon resupply stops documented at Guam from the 1670s, and gives no citation for the specific claim.
The strongest candidate for a text was found and could not be read. Verónica Peña Filiu's study of foodways and missionary accommodation in the Mariana Islands between 1668 and 1674 works directly from the Jesuit mission letters of exactly these years. Every route to it and to her fuller thesis returned an access block. It has not been read and its relevance to the chilli is not established either way; that is stated rather than guessed at. No Guam or Marianas specific Capsicum genetic study was found either, so this stop is thinner than a documentary or a genetic date would make it.
The words are certain. Donne' is the CHamoru hot pepper; donne' sali' the small, very pungent wild one; donne' ti'au a longer, backyard grown type. The dish is kelaguen, from Guam and the Northern Mariana Islands, meat or seafood cured in citrus with coconut and donne' sali', the plant's last stop on this route, where islanders now often take it for native. The route began with a plant so at home in Mexico that its absence marked the holy days. It ends with one so at home in the Marianas that its arrival has been forgotten.
Deliberately not asserted
The claims below are widely repeated and could not be stood up to the rule this story holds itself to. Each is named with who tells it.
The Yuste monastery story: that Columbus's seeds went by way of the monks of Guadalupe to the Monasterio de Yuste in La Vera, whose humid autumns forced the monks to invent smoke drying. Told by the Denominación de Origen Protegida Pimentón de la Vera on its own site and repeated by the brands. The one academic source that engages with it cites the historiographical majority holding that the Catholic Monarchs received Columbus in Barcelona, not at Guadalupe, and the smoke drying detail has no independent source at all.
The 1591 date for Gao Lian's chilli entry: disputed by Cheng Jie's comparison of the surviving printings. See stop 9.
"Circa 1522, via Sunda Kelapa" for Java, and "the 1600s" for Guam: both without a document naming the plant. See stops 12 and 13.
The Spanish occupation and Morisco migration claims for the Maghreb: neither cited to a primary document. See stop 6.
Purandara Dasa's chilli song: rests on K. T. Achaya alone; the pada was searched for in Kannada and not found. See stop 8.
That Sichuan has always eaten hot: its own gazetteers carry no chilli entry until 1749, and adoption dates from the nineteenth century. See stop 9.
Gochujang's "thousand year history", and a native pre Columbian Korean chilli: the first rests on a syllable for peppery fruits in general; the second is Kwon Dae-young's argument, real, addressed and not the mainstream position. See stop 10.
That Garcia de Orta's 1563 Colóquios names the chilli, and the "Pernambuco pepper" claim for Goa: the first attempted directly and not established; the second attributed to Lizzie Collingham, whose primary citation was not located. See stop 7.
That griot's name descends from West African griot storytellers, or that pikliz as made today existed in colonial Saint-Domingue: repeated across French language recipe sites with no historian behind them. See stop 2.
Kelaguen's debt to Filipino kilawin: plausible, given the Filipino presence in Spanish era Guam, but no CHamoru or Guam academic source makes the linguistic claim with a citation. And the claim that fina'denne' before the Spanish "was simply salt and pepper", incoherent for a New World plant and uncited where it appears.
Where this comes from
The research brief behind this chapter, with every source, the searches run in each language, and the count of local against Western sources at every stop, is on file as briefs/story-chilli.md and was written to the rules on How we research. Two stops have no source published in the place at all, Haiti and West Africa, and say so. Thirteen stops, seven with a text that names the plant, six that admit they have none: that is the honest shape of the chilli's history as it can be documented today, and the atlas would rather show you the shape than fill it in.
Read it in the atlas