The Incas: Quechua and the khipu

The realm of four parts

The Incas were originally one of many Andean peoples. According to the traditional chronology, their small state centred on Cusco in what is now southern Peru began to expand in 1438, when Pachacuti (Quechua Pachakutiy, “turn of the world”) took power. He and his son Túpac Yupanqui conquered most of present-day Peru within a few decades, and their successors went further. By the early 16th century the empire stretched from the southern tip of today’s Colombia through Ecuador, Peru and Bolivia to northwest Argentina and central Chile. It was the largest empire of pre-Columbian America.

The Incas called their realm Tawantinsuyu. Quechua tawa means “four” and the suffix -ntin marks a group that belongs together, so the name means roughly “the four parts together”. The four regions (suyu), Chinchaysuyu in the northwest, Antisuyu in the northeast, Kuntisuyu in the southwest and Qullasuyu in the southeast, met at Cusco. How many people lived in the empire is unknown; most estimates range between 6 and 14 million. The word inka referred to the ruler (the Sapa Inka) and to members of the ruling nobility, who are estimated to have numbered only 15,000 to 40,000. Today it is usually applied to all the inhabitants of the empire.

Runa simi: a language or a family?

Quechua is not a single language but a language family. Its varieties differ from one another roughly as much as the Slavic languages do, and speakers from distant regions cannot understand each other. Since the work of Gary Parker and Alfredo Torero in the 1960s, linguists have divided it into two main branches: Quechua I (Central), in the highlands of central Peru, and Quechua II (Peripheral), which includes the northern varieties of Ecuador and Colombia and the southern ones from Ayacucho through Cusco to Bolivia and northwest Argentina. The Central branch is the most diverse, which is why the homeland of the family is sought in central Peru.

Quechua was spreading long before the Incas. When they rose to power, many peoples already spoke it and it had already split into regional varieties; the Incas were just one of those peoples. The empire, however, carried it further. As the official language and lingua franca, a common language between people of different mother tongues, it was brought by administrators and resettled colonists (mitmaqkuna) to places where it had not been spoken before, such as present-day Ecuador and southern Bolivia. Many conquered peoples learned it while keeping their own languages.

Which Quechua served as the language of Inca administration is disputed. It was long assumed to be the speech of Cusco. A more widespread view today is that the administration relied on a coastal variety from the south-central coast of Peru, home to the rich and populous realms of Chincha and Pachacamac; this variety died out after the conquest. Traces of it seem to survive in the earliest loans into Spanish: inga was pronounced for inka and kundur for kuntur, hence Spanish cóndor. The name quichua was first applied to the language by Domingo de Santo Tomás in 1560; what its speakers called it before the conquest is unknown. The modern native name runa simi, “human speech”, also seems to have emerged only in the colonial period.

The languages of the empire and the language of the Incas

Tawantinsuyu was linguistically diverse. On the Altiplano in Qullasuyu Aymara prevailed, Puquina was spoken on the south coast, Mochica (above all in Lambayeque) and Quingnam on the north coast, and dozens of other languages on the fringes of the empire, many of them now extinct. Aymara shares much vocabulary and some striking structural features with Quechua, and the two families have therefore sometimes been joined as “Quechumaran”. Most specialists reject this hypothesis and explain the similarities by long contact and mutual borrowing. The ejective and aspirated consonants of Cusco Quechua, absent for instance in Ayacucho Quechua, are probably also of Aymara origin.

What language the Incas spoke among themselves is an open question. In the early 17th century Inca Garcilaso de la Vega wrote that the Incas had a particular language of their own, which they spoke only among themselves and which others were not allowed to learn. The evidence is scanty: a few words and proper names in colonial sources. One hypothesis, advanced above all by the Peruvian linguist Rodolfo Cerrón-Palomino, holds that it was a language close to Puquina; others suggest Aymara or an Aymara-influenced Puquina. None of these hypotheses has been proven. Most scholars agree, however, that Quechua was not the Incas’ original language, even though it is often called “the language of the Incas” today.

Khipu: records in knots

Writing in the narrow sense, signs that record spoken language, was apparently unknown to the Incas. Yet the administration of their empire relied on records: khipus (Quechua khipu or kipu, “knot”). A khipu consists of a main cord from which other cords hang, only a few on some khipus and several thousand on the largest; further subsidiary cords can be attached to the pendants. The cords were spun from cotton or from the wool of llamas, alpacas and other camelids. The information lies in the type and position of the knots, but also in the colour of the cords, their order, the direction of their twist and the way they are attached.

Khipus are older than the Incas. The earliest undisputed examples are khipus from about 600–1000 CE attributed to the Wari state. They are smaller than Inca khipus, with brightly wrapped cords and a system of knots that is not yet understood. Earlier finds from Caral (c. 3000–1800 BCE) and from the Late Paracas period are disputed as forerunners. The Incas used khipus to record censuses, taxes and labour service (mit'a), storehouse contents, the army and the calendar.

Khipus were tied and read by the khipukamayuq, “khipu specialists”. According to colonial chronicles, sons of the nobility learned to read them in the third year of the school called yachay wasi, “house of learning”. When Hernando Pizarro travelled along the royal road from the highlands to the coast in 1533, he recorded how khipukamayuq untied knots in one section of a khipu and retied them in another, probably as they issued supplies to the Spanish. This is probably the earliest European mention of khipus. Guaman Poma de Ayala later wrote that the khipukamayuq could “read” them with their eyes closed.

Did khipus record language?

That khipus carry numbers was shown in 1912 by the American Leslie Leland Locke, who described the decimal positional system of their knots. From the late 1960s Marcia Ascher and Robert Ascher analysed several hundred khipus and demonstrated that most knots record numbers that can be systematically interpreted: the value of one cord is often the sum of several others, sometimes even a sum of sums. Some groups of digits, however, apparently do not express quantities and serve as codes, rather like modern postal codes; the Aschers called them number labels.

Whether khipus also preserved language, such as names, narratives or histories, is unknown. According to colonial sources, the khipukamayuq supplied the Spanish with information on censuses, tribute, rituals, the calendar and genealogies, and Inca historians used khipus when recounting the empire’s past; whether the story lay in the knots or only in the teller’s memory cannot be told from the sources. Gary Urton regards khipus as a system of binary-coded signs that could also have carried linguistic information. In 2005 he and Carrie Brezine proposed that a sequence of three figure-eight knots at the start of khipus from Puruchuco near Lima may stand for the name of that place, or for the khipu maker or the subject. In 2011 a possible match was described between six colonial khipus from the Santa valley and a Spanish register from the same area.

Sabine Hyland argues that colonial “epistolary” khipus from the village of San Juan de Collata, probably from the time of an 18th-century rebellion against the Spanish, record the names of two kin groups (ayllus) phonetically, through the type of animal fibre and the colour of the cords. Since these are colonial objects, European writing may have influenced them. Frank Salomon, by contrast, sees khipus as a semasiographic system that conveys meaning without being tied to the sounds of a particular language, like musical notation or numerals. The authenticity of the so-called Naples manuscripts, according to which khipus recorded syllables, is questioned by specialists. There is no generally accepted decipherment, and khipus are therefore not usually regarded as writing in the narrow sense.

How many khipus survive is not known exactly. The Spanish destroyed many: in 1583 the Third Council of Lima ordered the burning of those that recorded offerings to the old gods. In 2003 Gary Urton estimated the number of surviving khipus at about 600; a 2021 survey counted just under 1,400 in museums and private collections. In some Andean villages, such as San Cristóbal de Rapaz, khipus are still kept as communal heritage, even though no one can read them any more.

Tocapu: decoration or signs?

The second Andean system in which scholars look for writing is tocapu (Quechua tuqapu): small geometric motifs set in squares or rectangles. They were woven into textiles, above all into the tunics (unku) of rulers and high nobles, and were also painted on wooden boards and on qiru ceremonial cups. On textiles they are repeated singly, in bands, for example along the lower edge of a tunic, or in no apparent order.

Because the same motifs appear on different objects and their selection does not look random, scholars have asked whether tocapu formed a system of signs: emblems of lineages and provinces, ideograms, or even writing. In the early 17th century the chronicler Juan de Santa Cruz Pachacuti Yamqui reported that Manco Capac introduced the custom whereby each province and town chose a sign of its origin, its paqarina (the sacred place from which a lineage emerged); some scholars link this to tocapu. According to Christiane Clados, the Incas borrowed some motifs from the art of the earlier Tiwanaku and Wari cultures. No one has proposed a decipherment, and whether tocapu recorded anything remains unclear.

Conquest and the first books

In November 1532 Francisco Pizarro with 168 men captured the Inca ruler Atahualpa at Cajamarca; Atahualpa had just won a civil war against his half-brother Huáscar. He paid a ransom in gold and silver for his release, but the Spanish executed him in 1533 all the same. The last Inca state, at Vilcabamba, fell in 1572, when Túpac Amaru was captured and executed. Quechua did not disappear with the empire: the Spanish adopted it as the “general language” (lengua general) for administration and above all for missionary work.

The Dominican Domingo de Santo Tomás came to Peru around 1540 and learned Quechua. In 1560 he published in Valladolid a grammar, the Grammatica o arte de la lengua general de los Indios de los Reynos del Perú, and a dictionary, the Lexicon o Vocabulario de la lengua general del Perú. They are the oldest substantial records of the language. They probably reflect a coastal variety, close according to some scholars to the language of Inca administration, with elements of other dialects. A chair of Quechua was set up at Lima Cathedral in 1571, and in the late 1570s knowledge of the language became compulsory for priests in Peru.

The turning point was the Third Council of Lima (1582–1583). It had basic church texts translated into Quechua and Aymara and so laid down a common written standard, now called Classical or Colonial Quechua. It was based on Southern Quechua but avoided words understood only in Cusco. In 1584 Antonio Ricardo, a printer who had come from Mexico, printed in Lima the trilingual Doctrina Christiana y catecismo, generally regarded as the first book printed in South America. An anonymous grammar and dictionary followed in 1586. The council’s translations of prayers are still used, with dialect adjustments, today.

Classical Quechua was used for sermons, confessional manuals, letters, petitions and land titles. The most valuable text is the Huarochirí Manuscript (dated to 1598 or 1608), written by Indigenous authors under the direction of the priest Francisco de Ávila: it records, directly in Quechua, the myths and rituals of the Andean province of Huarochirí. From the mid-17th century colonial policy turned towards Hispanicization, and after the suppression of the rebellion of Túpac Amaru II (1780–1782) the authorities banned Quechua literature and theatre and abolished the chair of Quechua at the University of San Marcos in Lima.

Two chroniclers

Inca Garcilaso de la Vega (1539–1616) was born in Cusco, the son of a Spanish conquistador and an Inca princess; Quechua was his mother tongue. In 1560 he left for Spain, where he stayed for the rest of his life. His Comentarios reales de los Incas (Royal Commentaries of the Incas, 1609) describe the history, religion and customs of the Incas, explain the meaning of Quechua words and also mention the numerical records on khipus. The book became the most widely read account of the Inca past; after the rebellion of Túpac Amaru II the Crown banned it.

Felipe Guaman Poma de Ayala (c. 1535 – after 1616; the year of his birth is uncertain and may have been as late as 1550) was not an Inca but came from another Indigenous noble family. He spoke several varieties of Quechua and Aymara. As an interpreter he accompanied the priest Cristóbal de Albornoz on a campaign against so-called idolatry, and later he illustrated the chronicle of the Mercedarian friar Martín de Murúa. Between 1600 and 1615 he wrote and drew El primer nueva corónica y buen gobierno (The First New Chronicle and Good Government), addressed to King Philip III of Spain: 1,189 pages, including 398 full-page drawings. The chronicle tells Andean history from before the Incas and is the most extensive critique of colonial rule written by an Indigenous author in the whole colonial period. It is written mostly in Spanish but with Quechua words, phrases and whole passages, which makes it a valuable source for the language as well.

The king never received the chronicle. The manuscript has been in the Royal Library in Copenhagen since the 1660s at the latest, where the German scholar Richard Pietschmann discovered it in 1908. A facsimile was published in Paris in 1936, a critical edition by John Murra and Rolena Adorno in 1980, and since 2001 the whole manuscript has been available online.

Quechua today

According to censuses from 2001–2017 taken together, just under seven million people speak Quechua in Peru, Bolivia, Ecuador, Colombia, Argentina and Chile (Ethnologue 2022 gives 7.2 million). Censuses probably undercount speakers; older estimates from around 2004 gave 8 to 10 million. It is the most widely spoken Indigenous language family of the Americas. In Peru’s 2017 census about 3.7 million people, just under 14 per cent of the population, gave it as their mother tongue. In 1975 Peru became the first country to make Quechua an official language; in Bolivia the 2009 constitution lists it among the official languages, and in Ecuador, where it is called Kichwa, it is the second most widely spoken language after Spanish.

After centuries of oppression Quechua is mainly a rural language, and many families send their children to Spanish-medium schools in the hope of better prospects. Peru, Bolivia and Ecuador have therefore introduced intercultural bilingual education and train interpreters for health care and the courts, and Peru’s national radio broadcasts in Quechua. In 2019 a doctoral thesis written in Quechua was defended at the University of San Marcos in Lima for the first time. Quechua has borrowed hundreds of Spanish words (waka “cow” from vaca, iskwila “school” from escuela), and Latin American Spanish has in turn borrowed from Quechua: papa “potato”, wawa “baby” in Bolivia and soroche “altitude sickness” in Peru.