Connect with us

Archaeology

Archaeologists use artificial intelligence (AI) to translate 5,000-year-old cuneiform tablets

Published

on

A team of archaeologists and computer scientists have created an AI program that can translate ancient cuneiform tablets instantly using neural machine learning translations.

In a paper published in the journal PNAS Nexus, from the Oxford University Press, the researchers have applied the AI program to translate Akkadian texts with a high level of accuracy.

Akkadian is an ancient East Semitic language, was once spoken in various regions of ancient Mesopotamia, including Akkad, Assyria, Isin, Larsa, Babylonia, and possibly Dilmun.

The language is preserved on clay tablets dating back to 2500 BC that was written using cuneiform, a script adopted from the Sumerians using wedge-shaped symbols pressed in wet clay.

According to the researchers: “Hundreds of thousands of clay tablets inscribed in the cuneiform script document the political, social, economic, and scientific history of ancient Mesopotamia. Yet, most of these documents remain untranslated and inaccessible due to their sheer number and limited quantity of experts able to read them.”

The AI program has a high level accuracy when translating formal Akkadian texts such as royal decrees or omens that follow a certain pattern. More literary and poetic texts, such as letters from priests or tracts, were more likely to have “hallucinations” – an AI term meaning that the machine generated a result completely unrelated to the text provided.

The goal of the neural machine translation (NMT) into English from Akkadian is to be part of a human–machine collaboration, by creating a pipeline that assists the scholar or student of the ancient language.

Currently, the NMT model is available on an online notebook and the source code has been made available on GitHub at Akkademia. The researchers are currently developing an online application called the Babylonian Engine.

PAP

https://doi.org/10.1093/pnasnexus/pgad096

Header Image Credit : Mbzt – CC BY 3.0

Continue Reading

Archaeology

Maya tomb with funerary offerings found during hotel construction

Published

on

By

A tomb with funerary offerings has been uncovered during the construction of the Tren Maya Hotel, in Palenque, Mexico.

Archaeologists from the National Institute of Anthropology and History (INAH) conducted rescue excavations following the discovery, revealing skeletal remains associated with the Maya city of Palenque.

Palenque, also known as Lakamha in the Itza Language (meaning “Flat-Place-River”), is located near the Usumacinta River in the Mexican state of Chiapas.

The city dates from 226 BC to AD 799, with most of the major construction works representing a rebuilding effort in response to attacks by the city of Calakmul and its client states in AD 599 and AD 611.

The population declined during the 8th century AD, instead becoming an agricultural population that led to the abandonment of the city zone. By 1520 following the Spanish conquest, contemporary Spanish accounts record the entire region being sparsely populated.

Excavations uncovered a stone lined funerary space sealed with limestone slabs, in which the researchers found the remains of a high-status individual who likely lived in the periphery of Palenque in a small settlement.

The burial is located at a depth of four metres, and also contained ceramic vessels and beads deposited as funerary offerings.

“The individual was placed face up with his legs extended and his head facing north,” said Diego Prieto Hernández from INAH.

The discovery was announced in a press conference reporting on the progress of the Program for the Improvement of Archaeological Zones (Promeza) in Palenque, Moral-Reforma and El Tigre, the three heritage sites that are served within Section 1 of the Mayan Train Project.

INAH

Header Image Credit : INAH

Continue Reading

Archaeology

Archaeologists unearth possible birthplace of King Henry VII at Pembroke Castle

Published

on

By

Archaeologists from the Dyfed Archaeological Trust may have discovered the possible birthplace of King Henry VII at Pembroke Castle in Pembrokeshire, Wales.

Pembroke Castle was founded during the 11th century by Roger de Montgomerie, the 1st Earl of Shrewsbury.

In 1452, Jasper Tudor was presented both the castle and the earldom by his half-brother, King Henry VI. In 1457, Henry VII was born at the castle, the only child of Lady Margaret Beaufort and Edmund Tudor, 1st Earl of Richmond.

Following Edward IV’s ascension to the throne in 1471, Henry VII endured 14 years of exile in Brittany. He eventually claimed the throne after his forces, with backing from France, Scotland, and Wales, emerged victorious over Richard III at the Battle of Bosworth Field, marking the climax of the Wars of the Roses.

Henry VII’s reign is credited with many administrative, economic, and tax reforms, having ruled for nearly 24 years until his death in 1509 at the age of 52. Henry VII was succeed by his second son, Henry, Duke of York, who ascended to the crown as Henry VIII.

Historians have long assumed that a 13th century tower on the outer ward (known today as Henry VII Tower) was the birth place of Henry VII. However, a recent study by the Dyfed Archaeological Trust has uncovered evidence of a late-medieval winged hall-house, broadly dating to the 15th century.

The walls of the structure extend to around 25 metres, with comparisons being drawn to medieval buildings found in England and East Wales, such as Penallt Mansion in Kidwelly. Historically, Pembroke Castle was situated in the English-speaking portion of Pembrokeshire, often referred to as “Little England beyond Wales.”

Speaking to the Western Telegraph, Neil Ludlow, a consultant to Pembroke Castle, said: “All our indications are pointing to a late-medieval building which was clearly of high status within Pembrokeshire, and it looks as if it was at least two-storeys, which possibly makes it a better candidate for the birthplace of a king rather that the tower that currently bears his name.”

Continue Reading

Trending

Generated by Feedzy