Faces of Ancient Rome: How DNA Recreated 16 People from Aquincum

Why reconstruct the people of Aquincum?

Archaeological reports can describe skeletons with precision while leaving the people behind them emotionally distant. The Aquincum Museum exhibition “Once We Were Like You” addresses that distance by presenting possible faces and historically grounded stories for people who lived on Rome’s Danube frontier almost two thousand years ago.

The characters include a blacksmith, stable worker, soldier, enslaved person and construction labourer. Their names and detailed biographies are not recovered facts. Curators combined biological evidence with well-established knowledge of Roman occupations and social conditions to create plausible museum narratives.

Aquincum as a frontier city

Aquincum stood in the northern part of modern Budapest and became an important military and civilian centre in the province of Pannonia. Forts, baths, workshops, houses and trade networks made it a busy Danubian settlement. Frontier cities were not isolated camps: soldiers, merchants, craftspeople, families and local communities moved through them.

The cemeteries of Aquincum therefore preserve evidence for mobility across the empire. Inscriptions and grave goods can suggest social roles, while bones record work stress, nutrition, disease, injury and ancestry.

How a face is built from a skull

Forensic facial reconstruction begins with anatomical constraints. The eye sockets, cheekbones, jaw, nasal opening and muscle-attachment areas limit the possible proportions of the face. Specialists place estimated soft-tissue depths at defined landmarks before modelling muscles and skin.

The result is not a photograph. Ear shape, hairstyle, facial expression and some aspects of body composition cannot be recovered with certainty. A reconstruction is a biologically plausible model, not a unique solution.

What DNA added to the models

According to the Associated Press report, DNA from the remains helped estimate probabilities for hair, eye and skin colour and the likelihood of freckles. Ancient samples may be incomplete, so these predictions must be presented with confidence ranges rather than as absolute facts.

DNA also suggested broad biogeographic connections. The exhibition includes people with ancestry linked to Italy, present-day Scotland and Syria, as well as Sarmatian and local Celtic populations. The diversity demonstrates that Roman identity was not a single ethnic category.

Evidence of work, hunger and injury

Curators reported that many of the studied skeletons carried inflammation and wear associated with regular physical labour. Some also showed signs consistent with periods of nutritional stress. Those findings informed the decision to emphasize ordinary and lower-middle-status residents rather than only elites.

A character called Respectus was imagined as a construction worker who plastered walls and split stone. Stress and injuries in his skeleton provide the factual core; the tavern incident and exact occupation are narrative reconstructions designed to make the evidence understandable.

Where science ends and storytelling begins

Age, biological sex, fractures, disease and some genetic traits can be inferred from remains. Names, precise jobs, family relationships and individual episodes usually cannot. The exhibition’s method depends on clearly labelling that boundary.

When the invented elements are transparent, storytelling can humanize the past without pretending to recover a complete biography. When they are presented as direct outputs of DNA, the method becomes misleading.

Mobility in the Roman Empire

The ancestry results fit a world connected by roads, military recruitment, trade and forced migration. Units could be transferred across provinces, merchants settled near forts, enslaved people were displaced and local groups adopted new legal and cultural identities.

“Roman” could describe citizenship, military service, urban life, language or participation in imperial institutions. A person with Syrian ancestry or a Sarmatian background could still be an integral resident of a Roman provincial city.

Why ordinary faces matter

Written history disproportionately preserves the voices of rulers and wealthy men. Labourers, children, enslaved people and the urban poor appear less often. Bioarchaeology partly corrects this imbalance because bodies preserve experience regardless of social rank.

A reconstructed face, combined with evidence for hard work or childhood illness, shifts attention away from monuments alone. It reminds visitors that Roman cities were maintained by people whose names rarely entered official history.

Silicone models and digital reconstructions

Six of the sixteen faces were produced as life-sized silicone models by artist Emese Gábor. Hair, clothing and jewellery were applied to create objects that can be viewed from every angle. Gábor contrasted this physical encounter with AI-generated faces that remain on a screen.

Digital tools are useful for testing multiple possibilities quickly, while physical models create a stronger spatial presence. Both can support education when the underlying data and artistic choices are documented.

Ethical and interpretive limits

Displaying human remains and reconstructed identities requires care. Museums should explain burial context, avoid stereotypes and mark fictional details. Genetic ancestry should not be converted directly into modern nationality or ethnicity.

Biogeographic affinity describes statistical similarity across reference populations. It does not reveal a person’s language, legal status, beliefs or self-identification. Those distinctions are especially important for a cosmopolitan frontier community.

What the Aquincum faces teach us

The exhibition brings Roman history closer to blacksmiths, children, soldiers and labourers rather than leaving it with emperors and marble portraits. Anthropology and DNA narrow the possibilities; historically informed art turns those limits into a visible encounter.

The reconstructions should be read as evidence-based possibilities. Their value lies not in claiming photographic certainty, but in showing the diversity, mobility and physical costs of life in an imperial city.

Frequently asked questions

Are the Aquincum faces exact likenesses?

No. Skull anatomy and DNA constrain the model, but hairstyle, expression and several soft-tissue details remain interpretive.

How many reconstructions are displayed?

The exhibition features 16 reconstructions, including six life-sized silicone models.

Are the names and occupations real?

No. They are historically grounded fictional narratives informed by the biological evidence.

Can DNA determine a Roman person’s ethnicity?

It can indicate broad ancestry connections, but not language, culture, citizenship or personal identity by itself.

Sources

• Associated Press: Face-to-face with ancient Rome
• Euronews report on the Aquincum exhibition
• Aquincum Museum, “Once We Were Like You” exhibition materials.