Science1 distinct publisher3 min readPublished
Eduardo Haddad and Inacio Araujo used the mint, date and findspot of coins recovered over two centuries of digging to argue that integration held the Republic together more than conquest did. The shape of the data sets the limits of that claim.
The Scientist · Science desk
Compiled by The ScientistSomething wrong?How this is made
The denominator does useful work here. Four million coins spread across the 157 years from 155 BCE to 2 CE works out to roughly 25,500 per year of the period covered [1][2][1][2], and measured against two centuries of excavation it is about 20,000 coins entering the record for every year of digging [3]. Neither average is evenly spread, because a coin becomes a findspot only when an excavation records it [1]. The geography of the sample therefore carries two centuries of fieldwork priorities inside it.
The main database is dedicated to hoards [7], and that matters for the question being asked. A hoard marks a place where coin came to rest together, not a sequence of hands it passed through. Getting from resting places to circulation is the load-bearing inferential step in a study like this, and the published account of the work does not show the bridge [12].
The stated conclusion, that consolidating Roman rule depended less on military conquest than on the economic integration of conquered territory [5], is a comparative causal claim about a century and a half of history [1] built from where objects sat in the ground. Coins arriving in a province establish that coins arrived. Whether they arrived through market exchange or through the state's own disbursements is an interpretive choice, and the summary reports the conclusion without a coefficient or a variable standing in for conquest [12].
The most convincing part of this is infrastructural rather than econometric. Records of Roman coins were long scattered across museums, libraries, private collections and researchers' own archives, which is precisely what blocked large-scale reconstruction [8]. Digitization projects coordinated by institutions including the American Numismatic Society imposed common protocols for recording archaeological information, and public databases now hold millions of standardized records from excavations in several countries [9]. Standardized fields are what make regional economics, spatial analysis and GIS applicable to the material at all [6]. It also explains how the analysis arrived from outside classical studies: Haddad photocopied a Roman coin catalog in Princeton's library in 2014, after a seminar presentation that reconstructed Mediterranean trade networks from shipwreck remains and pottery shards [10], then enrolled in a distance-learning graduate program on the ancient Mediterranean at the University of Leicester, where one of his papers became the basis for a published article [11].
What three fields cannot tell you is what anything cost. Mint, production date and findspot [3] support statements about where money went and how far it travelled; they do not carry prices, nor the speed at which coin changed hands. An integration result built on positions is a result about reach. Whether that reach raised living standards in the provinces is a separate measurement, and the fields recorded for each coin do not contain it [3].
Ranked by verification strength, evidence, and original report placement.
Two researchers from the Regional and Urban Economics Lab at the University of Sao Paulo (NEREUS-USP) in Brazil reconstructed the monetary circulation of the Roman Republic based on the remains of approximately 4 million coins found in excavations conducted over two centuries.
Haddad says each coin preserved by archaeology provides three key pieces of information: where it was minted, when it was produced, and where it was found some 2,000 years later.
The research was conducted by Eduardo Amaral Haddad, a professor at the School of Economics, Business and Accounting (FEA-USP), and Inacio Fernandes Araujo, currently a professor at the Luiz de Queiroz College of Agriculture (ESALQ-USP), who published an article in the journal Humanities and Social Sciences Communications.
The study shows that the consolidation of Roman rule depended less on military conquest and more on the economic integration of the territories Rome conquered.
The authors used techniques from regional economics, spatial analysis, geographic information systems and large international archaeological databases.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · September 3, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
science
A 324-million-year-old stem insect still carried paddles on nine abdominal segments4 distinct publishers
product
Princeton Critical Minerals' $16M buys its way into other people's ponds1 distinct publisher
science
PPPL's PACMAN framework closed a 20-millisecond AI control loop inside a real tokamak1 distinct publisher
science
NASA's COFFIES model flags solar active regions before they surface, averaging 9.24 hours2 distinct publishers
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Peer-reviewed core, single-outlet retelling
The inputs are unusually checkable for a story like this - CHRR, ORBIS and Pleiades are public, and the paper cleared review at Humanities and Social Sciences Communications. The output is not. Every figure reaching a reader here passes through one outlet quoting one of the two authors, and the step that matters most, turning buried hoards into a picture of circulation, is described only as a way to reduce preservation bias.
Borrowed infrastructure, no uptake of its own
Split the question and the picture clarifies. The infrastructure is genuinely adopted: standardised protocols, millions of pooled excavation records, four separate public platforms wired together for this project. The study itself has nothing yet - no replication, no citing work, no sign anyone outside NEREUS-USP has run the method on another dataset.
One-line conclusion, no magnitude behind it
'Less about conquest, more about integration' is a large claim about how an empire held together, and it is asserted once and then dropped in favour of method description. Scale does some rhetorical work too: four million coins is a striking number until you notice it covers 157 years of minting drawn from two centuries of digging, roughly twenty thousand coins a year of excavation. The overreach is in the framing, not the data collection, which is described soberly.
One lab, telling its own story in its own voice
This reads the way institutional science communication reads: departments named in full, Haddad quoted in four extended passages, a charming hobby-to-publication arc, and no one on the page positioned to say the finding is thinner than it sounds. That is a mild pull rather than a serious conflict - there is no product, no funding round and no commercial stake anywhere in the piece.
Firm on the how, agnostic on the how much
We can be fairly sure what was built and from what, because the datasets are named and public and the curation is spelled out. We cannot judge whether integration beat conquest, because nothing in this reporting quantifies either side. A second account, or the paper's own numbers, would move this quickly in either direction.